ارزیابی شبیه‌سازی بارش با استفاده از مدل wrf/wrf-hydro ( مطالعه موردی: حوضه ابوالعباس)

هر ساله وقوع بارش‌های سنگین در حوضه‌های سیل‌خیز کشور منجر به رخداد سیلاب و خسارت‌های ناشی از آن می‌گردد. بنابراین پیش بینی دقیق بارندگی برای انجام اقدامات پیشگیرانه از اهمیت بالایی برخوردار است. از این رو پیش‌بینی بارش یکی از مهمترین مسائل در حوزه مدیریت منابع آب می‌‌باشد. تاکنون روش‌های مختلفی برای پیش‌بینی بارش استفاده شده‌است. پژوهش‌های هیدرولوژیکی پیشرفته در پیش-بینی‌های کوتاه مدت هواشناسی نامشخص است و هنوز در دست بررسی و درک تاثیر آن است . انتظار می‌رود مفهوم اتصال مدل هیدرولوژیکی (wrf-hydro) با مدل جوی (wrf)عدم قطعیت های مربوط به توزیع مکانی و زمانی وقایع طوفان، به ویژه برای مناطق با خصوصیات پیچیده را کاهش دهد. در مطالعه حاضر به بررسی ارزیابی مدل wrf/wrf-hydro جهت پیش بینی 4 رویداد بارش که منتج به وقوع سیلاب شده اند، پرداخت شد. در شبیه‌سازی‌های بارش این مدل کم تخمین بود و مدل در حالت جفت‌‌شده نتایج بهتری ارائه داد. جهت اجرای مدل از داده های era5 استفاده شد که این داده ها عملکرد مطلوبی برای مدل در حوزه مذکور داشتند. همچنین برای پیکربندی مدل از سه طرح واره خرد فیزیک lin، thompson و wsm6 استفاده شد و با توجه به معیارهای ارزیابی خطا rmse و nse هر سه این طرحواره ها عملکرد مشابهی داشتند.

evaluation of rainfall simulation using wrf/wrf-hydro model (case study: abol-abbas basin)

introduction every year, the occurrence of heavy rains in the flood-prone basins of the country leads to the occurrence of floods and the resulting damages. therefore, accurate prediction of rainfall is of great importance to take preventive measures. so far, various methods have been used to predict rainfall. the concept of connecting the hydrological model (wrf-hydro) with the atmospheric model (wrf) is expected to reduce the uncertainties related to the spatial and temporal distribution of storm events, especially for areas with complex characteristics. in this study, the effectiveness of the wrf/wrf-hydro model in rainfall forecasting and the evaluation of this model in coupled and uncoupled mode for several flood events in the abu al-abbas catchment are investigated.methodology in this study, the wrf/wrf-hydro model was used to simulate four rainfall events in 1979, 1983, 1984 and 1993 that led to floods. wrf is designed as an open source and very flexible, so that it can be implemented on a wide range of computing environments from supercomputers to personal computers. in this study, the simulation area consists of three nested grids with horizontal grid spacing of 9, 3, and 1 km in the horizontal direction. the simulations performed in this study, initial and boundary conditions are provided by era5 reanalysis data. lin, thompson and wsm6 microphysics were used to configure the model. wrf-hydro is one of the free extensions of the wrf model, which is a distributed, physical and multi-scale hydrologic-hydraulic model. therefore, wrf-hydro is not an independent hydrological modelling structure, but a coupled system of hydrological models with atmospheric models, which were used in this study in two ways, coupled and uncoupled.results and discussion coupled simulation of all three microphysics had almost similar performance for all four events. in 1993, out of three rainfall peaks, only the first rainfall peak was simulated by the model, and the model performed relatively poorly for the next two peaks. the performance of the model in coupled mode is somewhat better than uncoupled mode. the 1979 performance in the coupled mode was almost identical to the real data and the performance of the model was very good. according to the precipitation curves, it can be concluded that in the general state of the wrf/wrf-hydro model, both in its coupled and uncoupled states, it is underestimated. the amount of error in the coupled and uncoupled state in all three schemas are almost equal and often the amount of error in the coupled state is slightly better than the uncoupled state. the error in 1993 was higher in both coupled and uncoupled modes than in other years, which could be due to errors in era5 data or observational data for that year.conclusion advanced hydrological research in short-term weather forecasting is uncertain and its impact is still being investigated and understood. the concept of connecting the hydrological model (wrf-hydro) with the atmospheric model (wrf) is expected to reduce the uncertainties related to the spatial and temporal distribution of storm events, especially for areas with complex characteristics. in this study, the results indicated that the wrf model has a good capability in predicting heavy rainfall and this model can be used to predict rainfall in mountain basins.