POPRECON 2 User`s Manual - Columbia Basin Research
Contents
1. But for the pooled data in Figure 27 both the N 3 and N cohorts can contribute to the individuals in N34 thus E N24 Ny3 1 py S1p2 N14 1 p1 S1 p2 Page 40 POPRECON 2 User s Manual ix B Abundance Variance Append Calculation AD Model Builder provides accurate estimates of the model parameters and functions of model parameters However in the case of abundance estimates it underestimates the variance For a given year the abundance of age class is calculated as Pir where x harvest counts of age class i p estimate of harvest probability of age class i r aging reporting probability AD Model Building treats x as a constant whereas it is actually a binomially distributed random variable Hence Program PopRecon estimates the variance as follows Var N Var Page 41 POPRECON 2 User s Manual Appendix B Using the Delta method Var x i pre Var p ar 4 Var r 2 x 733 Cov p r Var N Assuming Cov x p Cov x r 0 let 0 pr E x Var x E x N0 1 0 N 0 NO 1 8 292 pgz NOC 8 NO Var p Var r orp l 02 p T 9272 Var N ee ete 0 N 5 0 2Cov p r Var r Or Var p Var r and Cov p r are calculated by AD Model Builder Page 42 POPRECON 2 User s Manual Appendix B Annual Abundance Variance Calculation For a given year x harvest counts for age2. a tabbed document appears showing the data as shown in Figure 13 Harvest Data Aging Reporting Data oo Year Total Aged 1 3700 3626 2 6400 6353 3 4600 4585 4 4200 3950 5 5000 4886 Figure 13 Aging Reporting data displayed in a tabbed document Entering Auxiliary Data In order to obtain parameter estimates at least one source of auxiliary data must be included The three main types of auxiliary data that Program POPRECON accepts are 1 Harvest data 2 Survival data 3 Abundance data All data entry forms for the auxiliary data have a Paste from Clipboard button in the lower left corner which functions as described above for aging reporting data Page 15 POPRECON 2 User s Manual Chapter 3 Model Setup At least one set of auxiliary data must be entered in order to compute parameter estimates The user can enter multiple forms of auxiliary data if available The Model Summary panel shows what auxiliary data has been entered In Figure 14 harvest auxiliary data has been entered In order to remove the auxiliary data simply right click on it in the Model Summary and select Remove as shown in Figure 15 Model Summary Harvest Data File hary Lcsv Years 5 Age classes 5 Pooled No Harvest Probability Last distinct age Range Random effects Survival Probability Last distinct age Range Random effects Aging Reporting Assume 100 Auxiliary Data Harvest loaded
3. or Survival Probabilities brings up the plot of respective probabilities across years If the number of distinct age classes specified in the Harvest Survival Probability configuration is zero Figure 10 the plot will be like the one in Figure 22 If the last distinct age class is greater than zero the plot will appear as shown in Figure 23 Page 25 POPRECON 2 User s Manual Chapter 4 Estimation Results 4l Harvest Data Auxiliary Harvest Data_ Annual Harvest Plot X Estimates Harvest Estimate 0 1 2 3 4 5 6 Year one standard error 95 Confidence interval Display Al ages x Save to File Figure 23 Harvest Probabilities Plot for multiple age classes Notice the selection box in the lower right corner circled in Figure 23 The default is set for All Ages showing a line plot for each distinct age class The user can use the selection box to view the plot for only one class as shown in Figure 24 The plot for a single age class will then appear similar to Figure 22 with the user able to select which error interval to show for the selected age class Figure 24 Selecting an age class in the Harvest Probabilities plot Page 26 POPRECON 2 User s Manual Chapter 4 Estimation Results Random Effects Plot Harvest Data Auxiliary Harvest Data_ Harvest Random Effects Plot X gt Harvest Random Effects 0 01 Value 0 01 Save to File F
4. 2010 Using statistical population reconstruction to estimate demographic trends in small game populations Journal of Wildlife Management 74 310 317 Clawson M V J R Skalski and J J Millspaugh 2013 The utility of auxiliary information in statistical population reconstruction Wildlife Biology 19 147 155 Clawson M V J R Skalski J L Isabelle and J J Millsoaugh 2015 Trends in male wild turkey abundance and harvest following restoration efforts in the southeast region of Missouri 1960 2010 Wildlife Society Bulletin 39 1 116 128 Gast C M 2011 Fixed and random effects models and multistate estimation procedures for statistical population reconstructions Ph D dissertation University of Washington Seattle WA Gast C M J R Skalski and J Beyer D E 2013 Evaluation of fixed and random effects models and multistate estimation procedures in statistical population reconstruction Journal of Wildlife Management 77 6 1258 1270 Gast C M J R Skalski J L Isabelle and M V Clawson 2013 Random effects models and multistage estimation procedures for statistical population reconstruction of small game populations PLoS ONE 8 6 1 12 DOI 10 1371 journal pone 0065244 Gove N E J R Skalski P Zager and R L Townsend 2002 Statistical models for population reconstruction using age at harvest data Journal of Wildlife Management 66 310 320 Skalski J R M V Clawson and J J Millspau
