QUANTIFICATION OF CO2 EXCHANGE IN GRASSLAND ECOSYSTEMS OF THE WORLD USING TOWER MEASUREMENTS, MODELING AND REMOTE SENSING 1653

Authors: GILMANOV, T. - SOUTH DAKOTA STATE UNIV.; DEMMENT, M. - UC DAVIS; WYLIE, B. - USGS; LACA, E. - UC DAVIS; AKSHALOV, K. - BARAYEV KAZAKH RES. INST; BALDOCCHI, D. - ECOSYSTEM SCIENCE DIV; BELELLI, L. - UNIV. DI MILANO BICOCCA; Bradford, James; COULTER, R. - ENVIRONMENTAL RES. DIV; DUGAS, W. - TEXAS A & M; Emmerich, William; FLANAGAN, L. - UNIVERSITY OF LETHBRIDGE; FRANK, A. - RETIRED ARS; Haferkamp, Marshall; Johnson, Douglas; MEYERS, T. - NOAA/ARL; Morgan, Jack; NASYROV, M. - UNIV. OF PETROL. TECH.; OWENSBY, C. - AG. RES. GRASSLAND CENTR; PEKOUR, M. - ENVIRONMENTAL RES. DIV; PILEGAARD, K. - WIND ENERGY DEPT. DENMARK; SALIENDRA, N. - USFS; SANZ, M. - CEAM SPAIN; Sims, Phillip; SOUSSANA, J. - AG. RES. GRASSLAND CENTR; TIESZEN, L. - USGS; VERMA, S. - UNIVERSITY OF NEBRASKA

Submitted to: International Grasslands Congress
Publication Type: Proceedings
Publication Acceptance Date: 10/1/2004
Publication Date: 2/5/2005
Citation: Gilmanov, T.G., Demment, M.W., Wylie, B.K., Laca, E.A., Akshalov, K., Baldocchi, D.D., Belelli, L., Bradford, J.A., Coulter, R.L., Dugas, W.A., Emmerich, W.E., Flanagan, L.B., Frank, A.B., Haferkamp, M.R., Johnson, D.A., Meyers, T.P., Morgan, J.A., Nasyrov, M., Owensby, C.E., Pekour, M.S., Pilegaard, K., Saliendra, N.Z., Sanz, M.J., Sims, P.L., Soussana, J.F., Tieszen, L.L., Verma, S.B. 2005. Quantification of co2 exchange in grassland ecosystems of the world using tower measurements, modeling and remote sensing. In: Proceedings of the 20th International Grassland Congress. June 26 - July 1, 2005, University College, Dublin Ireland. p. 587.

Interpretive Summary: Data sets of continuous CO2 flux measurements at 18 flux towers on nontropical grasslands in North America, Europe and Asia were analyzed to estimate the major parameters of CO2 exchange and light-use efficiency, evaluate carbon sequestration potential, and determine their role in regional and global carbon budget. Partitioning of the tower-derived net ecosystem CO2 exchange (Fc) data into gross primary productivity (Pg) and ecosystem respiration (Re) components (Fc = Pg'Re) was achieved using the physiologically-based analysis of ecosystem-scale light-response functions. Maximum daily rates of gross primary production and total ecosystem respiration achieved the 63 g CO2 m-2 d-1 and 54 g CO2 m-2 d-1 respectively and were recorded not only in high productive tallgrass prairies, but also in southern arid grasslands in years with exceptionally high precipitation. Maxima of annual gross primary production (GPP) of 5200 g CO2 m-2 yr-1 and total ecosystem respiration (RE) of 4700 g CO2 m-2 yr-1 were observed in tallgrass prairies of Oklahoma and Texas; the lowest GPP < 400 g CO2 m-2 yr-1 was recorded on the grazed mixed prairie in Montana during a drought year; the lowest RE < 500 g CO2 m-2 yr-1 was estimated for the semidesert grassland in the foothills of Uzbekistan. Depending on hydrothermal conditions and management (grazing, fire), annual net ecosystem CO2 exchange (NEE) of grasslands varies from 2800 g CO2 m-2 yr-1 (early postfire succession in tallgrass prairie, strong carbon sink) to NEE < ' 1500 g CO2 m-2 yr-1 (grazed mixed/tallgrrass prairie, Oklahoma during a drought year, a strong carbon source). Maximum values of daily light use efficiency ' = Pg/(incident daily PAR) were achieved in warm temperate grasslands of Europe and eastern U.S. ('max = 34 ' 40 mmol CO2 (mol incident PAR)-1), followed by tallgrass and mixed prairies ('max = 32 mmol CO2 (mol incident PAR)-1), while the lowest light use efficiency values were recorded in the semidesert foothills of Uzbekistan ('max = 9 mmol CO2 (mol incident PAR)-1). Grassland CO2 flux data show significant positive correlation between gross primary productivity (Pg) and ecosystem respiration (Re) and the normalized difference vegetation index (NDVI), with correlation coefficient r(Pg, NDVI) in the range 0.55 to 0.92, and r(Re, NDVI) in the range 0.45 to 0.94, with maximum values achieved in ecosystems characterized by the highest aboveground phytomass (tallgrass prairies, warm temperate grasslands). Statistically significant relationships of gross primary productivity and ecosystem respiration to the remotely sensed NDVI and other factors-predictors {Xi} characterizing climate, soils, and vegetation were established for various grassland sites: Pg = f(NDVI, Xi1, ', Xin), Re = g(NDVI, Xj1, ', Xjm), with R2 for these relationships in many cases achieving values higher than 90%. Superposition of these functions on the spatio-temporal GIS data layers of NDVI and {Xi} for the areas ecologically similar to the tower sites represents a defensible method of scaling-up tower CO2 flux measurements allowing regional quantification of CO2 balance on grasslands with implications for continental and global carbon budgets.

