User's Guide for Biome-BGC, Version 4.1.1
Contents
1. Each change is described in detail below Thanks to Bev Law Peter Anthoni and Emil Cienciala for providing data and or assistance in the analysis of the model dynamics which made these improvements possible 1 Deployment of retranslocated leaf N for new growth In version 4 1 the entire pool of retranslocated N was available for new growth on each day which resulted in an unrealistic spike in productivity in the early part of the growing season for an N limited system In version 4 1 1 only a fixed proportion of the retranslocated N pool is available for new growth on each day For more details see the comments in the header of the source code file daily_allocation c 2 Handling of C pool deficit In version 4 1 any time there is a deficit in the C pool state variable e g as occurs in the spring for a temperate zone deciduous forest simulation due to zero photosynthetic potential and moderate respiration demands from live wood through the winter the early growing season production is used to eliminate this deficit This resulted in an slight but abrupt shift in the early growing season at the point when the deficit was alleviated since it is assumed that additions to the Cpool do not incur growth respiration costs In version 4 1 1 this mechanism is made slightly more complicated with allocations to eliminate a C pool deficit made at a rate that would alleviate the deficit if continued for a specified number of days For more details2. 0 for writing 1 or not writing 0 a restart file at the end of this simulation 3 flag 1 or 0 for met year from restart file 1 or met year reset to beginning of record 0 4 input restart filename including path if appropriate 5 output restart filename including path if appropriate lt EXAMPLE gt RESTART keyword start of restart control block 1 flag 1 read restart file 0 don t read restart file 0 flag 1 write restart file 0 don t write restart file T flag 1 use restart metyear 0 reset metyear restart test endpoint input restart filename restart test endpoint output restart filename lt END EXAMPLE gt NOTES Lines 3 and 4 are only relevant when line flag in line 1 is set to 1 Line 5 is only relevant when flag in line 2 is set to 1 In most cases the flag in line 3 should be set to 1 TIME_DEFINE block The next section begins with the keyword TIME_DEFINE It has the following five lines 1 number of years of data in meteorological input data file 2 number of years to run for this simulation 3 first simulation year e g 1998 4 flag 1 or 0 for spinup simulation 1 or normal simulation 0 5 maximum number of years to run in spinup mode lt EXAMPLE gt TIME DEFINE keyword do not remove 44 int number of meteorological data years 100 int number of simulation years 1950 int first simulation year 0 flag 1 spinup simulation 0 normal
3. 1988 288 0 148 L 4 280 5 0 0 70 3 518 1989 363 0 6 8 eed 368 2 0 0 141 4 13 3 1990 350 0 7 6 L 2 331 0 0 0 139 2 TSO 1991 308 0 Te 2 298 0 0 0 70 9 536 1992 300 0 8 0 il SIDE 0 0 142 2 14 6 1993 358 0 6 1 L 1 360 5 0 0 1 3 9 9 17 0 Looking at the example binary output in IDL For users with access to the IDL software there is an example IDL procedure in the main directory that will read and display the output stored in the enf_test1 dayout file For UNIX users start IDL in the unixbgc411 directory then enter the command example_plot on the command line For PC users after starting IDL it may be necessary to change the working directory under File preferences menu to the pcbgc411 directory Then open the example_plot pro file compile and execute In both cases a window will be created that will display three different plots for each of the 23 output variables requested in the enf _test1 ini file You scroll through these plots by pressing any key The first plot for each variable shows all the daily data for 44 years as a time series The second plot for each variable shows one average year of daily data the average of the 44 individual years The third plot shows the average value for each year as a 44 year timeseries This example IDL procedure is included to give you some ideas on how to read and display the binary data Compiling Biome BGC source code If you are working on an operating system other than IBM AIX or Windows 95 98
4. 4 0 0 169 1 LTsS5 1952 226 0 6 8 L 2 296 6 0 0 82 6 48 7 1953 302 0 8 2 Loe 295 2 0 0 96 0 28 5 1954 356 0 6 8 ork 358 1 0 0 150 6 22 8 1955 415 0 D2 L 1 345 6 0 0 105 5 17 0 1956 409 0 6 8 ee 456 8 0 0 237 8 57 4 L957 318 0 Gig L 3 318 5 0 0 91 6 33 8 1958 439 0 8 2 L 3 410 5 0 0 13249 EgS 1959 422 0 6 4 is 423 4 0 0 188 1 24 6 1960 261 0 Gad ie 293 9 0 0 78 6 49 2 1961 366 0 6 8 b3 337 2 0 0 110 4 20 9 1962 315 0 6 0 L 2 343 2 0 0 116 8 16 4 1963 388 0 6 6 L 2 3 95 3 0 0 189 3 SERES 1964 399 0 5 0 Led 346 6 0 0 126 1 L227 1965 374 0 6 0 P23 420 2 0 0 LO 2h 44 0 1966 291 0 TEB l3 299 7 0 0 88 6 41 1 1967 318 0 ed L 3 294 4 0 0 69 6 950 iD 1968 324 0 7 1 bd 33548 0 0 145 9 14 0 1969 341 0 6 0 L 2 354 6 0 0 135 9 2 0 1970 394 0 6 7 eZ 381 1 0 0 158 7 1 019 1971 343 0 TEA 3 328 8 0 0 108 9 28 0 1972 Sro p RAe 6 6 S 361 1 0 0 162 2 9 5 1973 233 0 7 6 L 2 183 6 0 0 3163 86 8 1974 283 0 7 6 L 2 360 7 0 0 183 7 26 7 T975 476 0 6 0 be 393 7 0 0 17548 349 1976 241 0 TAPAE L 3 333 6 0 0 109 2 24 0 1977 338 0 6 9 bi 249 2 0 0 68 2 43 9 1978 311 0 6 2 LeZ 368 0 0 0 181 0 31 5 1979 273 0 6 9 ee 281 6 0 0 77 4 51 7 1980 499 0 Tal l3 487 6 0 0 259 0 95 4 1981 450 0 7 6 L 4 418 2 0 0 171 5 43 6 1982 400 0 6 4 Pas 438 0 0 0 191 6 26 9 1983 433 0 6 8 L 6 409 3 0 0 180 0 65 3 1984 345 0 6 9 L 6 362 7 0 0 121 4 11 5 1985 328 0 5 8 L 5 321 3 0 0 T059 SDA 1986 432 0 7 4 LD 423 1 0 0 TOT 33 21 1987 265 0 7 6 L 5 302 3 0 0 1 0273 20 8
5. deficit 5 scaling factor for incoming shortwave radiation lt EXAMPLE gt CLIM_CHANGE keyword do not remove 0 0 deg C offset for Tmax 0 0 deg C offset for Tmin 1 0 DIM multiplier for Prep 1 0 DIM multiplier for VPD 1 0 DIM multiplier for shortwave radiation lt END EXAMPLE gt NOTES This section is included to facilitate simulation of simple climate change effects on ecosystem processes The default values are shown in the example above which result in no changes to the input meteorological data Increases or decreases in temperature Tmax and or Tmin can be introduced with lines 1 and 2 The values indicated are simply added to the daily values from the meteorological data file For precipitation VPD and radiation the daily values from the input file are multiplied by the values on lines 3 4 and 5 to get new daily values This mechanism does not allow for the simulation of seasonal differences in climate change CO2_CONTROL block The next section begins with the keyword CO2_CONTROL It has the following 3 lines 1 flag 0 1 2 controlling CO2 concentration 0 constant 1 varying using values from a special file 2 constant but use the CO2 values in the specified file to control levels of N deposition see below for N deposition options 2 the value to use for constant CO2 concentration ppm 3 the filename for annual CO2 levels see notes below for format information lt EXAMPLE gt CO2_CON
6. name of an initialization file Enter the following line of text and press enter ini enf_testl ini UNIX and PC At this point the program is launched and you should again see a progression of numbers scrolling down the left side of the screen or window This simulation is set to run once through the 44 years of meteorological data In this case the scrolling numbers start with 1950 and end with 1993 since 1950 was given as the starting year for the met data in the ini file This simulation will produce a complete set of output files in the outputs directory all having filenames beginning with enf_test Consult the enf_test1 ini file to see which variables are being output for the binary files The simple annual text file enf_test1_ann txt is also created and it is reproduced below in its entirety so that you can check your results Contents of enf_test1_ann txt Annual summary output from Biome BGC version 4 1 ann PRCP annual total precipitation cm yr ann Tavg annual average air temperature deg C max LAI annual maximum value of projected leaf area index m2 m2 ann ET annual total evapotranspiration mm yr ann OF annual total outflow mm yr ann NPP annual total net primary production gC m2 yr ann NPB annual total net biome production gC m2 yr year ann PRCP ann Tavg max LAI ann ET ann OF ann NPP ann NBP 1950 376 0 5 9 L 1 340 1 0 0 124 0 2 20 1951 390 0 5 8 ae 370
