============================================================================== DATA BASE: DB1015 (3-1996) TITLE: Global Patterns of Carbon Dioxide Emissions from Soils on a 0.5 Degree Grid Cell Basis CONTRIBUTORS: James W. Raich Dept. of Botany, Iowa State University, Ames, IA 50011 USA Email: jraich@iastate.edu Christopher S. Potter NASA Ames Research Center, MS 242-4, Moffett Field, CA 94035 USA Email: cpotter@gaia.arc.nasa.gov ============================================================================== This db1015.doc (README) file contains: (I) Introduction (II) Abstract (Raich and Potter, 1995) (III) Documentation of the contents of the DB1015 files (IV) FORTRAN program to access the data (V) Description of data checks performed by CDIAC (VI) Instructions for obtaining the files (VII) References ============================================================================== (I) Introduction This database contains global, spatially explicit (0.5 degree grid cells) and temporally explicit (monthly and annual) model output of soil CO2 emissions. The calculated emissions include the respiration of both soil organisms and plant roots. A map of the soil emissions is available as a soil.ps and a soil.gif file. The model (model B) used for these calculations is described in Raich and Potter (1995). The model is based on a stepwise linear multiple regression of 977 individual aggregated records of geographically referenced data on daily and/or monthly rates of CO2 flux rates from intact soils, temperature, precipitation, potential evapotranspiration, vegetation type, soil moisture status, and organic carbon and nitrogen contents of soil. These data points represent grid cells from every continent except Antarctica. Temperature and soil moisture status, as predicted by precipitation, were the only variables significantly correlated with soil CO2 efflux. Temperature and precipitation were, therefore, used as driving variables. The remaining model parameters for the mechanistic aspect of the model are (1) the soil CO2 efflux rate when temperature is zero and moisture not limiting, (2) the temperature coefficient and (3) the half-saturation coefficient of the precipitation function. These were obtained by Raich and Potter (1995) using least squares estimates with both Quasi-Newton and Simplex methods as described in Wilkinson (1990). To prevent extrapolation of the model beyond the range of observed temperatures, soil CO2 fluxes were presumed to be zero at average monthly air temperatures less than -13.3 degrees Celsius and equal to the rate predicted at 33.5 degrees Celsius for all warmer temperatures. Raich and Potter (1995) evaluated their model predictions against previously published annual soil respiration rates from specific locations and with published predictions by an annual model (Raich and Schlesinger, 1992) through linear regressions of predicted versus observed (or previously published predicted) soil emissions. Correlation coefficients were close to 0.5, intercepts not significantly different from zero and slopes not significantly different from one, indicating that the reliability of the predictions was not diminished by the monthly timescale and that the model provides quantitatively meaningful estimates of annual soil emissions. ============================================================================== (II) Abstract (Raich and Potter, 1995) Global Patterns of Carbon Dioxide Emissions from Soils. Global Biogeochemical Cycles 9(1)23-36. Abstract. We use semi-mechanistic, empirically based statistical models to predict the spatial and temporal patterns