Abstract: Shiyang River basin is a typical inland river basin in China. Its ecological characteristics are sensitive and fragile. It is an ideal area for understanding the responses and feedbacks of gross primary productivity (GPP) to climate change in arid regions. In this study, we developed a light efficiency model using satellite and ground-based data to simulate GPP in Shiyang River basin during 2000-2019, and analyzed the spatial distribution and interannual variation of GPP of different vegetation types under the influence of climate. The results showed that average GPP in Shiyang River basin was 256.52 g C·m^-2. The GPP of deciduous broadleaved forests, evergreen coni-ferous forests, shrublands, croplands, grasslands, wetlands, and desert was 676.38, 609.96, 144.42, 404.49, 314.07, 75.15, and 110.21 g C·m^-2, respectively. The GPP of deciduous broadleaved forests in Qilian Mountain was the highest in the southern part of the basin, and that of wetland in desert area was the lowest in the northern part. The GPP across the basin showed a growth rate of 6.99 g C·m^-2·a^-1 with a proportion of 92% for the area of increasing GPP during 2000-2019. The growth rate of GPP of different vegetations was in the order of deciduous broadleaved forests > evergreen coniferous forests > grasslands > croplands > shrublands > desert > wetlands. GPP was positively correlated with precipitation, while negatively correlated with evaporation. The GPP of natural vegetations was mainly affected by water availability, the closer to the arid area, the stronger the effect of water vapor pressure on GPP. Croplands had a high correlation with heat due to human interventions.

Key words: gross primary productivity, vegetation type, climatic factor, correlation, Shiyang River basin.