Abdelhakim Lahjouj, A. El Hmaidi, Boufala M'hamed, Bouhafa Karima
{"title":"Simulation of soil water and nitrate transport in wheat field under various nitrogen fertilizer rates and rainfed conditions using HYDRUS-1D","authors":"Abdelhakim Lahjouj, A. El Hmaidi, Boufala M'hamed, Bouhafa Karima","doi":"10.37501/soilsa/161944","DOIUrl":null,"url":null,"abstract":"In this study, we used HYDRUS-1D software to simulate soil water and nitrate (NO 3 -N) transport in a rainfed wheat fi eld under various nitrogen (N) fertilizer scenarios (0 to 126 kg ha –1 ) in Mo-rocco. We used inverse modeling to calibrate the input parameters involved in the simulation. The comparison between simulated and measured soil water (SWC) and NO 3 -N contents at different soil layers was carried out using the index of agreement ( d ), determination coe ffi cient ( R 2 ), RMSE , and MAE . By considering the soil pro fi le (0–100 cm), acceptable SWC simulation accuracies were obtained for the calibration and validation steps ( d =0.88–0.94, R 2 =0.67 to 0.80, RMSE =0.034–0.051 cm 3 cm –3 , and MAE =0.024–0.038 cm 3 cm –3 ), while NO 3 -N simulation was less accurate ( d =0.49–0.82, R 2 =0.20–0.58, RMSE =0.015–0.068 mg cm –3 , and MAE =0.012–0.046 mg cm –3 ). In addition, the observed NO 3 -N contents showed a lack of signi fi cant differences in the root zone (20–100 cm) between N fertilizer rates (p>0.05), which was consistent with the lack of N fertilizer effects on simulated NO 3 -N leaching below the soil pro fi le by HYDRUS-1D. The NO 3 -N leached amount accounted for 25 kg ha –1 and was derived mainly from the initial soil N contents. The simulated N balance of the soil pro fi le revealed that volatilization and denitri fi cation were the major pathways of N fertilizer loss, accounting for about 3.8 and 51.7% of the N fertilizer rates, respectively. We suggest further studies to improve the simulation accuracies of HYDRUS-1D using su ffi cient calibration data from long-term wheat experiments to ensure effective N fertilization management in the study area.","PeriodicalId":44772,"journal":{"name":"Soil Science Annual","volume":"75 1","pages":""},"PeriodicalIF":1.4000,"publicationDate":"2023-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Soil Science Annual","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.37501/soilsa/161944","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"SOIL SCIENCE","Score":null,"Total":0}
引用次数: 0
Abstract
In this study, we used HYDRUS-1D software to simulate soil water and nitrate (NO 3 -N) transport in a rainfed wheat fi eld under various nitrogen (N) fertilizer scenarios (0 to 126 kg ha –1 ) in Mo-rocco. We used inverse modeling to calibrate the input parameters involved in the simulation. The comparison between simulated and measured soil water (SWC) and NO 3 -N contents at different soil layers was carried out using the index of agreement ( d ), determination coe ffi cient ( R 2 ), RMSE , and MAE . By considering the soil pro fi le (0–100 cm), acceptable SWC simulation accuracies were obtained for the calibration and validation steps ( d =0.88–0.94, R 2 =0.67 to 0.80, RMSE =0.034–0.051 cm 3 cm –3 , and MAE =0.024–0.038 cm 3 cm –3 ), while NO 3 -N simulation was less accurate ( d =0.49–0.82, R 2 =0.20–0.58, RMSE =0.015–0.068 mg cm –3 , and MAE =0.012–0.046 mg cm –3 ). In addition, the observed NO 3 -N contents showed a lack of signi fi cant differences in the root zone (20–100 cm) between N fertilizer rates (p>0.05), which was consistent with the lack of N fertilizer effects on simulated NO 3 -N leaching below the soil pro fi le by HYDRUS-1D. The NO 3 -N leached amount accounted for 25 kg ha –1 and was derived mainly from the initial soil N contents. The simulated N balance of the soil pro fi le revealed that volatilization and denitri fi cation were the major pathways of N fertilizer loss, accounting for about 3.8 and 51.7% of the N fertilizer rates, respectively. We suggest further studies to improve the simulation accuracies of HYDRUS-1D using su ffi cient calibration data from long-term wheat experiments to ensure effective N fertilization management in the study area.
期刊介绍:
Soil Science Annual journal is a continuation of the “Roczniki Gleboznawcze” – the journal of the Polish Society of Soil Science first published in 1950. Soil Science Annual is a quarterly devoted to a broad spectrum of issues relating to the soil environment. From 2012, the journal is published in the open access system by the Sciendo (De Gruyter).