of global carbon dioxide emissions from terrestrial soils. Emissions include the respiration of both soil organisms and plant roots. At the global scale, rates of soil CO2 efflux correlate significantly with temperature and precipitation; they do not correlate well with soil carbon pools, soil nitrogen pools, or soil C:N. Wetlands cover about 3% of the land area but diminish predicted CO2 emissions by only about 1%. The estimated annual flux of CO2 from soils to the atmosphere is estimated to be 76.5 Pg C yr-1, 1-9 Pg greater than previous global estimates, and 30-60% greater than terrestrial net primary productivity. Historic land cover changes are estimated to have reduced current annual soil CO2 emissions by 0.2-2.0 Pg C yr-1 in comparison with an undisturbed vegetation cover. Soil CO2 fluxes have a pronounced seasonal pattern in most locations, with maximum emissions coinciding with periods of active plant growth. Our models suggest that soils produce CO2 throughout the year and thereby contribute to the observed wintertime increases in atmospheric CO2 concentrations. Our derivation of statistically based estimates of soil CO2 emissions at a 0.5o latitude by longitude spatial and monthly temporal resolution represents the best-resolved estimate to date of global CO2 fluxes from soils and should facilitate investigations of net carbon exchanges between the atmosphere and terrestrial biosphere. (Reproduced with permission of the American Geophysical Union.) ============================================================================== (III) Documentation of the contents of the DB1015 files (IIIa) The 'rp-modelb.*' files: Thirteen global 0.5 degree lat/lon output files were produced by Model B as described in Raich and Potter (1995): 12 files for the monthly CO2 flux from soil; 1 file for the summed annual CO2 flux from soil. Predicted CO2 flux from soil is expressed in units (integer format) of 100*grams carbon per square meter of soil area. Note: (100*) conserves 2 decimal places for unit grams. (IIIb) The 'landmask' file: A 0.5 degree lat/lon file with grid cell area information on land/water distinctions, used by the authors for the global summation of the soil emissions. (IIIc) File formats: The files are in Ascii text format; Six line headers appear in all files, e.g.: north: -0.5 south: -359.5 east: 719.5 west: 0.5 rows: 360 cols: 720 The array structure is as follows: 360 rows (r) x 720 columns (c); Undefined and Ocean = 0; First (r,c) is centered at 89.75 deg. N, 179.75 deg. W; (r) increases southward; (c) increases eastward (IIId) Data file names and definitions: 'rp-modelb.jan' is predicted soil CO2 flux for january 'rp-modelb.feb' is predicted soil CO2 flux for february 'rp-modelb.mar' is predicted soil CO2 flux for march 'rp-modelb.apr' is predicted soil CO2 flux for april 'rp-modelb.may' is predicted soil CO2 flux for may 'rp-modelb.jun' is predicted soil CO2 flux for june 'rp-modelb.jul' is predicted soil CO2 flux for july 'rp-modelb.aug' is predicted soil CO2 flux for august 'rp-modelb.sep' is predicted soil CO2 flux for september 'rp-modelb.oct' is predicted soil CO2 flux for october 'rp-modelb.nov' is predicted soil CO2 flux for november 'rp-modelb.dec' is predicted soil CO2 flux for december 'rp-modelb.ann' is predicted annual soil CO2 flux 'landmask' has information if grid cell is land or water ============================================================================== (IV) FORTRAN program to access the data C FORTRAN 77 CODE TO (1) READ GRIDDED (0.5 degrees) GLOBAL SOIL EMISSIONS AND C LAND-MASK INFORMATION AND (2) LABEL GRID-CELLS