Estimation of NPP in Western China Using Remote ... - Semantic Scholar
Estimation of NPP in Western China Using Remote Sensing and the C-Fix Model L.Lu and X.Li
F.Veroustraete and Q.H.Dong
Cold and Arid Regions Environmental and Engineering Research Institute, CAS Lanzhou, China [email protected]
Centre for Remote Sensing and Atmospheric Processes, Vito Mol, Belgium [email protected]
Abstract—Net Primary Productivity (NPP) is a key component of the terrestrial carbon cycle. The accurate estimation of NPP on regional and global scale is crucial for the studies of global change. In this paper, the Monteith type parametric model--CFix, the 1km SPOT4/VEGETATION data as well as the global meteorological data provided by Meteo France were used to estimate NPP of the terrestrial ecosystems in Western China (73°-112°E, 26°-50°N) for the year of 2002. The total yearly NPP of Western China was estimated at 0.96P (=1015)g C in 2002, but the total mean NPP was only equal to 168g C/m2/year over the study area of 4.5 million km2. The spatial pattern of the annual accumulated NPP as well as the monthly dynamics of NPP in Western China were illustrated and descript in detail. In addition, the NPP-values in the annual accumulated level and the mean level for the different ecosystems were evaluated by using the newest 1:1M land-use map of Western China. The results showed that the spatial and temporal patterns of NPP in Western China are attributable to the complex interaction between natural environment, various climates and human activities. Although Western China has a large area of land, the total mean NPP level is very low due to the hard natural conditions. Among them, the restricted water resource is the main element to control the NPP of Western China. Keywords-NPP; remote sensing;C-FIX model;Western China
I.
INTRODUCTION
NPP is an important variable in studies of the carbon cycle at regional and global scales, as it determines the rate of absorption of atmospheric carbon by land vegetation [1]. In recent years, several models of terrestrial NPP have been developed related to food security [2] and global warming [3]. The models can generally be classified into three broad categories as statistical climate-correlation models, canopy photosynthesis models and production efficiency models (PEMs) [4]. Because PEMs are based on the Monteith [5] approach and integrated satellite remote sensing data as well, they can estimate terrestrial NPP with high spatial and temporal resolution at regional and global scales. Moreover, with a new generation of advanced optical sensors such as SPOT4/VEGETATION and MODIS coming into operation recently, more attentions have been paid to PEMs [6]. Western China approximately covers 4.45million km2 area bounded by 26º－50ºN latitude and 73º－112ºE longitude. The
second largest desert-Taklimakan Desert and the highest plateau- Tibetan Plateau in the world are encompassed in this region. The climatic zone of this area ranges from a subtropical zone in the south to an ultra-arid zone in the center and a sub-frigid zone in the north. The ecosystem types are very diversified including forest, grasslands, agroecosystems, wetlands, shrub and desert-ecosystems. In addition, the ecosystems in Western China represent some of the most fragile ecosystem types in the nation and in the world [7]. With its ambitious economic plan on the development of Western China launched in March 2000 by the Chinese Government and the global warming as well, rapid changes in the terrestrial NPP of Western China are expected in the next few decades. The objectives in this study are to estimate the terrestrial NPP and display its spatial and temporal patterns of Western China by using a PEMs type model-C-Fix, which is an initial work for further research on the regional carbon budget responded to human activities and climate change. II.