5. Data The dialog for entering harvest release recapture data is shown in Figure 17 The data is entered at the bottom of the form circled in Figure 17 Marked Marked Harvested Figure 17 Data entry dialog for release recapture auxiliary harvest data For each release recapture record the user must enter 1 The year between 1 and the number of years 2 The age between 0 and the last distinct age class specified in the configuration dialog 3 The total number marked The total number of marked individuals that were harvested for survival data the total number that survived When the user has entered valid values in each field the Add button is enabled The record is added to the table when the Add button is pressed Page 19 POPRECON 2 User s Manual Chapter 3 Model Setup Figure 18 shows the entry dialog with two records already added and a third ready to be added Harvested 350 7 Paste trom Clipboard Figure 18 Data entry dialog for release recapture auxiliary harvest data with data entered Page 20 POPRECON 2 User s Manual Chapter 3 Model Setup Auxiliary Estimates with Standard Errors Figure 19 shows the entry from for auxiliary data that consists of estimates of the harvest probability and the corresponding standard errors Standard Error Harvest Prob Est Standard Error Paste from Clipboard Figure 19 Auxiliary data entry dialog
6. User Interface All parts of the user interface the Navigation Panel the Output Window the Model Summary and all active documents can be moved to a different part of the dialog or outside the POPRECON dialog itself simply by dragging the title bar to the desired location Page POPRECON 2 User s Manual Chapter 2 User Interface Description Page 8 POPRECON 2 User s Manual Chapter 3 Model Setup Load Harvest Data As mentioned in the previous chapter the only action available at startup is Load Harvest Data at the top of the Navigation Panel When the user double clicks on Load Harvest Data he or she will be presented with a dialog as shown in Figure 6 and must indicate whether or not the final age class is pooled A description of pooled vs non pooled data is given in the Appendix A After pressing Continue the user will be asked to select the harvest data file Last Age Class Pooled Not Pooled Pooled Continue Figure 6 Dialog asking whether the last age class of the harvest data is pooled or not The harvest data file is a comma separated value CSV file An example is shown in Figure 7 Page 9 POPRECON 2 User s Manual Chapter 3 Model Setup A E C D E F G H 1 1399 905 294 417 241 170 1 043681 F 2210 1642 1027 713 469 292 1 883665 3 1636 1133 706 480 333 231 1 587571 1464 918 618 460 295 195 1 511499 3 1740 1215 621 511 360 239 2 166131 6 Figure 7 A
7. from the clipboard Enter Survival Auxiliary Data il Enter Harvest Auxiliary Data Page 31 POPRECON 2 User s Manual Chapter 5 Numerical Example Enter Auxiliary Abundance Data Year Abundance Estimate Standard Error 1 21392 81 1058 133 4 Configure Probabilities Age 5 Total Annual Harvest Year 1 7 086 9 335 645 4 937 3 247 719 3 240 6 177 217 2 275 0 136 027 1 314 8 93 042 927 5 74 343 19 782 1 848 325 Survival r Year 2 7 053 4 287 318 5 240 6 220 288 3 277 8 147 233 2 275 6 109 432 1496 9 79 408 931 9 56 711 20 276 2 773 797 Parna Year 3 6 067 0 264 783 4 201 7 192 691 2840 7 139 526 1 802 3 98 099 1 234 9 74 649 856 7 58 252 17 003 3 683 767 Compute Year 4 5 706 8 255 931 3 578 4 172 127 2 409 0 125 349 1 793 1 100 179 1 149 9 72 970 760 1 55 447 15 397 4 633 196 4 View Results Year 5 5 015 1 211 801 3 501 9 154 482 2 366 3 111 099 1472 8 76 353 1 037 6 58 919 688 9 44 344 14 082 6 552 929 Harvest Probabilities Survival Probabilities Recruitment Harvest Random Effects Survival Random Effects 4 Plots Annual Abundance Harvest Probabilities Survival Probabilities Recruitment Harvest Random Effects Survival Random Effects 1 IVI 25210244 TOA y ULOOF J137 LLV UL Newton raphson 2 f 100 2794723216251 max g 5 828670879282072e 014 Exited normally loloma Model Summary vq Harvest Data File harvLesv Y
