Atmospheric parameter prediction using GNSS radio occultation data and machine learning: A case study in central Iran
محل انتشار: بیست و یکمین کنفرانس ژئوفیزیک ایران
سال انتشار: 1403
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 112
فایل این مقاله در 6 صفحه با فرمت PDF قابل دریافت می باشد
- صدور گواهی نمایه سازی
- من نویسنده این مقاله هستم
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
GCI21_195
تاریخ نمایه سازی: 1 بهمن 1403
چکیده مقاله:
This study employs GNSS radio occultation (GNSS-RO) data from the AWS GNSSro repository to predict atmospheric parameters including temperature, pressure, and water vapor pressure in the central area of Iran covering the period from January ۱, ۲۰۰۷, to September ۱۳, ۲۰۰۸. The geographical focus spans longitudes ۴۸ to ۵۳ and latitudes ۳۳ to ۳۵. Utilizing a machine learning approach, specifically a RF Regressor, we processed and analyzed the GNSS data to extract relevant atmospheric variables. The methodology included rigorous preprocessing steps, such as feature engineering and imputation of missing values, to enhance model accuracy. Our results demonstrate the effectiveness of the Random Forest (RF) model in predicting the specified atmospheric parameters, with Root Mean Squared Error (RMSE) values of ۴.۴۲ C for temperature, ۵۸۹.۷ Pa for pressure, and ۲۱.۰۲ Pa for water vapor pressure. Feature importance analysis revealed that refractivity and altitude are the most significant predictors for temperature, while pressure gradient as well as temperature gradients also contributed notably. The findings demonstrate the potential of GNSS radio occultation data combined with machine learning techniques to improve atmospheric parameter predictions. This approach provides a robust framework for enhancing weather forecasting models and contributes valuable insights into the field of atmospheric science.
کلیدواژه ها:
نویسندگان
S Barzegar
Assistant Professor of the Department of Space Physics, University of Tehran, Institute of Geophysics, Iran
S Sabetghadam
Assocciate Professor of the Department of Space Physics, University of Tehran, Institute of Geophysics, Iran