WITH LATITUDE/LONGITUDE C LOCATIONS C CDIAC - AB - 3/96 C==== REAL LAT,LONG,LAND C C ARRAY SF CONTAINS THE RAICH AND POTTER MODELED ANNUAL SOIL C CO2 EMISSIONS. C ARRAY TSF CONTAINS THE RAICH AND POTTER MODELED MONTHLY SOIL C CO2 EMISSIONS. C ARRAY SSF CONTAINS THE ANNUAL SUM OF THE RAICH AND POTTER C MODELED MONTHLY SOIL CO2 EMISSIONS. C DIMENSION LAND(360,720),SF(360,720) DIMENSION SSF(360,720),TSF(12,360,720) C==== C CALCULATE GRID CELL AREAS WITH THE PROGRAM FROM THE GODDARD INSTITUTE C FOR SPACE STUDIES PARAMETER (IM=720,JM=360) REAL DXYP(JM),DXV(JM),COSP(JM) TWOPI=6.283185 RADIUS=6375000. DLON=TWOPI/IM DLAT=.5*TWOPI/(JM-1) FJEQ=.5*(JM+1) COSP(1)=0. COSP(JM)=0. DO 10 J=2,JM-1 10 COSP(J)=COS(DLAT*(J-FJEQ)) DO 20 J=2,JM 20 DXV(J)=.5*RADIUS*DLON*(COSP(J-1)+COSP(J)) DXYP(1)=.25*DXV(2)*RADIUS*DLAT DXYP(JM)=.25*DXV(JM)*RADIUS*DLAT DO 30 I=1,IM DO 30 J=2,JM-1 DXYP(J)=.5*(DXV(J)+DXV(J+1))*RADIUS*DLAT 30 CONTINUE C==== C FOR LAND/WATER DISTINCTION: OPEN(8,FILE='landmask',STATUS='OLD') C FOR YEARLY INPUT: OPEN(10,FILE='rp-modelb.ann',STATUS='OLD') C FOR MONTHLY INPUT: OPEN(12,FILE='rp-modelb.jan',STATUS='OLD') OPEN(14,FILE='rp-modelb.feb',STATUS='OLD') OPEN(16,FILE='rp-modelb.mar',STATUS='OLD') OPEN(18,FILE='rp-modelb.apr',STATUS='OLD') OPEN(20,FILE='rp-modelb.may',STATUS='OLD') OPEN(22,FILE='rp-modelb.jun',STATUS='OLD') OPEN(24,FILE='rp-modelb.jul',STATUS='OLD') OPEN(26,FILE='rp-modelb.aug',STATUS='OLD') OPEN(28,FILE='rp-modelb.sep',STATUS='OLD') OPEN(30,FILE='rp-modelb.oct',STATUS='OLD') OPEN(32,FILE='rp-modelb.nov',STATUS='OLD') OPEN(34,FILE='rp-modelb.dec',STATUS='OLD') C FOR OUTPUT: OPEN(36,FILE='RP_ANN.SE',STATUS='UNKNOWN') OPEN(38,FILE='RP_ANN.ALL',STATUS='UNKNOWN') C==== C INITIALIZE ARRAYS: DO 40 I=1,360 DO 40 JJ=1,720 LAND(I,JJ)=0.D0 SF(I,JJ)=0.D0 SSF(I,JJ)=0.D0 DO 40 II=1,12 TSF(II,I,JJ)=0.D0 40 CONTINUE C==== C FILL LAND/WATER ARRAY OF 0.5*0.5 DEGREE GRID-CELLS BY READING VALUES C READ FIRST 6 LINES DO 50 I=1,6 50 READ(8,*) DO 60 I=1,360 C VALUES ARE ZEROS OR ONES 60 READ(8,*)(LAND(I,JJ),JJ=1,720) CLOSE(UNIT=8) C==== C FILL EMISSION ARRAY OF 0.5*0.5 DEGREE GRIDCELLS/MONTH BY READING VALUES C READ MONTHLY FILES DO 70 II=1,12 IF(II.EQ.1)INFILE=12 IF(II.EQ.2)INFILE=14 IF(II.EQ.3)INFILE=16 IF(II.EQ.4)INFILE=18 IF(II.EQ.5)INFILE=20 IF(II.EQ.6)INFILE=22 IF(II.EQ.7)INFILE=24 IF(II.EQ.8)INFILE=26 IF(II.EQ.9)INFILE=28 IF(II.EQ.10)INFILE=30 IF(II.EQ.11)INFILE=32 IF(II.EQ.12)INFILE=34 C READ FIRST 6 LINES DO 65 I=1,6 65 READ(INFILE,*) DO 70 I=1,360 C VALUES ARE IN UNITS OF '100 * G C/M2' == 100 TIMES GRAMS C READ(INFILE,*)(TSF(II,I,JJ),JJ=1,720) DO 70 JJ=1,720 C CHECK WITH LAND/WATER VALUES IF EMISSION VALUE SHOULD BE KEPT IF(LAND(I,JJ).LE.0.D0) TSF(II,I,JJ)=0.D0 C SUM MONTHLY VALUES TO YEARLY VALUES SSF(I,JJ)=SSF(I,JJ)+TSF(II,I,JJ) 70 CONTINUE CLOSE(UNIT=12) CLOSE(UNIT=14) CLOSE(UNIT=16) CLOSE(UNIT=18) CLOSE(UNIT=20) CLOSE(UNIT=22) CLOSE(UNIT=24) CLOSE(UNIT=26) CLOSE(UNIT=28) CLOSE(UNIT=30) CLOSE(UNIT=32) CLOSE(UNIT=34) C==== C READ YEARLY VALUES C READ FIRST 6 LINES DO 75 I=1,6 75 READ(10,*) LONG=-179.75D0 LAT=89.75D0 II=0 SUMA=0.D0 SUMM=0.D0 DO 90 I=1,359 II=II+1 C VALUES ARE IN UNITS OF '100 * G C/M2' == 100 TIMES GRAMS C READ(10,*)(SF(I,JJ),JJ=1,720) C SKIP AROUND THE EQUATOR IF(LAT.EQ.0.25D0) THEN LAT=-0.25D0 C LINE UP CALCULATED CELL AREA (DXYP) WITH LATITUDE II=II+1 ENDIF DO 80 JJ=1,719 C CHECK WITH LAND/WATER VALUES IF EMISSION VALUE SHOULD BE KEPT IF(LAND(I,JJ).LE.0.D0) SF(I,JJ)=0.D0 C SUM TO GLOBAL EMISSIONS: SUMA=SUMA+SF(I,JJ)*DXYP(II) SUMM=SUMM+SSF(I,JJ)*DXYP(II) C WRITE OUT > RP_ANN.SE: IF(SF(I,JJ).GT.0.D0) WRITE(36,'(7(2X,G12.6))') & LAT,LONG,SF(I,JJ),SSF(I,JJ),LAND(I,JJ),DXYP(II) C WRITE OUT ALL: WRITE(38,'(7(2X,G12.6))') LAT,LONG,SF(I,JJ) LONG=LONG+0.5D0 TLONG=TLONG+0.5D0 C SKIP AROUND THE GREENWICH TIMELINE IF(LONG.EQ.