8. viewed in the same way by double clicking on the appropriate action on the Navigation Panel All of the tabbed documents for the results will have Copy to Clipboard button in the lower left corner as shown in Figure 21 allowing the user to copy and paste the results into a spreadsheet or word processing document Viewing Plots Annual Abundance Plot Double clicking on the Annual Abundance action under the Plots heading will bring up a plot like the one in Figure 22 The X axis is the study year and the Y axis is the abundance estimate By default the black error bars show plus and minus one standard error for each point estimate The buttons at the bottom of the graph allow the user to select 95 Confidence interval instead and the error bars will change to show plus and minus 1 96 standard errors The Save to File button allows the user to save an image of the current plot to a PNG portable network graphics image file Page 24 POPRECON 2 User s Manual Chapter 4 Estimation Results Annual Harvest Plot yf Annual Abundance Plot Harvest Random Effects Plot Recruitment Estimates Plot Save to File Abundance Estimates with One Standard Error 10000 Abundance Estirnate 5000 0 1 2 3 4 5 6 Year one standard error W 95 Confidence interval Figure 22 Plot of Annual Abundance Harvest and Survival Probabilities Plot Double clicking on Harvest Probabilities
9. 0 0 10 0 30 No Figure 14 Model Summary with harvest auxiliary data loaded Page 16 POPRECON 2 User s Manual harvLcsv Years 5 Age classes Pooled No Harvest Probability Last distinct age Range Random effects Survival Probability Last distinct age Range Random effects Aging Reporting Assume 100 Auxiliary Data Chapter 3 Model Setup 0 10 0 30 No Figure 15 Removing auxiliary harvest data Page 17 POPRECON 2 User s Manual Chapter 3 Model Setup Auxiliary Harvest and Survival Data Auxiliary data for both harvest estimation and survival estimation take the same form the examples below demonstrate adding harvest auxiliary data but adding survival auxiliary data works the same way The two possible types of harvest and survival auxiliary data are 1 Release recapture data and 2 Estimates with standard errors To enter harvest or survival auxiliary data the user presses the Auxiliary Data button on the configuration dialog see Figure 10 This brings up a dialog asking the user to select the type of auxiliary data as shown in Figure 16 PR Select Auxiliary Data Select Type of Auxiliary Data Release Recapture Data Estimates with Standard Errors Continue Cancel i Figure 16 Dialog for selecting the type of auxiliary data for harvest or survival parameters Page 18 POPRECON 2 User s Manual Chapter 3 Model Setup Auxiliary Release Recapture
10. B with a user friendly interface for ease and convenience of use Similarly the traditional output from ADMB has been restructured for the specific purposes of statistical population reconstruction Program POPRECON can be installed without the installation of ADMB or a C compiler Future versions of the program will include the expanded ability to incorporate auxiliary study data and advanced model selection capabilities Further information on statistical population reconstruction can be found in the References section Chapter 6 A numerical example using the current software can be found in Chapter 5 Page 1 POPRECON 2 User s Manual Chapter 1 Introduction Page 2 POPRECON 2 User s Manual Chapter 2 User Interface Description eee When Program POPRECON Is started a splash screen is displayed as shown in Figure 1 The splash screen will disappear after five seconds or the user can dismiss it by clicking on it Statistical Population Reconstruction ADMB Version 2 Developed by Jim Lady and John Skalski Columbia Basin Research School of Aquatic and Fishery Sciences University of Washington ILLINOIS for Illinois Department of Natural Resources Department of Wildlife Midwest Furbearer Group Figure 1 The splash screen for Program PopRecon Figure 2 shows the POPRECON program dialog at startup The areas of the POPRECON dialog are as follows e The Navigation Panel along the left side displa
11. Harvest Probabilities Survival Probabilities Recruitment Harvest Random Effects Survival Random Effects Copy to Clipboard Random effects Survival Probability Last distinct age Range Random effects Aging Reporting Assume 100 Auxiliary Data None Model Output PopReconCompute version 2 0 5 Data loaded Figure 8 Program PopRecon after loading the harvest data Page 10 POPRECON 2 User s Manual Chapter 3 Model Setup Model Summary Figure 9 shows the Model Summary panel from Figure 8 after loading the harvest data Note that three headers are in red circled in Figure 9 Harvest Probability Survival Probability and Auxiliary Data The red highlights what components yet need to be configured in order to compute the parameter estimates The Compute action on the Navigation Panel will not be enabled until all components are configured and there is no red on the Model Summary panel Note that the user must enter at least one source of auxiliary data in order for the parameters to be estimable Model Summary Harvest Data File harvLcsv Years Age classes Pooled Range Random effects Aging Reporting Auxiliary Data Figure 9 Model Summary detail from Figure 8 Page 11 POPRECON 2 User s Manual Chapter 3 Model Setup Configure Harvest and Survival Probabilities To configure the harvest or survival pro
