You are here: Home / Research / Groups / Space Geophysics, Oceanography and Hydrology / Results / River and flooded areas by satellite altimetry

River and flooded areas by satellite altimetry

by GOHS last modified Jun 06, 2013 10:33 AM


Research on 'river and flooded areas by satellite altimetry' can be divided into three areas: validating altimetry river water levels collected during various satellite altimetry missions, interpreting time series in terms of climate variability, and integrating these series into hydrodynamic models.



Validating satellite altimetry river data

Subsequent to preliminary work conducted by GOHS team, which has revealed the specific problems of using river water levels obtained by radar altimetry, work has been initiated to build a database on continental surface waters (rivers, flooded areas and lakes), using Topex/Poseidon, Jason-1, ERS-2, ENVISAT and GFO missions (see HYDROWEB)

In this context, the team was responsible for producing water level time series calculated from satellite altimetry data gathered from several hundred 'virtual' stations located in ~ 10 major hydrographic basins around the world. Work addressed not only preprocessing problems, but also the difficulty of validating the atmospheric corrections that must be applied to the altimetric heights in order to calculate water levels. To test and validate corrections, GPS and ADCP campaigns were conducted in the Rio Negro basin of the Amazon region to compare the levels calculated from altimetry data against the levels measured in situ (Frappart et al. 2005a). The team is also working at validating ICESat laser altimetry river data (Calmant et al., 2005; Seyler et al.2005).

Interannual water level variability of rivers and flooded areas

Several studies based on water level time series calculated from satellite altimetry over several major river basins were conducted to study how surface water is affected by seasonal and interannual climate variability.Figure 1 below presents monsoon-induced seasonal flooding over the Mekong Delta plains. Although the monsoon season reaches its peak in August, flooding is the greatest in October.

Significant interannual variability was also observed (Frappart et al., 2005c). Another example is the La Plata basin in South America, which demonstrates the relationship between interannual water level variations on the Parana and Paraguay Rivers, rainfall and the 1997-1998 ENSO event (Maheu et al., 2003).

Hydrologie spatiale e-fig1

Figure 1 : Water level variations in the flooded plain of the Mekong Delta: the great flood of 2000 is related to a particularly intense monsoon season in 2000. In contrast, 1998 is characterised by little flooding (Frappart et al., 2005c). SPOT5 image of the Mekong Delta.

Study of river flows and surface water volumes

To quantify hydraulic flow, it is important to determine not only water level, but also discarge. Preliminary studies conducted by the GOHS team have been devoter to construct calibration curves to establish a relationship between in situ discharge measurements and altimetry water levels. The advantage of this approach is that with only 2-3 yeard of in situ measurements, calibration curve can be established which, in turn, can be used to calculate discarges time series over the duration of the satellite altimetry mission (currently 13 years for Topex/Poseidon) (Kouraev et al., 2004; Zakharova et al., 2005). An example for the Amazon River is shown in Figure 2.

Hydrologie spatiale e-fig2

Figure 2 : Estimate of the Amazon River flow rate at Juatarana based on Topex/Poseidon data (in red) and in situ measurements (in blue).
The bottom curve in green represents the difference between measured flow rates and measurements calculated from Topex/Poseidon data
(Zakharova et al., 2005)

Another aspect of studies conducted by the team (in cooperation with LMTG and CESBIO) involves determining time variations of surface water volume in large river basins.

Volume is obtained by combining altimetry to determine water levels in flooded areas and visible or radar satellite images to calculate surface area.

Two studies of this type were carried out: one on the Rio Negro Basin (Amazon region) (Frappart et al., 2005b) and the other on the Mekong basin (Frappart et al., 2006). Figure 3 shows evolution of water storage in the Mekong basin during the seasonal flood, based on combined analysis of water level by ENVISAT et NDVI data from Vegetation instrument on SPOT-4 (Frappart et al., 2006).

By using this method to determine spatio-temporal variations in surface water volume and by using GRACE data to compute total volume (soil water, plus surface water), it is possible to separate soil water and surface water, as well as any offset between surface water and deep water fluctuations.

A third aspect studied by the team is flow "routing", which entails incorporating altimetry water level data into hydrodynamic models (Leon et al., 2005) and measuring hydraulic gradients along rivers (Seyler et al., 2005).

Hydrologie spatiale e-fig3

Figure 3 : Evolution du volume d'eau dans le bassin du Mekong au cours de l'inondation saisonnière.

Contacts : S. Calmant, J.F. Crétaux, F. Frappart, K. Do Minh, A. Kouraev, E. Zakharova, M.C. Gennero,
N. Mognard-Cambell, A. Cazenave

Document Actions

logo cnes logo IRD Logo université de Toulouse Logo université Paul Sabatier Logo CNRS
Logo bibliothèque OBS Logo Observatoire Midi Pyrénées