A new approach for drought forecasting using wavelet-ANN model and satellite images
Publish Year: 1403
نوع سند: مقاله ژورنالی
زبان: English
View: 26
This Paper With 9 Page And PDF Format Ready To Download
- Certificate
- من نویسنده این مقاله هستم
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
JR_IJNAA-15-5_029
تاریخ نمایه سازی: 18 فروردین 1403
Abstract:
Due to different influencing factors, drought is difficult to forecast. Hence, robust and accurate forecasting methods are needed. A method was presented to improve the accuracy of drought forecasts using the wavelet neural network and proximity information in satellite images. Satellite precipitation and evapotranspiration data were applied to calculate drought indices. And the drought intensity in different months of the following year was forecasted using the wavelet neural network method. To increase forecast accuracy and discriminate random changes from drought signals, proximity data in satellite images were used to forecast drought at the East Isfahan climate station. The results showed that the wavelet neural network method is able to forecast drought with reasonable accuracy. Also, using adjoining data may improve forecasting precision. The correlation between the target and predicted values was ۰.۶۷۵.
Keywords:
Authors
Maedeh Behifar
Department of Remote Sensing and GIS, Faculty of Geography, University of Tehran, Tehran, Iran
Ata Abdollahi Kakroodi
Department of Remote Sensing and GIS, Faculty of Geography, University of Tehran, Tehran, Iran
Majid Kiavarz
Department of Remote Sensing and GIS, Faculty of Geography, University of Tehran, Tehran, Iran
Ghasem Azizi
Department of Geography, Faculty of Geography, University of Tehran, Tehran, Iran
مراجع و منابع این Paper:
لیست زیر مراجع و منابع استفاده شده در این Paper را نمایش می دهد. این مراجع به صورت کاملا ماشینی و بر اساس هوش مصنوعی استخراج شده اند و لذا ممکن است دارای اشکالاتی باشند که به مرور زمان دقت استخراج این محتوا افزایش می یابد. مراجعی که مقالات مربوط به آنها در سیویلیکا نمایه شده و پیدا شده اند، به خود Paper لینک شده اند :