7. vegetation type being simulated It is kept separate from the initialization file so that multiple initialization files can reference the same ecophysiology constants without cutting and pasting Researchers at NTSG have spent considerable effort summarizing a large number of ecophysiological studies from the literature to come up with a set of default parameterizations for a small number of highly aggregated vegetation classes These epc files are supplied with this distribution of the Biome BGC code and users are encouraged to use them as a first approximation of the appropriate parameters for a new simulation If you have good measurements from your site s relating to any of these parameters you should replace the defaults with your observations Users are cautioned that some of these parameters show strong covariance and so replacing some but not others with local observations may reduce the quality of results For example canopy average specific leaf area leaf C N and fraction of leaf N in Rubisco tend to covary so if you replace any of these default values you should consider replacing all of them Consult with your local ecophysiologist if you aren t sure about reparameterization of the default epc files Note that all epc files must have the same parameter lines in the same order but that not all lines are relevant to all vegetation types Lines marked with before the units in the default epc files indicate an irrelevant paramet
8. 1996 869 4 8 4 2 0 359 77 483 0 385 7 20 8 1997 475 6 8 4 2 0 347 7 182 8 405 5 11 1 1998 L219 8 6 2 0 380 3 340 7 405 7 PEL 1999 524 3 8 1 2 0 332 7 199 9 376 5 0 0 2000 869 4 8 4 22 0 399 7 483 0 385 6 S2 0in8 lt END EXAMPLE gt With just a little bit of simple programming the formatting of this file can be changed to include whatever variables might be of interest if the binary output options are not feasible RUNNING THE EXAMPLES This release of the Biome BGC code comes with a simple set of example input and output files that illustrate the recommended procedure for running a simulation at a new site The example simulation is for a site in Missoula MT using 44 years of meteorological data as processed by the MTCLIM program The vegetation type is assumed to be a generic evergreen needleleaf forest Instructions are given below for both PC and UNIX users The first step is to run a spinup simulation that will start with very low initial levels of soil carbon and nitrogen and loop through the 44 year meteorological data file many times until the total carbon levels stabilize This simulation does not use a restart input file but it produces a restart output file for use in the next step The second step is to run a normal simulation which runs for one pass through the 44 year met data file To run the spinup simulation UNIX From the unixbgc411 directory enter the following text at the command line and press enter bgc411 ini enf
9. 1996b Simulating effects of fire on northern Rocky Mountain landscapes with the ecological process model FIRE BGC Tree Physiology 16 319 331 Kimball J S Thornton P E White M A and Running S W 1997a Simulating forest productivity and surface atmosphere exchange in the BOREAS study region Tree Physiology 17 589 599 Kimball J S White M A and Running S W 1997b BIOME BGC simulations of stand hydrologic processes for BOREAS Journal of Geophysical Research 102 D24 29043 29051 Mahfouf J F et al 1996 Analysis of transpiration results from the RICE and PILPS workshop Global and Planetary Change 13 73 88 Parton W J Haxeltine A Thornton P Anne R and Hartman M 1996 Ecosystem sensitivity to land surface models and leaf area index Global and Planetary Change 13 89 98 Pierce L L and Running S W 1995 The effects of aggregating sub grid land surface variation on large scale estimates of net primary production Landscape Ecology 10 4 239 253 Pierce L L Running S W and Walker J 1994 Regional scale relationships of leaf area index to specific leaf area and leaf nitrogen content Ecological Applications 4 2 313 321 Pierce L L et al 1993 Ecohydrological changes in the Murray Darling Basin HI A simulation of regional hydrological changes Journal of Applied Ecology 30 283 294 Running S W 1994 Testing FOREST BGC ecosystem process simulations across a climatic gradient in Orego
10. 5 July 2000 Author Peter E Thornton User s Guide for Biome BGC Version 4 1 1 Source Numerical Terradynamic Simulation Group School of Forestry University of Montana Missoula MT 59812 USA General notes on Biome BGC version 4 1 1 Biome BGC is a computer model that simulates the storage and fluxes of water carbon and nitrogen within the vegetation litter and soil components of a terrestrial ecosystem Biome BGC is primarily a research tool and many versions have been developed for particular purposes The Numerical Terradynamic Simulation Group NTSG maintains benchmark code versions for public release and updates these benchmark versions periodically as new knowledge is gained on the research front The code and executables accompanying this file represent the most recent update to the benchmark version IMPORTANT NOTE COPYING The Biome BGC version 4 1 1 code is copyrighted You may NOT make copies of any part of the code for distribution to any other person or group However ANYONE can get a FREE copy of the code from the NTSG website www forestry umt edu ntsg go to the Models to Download section The reason for this restriction is that we want to keep track of who has what version of the public release benchmark code so that we can let our user community know when there are updates We appreciate your cooperation with this policy If you use Biome BGC in your research we request that you include the following acknowledgeme
11. ANNUAL_OUTPUT sections of the ini file lt EXAMPLE gt OUTPUT_CONTROL outputs enf_testl keyword do not remove text prefix for output files A flag 1 write daily output 0 no daily output 1 flag 1 monthly avg of daily variables 0 no monthly avg 1 flag 1 annual avg of daily variables 0 no annual avg il flag 1 write annual output 0 no annual output 1 flag for on screen progress indicator DAILY_OUTPUT keyword 3 int number of daily variables to output 43 5 wf soilw_trans 509 8 epv proj_lai 620 10 summary daily_npp ANNUAL_OUTPUT keyword 2 int number of annual output variables 545 0 annual maximum projected LAI 636 1 vegetation C lt END EXAMPLE gt Four binary output files will be created for this simulation and each file can be considered as a multi dimensional array of values The first dimension in each output file is the number of variables 3 in the case of daily variables 2 in the case of annual variables The last dimension is the number of years of simulation 100 in this case Daily and monthly files have another dimension which is either the number of days per year 365 or the number of months per year 12 The values in parentheses beside the filenames below show the dimensions for each file in this example outputs enf_test1 dayout 3 x 365 x 100 outputs enf_test1 monavgout 3 x 12 x 100 outputs enf_testl annavgout 3 x 100 outputs enf_test1 annout 2 x 100 Values a
12. N Steven W Running 1981 DAYTRANS Steven W Running 1975 Others who have contributed to the code in various ways and at various stages include Galina Churkina Tom Gower Kathy Hibbard Bob Keane John Kimball Lars Pierce Joseph White and Michael White Some relevant references Churkina G and Running S W 1998 Contrasting climatic controls on the estimated productivity of global terrestrial biomes Ecosystems 1 206 215 Cienciala E Running S W Lindroth A Grelle A and Ryan M G 1998 Analysis of carbon and water fluxes from the NOPEX boreal forest comparison of measurements with FOREST BGC simulations Journal of Hydrology 212 213 62 78 Hunt E R Jr et al 1996 Global net carbon exchange and intra annual atmospheric CO concentrations predicted by an ecosystem process model and three dimensional atmospheric transport model Global Biogeochemical Cycles 10 3 431 456 Hunt E R Jr and Running S W 1992 Simulated dry matter yields for aspen and spruce stands in the north american boreal forest Canadian Journal of Remote Sensing 18 3 126 133 Keane R E Morgan P and Running S W 1996a FIRE BGC A mechanistic ecological process model for simulating fire succession on coniferous forest landscapes of the Northern Rocky Mountains Res Pap INT RP 484 U S Department of Agriculture Forest Service Intermountain Research Station Ogden UT 122 PP Keane R E Ryan K C and Running S W
13. PREVIOUS VERSION version 4 1 This code release version 4 1 1 represents a relatively minor update from version 4 1 which was released in February 2000 Recent research efforts with Biome BGC have concentrated on comparisons of model estimates of carbon and water fluxes with fluxes measured by eddy flux methods at a number of sites in the U S and in Europe Through these comparisons a number of model process representations were identified that did not agree with the flux measurements Most of the changes from version 4 1 to version 4 1 1 are the result of these comparisons The changes are mostly in the handling of C and N allocation across multiple years for example in the deployment of N retranslocated from senescing leaves during subsequent growing seasons These changes represent the philosophy of simplest possible process representation in operation The original mechanisms version 4 1 were based on the simplest possible treatment in the absence of data to support more sophisticated mechanisms Now that several years of flux data are available from multiple sites we can see that the simple mechanisms capture the interannual dynamics but err in the representation of the seasonal dynamics under some circumstances On this basis the next logical steps toward increasing complexity in the process representation were added resulting in a much better representation of the seasonal cycles of fluxes without impairing the interannual behavior