METHOD
A. The C-FIX model C-FIX is a Monteith type parametric model to simulate carbon mass fluxes driven by temperature, radiation and fAPAR (fraction of Absorbed Photosynthetically Active Radiation). Its original form was presented by Veroustraete, Patyn, & Myneni in 1994 [8] and the advanced version were detailed by Veroustraete, Sabbe, & Eerens in 2002 [9]. Moreover, The C-FIX model has been successfully applied in estimating NPP over Europe continent and Africa as well. For a given point location, or an remote sensing pixel, the model uses (1) to estimate NPP (in g C/m²/d) on a daily (subscript d) basis: NPPd = ( p(Tatm ) * CO2 fert * ε * fAPAR * c * S g ,d ) * (1 − Ad ) (1)
Where p(Tatm) means the temperature dependency factor ranged from zero to unit [10]. CO2fert defined CO2 fertilization as the increase in carbon assimilation due to CO2 levels above the atmospheric background level [11]. ε is the radiation use efficiency taken equal to 1.1 (g C/MJ) [12]. fAPAR is extracted from NDVI according to NDVI- fAPAR relationship [13]. c is climatic efficiency giving the ratio of PAR to global radiation
0-7803-8743-0/04/$20.00 (C) 2004 IEEE
Sg,d. Ad is defined as an autotrophic respiratory fraction of gross primary productivity (GPPd). B. Input requirement for the model Meteo France provided the data set of the daily air temperature and incoming global radiation in the year of 2002. The resolution of the meteo raster is 1.5 at 1.5 degree. The data over the study region were extracted from the global data set and a re-sampling method of cubic convolution was performed to downscaling the grid resolution from 1.5 degree to 0.25 degree in favor of the C-FIX. We used standard 1km VGT-S10 products from SPOT4/VEGETATION sensor, which are generated by selecting a pixel with the maximum NDVI value in a 10-day period (http://free.vgt.vito.be). The 36 NDVI-composites coving the year of 2002 over the area of Western China were processed as inputs for the C-FIX model. The newest 1:1M land-use map of Western China was also used to evaluate different ecosystem’s NPP. This land-use map was recompiled to the geographic projection with a resolution of 1km to match the remote sensing data by using GIS software. The land use classification of Western China was determined to 18 types as showed in “Table.1”. III.
RESULTS AND DISCUSSION
A. The spatial pattern of annual-accumulated NPP in Western China in 2002 Annual terrestrial NPP of Western China in 2002 was estimated about 0.96PgC. But the total mean NPP was very low that only reached to 168gC/m2/year over the whole Western China. “Fig. 1” shows a strong regional annual NPP distribution. The fundamental spatial pattern of annual NPP in Western China is characterized by high NPP levels distributed in the southeast corner and northwest corner attributed to relatively rich precipitation. Then the NPP levels rapidly decrease from the two corners into the arid inland regions. The NPP value per square meter over the study area even ranges from 1714gC/m2/year in the southeast rainforest to 0gC/m2/year in the inland desert. B. The seasonal pattern of NPP in Western China in 2002 The seasonal pattern or temporal variation of NPP in Western China is pronounced as seen in “Fig. 2”. For most ecosystems, a maximum NPP growth rate can be found during the summer time, whereas a minimum can be noticed during the winter time. Furthermore, different seasonal variation of NPP manifests in different regional parts, which is strong relevant to different vegetation regimes in Western China. For example, in the southern corner grows with some sub-tropical rainforest, so the NPP level in any month shows high even in winter. In the northwest region grows with boreal forest and temperate grassland, there is a obvious difference between the period May-September with a pronounced NPP accumulation and January-March with even zero NPP level. The arid inland regions, especially in gobi and desert, have a very low seasonal variation with monthly NPP under 10gC/m2/month during the whole year.
C. NPP evaluation on different land use types in Western China in 2002 “Table.1” gives an overview of the yearly total NPP and mean total NPP per square meter in function of the different land use types in Western China in 2002. As occupying 48.28% of the total area of Western China, grasslands including closed grasslands, open grasslands and sparse grasslands have the highest annual accumulated NPP value reaching to 0.473PgC. Forestlands including original forests, secondary forests and shrubs are followed with the yearly total NPP value of 0.227PgC. Croplands places in the third, as the yearly total NPP is approximately to 0.142PgC. Other unexploited land use types, i.e. swamp lands, salinealkali lands, gobi, bore rock, bare soil, cold deserts, sandy lands, water bodies and glacier/permanent snow regions, although occupying about 40.32% of the total area of Western China, they only make 12% contributions to the yearly total NPP of the Western China. Compared to the mean NPP level with different ecosystems, the mean NPP of forests is the largest (from 680gC/m2/year to 711gC/m2/year), followed by that of crops (497gC/m2/a), and shrubs (473gC/m2/year). From closed grasslands to open grasslands and sparse grasslands, their mean NPP values rapidly decrease from 231gC/m2/year to 156gC/m2/year and 124gC/m2/year respectively, indicating a pronounced degrade of grasslands. Among the unexploited land use types, swamp ecosystem has the highest mean NPP level (222gC/m2/a) which is even closed to the closed grassland. Land use types with no or less vegetation, i.e. saline-alkali lands, gobi, bore rock, bare soil, cold deserts and sandy lands show very low mean NPP values ranged from 73gC/m2/year to zero. IV.