12. POPRECON 2 User s Manual POPulation RECONstruction Developed by James Lady and John R Skalski Columbia Basin Research School of Aquatic amp Fishery Sciences University of Washington 1325 Fourth Avenue Suite 1820 Seattle Washington 98101 2509 Prepared for Midwest Furbearer Group Illinois DNR Division of Wildlife One Natural Resources Way Springfield Illinois 62702 October 2015 POPRECON 2 User s Manual Acknowledgments Eo O This project was funded by the Illinois Department of Natural Resources Department of Wildlife Midwest Furbearer Group We would also like to acknowledge Dr Josh Millspaugh of the University of Missouri Columbia for his support of this contract and Mike Clawson Ph D at the University of Washington for testing this software Page i POPRECON Ur s Manual POPRECON 2 User s Manual Page il POPRECON 2 User s Manual POPRECON 2 User s Manual Table of Contents ACKNOWICOOMEONTS pansisronena t nanan AERAR E CRAWL AEH OCIMCTION aao E 1 Chapter 2 User Interface Description occ eesesecesesessssesesessssesesesceseseseseseeseseseeeesees 3 Managing Tabbed DoCuMe nts cccsssssssesssssessesessssssesesssesseseseeeseseseessesseseesenseees 4 VION M lg ee ee 6 Redisplaying a Hidden WINdOW Qu escssesssessssesecssesseseessesseseseeeseseseensesseseeeeeesees 7 Rearranging the User Interface sisisi 7 Chapter 3 Model SY 10 6 ent mt te ee en tne te ee te 9 L gd Harvest Dala
13. are enabled only if the appropriate random effects are enabled and Survival Probabilities under Plots is enabled only if survival random effects are enabled the survival plot would be a flat horizontal line without random effects Viewing Results To view the Abundance Estimates double click on Abundance under the View Results header bringing up the abundance estimates document as shown in Figure 21 For the abundance estimates standard errors are only shown for the total annual abundance estimates in the final column by default If the user toggles the Show all standard errors check box in the lower right corner circled all standard errors will be displayed Page 23 POPRECON 2 User s Manual Chapter 4 Estimation Results Abundance Estimates Annual Harvest Plot Harvest Data Harvest Estimates AgeO Agel Age2 Age3 Age4 Age Total Annual Year 1 6 450 8 4 480 3 2 940 6 2 064 4 1 193 1 841 6 17 950 86 1067 128 Year 2 6 545 1 4 661 4 3 040 6 2 110 9 1 388 5 664 5 16 609 1 922 566 Year 3 5 594 2 3 6 4 2 2 619 3 1 661 6 1 138 7 789 9 15 678 0 815 106 Year 4 5 237 4 3 264 1 2 210 9 1 645 6 1 055 3 697 6 14 130 59 767 254 Year 5 4 632 9 3 235 1 2 186 0 1 560 6 958 5 636 4 13 009 4 646 116 Copy to Clipboard C Show all standard errors Figure 21 Abundance Estimates results All other estimates Harvest Survival Recruitment Harvest Random Effects Survival Random Effects can be
14. babilities double click on Harvest or Survival under Configure Probabilities in the Navigation Panel This will bring up the appropriate configuration dialog Figure 10 shows the Configure Harvest Probability dialog The Configure Survival Probability dialog has the same layout The user is then required to enter the following before parameter estimation is allowed Last Distinct Age Range Min C Enable Random Effects Auxiliary Data Figure 10 Configure Harvest Probability dialog e The last distinct age class All age classes up to the last distinct age class will be modeled with unique harvest survival coefficients all age classes greater than the last distinct age class will be modeled with the same coefficient For example if the user leaves it at its default value of O zero then all age classes will be modeled with the same harvest survival coefficient if set to 1 age class 0 will be modeled with one coefficient and all older age classes will be modeled with another coefficient Note that for harvest probabilities the resulting harvest probability estimates are a function of both the estimated harvest coefficient and the harvest effort thus harvest probabilities for a given age class may differ across years due to different harvest efforts Page 12 POPRECON 2 User s Manual Chapter 3 Model Setup e A minimum and maximum value The numerical optimizer used to estimate the paramete
15. class i p estimate of harvest probability for age class i r aging reporting probability 0 pir Var N Var gt a Var gt 3 z X Var 2 X Cov N i lt j Using the Delta method as N N Cov N ry r cov pi D p rCov p r i j pirCov p r Cov p pj and Cov p r are calculated by AD Model Builder Page 43 POPRECON 2 User s Manual
