Water Resources Assessment in the Huai River Basin : Hydrological Modelling and Remote Sensing Mr. Charles Baubion(1), Prof. Emmanuel Ledoux(1) , Prof. Li Jiren(2) , Prof. Ghislain de Marsily(3) , Dr. Xin Jingfeng(2) , and Dr. Huang Shifeng(2) (1) UMR Sisyphe, Paris School of Mines, 35, rue Saint-Honoré, 77305 Fontainebleau, France (2) IWHR, Beijing, Beijing, China (3) UMR Sisyphe, Université Paris VI, 4 place Jussieu, 75 Paris, France
Click here to view symposium presentation Click here to download compressed PDF (recommended for printing)

Abstract

The research study site (the Shiguanhe watershed) was selected with the Chinese partners of the project as a good example of the complexities of a well developed Chinese watershed. This watershed is located within the transitional area between the southern monsoon climate and the northern continental one. Furthermore, this meteorological duality is increased by the topography, with large plains with irrigated agriculture to the north of the Huai River on one side, occupying two thirds of the basin, and a mountainous area in the southern part, covered with a forest. Thus with precipitation of 600 mm per year in the north, the long northern tributaries run slowly toward the Huai River, where they meet the southern rivers, coming quickly from the mountains where the high precipitation (average of 1600 mm per year) occurs for 70 % from June to September, which is the flood season. Two large dams store water at the base of the mountainous region, and many small dams exist in the northern plain. Modelling this complex watershed is thus fully consistent with the objectives of the project.

CONTRIBUTION TO THE DRAGON PROJECT

Because the Shiguanhe basin is mostly mountainous, and made of crystalline rocks, its geological structure does not favour groundwater flow. Only the surface hydrological part of the MODCOU model was therefore used. The Shuttle Radio Topographic Mission (SRTM) DEM with a 3 arcsec resolution was used for the geometry: the river network extraction, the boundaries of the watershed, and the selection of the grid of the MODCOU. We used four sizes of meshes (125, 250, 500 and 1 000 m). A grid of around 73,000 meshes was obtained, on which the hydrological information can be introduced to run the flow model. The EO-generated soil map and land-use data served to prescribe initial soil reservoir properties for the model, which were later calibrated. Four classes of production functions were used : rice-fields, other agricultural fields, forests, and hill grassland. Rainfall and evapotranspiration data were prescribed, and the model was run for the years 1982-1991, comparing measured and observed flow rates at the outlet.

The Shiguanhe basin was divided into 3 parts: the 2 sub-basins controlled by the 2 large dams, and the northern sub-basin. By adding the discharge from the dams to the meshes at their outlets, we can obtain the value of the discharge of the whole basin at the hydrological outlet of Jiangjiali.

RESULTS SINCE THE 2005 SANTORINI SYMPOSIUM

The results display the two main hydrologic components of this watershed, irrigation and floods, and their management on the Shiguanhe basin. The spring season is usually marked by an over-estimation of the flow by the model, which is due to the irrigation management not taken into account. Some water is withdrawn from the river or stored in the small dams, so that the measured discharge in Jiangjiali is lower than the calculated one. When the summer is coming, more rainfall occurs. The farmers release the stored water from their small dams. This explains the large under-estimation of the model during flood-peaks, especially if we observe that some water is withdrawn for irrigation just after such events. In autumn, the calculated discharge is very close to the observed one. Concerning floods, we have good results in 1986, 1988, 1989, 1990 and 1991, but otherwise we underestimate the discharge, particularly when the largest floods occurred: this is likely due to the overflowing of many reservoirs all over the basin, and thus the non-representativity of the measured discharge from the two large dams. There is also uncertainty about precipitation data during heavy rainfalls, and furthermore, in crisis situations, the floods over the whole Huai River basin are centrally managed, the irrigation network can be used to transfer water to the Shiguanhe basin. It is thus important to understand the role of the dams, the impact of irrigation, and the management rules in the Shiguanhe and Huaihe basins.

These first results show a reasonable agreement between the two discharges, indicating that the major hydrological processes are included and correctly represented in the model and could thus provide indications on the water management in the Shiguanhe basin. But particular features of the Chinese landscape and water management system need to be better taken into account : water transfers between adjacent basins for irrigation or flood mitigation, and the presence and operation of a very large number of medium and small-size dams providing water for irrigation.

STUDY EXCHANGES

The Young European Scientist Mr. Charles Baubion worked on the project in Beijing, hosted by the LIAMA, Sino-French joint Laboratory in Informatics, Automatics and Applied Mathematics, Chinese Academy of Sciences, Institute of Automation, Beijing, from February 2005 to December 2005. See also the PIs reports.

EO DATA INVESTIGATED, GROUND CAMPAIGNS : See the PIs reports.

PLANS TO COMPLETE THE PROJECT

Remote sensing data and shape recognition will be tested to improve the characterization of the land-use (e.g. extent of the free-water area along the year), to determine the position of the water transfer and irrigation channels, and to understand the operation rules of this very complex water management system. The channels might be distinguished from the natural rivers, as they usually are rectilinear. There are two kinds of irrigation networks. First, those where the withdrawals in the river system supply a canal, that provides water to fields outside the Shiguanhe basin. Second, the irrigation network that provides water for fields inside the basin. This water is also withdrawn from the river, but a part of it returns to it by infiltration and hypodermic runoff. Once the model will satisfactorily represent the the basin, different management options will be tested, under present climate conditions or for future climate scenarios. See also PIs abstracts.

Keywords: ESA European Space Agency - Agence spatiale europeenne, observation de la terre, earth observation, satellite remote sensing, teledetection, geophysique, altimetrie, radar, chimique atmospherique, geophysics, altimetry, radar, atmospheric chemistry