Home

ESAP-95 Version 2.01R 95 Version 2.01R User Manual User

1. Project Description My first field project Import Filename c US_Salinity_Lab esap2 demo_input_files bwd101p dat Project Subdirectory Demo Field ID Code BVWD Survey Type Transect of Survey Sites 1017 of Signal Columns 2 Signal Transformation none Signal Statistics N Mean Std Min Max Corr EMv 1017 1 158 0 225 0 681 1 929 0 9850 EMh 1017 0 749 0 145 0 424 1 239 27 3 3 Data Visualization The graphics routines in ESAP RSSD allow you to visualize your conductivity survey data using a number of different plotting techniques These various plotting techniques are designed to help you better understand and interpret your survey data and thereby assist you in the sampling design process In general you will create and view the majority of your graphs immediately after importing your conductivity survey data However some graphs can only be viewed after certain additional computations have been performed and other graphs may contain additional visual information after performing such computations as discussed below and in the on line help documentation 3 3 1 Description of Graphical Techniques There are six distinct types of graphs which the ESAP RSSD program can produce as described below Survey Grid This plot will display the x y location coordinates of your entire input data file allowing you to verify that the survey data coordinates were correctly scanned into the ESAP program Note that every survey gri
2. 4 0 3 5 Generating sampling designs 3 5 1 The manual sample site selection procedure 3 5 2 Generating spatial response surface SRS sampling designs 3 5 3 Practice module 3 6 Data Output 3 6 1 Text output 3 6 2 Graphical output 3 6 3 Creating an output svy data file 3 6 4 Practice module ESAP SaltMapper Software Program 4 1 ESAP SaltMapper program overview 4 1 1 Program description 4 1 2 Navigating the main menu 4 2 Data input and manipulation 4 2 1 Setting the current project and importing data 4 2 2 Renaming columns 4 2 3 Creating new data columns 4 2 4 Displaying basic column statistics 4 2 5 Practice module 4 3 Creating 1D line plots 4 3 1 Definition of a 1D line plot 4 3 2 Navigating the 1D line plot menu 4 3 3 Line plot initialization options 4 3 4 Adding a title axis labels and a legend to the plot 4 3 5 Practice module 4 4 Creating 2D raster maps 4 4 1 About the raster map creating process 4 4 2 Using the on screen interpolation controls 4 4 3 Navigating the raster map menu 4 4 4 Setting the map variable specification options 4 4 5 Practice module vi 47 49 49 51 52 52 53 53 53 59 59 59 60 62 62 63 63 65 65 66 66 67 68 69 70 71 71 12 74 76 77 5 0 4 5 Migrating data out of the ESAP SaltMapper program 4 5 1 Creating and saving ASCII text files 4 5 2 Practice module ESAP Calibrate Software Program 5 1 5 2 5 3 5 4
3. 5 5 5 6 ESAP Calibrate program overview 5 1 1 Program description 5 1 2 Navigating the main menu Importing previously saved ESAP data files 5 2 1 Importing ESAP RSSD survey data files 5 2 2 Importing ESAP Calibrate profile data files 5 2 3 Practice module Importing new data files 5 3 1 The two types of laboratory profile data 5 3 2 Importing data file type column structure specifications 5 3 3 Practice module Editing and validating new profile data 5 4 1 Validating new laboratory profile data 5 4 2 Editing laboratory profile data 5 4 3 Practice module Deterministic salinity conversion modeling 5 5 1 How does a deterministic conversion work 5 5 2 Navigating the DCS menu 5 5 3 Setting the conductivity to ECa input parameters 5 5 4 Setting the secondary soil input parameters 5 5 5 Practice module Performing a 1D profile analysis 5 6 1 What is a 1D profile plot 5 6 2 Navigating the 1D profile menu 5 6 3 Initializing a 1D profile plot for display 5 6 4 Practice module Vii 79 80 80 81 81 81 84 87 87 88 89 90 90 92 94 96 96 97 98 99 99 100 101 103 104 106 106 107 109 110 5 7 Performing a 2D standard correlation analysis 5 7 1 5 7 2 5 7 3 What is a standard correlation analysis Navigating the standard correlation analysis menu Practice module 5 8 Performing a DPPC correlation analysis 5 8 1 5 8 2 5 8 3 5 8 4 5 8 5 5 8 6 5 8 7 What i
4. About ESAP RSSD Select this option to display the ESAP information window ESAP Interface Controls ReSet default Text Editor You may use this option to change the default text editor used by ESAP to display and print all ESAP generated text files and or print any ESAP help file At program start up the default text editor package is set to c windows write exe OnLine Help What is ESAP RSSD Select this option to display the introductory help documentation This is the help documentation you should read first if you have never used the ESAP RSSD program before This documentation explains what the ESAP RSSD program does and how the program works OnLine Help Navigating the Main Menu This documentation explains how to use the Main menu bar and describes the menu bar features OnLine Help Frequently Asked Questions FAQ This is the help documentation you should read if need more detailed information about how to use the various program features For example refer to this documentation if you don t understand how the signal decorrelation and validation process works or what a transition analysis does or when to log transform your input conductivity survey data etc 3 2 Data File Input Specifications The following section describes how to import conductivity survey data into the ESAP RSSD software program Included in this section are directions for setting up your project and field ID code and formatting your input conductivi
5. Demo BVWDinfo txt 11 29 1999 10 20 04 AM inity_ inity_ My first field project iab esap2 demo_input_files bwd101p dat iab esap2 data Demo BVWDinfo txt FieldID Code BVWD File Type Transect rowcount 16 Survey Size 1015 of Signals 2 Signal Units dS m Sig Trnsfrm natural log Signal Statistics LnEMv LnEMh N 1015 1015 Mean 0 129 0 306 Std Oa TOT 0 190 Min 0 385 0 8 98 Max 0 657 0 214 Signal Corr 0 9839 Decorrelation Statistics Level Sites gt Level Masking STD 3 50 3 Outlier STD 4 50 0 Total of deleted Sites 2 of Validation iterations 1 Edge Buffering invoked Transition Analysis none 56 Table 3 2 Hard copy print out of the BVWDrsd1 txt text file Full Path Project File Name Date amp Time Field Desc Sample Size D Factor Val Opt Criteria Loop Count Target Site ID 934 247 99 168 324 888 73 437 876 612 943 557 C US_Salinity_Lab esap2 data Demo BVWDrsdl txt Demo BVWDrsdl txt 11 29 1999 1 28 15 PM My first field project 12 Total Survey Size 1015 Activ Survey Siz 1 00 1 20 9 Information for SRS Sampling Design 1 Design Levels Dsl STD Ds2 STD X Coordinate 0 75 0 0 73 0 04 270 1 75 L75 1 64 1 62 72 Fert breed As SLES 153 2203 36 TzS i 1 92 N 54 o 1D 1 34 1 67 108 25 0 2 63 0 12 252 eee 0 2019 0 56 36 0 2 5 0 28 2 17 126 0 Oo 0 34 2 46 252 0 75 0
6. a file with the extension pro which can in turn be easily imported back into the ESAP Calibrate program during any future analysis session As described above the primary purpose of the ESAP Calibrate program is to generate salinity predictions from conductivity survey data When you generate these predictions using either stochastic or deterministic modeling techniques you can save an output prediction file a file with the extension prd Therefore the ESAP SaltMapper program has been designed to import either svy or prd data files Hence you can easily generate either observed soil conductivity or predicted soil salinity maps by simply using the output ESAP RSSD or ESAP Calibrate program data files You can also use the ESAP SaltMapper program to export either type of imported ESAP data file svy or prd as an ASCII text file which can in turn be used as input to some other software program for example into Surfer or Arcview Each ESAP 95 program can also be used to generate a number of additional output text and or graphics files All three programs can generate graphical output 1 e plots or maps which can be printed and any plot or graph generated in either the ESAP Calibrate or ESAP SaltMapper program can be saved as a bitmap graphics file Additionally a number of text summary output files can be created and saved by both the ESAP RSSD and ESAP Calibrate programs For example the ESAP RSSD program will automatically c
7. decorrelation on this data in addition to the centering and scaling computations Keep in mind that you must open this window to initiate the decorrelation algorithm and this algorithm must be invoked before attempting to validate your decorrelated signal data Invoking the Signal Validation Algorithms After your survey data has been decorrelated your next step should usually be to open up the Signal Validation window by clicking on the Analysis gt Signal Validation main menu options The Signal Validation algorithm can be used to validate your decorrelated input conductivity survey data If outlier survey data was detected during the decorrelation analysis 43 then the validation algorithm will offer you two options You can either 1 delete the outlier survey data Sites and then perform a new signal decorrelation or 2 you can elect to mask these outlier Sites instead and end the validation process If desired you can also perform either an edge buffering or a transition analysis on your decorrelated survey data from this window provided you are working on a Transect type input file Handling Outlier Signal Data When one or more outlier survey data Sites are detected by the ESAP RSSD program the recommended approach is to delete the outlier Sites and perform a second decorrelation and validation iteration Indeed this is why the signal decorrelation and validation routines and windows have been designed the way they are From
8. re calculating the model and then using the re calculated model to predict the removed point You should use the 2D Prediction Scatter Plot Initialization window to initialize the specific type of prediction scatter plot you wish to display and to control which sub sets of predictions are displayed This window can be invoked and displayed from the 2D Prediction Scatter Plot program window which can in turn be invoked by clicking on the Predict gt View Prediction Plots SCM menu option Directions for navigating the 2D Prediction Scatter Plot menu are given below 2D Prediction Scatter Plot Menu Bar Layout The ESAP Calibrate 2D Prediction Scatter Plot menu bar is located in the upper left corner of the 2D Prediction Scatter Plot program window The full layout for this menu bar system is shown below Main level Sub Level 2 Sub Level 3 Plot gt Specify New Plot Variables gt Create Plot Output gt Print gt Obs vs Prd gt Obs vs JK Prd gt Both Plots gt Obs vs Prd gt Obs vs JK Prd Help gt Navigating the 2D Prediction Scatter Plot Menu gt General Tips how to plot a different depth 148 Exit gt Return to the SCM Menu 2D Prediction Scatter Plot Menu Bar Menu Item Descriptions The 2D prediction scatter plot menu contains 4 menu bar items Plot Output Help and Exit You can use the Plot menu to initialize and create different types of prediction scatter plots the Output menu t
9. survey area of interest typically an individual field Once this conductivity data has been acquired soil samples from a limited number of sites within the survey area are then collected The locations of these sample sites are usually based upon an analysis of the acquired soil conductivity data this is referred to as target or model based sampling This sample site selection and acquisition process represents the second step in the survey After the soil samples are removed from the field they are generally sent to a laboratory and analyzed for salinity content and or any other soil chemical or physical quantities of interest This laboratory data can then be compared to the measured soil conductivity data acquired at the corresponding sample sites and used to calibrate the conductivity data In other words a mathematical or statistical calibration model can be estimated using the survey and soil sample data associated with the sample site locations This calibration modeling represents the third step in the salinity survey process Once such a model had be estimated it is in turn used to predict the soil salinity levels at all of the remaining non sampled conductivity survey sites across the field Maps of this prediction salinity data are then created displayed and studied Such a model may also be used to make other types of predictions such as the average salinity level within the field or the expected degree of crop loss du
10. then you do not need to enter any additional plotting information However if you select one of the two custom panel graph options then you will need to specify additional plotting information as described below Standard Panel Select this option if you wish to display the calculated ECa against the measured ECe texture volumetric water content and bulk density by sampling depth This is referred to as a standard panel plot because these are the four soil variables which are used to determine the soil ECa levels If you did not acquire bulk density data then estimated bulk density levels calculated from the soil texture levels will be displayed in panel 4 Custom Panel 1 Calc ECa v s 4 Secondary Variables Select this option if you wish to display the calculated ECa against four user selected secondary soil variables by sampling depth If you choose this option then a second plot input information frame will appear within the Graph Options Window You should use the list boxes displayed within this frame to select the four secondary variables you wish to plot the ECa against You may also use the controls contained within this frame to indicate if either the calculated ECa or any all of the secondary soil variables should be log transformed Custom Panel 2 Ave Calc ECa v s depth specific Secondary Variable Select this option if you wish to display the calculated bulk average ECa against a single secondary soil variable acro
11. 0 71 0 06 180 support site 0 11 0 06 270 support site 0 22 0 12 162 Ordered Listing of Sample Sites Sample AANA BWNE erer NRO Site ID Row 73 2 99 2 168 3 247 4 324 6 437 7 557 9 612 10 876 14 888 14 934 t5 943 15 306 681 431 306 807 Y Coordinate 25 25 25 25 43 75 718 25 143 568 443 193 243 106 TS T T9 T3 TS 75 57 this page intentionally left blank 58 4 0 ESAP SaltMapper Software Program 4 1 ESAP SaltMapper Program Overview The following section gives an overview of the ESAP SaltMapper software program including a review of how the program functions and a description of the main menu bar layout 4 1 1 Program Description What is ESAP SaltMapper ESAP SaltMapper is a graphics program which can be used to generate display and plot high quality 1 D transect graphs and 2 D raster maps of either your observed soil conductivity data or predicted soil salinity or secondary soil variable data The ESAP SaltMapper program is part of the ESAP software package for Windows a multi program software package designed and distributed by the United States Salinity Laboratory for the sampling assessment and prediction of soil salinity and or other soil variables from electrical conductivity survey data The ESAP SaltMapper program is designed to read in i e import the output data files produced by the ESAP RSSD and ESAP Calibrate programs svy
12. 00 718 75 0 9246 0 6378 25 1017 288 00 731 25 0 9246 0 6652 Example 3 Transect file type no site ID column structure x y sl s2 row T800 TIZET 1 1414 0 7538 18 00 781 25 1 2086 0 7720 18 00 768 75 1 2848 0 8240 L 288 00 718 75 0 9246 0 6378 16 288 00 731 25 0 9246 0 6652 16 Example 4 Transect file type no 2nd signal data column structure site id x y sl row 1 138 00 193 75 1 1414 2 18 00 781 25 1 2086 3 18 00 768 75 1 2848 L 1016 288 00 718 75 0 9246 16 1017 288 00 731 25 0 9246 16 The File Structure and Import Window can be used to a define your conductivity survey data file column structure and b read in your input Grid or Transect data file This window is activated when you select the menu options Import a Grid Survey File or Import a Transect Survey File from the main ESAP menu 3 2 3 Practice Module Importing Data If you have not already done so start up the main ESAP 95 splash screen by clicking on the ESAP 95 prompt listed within the Programs drop down menu system i e click on Start gt Programs gt Esap95 Once the splash screen has displayed click once on the ESAP RSSD command button to invoke the RSSD program Once the ESAP RSSD main menu screen is fully displayed click on the File gt Set Create Project and Field ID menu options This will cause the Project amp Field ID window to be displayed Normally you would make i e set
13. 1 23 al 0 67 and a2 0 43 and you should enter these parameter values into the appropriate a0 al and a2 parameter text boxes etc If you wish to enter your own custom set of parameter values highlight the Single Eqn option Otherwise highlight the Rhodes Eqn option to use the 1992 Rhoades equations to convert EM 38 signal readings into depth specific soil conductivity If you highlight this latter option you can use the displayed command buttons shown at the bottom of Parameter Values frame to select from one of three prediction depths 1 e either 0 0 0 3 0 3 0 6 or 0 6 0 9 meters respectively Note that this option is only available if you have collected both horizontal and vertical EM 38 signal readings and you have specified a log linear formula type The two examples shown below represent the sort of information you might typically need to specify within the Conductivity to ECa window 102 Example 1 Suppose that conductivity survey data was collected using an automated trailer system containing 2 EM 38 units with both units positioned 0 1 meters above the soil surface with the Ist unit held in a horizontal mode and the second in a vertical mode Suppose also that we wish to estimate the soil salinity in the 0 6 0 9 meter depth range using the 1992 Rhoades equations and that the average soil temperature within this depth was 21 degrees C Then the following information should be specified 1 In the temperature facto
14. 133 5 9 2 Navigating the SCM Menu SCM Menu Bar Layout The ESAP Calibrate SCM menu bar is located in the upper left corner of the SCM program window This window can be invoked and displayed by clicking on the Calibrate gt Stochastic Methods gt Spatial MLR Analysis main menu option The full layout for this menu bar system is shown below Main level Sub Level 2 Sub Level 3 Model gt Select Response Variable gt Identify Model Parameters gt Estimate Calibration Equation gt Advanced Options gt View model summary l statistics gt View residual summary l diagnostics gt View residual plots Predict gt View prediction plots gt Calculate field summary statistics gt Save output predictions Test gt Calculate Net Flux tests Help gt What is a stochastic calibration model gt What information is provided by the l Advanced modeling options gt Navigating the SCM Menu gt Frequently Asked SCM Questions Exit gt Return to Main Menu 134 SCM Menu Bar Menu Item Descriptions The profile menu contains 5 menu bar items Model Predict Test Help and Exit You can use the Model menu options to identify estimate and if desired appraise your stochastic calibration model the Predict menu options to view prediction plots calculate field summary statistics and create and save your output predictions the Test menu option to perform Net Flux tests and the Help menu if you wish to acces
15. 158 Advanced Residual Summary Diagnostics If you wish to view any advanced residual summary diagnostics click on the Model gt Advanced Modeling Options gt View Residual Summary Diagnostics SCM menu option The ESAP Calibrate program will then display an MLR residual diagnostic summary sheet This sheet displays univariate r student residual summary statistics for each estimation depth HAT leverage and r student residual values for each depth and the residual correlation matrix If requested the Moran residual spatial correlation test statistics will also be shown at the bottom of this sheet As before this summary sheet will be displayed using your default text editor and you will also have the option of saving this summary sheet as a permanent ASCII text file Advanced Residual Plots In addition to the two summary sheets discussed above the ESAP Calibrate program can be used to produce a number of useful residual diagnostic plots These plots can be used to graphically access the adequacy of the fitted regression model If you are familiar with the use of various graphical residual diagnostic techniques then you will find these plots to be very useful for model validation purposes There are six types of diagnostic plots available for you to use and each plot can be displayed using the Residual Diagnostics Plot window To invoke and display this window click on the Model gt Advanced Modeling Options gt View Residual Plots SCM m
16. ESAP Calibrate generated prd data files If you attempt to import any other type of data file then the ESAP SaltMapper program will display an error message and halt the data file import process 4 2 2 Renaming Columns Once you have imported a data file you may change or rename any existing column labels To display the Rename Columns window click on the File gt Column Manipulation gt Change Column Labels menu options You can then use the options within this window to change the current column labels Note that if you requested that all log transformed data columns be back transformed during the data import process you will probably need to change your column labels i e to remove any In or log prefixes 4 2 3 Creating New Data Columns In addition to changing column labels the ESAP SaltMapper program also gives you the capacity to create one or more new data columns from two or more existing columns of data To create a new data column you need to invoke and display the Create Data Column window This window can be displayed by clicking on the File gt Column Manipulation gt Create a New Column menu options Any new data column you create must be generated from a combination of existing data columns already present in your data file Two types of data columns can be created by the ESAP SaltMapper program a new ratio column or a new linear column A new ratio column is created from a ratio of two or more exi
17. File button to save the ASCII text file You should click on the Finished command button when you are done viewing and or displaying your DPPC profile data correlation results Note that before the DPPC PDCA Window disappears you will be asked if you would like to add the calculated column of ECac data into your active profile data file Answer yes if you would like to be able to analyze this ECac data using any of the other ESAP Calibrate stochastic data analysis methods otherwise answer no 5 8 4 Understanding the DPPC Summary Report The DPPC PDCA summary report contains all of the results from the complete correlation analysis displayed in an easy to read table like format A standard summary report will contain the following 4 sub sections I ECac Correlation Structure II Bulk average ECac v s bulk average Primary Secondary soil variable Correlations Ill Bulk average and depth specific ECac v s depth specific Primary Secondary soil variable Correlations IV ECe ECac Signal Deterioration by depth A composite summary report will contain the 4 sub sections listed above along with a 5th sub section entitled V Bulk average amp depth specific Primary Secondary soil variables v s z1 signal Correlations The output associated with each of these sub sections is described in detail below I ECac Correlation Structure This section displays the correlation estimates between the depth specific and bulk average
18. V Bulk average amp depth specific Primary Secondary soil variables v s zl signal Correlations This final section will only be displayed if you are performing a composite DPPC analysis i e if you have merged your conductivity survey and soil profile data together This section displays the correlations between your zl signal conductivity data and each of the depth specific and bulk average soil profile variables In the example shown below the z1 signal data appears to be fairly well correlated with the log salinity measurements from the 0 45 0 75 and1 05 sample depths 0 742 0 770 and 0 714 as well as the bulk average measurement data 0 801 However the correlations between the z1 signal data and remaining soil variables range from marginal to poor V Bulk ave amp depth specific Pri Sec Soil Variables v s zl signal Correlations Pri Sec Soil Variable sample depth levels O15 0 45 0S 10 5 ave ln ECe 0 476 0 742 0 770 0 714 0 801 SP 0 273 0 574 0 592 0 293 0 498 Vol H2o0 0 114 0 641 0 565 0 421 07595 Bulk Den S0382 020 0AL 020 0 352 Note The correlations shown in section V above actually represent worst case scenarios for this survey sample data set In general an estimated regression model can produce a better correlation between the conductivity and calibration sample data due to the inclusion of 1 both conductivity signal readings rather than just a single average value and 2 additional trend surface
19. a current project by clicking once on the appropriate Project directory listed in the Set Current Project Directory frame If you have just installed the ESAP 95 Software Package you will only have two projects available Training and Training2 However for this practice session you will create a new project To create your new project click once on the New command button located within the New Project Directory frame type in the name Demo without the quotes and then click on the Create button Note that you have just declared your project to be Demo in the Project Field ID Information frame and the ESAP RSSD program has added this subdirectory to the project list 26 Now you need to enter a field description and a four alpha numeric character ID code Click on the field description text box and type in My first field project Next click on the ID text box and type in BVWD After you have done this you should see an OK command button appear within the Project Field ID Information frame click on this OK button The Project and Field ID window should now disappear and you should see a Project Status window appear on the ESAP RSSD main program form which lists your current project field ID and the field description Next click on the File gt Import Survey Data File gt Import a Transect Survey File main menu options This will cause the File Structure and Import window to be displayed which you can then use to impo
20. and 12 0 dS m and a sorghum crop grown under these predicted field salinity conditions would suffer an expected 53 5 yield loss Once you are done viewing the summary statistics you should save your output salinity predictions to a permanent ESAP prd prediction data file To do this click on the Predict gt Save Output Predictions SCM menu option When the Save Output Predictions window appears click on the Specify Output File Name command button type in your output file name for example ece_sk13 prd click on the windows Open command button and then the Save File command button You should then see an ESAP Calibrate message stating that your output file has been saved If you desire you can now leave the ESAP Calibrate program start up the ESAP SaltMapper program and create a 2D raster map of your predicted salinity data the ESAP SaltMapper program is described in detail in Chapter 3 An example map of the predicted bulk average salinity data created using the ESAP SaltMapper program is shown in figure 5 8 154 Table 5 2 Calibration modeling results field summary statistics for Training2 salinity prediction data Ts Field Average Poi depth eles 45 215 1 05 average OOO NNN DN LN nt mean 46941 38669 43849 x39909 4431 Estimates 1n ECe variance 95 Confidence Interval 0 00510 2 314 to 2 625 0 00640 2 212 to 2 561 0 00509 2 283 to 2 594 0 00449 2 253 to 2 545 0 00266 2331
21. and display a 2D raster map To invoke and display the 2D Raster Map window click on the Graphics gt 2D Raster Image Map main menu option Once this window displays you should begin the data interpolation process The start this process click on the Initialize Raster Window command button this re sets i e initializes the interpolation memory Next click on the Show I Grid Survey Sites command button You should now see your survey locations shown in red overlaid on the ESAP SaltMapper interpolation grid At this point if you wanted to you could either clip the interpolation grid or load up a previously saved boundary file However you will be able to adequately control this interpolation process by simply using a small kernel size Hence click on the Set Kernel Size command button and then use the two displayed horizontal scroll bars to set the displayed kernel height and width to 6 Next click on the Done command button and then on the Perform Interpolation button The ESAP SaltMapper program should now perform the interpolation process and then display the interpolated area as a green zone on the grid Since this interpolated zone closely matches the area containing the original survey sites go ahead and accept the interpolation grid i e highlight the Yes option and click on the OK button The ESAP SaltMapper program will then inform you that your grid interpolations have been accepted now you are ready to produce your
22. box and hence the area to be removed After doing this you will see that section of the interpolation grid disappear and the background color of the clipped area change from white to yellow This visually identifies this area as clipped and SaltMapper will avoid calculating any interpolated predictions within this region Please note that in this current version of the ESAP SaltMapper program you can only clip out rectangular regions from the interpolation grid However you can clip away as many sections of the grid as desired and your clipped regions can overlap each other Hence if you need to remove an odd shaped region from the interpolation grid you can do so by clipping away a series of appropriately shaped rectangles which cover the region etc When you are done clipping the grid simply click on the Done button and proceed to step 4 72 Step 3b Load a Clip Boundary File optional If you have previously worked on your survey data and saved a clip boundary file you may use this command button to import this previously saved boundary file When such a file is imported during the raster map initialization process all of the previously saved clipping actions which in turn defined the field boundaries are automatically recalled and restored The previous horizontal and vertical kernel values will also be restored see step 4 below however these values can be adjusted if desired This boundary recall feature can be v
23. button the map will be erased from memory and this printing option will not work correctly from within Help About the Raster Map Creation Process Select this option to display the About the Raster Map Creation Process help file This help file gives an introduction to how the map creation process works Using the On Screen Interpolation Controls Select this option to display the On Screen Interpolation Controls help file This help file explains how to properly use the on screen window controls to create the initial interpolation grid Navigating the Raster Map Menu Select this option to display the Navigating the Raster Map Menu help file This help file explains how to use the 2D raster map menu commands to create display print and or save the map images from within Exit Return to Main Menu Select this option to close down the 2D Raster Map Initialization window and return to the ESAP SaltMapper main menu 4 4 4 Setting the Map Variable Specification Options You should the controls contained within the Map Variable Specification window to choose the variable you wish to map and to specify the visual appearance of the displayed raster image To invoke and display this window click on the Plot gt Specify Map Variable raster map menu option By default all raster maps are displayed using four shaded zones you can specify the boundaries of each zone by specifying the values of the raster cut off levels Additionally
24. calculated conductivity ECac data In the example shown below the correlations between the log transformed bulk average ECac and the ECac associated with the 0 15 and 1 05 meter sample depths are 0 351 and 0 912 respectively 124 Additionally if you are performing a composite analysis then this section will also display the correlations between the calculated ECac and measured z1 signal conductivity readings If you acquire 2 conductivity readings at each survey site then the z1 signal data will represent an average of these 2 readings In the example shown below the z1 signal data is highly correlated with the bulk average In ECac data r 0 945 as well as the In ECac data from the 0 45 and 0 75 sampling depths r 0 944 and 0 915 respectively Te ln Calc ECa Correlation Structure and ln Calc ECa v s zl signal Correlation Structure depth specific 1n Calc ECa sample depth levels 0 15 0 45 079 1 05 ave ave in Calc ECa 0 351 0 891 0 956 0 912 1 000 zl signal data 02351 0 944 0 915 0 788 0 945 II Bulk average ECac v s bulk average Primary Secondary soil variable Correlations This section displays the pair wise correlations between the bulk average ECac and bulk average soil profile data columns Note that all pair wise correlations are displayed using a standard correlation matrix format In the example shown below the In ECac In ECe correlation estimate is 0 864 the In ECac SP correlation e
25. can be changed as necessary To edit calibration data first select the proper data record and then type in the new values Note that you can scroll through your data records using either the Increment Decrement command buttons or you can jump to a specific record using the Jump command button 97 Once you have finished editing your data or column labels click first on the appropriate OK command buttons and then on the Finished button to return to the Summarize Edit Save Profile Data You should then re validate your calibration sample data file by clicking on the Compute Summary Statistics button Note As stated above you can not edit either the sample site ID or sample depth data values This restriction has been built into the program on purpose to prevent you from accidently corrupting any site ID or sample depth records If one or more of these values are in error then you must exit the ESAP Calibrate program correct these values in your raw profile data file and then re import this data 5 4 3 Practice Module Validating Laboratory Profile Data This practice module demonstrates how you should validate new calibration sample data If you have not already performed practice module 5 3 3 you should do so now you need to import the sk13_97 lab data as a DPPC data file before you can begin this module All new calibration sample data files must be validated using the Summarize Edit Save SES Profile Data window To inv
26. chapter 83 5 1 2 Navigating the ESAP Calibrate Main Menu Bar Main Menu Bar Layout The ESAP Calibrate main menu bar is located in the upper left corner of the main program window The full layout for this menu bar system is shown below Main level Sub Level 2 Sub Level 3 File gt Import Data File l gt Import a Survey data file gt Import a Profile data file gt Edit or Validate Profile Data gt View Print Project Output Files gt Exit Calibrate gt Stochastic Methods gt Profile Shape Magnitude Analysis gt Standard Correlation Analysis gt DPPC Correlation Analysis gt Spatial MLR Analysis gt Deterministic Methods gt Conductivity to Salinity Help gt About ESAP Calibrate gt ESAP Interface Controls l l gt Reset default Text Editor l gt OnLine Help gt What is ESAP Calibrate gt Navigating the Main Menu gt Frequently Asked Questions FAQ Main Menu Bar Menu Item Descriptions The profile menu contains 3 menu bar items File Calibrate and Help You will use the File menu to access all the data input output and validation routines the Calibrate menu option to access all of the data analysis routines and the Help menu to access the main ESAP program 84 help files Additional help file documentation is available for nearly all ESAP program routines this documentation can be accessed by double clicking on the OnLine Help file button within whatever wind
27. conductivity data and salinity boron and or nitrate sample data In turn you would use this information to appraise the feasibility of performing a conductivity survey Example 2 Suppose that your sample profile data contains 1 salinity 2 texture 3 water content and 4 bulk density and that this profile data has been acquired from an ESAP generated sampling plan Additionally assume that EM 38 horizontal and vertical conductivity survey data has also been acquired at all your profile sites Hence your main project goal in this scenario is to determine the correlation s between the measured conductivity ECa which in this case is your EM 38 signal data calculated conductivity ECac and soil salinity ECe and to determine the degree of ECa ECe signal deterioration due to spatial variation in the soil texture water content and bulk density data In this example you should import your sample data as a DPPC profile data set import your EM 38 survey data as a pre saved svy file and then perform a composite DPPC profile correlation analysis a composite analysis represents an DPPC correlation analysis performed in conjunction with conductivity survey data The DPPC procedure will again automatically estimate a column of calculated conductivity readings i e ECac readings and then add these calculated readings back into your profile data file Next the procedure will calculate all of the cross correlation estimates
28. currently residing in any of your active project directories Exit Select this menu option to exit the ESAP Calibrate program from within Calibrate Stochastic Methods Profile Shape Magnitude Analysis Select this menu option to invoke and display the 1D Profile Display window which can in turn be used to create 1D profile plots A 1D profile plot is a graphical representation of your profile data where the magnitude of a specific soil sample variable is plotted against the 85 sampling depth and data points from specific soil cores are joined connected together by lines Stochastic Methods Standard Correlation Analysis Select this menu option to invoke and display the Standard Correlation Analysis window which can in turn be used to create bivariate correlation plots and statistics This analysis allows you to visualize and quantify the correlation structure between any two input columns of profile data Stochastic Methods DPPC Correlation Analysis Select this menu option to invoke and display the DPPC Profile Data Correlation Analysis window DPPC PDCA window which can in turn be used to perform a DPPC correlation analysis A DPPC correlation analysis can be performed on any valid DPPC profile data file either before or after merging such a profile data file with the corresponding survey data This analysis allows you to calculate i e estimate a set of soil conductivity readings based on your input sample salinity te
29. documentation This menu item only becomes enabled after you have successfully read in an conductivity survey data input file 20 from within Analysis Basic Statistics Select this option to open up the Signal Transformation Window which you can then use to apply or remove a natural log transformation to or from your input conductivity survey data You can also change the input survey data column labels from within this window and or change the conductivity measurement units between dS m and mS m respectively Signal Decorrelation Select this option to open up the Signal Decorrelation Window This window contains the routines used for centering scaling and decorrelating your input conductivity survey data These routines help you detect any survey data outliers and must be run before you can generate any SRS sample designs Signal Validation Select this option to open up the Signal Validation Window This window contains the routines used to validate your decorrelated conductivity survey data all decorrelated survey data must be validated before you can generate any SRS sample designs During this process you can also choose to perform a transition analysis on your decorrelated survey data if desired assuming that you are working with a Transect type conductivity survey file Note none of the menu items under the Analysis item become enabled until after you have successfully read in an conductivity survey data input fil
30. features contained within this window to select the sample design size adjust some of the design characteristics and or save any sample design files to your current active project Additionally if you wish you can generate up to 5 separate SRS sampling designs After you invoke the SRS algorithm the ESAP RSSD program will update you as the algorithm converges During this process you will see a number called the Opt Criteria or just Opt Cri printed to the screen this is short for optimization criteria This number represents a measurement of how uniform i e evenly spread across the field the sampling plan is Uniform sampling plans generally achieve a value of 1 15 or less highly non uniform sampling plans typically have values of 1 75 or more By design the ESAP RSSD program will try to achieve as uniform a sampling plan as possible However good uniformity is not always achievable all SRS sampling designs are constrained by the initial response surface design levels which take precedence over the uniformity criteria In general smaller survey sizes lend themselves to better optimization and smaller sample sizes i e n 12 as opposed to n 20 are easier to optimize In most situations you can employ the optimization criteria value to judge the degree of uniformity in a generated SRS design using the guidelines listed below for square or rectangular survey areas Opt Cri Value Uniformity lt 1 15 excellent uniformity 1