16. complex nonlinear models Optimization Methods and Software 27 233 249 Gast C M J R Skalski and J Beyer D E 2013 Evaluation of fixed and random effects models and multistate estimation procedures in statistical population reconstruction Journal of Wildlife Management 77 6 1258 1270 Page 37 POPRECON 2 User s Manual Chapter 7 Literature Cited Page 38 POPRECON 2 User s Manual Appendix A Pooled vs Non Pooled Data Cs When the user loads harvest data into Program POPRECON he or she is first asked if the last age class is Pooled on Not Pooled Figure 6 Figure 26 diagrams the age class structure of non pooled data In this case there are four age classes and an individual can only be observed in the final age class once before dying or exiting the study Figure 27 shows the age class structure for pooled data Unlike with non pooled data an individual can be observed in the final age class 4 years old and older for multiple years of a study Age Class Figure 26 Age class structure of non pooled data Page 39 POPRECON 2 User s Manual Appendix A Age Class Annual Na gt Neo N53 Niu Figure 27 Age class structure of pooled data To illustrate the difference let p probability of harvest in year i S probability of survival from year i to year i 1 The expected value of N3 in Figure 26 for non pooled data is calculated as E No4 Ny3 1 p1 S1 P2
17. ct age 0 Range 0 60 0 80 Pandom effects No Figure 4 Drop down menu showing all active documents View Menu The View Menu at the top of the POPRECON dialog maintains a list of all documents that have been created during the current session An example is shown in Figure 5 An active document can be hidden in two ways 1 Click on the X on the currently displayed document tab to close it as shown in Figure 3 or 2 From the View menu select the name of the document and un check it by clicking on its name as shown in Figure 5 Note that when a window is hidden the associated information is not removed from the program It is simply no longer displayed Page 6 POPRECON 2 User s Manual Chapter 2 User Interface Description Abundance Plot _ Recruit Auxiliary Harvest Data Recruitment 366 2 371 361 Abundance Estimates 545 6 333 641 Harvest Estimates 578 0 300 106 201 8 284 785 612 0 231 019 Survival Estimates Recruitment Estimates Annual Abundance Plot Harvat Probabilities Survival Probabilities Recruitment Harvest Random Effects Survival Random Effects 4 Plots Annual Abundance Figure 5 Closing an active window Auxiliary Harvest Data by unchecking it in the View menu Redisplaying a Hidden Window In order to display a window that was previously hidden select its name in the View menu and check it by selecting its name Rearranging the
18. ears 5 Age classes 6 Pooled No Harvest Probability Last distinct age 0 Range 0 20 0 40 Random effects Yes Survival Probability Last distinct age 0 Range 0 50 0 80 Random effects No Aging Reporting Assume 100 Auxiliary Data Page 32 POPRECON 2 User s Manual Chapter 5 Numerical Example J Harvest Estimates X _ _ Survival Estimates X _ All Ages All Ages Year 1 0 183 0 0074 Year 1 0 830 0 0129 Year 2 0 313 0 0115 Year 2 0 830 0 0129 Year 3 0 270 0 0103 Year 3 0 830 0 0129 Year 4 0 257 0 0099 Year 4 0 830 0 0129 Year 5 0 347 0 0130 Year 5 0 830 0 0129 Recruitment Estimates x AT SS ae x Year Recruitment Year Random Effect 1 7 086 9 335 645 2 7 053 4 287 318 l Baai rl 3 6 067 0 264 783 apmeeram 4 5 706 8 255 931 E 5 5 015 1 211 801 4 0 0037167321534 7674 5 0 00070546163023 7253 I Statistical Population Reconstruction ADMB cjB va _ Auwviliary Survival Data Auxiliary Abundance Data Abundance Estimates_ Annual Abundance Plot X Model Summary a 4 Configure Probabilities Abundance Estimates with 95 Confidence Interval paiga jn Harvest sv viva Years 5 Aging Reporting 20000 Age classes 6 Enter Auxiliary Data Pooled No or Harvest Probability Abundance Last distinct age 0 Harvest Probabilities Range 0 20 0 40 Survival Probabilities 15000 Random effects Yes Recruitment u Harvest Random Effects g Survival Pr
19. ected if the aging reporting data is for all years of the study 3 Year Specific This is selected if there is a unique aging reporting data for each year of the study If the user selects either Same for all years or Year Specific the Aging Reporting Data button on the dialog is enabled The user then presses the Aging Reporting Data button to bring up the form for entering the aging reporting data Figure 12 shows the entry from for a five year study with year specific data There is a data entry line for each year unless the user selected same for all years above in which case there would only be one row in the table In the first column the user enters the total harvest counts for the year In the second column the user enters the number actually reported and aged those that can actually be used for the study Done Cancel Figure 12 Aging Reporting Data Entry Form for year specific data for a five year study Page 14 POPRECON 2 User s Manual Chapter 3 Model Setup Notice the Paste from Clipboard button at the bottom left If the user wishes he or she can copy the data from a spreadsheet program and paste it directly into the table without having to manually enter it The data must conform to the table in this case it would have to consist of five rows and two columns of integer data After the aging reporting data has been entered and the user presses Done