14. TROL keyword do not remove 0 flag O constant l vary with file 2 constant file for Ndep 294 842 ppm constant atmospheric CO2 concentration XXXXXXXXXXX file annual variable CO2 filename lt END EXAMPLE gt NOTES When line 1 is set to 0 or 2 then the value on line 2 sets the constant CO2 level for the entire simulation When line 1 is set to 1 then the file named on line 3 is used to define the annual timeseries of CO2 concentration This file must have one line for each simulation year the number given on line 2 of the TIME_DEFINE section and the format of each line should be like this 1895 294 842 where the first value on the line is the year and the second value is the CO2 mole fraction ppm When line 1 is set to 2 the value on line 2 is used for the constant CO2 level for the entire simulation but the file on line 3 is used to control the time sequence of changes in N deposition SITE block The next section begins with the keyword SITE It has the following nine lines 1 rooting zone soil depth after accounting for the fraction of the rooting zone occupied by rocks 2 soil texture percent sand by volume in rock free soil 3 soil texture percent silt by volume in rock free soil 4 soil texture percent clay by volume in rock free soil 5 site elevation in meters above mean sea level 6 site latitude in decimal degrees negative values for southern hemisphere 7 site shortwave albedo 8 annual rat
15. University of Montana Missoula MT 59812 E mail peter ntsg umt edu
16. _testl_spinup ini PC In the pcbgc411 directory double click the icon for bgc411 bat NOT the bgc411 exe icon This opens a window that contains some text asking you to enter the name of an initialization file Enter the following line of text and press enter ini enf_testl_spinup ini UNIX and PC At this point the program is launched and you should see a progression of numbers scrolling down the left side of the screen or window These record the number of years of simulation that have been completed This particular spinup run should take 1584 years to reach a steady state after which the following message will be displayed and the program will exit SPINUP residual trend 0 000990 SPINUP number of years 1584 This indicates that the steady state requirement was met after 1584 years and that there is still a slight trend in the total system carbon in this case an increase of 0 000990 kg C m yr The spinup run does not produce any output files except a restart file that is stored in the restart subdirectory This is used in the next step to initialize a normal simulation that will produce a range of output files To run the 44 year simulation UNIX Again from the unixbgc411 directory enter the following line of text and press enter bgc411 ini enf_testl ini PC In the pebgc411 directory double click the icon for bgc411 bat NOT the bgc411 exe icon This opens a window which contains some text asking you to enter the
17. al pool lt END EXAMPLE gt NOTES As with W_STATE and C_STATE the N_STATE values are ignored when using an input restart file For the reasons outlineed above it is recommended that new simulations be started with a spinup run and then continued from the end of the spinup run with a restart file In the case of a spinup run it is recommended that the N_STATE values both be set to 0 0 as in the example shown here OUTPUT_CONTROL block The next section begins with the keyword OUTPUT_CONTROL It has the following 6 lines 1 text string giving the prefix for all output files including path if appropriate 2 flag 1 or 0 to write 1 or not write 0 a binary output file with daily values 3 flag 1 or 0 to write 1 or not write 0 a binary output file with monthly averages of the daily variables 4 flag 1 or 0 to write 1 or not write 0 a binary output file with annual averages of the daily variables 5 flag 1 or 0 to write 1 or not write 0 a binary output file with year end values 6 flag 1 or 0 to send 1 or not send 0 simultation progress information to the screen lt EXAMPLE gt OUTPUT_CONTROL keyword do not remove outputs test text prefix for output files 1 flag 1 write daily output 0 no daily output 0 flag 1 monthly avg of daily variables 0 no monthly avg 0 flag 1 annual avg of daily variables 0 no annual avg 0 flag 1 write annual output 0 no annual output 1 flag for on scr
18. and moderate temperatures Reciprocal of minimum stomatal resistance line 39 m s cuticular conductance projected area basis The conductance of the leaf cuticle to water vapor expressed on a projected area basis Assumed constant under all environmental conditions line 40 m s boundary layer conductance projected area basis Leaf boundary layer conductance to water vapor expressed on a projected area basis This is also refered to as the aerodynamic conductance It is defined as the conductance for water vapor entering the atmosphere from a free water surface on the leaf surface a raindrop on the leaf It is assumed constant under all environmental conditions although it is in reality a strong function of wind speed A constant windspeed of m s is assumed in defining values of this parameter for various leaf morphologies lines 41 and 42 MPa leaf water potential start of conductance reduction MPa leaf water potential complete conductance reduction These two parameters set the endpoints for a linear control on stomatal conductance due to leaf water potential The first parameter sets the leaf water potential at which conductance reduction begins and the second parameter sets the water potential at which stomatal conductance is reduced to 0 0 In the range between these values reduction of stomatal conductance is a linear function of leaf water potential lines 43 and 44 Pa vapor pressure deficit start of conducta
19. be invoked and 0 indicates that the C4 model should be invoked Although this flag can be set to 0 for any combination of the other parameters use of the C4 model should be restricted to grasses and herbaceous plants line 5 flag 1 MODEL PHENOLOGY O USER SPECIFIED PHENOLOGY An integer flag specifying how the phenological control for a simulation will be exercised A value of 1 invokes the internal phenology routine while a value of 0 indicates that the user will supply information on the yeardays for the start of new growth and the end of the litterfall period See below for more details on how to set these parameters for the case of user specified phenology lines 6 and 7 yday yearday to start new growth when phenology flag 0 yday yearday to end litterfall when phenology flag 0 Two integers specifying the yearday for the start of new leaf growth and the yearday for the end of the litterfall season respectively Relevant only when the phenology flag 0 see above There are several IMPORTANT NOTES about setting these values 1 To suppress new leaf growth entirely for example in the case of a simulation concerned with bare soil processes set both of these values to 1 2 Yearday values for these parameters start at 0 and go to 364 Note that BIOME BGC does not accept leap years i e all years are by definition 365 days long 3 Northern and Southern hemisphere yeardays are treated diff
20. cification line is ignored by the program It is recommended that as you add or delete output variable index values you also add a descriptive comment reminding you what variable is being requested The variables requested in this section are available for daily output for monthly average output and for annual average output specified by the flags in the OUTPUT_CONTROL section ANNUAL_OUTPUT block The next section begins with the keyword ANNUAL_OUTPUT It has exactly the same format as just described for the DAILY_OUTPUT section lt EXAMPLE gt ANNUAL_OUTPUT keyword 6 int number of annual output variables 545 0 annual maximum projected LAI 636 1 vegetation C 637 2 litter C 638 3 soil C 639 4 total C 307 5 soil mineral N lt END EXAMPLE gt NOTES The variables requested in this section are reported once each year yearday 365 and are stored in the annout file This provides a once yearly snapshot of the system state and activity and so it is most appropriate for recording system states that are changing relatively slowly such as soil carbon vegetation carbon etc Remember that if annual averages of system behavior or system state are desired then it is necessary to specify these variables in the DAILY_OUTPUT section and set the annual averaging of daily output flag to 1 line 4 in OUTPUT_CONTROL section END_INIT block The final section of the ini file consists only of the keyword END_INIT signaling the end of th
