اعتبارسنجی و صحت‌سنجی مدل aquacrop در شبیه‌سازی میزان عملکرد، ماده خشک و بهره وری آب ذرت دانه‌ای (zea mays l.) تحت روش‌های مختلف آبیاری و سطوح کود نیتروژنه

هدف از این مطالعه، ارزیابی مدل aquacrop از نظر شبیه‌سازی عملکرد دانه و زیست‌توده ذرت دانه‌ای تحت دو روش آبیاری (فتیله‌ای و سطحی) و کوددهی نیتروژن (کود سرک و کودآبیاری) بود. این آزمایش در دو سال 1399 و 1400 در دانشکده کشاورزی دانشگاه رازی انجام شد. تیمارهای آزمایشی شامل آبیاری فتیله‌ای و آبیاری سطحی (در سه سطح 100،70 و 50 درصد نیاز آبی گیاه) و دو سطح کود نیتروژنه 100 و 50 درصد نیاز کودی گیاه بود. نتایج نشان داد که مقدار ریشه میانگین مربعات خطا (rmse) در شبیه سازی روند توسعه پوشش گیاهی در تیمارهای مختلف آبیاری و کودی در مرحله واسنجی (1399) بین 1.5 تا 6.1 درصد و در مرحله صحت سنجی (1400) بین 2 و 6.4 درصد به‌دست آمد. مقدار ضریب کارآیی مدل (ef) در شبیه سازی پوشش گیاهی، در مرحله واسنجی بین 0.91 و 0.99 و در مرحله صحت سنجی بین 0.93 و 0.99 بود. مقدار rmse در شبیه سازی ماده خشک برای مرحله واسنجی بین 1.07 تا 2.1 تن در هکتار و در مرحله صحت سنجی بین 1.4 تا 2.9 تن در هکتار به‌دست آمد. همچنین نتایج شبیه سازی مدل aquacrop نشان داد که این مدل توانست شبیه سازی بهره وری آب را بهتر از عملکرد دانه و ماده خشک کل انجام دهد، به طوری‌که ضریب تبیین (r2) مربوط به بهره وری آب، عملکرد دانه و ماده خشک کل ذرت به‌ترتیب برابر با 0.82، 0.52 و 0.54 به‌دست آمد. در مجموع، aquacrop به‌خوبی برای منطقه مورد مطالعه انتخاب شد و توانایی aquacrop در زمینه ارائه یک مدل کم‌آبیاری مانند روش فتیله ای مناسب است.

calibration and verification of the aquacrop model in simulating the yield, dry matter and water productivity of grain corn (zea mays l.) under different irrigation methods and nitrogen fertilizer levels

introductionthe decrease in rainfall in recent decades and the occurrence of drought in iran, which is one of the arid and semi-arid regions, as well as the significant wastage of water in the agricultural sector, have caused many researchers to look for new irrigation methods that lead to improve water consumption. therefore, the importance of conducting this experiment is to reduce surface water evaporation, save water, reduce leaching and use of nitrogen fertilizer in maize cultivation. nowadays, due to the decrease in rainfall and the excessive water consumption in the agricultural sector, the importance of water resources management has received more attention from researchers. therefore, simulation models an effective role in evaluating irrigation management methods to improve water consumption in the agricultural sector. this study aimed to assess the aquacrop model in terms of simulating the yield and biomass of maize under two methods of irrigation (wick and surface) and nitrogen fertilization (topdressing and fertigation).materials and methodsthis experiment was carried out in 2020 and 2021 in the faculty of agriculture of razi university. experimental treatments included wick irrigation and surface irrigation (at three levels of 100, 70, and 50%) and two levels of nitrogen fertilizer 100 and 50% of the plant’s fertilizer requirement. the statistical design was factorial in the form of randomized complete blocks and was implemented in three replications. in the wick irrigation method, fertilizer was provided to the plant through tanks and in solution. in the surface irrigation method, fertilizer was topdressing on the soil surface. nitrogen fertilizer was applied in three stages. in each watering, the amount of water used will be measured and recorded through the meter. at the end of the experiment, the data measured in the first and second years were used for calibration and validation, respectively. for both stages of calibration and validation of the model, the measured (observed) and simulated values of yield, biomass, canopy cover, and water productivity were compared and statistically analyzed. for statistical evaluation, root mean square error (rmse), efficiency coefficient of the nash-sutcliffe model (ef), and wilmot agreement index (d) were used.results and discussionthe results indicated that the root mean square error (rmse) for simulating canopy cover development under various irrigation and fertilizer treatments ranged from 1.5% to 6.1% during the calibration stage (2020) and from 2% to 6.4% during the verification stage (2021). the model’s efficiency factor (ef) for canopy cover simulation was between 0.91 and 0.99 during the calibration stage and between 0.93 and 0.99 during the validation stage. for biomass simulation, the rmse values ranged from 1.07 to 2.1 tons/ha in the calibration stage and from 1.4 to 2.9 tons/ha in the validation stage. the model’s ef for biomass simulation ranged from 0.87 to 0.98 in the calibration stage and from 0.92 to 0.99 in the validation stage. these results demonstrate the model’s high accuracy and reliability in simulating both canopy cover and biomass development across different treatments. aquacrop simulation results showed that the model was able to simulate water productivity better than yield and total biomass so the coefficient of determination (r2) related to water productivity, yield, and biomass of maize was obtained as 0.82, 0.52, and 0.54 respectively.