31. gt Create Plot Output gt Print gt Raw Data Plot gt Smoothed Data Plot gt Both Plots gt Raw Data Plot gt Smoothed Data Plot Statistics gt Basic Profile Stats by depth 107 Help gt What is a 1D Profile Plot gt Navigating the 1D Profile Menu gt General Tips how to highlight sites etc Exit gt Return to Main Menu Profile Menu Bar Menu Item Descriptions The profile menu contains 5 menu bar items Plot Output Statistics Help and Exit You can use the Plot menu to initialize and create you 1D profile Plot the Output menu to print and or save your Plot the Statistics menu to generate basic summary statistics of all your input calibration sample data by depth and the Help menu if you wish to access any 1D profile Plot help files Brief descriptions of the sub level menu items located beneath the 5 main program menu bar items are given below from within Plot Specify new Plot Variable Select this option to specify the specific calibration sample data you wish to Plot When you select this menu option the 1D Profile Plot Initialization window will be displayed this window can then be used to select the plotting data and sampling depth units Create Plot Select this menu option to create your 1D profile data Plot as specified above from within Output Print Select the sub options listed under this menu item to print your 1D profile Plot Note that you can choos
32. has on the interpolation process etc EMy Conductivity Survey Data EMv dS m lt 1 006 1 006 1 159 1159 1 312 gt 1 312 Data Bounds x min amp max 16 288 Y min amp max 6 25 793 75 Figure 4 3 EMv raster map created in practice module 4 4 5 4 5 Migrating Data out of the ESAP SaltMapper Program There may be times when you want to analyze your processed survey or prediction data using some other software package For this reason the ESAP SaltMapper program contains an output ASCII text file feature which allows you to export your input data as a generic ASCII text 79 file This section describes how to use this export feature 4 5 1 Creating and Saving ASCII Text Files You can use the ASCII File Output window to generate a comma delimitated output ASCII text file of an ESAP SaltMapper data file To invoke this window click on the File gt Create Output Data File main menu option Before saving your ASCII text file you should choose the appropriate decimal degree accuracy for the output data columns The default is 2 decimal degrees i e the number 1 will be printed as 1 00 however you can set the accuracy from 1 to 4 decimal degrees respectively You should also specific data columns you wish to place into the output data file You need to select i e check at least one data column in order to create and save an ASCII output text file By default the site ID number x coordinate and y
33. left corner of the SCA program window This window can be invoked and displayed by clicking on the Calibrate gt Stochastic Methods gt Standard Correlation Analysis main menu option The full layout for this menu bar system is shown below Main level Sub Level 2 Data gt Specify Plot Data Plot gt By Depth gt All depths simultaneously gt Bulk average only 114 Statistics gt Calculate Correlations Output gt Print gt Save as Bitmap Help gt What is a standard correlation analysis gt Navigating the Standard Correlation Analysis Menu gt General Tips how to highlight sites plot a different depth etc Exit gt Return to Main Menu SCA Menu Bar Menu Item Descriptions The profile menu contains 6 menu bar items Data Plot Statistics Output Help and Exit You can use the Data menu to specify the two columns of calibration sample data you wish to plot the Plot menu to create the specific type of correlation plot you wish to display the Statistics menu to generate correlation summary statistics the Output menu to print and or save your correlation plot and the Help menu if you wish to access any SCA plot help files Brief descriptions of the sub level menu items located beneath the 6 main program menu bar items are given below from within Data Specify Plot Data Select this option to specify the specific columns of calibration sample data you wish to plot When you select this
34. measurements of 1 salinity 2 texture 3 water content 4 boron and 5 nitrate and that all this profile data has been acquired from a previous grid sampling survey Also suppose that no conductivity survey data has been acquired and that your primary project goal is to determine the feasibility of measuring and mapping salinity boron and nitrate levels from an as yet to be performed conductivity survey In this example you should import your calibration sample data as a DPPC profile data set and then perform a standard DPPC profile correlation analysis a standard analysis represents an analysis performed independently of any conductivity survey data The DPPC procedure will automatically estimate a column of calculated conductivity readings and then add these calculated readings back into your profile data file In practice bulk average calculated conductivity estimates and true conductivity readings tend to be highly correlated typical correlation values in actual survey applications usually exceed 0 9 Therefor the DPPC procedure will substitute the calculated conductivity data ECac for the missing true survey data and then compute the correlation between the ECac data and each column of soil sample data You can then examine these correlation values to obtain an approximate idea of how beneficial an actual conductivity survey would be in this field i e just how strong should the correlation 119 be between your acquired
35. menu option a Plotting Variables frame will appear within the SCA window You can then use the controls shown within this frame to select your x axis and y axis plotting variables and indicate if either data columns should be log transformed from within Plot By Depth Select this option if you with to display your correlation data by depth You may then use the increment and decrement buttons i e the lt lt and gt gt buttons to display each depth specific correlation plot 115 All depths simultaneously Select this option if you wish to create a single correlation plot which displays the calibration data associated with all of the sampling depths simultaneously Bulk average only Select this option if you wish to just create a correlation plot of the bulk average calibration data i e the average values across all sampling depths for each sample site from within Statistics Calculate Correlations Select this menu option to create and display the basic correlation statistics associated with the two currently displayed data columns These statistics will include the simultaneous bulk average and by depth correlation estimates corresponding to the three types of correlation plots mentioned above from within Output Print Select this option to print the currently displayed correlation data plot Save as Bitmap Select this option to save the currently displayed correlation data plot as a bitmap file from
36. need to use this selection procedure Actually there are three possible scenarios where this procedure is useful as described below The first is to re generate a design which was already created by the ESAP RSSD program during an earlier sampling period For example suppose that you ve already surveyed the field you are working in once before and used ESAP to generate the sampling design You now re survey this field by collecting new survey data at all of the previously established survey grid Sites and you d like to acquire your soil samples at the same set of earlier sample Sites Because the new survey data is different ESAP would probably not choose the exact same set of Sites during the second sampling iteration However you can force the program to re select the same set of Sites by creating a manual sampling design which exactly matches the first spatial response surface SRS design Along the same line of reasoning you may also need to re generate a sampling design which has already been performed just so that you can use the ESAP 95 software programs particularly ESAP Calibrate and ESAP SaltMapper to analyze your data As described above you can force the program to select a specific set of Sites by creating a manual sampling design which matches the already acquired sample Sites The second is for design comparison research purposes For example in your particular application it may be desirable to compare the efficiency
37. or prd files respectively ESAP SaltMapper preserves the structure of the survey or prediction file i e it can recognize the difference between a transect and grid file and process either file type accordingly Both 1 D transect graphs and 2 D raster maps can be created when transect files are imported into the ESAP SaltMapper program When grid files are imported only 2 D raster maps can be created In addition to its graphical capabilities the ESAP SaltMapper program can be used to create new data columns from the input data and or export any input data as a generic ASCII text file The data creation feature is especially useful when you wish to combine or average 2 or more existing columns of survey or prediction data Likewise the export feature can be used whenever you wish to employ another graphing or database program to read or process any ESAP output data file How does the ESAP SaltMapper Program Work To create any type of graphical output using the ESAP SaltMapper program you simply need to import the survey svy or prediction prd data file you wish to process Once you have successfully imported your data you can create new data columns produce graphics and or export your input data out to an ASCII text file Please note that the ESAP SaltMapper program has only been designed to process either ESAP RSSD output survey data or ESAP Calibrate output prediction data 1 e svy or prd data files At present these a
38. package through their work efforts Those deserving special mention include Dennis Corwin Lead Assessment Soil Scientist Robert LeMert and Nahid Vishteh Assessment Research support personal and Jessica Lin software support Partial funding for the development of this software package from the United States Bureau of Reclamation is gratefully acknowledged Program Flowchart input data conductivity survey data L input data ESAP RSSD calibration sample data L L output data ESAP Calibrate svy data file L TL 4 L input output data pro data file output data prd data file L L L ESAP SaltMapper L L output output graphics ASCII data output Figure 1 1 ESAP 95 software bundle program flowchart 2 0 ESAP 95 Software Overview 2 1 Software Program Design As explained in chapter 1 the ESAP 95 software package has been designed to help you perform field scale salinity estimation and prediction from soil conductivity survey data However to make effective use of this software package you first need to have a basic understanding of how a salinity survey is normally performed Once you acquire a feel for the steps involved in a typical salinity survey you will find the ESAP 95 software programs much easier to use In general a soil conductivity based field salinity survey is carried out in four steps In the first step a detailed grid of apparent soil conductivity is acquired across the
39. parameters when such parameters are found to improve the prediction accuracy 127 5 8 5 DPPC Data Plots After performing a DPPC profile data correlation analysis you will have the option to view the results using one or more types of panel graphs In the ESAP Calibrate program a panel graph is a series of 4 simultaneously displayed plots where each plot displays the same y axis variable against 4 different x axis variables In the DPPC panel graphs the y axis variable is always the calculated conductivity ECac Likewise one or more soil profile variables will always be shown along the x axis however the choice of which soil profile variable s to display is left to the user The purpose of the panel graph is to show how the ECac calculations correlate with and or are influenced by the displayed x axis soil variables All DPPC panel graphs can be created using the controls associated with the DPPC Graph Options window This Graph Options window can be invoked by clicking on the Plot Data command button located within the main DPPC Profile Data Correlation Analysis window note this command button will only become visible after the DPPC correlation analysis has been performed You should use the DPPC Graph Options window to specify the type of DPPC graph you wish to create and display You can select from three different types of panel graphs a standard panel or one of two custom panel graph types If you select a standard panel graph
40. prediction data This message is reminding you that any In or log prefixes should be removed from the imported signal column labels To do this close down the displayed ESAP SaltMapper message by clicking on the OK button Next click on the File gt Column Manipulation gt Change Column Labels main menu option this will invoke and display the Rename Columns window Now change the InEMv and InEMh labels to EMv and EMh and then click on the change 65 button Note that the new column labels are immediately re displayed in the Column Labels frame on the main SaltMapper window You have now successfully imported your BVWDdata svy data file and changed the EM column labels to reflect the fact that these columns were back transformed Now it is time to create a new data column For practice you should create a new column of EM ratio data defined as EMratio EMh EMh EMv To create this new column click on the File gt Column Manipulation gt Create a New Column main menu option After the Create New Data Column window is displayed highlight the ratio estimates option and enter the following numbers into the weight text boxes Column Labels Weights Weights EMv 0 1 EMh 1 1 Ist PCS 0 0 2nd PCS 0 0 Now type in EM ratio in the New Column Label text box and click on the Apply command button You should see a new column called EM ratio appear in the Column Label frame on the main program window If you like yo
41. program You will also learn how to change some the imported column labels and how to create and label a new data column If necessary start up the main ESAP 95 splash screen by clicking on the ESAP 95 prompt listed within the Programs drop down menu system i e click on Start gt Programs gt ESAP 95 Once the splash screen has displayed click once on the ESAP SaltMapper command button to invoke the SaltMapper program After the ESAP SaltMapper main menu screen is fully displayed click on the File gt Specify Project Input File Info menu option This will cause the Project amp Input File window to be displayed Make the Demo project your current project by clicking once on the Demo Project directory listed in the Set Current Project frame Next highlight the ESAP RSSD input file option you will be importing a svy data file and then check the back transform all log transformed data check box this tells the ESAP SaltMapper program to back transform the log transformed EM signal data columns Finally click on the Browse command button open the BVWDadata svy data file and click once on the OK command button You should now see 1015 lines of processed conductivity data imported into the ESAP SaltMapper program Once the data has been imported the ESAP SaltMapper program should print the following message to the screen Note don t forget to change the names i e labels of any columns associated with log transformed signal or
42. samples in the future at some or all of your calibration sites within your survey area you can use these new samples to statistically determine if the average magnitude of the predicted soil variable has changed over time Such tests are commonly referred to as net flux tests and these tests can be performed using the options available within the Net Flux Calculation window from within Help What is a stochastic calibration model Select this option if you would like to display the help file which briefly describes what a stochastic calibration model is and how such a model is commonly used for prediction purposes 136 What information is provided by the Advanced modeling options Select this option if you would like to display the help file which briefly describes some of the summary and graphical display features associated with the ESAP Calibrate advanced modeling options Navigating the SCM Menu Select this option to display the SCM Menu help file This help file explains how to use the SCM menu commands to invoke and or preform all of the various modeling and prediction procedures associated with the stochastic calibration modeling component of the ESAP Calibrate program Frequently Asked SCM Questions This is the help documentation you should read if you need more detailed information about how to use the various SCM procedures For example you should refer to this file for specific information about specifying and estim
43. survey site number 188 the sampling depths come from the 0 0 0 3 0 3 0 6 and 0 6 0 9 meter depths and the 3rd and 4th data columns represent your 2 soil measurements collected at each sample site Note once again that 1 the input data is sorted first by site and then by depth for each site and 2 arbitrary data formatting is acceptable 91 All DPPC or Generic type input data files must contain sample site ID numbers in column 1 and sample depth values in column 2 Furthermore all DPPC type files must contain soil salinity measurements in column 3 ECe readings in either dS m or mS m units texture measurements in column 4 either SP or Clay and water content measurements in column 5 either volumetric or gravimetric If it has been obtained then bulk density data should be placed into column 6 Any additional secondary soil measurements such as SAR boron nitrate cation or anion levels etc can be placed into columns 7 through 10 If no bulk density data is present then the first additional secondary soil measurement should appear in column 6 rather than column 7 In Generic type profile data files your soil measurements begin in column 3 and you can include up to a total of 8 soil measurement columns in your file placed into columns 3 through 10 You must include at least one soil measurement column in column 3 but columns 4 through 10 are optional Finally if your input soil data includes measurements of salinity tex
44. the plot you should click on the Plot gt Create Plot line plot menu option 4 3 4 Adding a Title Axis Labels and or Legend to a Transect Plot After you have displayed your line plot you can refine the visual appearance of this plot by adding titles and labels and or a figure legend Beneath the displayed plot there will be a frame called Title and Labels Inside this frame is an input text box a Figure Title button a Plot Title button a X Label button and a Y Label button You can type a unique text string in the text box and then use any of these buttons to assign this text string into the appropriate title or label slot on the graph Note adding a plot tile x label or y label will cause the displayed line plots to shrink slightly in size You can also display a legend in the upper right hand corner of the plot by clicking once on the Show button inside the Legend frame After you click on the Show button the button caption will change to Hide simply click on this button again if you wish to remove i e hide the displayed legend Note adding a legend to your graph will cause the displayed line plots to shrink in size 69 4 3 5 Practice Module Creating a 1D Transect Plot In this practice session you will learn how to create display and modify a 1D transect plot To invoke and display the 1D Line Plot Display window click on the Graphics gt 1D Line Transect Plot main menu option After this window ap
45. to 24555 Back Transformed Field Median Point Estimates ECe depth median 95 Confidence Interval 0 15 11 815 TOI to 13480 0 45 10 877 9 14 to 12 95 0275 11 456 9 31 CO 13 238 T05 11 013 9 52 to 12 74 average 11 509 10 29 to 12 88 II Field Range Interval Estimates depth range 1 range 2 range 3 range 4 0 15 22 94 9 41 8 59 7 66 0 45 214 09 9 64 8 52 7235 0 75 17 87 10 69 10 68 10 08 1 05 12 99 12 12 13 88 13 84 average 18 50 10 76 10 42 9 66 range 1 lt 6 800 range 2 6 800 to 8 363 range 3 8 363 to 9 925 range 4 9 925 to 11 488 range 5 gt 11 488 III Calculated Yield in computations Crop sorghum Yield 46 5 Yield Loss 53 5 with respect to Relative Yield under non conditions Note range 5 l 46 50 47 50 41 94 69 17 66 salin linear valued depth weighting vector mployed 155 Predicted bulk average ECe Pattern Training2 project ECe ave dS m lt 8 62 8 62 13 08 13 08 17 54 gt 17 54 Data Bounds X min amp max 414 96 737 56 Y min amp max 772 55 1115 44 Figure 5 8 Final predicted bulk average ECe pattern using Training2 salinity data raster map created by the ESAP SaltMapper program 5 9 14 Advanced Options Net Flux Testing The Net Flux window can be used to import and test for changes in the levels of calibration sample data collected across your survey area if you collect addit
46. window Now as before you should identify the name and location of your ASCII data input file and then import the data in the exact same manner as you just did during the generic file import process After the sk13_97 lab data has been re imported click on the ESAP Calibrate main window Column Info command button When the detailed data file information is displayed verify that the defined file type is DPPC and that the current data file status is reported as NOT validated Also verify that the main window information reports that this file is not yet saved In the next section section 5 4 you will learn how to validate and save this new DPPC profile data file 95 5 4 Editing and Validating New Profile Data The following section describes how to use the ESAP Calibrate software program to validate and if necessary edit a newly imported calibration sample data file This section also explains how to create a profile pro data file from imported calibration sample data 5 4 1 Validating New Laboratory Profile Data After you first import a new calibration sample data file of type DPPC or generic you will need to validate this data If you have imported a DPPC type data file this validation routine will perform a number of data checks on the input salinity texture and water content data columns and also the bulk density data column if present If you have imported a generic type data file then this validation routine wi
47. you can specify a map title a variable scale factor i e the measurement units of your mapping variable and the map appearance i e color or grayscale Select the variable to map by clicking on the mapping variable list box and highlighting the appropriate variable label Once you have selected the mapping variable the ESAP SaltMapper program automatically computes the raster cut off levels If you wish to you can 76 change these levels by typing in new values Note that the value of the 1 cut off level will control the boundary between the 1 and 2 raster zones on your displayed map Like the 2 and 3 levels control the boundaries between the 2 and 3 and 3 and 4 zones respectively After setting the cut off levels you should type in the appropriate scale factor into the Scale text box The scale factor should simply be the mapping variable units such as dS m for salinity or meq I for a laboratory cation measurement etc You should also type in a unique title for your map in the Map Title text box Note that any title you enter will be displayed above the raster map on all printed output Finally you can specify the appearance of the map by highlighting either the color or gray scale option button If you are printing to a black and white printer we recommend that you highlight the gray scale option 4 4 5 Practice Module Creating Raster Maps In this practice session you will learn how to interpolate create
48. 15 1 30 reasonable uniformity 1 30 1 50 moderate clustering 1 50 1 75 serious clustering gt 1 75 excessive clustering an unacceptable design 49 When possible you should always try to select a design with an optimization criteria of 1 30 or less When your survey area is non rectangular the above guidelines can not be directly applied In general highly non rectangular survey areas will produce significantly higher optimization criteria values and it will be necessary for you to adjust the above levels As a rule of thumb you can derive an approximate adjustment factor as follows First produce a printout of your survey zone and then draw the smallest possible rectangle you can around this zone Next calculate the approximate amount of area covered by your survey zone within that rectangle and then divide 100 by this value to produce the adjustment factor Finally multiply the above Opt Cri values by this factor to produce the new ranges For example if your survey zone was a circle then this circle would cover 78 54 of the smallest rectangle actually a square which could be drawn to completely enclose it Thus since 100 78 54 1 273 we would calculate that a design having excellent uniformity within this circular survey area should produce a value at or below about 1 46 1 e 1 273x1 15 1 464 Likewise a design with excessive clustering would produce a value above 2 28 etc Adjusting the Advanced Design Option
49. 1990 Determining soil salinity from soil electrical conductivity using different models and estimates Soil Sci Soc Am J 54 46 54 Rhoades J D D L Corwin and S M Lesch 1991 Effect of soil EC depth profile pattern on electromagnetic induction measurements Research Report 125 108 p Rhoades J D F Chanduvi and S M Lesch 1999 Soil salinity assessment Methods and interpretation of electrical conductivity measurements FAO Irrigation and Drainage Paper 57 Food and Agriculture Organization of the United Nations Rome Italy Slavich P G 1990 Determining EC depth profiles from electromagnetic induction measurements Aust J Soil Res 28 443 452 Weisberg S 1985 Applied linear regression Second Ed John Wiley amp Sons NY Williams B G and G C Baker 1982 An electromagnetic induction technique for reconnaissance surveys of soil salinity hazards Aust J Soil Res 20 107 118 Yates S R R Zhang P J Shouse and M Th van Genuchten 1993 Use of geostatistics in the description of salt affected lands pp 283 304 in Water Flow and Solute Transport in Soils Developments and Applications Advanced Series in Agriculture Series no 20 D Russo and G Dagan Eds Springer Verlag NY 161
50. 5 2 Importing Previously Saved ESAP Data Files This section describes how to import previously saved ESAP RSSD survey svy and ESAP Calibrate profile pro data files 5 2 1 Importing ESAP RSSD Survey svy Data Files You should use the Survey Data Import Window to import your svy data file into the ESAP Calibrate program This window can be invoked and displayed using the File gt Import Data File gt Import a Survey Data File main menu option In order to successfully import your data file you must perform the following two steps First you must set the current project i e you must identify which project your svy data file 87 resides in This can be done by clicking on the appropriate project name listed in the Set Current Project list box Second you must specify the path and filename of your input file This can be done by typing in the appropriate information into the Path Filename input text box or by using the Browse button to locate your input file Once you have performed these steps click on the OK button to import your svy data file If your input data file does not import properly then the error message listed below will appear and the data import process will be aborted Error Invalid path filename or file corruption error detected This error will be generated if you attempt to import any file type other than an ESAP generated svy data file The error will also occur if your input data file has bec
51. 7 and 1989 These methodologies represent efficient and practical field scale salinity estimation and prediction techniques and the ESAP 95 Software Package has been designed to help you appropriately use these techniques on your own conductivity survey data 1 2 The ESAP 95 Software Package Description Version 2 01R The ESAP 95 Software Package you are about to use currently contains three programs ESAP RSSD ESAP Calibrate and ESAP SaltMapper The ESAP RSSD program is designed to generate optimal soil sampling designs from conductivity survey information The ESAP Calibrate program is design to estimate both stochastic regression model and deterministic soil theory based calibration equations i e the equations which you will ultimately use to predict the spatial values of one or more soil variables from your conductivity survey data The final program ESAP SaltMapper can be used to produce high quality 1 D or 2 D graphical output of your conductivity survey data and or predicted soil variables Most importantly all three programs have been designed to work together in a seamless and efficient manner and each program employs a simple easy to learn graphical user interface Each program in this ESAP Software package contains a number of features designed to help you perform the various components of your soil salinity assessment process as described below ESAP RSSD Response Surface Sampling Design software Used to generate o
52. Calibrate gt Stochastic Methods gt Standard Correlation Analysis main menu option This will invoke and display the Standard Correlation Analysis window Once this window displays click on the Data gt Specify Plot Data sub menu option A variable selection frame will then appear you can use the controls within this frame to select your x axis and y axis data For this practice session select boron for the x axis ECe for the y axis and then click on the OK command button Now click on the Plot gt All Depths Simultaneously sub menu option a plot like the one shown in figure 5 4 should then appear This plot displays the overall correlation structure between your boron and ECe sample data If you wish to examine this relationship on a depth by depth basic click on the Plot gt By Depth sub menu option If you wish to study just the bulk average relationship click on the Plot gt Bulk Average Only sub menu option To calculate and display the correlation estimates for all of the above mentioned plots click on the Statistics gt Calculate Correlations option To gain more experience you should try creating these various plots and Statistics now In general a standard correlation analysis can be performed on any two columns of data present in your profile data file However this analysis is most commonly used to examine the correlation structure between soil salinity ECe and other primary profile data columns such as texture
53. D stands for response surface design When two input conductivity survey columns are present in your signal data file this option will produce a plot of your Ist z1 versus 2nd z2 principal component scores 28 When only one survey data column is available this plot will display the values i e magnitude of the 1st principal component scores along the x axis the y axis values will simply contain small random perturbations so that the 1 D principal component scores can be visualized in a 2 D plot All masked sites and any sites marked for deletion will be clearly displayed on either type of rsd plot In addition to the actual plots some pertinent conductivity survey statistics will be displayed in the lower left hand side of the Graphics Window These statistics will include the mean u and standard deviation s of whatever conductivity survey data is currently plotted along with the calculated correlation r between the two variables currently displayed in the scatter plot Note that by definition principal component scores have 0 mean unit variance and are uncorrelated Line Plots A line plot can be used to display the magnitude of the conductivity survey readings down an individual line transect These plots are most useful for detecting relationships between the bulk electrical conductivity and the transect coordinate position For example these types of plots can be used to check for the presence of tile lines i e t
54. ES AP 95 Version 2 01R User M anual and Tutorial Guide USDA ARS George E Brown Jr Salinity L aboratory R esearch Report No 146 J une 2000 ESAP 95 Version 2 01R User Manual and Tutorial Guide Scott M Lesch James D Rhoades and Dennis L Corwin Research Report No 146 June 2000 United States Department of Agriculture Agricultural Research Service George E Brown Jr Salinity Laboratory Riverside California DISCLAIMER The information in and or ESAP 95 software package associated with this document has been funded and developed by the United States Department of Agriculture Agriculture Research Service at the George E Brown Jr Salinity Laboratory Partial funding for the software package from the United States Bureau of Reclamation is also gratefully acknowledged Both this user manual and the ESAP 95 software associated with this manual are to be considered public domain software and as such may be used and copied free of charge Although the authors of this software have endeavored to produce accurate and error free program code this software including instructions for its use is provided as is without warranty expressed or implied Furthermore neither the authors nor the United States Department of Agriculture warrant guarantee or make any representations regarding the use or the results of the use of or instructions for use of this software or manual in terms of applicability reliability a
55. For Raw and Smoothed Data Plots option if you want ESAP to automatically determine the best scale for your smoothed profile data plot In general if you select this latter option then the data range shown along the x axis of the smoothed profile plot will not match the x axis data range of the raw profile plot 5 6 4 Practice Module Creating a Displaying 1D Profile Plots In this practice session you will learn how to create and display 1D profile plots using the demonstration profile data files in the Training and Training project directories i e the bwd_102lab pro and sk13_lab97 pro profile data files Therefore you should import the 110 bwd_102lab pro file now before continuing on with this session refer to practice module 5 2 3 if you ve forgotten how to do this After the Training profile data file has been imported click on the Calibrate gt Stochastic Methods gt Profile Shape Magnitude Analysis main menu option This will invoke and display the 1D Profile Plot window this may take a few seconds on some computers Once this window displays click on the Plot gt Specify New Plot Variable sub menu option to display the 1D Plot Initialization window After this window displays select ECe from the drop down variable list highlight the Yes option associated with the Apply In fitting algorithm prompt and then click on the Initialize Plot command button Now click on the Plot gt Create Plot sub menu option to displ
56. Help and Exit You can use Output menu to print and or save your panel plot s and the Help menu if you wish to access any DPPC Graphic help files Brief descriptions of the sub level menu items located beneath the 3 main program menu bar items are given below from within Output Print Select this option to print the currently displayed panel plots Note all displayed panel plots will be automatically printed on the same page of printer paper Save as Bitmap Select this option to save any one of the currently displayed panel plots as a bitmap file You should specify which plot you wish to save using the following layout guidelines Ist plot upper left hand plot 2nd plot upper right hand plot 3rd plot lower left hand plot 4th plot lower right hand plot from within Help Navigating the DPPC Graphics Menu Select this option to display the DPPC Graphics Menu help file This help file explains how to use the DPPC Graphics menu commands to print and or save panel plots General Tips how to highlight sites plot a different depth etc Select this option to display the General Tips help file This help file explains how you can use the various interactive plotting features associated with most ESAP Calibrate graphical displays such as highlighting sites plotting different sample depths etc from within Exit Return to DPPC Graph Options Menu Select this option to close down the DPPC Graphics Window and return
57. This will invoke and display the DPPC Graph Options window the controls within this window can then be used to select your graph type The panel graphs shown in figure 5 6 below show a standard panel for the 0 15 meter depth i e the relationships between the calculated In ECa and the measured In ECe SP volumetric H20 and bulk density for the 0 0 0 3 meter samples Note that you can view the data for each sample depth and or the bulk average data by simply clicking on the Increment Decrement command buttons 131 In Calc ECa vs In Salinity InfCalc ECa vs SP 0 6 0 4 0 2 00 0 2 No Point Selected No Point Selected In Calc ECa vs Yol H20 In Calc ECa vs Bulk Den 0 24 0 26 0 26 0 30 0 32 0 34 0 36 1 35 1 40 1 45 No Point Selected No Point Selected Figure 5 6 Standard panel graph for the 0 15 meter profile data from the bwd_102lab pro data file Note that the calculated ECa appears to be more correlated with the measured SP texture data than the measured salinity data 132 5 9 Spatial Regression Modeling Stochastic Calibration This section describes how to estimate validate and use a stochastic calibration model to predict the values of a response variable such a soil salinity from soil conductivity information There are many stochastic calibration model selection and validation techniques available for use in the ESAP Calibrate program Although most of these techniques are covered in detail in this
58. a programming perspective both the decorrelation and validation routines could actually be combined into a single process However these routines have been kept separate so that you can graphically analyze your decorrelated survey data before deleting any outliers Thus in most applications where outlier data Sites are detected you should visually display your decorrelated survey data using one or more of the graphical displays available in the Interactive Graphics window This in turn will allow you see both where the outliers are located within the survey area and how serious the outliers are with respect to the rest of the survey data 3 4 3 Additional Transect File Validation Options If you are processing a transect type signal data file then two additional post validation options will be available for use 1 edge buffering and 2 a transition analysis The edge buffering option which is located within the Signal Validation window can be used to mask out all your survey Sites located within approximately 4 of the edges of your field or survey area This buffering technique is useful when you wish to keep your final sample site locations off the edges of the survey zone Like edge buffering a transition analysis can be performed on a transect type survey data file A transition analysis quantifies the short scale signal variability i e this analysis identifies how different each conductivity survey reading is from its clo
59. al displays such as highlighting sites plotting different sample depths etc from within Exit Return to the SCM Menu Select this option to close down the 2D Prediction Scatter Plot Window and return to the SCM Stochastic Calibration Modeling menu 5 9 10 Calculating Field Summary Statistics The Calculate Field Summary Statistics window can be used to request the calculations of the mean survey area predictions and survey area interval estimates This window can be invoked and displayed by clicking on the Predict gt Calculate Field Summary Statistics SCM menu option These statistics are used to summerize the survey process and they must be calculated before an ESAP output prediction file i e a prd file can be created and saved To effectively perform the summary statistic calculations you will need to request the proper estimation options as described below Field Average Estimates After estimating your calibration model ESAP will compute the average level of these predictions across the entire survey area on a depth specific basis These averages are referred to as field average estimates For example if you have estimated a set of salinity calibration models for 4 sampling depths ESAP will calculate the field average estimate associated with each depth If your calibration model is based on log transformed response data then all calculated field average estimates will be automatically reported on a log scale Yo
60. alt tolerance curves command button This action will invoke and display the Salt Tolerance window which is explained in detail in section 5 9 11 However this Set Levels command button will only be visible if you request 4 cut off levels and that this feature should only be used if you are calculating soil salinity predictions Statistical Calculations After defining all of the calculation options click on the Calculate Statistics command button The ESAP Calibrate program will then calculate the mean survey area predictions and survey area interval estimates and display the View Output command button Click on this latter command button to view the ASCII output text file which contains your field summary Statistics 1 e the text file which contains all of the mean survey area predictions and survey area interval estimates If you wish to save the current ASCII output text file as a permanent text file highlight the Yes option under the Save Results to ASCII Text File frame and use the Specify Output File Name and Save File command buttons to name and save the file respectively You should click on the Specify Output File Name command button to display the File Open window which you can then use to specify a permanent output file name Likewise click on the Save File button to save your file After the file has been saved a message box will be displayed confirming the location of your permanent text file 5 9 11 Using the Salt Toleranc