21. during the first simulation year 2 peak stem carbon to be attained during the first year 3 initial coarse woody debris carbon dead trees standing or fallen 4 initial litter carbon labile pool 5 initial litter carbon unshielded cellulose pool 6 initial litter carbon shielded cellulose pool 7 initial litter carbon lignin pool 8 soil carbon fast pool 9 soil carbon medium pool 10 soil carbon slow pool 11 soil carbon slowest pool lt EXAMPLE gt C_STATE keyword start of carbon state variable initialization block 0 001 kgC m2 first year maximum leaf carbon 0 0 kgC m2 first year maximum stem carbon 0 0 kgC m2 coarse woody debris carbon 0 0 kgC m2 litter carbon labile pool 0 0 kgC m2 litter carbon unshielded cellulose pool 0 0 kgC m2 litter carbon shielded cellulose pool 0 0 kgC m2 litter carbon lignin pool 0 0 kgC m2 soil carbon fast microbial recycling pool 0 80 kgC m2 soil carbon medium microbial recycling pool 0 0 kgC m2 soil carbon slow microbial recycling pool 0 0 kgC m2 soil carbon recalcitrant SOM slowest lt END EXAMPLE gt NOTES As with the W_STATE section these values are ignored when using an input restart file since all the initial conditions are defined by the endpoint from the previous simulation usually a spinup run In starting a spinup run these values have to be supplied The above example shows the bes
22. e file This signal is used to make sure that the proper number of lines have been read from the ini file lt EXAMPLE gt END_INIT keyword indicates the end of the initialization file lt END EXAMPLE gt METEOROLOGICAL DATA FILE DETAILS As described above the MET_INPUT section of the ini file names a file containing the meteorological data used to drive a Biome BGC simulation This file can contain any number of header lines followed by numeric values for the meteorological data The number of header lines must be specified after the filename in the MET_INPUT section of the ini file Below is an example of the top of a meteorological data file having four header lines lt EXAMPLE gt Missoula 1950 1993 MTCLIM v4 3 OUTPUT FILE Mon Feb 7 10 15 00 2000 year yday Tmax Tmin Tday prcp VPD srad daylen deg C deg C deg C cm Pa W m 2 s 1950 1 Z309 0u T3 390 6 65 0 10 158 19 123 31 30229 1950 2 780 21 00 1s 62 0 00 136 27 183 78 30284 1950 3 Shoe 23 3 0 L808 0 00 D336 140 67 30344 1950 4 11 70 20 60 14 15 0 10 83 07 119 72 30408 1950 5 713 90 25 00 16 95 0 00 78 17 177 64 30476 1950 6 0 60 14 40 3 03 0 10 264 00 142 03 30549 1950 7 4 40 4 40 1 98 0 00 2973 158 19 30626 1950 8 2 20 11 70 4 81 0 10 182 84 126 15 30707 lt END EXAMPLE gt In this case the meteorological input file is one generated by the MTCLIM version 4 3 program as indicated in the header lines Meteo
23. e of atmospheric nitrogen deposition wet dry deposition 9 annual rate of symbiotic asymbiotic nitrogen fixation lt EXAMPLE gt SITE keyword start of site physical constants block 1 0 m effective soil depth corrected for rock fraction 30 0 3 sand percentage by volume in rock free soil 50 0 3 silt percentage by volume in rock free soil 20 0 3 clay percentage by volume in rock free soil 977 0 m site elevation 46 8 degrees site latitude for S Hem 0 2 DIM site shortwave albedo 0 0001 kgN m2 yr wet dry atmospheric deposition of N 0 0008 kgN m2 yr symbiotictasymbiotic fixation of N lt END EXAMPLE gt NOTES The values on lines 2 3 and 4 must sum exactly to 100 RAMP_NDEP block The next section begins with the keyword RAMP_NDEP It has the following three lines 1 flag 1 or 0 for variable nitrogen deposition 1 or constant nitrogen depostition 0 2 the reference year for industrial nitrogen deposition value 3 the reference value for industrial nitrogen deposition lt EXAMPLE gt RAMP_NDEP keyword do not remove 0 flag do a ramped N deposition run O no l yes 2099 int reference year for industrial N deposition 0 0001 kgN m2 yr industrial N deposition value lt END EXAMPLE gt NOTES when line 1 is set to 0 the nitrogen deposition level is constant and is determined by line 8 of the SITE section When line 1 is set to 1 the nitrogen deposition levels vary according to th
24. e soil carbon and nitrogen components for a site without adequate measurements of these characteristics Users need only know the names of input and output restart files and need not be concerned with the file contents INITIALIZATION FILE DETAILS There are several example initialization files called ini files for short in the ini directory of this code release and you should refer to one while reading the following discussion The initialization file is broken into sections with a keyword starting each new section and a special keyword at the bottom of the file This organization helps to ensure that the proper format for the initialization file is maintained while editing for new simulations The order of the sections and the order of the lines within each section is critical and changes in order will result in failed or flawed executions It is highly recommended that the example ini files provided with this release be copied to a safe directory for future reference on proper formatting There is a fair amount of error checking on the ini file format but it is still possible to scramble the order of lines in a way that the program does not detect In this case the model will run but the results will be garbage Copying the template is the easiest way to make a new ini file with assurance of the correct format but remember to replace ALL of the parameters with those for the new site It is not necessary to have blank lines between the end of
25. e time trajectory of CO2 mole fractions using two reference values to scale N deposition level to CO2 concentration For the case of varying N deposition the value on line 8 of the SITE section is taken as the N deposition level in the first simulation year and the value specified on line 3 of this section is taken as the N deposition level in the year specified on line 2 of this section In this case the values for CO2 concentration given in the file listed in line 3 of the CO2_CONTROL section are used to scale the N deposition value for each year according to changes in the CO2 concentration The assumption here is that N deposition increases as the anthropogenic CO2 source increases This assumption has some obvious limitations but it can provide a plausible time sequence of N deposition when only a historical and a current value are available EPC_FILE block The next section begins with the keyword EPC_FILE It has a single line 1 the name of the input ecophysiological constants file lt EXAMPLE gt EPC_FILE keyword do not remove epc enf epc file evergreen needleleaf forest lt END EXAMPLE gt NOTES There is a default set of ecophysiological parameter files supplied with this release of the Biome BGC code which has been developed through extensive literature searches and statistical summarization Details on the format and interpretation of these files are given in a later section Although there is a lot of careful work beh
26. een progress indicator lt END EXAMPLE gt NOTES As described below the user can select two different groups of variables for output one group that will be used for daily output and the monthly and annual averages of daily values and a second group that is used for writing year end day 365 values for each simulation year The flag for daily output on line 2 does not need to be set to 1 in order to get monthly and or annual averages of the daily output variables Setting any of the output flags to 1 lines 2 5 will result in the creation of an output file having the specified path and filename prefix from line 1 and the following suffix depending on which type of output was requested daily output suffix dayout monthly average output suffix monavgout annual average output suffix annavgout year end output suffix annout DAILY_OUTPUT block The next section begins with the keyword DAILY_OUTPUT The number of lines can vary depending on the number of output variables requested as follows 1 the number of daily output variables requested This value can be 0 2 the index number for the first requested daily output variable 3 etc lt EXAMPLE gt DAILY_OUTPUT keyword 3 int number of daily variables to output 43 wf soilw_trans 509 epv proj_lai 620 summary daily_npp lt END EXAMPLE gt NOTES The number of lines after line 1 must equal the number of daily output variables specified on line 1 There a
