The Remote Sensing Of Soil Moisture Based On ATI And TVDI Model In Hebei Province

Soil moisture is an important factor in climate, hydrology, ecology, agriculture and otherfields, has a very important role in land and atmospheric interactions. Surface soil moisture isan important factor to influence soil erosion[1], is of great significance to the study of naturaldust sources in atmospheric particulates. In addition, the soil moisture also plays a decisiverole in crop growth.Hebei province is an important producing grain area, mainly to group dry landagriculture and a population and economic density distribution,. The climate is temperatesemi-arid and semi humid continental monsoon climate, with abundant light resources, Hebeiprovince is mainly single cropping a year, two crops a year of farming system., Rainfall inHebei province are unevenly distributed in time and space, the rainfall in southeast area ismuch less than the north west in space distribution.. At the same time, with the rapiddevelopment of the second or third industry in Hebei Province, Hebei Province in recent yearshas also become one of the most serious national air pollution areas. the scientific monitoringof soil humidity in Hebei Province, will provide effective protection for the regionalcomprehensive management of the atmospheric pollution and the healthy development ofagricultural economy.There are many traditional methods of obtaining soil moisture, mainly on the basis ofeffective monitoring point on the ground to monitor soil moisture. In the aspect of agriculturalmeteorological monitoring, the weight of soil moisture is measured by means of drying andweighting soil mass, field soil moisture monitoring and automatic station measurements canuse the soil moisture meter. There are also indirect resistance monitoring method and indirectmeasurement method. the depth and accuracy of the traditional measurement method isrelatively high, but the traditional measurement method has certain drawbacks thatmeasurement of large area is not large scale, and dynamic monitoring can not last very longtime.The inversion of the soil moisture in the region based on remote sensing technologybased on remote sensing technology is a frontier research direction in the research process onsurface land. So far, in the aspect of the research on soil moisture monitoring region, the majority of scholars established many soil moisture inversion model. The inversion modelbased on apparent thermal inertia[2-6]and the temperature vegetation index[7-10]is widelyused in recent years and higher accuracy model.But using inversion of a certain modelignores the scope of application of the inversion model, For example, apparent thermal inertiaATI model is only suitable for bare soil and low vegetation coverage area, while thetemperature vegetation index TVDI model to inverse soil moisture in region is only suitablein high vegetation coverage area,[9]At present the domestic academic circles by adopting themodel of remote sensing of soil moisture are mainly concentrated in the drought evaluationaspects of growth season of winter wheat in Hebei area. Research on Remote SensingInversion of the soil moisture is less, the research in this field needs complementary.This paper takes Hebei Province as the study area, in view of the advantages anddisadvantages of ATI model and the temperature vegetation index TVDI model to retrievesoil moisture, based on these two models of partition on inversion of carry through the wholeyear, realizes the complementary advantages of two kinds of model and constructs soilmoisture retrieval model suitable for the study area, to achieve a comprehensive monitoringand evaluation of soil moisture in the study area. In addition, this paper also considers theinfluence of elevation factor on land surface temperature, with the increase of altitude, thetemperature gradually decreased. Firstly, this paper uses the digital elevation model (DEM)for correction of land surface temperature to mostly eliminate the influence of elevation oftemperature. then according to the soil moisture inversion of land surface temperature,vegetation index, reflectance parameters built remote sensing model based on ENVI4.6andArcGIS9.3technology platform support, NDVI value partition method to retrieve the value ofsoil moisture in2009ten days every month, Soil moisture inversion of this method especiallyin Bashang Plateau and mountainous northern Hebei Province, the relatively high elevation inthe inversion, can get good effect, can be closer to the actual distribution of soil moisture, alsohas certain practicability and promotion This paper uses MODIS remote sensing data and measured10cm depth soil moisturedata for the data source to retrieve soil moisture of Hebei Province by integrating ApparentThermal Inertia (ATI) model and Temperature Vegetation Dryness Index (TVDI)model. NDVI is employed as the division factor in March, April, May, October andNovember. In each of the ten days, TVDI model was used to inverse soil moisture whenNDVI is greater than0.2, inversely, ATI model is used. TVDI model is used alone to retrievesoil moisture in June, July, August and September. In January, February and December,because of the low vegetation coverage and the missing measured soil moisture data, thealternative ATI models are used to inverse soil moisture. That is, the soil moisture ofDecember and February is inversed from ATI regression models in November and March,respectively. The new ATI model is created from the average of ATI regression modelcoefficients in November and March to inverse soil moisture in January.。The models aretested through p value; p-values of all above-mentioned models are less than0.01. Theresults show that soil moisture of Hebei Province has two change cycles from increase todecrease in one year. The first period is from December to the next May, in which soilmoisture increases from December to the next March and decreases from March to May, themaximum value is in March. The second period is from May to December, in which soilmoisture increases from May to August and decreases from August to December, the peakvalue is in August. The annual average maximum value of soil relative moisture appeares inAugust, and the minimum average value is in May. The spatial distribution of soil moisture isinfluenced mainly by precipitation, irrigation and land use patterns，Soil moisture in the northof Hebei Province and Hebei Xishan area is relatively higher than that in other plains of thesame period, soil moisture in the basin of Northwest Hebei mountains and Hebei plateau areais relatively lower. soil moisture in Hebei Plain in summer is relatively higher than that inother seasons The results of the accuracy test show that soil moisture from remotesensing inversion has good correlation with the measured data. The inversion precision of soilmoisture is high. The inversion results reflect the temporal and spatial distribution of surfacesoil moisture area in Hebei province.