61. ample variable is plotted against the sampling depth and data 82 points from specific soil cores are joined connected together by lines For sample soil salinity data a 1D profile plot can be used to display the shape of the salinity profile with respect to the overall magnitude of the average salinity level for each sample core thus allowing you to qualitatively infer the net direction of water flow under equilibrium conditions Likewise a standard correlation analysis is designed to produce bivariate correlation plots and correlation statistics This analysis allows you to visualize and quantify the correlation structure between any two input columns of profile data For example you could plot and examine the correlation structure between your sampled soil salinity and soil texture One of the most useful procedures contained within the ESAP Calibrate program is the DPPC correlation analysis A DPPC correlation analysis can be performed on any valid DPPC profile data file either before or after merging such a profile data file with the corresponding survey data This analysis allows you to calculate i e estimate a set of soil conductivity readings based on your input sample salinity texture and water content data values These calculated soil conductivity readings can then be compared to each column of input soil data thus allowing you to see how well the calculated conductivity correlates with each soil variable If you have survey d
62. analysis features In the ESAP Calibrate program these features are separated into two general categories stochastic methods and deterministic methods as shown below Stochastic Methods methods which require profile data 1 profile shape magnitude analysis 1D profile plots 2 standard correlation analysis bivariate correlation plots 3 DPPC correlation analysis conductivity influence analysis 4 stochastic calibration modeling spatial regression modeling Deterministic Methods methods which do NOT require profile data 5 deterministic conversion of conductivity into salinity Four of the five primary analysis methods are stochastic statistical methods which in practice means that they can only be used if profile calibration data have been collected In contrast you do not need to acquire any profile data to use the deterministic conductivity to salinity conversion routine Table 5 1 shown below describes the type of data each analysis method is designed to work on Table 5 1 Data requirements for each analysis method Survey Profile Analysis Method conductivity data calibration data required required In the ESAP Calibrate program each analysis method is designed for a specific purpose For example a profile shape magnitude analysis is designed to produce 1D profile plots A 1D profile plot is a graphical representation of your calibration sample profile data where the magnitude of a specific soil s
63. anatory The sub level menu items located beneath these 5 main options are defined as follows from within Graphics Survey Grid This option can be used to produce a map of your survey grid i e a plot of the survey coordinates Scatter Plots This option can be used to produce scatter plots of the 1st s1 or 2nd s2 conductivity survey readings against the x or y coordinates or against each other This option can also be used to produce a response surface design rsd plot Line Plots This option can be used to produce line transect plots provided you imported a Transect type survey file You can plot the Ist s1 and or 2nd s2 conductivity survey readings against either the x or y row coordinates Note that this line plotting option must be initialized before any line plots can be displayed 32 Histograms This option can be used to produce histogram plots of the Ist s1 or 2nd s2 conductivity survey readings or the primary z1 or secondary z2 decorrelated principal component scores This option can also produce a histogram of the z1 transition standard deviation estimates provided you imported a Transect type survey file and requested ESAP to perform a transition analysis Note that this histogram plotting option must be initialized before any histogram plots can be displayed ColorScale Grids This option can be used to produce color scale grids which are simply survey grid plots where the plotting symbol col
64. and 3 the conductivity to ECa conversion formula type and parameter estimates You can use the horizontal scroll bar located within the Temperature Factor frame to define the average soil temperature within the depth interval you wish to estimate The default temperature is 25 degrees C and the valid input range is from 10 to 35 degrees C Information entered in the Signal Factors frame can be used to adjust your raw conductivity readings if specific types of adjustment is necessary For example if you have collected EM 38 survey data using an automated system then you more than likely had the EM 38 unit s mounted at some small distance above the soil surface This height above ground effect can be theoretically corrected for provided you know how high the EM 38 unit s are 101 mounted above the soil surface If you collect EM 38 signal data you need to indicate the exact height above the soil surface that both the horizontal EMh and vertical EMv signal readings are acquired at Additionally you need to indicate if the s1 signal readings represent EMv or EMh readings respectively Note that the height measurement is defined as the distance in meters from the soil surface to the base of the EM 38 unit in either orientation If your survey data consists of anything other than EM 38 signal data you have the option of specifying a multiplicative correction factor for each signal column ESAP can then automatically multiply the si
65. apper can be used to produce high quality 1D or 2D graphical output of conductivity survey data and or predicted soil variables This manual describes and documents to implementation and use of each of these three programs in detail 1 0 2 0 3 0 Table of Contents Introduction 1 1 General conductivity salinity assessment techniques 1 2 The ESAP 95 software package description 1 3 Software installation directions 1 4 ESAP 95 software training files 1 5 ESAP 95 software package development information ESAP 95 Software Overview 2 1 22 2 3 2 4 2 5 Software package design Data input output An overview of the 3 main ESAP program menus Supported surveying techniques and applications On line help documentation ESAP RSSD Software Program 3 1 3 2 3 3 3 4 ESAP RSSD program overview 3 1 1 Program description 3 1 2 Navigating the main menu Data file input specifications 3 2 1 Creating a project and field ID 3 2 2 Reading in survey data 3 2 3 Practice module Data visualization 3 3 1 Description of graphical techniques 3 3 2 Navigating the graphics menu 3 3 3 Invoking the GI and CT windows 3 3 4 Practice module Data analysis 3 4 1 Basic statistics 3 4 2 Using the decorrelation validation algorithms 3 4 3 Additional transect file validation options 3 4 4 Practice module NYA ARNK 11 12 15 15 17 17 17 19 22 23 23 26 28 28 31 34 35 42 42 43 44 45
66. aster map menu bar is located in the upper left corner of the Raster Map Initialization window The full layout for this menu bar system is shown below Main level Sub Level 2 Sub Level 3 Plot gt Save Clipping Boundaries l gt Specify clip boundary file name gt Save clip boundary file gt Specify Map Variable gt Create Map gt Enlarge Map for Output gt Send to Raster Map Window 1 gt View Raster Map Window 1 gt Send to Raster Map Window 2 gt View Raster Map Window 2 gt Print both Maps Help gt About the Raster Map Creation Process gt Using the On Screen Interpolation Controls gt Navigating the Raster Map Menu 74 Exit gt Return to Main Menu Raster Map Menu Bar Menu Item Descriptions The raster map menu contains 3 menu bar items Plot Help and Exit You can use the Plot menu to initialize create and print all of your 2D raster maps and you can use the Help menu if you wish to access any raster map help files However note that none of the Plot menu items will be enabled until after you have completed the on screen interpolation process Brief descriptions of the sub level menu items located beneath the 3 main program menu bar items are given below from within Plot Save Clipping Boundaries You should only use this menu option if you wish to save the clipping boundaries associated with your currently displayed survey data Select the submenu item Specify cli
67. ata appear to be approximately bell shaped then you probably don t need to apply a log transformation On the other hand when the distributions are strongly right skewed then a log transformation should definitely be applied If the histograms do not clearly look either bell shaped Normal or right skewed Lognormal then you will need to base your decision on your own personal survey experience 42 ESAP does not currently support any other type of transformation or than the natural log However you are free to apply any type of transformation you wish to your survey data before it is imported into the ESAP RSSD program 3 4 2 Using the Iterative Decorrelation Validation Algorithms The ESAP RSSD program uses a statistical technique known as a response surface sampling design to select the final sample Sites In order to facilitate this design all conductivity survey data must first be centered scaled and decorrelated To completely understand how the signal decorrelation and validation routines facilitate the development of a response surface sampling design you should refer to the research article by S M Lesch D J Strauss and J D Rhoades 1995b This scaling and decorrelation process is carried out by the signal decorrelation routine A by product of this decorrelation analysis is that outlier i e highly unusual survey readings become very easy to detect In ESAP RSSD detecting and removing outliers is referred to as sig
68. ata associated with these profile sites then you can also look at the correlation between your calculated and measured i e true conductivity readings The final stochastic procedure contained within the ESAP Calibrate program is the stochastic calibration procedure This procedure allows you to generate spatially referenced regression models which in turn use the acquired conductivity survey readings to predict the values of one or more soil variables at each survey site within your field or survey area These models can also be used to estimate the average level of each soil variable within the field and calculate a number of additional useful prediction statistics Furthermore these models can be used to generate predictions of your soil variable across multiple sampling depths provided you acquire multiple depth calibration data If you are only interested in estimating the soil salinity pattern across your survey area and you can not afford to acquire any calibration sample data then you can use the deterministic conversion algorithm This algorithm converts conductivity data into salinity data using a two step process First raw instrument conductivity is converted into depth specific soil conductivity ECa for a given depth Next this ECa data is converted into estimated soil salinity data using a linear version of the DPPC model All of the analysis procedures described above are explained in detail within the remainder of this
69. ata processing session For example if you input the alpha numeric code AB12 then all output files will begin with this code i e AB12data txt or AB12info txt etc Keep in mind that you should always use a unique i e different code for each input survey file and that project names and file ID codes can be used together to form an efficient 2 level organization structure for your ESAP output files 3 2 2 Reading in Survey Data Grid versus Transect Survey Data The ESAP RSSD program is designed to read in ASCII text data files which are formatted as one of two input file types a Grid file or a Transect file Conceptually the only difference between these to file types is that Transect survey data files contain a column of row numbers where as Grid files don t contain any row numbers However in practice there are actually significant differences between these to file types at least in the way the conductivity survey data is collected The brief discussion given below should clarify these differences In theory if you collect your conductivity survey data on the go then you are collecting a Transect type input file Usually this sort of file will contain a large number of readings associated with each transect or row but only a limited number of transects For example you might collect 10 transects across a field with 100 conductivity readings acquired within each transect Hence in this survey you would have acquired 1000 sur
70. ata usually results in more accurate and meaningful correlation estimates Important all log transformations requested using the Log Transform Columns option discussed above are TEMPORARY transformations only As soon as the DPPC correlation estimates have been calculated all log transformed columns are back transformed into their raw data units INCLUDING the calculated column of ECac data After requesting any desired log transforms you can click on the Perform DPPC Analysis command button to invoke the DPPC profile data correlation algorithm When the algorithm finishes the View Correlation Output and Plot Data command buttons will appear You can click on the View Correlation Output button to display the DPPC profile data correlation analysis summary report If desired you may print this report using the print command option s associated with your default text editor Additionally you should click on the Plot Data command button if you wish to invoke and display the DPPC Graph Options window You may then use the options associated with this window to produce various types of DPPC panel graphs 123 File Save Options The command buttons contained within the Save Results to ASCII Text File frame may be used to save the information contained within the DPPC summary report to a permanent ASCII text file You should use the Specify Output Filename command button to specify the name of your permanent output ASCII text file and then the Save
71. ating a regression model calculating various summary statistics and model predictions or using the net flux testing procedure etc from within Exit Return to Main Menu Select this option to close down the SCM Window and return to the Main program window 5 9 3 Response Variable Specification Window You should use the Response Variable Specification window to select the profile variable that you wish to predict i e use as the response variable in you stochastic calibration model This window can be invoked and displayed by clicking on the Model gt Select Response Variable SCM menu option There are three steps you must perform when you display this window 1 set the active project directory 2 select the appropriate response variable and 3 merge the survey and profile data files You should verify that your two data files were merged correctly If desired you can also create and view some preliminary response variable data plots and or delete one or more profile data sites To estimate a stochastic calibration model you first need to merge your active survey and profile files together into a single calibration data file If these two files are stored in different project directories then you can choose which project directory to save your regression modeling output results to using the Survey Project or Profile Project option buttons listed in the Set Active Project Directory frame Note if your survey and profile projec
72. ation explains how to use the Graphics menu bar and describes the menu bar features OnLine Help Graph Descriptions Information This is the help documentation you should read if you need help interpreting and or understanding any of the graphs mentioned above This documentation also contains some useful tips on when and how to best use each of the graphic procedures when running the ESAP RSSD program from within Exit Return to the Main Program Menu Select this option to close down the graphics window and return to the main program menu 3 3 3 Invoking and Using the GI and CT Windows The Graphics Initialization GI window should be used to initialize all Line Transect plots Histogram plots and ColorScale grids You can invoke and display the GI window by clicking on the Options gt Initialize Graphic Components menu option By default the plots specified above are not initialized when the Interactive Graphics Window is first activated Hence their corresponding graphic initialization procedures must be specifically requested before you can view these plots Both Histogram plots and ColorScale grids can be created from either a gird or transect survey data file However Line plots can only be created from a transect data file Additionally the transect direction north south or east west must be specified when requesting Line plots The Graphics Initialization window can also be used to turn off one or both symbol hi
73. ation resulted in a negative salinity estimate the ESAP Calibrate program will automatically re set all negative estimates to 0 Additionally a basic set of summary statistics are automatically displayed after the conversion process finishes In this example the mean standard deviation minimum and maximum calculated salinity values are 2 225 0 908 0 000 and 5 820 dS m respectively To estimate the salinity values across the survey area in the 0 3 0 6 and 0 6 0 9 meter depths repeat the above exercise using the appropriate 0 3 0 6 and 0 6 0 9 meter parameter values Provided you use the conductivity and soil information listed previously the following salinity summary statistics will be displayed for each prediction depth Prediction Depth 0 0 0 3m 0 3 0 6m 0 6 0 9m Mean 2 225 4 742 5 638 Std Dev 0 908 1 834 1 807 Minimum 0 000 0 644 1 792 Maximum 5 820 11 806 11 908 105 Note also that you can save an output file of the calculated salinity data values after each conversion process These output conversion data files could then be imported into and processed by the ESAP SaltMapper program if desired 5 6 Performing a 1D Profile Analysis This is the first of four sections which describe the various stochastic modeling and analysis features within the ESAP Calibrate software program This section describes how to create display and print 1D profile plots and how to effectively use these plots to perform an exploratory p
74. ay the 1D raw and smoothed profile plots Your 1D plots should look identical to the two plots shown in figure 5 1 Raw Profile Data ECe Smoothed Fitted Data r 9464 dS m 23 4 5 Figure 5 1 1D raw and smoothed ECe profile data plots generated from the bwd_102lab pro sample salinity data This profile plot reveals quite a bit about the soil salinity conditions within this field For example all of the individual profile shapes are regular and all of the near surface salinity levels within the first two feet are quite low This implies that the field is well managed and sufficient water is being applied to keep the soil salinity levels under control Indeed perhaps too much water is passing through the profile at certain locations given some of the very low 111 salinity levels at the 1 0 meter depth Also the correlation between the raw and smoothed salinity data is 0 9464 suggesting that a very strong relationship exists between the shape of the profile and the average magnitude of each sample core To see a completely different sort of situation you should now examine the profile data stored in the Training2 project i e the sk13_lab97 pro data file Exit out of the 1D Profile Plot module and import this Training2 profile data file you will automatically erase the bwd_102lab pro data and replace it with the sk13_lab97 pro data when you import this latter data file Now start up the 1D Profile Plot window once again a
75. being said the column structure of the input calibration data represents the primary difference between the two profile data file types By design all DPPC type input data files have the following general column structure DPPC Column Structure Column 1 site_ID required 2 sample depth midpoint required 3 salinity required 4 texture required 5 water content required 6 bulk density optional 7 10 secondary soil measurements optional Additionally the input data file must be double sorted first by the site _ID numbers and then by the sampling depths For example suppose you collected calibration sample data from 3 depths at each of 10 sites across the survey area Then your input data file might look like the abbreviated data file shown below of Owe 2533 6 1 2 he y 0 45 5 67 53 2 20 4 7 0 75 9 41 47 7 19 9 18 0 15 1 67 66 6 24 4 18 0 45 4 44 58 2 23 9 18 Or S 63 99 Vode de hh 90 L 181 Ow S696 E23 233 181 0245 6532 699 27 5 18 0 75 8 48 49 9 18 8 In the above example the first sample site comes from survey site number 7 the last sample site comes from survey site number 181 the sampling depths come from the 0 0 0 3 0 3 0 6 and 0 6 0 9 meter depths and the 3rd 4th and 5th data columns represent soil salinity dS m saturation percent and gravimetric water content Note that the input data is sorted first by site and then by depth for each site Howev
76. between ECa ECac and ECe and also compute the percentage of signal deterioration between the ECac and ECe readings by depth due to spatial variation in the secondary soil variables For example if the correlation between the ECa and ECac data is high but the correlation between the ECa or ECac and ECe data is rather poor then you should check the signal deterioration estimates If these estimates are also high i e gt than 15 then the poor ECa ECe correlation is probably due to excessive spatial variation in the texture water content and or bulk density data etc The above two examples represent only a brief introduction to the types of information available to you when you perform a DPPC data correlation analysis You should refer to the rest of section 5 8 to gain a more thorough understanding of this analysis procedure as well as the types of inference which can be drawn from it 5 8 2 Description of the DPPC Model The DPPC Dual Pathway Parallel Conductance Model is an electrical conductance model developed by Rhoades Rhoades et al 1989 which described the relationship between bulk soil electrical conductivity ECa volumetric water content Tw and the electrical conductivity of the soil water ECw Mathematically this model can be written as Ts Tws ECws ECs ECa Twc ECwc 5 1 Ts ECws Tws ECs 120 where Tws volumetric water content in the soil water path
77. bi z1 zl Ist OT bo bi z1 b2 x b3 y z1 x y same as above zl 2nd OT bo by z1 b2 x b3 y b4 xy bs x boy zl x y Xy a y same as above zl z2 bo b z1 b2 z2 zl z1 bo bi z1 b2 z1 zl z2 z1 x y y bo b z1 b2 z2 b3 z1 ba x bs y bey Restrictions All calibration models must contain an intercept term as well as a z1 signal parameter No parameter combination can be specified if the number of parameters including the intercept is greater than the number of sample sites minus 2 i e of parameters lt of samples 2 Additionally no z2 signal term can be specified if only one signal reading was acquired at each sample site nor can any higher order terms be entered into a model without also including the corresponding lower order term for example x can not be included without first including x etc Advanced Options associated with the 3 Model Identification Methods Standard Model frame Up to 12 different parameter combinations can be manually selected using the option buttons displayed in the Standard Model frame The ESAP generated default standard parameter combinations are as follows 143 of signal of calibration default parameter readings per site sample sites combination 1 lt 6 zl 1 gt 6 zl x y zl 1st OT 2 lt zl z2 2 SI zl z2 x y z1 z2 Ist OT These default selections usually generate
78. both point and conditional probability estimates 4 they could be used to test for changes in the geometric mean field salinity level over time and 5 they were shown to be theoretically equivalent to cokriging models provided the regression model residuals are spatially independent There are both advantages and disadvantages to using either modeling approach For example stochastic models are usually more accurate than deterministic models when the secondary soil properties are not known across all the survey sites However because stochastic models are dynamic some soil samples must be acquired during each survey expedition Additionally these models also tend to be both time and location dependent On the other hand deterministic models are at least in theory both time and location independent and these types of estimation techniques do not require calibration salinity data But deterministic techniques do require accurate secondary soil property information at every survey site and hence are not commonly well suited for use with automated assessment equipment etc This manual describes and documents a series of site selection and salinity modeling software programs contained within the ESAP 95 Software Package release version 2 01R The modeling techniques incorporated into this salinity software are based on the stochastic and deterministic modeling methodologies described in Lesch et al 1995a 1995b and Rhoades et al 1999 199
79. ccuracy or correctness The use and application of this software and manual is the sole responsibility of the user The mention of any trade names or commercial products is for the convenience of the user and does not imply any particular endorsement by the United States Department of Agriculture or its agents iii Technical Abstract Lesch S M J D Rhoades and D L Corwin 2000 The ESAP 95 version 2 01R user manual and tutorial guide Research Report No 146 USDA ARS George E Brown Jr Salinity Laboratory Riverside California This manual describes and documents a series of site selection and salinity modeling software programs collectively known as the ESAP 95 software package Release version 2 01R developed for the analysis and prediction of soil salinity from conductivity survey information It is designed to be used both as a software reference text and tutorial guide The ESAP 95 software package currently contains three programs ESAP RSSD ESAP Calibrate and ESAP SaltMapper The ESAP RSSD program is designed to generate optimal soil sampling designs from bulk soil electrical conductivity survey information The ESAP Calibrate program is design to estimate both stochastic regression model and deterministic soil theory based calibration equations i e the equations which are ultimately used to predict the spatial values of one or more soil variables from conductivity survey data The final program ESAP SaltM
80. cify that this analysis should be performed on the log transformed sample data by highlighting the Yes option button associated with the Apply In fitting algorithm option In general you should apply the In fitting algorithm when analyzing data which by definition must be greater than 0 and appears to be lognormally distributed as is often the case with soil salinity measurements In addition to providing for more accurate shape magnitude predictions this algorithm will also ensure that the predicted soil variable levels across all sampling depths remain positive The displayed 1D raw and smooth profile plots will always be shown on the raw i e untransformed data scale regardless of whether or not you apply the In fitting algorithm Also because the In fitting algorithm uses log transformed sample data the calculated shape magnitude correlation statistics produced with and without the use of this algorithm will usually not be the same although in practice the calculated r values should typically not be very different Display Scale ESAP automatically determines the plotting scale for your raw i e observed 1D profile data plot However you can control the scale of the smoothed i e fitted profile data plot You should highlight the Same Scale For Raw and Smoothed Data Plots option if you want the smoothed profile plot x axis data scale to match the raw profile x axis data scale Otherwise highlight the Different Scales
81. coordinate data columns will be initially selected the row number column will also be selected if you are working with a transect type survey or prediction file These columns can be de selected if you wish Finally to create the output data file you must specify the output path and file name and then save the file You can use the Specify Output File Name and Save File command buttons to perform these actions 4 5 2 Practice Module Exporting the BV10data svy Data File In this final practice module you will learn how to export and save the BVWDdata svy data as an ASCII text file To invoke and display the ASCII File Output window click on the File gt Create Output Data File main menu option Once this window appears you can set your decimal degree accuracy and then select the desired data columns For this example set the degree accuracy to 3 and select i e check the x coordinate y coordinate EMv EMh and EM ratio data column check boxes To de select the site ID and row number columns simply click on their corresponding check boxes Now click on the Specify Output File Name command button When the File Open window appears change the default directory to esap2 or some other directory if desired type in the name bvywd_out without the quotes and click on the Open button Then click on the Save File command button located back on the ASCH File Output window The ESAP SaltMapper program should then export 1015 line
82. cquired during the sampling process c ability to produce 1D profile data graphs and perform bivariate profile data correlation analysis d ability to fully automate the stochastic calibration regression modeling process ESAP SaltMapper 1 D Transect and 2 D raster mapping software Used to generate 1 D transect and 2 D raster maps of raw conductivity estimated soil salinity and or estimated secondary soil physical properties i e designed accept input files from either ESAP RSSD or ESAP Calibrate Figure 1 1 shows the software program flowchart for the ESAP 95 Software Package As shown in figure 1 1 the ESAP RSSD program should be used first to process your conductivity survey data The ESAP Calibrate program can then be used to import and process your soil sample data and calibrate this soil data to the conductivity survey data Additionally the ESAP SaltMapper program has been designed to produce graphical output of either your processed conductivity data or predicted soil properties Chapter 2 of this user guide discusses the data file input requirements and output file specifications in detail 1 3 Software Installation Directions If you are installing the ESAP 95 Software Package from floppy disks then insert disk 1 into you re a drive and click on the Windows Start button In the dialog box type a setup exe click on the Run button and then follow the on screen directions If you are installing a down loaded copy from t
83. ctories which contain demonstration input data for use in the ESAP Calibrate program Important Notes for Re Installs If you have previously installed an earlier version of the ESAP 95 software package i e a Beta version then you should follow the steps shown below a BACK UP YOUR DATA FILES This can be done by inserting a blank disk into the A drive and copying the C US_Salinity_Lab esap2 data subdirectory onto this disk By doing this you will automatically create a back up copy all of your unique project directories located off the data directory on the floppy disk b Un install your current beta ESAP software version This should be done by selecting START gt Settings gt Control Panel and then clicking on the Add Remove Programs icon Next highlight the currently installed ESAP software click on the Remove software command button and then follow the on screen directions c Install the latest version of ESAP 95 After the installation process is finished you can then restore your project directories by simply copying these sub directories back into the data directory Important Notes for Windows NT Users The ESAP 95 Software package will operate properly on an NT platform although it was not specifically designed for such an operating system You can install it in the same manner as you would on a 95 98 system platform However the default text editor used by ESAP write exe needs to be reset each tim
84. ctors which would be used to recreate the mathematical formulas shown above In general the easiest way to define your specific summation weights is by writing out the mathematical formula for the new data column you wish to create and then reading off the weights directly from this equation After you have defined an appropriate set of summation weights you should enter a label for your new data column in the new column label text box This can be any descriptive label which is 14 characters or less in length 64 4 2 4 Displaying Basic Column Statistics You can display basic summary statistics associated with each imported or created data column by clicking on the File gt Column Manipulation gt Column Statistics menu options The displayed statistics will include the minimum maximum mean and standard deviation for each column of data Among other things this feature is useful for determining the range and variability of newly created data columns 4 2 5 Practice Module Importing and Manipulating Data All the practice modules contained within Chapter 4 including this module will make use of the BVWDdata svy processed conductivity survey data file that was created during the last Chapter 3 practice session Hence if you have not already done so you will need to complete the Chapter 3 practice modules in order to continue on with this session In this practice session you will learn how to import data into the ESAP SaltMapper
85. d is shown true to scale i e ESAP does not distort or warp the x y coordinates in any manner The survey grid option is always active i e enabled whenever the ESAP Graphics Window is active Therefor you can produce a survey grid plot at any time After you have performed a signal decorrelation on the input conductivity survey data see section 3 4 any masked sites will be displayed as yellow points and any sites marked for deletion will be displayed as bright red points Once you delete a site its location will no be displayed on the survey grid since this data will no longer be contained within the conductivity data file Additionally any site selected as a sample site will be displayed as a light blue point on a survey grid plot You can turn these color display options off from within the Graphics Initialization GI window Scatter Plots Scatter plots provide you with an effective way to visualize the correlation structure between to variables In ESAP you may plot your s1 1st column or s2 2nd column survey data against either the x or y coordinate axis or against each other i e s1 versus s2 These plots will display the overall relationship between the above mentioned variables The scatter plot options above are always active i e enabled whenever the ESAP Graphics Window is active An additional scatter plot known as an rsd plot also becomes active after you have performed at least one signal decorrelation RS
86. data using a two step process First raw instrument conductivity is converted into depth specific soil conductivity ECa for a given depth Next this ECa data is converted into estimated soil salinity data using the linear version of the DPPC model The following information is required to convert raw conductivity into ECa readings 1 the average soil temperature 2 the conductivity instrument signal correction factors if applicable and 3 the conductivity to ECa conversion formula type and parameter estimates Then to convert ECa data into soil salinity you must additionally specify the field average SP bulk density and water content relative to field capacity Because this algorithm uses a deterministic approach no soil salinity prediction errors are generated during the conversion process Additionally caution is urged when using this approach since serious errors in one or more of the input parameters can lead to significant ECe estimation errors In general if you have acquired soil samples we recommend that you use the regression modeling approach to statistically predict the soil salinity levels from the conductivity survey data i e stochastic calibration modeling If you are unfamiliar with the DPPC model you may wish to refer the discussion of the DPPC equation given in section 5 8 Also please note that this deterministic conversion algorithm can only process i e estimate salinity data within a single depth If
87. display some histogram statistics These statistics include the mean u standard deviation s minimum min and maximum max of the requested variable In addition to the s1 s2 zl and z2 data you may also produce a histogram of the calculated z1 transition standard deviations if a transition analysis has been perform on the decorrelated signal data see section 3 4 This histogram of the calculated transition standard deviations can be used to indicate the overall distribution of short range variability associated with your input conductivity survey data ColorScale Grids A color scale grid is simply an enhanced survey grid plot e g a survey grid displayed using plotting symbol colors which correspond to the magnitude of the variable being plotted If you are working with survey data collected on a dense evenly spaced grid then you can use a color scale plot to create a quick and dirty raster map of the spatial conductivity survey response pattern If your input x y survey coordinates are un evenly spaced across the survey area or the survey grid is small less than 500 sites a color scale grid tends to be less visually appealing However you can still generally ascertain the spatial survey response pattern even under these latter conditions The color scale plotting option is disabled when the Graphics Window first activates you need to enable i e activate it by opening up the GI window and initializing it During the i
88. dow Select this option to close down the PRV Graphics Window and return the Response Variable Specification Window 5 9 5 Specifying the Calibration Model Parameters In order to estimate a stochastic calibration model you must specify the model parameters from within the MLR Model Identification window This window can be invoked and displayed by clicking on the Model gt Identify Model Parameters SCM menu option The model parameters can be specified one of four ways you can let ESAP choose a default standard model you can manually select a standard model you can manually define and specify a custom 140 model or you can have ESAP auto select an appropriate model Your calibration model parameters will typically consist of two components signal parameters and trend surface parameters All signal parameters are incorporated using the principal component scores keep in mind that these principal component scores simply represent linear combinations of your possibly log transformed survey signal data columns In a similar manner all trend surface parameters are incorporated using centered and scaled x y coordinate data where the x y coordinates are adjusted to lie between the bounds of 0 lt x lt 1 and 0 lt y lt 1 respectively These scaling techniques help insure that the matrix inversion techniques used by the regression modeling algorithm remain stable In general the goal of the model identification process is to identif
89. e Additionally the Basic Statistics menu item will de activate i e become disabled immediately after you perform the first signal decorrelation and both the Signal Decorrelation and Signal Validation menu items will de activate immediately after you perform the last signal validation from within Design Calculate SRS Sample Design Select this option to open the SRS Sample Design Window This window contains all of the routines for generating optimal SRS sampling designs based on your input conductivity survey data Within this window you can specify the design sample size either 6 12 or 20 sample sites invoke and or modify a number of advanced design features and interactively generate and save up to 5 different sampling plans for your survey area Manual Sample Site Selection If necessary you can use this menu option to open the Manual Sample Site Selection Window This window should only be invoked if you need to create and save a user specified sampling design i e a sampling design created or generated by some other program other than ESAP 21 Note neither of the menu items under the Design item become enabled until after you have successfully validated your conductivity survey data input file see the Signal Validation menu item above Additionally the Manual Sample Site Selection menu item will de activate 1 e become disabled immediately after you create your first SRS sample design from within Help
90. e R C W Chomistek and N F Clark 1989 Conversion of electromagnetic inductance readings to saturated paste extract values in soils for different temperature texture and moisture conditions Can J Soil Sci 69 25 32 McNeil J D 1980 Electromagnetic terrain conductivity measurement at low induction numbers Tech Note TN 6 Geonics Limited Ontario Canada McNeil J D 1986 Rapid accurate mapping of soil salinity using electromagnetic ground conductivity meters Tech No TN 18 Geonics Limited Ontario Canada Rhoades J D 1992 Instrumental field methods of salinity appraisal pp 231 48 in Advances in measurement of soil physical properties Bring theory into practice G C Topp W D Reynolds and R E Green Eds SSSA Special Publ no 30 ASA CSSA and SSA Madison WI 160 Rhoades J D 1996 Salinity electrical conductivity and total dissolved salts In Methods of Soil Analysis Part 3 Chemical Methods Soil Sci Soc Am Book Series 5 Am Soc of Agronomy Inc Madison WI pp 417 435 Rhoades J D N A Manteghi P J Shouse and W J Alves 1989 Estimating soil salinity from saturated soil paste electrical conductivity Soil Sci Soc Am J 53 428 433 Rhoades J D and D L Corwin 1990 Soil electrical conductivity effects of soil properties and application to soil salinity appraisal Commun Soil Sci Plant Anal 21 837 860 Rhoades J D P J Shouse W J Alves N A Manteghi and S M Lesch
