Nghiên cứu hiệu chỉnh sản phẩm dự báo mưa hạn nội mùa cho khu vực Bắc Trung Bộ bằng phương pháp phân vị với xấp xỉ Gamma (QM

Nghiên cứu hiệu chỉnh sản phẩm dự báo mưa hạn nội mùa cho khu vực Bắc Trung Bộ bằng phương pháp phân vị với xấp xỉ Gamma (QM–G)

0 Người đánh giá. Xếp hạng trung bình 0

Tác giả: Thị Thu Hương Hoàng, Văn Lượng Nguyễn, Thanh Hà Phạm, Văn Vinh Phan

Ngôn ngữ: vie

Ký hiệu phân loại:

Thông tin xuất bản: Tạp chí Khí tượng Thủy văn, 2023

Mô tả vật lý: tr.77-89

Bộ sưu tập: Metadata

ID: 339581

Thêm vào giỏ Liên kết toàn văn

Khoảng thời gian từ 10 ngày đến 2 tháng (hạn nội mùa) có ý nghĩa quan trọng trong quản lý nguồn nước, điều tiết hồ chứa, cả thuỷ lợi và thuỷ điện, cũng như lập nông lịch ở khu vực Bắc Trung Bộ. Tuy nhiên, cũng không thể phủ nhận được thực tế rằng, ở Việt Nam nói chung và Bắc Trung Bộ nói riêng bài toán dự báo mưa hạn nội mùa chỉ mới được đặt những viên gạch đầu tiên, kĩ năng dự báo vẫn còn thấp. Để có thể ứng dụng vào thực tiễn đòi hỏi phải có nhiều hơn các thử nghiệm, các công trình nghiên cứu hoặc tìm ra các công cụ dự báo mạnh mẽ hơn. Nhóm tác giả đã thực hiện nghiên cứu hiệu chỉnh dự báo mưa hạn nội mùa của mô ECMWF dựa trên phương pháp phân vị với xấp xỉ gamma (QM–G) cho độ phân giải thời gian 5 ngày với từng hạn dự báo (3–6 tuần). Nghiên cứu sử dụng hai bộ số liệu chính: bộ số liệu dự báo mưa hạn nội mùa của mô hình ECMWF cho 20 năm trong giai đoạn 2000–2019 và bộ số liệu mưa vệ tinh TRMM. Kết quả cho thấy: Sau khi hiệu chỉnh, các giá trị sai số như ME, MAE, RMSE giảm đi rõ rệt, đặc biệt là từ tháng 12 đến tháng 3 năm sau. Riêng tháng 7–9, tuy sai số đã giảm so với trước hiệu chỉnh nhưng vẫn còn nhiều vùng (chủ yếu là phía Nam khu vực) vẫn còn tồn tại sai số lớn. Tuy nhiên, giá trị tương quan CORR sau hiệu chỉnh không cải thiện, thậm chí có tháng còn giảm so với trước khi hiệu chỉnh.The timescale from 10 days to 2 months (sub–seasonal to seasonal) is important in water resource management, reservoir regulation, both irrigation and hydroelectricity, as well as agricultural scheduling in the North Central region. However, it is also impossible to deny the fact that, in Vietnam in general and the North Central region in particular, the sub season forecasting skills are still low. The authors corrected the sub–seasonal to seasonal rainfall forecast of the ECMWF model based on the quantile mapping method with gamma approximation (QM–G) for the total rainfall of 5 days with each forecast period (3–6 weeks). Two main data sets include: the ECMWF model's sub–seasonal rainfall forecast data set for 20 years in the period 2000–2019 and the TRMM satellite rain data set. The results show that: After correcting, the ME, MAE, RMSE values decreased significantly, especially from December to March next year. Besides, in July–September, although the bias has decreased compared to before the correction, there are still many areas (mainly in the southern region) that still have large bias. However, the correction by QM–G did not improve the correlation, even some month it decreased compared to before correction.

1. Đọc trực tuyến

1. Đọc trực tuyến

Tạo bộ sưu tập với mã QR