27. er for that vegetation type and any numeric value can be substituted in these places without effect The following section describes each line of an epc file in the required order The line number is followed by the units for the parameter in question and the short text description included in the default epc files some parameters are dimensionless and these units are given as DIM This is followed by a detailed description of the parameter s in question line 1 ECOPHYS keyword start of canopy ecophysiological constants block This is a keyword used by the code to interpret the start of a block of ecophysiological data The first line of each epc file must start with ECOPHYS line 2 flag 1 WOODY 0 NON WOODY An integer flag specifying the growth form where 1 for woody includes both tree and shrub vegetation types and 0 for non woody includes grasses as well as other primarily herbaceous plants line 3 flag 1 EVERGREEN 0 DECIDUOUS An integer flag specifying the leaf habit where 1 for evergreen includes leaf habits that retain at least some of their foliage year round and 0 for deciduous includes leaf habits in which all foliage is absent at some point during a year Either value for this flag can apply to both woody and non woody growth forms line 4 flag 1 C3 PSN 0 C4 PSN An integer flag specifying the photosynthetic pathway where indicates that the C3 photosynthesis model should
28. er labile and cellulose fractions are removed as described above Labile cellulose and lignin fractions for leaf litter must sum to 1 0 line 27 DIM fine root labile proportion The proportion of fine root mass in the labile fraction usually defined as that fraction soluble in hot water alcohol Labile cellulose and lignin fractions for fine root must sum to 1 0 line 28 DIM fine root cellulose proportion The proportion of fine root mass in the cellulose fraction usually defined as that fraction soluble in a mild acid solution after extraction of the water alcohol soluble fraction Labile cellulose and lignin fractions for fine root must sum to 1 0 line 29 DIM fine root lignin proportion The proportion of fine root mass in the lignin fraction usually defined as the remaining fraction after labile and cellulose fractions are removed as described above Labile cellulose and lignin fractions for fine root must sum to 1 0 line 30 DIM dead wood cellulose proportion The proportion of dead wood mass in the cellulose fraction usually defined as that fraction soluble in a mild acid solution after extraction of the water alcohol soluble fraction Cellulose and lignin fractions for dead wood must sum to 1 0 line 31 DIM dead wood lignin proportion The proportion of dead wood mass in the lignin fraction usually defined as the remaining fraction after labile and cellulose fractions are removed as described above Cel
29. erently for these parameters In the northern hemisphere yearday 0 Jan 1 while in the southern hemisphere yearday 0 July 2 This allows the same yearday values to be used to specify deciduous growth habit in the northern and southern temperate zones 4 If the leaf habit flag is set to evergreen 1 and the phenology model flag is set to user specified 0 these values do not have any effect lines 8 and 9 prop transfer growth period as fraction of growing season prop litterfall as fraction of growing season These two parameters determine the duration of the transfer growth and litterfall periods and are defined as proportions of the period between the start of new growth and the end of litterfall These parameters must be set by the user REGARDLESS of whether internal model phenology or user specified phenology is specified in the phenology model flag These parameters can take any values from 0 0 to 1 0 where a value of 0 0 indicates that all transfer growth or all litterfall occurs in a single day and a value of 1 0 indicates that transfer growth or litterfall occur throughout the growing season Transfer growth is the growth derived from carbon and nitrogen resources stored over the course of the previous growing season It is the growth that produces the first flush of new leaves in the spring for deciduous plants NOTE that when the leaf habit flag is set for evergreen 1 both transfer growth and litterfall are assumed to occur a
30. he sum of phloem and xylem with heartwood defined as the non conducting woody tissue The current treatment ignores the distinction between water conducting xylem and non conducting heartwood line 12 1 yr annual whole plant mortality fraction This parameter specifies the fraction of all plant pools that will be removed and sent to the litter compartments over the course of a year This is one mechanism by which woody material live and dead leaves the plant pool and enters the litter pools to be made available for subsequent decomposition It is the conceptual equivalent of wind throw since all plant pools living and dead are attenuated at the same rate This annual proportion is converted internally to a daily rate and whole plant mortality is assumed to go on at a constant rate throughout the year line 13 1 yr annual fire mortality fraction This parameter specifies the fraction of plant pools subject to fire on average each year The current treatment ignores the timing of individual fire events taking a long term view of the fire process in which some fraction of the community is subject to fire each year at a rate commensurate with the long term fire frequency For example in a system with a stand replacing fire return interval of 100 yrs this parameter would be set to 0 01 line 14 ratio ALLOCATION new fine root C new leaf C Sets the ratio of new fine root growth to new leaf growth This is a constant and applies
31. ind the default values that are supplied with the code these values are intentionally generalized to represent very coarse distinctions between vegetation types If you have reliable data for a particular species or functional type on or near your site you should use it to modify the default values that we have supplied W_STATE block The next section begins with the keyword W_STATE It has the following two lines 1 initial snowpack water content start of simulation 2 initial soil water content as a proportion of saturation lt EXAMPLE gt W_STATE keyword start of water state variable initialization block 0 0 kg m2 water stored in snowpack 0 5 DIM initial soil water as a proportion of saturation lt END EXAMPLE gt NOTES When using a restart file these values are ignored Otherwise they set the initial conditions for the water state variables storage components on the first day of simulation Line 1 sets the snowpack and is in water equivalent units where kg water m2 is equivalent to mm of water The second line controls the initial soil water content This is set as a proportion of saturation so that the user is not required to know the saturation water content of the site depends on texture and depth For a spinup run these values are used as the initial conditions C_STATE block The next section begins with the keyword C_STATE It has the following eleven lines 1 peak leaf carbon to be attained
32. l executable supplied with this release so that the original is not overwritten 5 In order to protect the code from accidental changes issue the protect_on command from the src directory after the new executable has been created PC compilation For the PC model release all of the source code is located in the src directory The executable provided with this release was generated by compiling the source code with the Microsoft Visual C C compiler version 6 0 All the project files used by MS Visual C C are located in the subdirectory src bgc411 and users with the Visual C C software installed can open the bgc411 dsw file to access this project Selecting rebuild all from the Build menu will recompile the code putting a new executable called bgc411 exe in the src bgc41 1 Release directory There are many other compilers to choose from contact your local PC system manager for more information Code History The Biome BGC ecosystem process model has a long heritage and many people have contributed extensively to its development over many years The following is a short synopsis of some of the most prominent code versions preceeding the current version 4 1 1 benchmark code for public release Biome BGC version 4 1 Peter E Thornton 2000 C C Biome BGC version 2 0 CRB BGC Peter E Thornton 1995 C C Biome BGC version 1 37 E Raymond Hunt Jr 1993 Pascal Forest BGC Joseph C Coughlan 1986 Pascal DAYTRANS PS