91. e Window The Salt Tolerance window which can only be invoked from the Calculate Field Summary Statistics window should be used to set the soil salinity range interval cut off levels corresponding to an appropriate salt tolerance equation These cut off levels are determined by the ESAP Calibrate program from the salt tolerance equation you select and in turn allow the ESAP program to estimate the relative yield loss occurring in your field You should first indicate which conductivity units your soil salinity samples are measured in by highlighting either the dS m or mS m Salinity unit options Next you should use the options shown in the Salt Tolerance frame to select or define the specific salt tolerance equation 151 You can select from a number of different crop types by using the Crop Type list box The selected crop type will in turn define the salt tolerance parameters 1 e the threshold and slope estimates Or if desired you can enter your own custom threshold and slope parameter estimates To do this select custom for your crop type and then type in your parameter estimates directly into the threshold and slope input text boxes Be sure to enter the threshold value in the correct units and note that the slope estimate is expressed as the reduction in yield given a 1 unit increase in ECe After you have specified the crop type you need to indicate the type of depth integration weights you would like to use in the yield l
92. e and the conductivity readings Note recall that when you have collected two signal readings at each site the z1 component simply represents an average of these two readings 139 Bivariate response variable plots Select this option to create and display a series of bivariate plots of your response variable across different sampling depths These plots are useful for determining the vertical correlation structure inherent in your response variable data Note if your sample profile data was collected from only a single sampling depth then this plotting option will not be available from within Output Print Select this option to print the currently displayed plot Save as Bitmap Select this option to save the currently displayed plot as a bitmap file from within Help Navigating the PRV Graphics Menu Select this option to display the PRV Graphics Menu help file This help file explains how to use the PRV Graphics menu commands to create display print and or save PRV plots General Tips how to highlight sights plot a different depth etc Select this option to display the General Tips help file This help file explains how you can use the various interactive plotting features associated with most ESAP Calibrate graphical displays such as highlighting sites plotting different sample depths etc Note sites displayed in PRV plots can not be highlighted from within Exit Return to Response Variable Specification Win
93. e create these plots plot the Vol H2O variable contained within the sk13_lab97 pro data file Note that these profile shapes are rather typical of most water content data i e the soil water content tends to increase with the sampling depth and the correlation between the shape of the water content profile and the average magnitude of the sample core tends to be high Raw Profile Data Yol H2o Smoothed Fitted Data r 9181 ratio ratio 0 30 No Point Selected Figure 5 3 1D raw and smoothed volumetric water content profile data plots generated from the sk13_lab97 pro sample water content data 5 7 Performing a 2D Standard bi variate Correlation Analysis A 2D scatter plot can be used to study the correlation between any two columns of your acquired calibration sample data This section describes how to create display and print 2D scatter plots as well as how to use these plots during a standard correlation analysis 113 5 7 1 What is a Standard Correlation Analysis A standard correlation analysis represents the second type of stochastic data analysis method available in the ESAP Calibrate program This analysis allows you to visualize and quantify the correlation structure between any two input columns of calibration sample data For example you could plot and examine the correlation structure between your sampled soil salinity ECe and soil texture either SP or clay Various menu options within the ESAP Calibrat
94. e data file The following material is covered in detail 1 a description of the DPPC model 2 the DPPC PDCA Profile Data Correlation Analysis window 3 the DPPC Graphics menu and 4 guidelines for understanding how to interpret the DPPC PDCA summary report and data plots 5 8 1 What is a DPPC Profile Data Correlation Analysis A DPPC Dual Pathway Parallel Conductance correlation analysis represents the third type of stochastic data analysis method available in the ESAP Calibrate program A DPPC correlation analysis can be performed on any valid DPPC profile data file either before or after merging such a profile data file with any corresponding survey data This analysis allows you to calculate i e estimate a set of soil conductivity readings based on your input sample salinity texture and water content data values These calculated soil conductivity readings can then be compared to the input soil data and or your measured i e true conductivity readings such as your input EM 38 readings etc If you use the ESAP Calibrate program much you will soon find that a DPPC correlation analysis represents a very powerful analytical procedure We recommend reading the two examples described below to develop a better understanding of how this procedure can be used to clarify the inter relationships between your profile soil sample and conductivity survey data Example 1 Suppose that your calibration sample data contains
95. e first sampling depth for 4 different transect lines Or you could overlay all 4 prediction depths for any given transect line etc If you wanted to you could even mix the above two plotting styles i e two lines with two variables displayed for each line 1D line plots are most useful for detecting non random data structure down survey passes For example such plots are commonly used to search for evidence of increasing head to tail conductivity trends in flood or furrow irrigated fields and or cyclic patterns which can indicate the effectiveness of any tile lines Such plots can also sometimes be used to detect abrupt changes in soil texture under non saline conditions 4 3 2 Navigating the 1D Line Plot Menu Bar 1D Line Plot Menu Bar Layout The 1D Line Plot menu bar is located in the upper left corner of the 1D Line Plot program window The full layout for this menu bar system is shown below Main level Sub Level 2 Plot gt Specify New Plot Variables gt Create Plot Output gt Print gt Save as Bitmap Help gt Navigating the 1D Line Plot Menu gt ID Line Plotting Tips Exit gt Return to Main Menu 1D Line Plot Menu Bar Menu Item Descriptions The 1D line plot menu contains 4 menu bar items Plot Output Help and Exit You can use the Plot menu to initialize the 1D line plotting options Specify New Plot Variables and display the initialized plot Create Plot You can use the Output menu if you wi
96. e program may be used to control how your profile correlation plots are displayed These display options allow you to create individual plots of the correlation data by depth create bulk average correlation plots and or create a correlation data plot which displays all of the depths simultaneously Correlation statistics for each of these plots can also be calculated and displayed and any of these plots can either be printed or saved as bitmap files For soil salinity sample data the correlation structure s between salinity and the other soil measurements can often be used to infer useful information about the underlying soil salinisation process For example if the correlation between salinity and soil texture is strong then this would suggest that the degree of water infiltration and hence leaching is being controlled primarily by the soil texture Likewise in the absence of a crop a strong positive salinity water content correlation would suggest that the field could be suffering from a drainage problem especially if the salinity water content correlation increases with depth Similar types of inference can often be drawn from correlation plots of other types of soil sample data depending on the physical and chemical attributes of the specific soil data being analyzed 5 7 2 Navigating the Standard Correlation Analysis Menu SCA Menu Bar Layout The ESAP Calibrate Standard Correlation Analysis SCA menu bar is located in the upper
97. e this your active project Now click on the file browse button when the File Browse window appears you should see a file called sk13data svy Double click on this file name and then click once on the OK command button after you return to the Survey Data File Import window The ESAP Calibrate program should then import this survey data file and print out the following summary information to the main program screen Field ID Code sk13 File Type Transect Sig Trans log of survey sites 100 of sig columns 2 Sig labels InEMv InEMh In addition to the above information the input file project and full file path information should also be displayed To import the corresponding profile data file information click on the File gt Import Data File gt Import a Profile Data File main menu option This will cause the Profile Data File Import window to be displayed Make sure the input file status is defined as previously saved profile data file and then set the current project to Training2 again by clicking once on the Training2 project name in the Project Directory display window Now click on the file browse button when the File Browse window appears you should see a file called sk13_lab97 pro Double click on this file name and then click once on the OK command button after you return to the Profile Data File Import window The ESAP Calibrate program should then import this profile data file and print out the following summary inf
98. e to print just the raw data Plot just the smoothed data Plot or both plots together i e side by side Save as Bitmap Select the sub options listed under this menu item to save either the raw or smoothed profile data plots as bitmap files 108 from within Statistics Basic Profile Stats by depth Select this menu option to create and display the basic summary statistics by depth for each column of calibration sample data currently contained in your profile data file Note that these calculations only need to be performed once typically at the beginning of your analysis since summary statistics for all possible plotting variables will be included in this displayed output from within Help What is a 1D Profile Plot Select this option to display the What is a 1D Profile Plot help file which explains what a 1D profile Plot is and how it should be interpreted Navigating the 1D Profile Menu Select this option to display the Navigating the 1D Profile Menu help file This help file explains how to use the 1D profile menu commands to create display print and or save 1D profile plots General Tips how to highlight sites etc Select this option to display the General Tips help file This help file explains how you can use the various interactive plotting features associated with most ESAP Calibrate graphical displays such as highlighting sites plotting different sample depths etc from within Exit Return to Main Men
99. e to the salinity levels etc In most situations the calculation display and interpretation of the predicted salinity data represents the fourth step in the salinity survey process Given the above scenario we can summarize a typical salinity survey process as shown on the next page Step Component Description 1 signal processing the collection of the conductivity survey data 2 soil sampling the collection of calibration soil sample data 3 data modeling the analysis of the soil sample data and the development of the calibration equation 4 salinity prediction the calculation display and interpretation of the predicted soil salinity data The programs in the ESAP 95 Software Package have been developed around these four survey steps The ESAP RSSD program is designed to perform the first two steps of the survey process In other words you use the ESAP RSSD program to process your soil conductivity data and generate your sampling design Likewise the ESAP Calibrate program has been designed to help you perform step three and most of step four This program can be used to estimate stochastic i e statistically based calibration models which will in turn use your conductivity survey data to make both individual site and field average salinity predictions The ESAP Calibrate program can also be used to analyze various relationships within your acquired laboratory sample data and or convert conductivity data into estimated salinity
100. e you run the software if you wish to view any output text files You can reset the text editor using the Help gt ESAP Interface Controls gt ReSet Default Text Editor menu commands located within the main RSSD Calibrate and SaltMapper program menus To locate the write exe program in NT look in the C Winnt system32 subdirectory You can also reset ESAP to use a different text editor if desired 14 ESAP 95 Software Training Files There are 3 demonstration conductivity survey files which are automatically installed by the ESAP setup program into the esap2 demo_input_files subdirectory These represent demonstration files which can be used with the ESAP RSSD software program In order to read these files into the ESAP RSSD software you need to know their format i e what sort of information each file contains This format information is given below Number Site ID Type of of Signal Column Conductivity File Name File Type Columns Present Signal Data bwd101p dat Transect 2 no EM 38 hol31 dat Grid 1 no EM 31 vertical frank 1 mws Transect 1 no USSL Mobile Wenner In addition to the above demonstration conductivity survey files there are 2 demonstration training projects which are automatically installed by the ESAP setup program into the esap2 data Training1 and esap2 data Training2 subdirectories These projects contain 3 files each an ESAP RSSD processed survey svy file and ESAP Calibrate processed profile pr
101. ear on the SCM window within the Response Variable Information frame To perform the second step click on the Model gt Identify Model Parameters SCM menu option This will invoke and display the MLR Model Identification window When this window appears highlight the Auto select an appropriate model option click on the Invoke Auto Selection Algorithm and then relax for a few moments while the ESAP Calibrate program performs the model optimization process Once this process is finished click on the Yes command button within the Accept Selection frame You have now just used the ESAP Calibrate program to auto select an appropriate set of regression model parameters and you should now see this information also appear on the SCM window the selected model should be In ECe bo bi z1 b2 z2 To perform the third step click on the Model gt Estimate Calibration Equation SCM menu option This will invoke and display the MLR Model Estimation window When this window appears check the Fit this model to the bulk average sample values check box and then click on the Estimate command button The ESAP Calibrate program will then estimate a regression model for each sample depth and inform you when the estimation process is finished After you click on the message box OK command button the Estimation window will disappear and some basic model summary statistics will appear within the Model Summary Statistics frame on the SCM windo
102. eate 2 D raster maps of your input survey or prediction data Note that the raster map interpolation routines allow you to control the smoothness of your final map as well as the mapping boundaries 61 from within Help About ESAP SaltMapper Select this option to display the ESAP 95 Software information window ESAP Interface Controls ReSet default Text Editor You may use this option to change the default text editor used by ESAP 95 to display and print all ESAP generated help files At program start up the default text editor package is set to c windows write exe OnLine Help What is ESAP SaltMapper Select this option to display the introductory help documentation This is the help documentation you should read first if you have never used the ESAP SaltMapper program before This documentation explains what the ESAP SaltMapper program does and how the program works OnLine Help Navigating the Main Menu This documentation explains how to use the Main menu bar and describes the menu bar features OnLine Help Frequently Asked Questions FAQ This is the help documentation you should read if need more detailed information about how to use the various program features For example refer to this documentation if you don t understand the difference between a svy and prd file or you wish to understand how a 2 D raster image is created etc 4 2 Data Input and Manipulation The following section describes how to impo
103. ectrode and electromagnetic induction EM survey instruments Within the last 15 years the adaptation of four electrode and EM sensors for soil electrical conductivity measurement has greatly increased both the speed and reliability of salinity reconnaissance survey work The efficient use of conductivity signal information requires the conversion of apparent soil conductivity EC into soil salinity EC A significant amount of research in recent years has been directed towards developing efficient conversion techniques Williams and Baker 1982 McNeill 1986 1980 McKenzie et al 1989 Rhoades and Corwin 1990 Rhoades et al 1999 1991 1990 1989 Rhoades 1996 1992 Slavich 1990 Cook and Walker 1992 Diaz and Herrero 1992 Yates et al 1993 Lesch et al 1998 1995a b 1992 These conversion techniques can generally be classified into one of two methodological approaches 1 deterministic and 2 stochastic In the deterministic approach either theoretically or empirically determined models are used to convert conductivity into salinity Deterministic models are Static i e all model parameters are considered known and no soil sample soil salinity data needs to be collected during the survey However these models typically require knowledge of additional soil properties e g soil water content texture temperature etc In the stochastic approach statistical modeling techniques such as spatial regression or cokrigi
104. enu option This window contains its own set of help file documentation you should read this documentation if you wish to learn more about interpreting the various residual diagnostic plots and or how to navigate the Residual Diagnostics menu 159 6 0 References Cook P G and G R Walker 1992 Depth profiles of electrical conductivity from linear combinations of electromagnetic induction measurements Soil Sci Soc Am J 56 1015 1022 Diaz L and J Herrero 1992 Salinity estimates in irrigated soils using electromagnetic induction Soil Sci 154 151 157 Lesch S M J D Rhoades L J Lund and D L Corwin 1992 Mapping soil salinity using calibrated electromagnetic measurements Soil Sci Soc Am J 56 540 548 Lesch S M D J Strauss and J D Rhoades 1995a Spatial prediction of soil salinity using electromagnetic induction techniques 1 Statistical prediction models A comparison of multiple linear regression and cokriging Water Resour Res 31 373 386 Lesch S M D J Strauss and J D Rhoades 1995b Spatial prediction of soil salinity using electromagnetic induction techniques 2 An efficient spatial sampling algorithm suitable for multiple linear regression model identification and estimation Water Resour Res 31 3387 398 Lesch S M J Herrero and J D Rhoades 1998 Monitoring for temporal changes in soil salinity using electromagnetic induction techniques Soil Sci Soc of Am J 62 232 242 McKenzi
105. er there are no other specific formatting requirements i e your individual data columns do not have to follow any special format nor do the numbers need to be column aligned like the data shown above If your laboratory data file you wish to import does not include measurements of salinity texture and soil water content then you should import it as a Generic file type By design all generic type input data files have the following general column structure Generic Column Structure Column 1 site_ID required 2 sample depth midpoint required 3 1st soil measurement required 4 10 additional soil measurements optional Hence in a Generic file your input data in columns 3 through 10 can represent any soil variable you may have measured However the input data file must still be double sorted first by the site_ID numbers and then by the sampling depths For example suppose you collected sample data from 3 depths at each of 10 sites across the survey area and measured two soil properties on each sample Then your input data file might look like the abbreviated data file shown below 17 OLD 23 24 1253 17 0 45 26 6 10 9 Li Dao S505 SES 36 28 Oto AL D9 28 0 45 47 7 22 2 28 USLI Daraa 2 lL L 188 0 15 17 4 10 188 0 45 14 3 11 VES Oe 225 2 13 coo N In the above example the first sample site comes from survey site number 17 the last sample site comes from
106. er Sites are from the rest of the decorrelated survey data Once you have finished viewing this plot exit out of the IG window RSD Scatter Plot z1 versus z2 4 66 Corr sl 82 r 0 0000 Figure 3 7 A rsd plot of the decorrelated bwd101p dat EM 38 survey data 46 Based on the rsd plot it would probably be best to delete these two Sites and re preform the signal decorrelation To do this click on the Analysis gt Signal Validation menu option make sure the delete all outliers option is high lighted and then click on the Invoke Validation command button This should cause the Signal Decorrelation window to be re displayed Now perform another decorrelation by clicking on the Perform Decorrelation command button Note that no further outliers are detected although the 3 masked Sites are still present At this point you have now determined that your conductivity survey information contains no further outlier data Sites Hence click on the Finished command button in the Signal Decorrelation window and then click on the Analysis gt Signal Validation menu option to re display the Signal Validation window You should now see that the no outliers detected option is high lighted implying that this data has been validated Although it is not necessary for this data set for practice you should go ahead and check the Invoke Edge Buffering Algorithm option and make sure that the mask all Sites in 1 and last rows option i
107. er first and within each row the conductivity survey data should also be sorted by either 1 time sequence if the survey data was collected in a sequential manner or 2 the x y survey coordinates if the survey data was collected in a non sequential manner Grid files do not have to be sorted in any manner unless you wish to use the data parsing option within the SRS Sampling Design Window If you do wish to use this option we recommend that you sort your Grid file data by the x y survey coordinates Finally neither type of input file is allowed to contain header or trailer records blank lines extra data columns other than the columns explicitly mentioned above character data or missing data values anywhere in the data file The abbreviated data files shown below represent 4 of the 8 possible file structures which the ESAP RSSD program can process In each example only the first three and last two rows of data are shown Note it is not necessary that your input column data values be format aligned as shown in these examples Example 1 Grid file type no site ID column no 2nd signal data column structure x y s1 18 00 793 75 1 1414 18 00 781 25 1 2086 18 00 768 75 1 2848 L 288 00 718 75 0 9246 288 00 731 25 0 9246 Example 2 Grid file type all columns present structure site id x y sl s2 1 18 00 793 75 1 1414 0 7538 2 18 00 781 25 1 2086 0 7720 3 18 00 768 75 1 2848 0 8240 L 1016 288
108. ery useful when you need to generate maps of multiple prediction variables and you wish to use the same boundary for each displayed map Note boundary files can be saved using the Save Clipping Boundaries option on the Raster Map Initialization menu Step 4 Set Kernel Size You can control the smoothness of the interpolation process by adjusting the interpolation kernel size In the ESAP SaltMapper program the size of the kernel simply defines the size of the surrounding area searched for each grid interpolation point In general you should make the horizontal and vertical kernel values large enough to encompass at least 6 to 8 individual survey or prediction points For sparse survey grids the kernel values may need to be set up to 20 or more Likewise for dense grids you may be able to set the values down to 6 Note also that the horizontal and vertical values do not have to be equal to one another To set the horizontal and vertical kernel values simply click on the upper and lower scroll bars Step 5 Perform Interpolation Once you have finished setting the kernel size you can initiate the interpolation process by clicking on the Perform Interpolation button When the ESAP SaltMapper program finishes the interpolation process you may see up to the following three colors displayed within the interpolation window green red and white All areas within the window displayed as green represent survey zones where the interpolati
109. ess may go by quite quickly if you are working on a fast computer In this example the final Opt Criteria value should be equal to 1 202 Since this is below 1 3 you can go ahead and save this design by clicking on the Save Current Design command button After you have done this you should see a message which says SRSD sampling design information has been saved to BVWDrsd1 txt This message is telling you that you have successfully saved your ESAP RSSD generated SRS sampling design information to the above referenced text file 51 That s all there is to it However for additional practice you might at this point try generating one or two additional designs Feel free to change the sample size setting to either 6 or 20 and perhaps adjust the design factor You might also try invoking the parsing option especially if you want to generate a sampling design of size 6 which achieves an Opt Criteria close to 1 3 3 6 ESAP RSSD Data Output This section describes how to view print and or save the various output text files and data plots produced by the ESAP RSSD program 3 6 1 Text Output Unlike ESAP Calibrate or ESAP SaltMapper the ESAP RSSD program automatically creates and saves output text files relating to your analysis If you have worked through the practice modules you now probably realize that these files are automatically created at specific points throughout the operation of the program For example immediately a
110. fic transects i e rows you wish to display in your 1D line plot This window can be displayed by clicking 68 on the Plot gt Specify New Plot Variables line plot menu option You can specify up to 4 transects per plot and the data from each transect can represent different signal or prediction variables Before a line plot can be created you must specify which set of axis coordinates to plot the data by i e either the x axis or y axis coordinates You will also need to select the variables and rows you wish to plot Additionally you can set the line and symbol plotting styles for each line plot you create as described below 1 If you plot 2 or more lines of data you can choose to display the lines using either the same color of different colors 2 If you plot 2 or more lines of data you can choose to display the lines using either the same line style solid or different line styles 3 You can choose to show or hide the symbols on each line of data you plot Note each symbol shown represents a unique data point along the line 4 If you plot 2 or more lines of data and choose to display the data point symbols you can specify whether the symbols should be the same or different across the different lines Once you have initialized all of the various options you should click once on the Initialize button to initialize the 1D line plot Note initializing the plot does not cause the plot to be displayed to display
111. fter you completed the data validation step the ESAP RSSD program created and saved a text file called xxxxinfo txt where xxxx corresponds to your 4 character alpha numeric field ID code Likewise the first time you saved an SRS sampling design the ESAP RSSD program created a text file called xxxxrsd1 txt If you saved a second SRS design then ESAP RSSD also created a text file called xxxxrsd2 txt Finally if you had created and saved a manually generated sampling design then ESAP RSSD would have created a text file called xxxxugsd txt ugsd is an abbreviation for user generated sampling design Any of the above mentioned files can be viewed or printed by clicking on the File gt View Print Output Files menu option You can then highlight and open these text files using your default text editor which will be WordPad unless you have reset it to something else If desired you can save any or all of these files under different file names using the save file as feature in your text editor The contents of each type of ESAP RSSD output text file are summarized below File Type Contents Xxxxinfo txt General data processing information including all information displayed in the Project Status frame except for any sample site design information XXXxrsd txt SRS sample design information including a list of the physical sample site locations and a summary of the SRS algorithm settings 52 Xxxxgps txt A sorted list of t
112. ghlighting options i e highlight all masked and outlier sites and highlight all selected sample sites By default both highlighting options are automatically turned on when the Interactive Graphics Window is first activated Additionally you can also change the ColorScale symbol plotting size and or change the ColorScale grid to a GrayScale grid provided you initialize the ColorScale grid graphing option The Coordinate Translation window can be used to adjust the survey location x y coordinates corresponding to the input conductivity survey data You can invoke and display the 34 CT window by clicking on the Options gt Perform Coordinate Translation menu option However you should be aware that all coordinate translation s are applied to the original x y input data Hence any translation will remain in effect until it is removed by applying a reverse translation Generally speaking the only time you should invoke this window is 1 if either the x and or y axis of your input coordinate system needs to be reversed or swapped or 2 you need to temporarily rotate the coordinate system in order to plot your transect survey lines along the north south or east west direction In either case the sample site selection algorithms in the ESAP RSSD program will not be affected by any of the translation options available in this window However in the second scenario you will need to apply a reverse translation in order to restore the
113. gnal column by this constant factor and thereby scale your data up or down in a proportional manner If you do not wish to use this feature you should simply leave the default factors set to values of 1 0 Your next step should be to set the conversion formula type When you use depth weighted electromagnetic induction survey instruments like an EM 38 the signal data generated by these instruments must be converted into depth specific soil conductivity values using some type of conversion formula ESAP allows for two formula types a linear type and a log linear type You are free to use either type provided you can specify the parameter values associated with the selected formula type However you will generally select your formula type based on the instrument you are using For example if you wish to use the 1992 Rhoades equations to convert your EM 38 signal data into depth specific soil conductivity data you should specify i e highlight the log linear formula type On the other hand if you have acquired depth specific soil conductivity data using a four electrode horizontal array system for example Veris 3100 data then you should specify the linear formula type Once you have chosen the conversion formula type you need to specify the actual parameter values for the selected formula For example suppose your own experimentally determined conversion formula is linear and defined to be ECa 1 23 0 67 s1 0 43 s2 Then a0
114. he Back transform mean field log predictions check box this instructs the ESAP Calibrate program to back transform the average field log salinity estimates into median estimates Next click on the Set Levels command button After a warning message displays the Salt Tolerance window will appear Now select sorghum from the drop down crop list highlight the linear weighting across depths option and then click on the OK command button You have now just set your range interval cut off levels using the threshold slope salt tolerance equation for sorghum This in turn will allow the ESAP Calibrate program to estimate the relative expected yield loss in this field in addition to the field median salinity levels and range interval estimates Finally click on the Calculate Statistics command button to perform the summary calculations Once the calculations are finished click on the View Output command button to view the summary statistics The statistical output in your summary file should be identical to the output shown in table 5 2 In most situations you should print out and save the generated summary statistics text file This output file will contain all of the pertinent summary statistics for your survey area including the predicted median salinity estimates by depth the range interval area estimates and the expected relative crop loss if requested In this training example the median ECe estimates all lie between 10 5
115. he Internet then move to the folder which contains the down loaded program code locate the setup exe program and then double click on this program to initiate the installation procedure If you have previously installed an earlier version of the ESAP 95 Software Package i e a 2 00 or 2 01 Beta version then be sure to read the Important Notes for Re Installs listed at the end of section 1 3 The only option you will be asked to specify during the installation procedure is the installation sub directory location The default location will be a sub directory called US_Salinity_Lab esap2 We recommend that you use this sub directory unless you have a good reason to choose another location note ESAP 95 will automatically create this sub directory if it does not already exist Please note that if you have other background programs running during the installation process then ESAP 95 may not install properly If you encounter an error message which states file access error occurred or something similar then you probably have some type of background program running on your Windows system which needs to be shut down before the installation process can work correctly This error will occur when another program using one or more dll or ocx files which ESAP 95 is attempting to install This usually means that you already have the dll or ocx file in question hence you could choose to ignore this error and continue with the installation proces
116. he plot will display a cyclic pattern or head to tail salinity redistribution effects the conductivity survey data will tend to increase in magnitude down the transect The line plotting option is only available if you input a transect type conductivity survey file Additionally this option is disabled when the Graphics Window first activates you need to enable i e activate it by opening up the GI window and initializing the line plotting option When you initially request a line plot ESAP will display a blank graph On the left hand side of this graph you will see a frame bar titled Line Plots with the following command buttons gt gt lt lt Plot Refresh s1 Stats s2 Stats and Done You can use the Plot button to create and display a line plot for any valid transect row number you request the transect row number will be displayed immediately to the right of the Plot button You can adjust the transect row number using the gt gt up and lt lt down buttons and you can request that ESAP compute signal statistics for the current transect row number using the s1 Stats and or s2 Stats buttons If you wish to clear i e erase the current graph you must use the Refresh button Finally when you are finished viewing the last line plot you should use the Done button to exit the line plotting mode The statistics produced in the line plotting mode include the mean u standard deviation s and c
117. he raster interpolation window This will initialize 1 e reset the interpolation memory array and clear the interpolation window Step 2 Show I Grid Survey Sites Click on this button to display the interpolation grid and plot the survey or prediction site locations within the raster interpolation window Step 3a Clip Grid optional Click on this button if you wish to clip away any sections from the interpolation grid else proceed directly to either step 3b or step 4 From time to time you may acquire survey data which does not originate from a rectangular field In these situations you may need to clip i e remove one or more sections of the interpolation grid so that the ESAP SaltMapper program does not attempt to interpolate predictions into these non surveyed areas If you wish to initiate the clipping feature you can do so as follows After clicking on the Clip Grid button move the mouse to the upper left corner of the area you wish to remove from the interpolation grid and click the left mouse button once This initiates the clipping process Now move the mouse to the lower right corner of the area to be clipped As you move the mouse you should see an expanding box appear on the screen the interior of this box represents the area you can clip Once you have enlarged the box to cover the area you wish to remove you should click the left mouse button a second time This second click defines the lower right corner of the
118. he sample site locations and ID numbers produced along with each rsd txt output file this text file can be used as input to a GPS data logger for automatically re locating the sample Sites xxxxugsd txt User generated i e manual sample design information 3 6 2 Graphical Output All graphs produced by the ESAP RSSD program are created using the Interactive Graphics window Additionally any graph you create and display can also be printed using the Print gt Print Current Graph menu options within this window Unlike the SaltMapper and Calibrate programs the ESAP RSSD program does not currently allow you to save any graphs as bitmap files 3 6 3 Creating an Output SVY Data File Creating the final output svy data file represents a critical step in your survey data analysis session you must create this file if you wish to continue processing your survey data using either the ESAP SaltMapper or ESAP Calibrate programs For this reason the ESAP RSSD program has been designed to automatically generate and save this file when you exit out of your analysis session However you MUST exit out of the program using the File gt Exit main menu options If you click instead on the upper right X close program button on the main RSSD window to exit the program then your svy data file WILL NOT BE CREATED Hence you should NEVER exit the ESAP RSSD program in this manner by clicking on the close program button unless you do not wi
119. how to use the various program interface features After collecting any type of conductivity survey data there are four basic steps that you must perform within the ESAP RSSD program in order to generate a sampling plan These steps are as follows 1 read your survey data into the program 2 visualize your survey data i e interactively examine the data using various graphical techniques 3 transform decorrelate and validate the data and 4 generate the final sampling design The entire ESAP RSSD program is designed around performing these 4 basic steps Indeed the main menu layout directly reflects this 4 step process File Graph Analysis and Design simply refer to the 4 steps shown above When you first start the ESAP RSSD program you will use the features under the File menu item to set or create your working project define your field ID code and then import your conductivity survey data file Next you will use the Graph menu item to open up the Interactive Graphics Window which contains all of the interactive graphics routines for visualizing your survey data After viewing your data you can then use the features under the Analysis menu item to decorrelate and validate your conductivity survey data and after this step you can employ the routines under the Design menu item to generate the final sampling plan 18 3 1 2 Navigating the ESAP RSSD Main Menu Bar Main Menu Bar Layout The ESAP RSSD Main menu bar is located in
