| In recent years,the global climate is experiencing a major change characterized by warming.Climate change not only changes the structure and function of water resources,but also has a direct impact on soil erosion by changing the amount of precipitation and its characteristics,wind speed and temperature.On the other hand,land use/cover change has changed the original surface vegetation type and its coverage,runoff status and physical and chemical properties of soil,and then the dynamic and anti erosion resistance system of soil erosion,becoming an inducement and a reinforcing factor of soil erosion.Systematic evaluation of their relative roles in soil erosion is of great significance for understanding the soil erosion process and carrying out comprehensive prevention and control of soil erosion.However,the soil erosion process is complex and the influencing factors are numerous and intertwined.It is difficult to assess the relative effects of climate change and land use/cover change on soil erosion under relatively independent conditions,as a weak link of soil erosion research.Moreover,previous studies have taken into account both climate change and land use/cover change that affect soil erosion.However,most of the temporal and spatial variation characteristics of soil erosion intensity are qualitatively analyzed,failed to give a quantitative estimate of their contribution,with the existence of comprehensive,poor data integrity and accuracy of the lack of single scale and interpretation of one-sided and so on.Therefore,the quantitative identification of the contribution of climate change and land use /cover changes to soil erosion needs to be strengthened.Based on the establishment of soil erosion prediction model,the future climate change is forecasted,and the land use/cover change scenarios is constructed.Taking quantitative research as the core and qualitative analysis as auxiliary method.The impacts of individual climate change,individual land use/cover change,climate andland use/cover change on soil erosion were revealed.Sediment yield is used as a public index to measure the relative effects of climate change and land use /cover change in soil erosion process,for quantitative evaluation of the contribution rate of both to sediment yield,and then determine the extent of the impact on soil erosion.In this paper,the Dongliao River Watershed is located in the middle part of the black earth area in Northeast China,is chosen as the study area.In view of the applicability and feasibility of soil erosion prediction model.Firstly,the soil erosion prediction model is constructed by using the radial basis function(RBF)neural network model and the distributed hydrological model SWAT,and the soil erosion prediction model suitable for the watershed is selected by comparing and analyzing.Secondly,using linear regression test,Spearman rank correlation method,Mann-Kendall rank correlation method,sliding T test and so on,to reveal the trend of climate change in 1960-2010 and the mutation point in the study area.The nonlinear principal component analysis method is used to deal with the large scale weather forecast factor data sets.A statistical down-scaling model of climate prediction is established to predict the climate change in the SRES,A2 and B2 emissions scenarios of the Dongliao River Watershed.Based on the soil erosion prediction model,the annual soil erosion modulus in the basin under the climate change background was calculated.The land use/cover change scenarios were constructed by using the hypothetical scenario method combined with the GIS method(LUCS1: the expansion of urban land;LUCS2: ecological restoration of farmland).Using soil erosion prediction model,the spatial distribution map of soil erosion area in two scenarios was obtained.The soil erosion intensity index was used to analyze the soil erosion intensity of different land use types.Finally,five kinds of climate and land use /cover changes simulation scenarios are constructed.Combined with soil erosion prediction model,the temporal and spatial distribution characteristics of soil erosion intensity under different scenarios were revealed by statistical analysis.According to the principle of sediment yield balance,the contribution rate of climate change and land use/cover change to the sediment yield of the basin was calculated quantitatively,so as to judge the influence degree of both on soil erosion.The main conclusions are as follows:(1)Both RBF neural network model and SWAT model can be used as a soil erosion prediction model in the Dongliao River Watershed.But the establishment of RBF neural network model needs a large amount of basic data as training samples.If the input data is not sufficient,it will affect the accuracy of the neural network or can not work.And it changes the characteristics of soil erosion into numbers,and the process of numerical calculation may lead to the loss of information.At the same time,by comparison with 2011-2012 in the Dongliao River Watershed average soil erosion modulus measured statistics,simulation results of RBF neural network model of the relative error is 0.04;the simulation results of SWAT model relative error is 0.02.The results show that the SWAT model is more accurate and is more suitable for the prediction model of soil erosion in the Dongliao River Watershed.