关键词: 参数优化, 最大羧化速率, BEPS模型, 最大电子传递速率, 总初级生产力

Abstract: Vegetation net primary productivity (NPP) is an important parameter to evaluate a terrestrial ecosystem. Based on MODIS NDVI data from 2000-2014, we improved CASA model with land cover type, classification resolution, radiation data selection and formula selection, and calibrated the model according to the local natural environment in Northwest China. Then we estimated NPP and analyzed its temporalspatial distribution pattern. The results showed that the improved model gave a good vegetation NPP estimation in the arid and semiarid areas and reflected the vegetation growth and distribution in the study area; the mean annual vegetation NPP was 191.63 g C·m^-2·a^-1. There was an obvious regional difference in NPP distribution in the arid area in Northwest China, basically being high in the northwest and southeast and low in the center. There was an increasing trend in annual NPP with an annual rate of 2.98 g C·m^-2·a^-1, which differed with vegetation types, being the largest in cropland, followed by shrub, and least in woodland. Vegetation NPP had a significant correlation with precipitation (r=0.538, P<0.05) and no significant correlation with temperature (r=0.394), indicating that vegetation NPP was more highly correlated with precipitation than with temperature; the correlation coefficient differed with vegetation types.

Key words: parameter optimization, maximum rate of carboxylation, gross primary productivity, BEPS model, maximum potential rate of electron transport.