CONCLUSION
Some general regional and temporal NPP trends could be found over Western China. High NPP values occur in the southeast corner and northwest corner, low values in the large arid inland regions. This spatial pattern is mainly attributable to the various climate zones and the different vegetation ecosystems over the western China. In addition, the terrestrial NPP in the western China has pronounced seasonal dynamics in a whole year, showing a high relative to the seasonal variation in the regional temperature, precipitation, irrigation and the farming activities as well. The total annual NPP of Western China in 2002 was estimated at 0.96PgC, but the total mean NPP per square meter was only 168gC/m2, indicating relatively low terrestrial NPP level due to the very limited water resource over the large arid inland regions. Different land use types and their corresponding NPP were also compared. Grasslands contribute the most on the total annual NPP of Western China. Forests have the highest mean NPP values. On the other hand, there are large areas with no vegetation (desert) in Western China. The C-FIX model is able to estimate and describe regional as well as temporal Net Primary Productivity evolutions and changes due to its simplicity and using of remote sensing data. Remote sensing is a powerful tool to extract fAPAR and monitor regional vegetation change with high resolutions.
0-7803-8743-0/04/$20.00 (C) 2004 IEEE
Figure 1. The spatial pattern of annual-accumulated NPP in Western China in 2002
Figure 2. The seasonal changes of NPP in Western China in 2002
0-7803-8743-0/04/$20.00 (C) 2004 IEEE
TABLE I. Land use type original forest secondary forest shrub and wood land closed grassland open grassland sparse grassland rice irrigated/non-irrigated crop swamp saline-alkali land gobi bare rock cold desert sandy land bare soil urban and built-up water bodies glacier and permanent snow
COMPARISON OF THE ANNUAL NPP OF DIFFERENT LAND USE TYPES IN WESTERN CHINA IN 2002
Mean NPP(gC/m2/year) 680 711 473 231 156 124 632 492 311 222 54 45 43 74 56 58 55 0
Numbers of pixels 219222 39534 105176 955491 970842 815765 9541 276081 2199 15225 110230 455430 647836 163578 670837 99430 63813 66638
Total of NPP（ Percentage of PgC/year） area（%） 0.149 3.85 0.028 0.70 0.050 1.85 0.221 16.80 0.151 17.07 0.101 14.34 0.006 0.17 0.136 4.85 0.00068 0.04 0.0033 0.27 0.006 1.94 0.020 8.01 0.028 11.39 0.012 2.88 0.037 11.80 0.0057 1.75 0.0034 1.12 0 1.17 [6]
ACKNOWLEDGMENT This work is supported by the National Science Foundation of China (NSFC) project (Grant number: 90202014) and two innovation projects of the Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences (Grant numbers: 2003102 and 2004111).
[7]
[8]
REFERENCES [1]
[2]
[3] [4]
[5]
A. Ruimy, L. Kergoat, A. Bondeau, et al., “Comparing global models of terrestrial net primary productivity (NPP): analysis of differences in light absorption and light-use efficiency”, Global Change Biology, 1999, 5(Suppl. 1): 56-64． S. D. Prince, C. D. Justice, and B. Moore III, “Monitoring and Modeling of Terrestrial Net and Gross Primary Production”, International Geosphere Biosphere Program (IGBP), Data and Information System, Global Analysis, Interpretation and Modeling (GAIM). Working Paper 1, 1994, p: 57. IPCC, “Second assessment report of the Intergovernmental Panel on Climate Change (IPCC)”, Environ. Policy Law, 1996, 26(5): 234-235. W. Cramer, D. W. Kicklighter, A. Bondeau, B. Moore, and G. Chrukina, “Comparing global models of terrestrial net primary productivity (NPP): overview and key results”, Global Change Biology, 1999, 5 (Supp. 1): 115. M. Kumar, and J. L. Monteity, “Remote sensing of crop growth”, In: Plants and the daylight spectrum (eds.Smith, H.), London New York Toronto Sydney San Francisco: Academic Press, 1981, 133-144.