20. ee 31 POP REC ONO UOT P 32 Chapter 6 References oc cccccsessssssessssesesesesscscsesesesscsesssesscsesescseeseseseseesssessseseeseseseeees 35 Chapter 7 Literature Cited wc cscsessssesesesesesesssssscscsesesesesesesesssscseseseseseseeessesess 37 Appendix A Pooled vs Non Pooled Data cc cccsesesesesesesesseseeseseseseeeeeeen 39 Appendix B Abundance Variance Calculation cccccecesesesesessseeseseseeeeeees 41 Annual Abundance Variance Calculation ccccsessesesssessesessessesseseesseseseeeees 43 Page iii POPRECON 2 User s Manual POPRECON 2 User s Manual Page iv POPRECON 2 User s Manual Chapter 1 Introduction POPRECON 2 POPulation RECONstruction http www cbr washington edu analysis apps PopRecon Program POPRECON 2 was developed to estimate population abundance based on full age at harvest data supplemented with harvest effort and auxiliary estimates of survival abundance or harvest vulnerability Maximum likelihood methods are used to estimate annual harvest probabilities which are in turn used to estimate age class and annual abundance based on Horvitz Thompson type estimators as described in Gast et al 2013 Random effects can be added to the survival and or harvest processes to account for annual stochasticity in the parameters Data processing is based on AD Model Builder Fournier et al 2011 which uses both symbolic differentiation and numerical integration Program POPRECON has combined ADM
21. for harvest estimates with standard errors Data entry works the same way as with release recapture data described above except each record consists of 1 The year 2 The age 3 A probability estimate between 0 and 1 4 The corresponding standard error Page 21 POPRECON 2 User s Manual Chapter 3 Model Setup Auxiliary Abundance Data In addition to harvest and survival data the user can enter auxiliary abundance estimates and standard errors by double clicking Abundance under Enter Auxiliary Data on the Navigation Panel Figure 20 shows the entry form for abundance auxiliary data Data is entered in the same way as other auxiliary data Year Abundance Estimate Standard Error Abundance Standard Error Paste from Clipboard Jone Figure 20 Auxiliary data entry form for abundance data Each record consists of 1 The year 2 An abundance estimate for the year 3 An estimate of the corresponding standard error Page 22 POPRECON 2 User s Manual Chapter 4 Estimation Results Once the harvest and survival probabilities have been configured and at least one form of auxiliary data has been entered the Compute action in the Navigation Panel is enabled The user double clicks the Compute action to estimate the model parameters If successful the actions under the View Results heading in the Navigation Panel become available with a few exceptions Harvest and survival random effects
22. fort Data Format Harvest and catch effort data are read into POPRECON using a CSV file Statistical Population Reconstruction ADMB File Edit View Help Load Harvest Data 4 Con Probabilities Survival Aging Reporting 4 Enter Auxiliary Data Abundance Compute 4 View Results Abundance Harvest Probabilities Survival Probabilities Recruitment Harvest Random Effects Survival Random Effects 4 Plots Annual Abundance Harvest Probabilities Survival Probabilities Recruitment Harvest Random Effects Survival Random Effects PopReconCompute version 2 0 5 Data loaded Year 1 Year 2 Year 3 Year 4 Year 5 _ Harvest Data X AgeO Agel Age2 Age3 Age4 AgeS 905 1642 1133 918 1215 1299 2210 1636 1464 1740 594 1027 766 618 821 417 713 486 460 511 241 469 333 295 360 170 292 231 195 239 Effort Total Catch Catch Effort 1 0437 1 8839 1 5876 1 5115 2 1661 3626 6353 4585 3950 4886 3474 2417 3372 2865 2888 0598 2613 2998 2255 6346 Years 5 Age classes 6 Pooled No Harvest Probability Last distinct age 0 Range 0 20 0 40 Random effects Yes Survival Probability Last distinct age Range a Random effects Aging Reporting Assume 100 Auxiliary Data Harvest loaded Model POPRECON 2 User s Manual Page 30 Chapter 5 Numerical Example Auxiliary Data Formats Auxiliary data can be typed in the boxes or pasted
23. gh 2012 Model evaluation in population reconstruction Wildlife Biology 18 225 234 Page 35 POPRECON 2 User s Manual Chapter 6 References Skalski J R and J J Millsoaugh 2002 Generic variance expressions precision and sampling optimization for the sex age kill model of population reconstruction Journal of Wildlife Management 66 1308 1316 Skalski J R J J Millsoaugh and M V Clawson 2012 Comparison of statistical population reconstruction using full and pooled adult age class data PLoS ONE 7 3 1 7 DOI 10 1371 journal pone 0033910 Skalski J R J J Millsoaugh M V Clawson J L Belant D R Etter B J Frawley and P D Friedrich 2011 Abundance trends of American martens in Michigan based on statistical population reconstruction Journal of Wildlife Management 75 1767 1773 Skalski J R K E Ryding and J J Millsoaugh 2005 Wildlife demography Analysis of sex age and count data Academic Press San Diego CA 656 PP Skalski J R R L Townsend and B A Gilbert 2007 Calibrating population reconstruction models using catch effort and index data Journal of Wildlife Management 71 1309 1316 Page 36 POPRECON 2 User s Manual Chapter 7 Literature Cited ee Fournier D A H J Skaug J Ancheta J Ianelli A Magnusson M N Maunder A Nielsen and J Sibert 2012 AD Model Builder using automatic differentiation for statistical inference of highly parameterized
