Using GRASS and PostgreSQL/PostGIS for the development of automatic preprocessing methods for a distributed vector-based hydrological model - INRAE - Institut national de recherche pour l’agriculture, l’alimentation et l’environnement
Communication Dans Un Congrès Année : 2012

Using GRASS and PostgreSQL/PostGIS for the development of automatic preprocessing methods for a distributed vector-based hydrological model

GRASS et PostgreSQL/Posgis pour le développement de méthodes de pré-traitements automatiques pour un modèle hydrologique à base vecteur

Résumé

Many GIS tools have been developed for the preparation of the geographical input data of distributed hydrological models such as the hydrological sciences toolbox of GRASS (http://grass.osgeo.org/wiki/Hydrological_Sciences), the raster terrain analysis plug-in of QGIS (http://www.qgis.org), the Taudem (http://hydrology.usu.edu/taudem/taudem5.0/index.html) extension of Mapwindow, the terrain analysis hydrology package of SAGA (http://sourceforge.net/apps/trac/saga-gis/wiki/ta_hydrology), etc. However, most of these preprocessing tools consider raster data, not vector data. Model meshes based on irregular vector geometries are mainly used by [1], [2] and [3]. Lagacherie et al. ([1]) developed the landscape discretization tool GEOMHYDAS in GRASS GIS, which allows the construction of vector-based HRU maps (Hydrological Response Units) based on a selective overlay of property layers such as land use, soil maps or sub-basins. Furthermore, special tools were developed to integrate man-made features such as ditches, roads and agricultural fields. Branger [2] simply used a land use map as model mesh. Rodriguez et al. [3] developed the preprocessing of the urban hydrological model URBS in MapInfo. Instead of HRUs, URBS uses Urban Hydrological Elements (UHEs) as model units, each consisting of one cadastral unit and half of the adjoining street. In this paper, we go one step further by combining the HRU model mesh with UHEs in order to create a model mesh adapted for peri-urban areas. This particular model mesh needed the development of several scripts for the extraction of the flow routing in a mixed urban and rural environment. In order to obtain realistic flow paths, mesh optimization methods were developed treating concave or too large polygons or polygons with holes. Furthermore, scripts were developed allowing the integration of raster information (e.g. slope) into the model mesh without creating a large number of small polygons.

Mots clés

Fichier non déposé

Dates et versions

hal-02597781 , version 1 (15-05-2020)

Identifiants

Citer

S. Jankowfsky, P. Sanzana, F. Branger, Isabelle Braud, Y. Paillé, et al.. Using GRASS and PostgreSQL/PostGIS for the development of automatic preprocessing methods for a distributed vector-based hydrological model. Open Source Geospatial Research & Education Symposium, OGRS 2012, Oct 2012, Yverdon les Bains, Switzerland. pp.6. ⟨hal-02597781⟩
23 Consultations
0 Téléchargements

Partager

More