Simulation Of Soil And Water Loss Variation Toward Terrace Practice In The Weihe River Basin

In recent years, runoff in many rivers of the world has been declined rapidly and even broken off due to population growth and the accelerated human activity influence. This problem has seriously affected regional social and economic development, which focuses people’s sight on the scientific question about the influence of human activity such as soil and water conservation measures toward rivers. Therefore, this research took the Weihe River as research object, which is a typical river with characteristics such as gross pollution, water shortage and high sediment concentration. Based on the research weakness of the soil and water conservation effects on rivers, this study developed a process-based terrace algorithm with independent intellectual property rights and incorporated it into the widely used Soil and Water Assessment Tool (SWAT) model. The response of soil and water loss and ecological base flow toward terraces of the Weihe River basin were detected using the newly developed model after its systematic verification. And an optimization technology was given based on the terrace model and the combined water and sediment analytical method and the economy analytical method. The above achievements preliminarily formed the basic theory for terrace simulation. The main conclusions were as follows:(1) A process-based terrace algorithm within the SWAT model was developed based on the preliminary "Watershed Self-response Theory" proposed in this research. This research proposed the concept of "Watershed Self-response Theory", which indicated that all parts of the watershed were connected and moving with the purpose of a balance statuse. The system will moving toward a new balance if the old one was interrupted. Based on this theory, a terrace simulation theory and method was developed with physical and mathematical models as the basic tools after analyzing the requirements of modelling terrace effects on river. The goal of the simulation theory was to reveal the water and nutrient cycle in the watershed level. Description method for different types of terraces and the segment simulation method were designed and incorporated into the SWAT model (version488with about18,000lines of newly added codes) using FORTRAN. The accomplishment of the terrace simulation technology fulfilled the tool needs of simulating the terrace effects toward soil and water losses in the watershed scale, which also improved the process-based effects simulation techology of soil and water conservation measures towards rivers.(2) The verification result indicted the satisfactory accuracy of the newly developed model and the feasibility of using the model for evaluating the soil and water loss response to terrace in both field and watershed scale. Observed runoff, sediment, nutrients and crop yield data of terraces in China and the U.S.A, and runoff and erosion data in typical small watershed in the Loess Plateau were employed for testing the newly developed model. The average balance testing indicators were less than0.1%which indicated that all functionalities of the model worked well. The field scale soil and water loss simulation Nash-Sutcliff Efficiency was usually higher than0.5which demonstrated that the model could reveal the runoff, sediment and nutrients loss processes in the field scale. The predicted runoff and erosion percent bias (PBIAS) in watershed level were between-16.2%and11.2%, and the modeled terrace regulation rate were similar to the measured data. Verification result proved the successful development of the new SWAT model and its accuracy in simulating runoff, sediment and nutrients yields from terrace in both field and watershed scale.(3) The newly developed SWAT model based on the "Watershed Self-response Theory" was used to simulated the runoff and sediment in the Weihe River and indicated the important effects of water conservation and conservancy measures on stopping river bed rising. The SWAT model was calibrated and validated using the highly varied runoff and sediment data collected in the1960s. The model could give satisfactory prediction for the main Weihe River runoff and sediment with NS and PBIAS ranging between0.52-0.93and-16.21%-28.35%. Based on the50years simulation result from1960to2009, the upstream and middle stream runoff in the drought season accounted for about89.4%of the whole river which was important for guaranteeing the ecological flow. The erosion mainly happened in the flood season which accounted for about86.4%of the annual sediment. The model also indicated that the up, middle and down-stream of the main river bed would have a4cm,10cm and20cm annual rise and lead to serious river deposition without human regulation, which was about1.5times of the observed value. This result indicated that the completed water conservation and other measures had great impact on stopping river bed rising.(4) Terrace in the main Weihe River basin could delay the flood and add the drought season runoff, prevent erosion and decrease river deposition, which were helpful for preventing soil and water loss and guaranteeing ecological base flow. Terrace in2000(about10%of the total basin area) could decrease about37million m3annual water yield in the whole watershed, and reduce16million tons of annual sediment transported in the Xianyang station. The most dry month runoff increased by3.5%and the short-term and long-term substandard days for ecological flow decreased by3.1and5.5days (accounted for26.3%and27.2%of the total substandard days) due to terrace practices. During1970s and2009, the construction of terraces had decreased upstream and middle stream sediment deposition by101and66million tons, which equals to9.1cm and22.8cm river bed lifting respectively. These effects were important for the main river controlling, flood preventing and water quality improving.(5) Optimization technology was achieved based on the terrace model and the combined water and sediment analytical method and the economy analytical method, which provided technical support for terrace distribution and structure optimizing, watershed planning and the smart decision. Based on the analysis result of4typical terrace optimization scenarios, the runoff and sediment reduction capacity of the bench terrace were higher than normal terrace, while the input-output ratio of the normal terrace was significant higher due to its cheaper price. It was recommended that erosion critical area should be firstly located for soil and water conservation measures optimization and reasonable distribution. Normal terrace should also be a consideration in the Loess Plateau eco-environmental construction for better erosion controlling effect and input-output ratio.