Study On Farmland Soil Moisture Information System Based On GIS And Soil Moisture Forecast Model

The soil moisture is the soil water content situation and the absorbing the moisture status of the crop,which is the important factors that affect crop root growth.At present most of the soil moisture monitoring method is time-consuming,poor timeliness,can't timely response the changes of soil moisture.Therefore,we need to apply information technology in monitoring soil moisture,improve the efficiency of the soil moisture data acquisition and transmission,and strengthen the organization and management of data of soil moisture.Meanwhile,it has to improve agricultural water utilization,provide basis for decision-making for the drought.Based on this,this paper carried out of soil moisture monitoring and prediction research,mainly includes the following contents:(a)ASW-4 is designed and developed the collection used to soil moisture of mobile terminal.The terminal can quickly measure the soil moisture content,collecting data through the GPRS network to data management center,and has the function of GPS,can locate the coordinates of sample point information.ASW-4 has low cost,simple operation,no limited by the monitoring location when using it,has a great deal of flexibility,ability to work with fixed soil moisture monitoring stations,also,it can normally cover more uniform comprehensive monitoring area.(b)The farmland soil moisture information system is based on the GIS technology.This system can send data,receive ASW-4 reserved the interface at the same time,also can receive the data of fixed soil moisture monitoring station.System combines set data acquisition,storage,management,display and output function,can be real-time monitoring sampling points of the change of soil moisture content,in the form of graphs,charts,intuitive displayed to the user.System can combine soil moisture and crop growth period,crop moisture analysis,get the contour,contour surface figure,from the overall understanding of the trend of soil moisture monitoring area.(c)Based on the feidong,badou countytown fixed soil moisture monitoring stations to collect data as learning samples,we build a BP neural network in short-term forecast models of soil moisture.Through the analysis of 0-20 cm,20-40 cm,40-60 cm depth of soil moisture,rainfall,sunshine time,air humidity and the average temperature as the input variables of the model,we use the second day 20-40 cm soil moisture as output variables of model,and it carried on the forecast model.In conclusion it used the genetic algorithm to optimize the BP neural network to improve the forecast precision of the model.In this results,part of the results have been used widely in the area of Anhui province,the system runs stably,to a certain extent,it improves the efficiency of soil moisture monitoring,has provided the safeguard for the agricultural production.