120. ically these maps are used by the sampling crew to navigate back to the soil sampling sites Like a survey grid all sample site maps are shown true to scale Additionally this plotting option is automatically enabled once one or more sampling designs are generated by the ESAP program 3 3 2 Navigating the ESAP RSSD Graphics Menu Bar Graphics Menu Bar Layout The ESAP Graphics menu bar is located in the upper left corner of the Graphics Window The full layout for this menu bar system is shown below Main level Sub Level 2 Sub Level 3 Graphics gt Survey Grid l gt Scatter Plots gt slvsx gt slvsy gt s2vsx l gt s2vsy gt slvs s2 gt rsd plot l gt Line Plots gt sl by row l gt s2 by row l gt sl and s2 by row l gt Histograms gt sl l gt s2 l gt zl l gt z2 l gt trans analysis l 31 gt ColorScale gt sl Grids gt s2 gt zl gt z2 gt trans analysis gt overlay sample sites gt Sample Site Map Options gt Initialize Graphic Components gt Perform Coordinate Translation Print gt Print Current Graph Help gt OnLine Help gt Navigating the Graphics Menu gt Graph Descriptions Information Exit gt Return to Main ESAP Program Menu Graphics Menu Bar Menu Item Descriptions The main level contains 5 menu bar items Graphics Options Print Help and Exit which are self expl
121. ics and a full range of graphical residual plots Note if you do not have any prior regression modeling experience you do not need to access any of these sub menu options 135 from within Predict View prediction plots Select this option to view the accuracy of the stochastic calibration regression model predictions When you select this option ESAP will display the 2D Prediction Scatter Plot window which you can use to create plots of the observed true versus model predicted soil variable data Calculate field summary Statistics Select this option to calculate your field summary statistics When you select this option ESAP will display the Field Summary Statistics window which you can use to produce summary statistics for your survey area These summary statistics include the field mean or median predictions survey area interval estimates and when estimating soil salinity the expected relative crop yield loss for a specific crop type due to the current salinity condition Save output predictions You should select this option to save your regression model predictions as an ESAP output prd data file By design any prd output file can be automatically imported into the ESAP SaltMapper program and used to construct spatial prediction maps of the soil variable s you have modeled from within Test Calculate Net Flux tests Select this option to display the Net Flux Calculation window If you acquire additional soil
122. iew article concerning the practical application of the DPPC equation for soil salinity estimation is Rhoades J D 1992 Instrumental field methods of salinity appraisal pp 231 248 In G C Topp et al ed Advances in measurements of soil physical properties bringing theory into practice SSSA Spec Publ 30 SSSA Madison WI 5 8 3 Using the DPPC PDCA Window All analysis routines associated with a Dual Pathway Parallel Conductance Profile Data Correlation Analysis can be accessed from the DPPC PDCA window which may be invoked and displayed by clicking on the Calibrate gt Stochastic Methods gt DPPC Correlation Analysis main menu option These routines include 1 merging your conductivity survey and profile data files 2 log transforming the calculated ECac ECe and or any other input profile data columns 3 displaying the DPPC correlation summary report and or graphics window and 4 saving the DPPC profile data correlation analysis summary report as an ASCII text file Each of these analysis options are described below in more detail 122 File Merge Options The options contained within this frame can be used to merge your profile and survey data files together provided that you have already imported a svy survey data file when you invoke the DPPC PDCA window You should first check the Merge active Survey Profile Data File information check box if you wish to merge your survey and profile data files together Af
123. ification errors can generally be fixed inside of the ESAP Calibrate program by re displaying the appropriate Format Sheet and changing the incorrect specifications However data sorting errors will need to be fixed outside of the ESAP Calibrate program using some other type of software 93 Note also that the ESAP Calibrate program does NOT have the ability to selectively import a subset of data columns from a larger data file In other words suppose your ASCII data file contains 12 columns but you only wish to import a subset of 6 columns In such a situation you can NOT import the 12 column data file directly into the ESAP Calibrate program Instead you must first create a new data file which only contains the 6 columns of data which you wish to use and then import this smaller data file 5 3 3 Practice Module Importing New Data For this practice module you will use the ASCII data file called sk13_97 lab which is located in the Training2 project As with previously saved profile data you first need to invoke and display the Profile Data File Import window in the identical manner as you did in Practice module 5 2 3 However once this window displays you need to perform the following extra steps when you import new calibration sample data for the first time First set the input file status to new profile data file Next define the project status i e decide whether you wish to create a new project directory or set the current p
124. ighting this check box Note this action will permanently replace the gravimetric water content data with estimated volumetric water content data values Estimate H20 Relative to FC Field Capability If desired ESAP can calculate the percent water content relative to estimated field capacity and 96 store this information into a new data column This option is available regardless of the type of input water content data volumetric or gravimetric Once you have indicated any desired conversion options if applicable click once on the Compute Summary Statistics command button to invoke the data validate routines If you are working with generic data then the ESAP Calibrate program will simply summarize your input calibration sample data If you are working with DPPC data and one or more data boundary errors are detected then you can view a listing of these errors by clicking on the View Summary Statistics command button Even if there are no boundary errors detected we recommend that you view this log file since this is where your profile summary statistics will be reported We also recommend that you generate a print out of this file for your records If you feel that one or more inputted data values are in error then you have the option of editing these values Note that a separate data editing window can be invoked and displayed by clicking on the Edit Data command button refer to section 5 4 2 for more details If all of y
125. il conductivity survey data represents the controlled input information and your soil variables salinity texture etc represent the process output 17 There are many additional statistical issues which must be addressed when you attempt to use an ordinary regression model to predict the values of a spatial data set For example spatial data is often highly correlated i e spatially correlated and hence the measured values of soil samples acquired close together are often more similar than those sampled farther apart Likewise the input conductivity information is almost always spatially correlated and highly collinear The ESAP RSSD program has been designed to deal with these statistical issues and to help you sample in a way which minimizes their impact Most importantly the ESAP RSSD program has been designed to accept survey data generated by nearly all types of commercially available conductivity instruments including both invasive direct contact and non invasive electromagnetic sensors Some examples of the type of bulk soil conductivity sensor data that ESAP can process include data generated by the Geonics EM 38 and EM 31 instruments Martek conductivity meters and Verris 3100 conductivity sensor systems How does the ESAP RSSD Program Work It is important to realize that you DO NOT need to be a statistician in order to effectively use the ESAP RSSD program You simply need to understand a few basic concepts and learn
126. ile shapes often imply that over leaching is occurring especially if the magnitude of the soil salinity levels are low Similar types of inferences can often be drawn from profile plots of other types of calibration sample data depending on the physical and chemical attributes of the specific calibration data being plotted One very useful characteristic that can be inferred from a 1D profile plot of any soil variable is the determination of the degree of dependency of the profile shape on the average magnitude of the corresponding sample levels In other words just how predictable is the profile shape given that we know the overall average level of the soil sample data across all the sampling depths This is referred to as a shape magnitude interaction analysis the more predictable the shape is given the average magnitude level the more deterministic the process is said to be This is equivalent to saying that we can accurately predict the value of the soil attribute at each sampling depth provided that we simply know the average level of the soil attribute across all the sampling depths In the ESAP Calibrate program every 1D profile graph contains two profile plots The 106 first plot shown on the left side of the 1 D Profile Plot window displays the raw sample data while a second plot shown on the right side of the window displays the fitted 1 e predicted sample data based on the calculated average soil core levels Addi
127. ility to produce 1D profile data graphs and perform bivariate profile data correlation analysis d the ability to fully automate the stochastic calibration regression modeling process The ESAP Calibrate program has been designed to operate on two types of input data survey data i e svy data files produced by the ESAP RSSD program and calibration sample data i e the laboratory analyzed soil sample data collected from a set of calibration sample sites The latter data is referred to as soil profile data or simply profile data in this manual because the ESAP Calibrate program can import and process calibration soil sample data from multiple sample depths In version 2 01 up to 6 sampling depths can be associated with each calibration sample site All calibration sample data imported into the ESAP Calibrate program should be saved as a permanent ESAP Calibrate pro profile data file This feature allows you to quickly and easily import the same data back into the program during any future data analysis sessions 81 Additionally whenever you generate output spatial prediction data you should save this data as a permanent prd prediction data file These prd data files can then in turn be imported into the ESAP SaltMapper program and used to create either 1D transect plots or 2D raster maps How does the ESAP Calibrate Program Work As discussed above the ESAP Calibrate program contains a number of modeling and
128. information using deterministic modeling techniques Finally the ESAP SaltMapper program can be used to display the predicted spatial salinity data i e to generate 2D salinity maps and or 1D salinity transect plots In other words this program can be used to perform the display component of the fourth step ESAP SaltMapper can also be used to display the acquired soil conductivity data if desired It is worthwhile to note that not all salinity surveying techniques employ all four of the steps shown above For example when deterministic calibration equations are used to predict salinity from conductivity one does not normally collect any soil sample data Hence such a survey would not include step 2 as described above nor would it rely on any sort of statistically based calibration equation Likewise there may be situations were previously collected soil sample data is available from a field but no conductivity survey data has yet been collected Itis actually possible to use the ESAP Calibrate program to estimate how effective a conductivity survey would be for predicting various soil properties providing the right type of soil sample data has been acquired without having any actual conductivity survey data available to analyze Such an analysis is known as DPPC Correlation Modeling and based on the above survey definition one would only be performing part of step 3 As you gain experience using the various ESAP 95 programs b
129. ining2 project directory Once you have opened this data file using the File Browse window click once on the Import window OK command button to initiate the data import process After the data has been successfully imported the ESAP Calibrate program will automatically display some summary file information on the main program window and also remind you that this file has not yet been saved as a permanent profile data file The directions given above should be followed whenever you need to import a generic type data file However the sk13_97 lab data file is actually a valid DPPC type data file so now you should re import this data as a DPPC file type To do this open up the Profile Data File Import window again and set or create your project Once you are ready to define the input file format click on the DPPC command button This will invoke and display the DPPC Profile Data File format sheet which should be filled in with the following information Number of sample sites 15 Number of sampling depths per site 4 Number of secondary variables measured per site 2 Column Units 2 meters 3 dS m Column 4 check the SP check box Column 5 check the Gravimetric check box Column 6 check the NOT present check box Column Label Units 6 SAR none or leave blank 7 Boron ppm After you have entered this information click again on the format sheet OK command button to return to the Profile Data File Import
130. ional sample data at one or more calibration sites in the future This window can be invoked and displayed by clicking on the Test gt Calculate Net Flux Tests SCM menu option A net flux testing situation could arise as follows Suppose you perform an electromagnetic survey on a salt affected field select 12 survey locations for soil sampling and then use the ESAP Calibrate program to generate predicted salinity data After examining the predicted salinity map the farmer decides to leach the field by applying 2 acre feet of irrigation water for a 30 day period After 30 days i e after the leaching process you re enter this field 156 re sample the 12 calibration sites acquiring 12 new soil cores and determine the new soil salinity levels at these 12 sites Then by using the options in the Net Flux window you can enter this data into the ESAP Calibrate program and test whether or not the salinity pattern has changed and if so by how much The above example focuses on soil salinity However the ESAP Calibrate program can perform net flux tests on any soil variable you sample The only requirement is that the locations of your new sample sites must coincide with your original calibration sites Mathematically it is possible to test for a change in a predicted soil variable over time provided that the monitoring site locations coincide with any of the original survey grid locations for example see Lesch et al 1998 Monitoring fo
131. ions Pri Sec bulk depth Soil depth ave specific Variable level ln Calc ECa ln Calc ECa 1n ECe 0 15 0 503 0 745 0 45 0 675 0 787 0 75 0 836 0 878 1 05 0 817 0 893 SP 0 15 0 267 0 948 0 45 0 612 0 602 0 75 0 643 0 641 L205 0 403 0 423 IV ECe ECac Signal Deterioration by depth This section displays the calculated ECe ECac signal deterioration by depth due to spatial variation in the remaining primary soil variables i e texture water content and bulk density if measured bulk density data is available ESAP calculates the signal deterioration as a percentage from 0 to 100 where 0 represents no deterioration and 100 represents complete deterioration In general calculated values over 15 indicate moderate signal deterioration and values over 25 represent serious deterioration Typically if the sample levels of the remaining primary soil variables are fairly constant across your survey area then the signal deterioration will be minimal On the other hand if these data values are highly variable and poorly correlated with the soil salinity levels then the signal deterioration can be extreme In the example shown below the signal deterioration in the 0 15 and 0 45 sampling depths is severe 44 6 and 40 8 respectively 126 IV ECe lt gt Calc ECa Signal Deterioration by depth depth Signal level Deterioration 0 15 44 6 0 45 40 8 0 75 22 6 1 05 19 8 ave 26 2
132. irst converting the raw instrument conductivity into depth specific soil conductivity ECa for a given depth and then by estimating the soil salinity levels from this ECa data using a linear version of the DPPC model 86 from within Help About ESAP Calibrate Select this option to display the ESAP information window ESAP Interface Controls ReSet default Text Editor You may use this option to change the default text editor used by ESAP to display and print all ESAP generated text files and or print any ESAP help file At program start up the default text editor package is set to c windows write exe OnLine Help What is ESAP Calibrate Select this option to display the introductory help documentation This is the help documentation you should read first if you have never used the ESAP Calibrate program before This documentation explains what the ESAP Calibrate program does and how the program works OnLine Help Navigating the Main Menu This documentation explains how to use the Main menu bar and describes the menu bar features OnLine Help Frequently Asked Questions FAQ This is the help documentation you should read if need more detailed information about how to use the various program features For example refer to this documentation if you don t understand what sort of input data each analysis procedure is designed to work on or if you need advice on where to find more detailed explanations for each procedure etc
133. isting of the file merging results Note you do not have to view this log file unless an error occurs during the file merging process Depending on the specific soil variable you are testing one or more of the following testing options may be enabled If necessary these options can be invoked as described below Log transform the input monitoring data values If your response variable has been log transformed you will be given the option of log transforming your input monitoring data If you wish to apply such a log transformation then highlight this log transform option 157 Calculate a Net Flux test on the bulk average estimates If you have estimated a bulk average stochastic calibration model and you have imported monitoring data corresponding to each original calibration sample depth then you will be given the option of testing for a bulk average change If you wish to test for such a change then highlight this bulk average option Calculate Net Flux estimates for each monitoring site If desired you can have ESAP calculate net flux estimates for each monitoring site on a depth by depth basis in addition to the field average net flux calculation which by default is always computed You should highlight this option if you wish to have ESAP calculate these additional individual net flux estimates Once you have specified any of the applicable testing options described above click on the Calculate Net Flux Test Result
134. ling designs in the ESAP RSSD program However unless you wish to generate either multi stage or composite sampling plans you probably will not need to use this feature 3 5 3 Practice Module Generating an SRS Sampling Design In this practice module you will learn how to use the basic features of the SRS Sample Site Selection window to generate spatial response surface sampling plans As with all the previous practice modules you will be using the bwd101p dat conductivity survey data as your training data set Therefor in order to complete this module you will need to have already performed practice modules 3 2 3 3 3 4 and 3 4 4 You can display the SRS Sample Site Selection window by clicking on the Design gt Calculate SRS Sample Design main menu option After this window first appears you should first set the sample size In the ESAP RSSD program the sample size will always default to n 12 and for this practice module you should leave it at this setting You should also note the Algorithm Status message shown in green type this message label will always display the current status of the sampling algorithm To generate your design click on the Invoke SRSS Algorithm command button The ESAP RSSD program will then display your survey grid cycle through a series of algorithm iterations until an optimal design is determined and then display the final optimization criteria value Opt Criteria in the status message label This proc
135. ll simply summarize your input data Regardless of the type of input calibration data file you import DPPC or generic you must run this validation routine once before you can save your calibration data as an ESAP Calibrate pro data file and or use any of the Stochastic Data Analysis Methods The validation routine can be invoked by clicking on the File gt Edit or Validate Profile Data main menu option This will invoke and display the Summarize Edit Save Profile Data window which can then be used to invoke the validation editing and file saving routines If you have imported a DPPC type data file you can perform up to three data column conversion routines when you validate your data Note if you are validating Generic profile data then all of the options discussed below will be disabled These conversion routines can be invoked by checking the appropriate check boxes shown in the DPPC Profile Data Options frame If selected they will perform the following actions Convert Clay To Estimated SP If your input texture data is percent clay then you can convert this data into estimated SP values by highlighting this check box Note this action will permanently replace the clay data with estimated SP data values Convert Gravimetric H2O To Estimated Volumetric H20 If your input water content data is measured on a gravimetric basis then you can convert this data into estimated volumetric water content values by highl
136. lp file In its entirety the on line documentation contains more detailed information on the ESAP 95 software package than this user manual However by the time you finish reading this manual you should understand how to use all of the basic features contained within each software program 16 3 0 ESAP RSSD Software Program 3 1 ESAP RSSD Program Overview The following section gives an overview of the ESAP RSSD software program including a review of how the program functions and a description of the main menu bar layout 3 1 1 Program Description What is ESAP RSSD ESAP RSSD is a statistical program which generates optimal soil sampling designs from bulk soil electrical conductivity survey information The ESAP RSSD program is part of the ESAP 95 software package for Windows a multi program software package designed and distributed by the Salinity Laboratory for the sampling assessment and prediction of soil salinity and or other soil variables from electrical conductivity survey data When you perform a soil electrical conductivity survey you typically want to use this survey information to estimate the values of one or more soil variables For example you may want to estimate the spatial soil salinity pattern across your survey area or perhaps the soil texture or water holding capacity etc In order to facilitate this estimation process you typically need to calibrate your soil variable s to the soil electrical conductivi
137. lue decreases Note that you will need to repeat this process for every available strata in order to determine which strata produces the lowest Opt Cri value you should not save any of the generated SRS designs while you determining what their Opt Cri values are Once you identify the best strata to use you should invoke the parse data option one last time select that particular strata generate the SRS sample design again and then save it The data parsing routine will often create more than one strata which can produce SRS Opt Cri design values below 1 30 Note that any of these strata can be used to produce your final design Also if none of the strata produce good designs then try changing the design factor value by 0 1 units and repeating the process 50 Additional Considerations If you have invoked the edge buffering option then the ESAP RSSD program will not select any sample Sites from the edge buffered area regardless of how you adjust any of the advanced design options If you have invoked the transition analysis option you also have the option of ignoring the TA information You should be aware that sometimes a transition analysis will improve the optimization criteria and sometimes this analysis will degrade it However in general we don t recommend employing and or removing the TA information to improve the optimization criteria Finally keep in mind that you can generate up to 5 separate i e non overlapping SRS samp
138. mand button then the Generic Profile Data File Format Sheet will appear As before you should then use the controls on this sheet to specify the exact structure of your input file Note that you will need to specify the following information 1 the number of sample sites i e sample cores 2 the number of sampling depths associated with each site from 1 to 6 3 the total number of input columns from 3 to 10 this includes the site ID and sample depth data columns 4 the sampling depth units such as inches feet or meters etc and 5 the remaining column labels and measurement units for each soil variable Once all of this information has been specified you can again use the Review Format command button to review your input information and the OK command button to save this information After you click on the OK command button in either Format Sheet you will return to the Profile Data File Import window You should now use the Browse command button to locate and open your calibration sample data file and then click on the Import window OK button to import your data As the ESAP Calibrate program imports your data it will perform some preliminary data sorting checks One or more input data read errors will occur if your data is not properly sorted or your data file format is mis specified If this occurs you should try to identify and fix the problem s before attempting to re import the file Note that file mis spec
139. n data quickly and easily However in order to use this program feature effectively you need to understand how the raster map creation process works ESAP generates all rater maps images through a two stage process The actual data interpolation is performed in the first stage by using the on screen interpolation controls in a systematic sequential manner Once the data interpolation is complete raster maps of any survey or prediction variable in the current data set can be quickly created from the Raster Map menu commands These menu driven map creation commands represent the second stage of the image generation process The remainder of this section describes the above mentioned two stage process in detail We recommend that you read the rest of this section and try completing the practice module 4 4 5 before you attempt to generate any 2D raster images 71 4 4 2 Using the Raster Map On Screen Interpolation Controls To invoke and display the Raster Map Initialization window click on the Graphics gt 2D Raster Image Map main menu option Once this window displays you can begin the data interpolation process using the on screen interpolation controls i e the command buttons contained on the left hand side of the Initialization window To interpolate your survey or prediction data you should sequentially perform the following 6 steps described below Step 1 Initialize Raster Window Click on this button to initialize t
140. n the Analysis gt Basic Statistics main menu options Normally the only reason you would need to recompute these statistics is if you decide to shift or transform your raw survey data The Signal Transformation Window allows you to either apply or remove a natural log transformation to from your input conductivity survey data and or change the data units mS m lt gt dS m However note that this window can only be activated if you have not already invoked the Signal Decorrelation Option There is a significant amount of contention within the soil salinity literature concerning the need or lack there of to log transform soil conductivity survey data It has been our experience that most conductivity survey data sets need to be log transformed We have seen some exceptions to this rule especially if the soil salinity levels are very low i e less than 1 0 to 2 0 dS m However in general we recommend that you apply the log transformation if your ultimate objective is the prediction of the spatial soil salinity pattern For other predictive purposes whether or not you choose to log transform your input conductivity data will depend primarily on the statistical distribution of conductivity readings You can generate histogram plots of your input data within the Interactive Graphics Window and we recommend that you use these histogram plots to determine if your conductivity data need a log transform If the histogram distribution s of your d
141. n the Plot gt Specify Map Variable mapping menu option and select the EMv variable from the drop down Map Variable combo box Accept the default raster cut off levels type in dS m in the Scale text box EMv Conductivity Survey Data in the Map Title text box and then click on the OK command button Now click on the Plot gt Create Map mapping option to create this new map and then click on the Plot gt Enlarge Map for Output gt Send to Raster Map Window 2 mapping option to enlarge the display Your second raster image should now look identical to Figure 4 3 EMh Conductivity Survey Data EMh dS m lt 651 651 75 75 848 gt 5458 Le Data Bounds x min amp max 18 288 Y min amp max 6 25 793 75 Figure 4 2 EMh raster map created in practice module 4 4 5 78 If you like you can again print this map or save it as a Windows bitmap graphics file Additionally you can print both of these maps on a single sheet of paper by clicking on the Plot gt Enlarge Map for Output gt Print both Maps mapping menu option Right now would be a good time to gain some additional practice by creating some more maps Try mapping some of the other variables and or changing the raster cut off levels and appearance settings Additionally you may wish to start the entire interpolation process over by closing down and then re invoking the Raster Map Initialization window and adjust the kernel size to see what effect this
142. n window 3 4 4 Practice Module Data Analysis In this practice module you will learn how to perform an iterative signal decorrelation and validation analysis on your soil conductivity data Note this practice session will make use of the previously imported bwd101p dat survey data file If you have not already done so you should go back and complete practice modules 3 2 3 and 3 3 4 before proceeding with this section Recall from figure 3 5 that the EM 38 survey data appears to be log Normally distributed Hence log transforming this data would be advantageous To do this click on the Analysis gt Basic Statistics menu option When the Signal Transformation window appears highlight i e check the Apply Natural Log Transformation option and change the signal labels from EMv and EMh to InEMv and InEMh Now click on the Compute command button to apply the log transformation At this point you would normally go back into the Interactive Graphics window and view the new InEMv and InEMh histograms If you like you can do this now Note that the new histograms appear to be much more bell shaped i e the log transformed data now appears to be nearly Normally distributed Once you have finished producing any desired graphics you should click on the Analysis gt Signal Decorrelation menu option This will invoke and display the Signal Decorrelation window To invoke the decorrelation algorithm click on the Perform Decorrelati
143. nal data validation In essence the signal validation routines are simply outlier removal routines To effectively use these routines you just need to remember that the centered scaled and decorrelated survey signal data is measured in standard deviation units std For example suppose you set the site outlier level within the Signal Decorrelation window to 5 0 Then what you are actually doing is telling ESAP RSSD to identify all survey Sites with conductivity readings more than 5 standard deviations away from the mean average level of the survey data In the ESAP RSSD program the default standard deviation levels for site masking and outlier detection are 3 5 and 4 5 std units respectively For most survey applications these values should be adequate Invoking the Signal Decorrelation Algorithms In practice you should use the signal decorrelation and validation routines in an iterative manner First you should open the Signal Decorrelation window by clicking on the Analysis gt Signal Decorrelation main menu options The decorrelation algorithm can be invoked this will automatically scale and decorrelate your input conductivity survey data If you have collected only one column of conductivity survey data then the decorrelation algorithm will simply center and scale this survey data If you have collected two columns or survey data i e two conductivity readings per site then the algorithm will perform a principal components
144. nd plot the Training2 ECe data follow the same directions as given for the Training data When you create these new plots they should look like the plots shown in figure 5 2 Raw Profile Data ECe Smoothed Fitted Data r 6233 dS m dS m 3 5 7 9 11 13 15 17 19 21 23 25 3 5 7 9 11 13 15 17 19 21 23 25 Figure 5 2 1D raw and smoothed ECe profile data plots generated from the sk13_lab97 pro sample salinity data Again this profile plot reveals quite a lot of information about the soil salinity conditions within this field Most of the individual profile shapes now appear to be either uniform or slightly inverted and the salinity levels throughout most of the profiles are quite high Hence this suggests rather poor management i e either insufficient water application or perhaps a drainage problem Whatever the reason it is clear that there is insufficient leaching occurring across much of this field Furthermore the correlation between the raw and smoothed salinity 112 data is now only 0 6233 suggesting that a the relationship is not that strong between the shape of the profile and the average magnitude of each sample core Although 1D profile plots are most useful for displaying and interpreting salinity data they can be created from any calibration sample variable within a profile data file For example figure 5 3 shows the raw and smoothed volumetric water content profile data associated with the Training2 project To r
145. ndard Correlation Analysis 5 8 all l gt DPPC Correlation Analysis 5 9 all gt Spatial MLR Analysis gt Deterministic Methods 5 5 all gt Conductivity to Salinity Help gt About ESAP Calibrate gt ESAP Interface Controls gt ReSet default Text Editor a OnLine Help gt What is ESAP Calibrate gt Navigating the Main Menu gt Frequently Asked Questions 14 2 4 ESAP 95 Supported Surveying Techniques and Applications The following surveying techniques and applications are supported by the ESAP 95 Software Package 1 salinity prediction from soil conductivity survey data using either stochastic or deterministic modeling techniques Calibrate along with the graphical display of the conductivity and or predicted salinity data SaltMapper 2 salinity monitoring strategies via repeated soil sampling where the sampling locations are optimally determined using the ESAP RSSD sample site selection algorithm RSSD Calibrate 3 the calculation of field average salinity estimates and range interval estimates i e the proportion of the field having salinity levels within a specific range interval for up to 6 separate sampling depths and or the calculation of the estimated relative crop yield loss due to the predicted salinity pattern for the surveyed field Calibrate 4 the prediction of secondary soil variables from acquired soil conductivity data via the stochastic calibration ap
146. ng are used to directly predict the soil salinity from conductivity survey data In this latter approach the models are dynamic i e the model parameters are estimated using soil sample data collected during the survey A unified deterministic model for describing the relationship between soil electrical conductivity and soil salinity was introduced in Rhoades et al 1989 This model which is now commonly referred to as the Dual Pathway Parallel Conductance DPPC equation described how soil salinity could be estimated from measurements of soil conductivity texture bulk density water content and temperature Additionally robust field measurement techniques were developed and described for acquiring the above conductivity and soil physical properties In Lesch et al 1995a b a comprehensive methodology was introduced for carrying out a field scale salinity survey using a stochastic dynamic modeling approach This methodology centered around the use of spatial regression models for predicting soil salinity from conductivity survey data when knowledge of the corresponding soil physical properties was either unavailable or too impractical to collect These models were shown to have a number of important advantages over other statistical modeling approaches including 1 they facilitated the use of rapid mobile conductivity surveying techniques 2 they could be estimated using a very limited number of soil samples 3 they could make
147. nitialization process you can set the color scale symbol plotting size and or request a gray scale grid rather than a color scale grid This latter option is useful if you wish to print a color scale grid plot out to a black and white printer When you request a color scale grid ESAP will automatically display a legend button to the right of the graph You can click on this button to display the color scale legend by default ESAP displays all color or gray scale grids using 3 program determined cut off values In addition to the s1 s2 z1 and z2 data you may produce a color scale grid of the calculated z1 transition standard deviations if a transition analysis has been perform on the decorrelated survey data see section 3 4 This color scale grid of the calculated transition 30 standard deviations can be used to display the spatial distribution of short range variability associated with your input conductivity survey data Additionally if you have already generated one or more soil sampling designs then the locations of these sample sites can also be overlaid on any of the grids described above Sample Site Maps Sample site maps can be produced from within the ESAP Graphics Window once one or more SRS sample designs have been generated and or a user defined sampling plan has been specified Sample sites for each requested sampling design are automatically labeled on these maps the labels are set equal to the site ID numbers Typ
148. ntly selected calibration model structure will be displayed Two additional model residual estimation options can be invoked before estimating your calibration equation First you can request that your calibration equation be fit to the bulk average response variable values in addition to each set of depth specific values Note if your calibration samples come from only one sampling depth then this option will be disabled This option is useful when you wish to automatically create estimates of bulk average response data such as the average salinity levels throughout the profile Second you can calculate Moran Residual Spatial Autocorrelation MRSA test statistics for each calibration equation These residual diagnostic test statistics can be used to help determine if the model residuals are spatially independent By default ESAP will not generate these test statistics In general we recommend that you only request these test statistics if you understand how they work After you have checked any desired options click on the Estimate command button to initiate the calibration model estimation algorithm When the estimation process is complete the MLR Model Estimation window will disappear and the basic calibration model summary statistics will be written to the lower left hand corner of the SCM program window 5 9 8 Practice Module 1 Specifying and Estimating a Salinity Calibration Equation In this practice session you will learn ho
149. o file and a soil sample data file used to create the profile data file These svy and pro data files can be imported into the ESAP Calibrate program and used to explore the various program modeling and analysis features refer to sections 5 2 and 5 3 of this manual for more details on how to import these files The format for each soil sample data file used to create the ESAP pro profile data files is listed below sk13_97 lab DPPC type format bwd_10296 lab DPPC type format column 1 site number column 1 site number column 2 sample depth m column 2 sample depth m column 3 ECe dS m column 3 ECe dS m column 4 SP column 4 SP column 5 water content grav column 5 water content grav column 6 SAR unitless column 6 bulk density g cm3 column 7 Boron ppm column 7 Clay 1 5 ESAP Software Package Development Information The ESAP 95 Software package has been developed for the Windows operating system 95 98 NT The 2 people responsible for developing ESAP 95 are Scott M Lesch Principal Statistician amp Lead Programmer Analyst statistical methodology software design and development James D Rhoades Research Soil Scientist amp past Laboratory Director soil theory amp methodology salinity measurement theory Additionally a number of other personal at the United States Salinity Laboratory have either directly or indirectly supported the development of this software
