Mapping of crop coefficient (kc) from remote information in avocado in Mediterranean Chile

Resumen

The current and future conditions of water scarcity in the central zone of Chile, foresee a scenario where irrigation requirements by fruit orchards must be supply optimally, especially in those with high water demand such as avocado. The present investigation had the objective of generating and evaluating cartographies of crop coefficient (Kc) to optimize avocado irrigation in Peumo, Cachapoal province, O’Higgins region. With this approach crop coefficient (Kc) and its components, basal crop coefficient (Kcb) and soil evaporation coefficient (Ke) were generated, accounting for the spatio-temporal variability of these variables and also generating local values for the species. Sentinel 2 satellite data and in situ measurements of leaf area index and radiation interception fraction were used. With these leaf area index (IAF) and fraction of radiation interception (FPAR) models were generated. Satisfactory results were obtained for IAF with an R2 = 0,87, from the vegetation index “Soil Adjusted Vegetation Index” (SAVI) and for the fraction of radiation interception (FPAR) with R2 = 0,85 from the Normalized Difference Vegetation Index (NDVI). Cartographies of Kcb, Ke and Kc were generated from the IAF and FPAR models, which were compared with results from simulations performed with the water transfer model, HYDRUS 1D, which was calibrated with soil moisture data and with which daily time series of transpiration, evaporation, percolation, Kcb, Ke and Kc were generated. For the potential crop coefficient an R2 = 0,75 was obtained between satellite estimates and simulations based on the HYDRUS 1D water transfer model. It is shown that the satellite method has high correlation with the water balance method in scenarios where irrigation management is optimal, not being so when there is any stress related to soil moisture. The time series of potential crop coefficient at the modeled site indicates that, on average, it was 0,78 for the whole season in an orchard where 1.334,8 mm were applied in the form of irrigation water and the water demand by the avocado crop was 868 mm, hence showing that irrigation could be decreased without reducing transpiration. The results show that it is possible to manage irrigation with local crop coefficient data measured periodically given the Sentinel 2 revisit of the territory, as long as it is combined with a method that allows detecting if the crop is under stress from the soil and with this, generate management zones associated to the crop coefficient and its components, basal crop coefficient and soil evaporation coefficient.