A study was conducted with a broad objective of developing and demonstrating a methodology for crop growth monitoring and yield forecasting which can provide periodical crop growth assessment with spatial information. The procedure was developed to generate grid-weather, link the point based simulation model WOFOST (World Food Studies) to spatial inputs like crop, soil and weather and predict wheat yield at grid and administrative scale. Two approaches were adopted to predict wheat yield; a) the regression approach, in which simulated potential yields were regressed with final estimated yields by Directorate of Economics and Statistics (DES) for each of the six major wheat growing states and b) forcing approach in which LAI for each grid (25km x 25km) derived from remote sensing was forced into the simulation model to divert the simulation output and final grain yield into right direction. The deviations between the estimated state yield and reported yield were more in case of the forcing (0.7 – 25.4 %) as compared to regression approach (0.5 – 9.2 %). However, the spatial variability at grid level was explained more in case of forcing approach. Results indicated that regression approach is suitable for in season yield forecasting at state level and forcing approach is better for spatial crop condition assessment and crop growth monitoring.