33. l of nitrogen from the leaves prior to litterfall The model does not consider the possibility of retranslocation of carbon out of leaves prior to litterfall line21 kgC kgN C N of fine roots Mass ratio of carbon nitrogen in fine roots The model assumes that there is no retranslocation of nitrogen out of fine roots prior to litterfall so this is the only C N parameter for fine roots line 22 kgC kgN C N of live wood Mass ratio of carbon nitrogen for live wood phloem and ray parenchyma This will typically be a much smaller value than the average C N for all woody parts and is typically found to be close to that for fine roots line 23 kgC kgN C N of dead wood Mass ratio of carbon nitrogen for dead wood bark xylem heartwood line 24 DIM leaf litter labile proportion The proportion of leaf litter mass in the labile fraction usually defined as that fraction soluble in hot water alcohol Labile cellulose and lignin fractions for leaf litter must sum to 1 0 line 25 DIM leaf litter cellulose proportion The proportion of leaf litter mass in the cellulose fraction usually defined as that fraction soluble in a mild acid solution after extraction of the water alcohol soluble fraction Labile cellulose and lignin fractions for leaf litter must sum to 1 0 line 26 DIM leaf litter lignin proportion The proportion of leaf litter mass in the lignin fraction usually defined as the remaining fraction aft
34. lulose and lignin fractions for dead wood must sum to 1 0 line 32 1 LAI d canopy water interception coefficient The proportion of daily rainfall that can be intercepted and retained on the canopy per unit of projected leaf area index line 33 DIM canopy light extinction coefficient The Beer s law extinction coefficient for attenuation of radiation in the canopy using a projected leaf area basis line 34 DIM all sided to projected leaf area ratio The ratio between the all sided area and the projected area for leaves Projected area for this and all other uses in BIOME BGC is the projected area of the leaf laid flat with its two longest dimensions parallel to the measurement surface while all sided area is the total leaf surface area line 35 m2 kgC canopy average specific leaf area projected area basis Projected area per unit of leaf carbon mass averaged over the canopy line 36 DIM ratio of shaded SLA sunlit SLA Ratio between specific leaf area for leaves in the shaded canopy fraction and specific leaf area for leaves in the sunlit canopy fraction line 37 DIM fraction of leaf N in Rubisco The fraction of total live leaf nitrogen occurring in the RuBisCO enzyme line 38 m s maximum stomatal conductance projected area basis The maximum stomatal conductance to water vapor expressed on a projected leaf area basis This is the conductance under saturating light low VPD leaf water potential near 0 0
35. n Ecological Applications 4 2 238 247 Running S W and Coughlan J C 1988 A general model of forest ecosystem processes for regional applications I Hydrological balance canopy gas exchange and primary production processes Ecological Modelling 42 125 154 Running S W and Gower S T 1991 FOREST BGC A general model of forest ecosystem processes for regional applications II Dynamic carbon allocation and nitrogen budgets Tree Physiology 9 147 160 Running S W and Hunt E R Jr 1993 Generalization of a forest ecosystem process model for other biomes BIOME BGC and an applicationfor global scale models In J R Ehleringer and C Field Editors Scaling Physiological Processes Leaf to Globe Academic Press San Diego CA pp 141 158 Running S W et al 1989 Mapping regional forest evapotranspiration and photosynthesis by coupling satellite data with ecosystem simlation Ecology 70 4 1090 1101 Schimel D S VEMAP Participants and Braswell B H 1997 Continental scale variability in ecosystem processes Models data and the role of disturbance Ecological Monographs 67 251 271 Thornton P E 1998 Regional ecosystem simulation combining surface and satellite based observations to study linkages between terrestrial energy and mass budgets Ph D Thesis The University of Montana Missoula 280 pp White J D et al 1998 Assessing simulated ecosystem processes for climate variability research at Glacier Nati
36. nce reduction Pa vapor pressure deficit complete conductance reduction These two parameters set the endpoints for a linear control on stomatal conductance due to the water vapor pressure difference VPD between the interior of the leaf and the air adjacent to the leaf The first parameter sets the VPD at which conductance reduction begins and the second parameter sets the VPD at which stomatal conductance is reduced to 0 0 In the range between these values reduction of stomatal conductance is a linear function of VPD OUTPUT FILES OVERVIEW There are two different styles of output produced by Biome BGC The first type includes the binary output files controlled through the initialization file with information from the OUTPUT_CONTROL DAILY_OUTPUT and ANNUAL_OUTPUT sections This is the most flexible output mechanism since the user can control exactly which model variables to include in the output files and what level of averaging to perform The second type of output is a very simple formatted text file that contains annual summary information for each year of the simulation This text output file is produced for all non spinup simulations and it uses a fixed list of output variables These two output types are described in greater detail below BINARY OUTPUT FILES DETAILS These are not text files and you will not be able to read them with a text editor or word processor Instead they are data files containing binary representations of
37. nt in the relevant manuscripts Biome BGC version 4 1 1 was provided by the Numerical Terradynamic Simulation Group NTSG at the University of Montana NTSG assumes no responsibility for the proper use of Biome BGC by others If you have made any significant modifications to the code please mention them in your manuscript This User s Guide is the only documentation provided with this release of Biome BGC Various model components have been described with varying degrees of detail in the literature and there is a list of relevant publications at the end of this guide The code itself contains extensive internal documentation and users with specific questions about the algorithms used to estimate particular processes should read the comments in the appropriate source code files The file bgc c contains references to all the core science routines and is an excellent starting point for this kind of inquiry The file bgc_struct h defines the data structures that are used to pass information between the process modules and includes both a short text description and the units for each internal variable We are always interested to get reports from users about new applications of Biome BGC including any problems that come up If you have questions about the code appropriate model applications possible programming errors etc please read this entire guide first then feel free to contact Dr Peter E Thornton peter ntsg umt edu UPDATES FROM
38. onal Park USA Ecological Applications 8 805 823 White M A 1999 Monitoring and modeling growing season dynamics Ph D Thesis University of Montana Missoula 231 pp White M A Running S W and Thornton P E 1999 The impact of growing season length variability on carbon assimilation and evapotranspiration over 88 years in the eastern US deciduous forest International Journal of Biometeorology 42 139 145 White M A Thornton P E and Running S W 1997 A continental phenology model for monitoring vegetation responses to interannual climatic variability Global Biogeochemical Cycles 11 2 217 234 White M A Thornton P E Running S W Nemani R R 2000 Parameterization and sensitivity analysis of the Biome BGC terrestrial ecosystem model net primary production controls Earth Interactions in press other references are listed online at www forestry umt edu ntsg Questions or comments If you have problems with the code or if you have suggestions on how it could be improved we d love to hear from you Also please send us copies of any manuscripts or reports that include the use of Biome BGC especially if you have compared Biome BGC estimates to measurements This information helps us to better understand model performance and improve it where we can For questions on how to parameterize or run the code please read everything in this file first Contact Dr Peter E Thornton NTSG School of Forestry
39. one section and the keyword for the next section but keeping them in makes the ini file more readable Each line can contain up to 100 characters of comment information after a keyword or after an input parameter This information is for your reference in keeping the correct format in the ini file and is ignored by the program Every line is required even if flags are set that mean the information on a line will be ignored The first line of the file is for header information that helps you keep track of which ini file is for which simulation when you are doing a sequence of simulations This line can contain up to 100 characters of information There is no keyword for this header line MET_INPUT block The next section begins with the keyword MET_INPUT It has the following two lines 1 name including path if appropriate of input meteorological data file 2 number of header lines in meteorological data file lt EXAMPLE gt MET_INPUT keyword start of meteorology file control block metdata miss5093 mtc41 meteorology input filename 4 int header lines in met file lt END EXAMPLE gt NOTES See the section of this guide titled METEOROLOGICAL DATA FILE DETAILS for more discussion of header lines and file format RESTART block The next section begins with the keyword RESTART It has the following five lines 1 flag 1 or 0 for reading 1 or not reading 0 a restart file from the end of a previous simulation 2 flag 1 or
