The objectives of this paper are to establish the climatically determined potential grain yields of wheat for different locations of India, to quantify the gap between actual and potential yields and to determine the optimal levels of irrigation and N required for given productivity levels. The analysis is based on simulations made with the crop growth model WTGROWS. Simulated potential grain yields, determined by solar radiation and temperature, varied between 2.56 and 8.25 t ha(-1) for 138 Iodations spread throughout India, In general, yields increased with latitude and inland locations had greater yields than the coastal locations at the same latitude. These trends were related to mean temperature differences over latitude/location. The results indicate a strong linear decline in grain yield as mean temperature increased. Late sowings had smaller yields as well as increased variability. The decrease for each day's delay in sowing was more when potential yield was high. The yield gap was at least 2 t ha(-1) irrespective of location and a significant portion of this was due to delayed sowing. Crop simulation with different amounts of nitrogen and irrigation inputs showed significant interaction between water and N availability as well, as inter-seasonal climatic variability, particularly with low input of water. The optimal N application depended on the amount of water availability. Yield variance of stressed wheat crops is moderated greatly by irrigation but reduced N fertilizer application may modify the response.