150. o initialize these graphics options At this point if you click once on the Graphics menu option within the ISG menu screen you will see that there are five types of graphical procedures now available for you to use These five procedures are Survey Grid Scatter Plots Line Plots Histograms and ColorScale Grids respectively You should try using each of these procedures now and explore the various options available in each graphical routine As practice you should try recreating figures 3 1 through 3 6 shown on the following pages using the directions listed below each figure You may also wish to review section 3 3 1 again before proceeding if you are unsure of how to interpret any of these graphical displays 35 EMv versus Y 1 93 1 30 gt o E Q s 793 75 400 00 click on the Graphics gt Scatter Plots gt s1 vs y menu options This plot displays all of the EM 38 vertical conductivity survey data plotted against their y axis locations The EMv versus Y axis coordinate locations To create this figure Figure 3 1 mean 1 158 and standard deviation 0 225 of the EMv data are also displayed as is the EMv y axis correlation 0 2993 36 EMv versus EMh 1 93 Corr sl s2 0 42 0 83 1 24 Figure 3 2 EMv versus EMh scatter plot To create this figure click on the Graphics gt Scatter Plots gt sl vs s2 menu options This plot displays the correlation between the two conductivi
151. o print and or save these plots and the Help menu if you wish to access any 2D prediction scatter plot help files Brief descriptions of the sub level menu items located beneath the 4 main program menu bar items are given below from within Plot Specify New Plot Variables Select this menu option to invoke and display the 2D Prediction Scatter Plot Initialization window which you can then use to initialize your specific scatter plot Create Plot Select this menu option to create and display the most recently initialized scatter plot from within Output Print Select the sub options listed under this menu item to print the most recently displayed observed versus predicted and or jack knife predicted scatter plots Save as Bitmap Select the sub options listed under this menu item to save the most recently displayed observed versus predicted and or jack knife predicted scatter plots as a bitmap file from within Help Navigating the 2D Prediction Scatter Plot Menu Select this option to display the Navigating the 2D Prediction Scatter Plot Menu help file This help file explains how to use the menu commands to create display print and or save the various types of 2D prediction scatter plots 149 General Tips how to plot a different depth Select this option to display the General Tips help file This help file explains how you can use the various interactive plotting features associated with most ESAP Calibrate graphic
152. of the BVWDrsd1 txt and BVWDinfo txt text files click on the File gt View Print Output Files main menu option This should cause the Windows File View window to be displayed now simply high light and open the desired file Which ever file you select will then be displayed in WordPad or what ever default text editor you are currently using You should then be able to print out copies of these text files using your text editor menu options etc If you have followed the instructions given in each of the preceding practice modules the hard copy of your BVWDinfo txt text file should match the information shown in table 3 1 Likewise your BVWDrsd1 txt text file should match the information shown in table 3 2 54 The final step you still need to perform is to create the BVWDdata svy processed conductivity survey file To do this simply click on the File gt Exit main menu option The Data File Export window should then appear and 1015 lines of conductivity data should automatically be exported to this file You can now safely exit the program by clicking on the Yes command button contained within the Confirm Program Exit frame Congratulations You have just finished processing your first conductivity survey data file using the ESAP RSSD software program 55 Table 3 1 Hard copy print out of the BVWDinfo txt text file Input File Output File Project O File Name Date amp Time Field Desc C US_Sal C US_Sal
153. of the design s created by ESAP to a more traditional sampling plan such as a simple random sampling design or stratified design etc In these situations you can generate the more traditional design using some other software package and then identify these Sites within ESAP using the manual sample site selection procedure This technique can be used to insure that none of the ESAP generated SRS designs overlap with the traditional design ESAP will not include any user specified sample Sites in any SRS generated designs The third scenario not discussed in the on line help is unusual but very important The ESAP RSSD program is designed to create and output your svy data file automatically once you are finished processing your survey data recall that this is the processed survey data file used by the ESAP Calibrate and ESAP SaltMapper programs However this automatic svy file creation process can only occur if at least one sampling design has been first created and saved In other words you can not create the svy output data file unless you create a sampling design This issue becomes important in the following situation Suppose you intend to collect conductivity data only and then use the deterministic calibration technique available in the ESAP Calibrate program to convert this conductivity information into calculated salinity levels Hence you would not be collecting any soil sample data in this particular survey process jus
154. oke and display the SES window click on the File gt Edit or Validate Profile Data main menu option If you are validating a DPPC data file as in this practice example you will see up to 3 data conversion check box options enabled and displayed within the SES window In this example two options are enabled 1 convert gravimetric H20 to estimated volumetric H20 and 2 estimate H20 relative to field capacity These conversions are optional but for this example you should go ahead and activate 1 e check both of these check boxes This will instruct the ESAP Calibrate program to replace your gravimetric water content data with estimated volumetric readings and also to produce a new column of estimated water content relative to field capacity Hz20 FC readings After you ve activated any applicable check boxes you simply need to click once on the Compute Summary Statistics command button If no boundary error conditions are detected then both a View Summary Statistics command button and a File Save Options frame will appear If one or more errors do occur the ESAP Calibrate program will notify you with a warning message and inform you to use the View Summary Statistics command button to display the error messages If the reported errors can be fixed by editing specific laboratory data readings then you should click on the Edit Data command button to invoke and display the Data Editing window see section 5 4 2 Note that during thi
155. ome accidently corrupted in this latter case you should use the ESAP RSSD program to re generate your input data file 5 2 2 Importing ESAP Calibrate Profile pro Data Files In the ESAP Calibrate program the term profile data is synonymous with calibration sample data In other words a profile data file is actually a file containing laboratory measurements made on your calibration soil samples such as soil salinity SP gravimetric or volumetric water content etc In practice this data file should contain at least some measurements which you wish to calibrate your soil conductivity survey data to Additionally the sampling locations of the analyzed soil samples should generally correspond to one or more sampling designs generated within the ESAP RSSD program You can use the Profile Data File Import Window to import either a previously saved profile data file or new calibration sample data stored in a suitably formatted ASCII text file This window can be invoked and displayed by clicking on the File gt Import Data File gt Import a Profile Data File main menu option To import a previously saved ESAP Calibrate profile data file a file having a pro extension you simply need to set the current project directory and then specify the path and filename of your input file To specify the current project click on the appropriate project name listed in the Set Current Project Directory list box Next make sure that the input file
156. on process was successful Likewise all red areas represent zones where interpolated values could not be generated i e zones too far removed from the survey data for the algorithm to generate any interpolated data Finally any areas that had been clipped will be displayed as white zones You should note that the final map images will not cover any red or white areas The ESAP SaltMapper program displays these colored areas within the interpolation window in order to help you decide if you are satisfied with the interpolation process 73 Step 6 Accept Interpolation Grid After the ESAP SaltMapper program color shades the interpolation window you must indicate if you are satisfied with the interpolation process If you are satisfied with this process then you should highlight the Yes option and click the OK button If you are not satisfied with the results then you should highlight the No option and click the OK button This will allow to start the process over again If you wish to repeat the interpolation process keep the following points in mind Excessive red zones typically indicate that you set the size of the kernel boundaries too small to correct this problem you will need to increase the horizontal and or vertical boundary sizes Likewise you can change the locations of any white zones by simply changing the clipping boundaries 4 4 3 Navigating the Raster Map Menu Bar Raster Map Menu Bar Layout The ESAP SaltMapper r
157. on command button Note that the ESAP RSSD program has found 5 Sites above the masking STD level currently set to 3 5 standard deviations and 2 of these Sites exceed 4 5 standard deviations these are flagged as outlier Sites Next click on the List all Sites above Outlier STD level command button You should now see the following information printed to the screen 45 Site ID X Coordinate Y Coordinate Sig 1 Sig 2 STD 897 270 00 768 75 0 119 0 389 4 74 988 288 00 368 75 0 224 0 494 4 86 These are your two outlier Sites To better determine where these Sites are located within the survey area and just how far removed from the rest of the survey data they are open up and display the Interactive Graphics window To do this first close out the Listing of Outlier Sites window and click on the Finished command button in the Signal Decorrelation window Then click on Graph gt Open Graphics Window to display the IG window Now click on the Graphics gt Survey Grid graph menu option This should produce a grid map of the survey site locations with the 2 outlier Sites colored red and the 3 other masked Sites colored yellow Note that the 2 outlier Sites occur on the far right hand side of the survey zone Next click on the Graphics gt Scatter Plots gt rsd plot graph menu option This should reproduce the response surface design rsd plot shown in figure 3 7 below Note that this plot clearly identifies how far removed the 2 outli
158. onductivity survey transect coordinate correlation coefficient r However note that the statistics calculated in the line plotting routines are based only on the currently displayed transect rather than the entire survey data file Also if you display multiple conductivity transects on the same graph and then request any signal statistics these statistics will only reflect the last transect shown on the graph 29 Histograms A histogram represents a useful way to display the statistical distribution of a set of data values In ESAP you should use the histogram plots to examine the input conductivity survey data distributions Typically if the conductivity survey data is approximately Normally distributed then the conductivity histograms will appear approximately bell shaped On the other hand if the conductivity survey data is log Normally distributed then the histograms will appear to be right skewed or if the signal data is bi modal then the histogram will also appear bi modal etc Histogram plots may sometimes also reveal data outliers although this is not their primary purpose In ESAP the main purpose of the histogram plots is to help you determine if your input conductivity survey data should be log transformed The histogram option is disabled when the Graphics Window first activates you need to enable i e activate it by opening up the GI window and initializing it Also when you request a histogram plot ESAP will also
159. onships between the salinity ECe soil texture SP amp Bd and soil water content pW influence the measurement of bulk soil electrical 121 conductivity ECa When the variation in soil salinity across the survey area is significant then equation 5 2 tends to be dominated by the ECe values However when the salinity variation is minimal other soil properties can significantly influence the conductivity readings Among other things equation 5 2 demonstrates why conductivity survey information can often be used in non saline areas to accurately map soil texture water content and or other physical soil properties If the soil salinity levels across the survey exceed about 1 5 dS m then equation 5 2 can be further simplified to Ts Tws ECs ECA as eee nee mee te mere Tw Tws ECw 5 3 since the product Ts ECw tends to be significantly larger than the product Tws ECs and thus the latter term can be ignored Equation 5 3 is often referred to as the linear version of the DPPC model since it expresses the relationship between conductivity ECa and salinity ECe in a linear manner A complete description of the theoretical development of the DPPC equation can be found in the following technical article Rhoades J D N A Manteghi P J Shouse and W J Alves 1989 Soil electrical conductivity and soil salinity new formulations and calibrations Soil Sci Soc Am J 53 433 439 Additionally a good rev
160. ontained within the Titles and Labels frame type in Transect 2 without the quotes and click once on the Plot Title command button Your plot should now have a title located in the upper left hand corner that says Transect 2 Next type in Distance down furrow meters in the text box and click on the X Label command button Then type in EM 38 dS m and click on the Y Label button You have now just added your X and Y axis labels and your displayed plot should look like the plot shown in Figure 4 1 For additional practice you may now wish to experiment with the 1D transect plotting features some more on your own Feel free to modify the look of your plot and or create one or more new transect plots of different line data Keep in mind that you can plot up to 4 rows of data on the same plot and that each row can represent a unique variable and transect pass You should also be able to print any transect plots you create using the Options gt Print line plot menu option 70 Transect 2 EM 38 dS m 1 4 1 2 1 0 0 84 100 200 300 400 500 600 Distance down furow meters Figure 4 1 1D transect plot created in practice module 4 3 5 4 4 Creating 2D Raster Maps The following section describes how to initialize create display and print a 2D raster map 4 4 1 About the Raster Map Creation Process The ESAP SaltMapper program has been designed to allow you to generate 2D raster images of your survey or predictio
161. or water content With this in mind you might now wish to examine the correlation structure between 1 ECe and SP 2 ECe and volumetric H20 and 3 ECe and H 2OIFC In particular note that the bulk average correlation estimates between salinity and water content are quite high 0 784 and 0 845 for the volumetric H2O and H2OlFC correlations respectively Figure 5 5 displays the bulk average ECe versus the bulk average H2OIFC data Recall that the 1D salinity profile plot from this field indicated that nearly all of the salinity profile 117 ECe dS m 26 boron ppm Figure 5 4 Overall correlation structure between boron and ECe sample data in sk13_lab97 pro data file shapes were either uniform or inverted and most of the salinity levels were also quite high This current salinity water content correlation plot indicates that the percent water content relative to field capacity increases as the bulk average salinity increases In the absence of an established crop this information strongly suggests that the elevated salinity conditions in this field are due at least in part to insufficient drainage ECe dS m 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 H20 FC No Point Selected Figure 5 5 Bulk average ECe v s H2OIFC correlation structure within the sk13_lab97 pro data 118 5 8 Performing a DPPC Correlation Analysis This section describes how to perform a DPPC correlation analysis on a DPPC type profil
162. original coordinate system 3 3 4 Practice Module Graphing Data In this practice session you will learn how to use the ESAP RSSD interactive graphics routines to display various features of your soil conductivity data Note this practice session will make use of the previously imported bwd101p dat survey data file If you have not already done so you should go back and complete practice module 3 2 3 before proceeding with this section From the ESAP RSSD main menu screen click on the Graph gt Open Graphics Window menu options This will cause the Interactive Survey Graphics ISG window to be displayed Now click on Options gt Initialize Graphics Components from within the ISG menu screen This in turn will cause the Graphics Initialization window to be displayed You can now use the options on this window to initialize the Line Histogram and ColorScale graphics options To initialize the above options click on the appropriate check boxes a graphics option will be initialized when the check box becomes checked and the color of the option description changes from black to blue For the Line plot you should also click on the North South transect option this tells the ESAP RSSD program to plot your conductivity transect lines against the y axis coordinates Next click on the change ColorScale to GrayScale check box and change the ColorScale symbol plotting size from 4 to 6 Finally click on the Initialize command button t
163. ormation to the main program screen of sample sites 15 of sample depths 4 of data columns 8 Furthermore you can display additional information about the profile data file by clicking on the Column Info command button This additional information will include the profile data file type individual column labels and units and the data file status 89 As should be obvious it is relatively simple to import previously saved survey or profile data into the ESAP Calibrate program All you really need to remember are the file names and where each file resides i e which project 5 3 Importing New Data Files This section describes how to import new calibration sample data into the ESAP Calibrate program Included here is a description of the two types of profile data files which can be created from calibration sample data and the formatting and column structure requirements of each input file type 5 3 1 The Two Types of Laboratory Profile Data In the ESAP Calibrate program calibration sample data can be imported as one of two profile data file types DPPC or Generic However in either case your calibration data must first be stored in a comma or space delimited ASCII text file Note also that you must create this input file using some other program such as a text editor or spreadsheet software before importing it into the ESAP Calibrate program The ESAP Calibrate program can not create this ASCII data file for you This
164. ors correspond to the magnitude of the data being plotted ColorScale grids can be produced for the Ist s1 or 2nd s2 conductivity survey readings the primary z1 or secondary z2 decorrelated principal component scores and the z1 transition standard deviation estimates again assuming this data is available If sample sites have been selected then these sites can also be overlaid on the grid Like the line and histogram plots the color scale plotting option must be initialized before any color scale plots can be produced Sample Site Map This option can be used to produce the final ESAP generated sample site maps for your survey data Note that this option only becomes available after you have generated one or more sampling designs from within Options Initialize Graphic Components This option should be used to open the Graphics Initialization window which can in turn be used to initialize the line histogram and color scale plotting routines discussed above Perform Coordinate Translation This option can be used to open the Coordinate Translation window which can in turn be used to adjust and or rotate you x y survey data coordinates from within Print Print Current Graph Any graph produced within the graphics window can be printed using this option Note that the print out will simply be a screen dump of the entire graphics window 33 from within Help OnLine Help Navigating the Graphics Menu This document
165. oss calculations These weights can be thought of as root distribution weights sense they allow you to specify which sampling depths should have the greatest effect on the crop yield For example suppose you have sampled down to 4 feet in 1 foot intervals and the field crop is cotton Then you might use a linear weighting scheme 0 4 0 3 0 2 0 1 to indicate that the salinity levels in the 1st foot should account for 40 of the overall yield loss calculations since you expect 40 of the root distribution to lie within the 1st foot of topsoil etc The Salt Tolerance window supplies two standard depth integration weighting options Equal weighting and Linear weighting Additionally you can highlight the User Defined Weighting option if you wish to specify your own set of depth integration weights Once you highlight this option a series of input weight text boxes will appear you should type your weight values directly into these text boxes Please note that all values must be greater than or equal to 0 and these weight values must sum to 1 Once you have finished initializing the various salt tolerance and depth weighting options discussed above click on the OK command button to return to the Calculate Field Summary Statistics window After you request the summary calculations your ASCH output text file should contain an estimated relative yield loss value in addition to the standard field average and range interval estimates This latter
166. our calibration sample data passes the data validation tests then you should save this data as an ESAP Calibrate profile pro data file using the command buttons contained in the File Save Options frame First click on the Specify Output Filename button to specify a unique name for your pro data file Once you have done this you should then click on the Save Data command button to save i e create a permanent copy of this profile data file In general we recommend that you always save any input calibration sample data you wish to analyze as a permanent pro data file Previously saved profile data files can be quickly imported into the ESAP Calibrate program because ESAP can automatically recognize their format structure see section 5 2 2 5 4 2 Editing Laboratory Data The Data Editing window can be used to edit any input calibration sample data values and or change the data column labels or column units To invoke and display this window click once on the Edit Data command button from within the Summarize Edit Save Profile Data window You can edit both your calibration sample data values and your data column labels or units if necessary You are free to change any or all of your column labels and or column units simply select the appropriate text box and type in the new label or unit When editing calibration sample data note that you can not change any site ID or sample depth data values However all other data values
167. our project you can then define your input data file type If you wish to import a DPPC file type then click once on the DPPC command button located inside the Define Input File Format frame Else click on the Generic command button to define a Generic file type If you click on the DPPC command button then the DPPC Profile Data File Format Sheet should appear You can then use the controls on this sheet 1 e contained within this window to specify the exact structure of your input file Note that you will need to specify the following information 92 1 the number of sample sites i e sample cores 2 the number of sampling depths associated with each site from 1 to 6 3 the number of secondary variables from 0 to 4 measured on each sample i e variables other that salinity texture water content and bulk density 4 the sampling depth units such as inches feet or meters etc 5 the salinity ECe units which should be either dS m or mS m 6 the type of soil texture SP or Clay and water content measurements gravimetric or volumetric 7 whether or not bulk density measurements have been acquired and 8 the column labels and measurement units for each secondary variable if any Once all of this information has been specified you can use the Review Format command button to review your input information and the OK command button to save this information If you click instead on the Generic com
168. ow is currently activated the OnLine Help button is the small button displaying a capital H usually located in the lower right hand corner of the window Brief descriptions of the sub level menu items located beneath the 3 main program menu bar items are given below from within File Import Data File Import a survey data file Select this option to invoke and display the Survey Data File Import window which can in turn be used to import a previously created ESAP svy survey data file Import Data File Import a profile data file Select this option to invoke and display the Profile Data File Import window which can in turn be used to import either a previously created ESAP profile pro data file or new calibration sample data stored in a suitably formatted ASCII text file Edit or Validate Profile Data Select this option to invoke and display the Summarize Edit Save Profile Data window The options contained within this window can be used to validate and save newly imported calibration sample data as an ESAP profile pro data file All calibration sample data must be validated i e examined by ESAP the first time it is imported into the program Additionally once you save your data as a pro data file you will never need to validate it again or specify the file structure etc View Print Project Output Files Select this option if you would like to view and or print any of the ESAP generated output text files
169. p boundary file name to display the File Open window you can then use the features in this window to specify the boundary file name Once you have done this select the Save clip boundary file submenu option the save the file You can import a previously saved boundary file during the raster map initialization process by using the Load a Clip Boundary File command button Refer to section 4 4 2 Using the On Screen Interpolation Controls for more information about this feature Specify Map Variable Select this menu option to open up the Map Variable Specification window which is where you can select the specific variable to map as well as control the visual appearance of the created image Create Map Select this menu option to create and display the map image within the survey interpolation window Enlarge Map for Output You can use this menu option to enlarge the raster image into a full screen raster map window You may then use the menu commands within this window to print or save the raster map Note that if you display two different raster maps in raster map windows 1 and 2 you can also print both of these maps on a single sheet of paper by selecting the Print both Maps menu sub option 75 Special Note In order to use the Print both Maps feature you must close down each raster map window using the windows Return menu option If you close down a raster map window using the upper right X close window
170. pears click on the Plot gt Specify New Plot Variables line plot menu option This will invoke and display the Line Plot Initialization window which you can then use to initialize your 1D line plotting options The first thing you must always do is specify which axis to plot your data by this can be done by highlighting either the X or Y coordinate location option in the X Axis Variable frame Then you should specify the data you wish to plot This is done by using the different drop down Variable list boxes selecting the variable to plot and then specifying the corresponding transect row number Finally you can use the options contained within the Line Color Line Style Show Symbols and Symbol Style frames to customize the look of the plot For this practice module highlight the Y coordinate location option Next click on the first Variable list box and select EMv then click on the second Variable list box and select EMh Now click once on each gt gt command button to the right of the Variable list boxes this should increment the Row values up to 2 in each row number text box Finally highlight the Same style for each line option in the Line Style frame and then click on the Initialize Plot command button After the Line Plot Display window disappears click on the Plot gt Create Graph line plot menu option A 1D line plot of the EMv and EMh signal data down the 2nd transect should now appear Next locate the text box c
171. presents the 2nd column of signal data Grid File site ID x cord y cord sl s2 Transect File site ID x cord y cord sl s2 row In both file types the site ID and s2 signal data columns are optional However both file types must honor the following data column restrictions Data Column Restrictions site ID distinct unique integer values only x y coordinates Cartesian coordinates only i e UTM coordinates are acceptable Lat Long are not sl s2 any real value bulk soil electrical conductivity measurement i e EM 38 EM 31 Mobile Wenner Martek SCT 10 or Verris 3100 data etc measured in either dS m or mS m conductivity units row_number distinct integer values only numbered sequentially from 1 to max_row_number Transect files only data formats free format is acceptable however the input data must be either comma or space delimited 24 Additionally both file types must also honor the following file size restrictions minimum of conductivity survey sites per file 50 sites maximum of conductivity survey sites per file 10 000 sites minimum of conductivity survey readings per site 1 maximum of conductivity survey readings per site 2 minimum of rows Transect files only 1 maximum of rows Transect files only 250 If you create a Transect input file you should also make sure that the file is properly sorted Transect files must be sorted by ascending row numbers lowest row numb
172. proach for precision farming applications Calibrate along with the graphical display of this prediction data SaltMapper 5 DPPC correlation modeling techniques Calibrate and 6 multiple graphical and statistical diagnostic techniques which are useful for examining and analyzing your acquired conductivity or soil sample data RSSD Calibrate Various examples of these techniques and applications are given in the practice modules located throughout sections 3 4 and 5 2 5 ESAP 95 On Line Help Documentation In addition to this user manual each ESAP software program contains a complete set of on line help documentation You should refer to this help documentation when you have specific questions concerning any program interface procedure not covered in this user manual i e how or when to use specific interface controls on any displayed program window All on line help documentation can be accessed from within each program at any point during your analysis If the window you have currently displayed contains a menu bar look for a Help menu option You should be able to access all of the relevant help files using this menu option 15 If the window does not contain a menu bar then you should look for a small square command button with a capital H label this button will usually be located in the lower right hand corner of the window After you locate this button click once on it to access and display the corresponding he
173. ptimal sampling designs for stochastic calibration models based on conductivity survey data The following capabilities listed below have been incorporated into the RSSD program a ability to process grid or transect data i e can be used to process EM 38 EM 31 Verris 3100 or Mobile 4 Electrode types of signal data b ability to handle arbitrarily large survey sizes up to 10 000 sites per field c allows for the interactive display and validation of signal data d ability to handle either 1 or 2 signal readings per survey site e can be used to generate calibration sample sizes of 6 12 or 20 sites per field or allow user to enter and record a custom sampling design f can adjust the sampling design based on signal variability i e a transition analysis ESAP Calibrate Conductivity to Salinity Calibration software Used to convert conductivity survey data to soil salinity via either stochastic calibration or deterministic techniques i e direct multiple linear regression estimation and or the Dual Pathway Parallel Conductance equation Additional capabilities currently include a ability to use stochastic calibration models to predict levels of secondary soil properties provided secondary sample data has been acquired b ability to use deterministic DPPC model to estimate theoretical strength of correlations between raw conductivity and salinity SP volumetric H20 and or other secondary sample data which may have been a
174. r example you can request that ESAP only consider standard types of parameter combinations or only models without any trend surface parameters etc Analyze All Possible Models Select this option if you wish ESAP to analyze all possible parameter combinations this is the default 144 Eliminate All Models Which Use x Location Trend Surface Components Select this option if you wish to eliminate all models which use any x location trend surface components Eliminate All Models Which Use y Location Trend Surface Components Select this option if you wish to eliminate all models which use any y location trend surface components Eliminate All Models Which Use 2nd Order Trend Surface Components Select this option if you wish to eliminate all models which use either quadratic x y or interaction xy trend surface components Eliminate All Models Which Use Any Trend Surface Components Select this option if you wish to eliminate all models which use any trend surface components Eliminate All Non Standard Type Models Select this option if you wish to limit the analysis to only those parameter combinations displayed in the Standard Model frame Click on the Invoke Auto Selection Algorithm command button when you are ready to invoke the auto selection algorithm Note that this algorithm may take some time to process all the various models possible up to a few minutes in some cases Once the auto selection algorithm ha
175. r frame set the soil temperature value to 21 0 In the signal factors frame select the EM 38 option enter 0 1 in the s1 and s2 text boxes and select the s1 EMh option 3 In the conversion formula frame select the Log Linear option 4 In the parameter values frame select the Rhoades Eqn option and then click on the 0 6 0 9m command button to define the parameter values depth prediction range and range midpoint Example 2 Suppose that conductivity survey data was collected using a single direct contact four electrode horizontal array system which measures the average bulk soil conductivity ECa in the 0 0 1 0 meter depth Note that since this system acquired ECa data directly we do not need to convert it per se but we must still adjust for temperature effects Suppose the average temperature within the 0 0 1 0 meter depth range was 18 5 degrees C Then the following information should be specified In the temperature factor frame set the soil temperature value to 18 5 In the signal factors frame select the Other option leave the MCF value set to 1 In the conversion formula frame select the Linear option In the parameter values frame select the Single Eqn option and then enter 0 in the a0 text box and 1 in the al text box this is equivalent to defining the following formula ECa 0 1 ECa which is the same as saying ECa ECa Poo 5 5 3 Setting the Secondary Soil Input Parameters You should use the Seconda
176. r right hand corner of the DCS window If desired you may then save these calculated salinity values as a permanent ESAP prd prediction file using the Specify Output Filename and Save File command buttons 5 5 4 Practice Module Converting Conductivity to Salinity This practice module will teach you how to use the deterministic conversion modules within the ESAP Calibrate program to process EM 38 survey data Before beginning this module you should import the BVWDdata svy survey data file from the Demo project Recall that this was the output survey data file created during the Chapter 3 practice sessions If you have not yet created this data file you will need to return to Chapter 3 and do so now If you have forgotten how to import a survey data file into the ESAP Calibrate follow the directions given in practice module 5 2 3 After the BVWDdata svy survey data file has been imported into the program click on the Calibrate gt Deterministic Methods gt Conductivity to Salinity main menu option to invoke and display the Deterministic Conversion window Once this window appears you may proceed to specify the conductivity and soil parameters You should use the following input conductivity and soil survey information during this practice session Conductivity Information Temp deg C 17 0 0 0 0 3m 18 0 0 3 0 6m 19 0 0 6 0 9m Height above ground EMv 0 05 meters EMh 0 10 meters Equation use appropriate Rhoades equa
177. r temporal changes in soil salinity using electromagnetic induction techniques Soil Sci Soc of Am J 62 232 242 Future releases of the ESAP 95 software package will support this technique however version 2 01 currently does not support this testing methodology Using the Net Flux window options You should use the options shown in the Data File Structure frame to define the number of monitoring sites contained in your monitoring data file as well as the specific sampling depths If desired monitoring data files can contain less sample depths than the original calibration data and or not be present for every calibration site However it is required that your monitoring data file have the following column format site_ID depth 1 depth 2 depth k where k represents the deepest monitoring depth the columns of monitoring data are arranged in order from shallowest to deepest and the site_ID column is the first column in the text file The data for each monitoring site i e each record can be either space or comma delimitated Note also that the ESAP Calibrate program assumes that your input monitoring soil variable is the same as the currently active calibration variable in the MLR module You should use the Specify Input File Name and Import Data command buttons located within the Data File Input frame to import your monitoring data file Additionally you can click on the View Log File command button if you would like to view a l
178. raster map s To begin the map creation process click on the Plot gt Specify Map Variable mapping menu option This will invoke and display the Map Variable Specification window which you can then use to select your mapping variable and control the final raster map appearance Select the EMh variable by clicking on the drop down Map Variable list box and for now simply accept the default raster cut off levels Next type in dS m in the Scale text box and EMh Conductivity Survey Data in the Map Title text box Finally make sure that the display map as a GrayScale image appearance option is highlighted and then click on the OK command button 77 To display the map you ve just specified click on the Plot gt Create Map mapping menu option You should now see the interpolated raster map appear within the ESAP SaltMapper interpolation grid frame To create an enlarged printable version of this map click on the Plot gt Enlarge Map for Output gt Send to Raster Map Window 1 mapping menu option This should create a raster image which looks identical to Figure 4 2 If desired you can now print this map or save it as a Windows bitmap graphics file Now lets create a second map First use the Return gt Return the Raster Map Initialization Window display menu option to close Output Window 1 Note do NOT click on the upper right hand X close window button to close this window this will erase the map from memory Next click o
179. re the only two file types that the ESAP SaltMapper program can operate on 59 4 1 2 Navigating the ESAP SaltMapper Main Menu Bar Main Menu Bar Layout The ESAP SaltMapper Main menu bar is located in the upper left corner of the main program Window The full layout for this menu bar system is shown below Main level Sub Level 2 Sub Level 3 File gt Specify Project Input File Info gt Column Manipulation gt Change Column Labels gt Create a New Column gt Column Statistics gt Create Output Data File gt Exit Graphics gt ID Line Transect Plot gt 2D Raster Image Map Help gt About ESAP SaltMapper gt ESAP Interface Controls gt ReSet default Text Editor gt OnLine Help gt What is ESAP SaltMapper gt Navigating the Main Menu gt Frequently Asked Questions FAQ Main Menu Bar Menu Item Descriptions The main level contains 3 menu bar items File Graphics and Help You should use the File menu to access all the data input output and column manipulation routines the Graphics menu item to access either the 1D transect plotting routines or 2D raster mapping routines and the Help menu to access the main ESAP program help files Additional help file documentation is available for nearly all ESAP program routines this documentation can be accessed by double clicking on the OnLine Help file button within whatever window is currently activated the OnLine Help button is the small bu
