Use of a root zone soil moisture model and crop spectral characteristics to estimate sorghum yields in a dryland Alfisol toposequence Uttam Kumar Mandal*, U.S. Victor, N.N. Srivastava, K.L. Sharma, V. Ramesh, M. Vanaja, G.R. Korwar, Y.S. Ramakrishna Central Research Institute for Dryland Agriculture, Santoshnagar, P.O.-Saidabad Hyderabad-500059, India *corresponding author: uttamkm@crida.ernet.in ABSTRACT This study investigated the relationship between sorghum grain yield over range of soil depth with seasonal crop water stress index based on relative evapotranspiration deficits and spectral vegetation indices. A root zone soil moisture model has been used to evaluate the seasonal soil moisture fluctuation and actual evapotranspiration within a toposequence having varying soil depth of 30 to 75 cm as well as different available water capacity ranging from 6.9% to 12.6% (V/V%). The higher r 2 values between modeled and observed values of soil water (r 2 > 0.69 significant at <0.001) and runoff (r 2 = 0.95, significant at P<0.001) indicated good agreement between model output and observed values. The spectral vegetation indices like simple ratio, normalized difference vegetation index (NDVI), green NDVI, perpendicular vegetation index, soil adjusted vegetation index (SAVI) and modified SAVI (MSAVI) was recorded through out the growth period of sorghum. The vegetation indices except perpendicular vegetation index measured during booting to anthesis stages were positively correlated (P<0.05) with leaf area index and yield. The MSAVI measured during booting to milk-grain stage have the highest positive correlation with yield. Variation was noticed when additive and multiplicative forms of water-production functions calculated from water budget model were used to predict crop yield. But the yield estimation was improved when spectral vegetation indices measured during booting to milk-grain is incorporated along with water production functions. The water budget model along with spectral vegetation indices gave satisfactory estimates of sorghum grain yields and appears to be a useful tool to estimate yield as a function of soil depth and available soil water. Keywords: Evapotranspiration, Water production function, vegetation indices, water balance model 1. INTRODUCTION The ability to accurately predict yield of field crops allows producers, economic agencies, and buyers to make decisions with respect to crop management, pricing, and available markets. Other than genetic factor the factor associated with grain yields includes soil characteristics (particle size distribution, bulk density, organic matter, nutrient levels), agronomic inputs (fertilizers and soil amendments), field scale management (tillage, drainage and irrigation) and meteorological effects. However, while simulation models can predicts yield relatively accurately under ideal conditions they are much less accurate when the plant suffers stress due to disease and pest, weed influence and nutrient and moisture deficiency. In dryland/rainfed region water is long been considered to be the main limiting resource for crop growth and yield. Dryland crops frequently suffer crop water stress because of uneven seasonal distribution of rainfall, which may subsequently affect the yield adversely. The magnitude of crop water stress/deficit is assessed in terms of the extent by which the actual evapotranspiration (AET) falls short of its potential value (PET) or the actual soil moisture content is short of a critical threshold value. A simple water budget model is effective to estimate the availability of water to the crop to meet evapotranspiration. The specific indices used to quantify water stress to crop are relative evapotranspiration (AET/PET), relative evapotranspiration deficit (1-AET/PET) or soil moisture deficit (SMD). The effects of stress, as defined by these indices, in different periods of the growing season interact in a complex manner. The combined effects of stress effects in several periods are evaluated by postulating that these effects are additive or Agriculture and Hydrology Applications of Remote Sensing, edited by Robert J. Kuligowski, Jai S. Parihar, Genya Saito, Proc. of SPIE Vol. 6411, 64110B, (2006) · 0277-786X/06/$15 doi: 10.1117/12.707846 Proc. of SPIE Vol. 6411 64110B-1