[9]
[10]
[11]
[12]
[13]
Percentage of NPP（%） 15.52 2.92 5.21 23.02 15.73 10.52 0.63 14.17 0.07 0.34 0.63 2.08 2.92 1.25 3.85 0.59 0.35 0.00
S. W. Running, R. Nemani, J. M. Glassy, and P. E. Thornton, “MODIS daily photosynthesis (PSN) and annunal net primary production (NPP) product (MOD17)”, Algorithm Theoretical Basis Document, Version 3.0, 1999. P. Gong, M. Xu, J. Chen, J. M. Cheng, Y. Qi, B. Greg, J. Y. Liu, and S. Q. Wang, “A preliminary study on the carbon dynamics of China’s terrestrial ecosystems in the past 20 years”, Earth Science Frontiers, 2002, 9(1): 55-61. F. Veroustraete, J. Patyn, and R. B. Myneni, “Forcing of a simple ecosystem model with fAPAR and climatic data to estimate regional scale photosynthetic assimilation”, In: VGT, Modelling and Climate Change Effects, eds. Veroustraete F. et al., Academic Publishing, The Hague, the Netherlands, 1994a, 151-177. F. Veroustraete, H. Sabbe, and E. Herman, “Estimation of carbon mass fluxes over Europe using the C-FIX model and Euroflux data”, Remote Sensing of Environment, 2002, 83: 376-399. F. K. Y. Wang, “Canopy CO2 exchange of Scots pine and its seasonal variation after four year exposure to elevated CO2 and temperature”, Agricultural and Forest Meteorology, 1996, 82: 1 –27. F. Veroustraete, “On the Use of Ecosystem Modelling for the Interpretation of Climate Change Effects at the Ecosystem Level”, Ecological Modelling, 1994b, Vol 75-76: Issue Sept. '94, pp.221 – 237. F. S. C. Wofsy, F. M. L. Goulden, F. S. M. Fan, P. S. Bakwin, F. B. C. Daube, F. S. L. Bassow, and F. A. Bazzazz, “Net exchange of CO2 in midlatitude forests”, Science, 1993, 260: 1314 – 1317. R. B. Myneni, and D. L. Williams, “On the relationship between fAPAR and NDVI”, Remote Sensing of Environment, 1994, 49: 200 – 211.
0-7803-8743-0/04/$20.00 (C) 2004 IEEE
F.Veroustraete and Q.H.Dong
Cold and Arid Regions Environmental and Engineering Research Institute, CAS Lanzhou, China [email protected]
Centre for Remote Sensing and Atmospheric Processes, Vito Mol, Belgium [email protected]
Abstract—Net Primary Productivity (NPP) is a key component of the terrestrial carbon cycle. The accurate estimation of NPP on regional and global scale is crucial for the studies of global change. In this paper, the Monteith type parametric model--CFix, the 1km SPOT4/VEGETATION data as well as the global meteorological data provided by Meteo France were used to estimate NPP of the terrestrial ecosystems in Western China (73°-112°E, 26°-50°N) for the year of 2002. The total yearly NPP of Western China was estimated at 0.96P (=1015)g C in 2002, but the total mean NPP was only equal to 168g C/m2/year over the study area of 4.5 million km2. The spatial pattern of the annual accumulated NPP as well as the monthly dynamics of NPP in Western China were illustrated and descript in detail. In addition, the NPP-values in the annual accumulated level and the mean level for the different ecosystems were evaluated by using the newest 1:1M land-use map of Western China. The results showed that the spatial and temporal patterns of NPP in Western China are attributable to the complex interaction between natural environment, various climates and human activities. Although Western China has a large area of land, the total mean NPP level is very low due to the hard natural conditions. Among them, the restricted water resource is the main element to control the NPP of Western China. Keywords-NPP; remote sensing;C-FIX model;Western China
I.
