| Water resources are the irreplaceable material basis supporting all life on the earth and sustainable development of human society and economy. Runoff variation plays a leading role in the evolution of whole hydrology and water resources system, and also greatly affects regional resource environment as well as social and economic development. The research on runoff evolution rule is the premise and foundation for rational development and effective utilization of regional water resources. Because of relatively long forecast period, medium and long term runoff forecast can provide relevant department with overall planning basis for carrying out water project operation and water resources optimal allocation. However, under the integrated influence of general atmospheric circulation, solar activity, hydro meteorology factors, physical geography and many other uncertain factors, space –time evolution features of river runoff are diverse, compared with short term runoff forecast, the medium and long term runoff forcast is more difficult, severely lagging behind the need of actual production. However, China natural disasters like flood and drought happen frequently, how to improve the precision of the medium and long term runoff forecast becomes a serious problem.The statistical features of medium and long term runoff drive factors are relatively complicated in terms of time and space, and the practical application of factor forecast model is difficult to some extent. But, the runoff forecast model based on solid physical causes is recognized as an important development trend of the medium and long term runoff forecast. With the development of modern science and technology, it has been possible to collect hydrological data efficiently and in large area, water circulation system theoretical method emerges through establishing hydrological remote sensing and measuring system, by virtue of satellite, radar, and other spatial information technologies, with modern science and technology as well as computer as tools, which is one of hotspots of the medium and long term runoff forecast research at present.As everyone knows change of river runoff is not only related to climatic factors(rainfall, temperature, etc.), but also affected by land surface vegetation. The forest ecosystem hold up detention storage rainfall through canopy layer, litter layer and soil layer to effectively control soil moisture and regulate river runoff. Forest water conservation function is one of the major drive factors for river runoff variation. The traditional hydrology research usually conducts independent typical sample region test for canopy layer, litter layer and soil layer to analyze their hydrology effect. However, such method has complicated operation and low precision, and the point replacing surface method cannot show the water conservation situation of the forest of the whole drainage basin completely, concretely and truly. The research of the effect of forest on runoff drive needs the support of modern science and technology urgently.This paper introduces vegetation eco-water(layer) remote sensing quantitative inversion theory and method proposed by this scientific research team to river runoff forecast research. Vegetation eco-water(layer)(eco-water layer or eco-water for short) refers to the water body closely related to the earth surface plant material, earth surface vegetation layer specifically, including blade surface, root system, humus layer, and the water held back or conserved in surface layer vegetation root system soil layer and the plant material. Such water forms a special conversion or transition belt of water circulation around vegetation layer, with the main function of evaporation and transpiration as well as regulating and supplying surface water and ground water. According to the existing research achievements, in the original forest, the water held up by eco-water layer accounts for 30%-40% of the precipitation, even up to 60%, which cannot be underestimated. In the process of hydrological circulation, the amount of eco-water directly affects water resources allocation and the time of detention in storage body, thus influencing the change of river flow, in wet season, hold up and buffer the rainstorm to avoid flood and peak flow; in dry season, supply surface water and ground water to avoid rapid dry or cutoff of river valley.As the water–rich features of eco-water are special, it is difficult to make extraction and quantization with the conventional method, after many years of research by this project team, quantization of land surface forest hydrological effect is realized with 3S technology focusing on remote sensing quantitative inversion technology. Due to advanced eco-water(layer) research theory and technology, it is possible to reflect the water conservation function of forest more comprehensively, macroscopically and quantitatively, thus providing strong support for researching the influence of land surface forest on runoff change and establishing medium and long term runoff forecast factor model. Taking the upper reaches of Minjiang River for example, this paper performs inversion of eco-water resources quantity with remote sensing method, and establishes Eco-driven runoff forecast model.Main contents and innovations of this paper are shown below,(1) Based on the early research results of project team as well as establishment of eco-water conservation module model, with the upper reaches of Minjiang River as the research object, build eco-water remote sensing inversion database of this drainage basin from 1992 to 2005, realizing the quantitation of forest water conservation ability.(2) On the basis of establishment of eco-water remote sensing inversion database, analyze eco-water variation trend of the research area, eco-water reduces on the whole, and the reducing trend is relatively obvious in the period from 1992-1998, and the reduction becomes slow from 1998-2005. The change of eco-water resources quantity is significantly related to variation of forest vegetation area. Deforestation is the key reason for continuous reduction of eco-water resources quantity from 1992-1998, while from 1998-2005, in addition to natural restoration of vegetation, the implementation of policy of grain for green and forest planting is the reason for slow reduction of eco-water resources quantity.(3) Establish eco-water driven runoff forecast model. Analyze runoff reason from the point of view of drive, with rainfall, air temperature and eco-water as the three major drive factors for runoff, build eco-water driven runoff database for the upper reaches of Minjiang River, research and analyze the correlation between drive factors and runoff variation, and build eco-water driven runoff forecast model for the upper reaches of Minjiang River. Through conducting verification, precision analysis and significance test for the forecast model, the effectiveness and practicability of the model are proved, the results show that the annual runoff is mainly led by annual precipitation, while the runoff in dry season is closely related to the eco-water resources quantity. |
PDF Full Text Request