HR: 1330h
AN: H22D-0960 [PDF]
TI: Application of Remote Sensing Data in a Distributed Hydrological Model for the Gambia River Basin
AU: * Stisen, S
EM: simonstisen@wanadoo.dk
AF: Institute of Geography, University of Copenhagen
Oster Voldgade 10, Copenhagen K., DK-2800
Denmark
AU: Sandholt, I
EM: is@geogr.ku.dk
AF: Institute of Geography, University of Copenhagen
Oster Voldgade 10, Copenhagen K., DK-2800
Denmark
AU: Jensen, K H
EM: khj@geo.geol.ku.dk
AF: Geological Institute, University of Copenhagen
Oster Voldgade 10, Copenhagen K., DK-2800
Denmark
AB:
Distributed hydrological models have an extensive demand of high resolution spatial and temporal data for driving and
validating the models. Most of the variables are only available as point measurements which impose serious constraints to the
applicability and the credibility of such models particularly for large regional scales and in areas where the availability
and quality of hydrological data are limited. Remote sensing information appears to offer useful data that not only can fill
some of the gaps in data availability but also can supply data at the appropriate scale for distributed hydrological models.
In this study we tested three types of remote sensing derived variables in a distributed hydrological model of the 42,000 km2
Gambia River Basin in West Africa: (1) potential evapotranspiration estimated by the Makkink equation and based on daily
global radiation fields derived from the geostationary meteorological satellite Meteosat, (2) leaf area index (LAI) based on
data from the MODIS satellite, and (3) Temperature Vegetation Dryness Index (TVDI) derived form NOAA AVHRR images. The remote
sensing derived time series of potential evapotranspiration and LAI were used as input to the model while TVDI was used for
validating the spatial simulations of soil moisture.
The effects of introducing remote sensing based input were evaluated for both discharge simulations and spatial outputs by
comparing the model simulations to those based on traditional data. Application of remote sensing based input of potential
evapotranspiration and LAI had in both cases little effects on the simulated discharges while some effects were seen on the
spatial and temporal variation of variables like actual evapotranspiration and soil moisture. Improved spatial simulations of
these variables may potentially allow for better design of e.g. irrigation schemes. The comparative analysis of TVDI
estimates and spatial model simulations of soil moisture content in the root zone was however inconclusive.
DE: 1836 Hydrologic budget (1655)
DE: 3360 Remote sensing
SC: Hydrology [H]
MN: 2003 Fall Meeting