180. reasonable calibration models However you are free to choose any parameter combination associated with an active standard model option button Custom Model frame The Custom Model Definition frame will be displayed if you select the custom model identification method This frame contains two sub frames You should use the option buttons displayed in the Signal Parameters sub frame to select an appropriate combination of signal parameters Likewise you can use the option buttons displayed in the Trend Surface Parameters sub frame to select a specific combination of trend surface parameters Your final model will then contain both sets of selected parameters i e signal trend Note that if you have acquired 2 signal readings per site and you have collected samples from at least 11 calibration sites then you can potentially create a total of 36 separate regression model parameter combinations Automatic Model Identification frame The Automatic Model Identification frame will be displayed if you request the auto selection model identification method The options within this frame may be used to control the types of parameter combinations which are analyzed as well as to invoke the auto selection procedure and display the results Analysis Options Unrestricted v s Restricted Before you invoke the auto selection algorithm you have the option of restricting the possible parameters to specific parameter combination subsets Fo
181. reate and save text files which describe the generated sampling design and list the selected sample site locations Likewise the ESAP Calibrate program can generate a number of different output text files which document the various analyses performed on the conductivity survey and or profile sample data More detailed information about these additional output file features can be found in chapters 3 4 and 5 11 2 3 An Overview of the 3 Main ESAP Program Menus The following section shows the main menu layouts for each ESAP program along with references to the appropriate manual sub sections which describe each menu option in detail ESAP RSSD Main Menu Layout Refer to section 3 1 for a general program overview Refer to the manual sub sections listed below for specific menu option details Manual Sub section Main level Sub Level 2 Sub Level 3 3 2 1 File gt Set Create Project and Field ID 3 22 2 Import Survey Data File PN Import a Grid Survey File l gt Import a Transect Survey File 3 6 1 gt View Print Output Files 3 6 3 us Exit 3 3 all Graph gt Open Graphics Window 3 4 1 Analysis gt Basic Statistics 3 4 2 a Signal Decorrelation 3 4 2 3 4 3 Signal Transformation 3 5 2 Design gt Calculate SRS Sample Design 3 5 1 a Manual Sample Site Selection Help gt About ESAP RSSD gt ESAP Interface Controls l l gt ReSet default Text Editor l 12 gt OnLine Help gt Wha
182. redictor variables in the regression equation Additionally different fields may require different numbers of signal or trend surface parameters and even if the form of the regression equation is the same for two different fields the actual parameter estimates typically will not be Variation in the model parameter estimates should be expected this variation arises naturally due to changing survey conditions i e due to differences between fields in soil temperature texture water content crop cover bed furrow shape etc In particular the stochastic calibration modeling procedures contained within the ESAP Calibrate program are designed to help you 1 determine the best regression model parameter form and 2 optimally estimate these model parameters Once this regression model has been estimated you can then use the stochastic calibration prediction procedures to predict the levels of your soil variable across the entire field or survey area estimate the average field level and calculate a number of additional useful prediction statistics You can also generate and save an output file of your predictions which can then be imported into the ESAP SaltMapper program and used to generate a spatial prediction map Furthermore all of the modeling and prediction techniques described above can be used to generate predictions of your soil variable across multiple sampling depths provided you acquire multiple depth calibration sample data
183. rmation consists of 1 the average soil temperature 2 conductivity instrument signal correction factors if applicable and 3 the conductivity to ECa conversion formula type and parameter estimates Set Soil Parameters Select this option to define the values of the soil parameters for use in the DPPC 100 equation i e to define the soil texture bulk density and water content level from within Calculate Convert Conductivity to Salinity Select this option to convert your conductivity survey data into estimated soil salinity data from within Help How does a Deterministic Conversion work Select this option to display the DCS explanation help file This help file describes how the deterministic conversion routine works as well the DPPC equation used within the conversion routine Navigating the DCS Menu Select this option to display the DCS Menu help file This help file explains how to use the DCS menu commands to covert conductivity to salinity from within Exit Return to Main Menu Select this option to close down the DCS Window and return the ESAP Calibrate main program window 5 5 3 Setting the Conductivity to ECa Input Parameters You should use the Conductivity to ECa Parameters window to define and or input your deterministic conductivity ECa conversion information This information consists of 1 the average soil temperature 2 conductivity instrument signal correction factors if applicable
184. rofile analysis on your acquired calibration sample data 5 6 1 What is a 1D Profile Plot Four types of stochastic data analysis techniques are available in the ESAP Calibrate program the first of these techniques being the 1D Profile Plot A 1D profile plot is a graphical representation of your calibration sample data where the magnitude of a specific soil sample variable is plotted against the sampling depth and data points from specific soil cores are joined connected together by lines In a profile plot the sample variable is plotted along the x axis and the sampling depth is plotted along the y axis with the shallowest depth shown at the top of the y axis and the deepest depth shown at the bottom Such a plot can be used to create a visual image of the shape of your calibration sample data with respect to the sampling depth The interpretation of this sort of shape data is typically referred to as a 1D profile analysis For soil salinity sample data a 1D profile plot can be used to display the shape of the salinity profile with respect to the overall magnitude of the average salinity level for each sample core This information can in turn be used to infer the direction of water flow through the soil profile across your survey area regular shaped salinity profiles imply downward water flow while inverted shaped profiles typically imply that a net upward water flux is occurring often from a shallow water table Likewise uniform prof
185. roject directory If you would like to make a new project then select the create a new Project option type in a valid Project name and then click on the Create command button If you wish to use an existing project then select the set the current Project option and click on the desired project name in the Set Project display window Note use the Training2 project for this practice module if you decide not to create a new project Once your project has been identified you next need to define the input file format If you intend to import a DPPC type data file then click once on the DPPC command button otherwise click on the Generic command button to import a generic type data file Lets begin by importing the sk13_97 lab data file as a generic file so click once on the Generic button This should invoke and display the Generic Profile Data File format sheet which should be filled in with the following information Number of sample sites 15 Number of sampling depths per site 4 Total number of input columns 7 Column Units 2 meters Column Label Units 3 ECe dS m 4 SP 5 grav H2O 6 SAR none or leave blank 7 Boron ppm 94 After you have entered this information click once on the format sheet OK command button to return to the Profile Data File Import window You should now identify your input ASCII data file by clicking on the Browse command button you will find the sk13_97 lab data file in the Tra
186. rt ESAP RSSD svy or ESAP Calibrate prd generated data files into the ESAP SaltMapper program and how to manipulate this data once it has been imported 4 2 1 Setting the Current Project and Importing Data You should use the Project and Input File window to import your svy or prd data file into the ESAP SaltMapper program This window can be displayed by clicking on the File gt Specify Project Input File Info main program menu options In order to successfully import your data file you must perform the following three steps 62 First you must set the current project i e you must identify which project your data file resides in This can be done by clicking on the appropriate project name listed in the Set Current Project list box Second you must specify your data file type either svy or prd using one of the appropriate Input File type option buttons Finally you must specify the path and filename of your input file This can be done by typing in the appropriate information into the Path Filename input text box or by using the Browse button to locate your input file If you wish you can also request that any natural log transformed data columns be back transformed using the check box option under Data Scale If you request this option and your data has not been log transformed then this request will be ignored Please note that the ESAP SaltMapper program is only designed to read in ESAP RSSD generated svy or
187. rt your survey data file For this practice session you will use one of the pre supplied demonstration input files bwd101p dat already installed on your computer First click once on the s2 2nd signal column check box this tells the ESAP RSSD program that the input file contains two columns of conductivity survey data Next change the conductivity units from mS m to dS m by clicking on the dS m check box the conductivity readings in this data file are in dS m units Finally click once on the Browse command button which will open up the standard Windows file browser window click on the bwd101p dat data file and then click on the Open command button You have now just told the ESAP RSSD program the name of your input conductivity data file and where it resides on your computer Now click once on the OK command button The ESAP RSSD program will now scan and import all 1017 lines of conductivity survey data from the bwd101p dat data file and then display some useful file summary information within the Project Status window This information will include 1 the full path of the input file 2 the file survey type 3 the number of survey sites 4 the number of signal readings per survey site 5 the signal transformation if any and 6 some basic signal statistics in addition to the project information listed previously Assuming that you ve followed the directions given in this practice session you should see the following information
188. ry Soil Parameters window to define the values of the soil parameters for use in the DPPC equation 1 e to define the soil texture bulk density and water content level These values must be specified a priori in order to convert soil conductivity in soil salinity You can specify the field average soil texture using the horizontal scroll bar s contained in the Soil Texture frame You can specify the average soil texture in either SP or clay units You can also specify the field average bulk density using the controls contained in the Soil Texture frame Note that you can either specify the average bulk density directly using the bulk 103 density scroll bar or estimate the bulk density value from the average SP or clay data You should specify the average water content relative to field capacity using the horizontal scroll bar contained in the Water Content frame Note The default acceptable minimum and acceptable maximum values for each of the soil parameters discussed above are as follows Parameter default minimum maximum SP 60 25 105 clay 43 0 100 bulk density 1 33 0 95 1 95 H201 FC 75 30 100 Once your soil input values have been saved you should click on the Calculate gt Convert Conductivity to Salinity option on the main DCS menu to perform the conversion The ESAP Calibrate program will then convert each survey point into a calculated salinity value and display the salinity summary statistics in the lowe
189. s In most situations you should not need to adjust the advanced design options However if your generated SRS design does not produce an optimization criteria value less than 1 30 then we recommend that you change the design factor and re run the SRS algorithm Try setting the design factor to values of 0 90 0 95 1 05 and or 1 10 re generate the sampling design and see if any of these new designs produce optimization criteria values less than 1 30 Note that lowering the design factor shrinks the response surface design levels resulting in a more conservative design Likewise raising the design factor causes the response surface design levels to expand resulting in a more aggressive design In general we recommend that you try lowering the design factor first If changing the design factor does not solve the site clustering problem i e the Opt Cri value is still gt 1 30 then you should invoke the parsing option Data parsing can be employed on any conductivity survey data set with more than 400 survey Sites data sets with only a few hundred observations rarely exhibit poor optimization criteria values When you invoke the parsing option ESAP will separate your survey data into 2 or more distinct strata the number of strata ESAP uses will increase as your sample size increases You should then select one of these strata to use from within the Data Parsing Stratification Window invoke the SRS algorithm and see if the Opt Cri va
190. s However we recommend that you first try to locate and shut down the background program which is causing this error Assuming that you use the default installation location the sub directory structure shown on the next page will exist on your computer once the installation process finishes C US_Salinity_Lab esap2 l bitmap l data Training 1 Training2 demo_input_files helpdocs All ESAP 95 software programs will reside in the esap2 directory Additionally the bitmap data demo_input_files and helpdocs sub directories located off the esap2 directory are used by the software programs for storing and or retrieving various data files In particular the helpdocs sub directory contains all of the on line help documentation the bitmap sub directory can be used to store any graphical bitmap files created by any ESAP 95 graphical procedures created in either the ESAP Calibrate or ESAP SaltMapper programs and the demo_input_files sub directory contains some demonstration input data files for use with the ESAP RSSD program Likewise the data sub directory is used by each program for the storage and or retrieval of all survey project data files In ESAP 95 different survey projects can be created to hold different sets of survey data these projects are actually just sub directories created off the data directory The Training and Training2 sub directories are simply two project dire
191. s a DPPC profile data correlation analysis Description of the DPPC model Using the DPPC PDCA window Understanding the DPPC summary report DPPC data plots Navigating the DPPC graphics menu Practice module 5 9 Spatial Regression modeling Stochastic Calibration 5 9 1 5 9 2 5 9 3 5 9 4 5 9 5 5 9 6 5 9 7 5 9 8 5 9 9 5 9 10 5 9 11 5 9 12 5 9 13 5 9 14 5 9 15 6 0 References What is a stochastic calibration model Navigating the SCM menu Response variable specification window Navigating the PRV graphics menu Specifying the calibration model parameters Understanding the model parameter options Estimating the calibration model Practice module 1 Viewing 2D prediction scatter plots Calculating field summary statistics Using the salt tolerance window Saving your output predictions Practice module 2 Advanced options Net flux testing Advanced options Advanced modeling information viii 113 114 114 117 119 119 120 122 124 128 129 131 133 133 134 137 139 140 142 146 146 148 150 151 152 153 156 158 160 1 0 Introduction 1 1 General Conductivity Salinity Modeling and Assessment Techniques Accurate soil salinity assessment is needed for the design of efficient agricultural management practices and irrigation water allocation strategies Fortunately the ability to diagnose and monitor field scale salinity conditions has been significantly improved through the use of both four el
192. s any SCM help files Brief descriptions of the sub level menu items located beneath the 5 main program menu bar items are given below from within Model Select Response Variable Select this option to identify your stochastic calibration model response variable 1 e the soil variable you wish to predict When you select this option ESAP will display the Response Variable Specification window which can in turn be used to 1 merge your survey and profile data files together 2 identify your response variable and 3 create preliminary plots of the response variable versus the conductivity instrument readings Identify Model Parameters Select this option to identify the form of your stochastic calibration regression equation When you select this option ESAP will display the MLR Model Identification window which can be used to either manually or automatically identify an optimal regression equation Estimate Calibration Equation When you select this option ESAP will display the MLR Model Estimation window which you can use to estimate your specified stochastic calibration regression equation Advanced Options At any time after ESAP estimates a regression equation you can access a full set of model and residual error diagnostics using the sub menu items located under the Advanced Options menu option The information which can be displayed includes a complete set of model summary statistics residual error summary diagnost
193. s command button to calculate the requested net flux test statistics Once the calculations are completed you can use the View Test Results command button to view the calculated net flux test statistics As with most other ESAP Calibrate procedures the command buttons contained within the Save Results to ASCII Text File frame may be used to save the net flux testing information to a permanent ASCII text file You should use the Specify Output Filename command button to specify the name of your permanent output ASCII text file and then the Save File button to save the ASCII text file 5 9 15 Advanced Options Advanced Modeling Information If you are familiar with the statistical details behind regression modeling you can use the ES AP Calibrate program to access three different types of advanced model summary displays Each of these displays are briefly described below Advanced Model Summary Statistics If you wish to view any advanced regression model summary statistics click on the Model gt Advanced Modeling Options gt View Model Summary Statistics SCM menu option The ESAP Calibrate program will then display an MLR Analysis and Statistics summary sheet which will include an analysis of variance AOV table and a listing of the parameter estimates for each estimation depth This summary sheet will be displayed using your default text editor and you will also have the option of saving this summary sheet as a permanent ASCII text file
194. s finished you can click on the View Results command button to view a tabular listing of all the PRESS score results These results will include the value of the minimum PRESS score each models relative PRESS score ratio i e that models PRESS score divided by the minimum PRESS score and the ranking of each analyzed model with respect to the minimum PRESS score After the auto selection algorithm has finished you can choose to either accept or reject the auto selected model Typically you should view the auto selection results before making this decision If you accept the selection ESAP will record the parameter combination which produced the minimum PRESS score close down the MLR Model Identification window and then return to the Stochastic Calibration Modeling window If you reject this selection then the Automatic Model Identification frame will disappear but the Model Identification window will still remain active You will then need to select a different parameter combination using one of the other model identification options 145 5 9 7 Estimating the Calibration Model You should invoke the MLR Model Estimation window once you are ready to estimate your stochastic calibration equation i e after you have selected your response variable and identified your model parameters This window can be invoked and displayed by clicking on the Model gt Estimate Calibration Equation SCM menu option After you invoke this window the curre
195. s of data to your output file When the export process is finished you should see the following message Output ASCII file saved to C US_Salinity_Lab esap2 bvwd_out txt This message is reminding you where your output data file is on your hard drive and confirming that the export process was completed successfully 80 5 0 ESAP Calibrate Software Program 5 1 ESAP Calibrate Program Overview The following section gives an overview of the ESAP Calibrate software program including a review of how the program functions and a description of the main menu bar layout 5 1 1 Program Description What is ESAP Calibrate ESAP Calibrate is a comprehensive data analysis program which can be used to convert raw conductivity data to soil salinity data via either stochastic calibration or deterministic techniques i e by using an empirically fit regression model or the DPPC Dual Pathway Parallel Conductance equation a k a the Rhoades conductivity model Additional capabilities of the ESAP Calibrate program currently include a the ability to use stochastic calibration models to predict levels of secondary soil properties provided secondary sample data has been acquired b the ability to use the deterministic DPPC model to estimate the theoretical strength of correlation s between raw conductivity and ECe SP volumetric water content and or other secondary sample data that may have been acquired during the sampling process c the ab
196. s practice session no errors should occur when you compute the summary statistics on the sk13_97 lab data file We strongly recommend that you make it a practice to save all new calibration sample data files as permanent pro profile files You should use the Specify Output Filename command button to specify your permanent output file name and the Save File command button to save the file As you have already learned previously saved profile data files can be imported back into the ESAP Calibrate program much more quickly than new calibration sample data files during any future analysis sessions You should also print out the summary statistics report if 98 you want to retain a hard copy record of when your calibration data was first imported into the ESAP Calibrate program 5 5 ESAP Calibrate Deterministic Salinity Conversion Modeling The ESAP Calibrate program allows you to covert conductivity to salinity via either deterministic or stochastic modeling techniques This section explains how the deterministic conversion routines work and what sort of survey information you need to collect in addition to your soil conductivity data in order to achieve the highest possible conversion accuracy 5 5 1 How Does a Deterministic Conversion Work In the ESAP Calibrate program soil conductivity can be converted into estimated soil salinity data using a deterministic conversion algorithm This algorithm converts conductivity data into salinity
197. s set to Yes Now click on the Accept Configuration command button You should then see the following two messages printed to the screen Note 217 Sites were masked by the edge buffering algorithm Data file information saved to BVWDinfo txt These messages are telling you that 217 Sites around the edge of the survey zone have been masked 1 e removed as potential sample site locations and that all of the information currently displayed in the main ESAP RSSD window has just been saved to a text file called BVWDinfo txt They also indicate that you have now finished your interactive signal decorrelation and validation analysis and prepared your conductivity survey data for the ESAP spatial response surface SRS sampling design algorithm 3 5 Generating Sampling Designs This section reviews the two ESAP RSSD procedures for generating sampling designs and describes when and how to use each procedure 3 5 1 The Manual Sample Site Selection Procedure The manual sample site selection procedure can be used to manually create a custom e user specified sampling design This procedure should be used if you wish to specify your own set of sample Sites To display the Manual Sample Site Selection window which can be used to invoke this manual selection procedure click on the Design gt Manual Sample Site Selection menu option 47 The ESAP RSSD on line help documentation states that there are normally just two scenarios where you would
198. section some of these represent advanced procedures which the beginning user may wish to skip over at first Note all sub sections covering advanced material are clearly identified 5 9 1 What is a Stochastic Calibration Model In the ESAP Calibrate program a stochastic calibration model is a spatially referenced multiple linear regression model Typically such a regression model would include both conductivity and trend surface parameters For example suppose that we wish to predict the log soil salinity levels within a field from log transformed EM38 conductivity survey readings acquired across the field Then we might wish to use the following regression model In ECe bo b In EMh b2 In EMv b3 x baLy 5 4 where EMh and EMv represent the EM 38 survey readings and x and y represent the spatial coordinate locations of the EM 38 survey data Hence the purpose of the regression fitting procedure would be to optimally estimate the regression parameters i e to estimate the values of bo bi b2 b3 and b4 Once the parameter estimates were acquired we could then predict the log salinity levels at every survey site as well as estimate the accuracy associated with any or all of these site predictions The regression equation shown above represents only one hypothetical example In practice transformed and decorrelated signal data i e the principal component scores are often used in place of the raw signal readings as p
199. sest neighbors This transition variability information can then be optionally incorporated into the SRS sampling algorithm in order to choose Sites with low variation In practice there is really only one situation when you might wish to incorporate transition variability information into the design e g if you are using a low grade GPS unit to relocate the sample Sites Low grade GPS units typically can not achieve sub meter accuracy Hence when using such units you would naturally want to choose regions within your survey area which do not exhibit large conductivity variations over short distances In such a scenario a transition analysis may help you avoid degrading your calibration equation i e prediction model by minimizing the negative effects of any relocation errors 44 You may also perform a transition analysis if you wish to visually detect any zones within your survey area having rapidly changing conductivity levels Transition variability information can be displayed graphically using the Interactive Graphics routines discussed earlier If you perform a transition analysis solely for this purpose you should make sure that you tell the ESAP RSSD program not to use the transition variability information when generating any sampling designs see section 3 5 Additional information about both the edge buffering and transition analysis options can be found within the on line help documentation associated with the Signal Validatio
200. sh to create an output svy data file Note also that unless you have generated and saved at least on sampling design either an SRS or user generated sample design the ESAP RSSD program will NOT generate an output svy data file 3 6 4 Practice Module Viewing and Printing Output Data This is the final practice module for Chapter 3 If you have performed all of the previous modules you should be able to re create the output data shown on the next few pages In most survey situations there will be three items files or graphs that you should always make hard copy prints of 1 your sample site map 2 your rsd sample site documentation information file and 3 your general information log file To produce a plot of the sample site locations generated in practice module 3 5 3 invoke and display the Interactive Graphics window Next click on the Graphics gt Sample Site Map graph menu option to display the Sample Site Labeling window High light the RSD sample Sites from Design 1 check box 53 option and then click on the Create Map command button This should produce a map like the one shown in figure 3 8 below Now click on the Print gt Print Current Graph menu option to print a copy of this map and then exit out of the Interactive Graphics window Design 1 ESAP Sample Site Locations 833 13 400 00 33 13 280 13 153 00 586 13 Figure 3 8 Response surface sampling design map for Design 1 To print out copies
201. sh to print the currently displayed line plot and or save this plot as a bitmap file Likewise you should use the Help menu if you wish to access any 1D line plot help files Brief descriptions of the sub level menu items located beneath the 4 main program menu bar items are given below 67 from within Plot Specify New Plot Variables Select this menu option to open up the 1D Line Plot Initialization window You will use the controls within this window to initialize each line plot you wish to create Create Plot Select this menu option to create and display your 1D line plot from within Output Print Select this menu option to print the currently displayed line plot Save as Bitmap Select this menu option to open up the File Save window You can then use this window to specify the path and name of your bitmap file and then save this file from within Help Navigating the 1D Line Plot Menu Select this option to display the Navigating the 1D Line Plot Menu help file ID Line Plotting Tips Select this option to display the Line Plotting Tips help file This help file describes how you can add figure titles labels and or a figure legend to your 1D line plot from within Exit Return to Main Menu Select this option to close down the 1D Line Plot Display window and return to the ESAP SaltMapper main menu 4 3 3 Line Plot Initializing Options You should use the 1D Line Plot Initialization window to identify the speci
202. sponse Variable PRV Graphics Menu PRV Graphics Menu Bar Layout The ESAP Calibrate PRV Graphics menu bar is located in the upper left corner of the PRV window The full layout for this menu bar system is shown below Main level Sub Level 2 Plot gt Response Variable v s z1 primary decorrelated signal data gt Bivariate response variable plots Output gt Print gt Save as Bitmap Help gt Navigating the PRV Graphics Menu gt General Tips how to highlight sites plot a different depth etc Exit gt Return to Response Variable Specification Window PRV Graphics Menu Bar Menu Item Descriptions The profile menu contains 4 menu bar items Plot Output Help and Exit You can use the Plot menu to select the type of plots to create and display the Output menu to print and or save your plot s and the Help menu if you wish to access any PRV Graphic help files Brief descriptions of the sub level menu items located beneath the 4 main program menu bar items are given below from within Plot Response variable v s zl primary decorrelated signal data Select this option to create and display plots of your selected response variable against the primary z1 principal component score by depth These plots are useful for determining if your response variable data responds to your conductivity signal data in a linear manner and also for determining the strength of correlation between the response variabl
203. ss multiple sampling depths up to 4 depths may be displayed simultaneously If you choose this option then a second input information frame will appear 128 within the Graph Options window You should use the list box displayed within this frame to select the secondary variable you wish to plot the ECa against You should also use the controls contained within this frame to indicate if either the calculated ECa or the secondary soil variable should be log transformed as well as to specify the secondary variable sampling depths you wish to display Note that this last type of panel plot can be used to determine if the calculated bulk average ECa correlates well to depth specific soil variable information such as depth specific soil ECe data etc 5 8 6 Navigating the DPPC Graphics Menu DPPC Graphics Menu Bar Layout The ESAP Calibrate DPPC Graphics menu bar is located in the upper left corner of the DPPC Graphics program window The full layout for this menu bar system is shown below Main level Sub Level 2 Sub Level 3 Output gt Print Current Graph s gt Save as Bitmap gt Save Ist Plot gt Save 2nd Plot gt Save 3rd Plot gt Save 4th Plot Help gt Navigating the DPPC Graphics Menu gt General Tips how to highlight sites plot a different depth etc Exit gt Return to DPPC Graph Options Menu 129 DPPC Graphics Menu Bar Menu Item Descriptions The profile menu contains 3 menu bar items Output
204. status is set to previously saved Profile data file the file status frame should default to this option Then type in the appropriate information into the Path Filename input text box or use the Browse button to locate your input file Once you have performed these steps click on the OK button to import your pro data file If a file import error message occurs when you attempt to read in a previously saved profile data file then this pro data file has become corrupted In such a scenario you will need to re import the calibration sample data associated with this file and re save it Note previously saved pro files can be corrupted if you attempt to edit these data files outside of the ESAP Calibrate program 88 5 2 3 Practice Module Importing ESAP Data Files If you have not already done so start up the main ESAP 95 splash screen by clicking on the ESAP 95 prompt listed within the Programs drop down menu system i e click on Start gt Programs gt Esap95 Once the splash screen has displayed click once on the ESAP Calibrate command button to invoke the Calibrate program Once the ESAP Calibrate main menu screen is fully displayed click on the File gt Import Data File gt Import a Survey Data File main menu option This will cause the Survey Data File Import window to be displayed You will be using the data stored in the Training2 project during this practice module so click once on the Training2 project name and mak
205. stimate is 0 551 and the In ECe SP estimate is 0 084 II Bulk ave ln Calc ECa v s bulk ave Pri Sec Soil Variable Correlations Column Labels QO In Carce ECa 1 I1n ECe 2 SP 3 Vol H20 4 Bulk Den 0 1 2 3 4 0 1 000 0 864 0 551 Op593 06377 1 0 864 1 000 0 084 0 132 0 043 2 0 551 0 084 1 000 0 920 0 896 3 0 593 0 132 0 920 1 000 0 824 4 0 377 0 043 0 896 0 824 1 000 125 MI Bulk average and depth specific ECac v s depth specific Primary Secondary soil variable Correlations This section displays the correlations between the depth specific soil profile data versus both the bulk average and depth specific ECac data In general depth specific profile data will correlate better with depth specific ECac readings as opposed to the bulk average ECac readings The differences between these two sets of correlation estimates represent the degree of improvement in the prediction accuracies you would achieve if you could acquire depth specific conductivity data as opposed to conductivity data averaged across the entire soil profile Example correlations for salinity In ECe and texture SP are shown below In particular note that the prediction of the SP data in the 0 0 0 3 m sample depth could be greatly improved using depth specific conductivity information i e the correlation jumps from 0 267 to 0 948 LILE Bulk ave amp depth specific ln Calc ECa v s depth specific Pri Sec Soil Variable Correlat
206. sting data columns a new linear column is created from a linear combination of two or more existing data columns In order to create either type of data column you must define a set of appropriate summation weights The example given below describes how to define appropriate summation weights for a typical modeling application Assume for the moment that you have imported a prd data file which contains salinity 63 predictions across 4 sampling depths i e you have 4 existing data columns labeled ECe 1 ECe 2 ECe 3 and ECe 4 Suppose further that you wish to create one new linear data column and one new ratio data column defined as follows New linear column an average of the existing salinity columns ECe ave 0 25 ECe 1 0 25 ECe 2 0 25 ECe 3 0 25 ECe 4 New ratio column a salinity profile shape column ECe shape ECe 1 ECe 2 ECe 1 ECe 2 ECe 3 ECe 4 To create the ECe ave column you would select the Linear Estimates option under the New Column Type and then enter the following summation weights Label Weights ECe 1 0 25 ECe 2 0 25 ECe 3 0 25 ECe 4 0 25 Likewise to create the ECe shape column you would select the Ratio Estimates option and then enter the following numerator and denominator weights Label Weights Weights ECe 1 1 1 ECe 2 1 1 ECe 3 0 1 ECe 4 0 1 Note that in both cases the summation weights simply reflect the appropriate algebraic multiplication fa
207. t conductivity information However to generate the svy data file which the ESAP Calibrate program needs you must generate a sampling design The solution here is to invoke the manual sample site selection procedure select a single sample site any site will do and then save this sampling design This will then trick the ESAP RSSD program into generating the output 48 svy data file thus creating an input file which the ESAP Calibrate program can process A few other points about the manual site selection procedure should be kept in mind First the Manual Sample Site Selection window can no longer be activated after you create and save one or more spatial response surface SRS sampling plans So if you need to generate a manual design you should do it first before you generate any ESAP RSSD SRS designs And second please note that ESAP uses the site ID numbers to identify all manually selected sample Sites Therefor if you wish to use this procedure you should include a site ID column in your input conductivity survey data file and have access to these site ID numbers in order to properly identify your set of sample Sites 3 5 2 Generating ESAP RSSD Spatial Response Surface SRS Sampling Designs The SRS Sample Site Selection window can be used to invoke the SRS spatial response surface sampling algorithm This window can be displayed by clicking on the Design gt Calculate SRS Sample Design menu option You can then use the
208. t is ESAP RSSD gt Navigating the Main Menu gt Frequently Asked Questions ESAP SaltMapper Main Menu Layout Refer to section 4 1 for a general program overview Refer to the manual sections listed below for specific menu option details Manual Sub section Main level Sub Level 2 Sub Level 3 4 2 1 File gt Specify Project Input File Info l gt Column Manipulation l l 4 2 2 l gt Change Column Labels 4 2 3 l gt Create a New Column 4 2 4 l gt Column Statistics l l 4 5 1 gt Create Output Data File l gt Exit 4 3 all Graphics gt ID Line Transect Plot 4 4 all gt 2D Raster Image Map Help gt About ESAP SaltMapper gt ESAP Interface Controls l l gt ReSet default Text Editor l gt OnLine Help gt What is ESAP SaltMapper gt Navigating the Main Menu gt Frequently Asked Questions 13 ESAP Calibrate Main Menu Layout Refer to section 5 1 for a general program overview Refer to the manual sections listed below for specific menu option details Manual Sub section Main level Sub Level 2 Sub Level 3 File gt Import Data File l l 52 1 l gt Import a Survey data file 5 2 2 5 3 all l gt Importa Profile data file l 5 4 all gt Edit or Validate Profile Data gt View Print Project Output Files l gt Exit Calibrate gt Stochastic Methods l l 5 6 all gt Profile Shape Magnitude Analysis 5 7 all l gt Sta
