Estimation of NPP in Xuzhou Based on Improved CASA Model and Remote Sensing Data

In order to explore the distribution and change of NPP at urban scale, and in view of the high spatial heterogeneity of cities, this paper improves the CASA model, estimates the NPP in the central urban area of Xuzhou in March 2018 based on MODIS and Landsat 8 remote sensing data, analyses the spatial distribution characteristics of NPP in the study area and compares the NPP estimates under different models. The results show that: 1) the NPP values of the eastern, southern parts of the study area are higher, while the NPP values of the western part of the central region are lower, and the NPP values of the outward parts of the central region tend to increase gradually; 2) without considering the construction land, the NPP values of cultivated land in the study area are the highest, followed by grassland, forest land and water body, and the NPP values of unused land are the lowest; 3) Compared with CASA model, the improved CASA model is better. It highlights the changes in the distribution of construction land, and reflects the impact of construction land on the results of NPP estimation at the urban scale. In addition, under this model, NPP estimation based on Landsat 8 remote sensing data is more advantageous in urban scale, and the estimation results are more accurate.