Performance analyses of effective rainfall estimation methods for accurate quantification of agricultural water...

Water footprint (WF) assessments have become a significant tool for the sustainable management in recent years. Effective rainfall (P_eff) is a critical indicator for characterizing soil moisture (green water, WF_green) and calculating irrigation requirements (blue water, WF_blue). However, majority of the water footprint analyses employ empirical or numerical models to predict P_eff, and the number of studies for experimental validation of these models are quite insufficient. The main scope of this study is to test the performance of commonly used P_eff estimation models in relation to the soil water balance (SWB) of an experimental site. Accordingly, the daily and monthly soil water budget is estimated from a maize field which is characterized as semi-arid land with continental climate (Ankara, Turkey), equipped with moisture sensors. Then, P_eff, WF_green, and WF_blue parameters are calculated using FP, US-BR, USDA-SCS, FAO/AGLW, CROPWAT, and SuET methods and compared with SWB method. Employed models were highly variable. CROPWAT and US-BR predictions were the most accurate. In majority of months, the CROPWAT method estimated the P_eff with a maximum deviation of 5% from the SWB method. In addition, the CROPWAT method predicted blue WF with an error less than 1%. The widely utilized USDA-SCS approach did not produce expected results. The FAO-AGLW method provided the lowest performance for each parameter. We also find that the errors in estimating P_eff in semi-arid conditions cause green and blue WF outputs to be quite less accurate than the dry and humid cases. This study provides one of the most detailed assessments about the impact of effective rainfall on the blue and green WF results with high temporal resolution. The findings of this study are important for the accuracy and performance of the formulae used in P_eff estimations and to develop more precise blue and green WF analyses in the future.