This file describes the procedures for creating transient land cover datasets. The data was created on Poorman (linux) at NCAR with pgf90 compiler Created by Johannes Feddema Department of Geography University of Kansas Lawrence KS 66045 feddema@ku.edu Datafile contents: Annual datafiles are stored in 10 time-stamped directories there are 2 nearly duplicate datasets. The difference between the CN version and the regular version is that the CN version converts the following PFTs in the tropics (-23.5 to 23.5): PFT 7 ==> PFT 6 PFT 10 ==> PFT 6 PFT 8 ==> PFT 13 PFT 11 ==> PFT 13 NOTE: GRAZING FIGURES NO LONGER APPLY WITH THE CN CONVERSION All these files are time-stamped at the directory level by yymmdd - historical_yymmdd - historical_CN_yymmdd within this directory files are names mksrf_pft_xxxx_hist.nc where xxxx is a year between 1700 and 1990 agriculture is based on Ramankutty and Foley (1999) grazing is based on Klein Goldewijk (2001) - Future_A2_yymmdd - Future_A2_CN_yymmdd within this directory files are names mksrf_pft_xxxx_A2.nc where xxxx is a year between 1990 and 2100 agriculture and grazing are based on IMAGE 2.2 (IMAGE-team, 2001) - Future_B1_yymmdd - Future_B1_CN_yymmdd within this directory files are names mksrf_pft_xxxx_B1.nc where xxxx is a year between 1990 and 2100 agriculture and grazing are based on IMAGE 2.2 (IMAGE-team, 2001) - Future_B2_yymmdd - Future_B2_CN_yymmdd within this directory files are names mksrf_pft_xxxx_B2.nc where xxxx is a year between 1990 and 2100 agriculture and grazing are based on IMAGE 2.2 (IMAGE-team, 2001) - Future_A1B_yymmdd - Future_A1B_yymmdd within this directory files are names mksrf_pft_xxxx_A1B.nc where xxxx is a year between 1990 and 2100 agriculture and grazing are based on IMAGE 2.2 (IMAGE-team, 2001) In each directory are netCDF files with the following information: Variables 1. LONGXY -two D array index same as in the original CCSM mksrf_pft.nc file 2. LATXY -two D array index same as in the original CCSM mksrf_pft.nc file 3. LANDMASK - 2 d array of half degree land mask values (100.0 for land) 4. PFT_PCT - 3 dimensional array (x,y,0:16) of percent pft in each grid cells 5. PCT_TREE_TOTAL - total percent of grid cells covered by trees (PFT 1-8) 6. PCT_GRASS_TOTAL - total percent covered by grass and shrubs (PFT 9-14) 7. TOT_GRASS_GRAZED - percent of PCT_GRASS grazed 8. PCT_PFT_GRAZED - 3D array (x,y,0:16) of each PFT that is grazed To create the files: The data are fists created as ASCII files and then converted to NetCDF. For this intermediate step several additional directories are required as follows: input: contains all the files required to build the transient pft datasets output: empty to begin with and stores all the ASCII pft files for all scenarios in_LAI: contains the CLM original ASCII monthly LAI files to reconstruct the transient LAI values out_LAI: empty to begin with and stores the intermediate monthly ASCII LAI files and the final mksrf_trans_LAI.nc file You will also need to create the following directories before you start - historical_yymmdd - Future_A2_yymmdd - Future_B1_yymmdd - Future_B2_yymmdd - Future_A1B_yymmdd - historical_CN_yymmdd - Future_A2_CN_yymmdd - Future_B1_CN_yymmdd - Future_B2_CN_yymmdd - Future_A1B_CN_yymmdd where yymmdd is the time stamp you wish to put on the dataset You only need to create the directories for the set of interest You will be prompted for the set to construct and the time-stamp 1. Run Pot_to_pft.f90 (OK) This creates creates a potential vegetation file using the Ramankutty and Foley (R&F) potential vegetation information along with the original CLM mksrf-igbp.5x.5 datafiles. Climate rules are applied as for the original dataset Input Files are: input/air_t - Willmott and Matsuura monthly average air temp. input/meas_precip - Willmott and Matsuura monthly average precip. input/pft-igbp.5x.5 - Original CLM3.0 surface dataset input/rf_pot_5min.asc - Ramankutty and Foley (R&F) 10 min. potential veg Output File: output/pft-pot.5x.5 - PFT Potential vegetation - no grazing info 2. Run init1990.f90 This modifies the original CLM pft data create grazing infromation and to use the R&F agriculture information Files required are: input/h1990.asc - ASCII raster HYDE 2.0 1990 land cover file input/glcrop1990_0_5.asc - ASCII raster R&F agriculture file for 1990 input/pft-igbp.5x.5 - Original CLM3.0 surface dataset output/pft-pot.5x.5 - PFT Potential vegetation Output File: output/pft-1990-hist.5x.5 - 1990 land cover ASCII format 3. Run hist.f90 This works back in time from 1990 to create pft files for dates which have both grazing and agricultural information Files required are: input/h1990.asc - ArcGIS ASCII raster HYDE 2.0 land cover file (graz) input/h1970.asc - ASCII raster HYDE 2.0 land cover file (graz) input/h1950.asc - ASCII raster HYDE 2.0 land cover file (graz) input/h1900.asc - ASCII raster HYDE 2.0 land cover file (graz) input/h1850.asc - ASCII raster HYDE 2.0 land cover file (graz) input/h1800.asc - ASCII raster HYDE 2.0 land cover file (graz) input/h1750.asc - ASCII raster HYDE 2.0 land cover file (graz) input/h1700.asc - ASCII raster HYDE 2.0 land cover file (graz) input/glcrop1990_0_5.asc - ASCII raster R&F agriculture file input/glcrop1970_0_5.asc - ASCII raster R&F agriculture file input/glcrop1950_0_5.asc - ASCII raster R&F agriculture file input/glcrop1900_0_5.asc - ASCII raster R&F agriculture file input/glcrop1850_0_5.asc - ASCII raster R&F agriculture file input/glcrop1800_0_5.asc - ASCII raster R&F agriculture file input/glcrop1750_0_5.asc - ASCII raster R&F agriculture file input/glcrop1700_0_5.asc - ASCII raster R&F agriculture file output/pft-pot.5x.5 - PFT Potential vegetation output/pft-1990-hist.5x.5 - 1990 land cover ASCII format Output Files: output/pft-1970-hist.5x.5 - land cover ASCII format output/pft-1950-hist.5x.5 - land cover ASCII format output/pft-1900-hist.5x.5 - land cover ASCII format output/pft-1850-hist.5x.5 - land cover ASCII format output/pft-1800-hist.5x.5 - land cover ASCII format output/pft-1750-hist.5x.5 - land cover ASCII format output/pft-1700-hist.5x.5 - land cover ASCII format 4. Run fill.f90 This fills in the gaps for dates where there is agricultural information, but no grazing. Grazing is interpolated linearly. Files required are: input/glcrop1930_0_5.asc ASCII raster R&F agriculture file input/glcrop1910_0_5.asc ASCII raster R&F agriculture file input/glcrop1890_0_5.asc ASCII raster R&F agriculture file input/glcrop1870_0_5.asc ASCII raster R&F agriculture file output/pft-1950-hist.5x.5 land cover ASCII format output/pft-1900-hist.5x.5 land cover ASCII format output/pft-1850-hist.5x.5 land cover ASCII format Output Files: output/pft-1930-hist.5x.5 land cover ASCII format output/pft-1910-hist.5x.5 land cover ASCII format output/pft-1890-hist.5x.5 land cover ASCII format output/pft-1870-hist.5x.5 land cover ASCII format 5. Run future.f90 Creates the ten year pft files from 2000 to 2100 for various IPCC SRES scenarios (A2, B1, B2, A1B) Files required are: input/worldnc.grd - IMAGE 2.2 grid location file input/landcovera2.dat - IMAGE 2.2 datafiles input/landcoverb1.dat - IMAGE 2.2 datafiles input/landcoverb2.dat - IMAGE 2.2 datafiles input/landcoverA1B.dat - IMAGE 2.2 datafiles input/gradient05.asc - GIS ASCII half degree average gradient file output/pft-pot.5x.5 - potential vegetation output/pft-1990-hist.5x.5 - 1990 land cover ASCII format Output Files: output/pft-19x0-yy(y).5x.5 - land cover ASCII format where x is decade years yy(y) is the scenario 2 or 3 letter id 6. Run interp.f90 Linearly interpolates all PFTs, grazing etc. between two dates. Files required are: output/pft-xxxx-yy(yy).5x.5 land cover ASCII format output/pft-yyyy-yy(yy).5x.5 land cover ASCII format where xxxx is the starting and ending years of the interpolation sequence where yy(yy) is the ending scenario indicator (hist or SRES) Output Files: output/pft-zzzz-yy(yy).5x.5 land cover ASCII format . . . output/pft-zzzz-yy(yy).5x.5 land cover ASCII format where zzzz represents all the years between the anchor dates 7. Run convert_nc.f90 -- use Makefile_convert_nc To convert the ASCII files to netCDF Files required are: output/pft-xxxx-yy(yy).5x.5 land cover ASCII format Output Files: aaaaa/mksrf_pft_xxxx.nc land cover netCDF file where aaaaa is the appropriate historical or future directory OR if the CN option is selected then aaa_CN_aa/mksrf_pft_xxxx_CN.nc land cover netCDF file where aaa_CN_aa is the appropriate historical or future time-stamped directory 8. run convert_pot_nc.F90 --- use Makefile_pot_nc converts the ASCII potential data to netCDF file Files required are: output/pft-pot.5x.5 -- ASCII potential vegetation file where xx is the month number 01 to 12 Output Files: historicalxxxxx/mksrf_pft_potv.nc potential vegetation netCDF file where xxxx is the time stamp there is no grazing, format is as the original igbp file for CLM OR historical_CN_xxxxx/mksrf_pft_potv.nc potential vegetation netCDF file where xxxx is the time stamp there is no grazing, format is as the original igbp file for CLM References: Ramankutty, N., and J.A. Foley (1999). Estimating historical changes in global land cover: croplands from 1700 to 1992, Global Biogeochemical Cycles 13(4), 997-1027. Klein Goldewijk, K. , 2001. Estimating global land use change over the past 300 years: the HYDE database. Global Biogeochemical Cycles, Vol 15(2): 417-434. IMAGE-team, 2001. The IMAGE 2.2 implementation of the SRES scenarios. A comprehensive analysis of emissions, climate change and impacts in the 21st century. RIVM CD-ROM publication 481508018, National Institute for Public Health and the Environment, Bilthoven, the Netherlands.