Predicting biomass and yield of sweet pepper grown with and without plastic film mulching under different water supply and weather conditions

Abstract

The applicability of the FAO AquaCrop model to simulate sweet pepper growth with and without plastic film mulching under different water supply and weather conditions is assessed. Pepper yield and biomass are simulated using the results of a two-year (2012 and 2013) field experiment conducted in Stara Pazova and a three-year (2011, 2012 and 2013) field experiment in Sombor, both in Serbia. The year 2012 was much warmer and drier than the two other years. At the Stara Pazova site, black plastic mulch was used and three different irrigation treatments applied: 1) full irrigation (F), at 100% of ETc (crop evapotranspiration); 2) deficit irrigation (R), at 80% of ETc; and 3) deficit irrigation (S), at 70% of ETc. In Sombor, only full irrigation (F*) was applied and there was no mulching. Two different cultivars (cv. Elephant Ear and Mišina) were grown in Stara Pazova and Sombor, respectively.The model was calibrated using data from the F-treatment in 2012 at Stara Pazova and F*-treatment in 2011 at Sombor, given that the highest yields were recorded in these two cases. The difference between the measured and simulated values was 2.9% and 7.5% for yield and 7.6% and −14.3% for biomass, in Stara Pazova and Sombor, respectively. The calibration results showed a very good agreement between the measured and simulated values. The deviation of simulated yields from the measured values in the validation data set depended on weather and soil conditions. Namely, the match was very good in the case of well-watered pepper grown without plastic film mulching in the average year at Sombor (1.2%), and was the worst in the extremely warm year (−45.2%). Similar differences were noted in biomass. Based on statistical indicators, the coefficient of determination (R2) of pepper yield at Stara Pazova and Sombor was 0.61 and 1.00 and the Willmott index of agreement (d), 0.79 and 0.89, respectively. Thus the AquaCrop model simulates the yield of pepper grown with and without plastic film mulching quite well. Biomass simulations resulted in slightly lower values of R2 and d. The AquaCrop model can also predict the amount of water needed for pepper grown with and without black plastic mulch.