INTRODUCTION
NPP is an important variable in studies of the carbon cycle at regional and global scales, as it determines the rate of absorption of atmospheric carbon by land vegetation [1]. In recent years, several models of terrestrial NPP have been developed related to food security [2] and global warming [3]. The models can generally be classified into three broad categories as statistical climate-correlation models, canopy photosynthesis models and production efficiency models (PEMs) [4]. Because PEMs are based on the Monteith [5] approach and integrated satellite remote sensing data as well, they can estimate terrestrial NPP with high spatial and temporal resolution at regional and global scales. Moreover, with a new generation of advanced optical sensors such as SPOT4/VEGETATION and MODIS coming into operation recently, more attentions have been paid to PEMs [6]. Western China approximately covers 4.45million km2 area bounded by 26º－50ºN latitude and 73º－112ºE longitude. The
second largest desert-Taklimakan Desert and the highest plateau- Tibetan Plateau in the world are encompassed in this region. The climatic zone of this area ranges from a subtropical zone in the south to an ultra-arid zone in the center and a sub-frigid zone in the north. The ecosystem types are very diversified including forest, grasslands, agroecosystems, wetlands, shrub and desert-ecosystems. In addition, the ecosystems in Western China represent some of the most fragile ecosystem types in the nation and in the world [7]. With its ambitious economic plan on the development of Western China launched in March 2000 by the Chinese Government and the global warming as well, rapid changes in the terrestrial NPP of Western China are expected in the next few decades. The objectives in this study are to estimate the terrestrial NPP and display its spatial and temporal patterns of Western China by using a PEMs type model-C-Fix, which is an initial work for further research on the regional carbon budget responded to human activities and climate change. II.
METHOD
A. The C-FIX model C-FIX is a Monteith type parametric model to simulate carbon mass fluxes driven by temperature, radiation and fAPAR (fraction of Absorbed Photosynthetically Active Radiation). Its original form was presented by Veroustraete, Patyn, & Myneni in 1994 [8] and the advanced version were detailed by Veroustraete, Sabbe, & Eerens in 2002 [9]. Moreover, The C-FIX model has been successfully applied in estimating NPP over Europe continent and Africa as well. For a given point location, or an remote sensing pixel, the model uses (1) to estimate NPP (in g C/m²/d) on a daily (subscript d) basis: NPPd = ( p(Tatm ) * CO2 fert * ε * fAPAR * c * S g ,d ) * (1 − Ad ) (1)
Where p(Tatm) means the temperature dependency factor ranged from zero to unit [10]. CO2fert defined CO2 fertilization as the increase in carbon assimilation due to CO2 levels above the atmospheric background level [11]. ε is the radiation use efficiency taken equal to 1.1 (g C/MJ) [12]. fAPAR is extracted from NDVI according to NDVI- fAPAR relationship [13]. c is climatic efficiency giving the ratio of PAR to global radiation
0-7803-8743-0/04/$20.00 (C) 2004 IEEE
Sg,d. Ad is defined as an autotrophic respiratory fraction of gross primary productivity (GPPd). B. Input requirement for the model Meteo France provided the data set of the daily air temperature and incoming global radiation in the year of 2002. The resolution of the meteo raster is 1.5 at 1.5 degree. The data over the study region were extracted from the global data set and a re-sampling method of cubic convolution was performed to downscaling the grid resolution from 1.5 degree to 0.25 degree in favor of the C-FIX. We used standard 1km VGT-S10 products from SPOT4/VEGETATION sensor, which are generated by selecting a pixel with the maximum NDVI value in a 10-day period (http://free.vgt.vito.be). The 36 NDVI-composites coving the year of 2002 over the area of Western China were processed as inputs for the C-FIX model. The newest 1:1M land-use map of Western China was also used to evaluate different ecosystem’s NPP. This land-use map was recompiled to the geographic projection with a resolution of 1km to match the remote sensing data by using GIS software. The land use classification of Western China was determined to 18 types as showed in “Table.1”. III.
RESULTS AND DISCUSSION
A. The spatial pattern of annual-accumulated NPP in Western China in 2002 Annual terrestrial NPP of Western China in 2002 was estimated about 0.96PgC. But the total mean NPP was very low that only reached to 168gC/m2/year over the whole Western China. “Fig. 1” shows a strong regional annual NPP distribution. The fundamental spatial pattern of annual NPP in Western China is characterized by high NPP levels distributed in the southeast corner and northwest corner attributed to relatively rich precipitation. Then the NPP levels rapidly decrease from the two corners into the arid inland regions. The NPP value per square meter over the study area even ranges from 1714gC/m2/year in the southeast rainforest to 0gC/m2/year in the inland desert. B. The seasonal pattern of NPP in Western China in 2002 The seasonal pattern or temporal variation of NPP in Western China is pronounced as seen in “Fig. 2”. For most ecosystems, a maximum NPP growth rate can be found during the summer time, whereas a minimum can be noticed during the winter time. Furthermore, different seasonal variation of NPP manifests in different regional parts, which is strong relevant to different vegetation regimes in Western China. For example, in the southern corner grows with some sub-tropical rainforest, so the NPP level in any month shows high even in winter. In the northwest region grows with boreal forest and temperate grassland, there is a obvious difference between the period May-September with a pronounced NPP accumulation and January-March with even zero NPP level. The arid inland regions, especially in gobi and desert, have a very low seasonal variation with monthly NPP under 10gC/m2/month during the whole year.
