Thông tin chung
  English
  Đề tài NC khoa học
  Bài báo, báo cáo khoa học
  Hướng dẫn Sau đại học
  Sách và giáo trình
  Các học phần và môn giảng dạy
  Giải thưởng khoa học, Phát minh, sáng chế
  Khen thưởng
  Thông tin khác
  Tài liệu tham khảo
  Hiệu chỉnh
 
Số người truy cập: 112,298,152

 Flood magnitude detection using satellite-based land surface temperature
Tác giả hoặc Nhóm tác giả: Hung T. Pham, Lucy Marshall, Fiona Johnson
Nơi đăng: General Assembly 2019 of the European Geosciences Union, Vienna, Austria; Số: 11705;Từ->đến trang: 31284;Năm: 2019
Lĩnh vực: Chưa xác định; Loại: Báo cáo; Thể loại: Quốc tế
TÓM TẮT
Floods can result in significant damages and loss of life especially in remote communities and in developing countries. People living in such areas frequently lack flood warning information to prepare for floods in advance due to sparse hydrological measuring networks and delays in accessing field data. Recently, it has been suggested that the large-scale spatial coverage and near-real time availability of remote sensing data may be suitable to identify potential flooding during extreme events and in data-scarce regions. While water levels, river morphologies and inundation extents can be measured directly from space, river discharge cannot be monitored directly from satellite sensors. Tremendous efforts have been devoted to estimate river discharge from these variables using rating curves, hydraulic models or hydraulic geometry. The ratio between brightness temperature at a land pixel and at a water pixel has also been used to estimate river discharge. However, an accurate, fast and universally robust method to estimate river discharge or flood flows from satellites has not yet been developed for use in ungauged and sparsely gauged areas. Here we demonstrate a new approach that uses the difference in day and night land surface temperature (DLST) derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) products for flood detection. In previous work, DLST has been assessed along with flood inundation qualitatively demonstrating that low DLST values are associated with flood events while dry seasons typically had high DLST values. We extend this analysis to quantify the relationship between flood flows and DLST values and understand the flood prediction potential of DLST. Flood flows were derived from 222 hydrological reference stations at unregulated catchments across contrasting hydro-climate conditions in Australia. The relationship between flood flows and DLST values was modelled using a regression tree technique governed by climate conditions, catchment properties and hydrograph characteristics. The results reveal that the DLST – flow relationships mainly depend on local climate conditions and catchment properties. Strong and moderate negative correlations were found in hot humid and cool temperate regions. Small positive correlations happened in mild temperate regions while strong positive correlations occurred in hot dry regions. The findings highlight conditions where DLST will be useful for river discharge estimation, flood predictions and flood warning system in sparsely gauged and ungauged regions.
ABSTRACT
Floods can result in significant damages and loss of life especially in remote communities and in developing countries. People living in such areas frequently lack flood warning information to prepare for floods in advance due to sparse hydrological measuring networks and delays in accessing field data. Recently, it has been suggested that the large-scale spatial coverage and near-real time availability of remote sensing data may be suitable to identify potential flooding during extreme events and in data-scarce regions. While water levels, river morphologies and inundation extents can be measured directly from space, river discharge cannot be monitored directly from satellite sensors. Tremendous efforts have been devoted to estimate river discharge from these variables using rating curves, hydraulic models or hydraulic geometry. The ratio between brightness temperature at a land pixel and at a water pixel has also been used to estimate river discharge. However, an accurate, fast and universally robust method to estimate river discharge or flood flows from satellites has not yet been developed for use in ungauged and sparsely gauged areas. Here we demonstrate a new approach that uses the difference in day and night land surface temperature (DLST) derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) products for flood detection. In previous work, DLST has been assessed along with flood inundation qualitatively demonstrating that low DLST values are associated with flood events while dry seasons typically had high DLST values. We extend this analysis to quantify the relationship between flood flows and DLST values and understand the flood prediction potential of DLST. Flood flows were derived from 222 hydrological reference stations at unregulated catchments across contrasting hydro-climate conditions in Australia. The relationship between flood flows and DLST values was modelled using a regression tree technique governed by climate conditions, catchment properties and hydrograph characteristics. The results reveal that the DLST – flow relationships mainly depend on local climate conditions and catchment properties. Strong and moderate negative correlations were found in hot humid and cool temperate regions. Small positive correlations happened in mild temperate regions while strong positive correlations occurred in hot dry regions. The findings highlight conditions where DLST will be useful for river discharge estimation, flood predictions and flood warning system in sparsely gauged and ungauged regions.
© Đại học Đà Nẵng
 
 
Địa chỉ: 41 Lê Duẩn Thành phố Đà Nẵng
Điện thoại: (84) 0236 3822 041 ; Email: dhdn@ac.udn.vn