Modeling changes of net primary productivity of karst vegetation in southwestern China using the CASA model

Modeling large-scale net primary productivity(NPP) using remote sensing data and ecosystems model is a great challenge of global change study.The processes-based model of Carnegie-Ames-Stanford Approach(CASA) has been widely used in regional and global NPP estimates.However applying this model to specific region with harsh topography such as the karst terrain is few reported because of the difficulty in accurately estimating the maximum Light Use Efficiency(LUE).Estimating precisely the karst vegetation NPP is a foundation of rocky-desertified ecosystem restoration in harsh karst habitats.In this study,annual NPP from 1999 to 2003 was simulated using the CASA model driven by SPOT Normalized Difference Vegetation Index(NDVI) at 1km spatial resolution,digital elevation model(DEM),climate variables and vegetation-based parameters.The CASA model was improved by calculating the maximum LUE for 19 vegetation types based on the field observations of NPP.Results showed that(1) the simulated NPP by the improved CASA model has significant relationship with the observed NPP.This model could be better used in estimating NPP of karst vegetation in SW China.(2) From 1999 to 2000 the NPP of both karst and non-karst vegetation increased slightly in SW China,but the spatial change was not significant.The area with lower NPP enlarged in 2001,but the higher NPP area in 2002 enlarged significantly.The NPP decreased in 2003,but the values were still higher than those in 2001.(3) Annual NPP of karst vegetation in SW China ranged from 381.7-439.9 gC m-2 a-1(average 402.34 gC m-2 a-1) during 1999-2003.The annual NPP fluctuated in this period,but there was an overall increase trend with annual increment of 9.93 gC m-2 a-1.The total amount of increment was 11 TgC during those five years.The averaged annual NPP and its increment of non-karst vegetation were all higher than those in karst vegetation.(4) Annual NPP of tropical and subtropical forests,shrub and grassland in karst region during five years are less than those in non-karst region,but the temperate forests and agricultural vegetation have the opposite trend.All of six types of vegetation have increased their NPP both in karst and non-karst regions,along with the highest increment in tropical forest and the lowest one in grassland.The NPP increments of each vegetation type in non-karst region are higher than those in karst region.In general,the spatial-temporal changes of vegetation NPP in SW China are closely related to regional changes of temperature,precipitation and solar radiation.Karst vegetation has lower NPP than non-karst vegetation is mainly due to the water stress and poor soils in harsh karst habitats that restrict plant growth.