24. h the one labeled Harvest Data currently being displayed The user can display any active document by simply clicking on the appropriate tab across the top Page 4 POPRECON 2 User s Manual Chapter 2 User Interface Description Annual Harvest Plot Annual Abundance Plot Harvest Data X Auxiliary Harvest Data Abundance Estimates Age2 Age3 Age4 Age5 Effort Total Catch Catch Effort 417 241 170 1 0437 3626 3474 2417 713 469 292 1 8839 6353 3372 2865 486 333 231 1 5876 4585 2888 0598 460 295 195 15115 3950 2613 2998 239 2 1661 4886 2255 6346 Display an active document by clicking on its tab Click on drop down menu for all active documents copy to Clipboard Figure 3 Main part of PopRecon dialog with several active documents In some cases there may be a large enough number of active documents so that there is not enough room to display all the tabs with the document names across the top In that case as shown in Figure 3 above the user can select the drop down menu displaying all active documents as shown in Figure 4 and select the desired document Page 5 POPRECON 2 User s Manual Chapter 2 User Interface Description Sax Model Summar vi Abundance Estimates Annual Abundance Plot _ Auxiliary Harvest Data Abundance Estimates EN Auxiliary Harvest Data Harvest Data Harvest Estimates Recruitment Estimates Survival Estimates random GMects Survival Probability Last distin
25. igure 25 Plot of the Harvest Random Effects Figure 25 shows the plot of the harvest random effects This plot is available if harvest random effects have been enabled similarly with survival random effects The two horizontal red lines are at 1 96 standard errors from zero If the selected model adequately fits the data the random effects should have a normal distribution with a mean of 0 and 95 of the standard errors should lie between the red lines Page 27 POPRECON 2 User s Manual Chapter 4 Estimation Results Write Estimates to File Once the estimates have been computed the user can save the estimates to a CSV comma separated value file by going to the File menu and selecting Write Estimates to File The user is then prompted for an output file name Resetting the Estimates Once the parameter estimates have been computed the user cannot change any of the inputs add or remove auxiliary data change the parameter configurations etc without resetting the estimates To reset the estimates go to the Edit Menu and select Reset Estimates This will remove all displayed results and plots the Compute action will once again be enabled and the user can then change the model setup Page 28 POPRECON 2 User s Manual Chapter 5 Numerical Example E Examples of data inputs and outputs using Program POPRECON Page 29 POPRECON 2 User s Manual Chapter 5 Numerical Example Harvest and Ef
26. n example of a harvest data file opened with Microsoft Excel 2013 Each row represents one year of harvest data the harvest data in Figure 7 has 5 years of data Each column represents one age class with column A representing age class 0 column B age class 1 and so on The final column is the measure of the harvest effort for each year One note on the harvest effort The estimation seems to work best if the harvest effort is somewhere in the neighborhood of 1 0 This can be achieved by multiplying the harvest effort values by a constant Once the user has loaded the harvest data a tabbed document appears showing the harvest data as shown in Figure 8 The Harvest Data tab shows the harvest data as well as the Total Catch total number harvested in a year and the Catch Effort Total Catch divided by Effort Model Summary AgeO Agel Age2 Age3 Age4 AgeS Effort Total Catch Catch Effort Harvest Data 4 Configure Probabilities Harvest Year 1 1299 417 170 1 0437 3626 3474 2417 File harvL csv Survival Year 2 2210 713 292 1 8839 6353 3372 2865 Aging Reporting ge classes Pa Sea Year 3 1636 231 1 5876 4585 2888 0598 aie Pa Abendence Year 4 1464 195 1 5115 3950 2613 2998 ne Compute Year 5 1740 239 2 1661 4886 2255 6346 Paraat nOn 4 View Results Last distinct age Abundance Range Harvest Probabilities Survival Probabilities Recruitment Harvest Random Effects Survival Random Effects 4 Plots Annual Abundance