C. NPP evaluation on different land use types in Western China in 2002 “Table.1” gives an overview of the yearly total NPP and mean total NPP per square meter in function of the different land use types in Western China in 2002. As occupying 48.28% of the total area of Western China, grasslands including closed grasslands, open grasslands and sparse grasslands have the highest annual accumulated NPP value reaching to 0.473PgC. Forestlands including original forests, secondary forests and shrubs are followed with the yearly total NPP value of 0.227PgC. Croplands places in the third, as the yearly total NPP is approximately to 0.142PgC. Other unexploited land use types, i.e. swamp lands, salinealkali lands, gobi, bore rock, bare soil, cold deserts, sandy lands, water bodies and glacier/permanent snow regions, although occupying about 40.32% of the total area of Western China, they only make 12% contributions to the yearly total NPP of the Western China. Compared to the mean NPP level with different ecosystems, the mean NPP of forests is the largest (from 680gC/m2/year to 711gC/m2/year), followed by that of crops (497gC/m2/a), and shrubs (473gC/m2/year). From closed grasslands to open grasslands and sparse grasslands, their mean NPP values rapidly decrease from 231gC/m2/year to 156gC/m2/year and 124gC/m2/year respectively, indicating a pronounced degrade of grasslands. Among the unexploited land use types, swamp ecosystem has the highest mean NPP level (222gC/m2/a) which is even closed to the closed grassland. Land use types with no or less vegetation, i.e. saline-alkali lands, gobi, bore rock, bare soil, cold deserts and sandy lands show very low mean NPP values ranged from 73gC/m2/year to zero. IV.
CONCLUSION
Some general regional and temporal NPP trends could be found over Western China. High NPP values occur in the southeast corner and northwest corner, low values in the large arid inland regions. This spatial pattern is mainly attributable to the various climate zones and the different vegetation ecosystems over the western China. In addition, the terrestrial NPP in the western China has pronounced seasonal dynamics in a whole year, showing a high relative to the seasonal variation in the regional temperature, precipitation, irrigation and the farming activities as well. The total annual NPP of Western China in 2002 was estimated at 0.96PgC, but the total mean NPP per square meter was only 168gC/m2, indicating relatively low terrestrial NPP level due to the very limited water resource over the large arid inland regions. Different land use types and their corresponding NPP were also compared. Grasslands contribute the most on the total annual NPP of Western China. Forests have the highest mean NPP values. On the other hand, there are large areas with no vegetation (desert) in Western China. The C-FIX model is able to estimate and describe regional as well as temporal Net Primary Productivity evolutions and changes due to its simplicity and using of remote sensing data. Remote sensing is a powerful tool to extract fAPAR and monitor regional vegetation change with high resolutions.
0-7803-8743-0/04/$20.00 (C) 2004 IEEE
Figure 1. The spatial pattern of annual-accumulated NPP in Western China in 2002
Figure 2. The seasonal changes of NPP in Western China in 2002
0-7803-8743-0/04/$20.00 (C) 2004 IEEE
TABLE I. Land use type original forest secondary forest shrub and wood land closed grassland open grassland sparse grassland rice irrigated/non-irrigated crop swamp saline-alkali land gobi bare rock cold desert sandy land bare soil urban and built-up water bodies glacier and permanent snow
COMPARISON OF THE ANNUAL NPP OF DIFFERENT LAND USE TYPES IN WESTERN CHINA IN 2002
Mean NPP(gC/m2/year) 680 711 473 231 156 124 632 492 311 222 54 45 43 74 56 58 55 0
Numbers of pixels 219222 39534 105176 955491 970842 815765 9541 276081 2199 15225 110230 455430 647836 163578 670837 99430 63813 66638
Total of NPP（ Percentage of PgC/year） area（%） 0.149 3.85 0.028 0.70 0.050 1.85 0.221 16.80 0.151 17.07 0.101 14.34 0.006 0.17 0.136 4.85 0.00068 0.04 0.0033 0.27 0.006 1.94 0.020 8.01 0.028 11.39 0.012 2.88 0.037 11.80 0.0057 1.75 0.0034 1.12 0 1.17 [6]
ACKNOWLEDGMENT This work is supported by the National Science Foundation of China (NSFC) project (Grant number: 90202014) and two innovation projects of the Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences (Grant numbers: 2003102 and 2004111).
