حساسیت مدل CERES – Maize نسبت به پارامترهای فیزیکی خاک برای تخمین محصول ذرت علوفه‌ای تحت دو سطح کود نیتروژنه

نویسندگان

چکیده

استفاده گسترده و مفید از مدل‌های گیاهی،‏ مستلزم شناخت رفتار آن‌ها در پاسخ به خطاهای پارامترهای ورودی است. هدف از انجام این پژوهش،‏ ارزیابی حساسیت در شبیه‌سازی محصول ذرت علوفه‌ای با مدل CERES – Maize نسبت به پارامترهای فیزیکی خاک تحت دو سطح کود نیتروژنه است. در این مطالعه با استفاده از داده‌‌های یکسال کشت ذرت در منطقه ورامین،‏ حساسیت مدل نسبت به پارامترهای رطوبت ظرفیت زراعی (FC)‎،‏ رطوبت اشباع (SAT)‎،‏ شماره منحنی (CN)‎،‏ ضریب زهکشی (DR)‎ و ضریب بازتابش (ALB)‎ برای دو سطح صفر و 150 کیلوگرم در هکتار کود نیتروژنه بررسی شد. برای تعیین تابع پاسخ مدل به تغییرات پارامترهای ورودی،‏ مدل 80 بار اجرا شد و پس از جمع آوری نتایج به کمک برنامه نوشته شده در محیط نرم‌افزار متلب،‏ حساسیت نرمال شده هر پارامتر به دست آمد. نتایج نشان داد که مدل،‏ بیشترین حساسیت را در شبیه‌سازی محصول در هر دو تیمار به پارامتر FC از خود نشان می‌دهد،‏ اما پاسخ مدل به تغییرات پارامترهای ورودی به شدت تحت تأثیر میزان کود مصرفی قرار دارد. با استفاده از مقادیر حساسیت نسبی محاسبه شده،‏ رابطه بین مقدار خطای محتمل در شبیه‌سازی محصول و میزان خطا در تخمین داده‌های ورودی برای هر دو تیمار ارائه شد.

کلیدواژه‌ها


عنوان مقاله [English]

The sensitivity of predicted silage-corn yield using CERES-Maize model to physical soil parameters under two nitrogen fertilizer levels

نویسندگان [English]

  • h d
  • m gh
  • f m
چکیده [English]

Useful and extensive use of crop models depends on understanding their behavior in response to probable errors in input parameters. The aim of this study was to assess the sensitivity of the simulated silage-corn yield by the most commonly used crop model, CERES – Maize, to physical soil parameters under two nitrogen fertilizer levels. In this study, by using one year field data of silage-corn in Varamin, the model-output sensitivity to field capacity, saturated water content, curve number, root growth factor, drainage coefficient and albedo was analyzed for two N-fertilizer levels (0 and 150 kg N/ha). To determine the model response to changes of input parameters, the model was run 80 times. After collecting data, the normalized sensitivity of each parameter was calculated using the program written in MATLAB environment. The results showed that the highest model sensitivity in the simulation of the yield in both treatments was shown to field capacity. But the response of the model to changes in input parameters was strongly influenced by the amount of fertilizer. Finally, by using the calculated relative sensitivity values, the relationship between probable error in estimating yield for both fertilizer treatments and error in estimation input data was presented.
Useful and extensive use of crop models depends on understanding their behavior in response to probable errors in input parameters. The aim of this study was to assess the sensitivity of the simulated silage-corn yield by the most commonly used crop model, CERES – Maize, to physical soil parameters under two nitrogen fertilizer levels. In this study, by using one year field data of silage-corn in Varamin, the model-output sensitivity to field capacity, saturated water content, curve number, root growth factor, drainage coefficient and albedo was analyzed for two N-fertilizer levels (0 and 150 kg N/ha). To determine the model response to changes of input parameters, the model was run 80 times. After collecting data, the normalized sensitivity of each parameter was calculated using the program written in MATLAB environment. The results showed that the highest model sensitivity in the simulation of the yield in both treatments was shown to field capacity. But the response of the model to changes in input parameters was strongly influenced by the amount of fertilizer. Finally, by using the calculated relative sensitivity values, the relationship between probable error in estimating yield for both fertilizer treatments and error in estimation input data was presented.
Useful and extensive use of crop models depends on understanding their behavior in response to probable errors in input parameters. The aim of this study was to assess the sensitivity of the simulated silage-corn yield by the most commonly used crop model, CERES – Maize, to physical soil parameters under two nitrogen fertilizer levels. In this study, by using one year field data of silage-corn in Varamin, the model-output sensitivity to field capacity, saturated water content, curve number, root growth factor, drainage coefficient and albedo was analyzed for two N-fertilizer levels (0 and 150 kg N/ha). To determine the model response to changes of input parameters, the model was run 80 times. After collecting data, the normalized sensitivity of each parameter was calculated using the program written in MATLAB environment. The results showed that the highest model sensitivity in the simulation of the yield in both treatments was shown to field capacity. But the response of the model to changes in input parameters was strongly influenced by the amount of fertilizer. Finally, by using the calculated relative sensitivity values, the relationship between probable error in estimating yield for both fertilizer treatments and error in estimation input data was presented.
Useful and extensive use of crop models depends on understanding their behavior in response to probable errors in input parameters. The aim of this study was to assess the sensitivity of the simulated silage-corn yield by the most commonly used crop model, CERES – Maize, to physical soil parameters under two nitrogen fertilizer levels. In this study, by using one year field data of silage-corn in Varamin, the model-output sensitivity to field capacity, saturated water content, curve number, root growth factor, drainage coefficient and albedo was analyzed for two N-fertilizer levels (0 and 150 kg N/ha). To determine the model response to changes of input parameters, the model was run 80 times. After collecting data, the normalized sensitivity of each parameter was calculated using the program written in MATLAB environment. The results showed that the highest model sensitivity in the simulation of the yield in both treatments was shown to field capacity. But the response of the model to changes in input parameters was strongly influenced by the amount of fertilizer. Finally, by using the calculated relative sensitivity values, the relationship between probable error in estimating yield for both fertilizer treatments and error in estimation input data was presented.

کلیدواژه‌ها [English]

  • DSSAT.-Corn-Sensitivity analysis-Crop model-