209. tand what these abbreviations mean The definitions shown below may be used as a reference for these abbreviations Definitions Raw Signal Readings sl and s2 note s1 and s2 may be log transformed Decorrelated Signal readings zl and z2 Relationship zl a sl mean s1 a2 s2 mean s2 z2 a3 sl mean s1 a4 s2 mean s2 where a1 a2 a3 and ay are determined by the principal components algorithm If only one signal reading has been collected at each site then z1 a s1 mean s1 where a 1 std dev s1 Raw Location Coordinates u and v Scaled Location Coordinates x and y Relationship x u min u k y v min v k where k the greater of max u min u or max v min v 142 Abbreviations Signal parameters are shown using the abbreviations z1 z2 or z1 i e z1 squared where z1 z1 z1 Trend surface parameters are shown as x y xy xy x y x x x x and y y y y Additionally Ist OT is an abbreviation for first order trend surface i e the parameters x and y and 2nd OT is an abbreviation for second order trend surface i e the parameters x y xy x and y Various numbers of signal and trend surface parameters can be mixed together to form the final model parameter combination The combinations shown below represent a few of the possible standard and custom model definitions Abbreviation Corresponding Regression Model zl bo
210. ter Plot window click on the Predict gt View Prediction Plots SCM menu option After this window appears click on the Plot gt Specify New Plot Variables sub menu option After the Plot Initialization window appears highlight the All Depths Simultaneously option check the Back Transform Data check box and then click on the Initialize Plot command button You have just instructed the ESAP Calibrate program to plot all the sample depths together and to back transform the observed and predicted In ECe data values Now click on the Plot gt Create Plot sub menu option the prediction plot which appears should look like figure 5 7 shown below Note the title will not appear unless you manually add it Observed v s Predicted Training2 ECe Data Predicted oe Sen ee ee ee a a a ee 3 94 5 6 7 8 9 1011 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Observed Figure 5 7 Observed versus predicted Training salinity data As suggested by the regression model summary statistics this plot confirms that the fitted calibration equation can adequately predict the soil salinity levels from the acquired conductivity survey data Hence you can now proceed to calculate the field summary statistics 153 To invoke and display the Field Summary Statistics window click on the Predict gt Calculate Field Summary Statistics SCM menu option Once this window displays you should perform the following steps First check t
211. ter you highlight this check box the Merge Data command button will appear Click on this button to merge your survey and profile data files together After clicking on this command the ESAP Calibrate program will merge your two data files and display the View Log File command button Next click on this button to display and view a detailed description of the file merging results In general you should always view this log file to verify that the survey and profile data files were merged successfully DPPC Analysis Options The command buttons contained within this frame should be used to perform the actual DPPC profile data correlation analysis and to display the analysis results If desired you can log transform one or more of your profile data columns before you perform the DPPC correlation analysis You can also have the ESAP Calibrate program log transform the calculated column of ECac data before computing any correlation statistics Log transformations may be specified by clicking on the Log Transform Columns command button This will invoke and display the Log Transform Columns Window which in turn can be used to specify the columns of data to log transform In general we recommend that you log transform the salinity ECe and calculated conductivity ECac data columns You may also wish to log transform any profile data which is highly correlated with salinity such as SAR boron various cation levels etc Log transforming such d
212. the upper left corner of the main program Window The full layout for this menu bar system is shown below Main level Sub Level 2 Sub Level 3 File gt Set Create Project and Field ID gt Import Survey Data File l l l gt Importa Grid Survey File l gt Import a Transect Survey File l l gt View Print Output Files gt Exit Graph gt Open Graphics Window Analysis gt Basic Statistics gt Signal Decorrelation gt Signal Transformation Design gt Calculate SRS Sample Design gt Manual Sample Site Selection Help gt About ESAP RSSD gt ESAP Interface Controls gt ReSet default Text Editor a OnLine Help gt What is ESAP RSSD gt Navigating the Main Menu gt Frequently Asked Questions FAQ Main Menu Bar Menu Item Descriptions The main level contains 5 menu bar items File Graph Analysis Design and Help You 19 will use the File menu to access all the data input output routines the Analysis menu option to access all of the signal processing routines and the Design menu option to access the sample design routines Additionally you can use the Graph menu item to bring up an interactive Graphic Window and the Help menu to access the main ESAP program help files Additional help file documentation is available for nearly all ESAP program routines this documentation can be accessed by double clicking on the OnLine Help file button within whatever window is currentl
213. tionally a correlation r Statistic is displayed in the second plot this statistic represents a measurement of the correlation between the observed and fitted sample levels across all the sampling depths The closer this r statistic is to 1 the more predictable the profile shapes are from knowledge of the average magnitude values In general you can expect to produce more accurate stochastic calibration models on sample data with profile shape magnitude correlation statistics near 1 as opposed to statistics near 0 which would indicate weak shape magnitude dependencies This usually occurs because as the shape magnitude correlation approaches 1 the correlation between the depth specific and bulk average soil levels also approaches 1 and bulk average sample data can usually be modeled much more accurately than depth specific sample data especially if you are using a conductivity meter which is acquiring some type of weighted bulk average conductivity reading s like an EM 38 meter 5 6 2 Navigating the 1D Profile Menu Profile Menu Bar Menu Layout The ESAP Calibrate Profile menu bar is located in the upper left corner of the 1D Profile Plot program window To invoke and display this window click once on the Calibrate gt Stochastic Methods gt Profile Shape Magnitude Analysis main menu option The full layout for this menu bar system is shown below Main level Sub Level 2 Sub Level 3 Plot gt Specify new Plot Variable
214. tions for 0 0 0 3m 0 3 0 6m or 0 6 0 9m depths 104 Soil Information by depth 0 0 0 3m 0 3 0 6m 0 6 0 9m SP 50 55 60 Bulk Den use Estimate from SP option H20 FC 65 80 90 To enter the 0 0 0 3 meter conductivity information click on the Parameters gt Set Conductivity Parameters sub menu option When the Conductivity to ECa window displays set the temperature value to 17 0 the EMv and EMh height above ground values to 0 05 and 0 10 respectively and the conversion formula type to Log Linear Next highlight the Rhoades Eqn option click on the 0 0 0 3m command button and then on the OK command button The above specified conductivity information will then appear on the main Deterministic Conversion window indicating that these parameter values have been initialized To enter the 0 0 0 3 meter soil information click on the Parameters gt Set Soil Parameters sub menu option When the Soil Parameters window appears set the SP value to 50 and the H20 I FC value to 65 After clicking in the OK command button this information should also appear on the Deterministic Conversion window To perform the calculations click on the Calculate gt Convert Conductivity to Salinity sub menu option The ESAP Calibrate program will then invoke the deterministic conversion routine which will in turn convert each set of EM survey readings into a calculated salinity value In this specific example the program reports that one calcul
215. to the DPPC Graph Options Window If desired you can select a new i e different panel plot to display using the options associated with this latter window 130 5 8 7 Practice Module Performing a DPPC Correlation Analysis Section 5 8 4 describes in detail the correct way to interpret the results from a DPPC correlation print out To recreate the output data shown in that section you should import both the survey and profile data files associated with the Training project 1 e the b102data svy and bwd_102lab pro data files Once you have imported the Training data files click on the Calibrate gt Stochastic Methods gt DPPC Correlation Analysis menu option to invoke and display the DPPC PDCA window After this window displays check the Merge active Survey Profile data file check box and then click once on the Merge Data command button you just merged the svy and pro data file information together Next click on the Log Transform Columns command button check the Calc ECa and ECe check boxes and then click on the OK command button this indicates that you would like to log transform these two data columns Now click on the Perform DPPC Analysis command button to perform the analysis and then the View Correlation Output command button to display the results These results should match the results shown in section 5 8 4 If you wish to create any DPPC data plots click on the Plot Data command button
216. togram To create this figure click on the Graphics gt Histogram gt s2 menu options This plot displays a histogram of the EM 38 horizontal survey data along with some basic summary statistics 40 793 75 400 00 6 25 240 75 546 75 Figure 3 6 GrayScale Map of EMv Survey Data To create this figure click on the Graphics gt ColorScale Grids gt s1 menu options This plot displays a type shaded relief map where the shades of gray represent different conductivity zones 4 zones from low to high 41 Figures 3 1 through 3 6 represent only a portion of the plots that you can create in the ESAP RSSD Interactive Graphics window Hence before leaving this practice module we recommend that you try creating some additional plots using each of the five graphical procedures discussed above 3 4 Data Analysis The following section describes how the ESAP RSSD software program converts your conductivity survey data into transformed and decorrelated signal data which is done primarily for the purpose of facilitating the generation of the sampling design s 3 4 1 Basic Statistics When the ESAP RSSD program imports your conductivity survey data file it automatically generates and displays a set of basic survey statistics on the main program frame However if desired you can recompute these statistics using the options contained within the Signal Transformation Scaling window This window can be displayed by clicking o
217. ts directories are one in the 137 same then it does not matter which directory you select Next you should use the list box shown in the Select Response Variable frame to select the appropriate response variable for your calibration model If desired you can also log transform this variable by highlighting the log transform check box shown to the immediate right of the list box Finally you should click on the Merge Data command button once you are ready to merge your survey and profile data files together Once these two files have been processed a File Merge Results frame will appear which displays the merging results i e which sampling designs your profile data associates with If the displayed results appear to be in error then you should click on the View Log File command button to display a detailed summary of the file merging results Otherwise you can accept the merging results by simply clicking on the Accept Results command button unless you wish to use one of the additional features described below Note once you click on the Accept Results Reject Results or Cancel command buttons the Response Variable Specification window will disappear and the Stochastic Calibration Modeling window will be displayed allowing you to once again access the main stochastic calibration modeling menu items Plot Data Click on this command button if you would like to create and view any preliminary response variable PRV plots This
218. tton displaying a capital H usually located in the lower right hand corner of the window 60 Brief descriptions of the sub level menu items located beneath the 3 main program menu bar items are given below from within File Specify Project Input File Info Select this menu option to open up the Project and Input File window which is where you can set the current project and identify the input file Note that you must define the input file type either a svy or prd file before you can import your data Column Manipulation Select this menu option to manipulate one or more data columns Your column manipulation choices include 1 changing one or more column labels 2 creating one or more new data columns from a combination of existing data columns and 3 computing the univariate column statistics respectively Create Output Data File Select this menu option to open up the ASCII File Export window which can then be used for data export purposes Exit Select this menu option to exit the ESAP SaltMapper program from within Graphics ID Line Transect Plot Select this menu option to open up the 1D Line Plot window This window contains the routines you will use to create and output various types of 1D line transect plots of your survey or prediction data 2D Raster Image Map Select this menu option to open up the Raster Map Initialization window From within this window you can then access the routines needed to cr
219. ture and water content then you should import this data as a DPPC type profile data file This in turn will allow you to perform a DPPC correlation analysis on your input profile data a DPPC correlation analysis can not be performed on generic type profile data files On the other hand if your input data does not contain salinity texture and water content readings at every sample site then you can not import it as a DPPC type profile data file you must use the Generic file format instead 5 3 2 Importing Your Data Setting File Type and Column Structure Specifications Once you have created and saved your ASCII data file which contains your calibration sample data you can import it into the ESAP Calibrate program To import a new calibration sample data file you should again use the Profile Data File Import window This window can be invoked and displayed by clicking on the File gt Import Data File gt Import a Profile Data File main menu option After the Profile Data File Import window has displayed you should perform the following steps First change the Input File Status to new Profile data file After you do this a Project Status frame will appear If you wish to create a new project then click on the create a new Project Directory option else move down to the Set Current Project Directory frame and select your desired project 1 e the project you wish to store your laboratory data in Once you have either set or created y
220. ty data file s 22 3 2 1 Creating a Project Directory and Assigning a Field ID Code The first thing you need to do after starting the ESAP RSSD program is to set or create the working project and input a field ID code The ESAP RSSD program has been designed to organize and track your various field survey output files using projects and field ID codes You can group output data files from similar surveys together within the same project and differentiate between these survey files within the same project using unique field ID codes You should use the Project amp Field ID Window to create or set the project name and enter the 4 character field ID code along with an optional field description if desired This window is activated when you select the menu option Set Create Project and Field ID from the main ESAP RSSD menu bar In ESAP the project name you select or create actually represents or becomes a sub directory branch off the esap2 data directory For example if you select Training as your current project then all ESAP RSSD output files will go to the esap2 data Training 1 subdirectory Likewise if you create a new project called MyProject then ESAP will instruct Windows to create the sub directory branch esap2 data MyProject and subsequently place all output files into this MyProject sub directory The 4 character field ID code is used by ESAP to name and or identify the output files associated with the current d
221. ty survey information Hence you will need to sample at a few 6 to 20 sites across the area in order to develop a prediction model i e an equation which can be used to predict the value s of the soil variable s from the conductivity survey information Thus the idea is to select a set of sample sites which in some way optimizes the prediction model In other words you need to select a set of sample sites which gives you the best possible information for estimating your prediction model ESAP optimizes your sampling design by selecting sample sites which in theory optimize the estimation of your prediction model The ESAP RSSD program does this by examining your input conductivity survey data and selecting conductivity survey sites which represent statistically optimal sample sites using a statistical methodology known as a response surface sampling design Response surface designs have been used for many years in industrial applications to estimate statistical regression models These regression models are typically used to predict the value of a process output from one or more controlled inputs For example they could be used to predict the durability or strength of a manufactured product based on the quality of the raw material inputs or the temperature pressure and or speed of the manufacturing process etc The ESAP RSSD program essentially uses this same statistical methodology to design your soil sampling plans In essence your so
222. ty survey readings in this case EMv versus EMh Basic summary statistics are also computed and displayed within the Statistics frame to the left of the plot 37 EMv amp EMh versus Y for Row 1 Statistics EMv u 1 099 s 0 203 EMh u 0 710 s 0 123 sl s2 Corr r 0 3292 r 0 3214 0 42 6 25 400 00 793 75 Figure 3 3 EMv amp EMh versus Y axis for Row 1 This figure shows a line transect plot which represents an individual transect of your soil conductivity survey data To create this figure first click on the Graphics gt Line Plots gt s1 and s2 by row menu options This will cause the Line Plot frame to be displayed the options within this frame can now be used to create line plots Now click on the Plot button then on the s1 Stats button and finally on the s2 Stats button to create the graph and statistics 38 EMv amp EMh versus Y for Row 7 sl s2 Corr r 0 4263 r 0 5148 6 25 400 00 793 75 Figure 3 4 EMv amp EMh versus Y axis for Row 7 If the Line Plot frame is still displayed this figure can be created as follows Click on the Refresh button to clear the screen and then use i e repeatedly click on the gt gt button to scroll up to transect number 7 Now click on the Plot button then on the s1 Stats button and finally on the s2 Stats button to create the graph and statistics 39 EMh Histogram 141 0 70 5 0 42 0 83 1 24 0 0 Figure 3 5 EMh His
223. u Select this option to close down the 1D Profile Plot Display Window and return to the ESAP Calibrate main menu 5 6 3 Initializing a 1D Profile Plot for Display The 1D Profile Plot Initialization window should be used to select and initialized the calibration sample data you wish to plot This window can also be used to define and set a few of the plot display features such as the sampling units depth units and plotting scale To invoke and display this window click once on the Plot gt Specify New Plot Variable menu option You can use the features listed within the Profile Initialization frame to initialize your 1D profile data plot Simply select the column of calibration sample data you wish to plot from the 109 list of data column labels shown in the combo list box and then click once on the Initialize Plot command button This will initialize the plot data and then return you to the 1D Profile Plot Display menu where you can then use the Plot gt Create Plot menu option to create and display your profile data plot If you need to adjust the way your profile plot s look you may do so by changing either the display scale or by invoking or disabling the In fitting algorithm These features are explained in more detail below Apply In fitting algorithm Yes No When ESAP perform the profile shape magnitude analysis these calculations are normally performed on the untransformed calibration sample data However you can spe
224. u can view the statistics associated with this new data column by clicking on the File gt Column Manipulation gt Column Statistics main menu option If you have created your new EM ratio data column as instructed above then these data should range between 0 371 to 0 421 and have a mean and standard deviation of 0 393 and 0 008 respectively 4 3 Creating 1D Line Transect Plots The following section describes how to initialize display modify and print a 1D line transect plot 4 3 1 Definition of a 1D Line Plot A 1D line plot is simply a graph of your conductivity or prediction data down a specific transect or survey pass This type of plot can be used to display your data on a transect by transect basic or data associated with multiple transects can be overlaid onto a single graph 1D line plots can be created from within the 1D Line Plot window which can be invoked by clicking on the Graphics gt 1D Line Transect Plot menu option In the ESAP SaltMapper program you can overlay up to 4 lines of data within a single plot These lines can either be data arising from up to 4 different transects or from 4 different variables associated with the same transect For example suppose you have imported an ESAP Calibrate prediction data file that contains 4 columns of predicted salinity data 1 e predicted 66 salinity data at 4 sampling depths You could then create a 1D line plot which overlays the predicted salinity data within th
225. u should highlight the Back Transform Mean Field Log Predictions check box if you wish to back transform these log average estimates into field median estimates Range Interval Cut off Levels In addition to calculating average level predictions ESAP can also calculate what percentage of your survey area falls within specific range intervals In other words if you have estimated a soil salinity calibration model you can request ESAP to calculate the proportion of your survey area with soil salinity levels falling between certain threshold values You should use the option controls located within the Range Interval Cut off Levels frame to specify both 150 the number of cut off levels and the value of each cut off level For example if you wanted to define four salinity intervals as lt 2 2 4 4 8 and gt 8 dS m you would select 3 cut off levels and enter 2 4 and 8 into the Ist 2nd and 3rd input text box respectively If your calibration model is based on log transformed response data but your cut off levels have been entered using non transformed units then you should highlight the Log Transform cut off levels check box Note that by highlighting this check box you are telling ESAP to log transform your input cut off levels If desired you can also have the ESAP Calibrate program automatically set the cut off levels for you using a Salinity Lab salt tolerance threshold slope equation To do this click on the Set Levels using s
226. ue The prediction errors which represent the difference between the predicted and true values are then squared and summed together to create the PRESS score Manually Select a Standard Model This is the default model identification method You may be able to choose from up to 12 different parameter combinations depending on your profile data calibration sample size and the number of signal readings you collected at each site i e either 1 or 2 If you have no prior experience fitting regression models we recommend that you use this identification method and simply accept the default selection that ESAP generates when the MLR Model Identification window is first displayed 141 Define and Specify a Custom Model You should select this option if you wish to specify a custom i e non standard parameter combination This option may allow you to choose from up to 36 different parameter combinations depending on your profile data calibration sample size and the number of signal readings you collected at each site Once you have selected an appropriate parameter combination click on the OK command button to return to the SCM window 5 9 6 Understanding the Model Parameter Specification Options The ESAP Calibrate program uses a set of standard abbreviations to indicate the various regression model parameter combinations you may choose from In order to use the advanced features of the model identification options you need to unders
227. value represents the projected yield loss incurred due to the current soil salinity pattern as compared to the yield one would expect under ideal i e non saline conditions 5 9 12 Saving your Output Predictions You can save your output prediction data at any point after you have calculated the field summary statistics To create a permanent prd prediction data file simply click on the Predict gt Save Output Predictions SCM menu option This action will invoke and display a File Save window which can then be used to name and save your output file Note that you must create a permanent ESAP Calibrate prediction data file if you wish to use the ESAP SaltMapper program to create any prediction transect plots or raster maps In general we recommend that you always create such a file 152 5 9 13 Practice Module 2 Analyzing Summarizing and Saving Prediction Data In this last practice session you will learn how to analyze summarize and save regression based salinity prediction data This session will make use of the output from Practice Module 1 section 5 9 8 If you have not already done so you should you should complete Practice Module now Once your calibration equation i e regression model has been estimated you can proceed to analyze and summarize your output prediction data This should be done by viewing the prediction plots and calculating the field summary statistics To invoke and display the 2D Prediction Scat
228. vey readings but these 1000 readings would be associated with only 10 unique rows of data rows 1 through 10 In ESAP you could read this survey data in as a Transect file provided the proper row number was associated with each input line of conductivity data 23 On the other hand Grid files are usually created from a survey conducted in a stop and go manner In other words you go out to a pre specified location within your survey area stop and acquire a conductivity reading move onto to next location stop and acquire another reading etc until the entire survey process is completed Often gird data is therefore acquired in a more uniform manner across the survey area i e evenly spaced in 2 dimensions For example you might collect a 20 by 20 grid of conductivity readings across the area producing a total survey size of 400 readings Furthermore depending on your survey design each set of 20 readings might not be perfectly aligned in either the x or y direction For example this would occur if you used a stratified random survey design etc File Formatting Requirements Regardless of which input file type you chose your input conductivity data file will need to be correctly structured as described below in order for the ESAP RSSD input routine to work properly The order specific column structures for a Grid and Transect survey file are as follows note sl represents the Ist column of conductivity signal data and s2 re
229. w If you have followed the above directions correctly your summary statistics should be identical to the statistics shown below Depth R square Est CV 0 15 0 9025 23 850 0 45 0 8828 26 822 0 75 0 8585 23 841 1 05 0 7969 22 344 bulk ave 0 9263 17 101 Note that the R square column reflects the regression model fitted R values and the Est CV column reflects the estimated model coefficient of variation CV values on the back transformed data If you had not log transformed the salinity data then the second column would have instead shown the regression model root mean square error estimates 147 5 9 9 Viewing 2D Prediction Scatter Plots Once the calibration equation has been estimated you can view 2D scatter plots of the observed versus predicted response data These plots give you a convenient way to judge the adequacy of your calibration equation in addition to the basic model summary statistics automatically displayed in the SCM program window Two plot types are available for viewing an ordinary prediction plot and a jack knifed prediction plot An ordinary observed versus predicted plot shows the correlation between the actual i e true versus regression predicted data values An observed versus jack knifed predicted plot shows the correlation between the true data against the jack knifed predicted data where the jack knifed estimates are calculated by interactively removing each data point from the regression model
230. w to estimate a salinity calibration equation using the Training2 demonstration data i e the sk13data svy survey and sk13_lab97 pro profile data files The results from this exercise will also be used in the next practice module section 5 9 13 Practice Module 2 If you have not already done so you should import both files now before continuing on with this session After the Training2 data files have been imported click on the Calibrate gt Stochastic Methods gt Spatial MLR Analysis main menu option This will invoke and display the Stochastic Calibration Modeling SCM window To estimate your salinity calibration equation you must perform the three sequential steps listed below Step 1 Select the response variable Step 2 Identify the regression model parameters Step 3 Estimate the equation 146 To perform the first step click on the Model gt Select Response Variable SCM menu option This will invoke and display the Response Variable Specification window Now select ECe from the drop down response variable list check the log transform check box and click on the Merge Data command button Verify that the file merge results appear correct in this example you should see 15 sample sites associated with the User Specified SD label and then click on the Accept Results command button You have just selected your response variable i e log transformed salinity and you should see this response variable information app
231. way Twe volumetric water content in the continuous liquid pathway Ts volumetric content of the solid phase of the soil ECws electrical conductivity of the soil water pathway ECwc electrical conductivity of the continuous liquid pathway ECs electrical conductivity of the solid soil particles Additionally if we define Tw Tws Twc total volumetric water content ECw average electrical conductivity of the soil water and assume that under equilibrium ECw ECws ECwc then the above specified model can be written as Ts Tws ECw ECs ECa Tw Tws ECw 5 2 Ts ECw Tws ECs Now it is possible to either measure or estimate all the parameters on the right hand side of equation 5 2 provided that measurements of the soil ECe SP gravimetric water content pW and bulk density Bd are acquired These relationships as shown in Rhoades et al 1989 are as follows Tw pW Bd 100 Tws 0 639 Tw 0 011 Ts Bd 2 65 ECs 0 019 SP 0 434 ECw ECe Bd SP 100 Tw Additionally note that when bulk density measurements are not available the bulk density can be estimated from the SP measurements as follows Bd 1 73 0 0067 SP Hence it is possible to estimate the ECa given knowledge of the ECe SP pW and optionally Bd Such estimates are typically referred to as calculated ECa or ECac data values Equation 5 2 shows how the inner relati
232. will invoke the PRV window which you may then use to produce either 1 bi variate response variable plots or 2 plots of the response variable against the primary z1 principal component score These plots are explained in more detail in section 5 9 4 which describes the PRV graphics menu layout and features Delete a Site The Site Deletion window is invoked and displayed when you click on the Delete a Site command button The options within this window can be used to remove i e delete the profile data core associated with a specific sample site from the current calibration data file You should invoke the Site Deletion window ONLY if you wish to remove a profile data core from your current calibration sample data file BEFORE estimating the regression model In general you should not use this feature unless you have already tried to model the current response variable and discovered one or more extreme outliers which must be removed from the calibration sample data To delete a site select the appropriate sample site number from the displayed sample site list box and then click on the Delete this Site command button Note that you can only delete one site at a time i e you can not highlight and delete multiple sites at once If you wish to delete more than one site you will need to interactively re invoke this Site Deletion window from within the Response Variable Specification window 138 5 9 4 Navigating the Preliminary Re
233. within Help What is a standard correlation analysis Select this option to display the Standard Correlation Analysis help file which explains what a standard correlation plot is and how it can be interpreted Navigating the Standard Correlation Analysis Menu Select this option to display the Navigating the Standard Correlation Analysis Menu help file This help file explains how to use the SCA menu commands to create display print and or save correlation plots 116 General Tips how to highlight sites plot a different depth etc Select this option to display the General Tips help file This help file explains how you can use the various interactive plotting features associated with most ESAP Calibrate graphical displays such as highlighting sites plotting different sample depths etc from within Exit Return to Main Menu Select this option to close down the SCA window and return to the ESAP Calibrate main menu 5 7 3 Practice Module Performing a Standard Correlation Analysis In this practice session you will learn how to perform a standard correlation analysis using the Training2 demonstration calibration sample data i e the sk13_lab97 pro profile data file If you have not already done so you should import the sk13_lab97 pro file now before continuing on with this session refer to practice module 5 2 3 if you ve forgotten how to do this After the Training profile data file has been imported click on the
234. xture and water content data values These calculated soil conductivity readings can then be compared to each column of input soil data thus allowing you to see how well the calculated conductivity correlates with each soil variable If you have survey data associated with these profile sites then you can also look at the correlation between your calculated and measured i e true conductivity readings Stochastic Methods Spatial MLR Analysis Stochastic Calibration Select this menu option to invoke and display the Stochastic Calibration Modeling window The various routines and procedures contained within this window can be used to generate spatially referenced regression models which in turn use the acquired conductivity survey readings to predict the values of one or more soil variables at each survey site within your field or survey area These models can also be used to 1 estimate the average level of each soil variable within the field 2 calculate various additional prediction statistics and 3 generate predictions of your soil variable across multiple sampling depths provided you acquire multiple depth calibration data Deterministic Methods Conductivity to Salinity Select this menu option to invoke and display the Deterministic Conductivity to Salinity Conversion window which can in turn be used to initialize and invoke the deterministic conversion algorithm This algorithm converts conductivity data into salinity data by f
235. y activated the OnLine Help button is the small button displaying a capital H usually located in the lower right hand corner of the window Brief descriptions of the sub level menu items located beneath the 5 main program menu bar items are given below from within File Set Create Project and Field ID Select this menu option to open up the Project and Field ID Window which is where you can set or create the current project and define the current field identification code Note that you must define both the project and field ID before you can perform any other actions in the ESAP RSSD program Import Survey Data File Grid or Transect Select this menu option to import a Grid or Transect conductivity survey data file All ESAP input files must be specified as one of these two types respectively View Print Output Files Select this menu option to view or print an output ESAP text file typically either a sample design file or survey information file using WordPad Note that this option only becomes enabled after you have validated your input conductivity survey data Exit Select this menu option to exit the ESAP RSSD program from within Graph Initialize Graphic Components Select this menu option to open and initialize the Interactive Graphics Window Note that this window contains its own separate menu bar which can be used to produce a number of different plots This window also contains its own set of on line help file
236. y which signal parameters and which if any trend surface parameters should be incorporated into the final calibration model If you are unfamiliar with general regression modeling techniques then you should let ESAP decide the optimal parameter combination However if you have training and or experience fitting statistical regression models then we recommend that you use standard modeling diagnostic techniques to determine an appropriate parameter combination for your particular data set To determine the parameter combination highlight one of the appropriate model identification methods shown in the Model Identification Method frame this should correspond to the identification method you wish to employ for selecting the model calibration parameters You may choose from any one of the three methods listed below Auto Select an Appropriate Model You should select this option if you would like ESAP to automatically examine a number of possible parameter combinations and then select the parameter combination which generates the lowest cumulative PRESS score In regression modeling a PRESS score represents the sum of squares of the jack knifed prediction errors and low PRESS scores generally indicate more accurate models Jack knifing is a technique where each data point is temporarily removed from the data set the model is estimated using the remaining data points and then the predicted value at the removed site is compared to the true val
237. y working through this user manual you will find that each program in the ESAP 95 Software Package actually represents a rich toolbox of modeling and prediction algorithms which can be used to perform many different types of soil conductivity and salinity analyses Most of the details associated with using each program will be covered in chapters 3 4 and 5 The remainder of this chapter will focus on more general principles such as the standard 10 data input output requirements for each program an overview of the three main program menus and a brief summary of the various survey situations and applications which the ESAP 95 Software Package is designed to accommodate 2 2 Data Input Output Figure 1 1 gives a good visual overview of the main data input and output files for each ESAP 95 program As shown in figure 1 1 and described above the ESAP RSSD program is designed to process your conductivity survey data Hence the input to this program is your appropriately formatted soil conductivity data the data format requirements for the ESAP RSSD program are described in chapter 3 Likewise the output from this program is the processed soil conductivity data a file with the extension svy The ESAP Calibrate program is designed to read in an ESAP RSSD svy data file This program can also read in your appropriately formatted soil sample data as described in chapter 4 This sample data can then be saved as a processed profile data file
238. you wish to calculate the salinity levels across multiple profile depths then you will need to run this algorithm in an iterative manner 99 5 5 2 Navigating the Deterministic Conductivity to Salinity DCS Menu DCS Menu Bar Layout The ESAP Calibrate Deterministic Conductivity to Salinity menu bar is located in the upper left corner of the DCS program window This window can be invoked and displayed by clicking on the Calibrate gt Deterministic Methods gt Conductivity to Salinity main menu option The full layout for this menu bar system is shown on the following page Main level Sub Level 2 Parameters gt Set Conductivity Parameters gt Set Soil Parameters Calculate gt Convert Conductivity to Salinity Help gt How does a Deterministic Conversion work gt Navigating the DCS Menu Exit gt Return to Main Menu DCS Menu Bar Menu Item Descriptions The DCS menu contains 4 menu bar items Parameters Calculate Help and Exit You can use the Parameters menu to set your conductivity and soil parameters the Calculate menu to convert your conductivity data to estimated salinity data and the Help menu if you wish to access any DCS help files Brief descriptions of the sub level menu items located beneath the 4 main program menu bar items are given below from within Parameters Set Conductivity Parameters Select this option to define and input your raw conductivity to ECa conversion information This info

ESAP-95 Version 2.01R 95 Version 2.01R User Manual User

Contents

Download Pdf Manuals

Related Search

Related Contents