In:
ISPRS International Journal of Geo-Information, MDPI AG, Vol. 9, No. 11 ( 2020-10-26), p. 634-
Abstract:
Catchment division constitutes the foundation for urban water flood forecasting but represents a technically challenging task. The accurate division of catchments is significant for precisely forecasting urban waterlogging. However, existing catchment division methods usually lead to produce results that do not accurately reflect the actual land-use distributions. In recent years, most research has been performed in smaller study areas (less than 10 km2), in residential areas, parks and campuses, and usually focused on a single landscape type. However, for large highly urbanized areas with complex land uses, due to the spatial heterogeneity and complexity of such areas in terms of building, traffic network and hydrology, etc., there is few studies on sub-catchment division. Moreover, the division results by using existing method usually have deviate with the actual land-type distributions. To address the above-mentioned issues, a sub-catchment division method was here proposed that accounts for land-use types and flow directions, and it is suitable for large urban areas by introducing an auto-adaptive threshold adjustment in a novel algorithm. First, the study area is divided into first- and second-level (FL and SL, respectively) catchments according to the macroscale features such as natural landforms, canals, and pipe network. Second, an amended DEM (Digital Elevation Model) and flow direction data are used to divide the SL catchments into third-level direction-based (D-B) catchments. Finally, a novel land use-based algorithm is proposed to divide the D-B catchments into the “smallest” catchments (S-catchments). A large-scale area (44 km2) in Dongying City of China was employed to validate the proposed method. The experiment showed that the proposed method is suitable for subcatchment divisions in large regions and can ensure that the subcatchments are consistent with the actual distribution of land uses and runoff directions.
Type of Medium:
Online Resource
ISSN:
2220-9964
Language:
English
Publisher:
MDPI AG
Publication Date:
2020
detail.hit.zdb_id:
2655790-3
Permalink
