نوع مقاله : مقاله پژوهشی

نویسندگان

• علیرضا نوری ¹
• سمیرا نوری ²
• جواد امیدوار ³
• حسین بانژاد ⁴
• فرشته مدرسی ⁵

¹ گروه آگرو تکنو.لوژی، دانشکده کشاورزی ، دانشگاه فردوسی مشهد، مشهد-ایران

² گروه علوم و مهندسی آب، ، دانشکده کشاورزی ، دانشگاه فردوسی مشهد، مشهد- ایران

³ 'گروه علوم و مهندسی آب، دانشکده کشاورزی- دانشگاه فردوسی مشهد. مشهد. ایران

⁴ گروه علوم و مهندسی آب- دانشکده کشاورزی - دانشگاه فردوسی مشهد- مشهد - ایران

⁵ گروه علوم و مهندسی آب - دانشکده کشاورزی -دانشگاه فردوسی مشهد- مشهد- ایران

چکیده

اندازه‌گیری دقیق مقدار بارش به عنوان یکی از مهم‌ترین متغیرهای هواشناسی در مطالعات هیدرولوژیکی و کشاورزی است که تأثیر به سزایی در مدیریت بهینه منابع آبی، کشت محصولات کشاورزی، مدیریت شهری و شناسایی و دسته بندی مناطق از نظر میزان ریسک وقوع سیلاب یا قابلیت استحصال آب دارد. عدم پراکندگی مناسب ایستگاه‌های اندازه‌گیری بارش در مناطق مختلف، سبب گردیده است تا محققان و پژوهشگران به دنبال ایجاد مدل‌‌هایی باشند که بتوانند مقدار بارش را در مناطق فاقد یا دچار کمبود ایستگاه باران‌سنجی برآورد نمایند. در پژوهش حاضر از داده‌های مدلسازی شده Era5 که جدیدترین محصول مرکز اروپایی ECMWF می‌باشد و دقت خروجی‌های آن در منطقه خاورمیانه و ایران چندان مورد بررسی قرار نگرفته ، استفاده شده است. به منظور ارزیابی داده‌های بارش Era5، از داده‌های بارش 51 ایستگاه باران‌سنجی در استان خراسان رضوی طی سال‌های 1376 الی 1396 استفاده شد. برای بررسی خروجی‌های مدل مذکور، دو گروه شاخص ارزیابی کیفی (POD، FAR و CSI) و کمی (RMSE، NSE و R) مورد استفاده قرار گرفتند. نتایج نشان داد که خروجی‌های Era5 در مقیاس روزانه دارای خطای نسبتاً زیادی هستند (میانگین منطقه‌ای شاخص‌های کمی: 54/0 = R، 12/0=NSE، 47/2= و RMSE و میانگین منطقه‌ای شاخص‌های کیفی: 96/0=POD، 79/0=FAR، 21/0=CSI). در حالی‌که این داده‌ها در مقیاس‌های ماهانه و فصلی به خصوص فصلی (میانگین منطقه ای: 89/0= R، 45/0= NSE و 87/33=RMSE) از عملکرد خوبی برخوردار بوده و در صورت حذف خطای اریبی، می‌توانند در تحلیل‌های مختلف مورد استفاده قرار گیرند.

کلیدواژه‌ها

• ارزیابی بارندگی
• مدلسازی بارندگی
• ECMWF
• NSE
• POD

موضوعات

• سنجش از دور

عنوان مقاله [English]

A Qualitative- Quantitative Evaluation of Era5 Rainfall Data in Identifying the Occurrence and Amount of Rainfall in Razavi Khorasan Province

نویسندگان [English]

• Alireza Nouri ¹
• Samira Noori ²
• javad omidvar ³
• Hossein Banejad ⁴
• Fereshteh Modaresi ⁵

¹ Agrotecology Dept. , Faculty of Agriculture, Ferdowsi University of Mashhad, Mashhad- Iran.

² Water Sciences and Engineering Dept. Faculty of Agriculture, Ferdowsi University of Mashhad, Mashhad-Iran.