40. or if you want to make any changes to the code you will need to recompile to get a new executable The process is different on PC and UNIX systems so two sets of instructions are given below UNIX compilation This code release comes with a set of makefiles that need some alteration before you use them The main makefile is in the subdirectory src and this should not have to be changed There are two other makefiles one in the src bgclib_v4 1 1 subdirectory and another in the src pointbgc_v4 1 1 subdirectory that both need to be modified as follows 1 Change the ROOTDIR variable so that it reflects the pathname for the src subdirectory For example if the src subdirectory is located at ul yourname bgc src then you would replace the current text ntsg peter pointbgc_v4 1 1 src with the new path 2 Change the name of the C compiler to a valid name for your system Do this by editing the line in both subdirectory makefiles that begins with CC replacing the text xlC with the name of your C C compiler Some possible examples for the compiler name are cc and gcc 3 From the sre directory issue the protect_user_off command which removes write protections 4 From the src directory issue the command make This executes the makefiles in both subdirectories and if successful produces a new executable called pointbgc in the parent directory to the src directory The name of the new executable is different from the origina
41. r files NOTE Biome BGC expects each year to have 365 days of data so leap years will have to be truncated by eliminating one day We suggest eliminating December 31 of a leap year since then the yearday numbering doesn t have to be adjusted We have found that this truncation has negligible effects on most simulations The rationale for requiring 365 day years is that the met data files are commonly of shorter duration than the intended simulation so they must be recycled enough times to match the number of requested simulation years Recycling is also essential in the spinup runs If the number of years in the met data file is not a multiple of four then the handling of leap days in the code becomes very tedious and introduces several layers of testing and manipulation between the input and the process algorithms We have accepted the small errors are introduced by eliminating one day from leap years in exchange for code that is clear and easier to maintain Meteorological data files can be assembled from observations if all of the required parameters have been measured for the site of interest They can also be generated using the MTCLIM program using only observations of temperature and precipitation The MTCLIM code and documentation is available from the NTSG website www forestry umt edu ntsg go to the section Models to Download ECOPHYSIOLOGICAL CONSTANTS FILE DETAILS This input file defines the ecophysiological characteristics of the
42. re more than 500 possible output variables and the use of an index is the simplest way to allow the user access to all of them without introducing complicated parsing routines that would tend to clutter the code The index value for each variable is defined in the file output_map_init c where variables are organized into logical groups located in directory src bgclib_v4 1 1 for UNIX release and in directory src for PC release For example all the carbon flux variables are listed together and all the water flux variables are listed together The organization of the output index mapping is guided by the data structures that are used to pass information among process modules within the code These structures are defined in the file bgc_struct h located in directory src include for UNIX release and directory src for PC release This indexing system is admittedly awkward for new users since it requires first knowing which data structure element to reference and then getting the index value for that element On the other hand it has the advantages of being unambiguous and providing a direct user interface to the logical organization of the data structures used in the code Consulting the output_map_init c and bgc_struct h gives a comprehensive view of the possible output variables and provides a useful introduction to the logical structure of process representation within Biome BGC As in all ini file lines all text after the first value on each output spe
43. re stored sequentially in the file as dictated by these dimensions As an example for the dayout file the first three values in the file are the three daily output variables for the first day of the first year of the simulation The next three values in the file will be the three daily output variables for the second day of the first year of simulation etc The units for all variables are defined in the bgc_struct h file ANNUAL TEXT OUTPUT DETAILS Because it can take some time to get used to the binary output formats we also include a formatted text output file with annual summary information For many applications this may be all the information required In any case it allows a quick look at the results before proceeding with more detailed analyses The name of the file is the user supplied output prefix plus the following suffix _ann txt The file is self documenting and an example is shown below lt EXAMPLE gt Annual summary output from Biome BGC version 4 1 1 ann PRCP annual total precipitation cm yr ann Tavg annual average air temperature deg C max LAI annual maximum value of projected leaf area index m2 m2 ann ET annual total evapotranspiration mm yr ann OF annual total outflow mm yr ann NPP annual total net primary production gC m2 yr ann NPB annual total net biome production gC m2 yr year ann PRCP ann Tavg max LAI ann ET ann OF ann NPP ann NBP
44. rological data files generated by MTCLIM version 4 3 will always have four header lines with the last two header lines describing the variables and their units This example also illustrates the required input variables their units and their order Biome BGC requires these variables using these units and in this order with no additional variables for every simulation The spacing between variables is not important as long as there is some white space between each value on a line Do not use commas or other non white space separators The nine values required for each day are as follows 1 year the numerical year repeated for each yearday 2 yday the numerical day of the year values must start with 1 and end with 365 3 Tmax the daily maximum temperature C 4 Tmin the daily minimum temperature C 5 Tday the average daytime temperature sunrise to sunset C 6 prcp the daily total precipitation cm 7 VPD the daylight average vapor pressure deficit Pa 8 srad the daylight average shortwave radiant flux density W m 9 daylen the daylength sunrise to sunset seconds The meteorological input data file can contain any number of years of data and each year must have exactly 365 days with one line of data per day and no separations between years This release of the code has an example met data file containing 44 years of daily weather values for Missoula MT which shows the correct formatting for multiple yea
45. see the comments in the header of the source file daily_allocation c 3 Accumulation of soil mineral N In version 4 1 the only explicit controls on the accumulation of mineral forms on N in the soil were due to leaching fire plant and microbial N uptake and a fractional loss to denitrification as part of the mineralization process Under non limiting N conditions mineral N could rise to unrealistic levels A very simple additional denitrification pathway was added in version 4 1 1 which removes a constant fraction of the mineral N remaining on each day after all the other processes have been reconciled For more details see the comments in the header of the source file daily_allocation c 4 Default combustion of dead woody material This change is related to the very simple treatment of fire effects included in version 4 1 5 of the dead non respiring stem material was assumed to be consumed in a fire This default has been changed to 20 See further comments in the header for the source code file mortality c 5 New variables in the default ASCII text output file Output for annual total precipitation and annual average air temperature have been added to the standard ascii text output file see further discussion later in this User s Guide INPUT FILES OVERVIEW Biome BGC uses at least three input files each time it is executed A brief description of all files is given first followed by detailed discussions of each file The firs
