Study On Soil And Water Loss Law And Prediction Model Of Pinus Massoniana Forest With Different Canopy Density In Red Soil Erosion Area Of Changting County

Changting of Fujian provinces one of the counties with the most serious soil erosion in the red soil hilly region in the south of China.After decades of continuous treatment,a large-scale preliminary management area dominated by pure Pinusmassoniana forest has been formed,there are widespread problems,such as large stand density,simple structure of understory vegetation,lack of undergrowth and etc.,and water and Soil loss under the Pinusmassoniana forest is still severe.It is a feasible measure to change the canopy density of the Pinusmassoniana forest by thinning,providing suitable conditions for the restoration of undergrowth vegetation and increasing the types and diversity of undergrowth vegetation,so as to reduce the problem of soil erosion under the Pinusmassoniana forest.However,the laws of water and soil loss of the Pinusmassoniana forest with different canopy density at present are still unclear,especially the canopy of the Pinusmassoniana forest with the best soil and water loss control effect was still not clear,and it has affected the continuous control of soil and water loss in the Pinusmassoniana forest in Changting red soil erosion management areas.Therefore,studying how to screen the canopy density of the Pinusmassoniana forest with the best soil erosion control effect in Changting red soil erosion area,to reveal the law of soil and water loss of the Pinusmassoniana forest with different canopy density,and how to construct the prediction model of soil and water loss of the Pinusmassoniana forest,which have important practical significance for the transformation of the Pinusmassoniana forest stand,the continuous management of soil erosion,the improvement of ecological functions,the construction of ecological civilization.In view of this,taking the Changting red soil erosion control area as the research object,and different canopy density?0.2,0.4,0.6,0.8?were established by thinning.Using the standard runoff plot observation method to investigate the Pinusmassoniana forest with different canopy density,to do dynamic observation of soil and water loss and meteorological factors in the Pinusmassoniana forest with different canopy density,and to compare annual and monthly changes in the yield,sediment yield,and nutrient loss of the Pinusmassoniana forest with different canopy density,to analyze the change laws of soil physical and chemical characteristics of the Pinusmassoniana forest with different canopy density,and to analyze the internal relationship between runoff and sediment yield,nutrient loss and meteorological factors.At the same time,using the dynamic observation data of soil erosion and rainfall factors of field rainfall in Pinus massoniana forest with different canopy density for one consecutive year to establish the natural rainfall gradient,multiple regression analysis and neural network analysis are used to analyze the runoff and sediment yield of the Pinusmassoniana forest with different density to carry out simulation and prediction analysis,and construct a prediction model of soil erosion for the Pinusmassoniana forest in Changting red soil erosion preliminary management area,which provide a scientific basis for the sustainable management of soil erosion and the accurate improvement of ecological function in the Pinusmassoniana forest in Changting red soil area.The main research conclusions are as follows:?1?Soil bulk density,porosity,water content and aeration of the Pinusmassoniana forest with different canopy density were showed:those of 0.4 and 0.6 are better than those of 0.2 and 0.8,especially those of 0.6 is generally better than other canopy density.The contents of total carbon,total nitrogen,total potassium and total phosphorus?2019?in the0-10cm soil layer of the Pinusmassoniana forest with canopy density of0.4 and 0.6 were significantly greater than those of 0.2 and 0.8,while there was no significant difference in other soil layers.?2?Species richness of pinusmassoniana forest with different canopy density was 0.6>0.4>0.2>0.8.The pinusmassoniana forest with the canopy density of 0.6 has the most plant species and the highest species richness,and it also has the highest species diversity index and richness index,and the best understory vegetation restoration effect.The canopy density of 0.6 is beneficial to vegetation restoration under the pinusmassoniana forest in Changting red soil erosion area.?3?The runoff,sediment loss and nutrient loss of the Pinusmassoniana forest with different canopy density in the initial treatment area of red soil erosion in Changting showed a law of increasing first and then decreasing with the increase of canopy density,and the change law of soil and water loss in the Pinusmassoniana forest with different canopy density was basically consistent with that of rainfall.Soil and water loss in the Pinusmassoniana forest with canopy density of 0.2 was the largest,followed by that with canopy density of 0.8 and that with canopy density of 0.6.During the monitoring period from 2017 to 2019,the runoff and sediment loss of the Pinusmassoniana forest with canopy density of 0.2,0.4 and 0.8 were fluctuated,while the canopy density 0.6 showed a trend of continuous decrease,indicating that the soil erosion control effect of the Pinusmassoniana forest with canopy density of 0.6 was better.?4?Under the condition of natural rainfall gradient,the runoff and sediment yield of the Pinusmassoniana forest with different canopy density in Changting red soil erosion preliminary management area was significantly positively correlated with rainfall,I30 and rainfall erosion,and are weakly correlated with rainfall intensity.The runoff and sediment yield with the canopy density of 0.4 and 0.6 and rainfall duration were extremely significantly positively correlated,and these with the canopy density of 0.2 and 0.8 were significantly positively correlated with rainfall duration.?5?Multiple regression analysis was used to analyze the relationship between soil erosion and rainfall factors of the Pinusmassoniana forest with different canopy density in the preliminary soil erosion area of Changting.The results showed that:the main driving factors of precipitation of abortion sands in the Pinusmassoniana forest with different canopy density were different,and the main variables of the regression equation of yield flow were rainfall and I₃₀,while the main variables of the regression equation of sediment yield were rainfall and rainfall erosion,and their fitting degree is higher.?6?The artificial neural network was used to simulate and predict the miscarriage sand of the Pinusmassoniana forest with different canopy density.The results showed:during the runoff simulation in the process of prediction,the relative errors of training and testing were 22.20%.During the sediment loss simulation process,the relative errors of training and testing were 8.6%.?7?The verification results of the two prediction models show that the fitting R square of the measured and predicted multiple regression runoff is 0.838,and the fitting R square of the measured and predicted sediment loss is 0.691.The fitting R square of the predicted value of runoff and measured value of the neural network model is 0.771,and the fitting R square of the predicted value and measured value of sediment loss is0.898.The fitting degree of the neural network model is relatively high.In summary,the soil and water loss of the Pinusmassoniana forest in the preliminary soil erosion control area of Changting showed a trend decreased first and then increased with the increase of canopy density.The canopy density of 0.4 and 0.6 had a better effect on reducing soil erosion;The law of nutrient loss was consistent with the law of runoff and sediment loss,the canopy density of 0.6 is the best value for vegetation restoration and soil and water conservation under the Pinusmassoniana forest;The runoff and sediment yield and rainfall,I₃₀and rainfall erosivity were both significantly and positively correlated;The verification results of soil erosion prediction models of the pinusmassoniana forest with different canopy density indicate that the neural network BLP model had a good effect on runoff and sediment yield of the Pinusmassoniana forest in the Changting red soil erosion primary control area,which was a reference prediction model.