Response Simulation Of Hydrological Processes And Soil Erosion On Slope And Watershed Scales To Forest Fires

Forest fire is very destructive to the forest,not only destroying the original forest ecological environment,but also causing huge economic losses.The annual frequency of forest fires in the world can reach several hundred thousand times,with an area of millions of hectares,.In the past 30 years,forest fires have been on the rise year by year.Although the investment in forest fire prevention in the world is increasing,the effect is not significant,.Especially in the late 1990s,millions of hectares of tropical rainforests were destroyed and burned by forest fires,which seriously damaged the global ecological balance.In recent years,more and more attention has been paid to the influence of forest fire on climate and weather,ecological environment and function,human and wildlife.Therefore,it is urgent to carry out a strong computer model to simulate and evaluate the forest soil erosion and management after fire,so as to provide decision-making and support for forest erosion risk prediction and forest restoration after fire.The water erosion prediction project(WEPP)based on the physical process has been widely verified to evaluate the soil erosion after fire,but the basin scale WEPP model has not been applied to the evaluation and verification of forest soil erosion after fire,which limits the scope and effectiveness of forest management.In this study,remote sensing technology is combined with the existing WEPP model to develop a new framework method.The framework method extends the ability of WEPP model to simulate soil erosion and hydrology at different scales(slope scale and watershed scale)after forest fire.The three decomposition objectives are as follows:(1)using the revised slope version of WEPP model to simulate runoff and erosion,and to evaluate the burning severity The influence of heavy degree on runoff and erosion on the scale of slope;(2)the observation data of runoff and erosion after fire are used to verify the WEPP Model of drainage basin version,to evaluate the impact of combustion severity disaster on runoff and erosion on the scale of drainage basin;(3)the spatial information related to road flow is extracted,divided and analyzed through the new framework based on GIS and lidar terrain data with high resolution,and then the spatial information related to road flow is evaluated To estimate the impact of roads on watershed scale runoff and soil erosion after forest fires.The results show that:Aiming at target 1,a structural framework was established to transform terrain discretization into representative slope used in WEPP model by using multiple runoff elements(ofe:overland flow element).The framework was applied to the watershed(study area)after forest fire.59%of the slope sediment yield in the watershed was greater than 1 Mg·ha-1·yr-1,and the runoff was greater than 5mm yr-1.Using the ofe structure framework developed in this study,the average sediment yield increased by 40%in general,.The reason for such a big difference is that before the ofe structure framework developed in this study was adopted,the small area of high combustion severity area included in the low combustion severity area was ignored,which led to the large error in the basin due to the small area of high combustion severity included in the low combustion severity area,leading to the lack of representativeness.The huge spatial difference between runoff and erosion indicates that the decision support tool of soil erosion simulation at slope scale should first define the spatial pattern of combustion severity.It is very important not only to capture the infiltration and deposition of the downhill in the slope with high combustion severity and small area,but also to capture the high combustion severity in the slope with low combustion severity.For target 2,based on 163 discrete hillsides in the basin,compared with the default WEPP treatment methods,and applied these methods(target 1)to the verification of runoff and erosion after fire in a 117 ha forest watershed in Eastern Arizona,USA.Compared with the default WEPP method,after calibration,the model can accurately simulate the runoff in the first two years after fire(NASH Sutcliffe efficiency,NSE:0.70-0.75).In the first year after the fire,the peak flow is slightly underestimated,but the simulation effect is better in the following years.The maximum daily peak value of the first and second year after the fire was 54.1mm and 74.8mm(NSE:0.30 and 0.76;R2:0.60 and 0.78).Slope erosion rate and sediment yield were simulated in the same way,R2 were 0.73 and 0.82 respectively.In the first year of the simulation,the hillside erosion rate leading to high combustion severity is 70-85 mgha-1,and the sediment yield is 2.7 Mg·ha-1·yr-1.The corrected model is further applied to the forest watershed of 207ha nearby,and performs well(NSE:0.57).Finally,the visualization strategy is developed to represent the WEPP processing structure,which can promote the practicability of this kind of model.Aiming at target 3,the road effect of forest soil erosion in the burning area of Emerald Bay on the southwest Bank of Lake Tahoe,California,USA was evaluated by using GIS technology and soil erosion model,and the possibility of quality damage caused by debris flow and translational landslide was studied.The results show that the road can change the flow path and sediment transport route,which can lead to the deposition of sediment and reduce soil loss after forest fire.The deposition pattern of pavement and depression predicted by WEPP is similar to that observed.The probability of debris flow is estimated to be about 35%,and no debris flow has been observed.However,with the root of dead trees rotting,it is likely that a translational landslide will occur within 4 to 6 years.In order to promote the application of the input and processing framework of WEPP model,an online program framework is developed,which allows the model interface to provide richer interaction between users and the front end of the model.In addition,the interface display of many model attributes(including topographic map representation)can ensure that the user can process the model correctly.The framework can be extended to input for other models,thus providing a better user experience.In this study,remote sensing technology is combined with forest soil erosion to study the impact of forest fire on soil erosion and hydrological characteristics,which provides a basis for forest soil erosion estimation based on remote sensing technology and soil erosion computer model.This study also shows that remote sensing data can be used to estimate forest soil erosion and generate scientific post fire assessment information on watershed scale.