Prediction models were developed for wheat to assess crop growth in terms of leaf area index, dry matter production and grain yield from remotely-sensed temperature and spectral indices. The cumulative stress degree days (SDD) for the period of flowering to grain formation stage showed significantly higher correlation with dry matter (r= — 0940) and grain yield (r= —0-939) whereas that, for the period grain formation to harvest stage, showed significantly higher correlation lpar;r= —0-967) for crop water use. Significant and positive correlations between dry matter, leaf area and grain yield with infrared/red, normalised difference (ND), transformed vegetation index and greenness index were attained with the latter providing the highest degree of predictability. Spectral indices measured between flowering to milking stages gave the best prediction indicating the suitability of this period for crop growth assessment by this technique. Inter-stage sensitivity analysis by using multiple regression approach also revealed that greenness and transformed vegetation indices could provide better prediction of dry matter and grain yield. From the values of regression coefficients the jointing to beginning of milk formation period of the crop was found to be the most sensitive stage influencing the yield of crop.