| Soil moisture is one of essential parameters in the study of climatology, hydrology,ecology and agriculture,which dominates the transportation and balance of water and heat between lands and air directly. The change of the soil moisture will lead to change of the thermotics characteristic of soil and optical properties of surface, even to the change of climate. The information of the change of land soil moisture in regional and large-scale is important to the research as follows: balance of land-atmosphere interaction, land hydrology, improving the forecast accuracy of regional and GCM, monitoring of flooding and drought, assessment of the growth condition of corps, the natural and ecology problems. Therefore the research on change and estimation of the soil moisture in regional and large-scale is very important, which also is a key international problem.The traditional surface observing network can not meet the study on soil moisture dynamic and successive variation in temporal and spatial scale. As to optical remote sensing, by means of visual spectrum remote sensing and thermal infrared remote sensing, the soil moisture information can be got with help of optical sensors,which are run at the present time. Though they can not penetrate cloud layer and vegetation to observe the soil moisture in shortwave range, which can be done easily by long-wave remote sensing. Besides the observation is disturbed by atmosphere badly, which limits the application of optical remote sensing to monitoring the soil moisture.The study shows that the microwave remote sensing is the only frequency band in electromagnetic spectrum to quantitatively and actually estimate the soil moisture. The microwave remote sensing can weaken the impacts from air,cloud and rain because of it's longwave,besides, the microwave can penetrate the objects, and get the characteristic information of the underlaying surface in certain thickness.On the other hand,the microwave can observe the soil moisture 24-hourly in all weather without the impact of the irradiation, by which the microwave remote sensing distinguished itself in monitoring the soil moisture. The multi-frequency and multi-polarization operating mode of the microwave remote sending sensor makes it proper to get surface parameters. Therefore, the microwave remote sensing is the best measuring means to observe the soil moisture.Based on satellite-borne microwave radiometer-AMSR-E data, this paper focuses on the quantitative retrieval of soil moisture by means passive microwave remote sensing in the case of Xinjiang. The algorithm of soil moisture is modified to analyze the surface vegetation cover condition by using MPDI(microwave polarization difference index ), and the vegetation opacity is calculated. Based on the estimation of soil roughness, the radiactive transfer equation is solved. In the end, the retrieved soil moisture data is applied to monitor soil moisture content, according to which drought rank is created. To summarize, main findings of this study are:1. Several common microwave radiation models above randomly rough surface are introduced, which including small disturbance model, physical optics model, geometrical optics model, AIEM integral equation model, empirical model-Q/H model, Hp model and newly-developed Qp model. Besides, the water-cloud model and discrete model MIMICS, which are widely used on vegetation-covered surface, are also presented. The merits and faults of every model are comparatively analyzed, so do the sphere of application.2. According to statistical distribution and variation of MPDI, the character distribution of soil moisture on the surface is analyzed. Based on AMSR-E multitemporal data in the last 5years, the statistical distribution of MPDI and departure value of July in Xinjiang are discussed, the soil moisture variation distribution is also mapped.The results show that, the soil moisture of 2002 and 2003 in most parts in Xinjiang increased and the departure value is above zero, while, the departure value from 2004 to 2006 is below zero, and the soil moisture is less than the synchronous average value, which shows that the damage to crops by drought worsen in the last three years.3. Based on the analysis on the character of MPDI, the relationship between NDVI and MPDI is studied, and the threshold value of MPDI under densely vegetation covered, middle vegetation covered, sparse vegetation covered and nudation are built. MPDI＞0.12 means the surface is desert,sand or rocks; 0.01＜MPDI＜0.12 means the sureface is covered by middle vegetation, and MPDI＜0.01 means the earth is covered by densely vegetation.4. Based on the expirical model relation between MPDI and vegetation optical thickness, the latter is computed by the modified algorithm.5. The characteristic of MPDI affected by soil moisture, surface roughness and moisture content in vegetation stratum is analyzed. Considering the surface roughness can be approximately thought as a variable which does not vary with time, the soil moisture can be removed from MPDI and the index—Γrelated to surface roughness can be deduced, the latter is applied to compute the surface roughness.6. Radiative transfer equation cooperated with AMSR-E data is employed to retrieve the soil moisture, and the impact of vegetation and roughness is removed. The error test shows that the dependency between the retrieval data and the actually measured data is 0.6881.7. Based on the soil moisture retrieved from AMSR-E data, the seasonal variation and interannual variation of soil moisture content in Xinjiang are analyzed. In consideration of field moisture capacity, the soil moisture content index is computated. The grading standard of soil content is used to rank the degree of drought in Xinjiang, which can forcast the drought.The soil moisture content is used to rank the drought of Xinjiang, which can confirm the degree of drought.Based on the discussion of characteristic of microwave radiation in Xinjiang, the impact of vegetation and surface roughness is removed by means of MPDI.Radiative transfer equation cooperated with soil dielectric model is employed to retrieve the soil moisture. And the retrieval result is ideal, and the modeled standard deviation is 0.6881. The retrieval soil moisture is used to analyze the soil moisture in present status, which is applied to rank the drought in Xinjiang. The retrieval soil moisture will provide a new data source to monitor the drought in Xinjiang.
|
PDF Full Text Request