[7]
[8]
REFERENCES [1]
[2]
[3] [4]
[5]
A. Ruimy, L. Kergoat, A. Bondeau, et al., “Comparing global models of terrestrial net primary productivity (NPP): analysis of differences in light absorption and light-use efficiency”, Global Change Biology, 1999, 5(Suppl. 1): 56-64． S. D. Prince, C. D. Justice, and B. Moore III, “Monitoring and Modeling of Terrestrial Net and Gross Primary Production”, International Geosphere Biosphere Program (IGBP), Data and Information System, Global Analysis, Interpretation and Modeling (GAIM). Working Paper 1, 1994, p: 57. IPCC, “Second assessment report of the Intergovernmental Panel on Climate Change (IPCC)”, Environ. Policy Law, 1996, 26(5): 234-235. W. Cramer, D. W. Kicklighter, A. Bondeau, B. Moore, and G. Chrukina, “Comparing global models of terrestrial net primary productivity (NPP): overview and key results”, Global Change Biology, 1999, 5 (Supp. 1): 115. M. Kumar, and J. L. Monteity, “Remote sensing of crop growth”, In: Plants and the daylight spectrum (eds.Smith, H.), London New York Toronto Sydney San Francisco: Academic Press, 1981, 133-144.
[9]
[10]
[11]
[12]
[13]
Percentage of NPP（%） 15.52 2.92 5.21 23.02 15.73 10.52 0.63 14.17 0.07 0.34 0.63 2.08 2.92 1.25 3.85 0.59 0.35 0.00
S. W. Running, R. Nemani, J. M. Glassy, and P. E. Thornton, “MODIS daily photosynthesis (PSN) and annunal net primary production (NPP) product (MOD17)”, Algorithm Theoretical Basis Document, Version 3.0, 1999. P. Gong, M. Xu, J. Chen, J. M. Cheng, Y. Qi, B. Greg, J. Y. Liu, and S. Q. Wang, “A preliminary study on the carbon dynamics of China’s terrestrial ecosystems in the past 20 years”, Earth Science Frontiers, 2002, 9(1): 55-61. F. Veroustraete, J. Patyn, and R. B. Myneni, “Forcing of a simple ecosystem model with fAPAR and climatic data to estimate regional scale photosynthetic assimilation”, In: VGT, Modelling and Climate Change Effects, eds. Veroustraete F. et al., Academic Publishing, The Hague, the Netherlands, 1994a, 151-177. F. Veroustraete, H. Sabbe, and E. Herman, “Estimation of carbon mass fluxes over Europe using the C-FIX model and Euroflux data”, Remote Sensing of Environment, 2002, 83: 376-399. F. K. Y. Wang, “Canopy CO2 exchange of Scots pine and its seasonal variation after four year exposure to elevated CO2 and temperature”, Agricultural and Forest Meteorology, 1996, 82: 1 –27. F. Veroustraete, “On the Use of Ecosystem Modelling for the Interpretation of Climate Change Effects at the Ecosystem Level”, Ecological Modelling, 1994b, Vol 75-76: Issue Sept. '94, pp.221 – 237. F. S. C. Wofsy, F. M. L. Goulden, F. S. M. Fan, P. S. Bakwin, F. B. C. Daube, F. S. L. Bassow, and F. A. Bazzazz, “Net exchange of CO2 in midlatitude forests”, Science, 1993, 260: 1314 – 1317. R. B. Myneni, and D. L. Williams, “On the relationship between fAPAR and NDVI”, Remote Sensing of Environment, 1994, 49: 200 – 211.
0-7803-8743-0/04/$20.00 (C) 2004 IEEE