-0.25D0) THEN LONG=0.25D0 ENDIF 80 CONTINUE LAT=LAT-0.5D0 LONG=-179.75D0 90 CONTINUE CLOSE(UNIT=10) CLOSE(UNIT=36) c WRITE OUT SUMS AS OVERALL CHECK (executed on a DEC-ALPHA) TYPE *,SUMA,SUMM C SUMA= 7.7761701E+18 SUMM= 7.7761701E+18 >> 77.8 P gC/yr STOP END ============================================================================== (V) QA performed by CDIAC One of the roles of the Carbon Dioxide Information Analysis Center (CDIAC) is quality assurance (QA) of data. The QA process is an important component of the value-added concept of assuring accurate, usable information for researchers, because data received by CDIAC are rarely in condition for immediate distribution, regardless of source. The following summarizes the QA checks performed on the Model B output presented in the data files: 1) To get correct latitude/longitude locations for soil emissions, we plotted the predicted emissions over maps of vegetation types at the one degree grid cell level by Matthews (1985), and on the half degree grid cell level by Post et. al. (1996), and Olsen et al. (CDIAC, 1985). For the soil emission plots the 0.5 degree latitude band containing the Greenwich meridian and the 0.5 degree longitude band around the equator had to be skipped to achieve proper coastline alignment. 2) When Model B output per grid cell (g C/m2/yr or /month) was multiplied by grid cell area (m2) (calculated according to the program from the Goddard Institute for Space Studies (see IV)), and these products summed, total global soil emission was found to be 80.4 Pg C/yr, while the published Model B value was 77.1 Pg C/yr (Table 1 & p 27 in Raich and Potter, 1995). Using the landmask information, provided by the authors, reduced the total global soil emission to 77.8 Pg C/yr. 3) The contributors of the model output (authors of the paper) are in agreement with CDIAC analysts about the necessity of reporting the above mentioned two points. Note: 1 Pg C = 1.e15 g C ============================================================================== (VI) How to obtain the model predictions The 14 data files (~ 9 MB total) and this db1015.doc file are available from CDIAC's anonymous FTP (file transfer protocol) area: > FTP to CDIAC.ESD.ORNL.GOV > enter 'anonymous' as user id > enter your e-mail address as password > cd pub/db1015 > get filenames as in (III) > quit ============================================================================== (VII) References Matthews, E. 1985. Atlas of Archived Vegetation, Land-Use and Seasonal Albedo Data Sets. Goddard Space Flight Center Institute for Space Studies. N.Y, N.Y. Olson, J. S., J. A. Watts and L. J. Allison. 1985. Major World Ecosystem Complexes Ranked by Carbon in Live Vegetation: A Database. NDP017. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory. Oak Ridge, Tennessee. Post, W. M., A. W. King and S. D. Wullschleger, 1996. Soil Organic Matter Models and Global Estimates of Soil Organic Carbon. Nato ASI Series Vol 138: 201-222. Raich, J. W. and C. S. Potter. 1995. Global Patterns of Carbon Dioxide Emissions from Soils. Global Biogeochemical Cycles 9(1)23-36. Raich, J. W. and W. H. Schlesinger. 1992. The Global Carbon Dioxide Flux in Soil Respiration and its Relationship to Vegetation and Climate. Tellus. Ser. B 44:81-99. Wilkinson, L. 1990. SYSTAT: The System for Statistics. version 5.0, SYSTAT: Evanston, Ill. ==============================================================================