Evaluation of physiological and biometric parameters of potato (Solanum tuberosum L.) Under different water regimes by proximate sensing

A field experiment has been conducted in Southern Italy (Valenzano, Bari) under openfield conditions to evaluate the use of proximate sensing techniques for monitoringpotato (Solanum tuberosum L.) development and growth under different water regimes.Three water levels have been applied, rainfed (I0), 50 % of irrigation requirements (I50)and full irrigation (I100), in order to assess the impact of water stress on the crop growthand physiology. The irrigation scheduling has been managed via an Excel-basedirrigation tool that employed meteorological, soil and crop data for a day-by-dayestimation of the soil water balance in the effective root zone. This study aims toestablish a comprehensive dataset with different types of measurements in order to linka set of spectral vegetative indexes (such as Simple Ratio, Normalized DifferenceVegetation Index, Soil Adjusted Vegetation Index and the Water Index) as well as thethermal crop water stress index (CWSI), with some eco-physiological and biometriccrop parameters. The variations of the physiological functions of potato at differentphenological stages have been monitored under the three water regimes. Thepreliminary results indicated that under the water stress a reduction of stomatalconductance is observed which affected the rate of photosynthesis, canopy cover andbiomass production and yield. Both the CWSI and the Water Index have shown a highsensitivity to the level of plant water status, with a clear correlation with the level ofstomatal conductance and other physiological parameters, confirming the reliability ofthe indexes to detect the relative level of water stress when properly applied at the leafscale. On the other side, with respect to the VIs applied at canopy scale, the spectralvegetation indices NDVI and SAVI showed generally a good level of correlation withleaf area index and fraction of ground cover, although NDVI seemed to reach asaturation level for values of LAI around 3, while SAVI seems to require a betterparameterization of the effect of the soil background under the different levels ofeffective ground cover. Based on these preliminary findings, it can be confirmed thatproximate sensing techniques and derived indexes could be powerful and effectivemethods to support the real-time monitoring of crop water status for a more appropriate irrigation scheduling