Inversion And Dynamic Monitoring Of Soil Moisture In The Daxing’anling Burnt Area,Inner Mongolia

In this paper,the soil moisture of three burnt areas in the Daxinganling forest area of Inner Mongolia in different periods and different vegetation coverage is taken as the research object,and the remote sensing images of the Landsat series are used as the data source to extract the corresponding remote sensing drought index respectively,and then compare the three burnt areas with the soil moisture.A total of 5 periods of soil moisture measured data were analyzed by regression model,and the vertical drought index(PDI)model,the improved vertical drought index(MPDI)model and the vegetation-adjusted vertical drought index(VAPDI)model based on the remote sensing index were obtained.The accuracy of the regression model constructed by the drought index was verified and the suitability was evaluated,and the suitability of the three models for soil moisture inversion in the burned area was discussed.On this basis,the optimal model for inversion of soil moisture in the burned area with different vegetation coverage conditions was selected,and then combined with the Landsat series of remote sensing image data sources(with the advantage of long-term series),the surface soil moisture of the burned area was inverted,and the result was obtained.Based on the inversion results of soil moisture in the burned areas,the dynamic monitoring of soil moisture in the burned areas was carried out,and then the temporal variation trend and spatial distribution characteristics of soil moisture in the burned areas were analyzed.The main conclusions are as follows:(1)The measured data of soil moisture in the burned area showed a certain correlation with the three remote sensing indices of PDI,MPDI and VAPDI.Among them: the PDI model based on the PDI model showed poor fitting effect in the inversion of the burned area in different periods and under different vegetation coverage;the MPDI model based on the MPDI model showed a poor fitting effect in the inversion of the burned area in different periods and under the different vegetation coverage conditions.The model based on VAPDI shows better fitting effect than MPDI in the case of high vegetation coverage,but the model fitting effect is better in the soil moisture inversion of the burned area with low vegetation coverage.Difference.In the verification and evaluation of the accuracy of the soil moisture inversion model of the burnt area in different periods and with different vegetation coverage,it is found that PDI does not show good inversion accuracy in either the high vegetation coverage area or the low vegetation coverage area.VAPDI has the highest inversion accuracy in the burned area with high vegetation coverage,but poor inversion accuracy in the burned area with low vegetation coverage;while MPDI shows good inversion accuracy in both high and low vegetation coverage areas,the inversion accuracy is the highest in the low vegetation coverage area,and the accuracy is slightly lower than that of VAPDI in the high vegetation coverage area.(2)Based on the MPDI model and the VAPDI model,the soil moisture in the third phase of the "Hands-Horse-burned Land" was inverted respectively.The results showed that the spatial distribution of soil moisture inversion by the two models was basically the same,but the MPDI index(which was more sensitive to vegetation differences)was used.The results of the soil moisture inversion(compared to VAPDI)in the burned-out sites are more precise,and the spatial distribution characteristics of soil moisture obtained are more hierarchical,which can reflect the heterogeneity of soil moisture in similar plots,although in high vegetation The inversion effect of VAPDI remote sensing index in the coverage area is slightly better,but the comprehensive inversion effect is still better than MPDI.(3)The analysis of the long-term phase of soil moisture in the "Jinhe-Genhe fire site" shows that the soil moisture in this area was evenly distributed before the forest fire,and suddenly became polarized after the fire.The distribution characteristics gradually changed from polarized distribution to uniform distribution in space,and roughly reverted to the law of uniform distribution after the fifth year after the fire.(4)In the year of the fire,the average value of soil moisture increased sharply in the burned area in the primeval forest area in the northern part of the Daxing’an Mountains,Inner Mongolia,from 45.44% to 61.27%.This type of burnt area belongs to the perennial underground permafrost area.The forest fire causes the surface temperature to rise sharply in a short period of time,which may lead to the melting of the underground permafrost,thus causing the soil moisture in the burned area to increase suddenly after the fire.In the second year after the fire,the average value of soil moisture in this type of burnt area was still high,and after the third year,it tended to the normal level of forest soil moisture in this area.(5)The spatio-temporal dynamic monitoring and analysis of soil moisture in the“Jinhe-Genhe burned area” from 2000 to 2020 shows that the vegetation coverage has a greater impact on the stability of the spatiotemporal pattern of soil moisture in the burned area;The change characteristics of soil moisture over time are also more complex.The soil moisture in the burned area is not only affected by vegetation coverage,but also determined by the specificity of the environment and region.But on the whole,the soil water content of the burned area and the restoration of vegetation mutually restrict and promote each other.