(2)In the 1960-2010 years,the average annual precipitation in the basin showed an obvious downward trend,and the precipitation tendency rate was-7.13mm/10a;Generally,the temperature in the basin shows an upward trend,with the average daily maximum temperature linear tendency rate being 0.02 ℃ /10 a,the average annual daily minimum temperature of 0.05 ℃ /10 a linear trend rate,the annual average temperature of 0.04℃/10 a linear tendency rate.Compared with the baseline period,the A2 scenario model decreased by 4.68%,9.94% and 18.32% in 2020 s,2050s and2080 s,respectively;In the scenario model of B2,the decline in water was 2020 s and2080s respectively,increased by 3.51%,7.74%,in the 2050 s showed a downward trend,with the decrease of 2.34%.Under the A2 scenario model,the average annual temperature in 2020 s,2050s and 2080 s increased by 4.45%,8.40% and 15.28%respectively;Under the B2 scenario model,the annual warming rates of 2020 s,2050s and 2080 s are 3.46%,5.27% and 5.77%.,respectively.As a whole,the precipitation and temperature in the Dongliao River Watershed both showed an upward trend under the scenarios of A2 and B2.(3)Compared with the baseline period,under the A2 scenario,the average annual soil erosion modulus increased by 2.68%,5.51%,14.20% in the next three periods of 2020 s,2050s and 2080s;Under the B2 situation,the average soil erosionmodulus of 2020 s and 2080 s increased by 0.04%,4.55%,and 2050 s by 5.77%.On the whole,it can be found that the impact of climate change on soil erosion in the future is increasing,and the A2 situation is more influential than the B2 situation on the soil erosion in the watershed.(4)The spatial distribution characteristics of soil erosion intensity under different scenarios of land use/cover change are obviously different.Based on the present situation of land use in the Dongliao River Watershed in 2000,the expansion of urban land led to lower reaches of micro erosion is mild erosion was replaced by the basin in the middle and upper reaches of moderate erosion,strong erosion and severe erosion area increased significantly;ecological restoration of farmland in the lower reaches of river basin caused by micro erosion area increased,watershed upstream area moderate erosion,strong erosion and severe erosion area reduced.This indicates that the improvement of vegetation coverage can effectively control the occurrence and development of soil erosion in the watershed.Through the calculation of soil erosion intensity index,dry land,township land types of soil erosion intensity index is higher,should focus on regional soil erosion as the Dongliao River Watershed comprehensive management.(5)Through comparing of the annual soil erosion modulus in simple climate change,simple land use / cover change and climate and land use / cover change,the combined effects of climate change and land use/cover change on soil erosion are not a simple superposition of the individual effects.The temporal and spatial distribution of soil erosion intensity can be seen,compared with the baseline state,under the LUCS1+A2 scenario,will lead to an increase in the overall soil erosion intensity;Under the LUCS2+A2 scenario,the soil erosion intensity in some areas of the lower reaches of the basin is mild and moderate erosion.Under the LUCS1+B2 scenario,can cause soil erosion in the upper reaches of the basin,The soil erosion intensity of the lower reaches of the basin shows moderate erosion,and the soil erosion in the upper reaches of the basin presents a very strong and severe erosion situation;Under the LUCS2+B2,the intensity of soil erosion in the middle and upper reaches of the basin is slight and slight erosion.However,the soil erosion intensity in the lowerreaches of the basin is at a slight level,which tends to moderate to moderate erosion.By calculating the contribution rate of sediment yield,it is found that,Under the LUCS1+A2 scenarios,the average contribution of land use /cover change on sediment yield increase rate is 57.30%,the average contribution of climate change rate is 42.7%;LUCS2+A2 scenarios,land use/cover change to reduce sediment yield contribution rate is 81.58%,the average contribution of climate change rate is 18.42%;LUCS1+B2 scenarios,the average contribution of land use /cover change on sediment yield increase rate is 65.92%,the average contribution of climate change rate is34.08%;LUCS2+B2 scenarios,the average contribution of land use/cover change to reduce sediment yield rate is 77.32%,the average contribution of climate change rate is 22.68%.Thus it can be seen that the land use/cover change plays a leading role in the soil erosion under the combined effects of both.Therefore,the emphasis of soil erosion control and soil and water conservation management in the Dongliao River Watershed lies in the rational regulation of land use patterns. |
PDF Full Text Request