Nghiên cứu khai thác dữ liệu dự báo mưa hạn dài cho lưu vực sông Trà Khúc

0 Người đánh giá. Xếp hạng trung bình 0

Tác giả: Vi Nghiêm Đặng, Lê An Ngô, Thị Thu Hà Nguyễn

Ngôn ngữ: vie

Ký hiệu phân loại:

Thông tin xuất bản: Tạp chí Khoa học Kỹ thuật Thủy lợi và Môi trường, 2022

Mô tả vật lý: tr.112

Bộ sưu tập: Metadata

ID: 332591

Thêm vào giỏ Liên kết toàn văn

Dự báo tài nguyên nước hạn dài đóng vai trò quan trọng trong các bài toán lập kế hoạch sử dụng nước, quản lý tài nguyên nước và các hoạt động khác. Cùng với sự phát triển của khoa học công nghệ, các sản phẩm mưa dự báo số trị hạn dài ngày càng được nghiên cứu nâng cao chất lượng cũng như mức độ chi tiết. Nghiên cứu này tập trung đánh giá và nâng cao khả năng khai thác dữ liệu dự báo mưa mùa ECMWF-System5 thuộc Trung tâm dự báo hạn vừa châu Âu (ECMWF) ứng dụng cho lưu vực Trà Khúc với thời gian dự kiến dự báo 6 tháng thử nghiệm cho giai đoạn 1993-2016. Hai phương pháp hiệu chỉnh được xem xét là phương pháp hệ số tỷ lệ và phương pháp hồi quy tuyến tính. Kết quả cho thấy, cả hai phương pháp đều giúp cải thiện sai số tuyệt đối trung bình MAE giảm từ trung bình trên 50mm/tháng xuống dưới 10mm/tháng. Tuy nhiên, phương pháp hệ số tỷ lệ lại làm giảm hệ số tương quan R. Còn phương pháp hồi quy tuyến tính không làm thay đổi nhiều hệ số này. Nghiên cứu chỉ ra rằng, phương pháp hồi quy tuyến tính phù hợp để làm tăng chất lượng dữ liệu mưa dự báo hạn dài cho lưu vực Trà Khúc.Seasonal water resource forecasts play an increasingly important role in decision-making systems, especially in the agriculture and water sectors. Along with the rapid development of science and technology, seasonal forecasting has made progress in recent years, and global ensemble prediction systems provide increasingly accurate and reliable seasonal forecasting with up to 6–9 months’ lead time. This study focuses on evaluating the applicability of seasonal precipitation re-forecasts from the new ECMWF seasonal forecast system 5 (ECMWF-System5) for the Tra Khuc river basin at six-monthly lead times over the period 1993-2016. In addition, two calibration methods are used to bias correct the seasonal ensemble precipitation forecasts, including the scaling method and the regression-based method. A comparative evaluation of both raw and bias-corrected reforecasts is performed using mean absolute error (MAE) and correlation coefficient (R). According to MAE, both bias correction methods are able to reduce the MAE value from over 50mm/month to less than 10mm/month averaged over the basin and the forecasted months considered. In terms of the R evaluation metric, the scaling method decreases the R-value from 0.63 (for the raw reforecasts) to 0.55 (for the calibrated reforecasts) on average, while the regression-based method does not change this coefficient significantly...

1. Đọc trực tuyến

1. Đọc trực tuyến

Tạo bộ sưu tập với mã QR