27. obability Survival Random Effects Last distinct age 0 Pos S Range 0 50 0 80 10000 Random effects No Harvest Probabilities e Survival Probabilities 3 Aging Reporting Harvest Random Effects Assume 100 Survival Random Effects Auxiliary Data 5000 Harvest loaded Survival loaded Abundance loaded Model 0 parameters 3 0 1 2 log likelihood 965 9594 AIC 199 9188 er one standard error 95 Confidence interval PS AUVs 95k 20S Ss TIGA y ULOJ JDI 24 UV OE VLE Newton raphson 2 f 100 2794723216251 max g 5 828670879282072e 014 Page 33 POPRECON 2 User s Manual Chapter 5 Numerical Example I Statistical Population Reconstruction ADMB mer Harvest Data Harvest Estimates with 95 Confidence Interval a Configure Probabilities Harvest Survival Aging Reporting b Enter Auxiliary Data Compute 4 View Results Abundance Harvest Probabilities Survival Probabilities Recruitment Harvest Random Effects Survival Random Effects 4 Plots E02 Survival Probabilities Recruitment Harvest Random Effects Survival Random Effects one standard error 95 Confidence interval PS LUM 2922S 44 TIGA Y VLOOI JLo de OE Use Newton raphson 2 f 100 2794723216251 max g 5 828670879282072e 014 Exited normally Page 34 POPRECON 2 User s Manual Chapter 6 References EC Broms K M J R Skalski J J Millsoaugh C A Hagen and J H Schulz
28. rs needs starting values as input Program POPRECON will use the mean of the range specified by the user as the starting value If random effects are enabled the range will be used as a best guess as to a 95 confidence interval In addition to these required entries the user may also 1 enable random effects by checking the checkbox labeled Enable Random Effects or 2 Enter auxiliary harvest or survival data by clicking on the Auxiliary Data button The entry of auxiliary data is discussed later in this chapter Configure Aging Reporting Data In some cases not all harvested animals are aged or reported Animals that are not aged or reported are not included in the age at harvest matrix However accounting for their harvest is important Therefore an aging reporting parameter can be included assuming the animals that are aged reported are a random sample from the total harvest To enter aging reporting data double click on Aging Reporting under Configuration on the Navigation Panel This will bring up the Configure Aging Reporting Probability Dialog as shown in Figure 11 Assume 100 Same for all years D Year Specific Done Cancel Figure 11 Aging Reporting Dialog Page 13 POPRECON 2 User s Manual Chapter 3 Model Setup The user selects one of three options 1 Assume 100 This is the default No other action is needed if this is selected 2 Same for all years This is sel
29. si rere ten err tere Rrnr rT ere teeter renner tener Aai 9 Model Summa siiin neste snes teeter tre tt nnn inn nee er serra wer aarts 11 Configure Harvest and Survival Probabilities cesses 12 Configure Aging Reporting Data se sesessoesseooso1sseoosroossetnsstesseenssrnssresseessernss 13 Entering Auxiliary Data ecssessssssssesessssssssessesssssessesssssesssssssssesssssesseseseesses 15 Auxillary Harvest and Survival Data ou ccesessesesessssesssssesesssssesesesseeseseees 18 Auxillary Abundance Data uu sessssssssesesssssesscsesssesessssssesessesssesessessessesseseens 22 Chapter 4 Estimation R SUItS 0 cc cscscsesesesesesesessssesesesesesssessssseseseseseseseesssesees 23 Viewing Results nen RO nen eee er re ee nen Pe eT even ere re rt ner ener ete oreerer vere 23 VOND PIO eens cnr ren ne Te ne renee er rn ere 24 Annual Abundance Plt esesessesssesessesssssessesssssesessssseseesessessessesesssesseseens 24 Harvest and Survival Probabilities Plot csssesseeseeesesseseeseeeeseeeeen 25 Random Effects Plot ccssessssessssssesecessesessssesessssesessessecsscsesesseseessesessseeeeseees 27 Wite Estimates TO Fil Osorereicrnir n nd nnana EnA 28 PRE SEU LINC the ESUMIATES iioc einan E r ar 28 Chapter 5 Numerical Example s sssssessessesssessessesssessesseessrnsenseessrnsenssensrnsenseessrnseesees 29 Harvest and Effort Data Format cccsssssssssessssssesessssessssesessssesssessesseesseeseseaes 30 AS Data i One Sx eceeraereee nee T neneeee
30. ys a hierarchical arrangement of actions Actions currently available are highlighted in blue unavailable commands are in gray The user executes an available action by double clicking on it At startup the only available action is Load Harvest Data Page 3 POPRECON 2 User s Manual Chapter 2 User Interface Description File Edit View Help Navigation Panel Model Summary Load Harvest Data No Data Loaded 4 Configure Probabilities Harvest Survival Aging Reporting 4 Enter Auxiliary Data Abundance Compute 4 View Results Abundance Harvest Probabilities Survival Probabilities Recruitment Harvest Random Effects Survival Random Effects 4 Plots Annual Abundance Harvest Probabilities Survival Probabilities Recruitment Harvest Random Effects Survival Random Effects Output PopReconCompute version 2 0 5 Figure 2 Program PoPRECON Main Dialog at startup e The Model Summary along the right side summarizes the current state of the model development e The Output at the bottom of the dialog displays all informational messages along with output from AD Model Builder e The main part of the POPRECON Program dialog is the center section where all program data and output are displayed Managing Tabbed Documents When data is displayed a tabbed document is created and displayed in the main part of the POPRECON main dialog Figure 3 shows an example with several active documents wit
Download Pdf Manuals
Related Search