Vegetation temperature condition index (VTCI) takes the effects of NDVI and LST into account, and is applicable to drought monitoring in the Guanzhong Plain, Shaanxi, China. VTCI provides a scientific basis for drought relief and crop yield estimation in the plain. Based on the research of correlation of the weighted VTCI and winter wheat yield, and SARIMA drought forecasting results, the winter wheat yield forecasting models were developed at intervals of 1 ten-day, 2 ten-day and 3 ten-day before the harvest. The results showed that the yield forecasting results and yield monitoring results had a good agreement. The forecasting accuracy was reduced with the increase of the forecasting interval and the maximum relative error of the yield forecasting results of the 3 ten day was 3.27%, which indicated that the developed models could be used to forecast winter wheat yield at 30 days before the harvest.