Technical Abstract: Data sets of continuous CO2 flux measurements at 18 flux towers on nontropical grasslands in North America, Europe and Asia were analyzed to estimate the major parameters of CO2 exchange and light-use efficiency, evaluate carbon sequestration potential, and determine their role in regional and global carbon budget. Partitioning of the tower-derived net ecosystem CO2 exchange (Fc) data into gross primary productivity (Pg) and ecosystem respiration (Re) components (Fc = Pg'Re) was achieved using the physiologically-based analysis of ecosystem-scale light-response functions. Maximum daily rates of gross primary production and total ecosystem respiration achieved the 63 g CO2 m-2 d-1 and 54 g CO2 m-2 d-1 respectively and were recorded not only in high productive tallgrass prairies, but also in southern arid grasslands in years with exceptionally high precipitation. Maxima of annual gross primary production (GPP) of 5200 g CO2 m-2 yr-1 and total ecosystem respiration (RE) of 4700 g CO2 m-2 yr-1 were observed in tallgrass prairies of Oklahoma and Texas; the lowest GPP < 400 g CO2 m-2 yr-1 was recorded on the grazed mixed prairie in Montana during a drought year; the lowest RE < 500 g CO2 m-2 yr-1 was estimated for the semidesert grassland in the foothills of Uzbekistan. Depending on hydrothermal conditions and management (grazing, fire), annual net ecosystem CO2 exchange (NEE) of grasslands varies from 2800 g CO2 m-2 yr-1 (early postfire succession in tallgrass prairie, strong carbon sink) to NEE < ' 1500 g CO2 m-2 yr-1 (grazed mixed/tallgrrass prairie, Oklahoma during a drought year, a strong carbon source). Maximum values of daily light use efficiency ' = Pg/(incident daily PAR) were achieved in warm temperate grasslands of Europe and eastern U.S. ('max = 34 ' 40 mmol CO2 (mol incident PAR)-1), followed by tallgrass and mixed prairies ('max = 32 mmol CO2 (mol incident PAR)-1), while the lowest light use efficiency values were recorded in the semidesert foothills of Uzbekistan ('max = 9 mmol CO2 (mol incident PAR)-1). Grassland CO2 flux data show significant positive correlation between gross primary productivity (Pg) and ecosystem respiration (Re) and the normalized difference vegetation index (NDVI), with correlation coefficient r(Pg, NDVI) in the range 0.55 to 0.92, and r(Re, NDVI) in the range 0.45 to 0.94, with maximum values achieved in ecosystems characterized by the highest aboveground phytomass (tallgrass prairies, warm temperate grasslands). Statistically significant relationships of gross primary productivity and ecosystem respiration to the remotely sensed NDVI and other factors-predictors {Xi} characterizing climate, soils, and vegetation were established for various grassland sites: Pg = f(NDVI, Xi1, ', Xin), Re = g(NDVI, Xj1, ', Xjm), with R2 for these relationships in many cases achieving values higher than 90%. Superposition of these functions on the spatio-temporal GIS data layers of NDVI and {Xi} for the areas ecologically similar to the tower sites represents a defensible method of scaling-up tower CO2 flux measurements allowing regional quantification of CO2 balance on grasslands with implications for continental and global carbon budgets.