³ Water Sciences and Engineering Dept. Agriculture, Ferdowsi University of Mashhad, Mashhad, Iran.

⁴ Water Science and Engineering Dept. Faculty of Agriculture,, Ferdoqwsi University of Mashhad, Mashhad, Iran

⁵ Water science and Engineering Dept. Faculty of Agriculture,, Ferdowsi University of Mashhad-= Mashhad- Iran

چکیده [English]

Extended Abstract

Introduction:

Accurate measurement of the rainfall, as one of the essential meteorological variables in hydrological and agricultural studies, has a significant impact on the optimal management of water resources, cultivation of agricultural products, urban management, and identification and classification of areas in terms of the occurrence of floods risk or the ability to extract water. The inappropriate distribution of rainfall measurement stations in different regions has caused researchers to seek to create models that can estimate the amount of rainfall in areas that lack rain gauge stations or have Insufficient of them.

Methods:

In the current research, the data from Era5, the newest product of the ECMWF, has been used. Era5 data is the fifth generation of ECMWF reanalysis data for climate over the past several decades. These data provide hourly estimates of important atmospheric variables at the surface of the earth and other pressure levels. Era5 data is updated with a delay of 5 days and is available in a gridded form with an accuracy of 0.25 degrees (equivalent to 31 km). In the present study, hourly data in NetCDF format were used during the period from 1997 to 2018. The accuracy of Era5 in the Middle East and Iran has yet to be investigated much. In order to evaluate the accuracy of Era5 model data, the data of 51 rain gauge stations of Iran Water Resources Management Company in Razavi Khorasan province from 1997 to 2017 were used were used. The mentioned stations have a suitable distribution and length of statistical period compared to other types of stations. To evaluate the data of Era5 model, first using Python programming and the nearest neighbor method, NetCDF data was processed and rainfall values for each station were extracted. Then the hourly data were converted to daily. The modeled values compared to the observed values were evaluated using two groups of quantitative and qualitative indicators. The first group of indicators that were used for quantitative evaluation of observed and modeled rainfall values include root mean square error (RMSE), coefficient of explanation (R2) and Nash‐Sutcliffe Efficiency (NSE). The second group of indicators was used to evaluate the quality of Era5 model outputs. Using these indicators, the occurrence or non-occurrence of precipitation (regardless of the amount of recorded precipitation) was investigated in both data groups. The indicators of the second group include probability of detection (POD), false alarm ratio (FAR) and critical success index (CSI). In this group of indicators, the occurrence or non-occurrence of recorded precipitation is recorded as yes or no by observational data and modeling. Using these indicators, it is possible to determine the occurrence or non-occurrence of precipitation by the model.

Results:

According to the quality indicators of Era5 rainfall data on a daily scale, the results showed that the said data have a high ability to identify the days that rainfall occurred at the station. It is worth noting that qualitative indicators can only be calculated and evaluated for the daily scale. The results of the investigations show that the Era5 model recorded the maximum rainfall values on a daily scale with a relatively large difference lower than the observed values. The results of the statistical evaluation of the monthly rainfall of the stations in comparison with the estimated values of the model show that unlike the low coefficient of explanation of the observed and modeled data on a daily scale, the coefficient of explanation has increased significantly on a monthly scale so that the range Its fluctuation ranges from 0.45 in Ferizi station to 0.86 in Sarakhs station. As the time scale increases, the accuracy of Era5 model estimates also increases. The results show that the dynamic and numerical methods used in the Era5 model have the ability to provide estimates that are close to reality with the least error and have the ability to improve in providing more optimal results in other areas. These results show that Era5 model estimates have performed best on monthly and seasonal scales in regions with a warmer climate and less rainfall in Razavi Khorasan province. These results show that Era5 model estimates performed best in monthly and seasonal scales in areas with warmer climate and less rainfall in Razavi Khorasan Province. Therefore, if the skew error is fixed, the data of this model can be used as input data to agricultural and hydrological models.

کلیدواژه‌ها [English]

• Precipitation assessment
• Precipitation modeling
• ECMWF
• NSE
• POD