Integration of fruit fly and firefly optimization algorithm with support vector regression in estimating daily pan evaporation

The purpose of the present study was to predict the pan evaporation values at four stations including Urmia, Makou, Mahabad, and Khoy, located in West Azerbaijan, Iran, using support vector regression (SVR), SVR coupled by fruit fly algorithm (SVR-FOA), and SVR coupled with firefly algorithm (SVR-FFA). Therefore, for the first time, this research has used the combined SVR-FOA to predict pan evaporation. For this purpose, meteorological parameters in the period of 1990-2020 were gathered and then using the Pearson's correlation coefficient, significant inputs for pan evaporation estimation were determined. The correlation evaluation of the parameters showed that the two parameters of wind speed and sunshine hours had the highest correlation with the pan evaporation values, and in addition, these parameters, as input to the models, improved the results and increased the accuracy of the models. The obtained results indicated that at Urmia station, SVR-FFA with the lowest error was the best model. The SVR-FOA also had better performance than the SVR model. Additionally, the result showed that SVR-FOA with the lowest errors had the best capability in pan evaporation estimation at other studied stations. Therefore, it was concluded that FOA with advantages such as simplicity, fewer parameters, easy adjustment, and less calculation can significantly increase the capability of independent SVR models. Hence, based on the overall results, SVR-FOA may be recommended as the most accurate method for pan evaporation estimation.

Long-term numerical modeling of nitrate leaching into groundwater under surface drip irrigation of corn

Proper management of fertigation is necessary to deal with the harmful impacts of fertilizers. This research aimed to investigate the nitrate leaching rate into groundwater in different fertigation management under the climate change impact in drip irrigation of corn. For this purpose, HYDRUS-2D was calibrated by performing field experiments. Plant water requirement and rainfall were projected until 2050 using LARS-WG6 under the RCP85 scenario. Then, nitrate leaching up to groundwater at the depth of 5 m was simulated in the growing season of corn and the like until 2050 in three fertigation scenarios, including S1 (three regional fertigation splits with irrigation efficiency of 85%), S2 (weekly fertigation with irrigation efficiency of 85%), and S3 (optimum fertigation with irrigation efficiency of 100%). Finally, the annual nitrate leaching rate to groundwater and leached amount were compared in the studied scenarios. The results demonstrated that nitrate penetrated to the depth of 117 and 105 cm at the end of the first year in S1 and S2 scenarios, respectively. In these scenarios, nitrate will reach groundwater in 2031, but nitrate concentrations will not be the same. In the S3 scenario, the nitrate will reach a depth of 180 cm by 2050. Total leached nitrate to groundwater up to 2050 will be 1740, 1200, and zero kg/ha in S1, S2, and S3 scenarios, respectively. Based on the approach of this study, the vulnerability of groundwater to nitrate contamination in different agricultural areas can be evaluated, and appropriate strategies with minimum environmental impacts of fertilizer abuse can be selected accordingly.

Evaluation of AquaCrop and intelligent models in predicting yield and biomass values of wheat

AquaCrop is one of the dynamic and user-friendly models for simulating different conditions governing plant growth in the field. But this model requires many input parameters such as plant information, soil, climate, groundwater, and management factors. In this research, to solve this problem and develop a model with fewer input data, artificial neural network (ANN), support vector regression (SVR), and combined support vector regression with firefly algorithm (SVR-FFA) were used. For this purpose, 440 scenarios were created in 2 farms located in Iran, and the values of yield and biomass obtained by the AquaCrop model were compared with intelligent models. Also, consumable seed and irrigation were considered as inputs of intelligent models. The 99WestW2 farm in Miandoab had a seed yield of 6.588 t/ha, and the WestW10 farm in Mahabad had a seed yield of 5.05 t/ha. The results of this research showed that for both 99WestW2 and WestW10 farms, the SVR-FFA3 model was able to have the lowest amount of error so that for the amount of grain yield, the error values ​​for the farms were 0.033 and 0.069 t/ha, respectively. The error value of biomass was obtained for farms as 0.057 and 0.066 t/ha respectively. After SVR-FFA model, SVR and ANN models also showed good performance due to proper accuracy and saving time. Finally, SVR-FFA, SVR, and ANN models were able to predict yield and biomass values in the shortest time and with the highest accuracy with only two inputs.

An analysis of optimal fertigation implications in different soils on reducing environmental impacts of agricultural nitrate leaching

Excessive and incorrect use of nitrogen (N) fertilizers in agriculture leads to high nitrate leaching to groundwater and harmful effects on the environment. The main objective of this research was to optimize the N fertigation scheduling for a surface micro-irrigation system in different soils. N uptake by corn and its losses were investigated for two fertigation scheduling scenarios including regional recommendation scheduling with three fertigation events and a weekly application schedule. The fertigation scheduling was then optimized to achieve both environmental objectives (minimizing nitrate losses) and corn N requirements (maximizing N uptake sufficiency). For this purpose, the HYDRUS-2D model, simulating water flow and N transport in soil, was linked to an optimization algorithm. In both scenarios, N uptake by plant was not adequate at different stages of growth in all three soil types, especially in the sandy loam soil. Optimization produced a decrease in nitrate leaching and an increase in N uptake as well as fully supplied plant requirements at different stages of corn growth. Optimization framework presented in this study and optimum fertigation scheduling in various soil textures can be applicable as a guideline for operators of micro-irrigation systems which reduce nitrate leaching and increase N uptake sufficiency.

Evaluation of fertigation management impacts of surface drip irrigation on reducing nitrate leaching using numerical modeling

The objective of this study was to investigate the impacts of fertigation strategies on nitrate leaching and its uptake into maize plants. Field experimental data were employed to calibrate a numerical model (HYDRUS 2D/3D) for a surface drip irrigation system in a sandy clay loam soil. The calibrated model was used to simulate nitrate plant uptake and its leaching in different fertigation scenarios based on various fertigation durations and different start times of fertigation. Finally, nitrogen plant uptake was compared with maize N requirement during growth stages in two fertigation frequency scenarios. These simulations were also performed in sandy loam soil. The results show that, if fertigation is done at the end of irrigation, nitrate leaching in shorter fertigation duration will be less than the leaching in longer fertigation duration. However, in the case of fertigation at the beginning of irrigation, the nitrate leaching is higher if the fertigation duration is short, and vice versa. Furthermore, reducing the number of fertigation events in the sandy clay loam soil increases the nitrate plant uptake. However, in the sandy loam soil, a lesser number of fertigation events reduce nitrate uptake.