Estimation of winter wheat yield in hebei plain of China by improved CASA model

In this study, we use an improved and validated Carnegie Ames Stanford Approach (CASA) model to predict NPP. In this model, NPP can be estimated just using monthly meteorological data and monthly NDVI data. It is relatively easier to acquire data than other model and its application can be enhanced. And we build regression model between NPP and yield got from national statistical almanac, so we can go further to estimate yield. Results showed that the square of regression coefficients of the model between NPP and yield was 0.985, and significance level of this model passed 0.001. And there are clearly strong spatial variations in Hebei Plain winter wheat NPP, and the spatial changes of yield were notable. The total annual yield of winter wheat in Hebei Plain in 2004 was 12.3 Mt.