Meso-scale modelling of runoff and water balances using remote sensing and other GIS data*

For about two decades remote sensing (RS) data in the form of multi-spectral air photography, ground based weather radar as well as geostationary and polar orbiting satellite data have been available for use in hydrology. It is the intention of this paper to advocate the use of RS data in the field of hydrological modelling. It is shown that the main field of application of RS data in hydrological modelling lies in the mesoscale. The potential of RS in the framework of GIS is shown along with three examples. The first example shows the use of RS data for rainfall/runoff modelling in the lower meso-scale. A model was developed which used the high spatial resolution of Landsat polar orbiting satellites (30m X 30 m) for model parameter estimation. The second example shows the use of geostationary satellite data for the computation of long time series of monthly runoff values in the upper meso-scale. It was applied in West Africa for the purpose of water resources planning. In the third example, a water balance model in the upper meso-scale is presented which computes daily values for all the variables in the water balance equation along with an example of the model's application in the international Mosel River basin. Here also the detection of land use changes by multi-temporal satellite imagery and their impact on hydrological parameters is discussed. In this example, emphasis is put on die combination of RS data with other information, such as digital elevation models or digitized maps, all within the framework of GIS. In conclusion, not only the potential but also the limitations of the use of RS in hydrological modelling are discussed. Utilisation de données provenant de la télédétection et de systèmes d'information géographiques pour la modélisation à méso-échelle des écoulements et des bilans en eau Résumé Depuis environ vingt ans des données télédétectées sont à la disposition de l'hydrologie, qu'il s'agisse de photographies aériennes selon différentes gammes de longueur d'onde, d'images fournies par les radars météorologiques au sol ou par des satellites géostationnaires ou à orbite polaire. L'objet de cet article est de plaider en faveur de l'utilisation des données télédétectées dans le domaine de la modélisation hydrologique. On montrera que leur principal champ d'application se situe à méso-échelle. Les possibilités associées aux données télédétectées *Keynote Paper from Symposium J1.1 : Remote Sensing Applications to Large-scale Hydrological Modelling (Yokohama, July 1993). Open for discussion until 1 October 1994