46. simulation 6000 int maximum number of spinup years if spinup simulation lt END EXAMPLE gt NOTES If line 2 is greater than line 1 then the meteorological data file is recycled enough times to satisfy the requested number of simulation years Line 3 defines the starting point for the simulation output and is mainly used in the simple text output file described below Line 4 sets the simulation mode for either a spinup run or a normal run If spinup mode is selected the meteorological data file will be recycled enough times to establish steady state conditions in the soil carbon and nitrogen pools The number of years of simulation for a spinup run will depend on the climate and vegetation characteristics but will not exceed the number of years specified on line 5 Line 2 has no effect for a spinup run If a spinup run reaches the maximum number of years specified in line 5 it is likely that the resulting final soil carbon and nitrogen pools are not equilibrated with the climate usually indicating a long term net sink of carbon This is observed for example in some boreal climates where the model predicts long term accumulation of organic matter peat formation CLIM_CHANGE block The next section begins with the keyword CLIM_CHANGE It has the following five lines 1 temperature offset for maximum temperature 2 temperature offset for minimum temperature 3 scaling factor for precipitation 4 scaling factor for vapor pressure
47. t approach to these initial conditions for a spinup run start with very low initial leaf area and no carbon in any of the other pools This is essentially a primary succession simulation starting with no organic matter and a very meager colonizing plant cover The development of soil organic matter depends on the site s climate and external fluxes of nitrogen deposition fixation leaching volatilization as well as on the vegetation type The end result is that soil and litter pools are in equilibrium with climate and N deposition loss rates If these values are assigned without performing a spinup run then the initial model dynamics are likely to represent a transient response to disequilibrium conditions between the specified initial conditions and the specified climate and N deposition loss This can produce very misleading signals in net ecosystem exchange of carbon that can persist for many hundereds of years For this reason it is recommended that new simulations be started with a spinup run then a restart from the spinup conditions as in the examples described later in this guide N_STATE block The next section begins with the keyword N_STATE It has the following two lines 1 litter nitrogen associated with labile litter carbon pool 2 soil mineral nitrogen pool lt EXAMPLE gt N_STATE keyword start of nitrogen state variable initialization block 0 0 kgN m2 litter nitrogen labile pool 0 0 kgN m2 soil nitrogen miner
48. t constant rates throughout the year and the specification of these two parameters has no effect line 10 1 yr annual leaf and fine root turnover fraction Determines leaf and fine root turnover for evergreen plants This is the fraction of the annual maximum leaf and fine root mass that will be dropped in the following year as litter It is the reciprocal of the leaf longevity so a plant that retains its leaves an average of two years would have a leaf fine root turnover of 0 5 Note that leaf and fine root phenology are assumed to be entirely synchronized for all vegetation types ALSO NOTE that when the leaf habit is specified as deciduous 0 this parameter is always assumed to be 1 0 and will be reset inside the code if the user specifies any other value line 11 1 yr annual live wood turnover fraction Determines livewood turnover to deadwood for all woody types deciduous and evergreen IMPORTANT NOTE about the definition of livewood and deadwood in BIOME BGC Livewood is defined as the actively respiring woody tissue that is the lateral sheathing meristem of phloem tissue plus any ray parenchyma extending radially into the xylem tissue Deadwood consists of all the other woody material including the heartwood the xylem and the bark It has been common in many tree models including previous versions of BIOME BGC to divide the woody tissue into two compartments called sapwood and heartwood where sapwood is usually defined as t
49. t required input file is called the initialization file It provides general information about the simulation including a description of the physical characteristics of the simulation site a description of the time frame for the simulation the names of all the other required input files the names for output files that will be generated and lists of variables to store in the output files The second required input file is the meteorological data file It contains daily values for temperature precipitation humidity radiation and daylength at the simulation site It can contain any number of years of data IMPORTANT NOTE LEAP YEARS The Biome BGC code assumes that all years have 365 days so meteorological data files should be edited to remove one day from leap years The third required input file is the ecophysiological constants file It contains an ecophysiological description of the vegetation at a site including parameters such as leaf C N ratio maximum stomatal conductance fire and non fire mortality frequencies and allocation ratios There is also an optional input file called a restart file This file provides all the information required to start one simulation from the endpoint of a previous simulation The results of the restarted simulation are exactly the same as if the original simulation had been carried out for additional years This option is typically used in conjunction with the spinup mode described below to initialize th
50. the values of the output variables Each value is written as a single precision IEEE floating point binary number using 4 bytes of storage per number These values can be read directly using simple code written in C C FORTRAN BASIC PASCAL and other programming languages They can also be read by many commercially available software packages typically those that specialize in manipulating large multi dimensional data sets One good example of a commercially available package for display and analysis of binary data files is IDL from Research Systems Inc www rsinc com The current cost is about 500 for a single PC Mac license and about 900 for a single workstation license U S educational institution pricing We will be making a package of simple plotting routines for IDL users available in the near future on the NTSG website There are lots of other possibilities for reading these files and you might want to consult with a computer programmer about the methods best suited for the kind of analysis you want to do In order to read the binary files you need to know how output values are ordered in the file Consider a single simulation over 100 years for which the user requested three daily output variables and two annual ouput variables and requested that in addition to the daily output the three variables also be summarized as monthly averages and annual averages Below is a corresponding example of the OUTPUT_CONTROL DAILY_OUTPUT and
51. to both current year growth and growth resources stored for the next year s transfer growth line 15 ratio ALLOCATION new stem C new leaf C Sets the ratio of new stem growth to new leaf growth Stem growth includes both the live and dead woody components as described above This ratio is constant and applies to both current year growth and growth resources stored for the next year s transfer growth line 16 ratio ALLOCATION new live wood C new total wood C Sets the ratio of new live wood to new total wood This ratio is constant and applies to both current year growth and growth resources stored for the next year s transfer growth line 17 ratio ALLOCATION new croot C new stem C Sets the ratio of new coarse root growth to new stem growth This ratio is constant and applies to both current year growth and growth resources stored for the next year s transfer growth line 18 prop ALLOCATION current growth proportion Sets the proportion of daily production that is displayed immediately as new growth with the remainder stored for the next year s transfer growth line 19 kgC kgN C N of leaves Mass ratio of carbon nitrogen in live leaves line 20 kgC kgN C N of leaf litter after retranslocation Mass ratio of carbon nitrogen in freshly fallen leaf litter This can only be higher than or equal to the C N for live leaves i e retranslocation can only be positive or zero Retranslocation is the remova
Download Pdf Manuals
Related Search