|
Countryside water body is a sort of important water resources, which influences agricultural production in our country. As the fast increment of the economics in our country, urbanization goes faster, which leads to a serious pollution in agricultural areas, causing bad effects on production as well as decrement on quality, and at last harms people’s health. Therefore, it’s of great importance to get to know the status of water body, which helps water body processing and production safety.Countryside water body ranges wide, but meanwhile small, shallow and not fluid, which complicates inspection on pollution. Traditional methods need sampling and then testing water qualities, of which limitations vary in aspects such as manpower, material resources and hydrological conditions. Remote sensing is a new technology which develops recently, and is more and more widely used in environment monitoring. Remote sensing is real-timed, of great efficiency, long-lasting, with large amount of data and wide range of area, which plays a critical role on water body monitoring. Nowadays, remote sensing becomes an important technical way of rural water body pollution monitoring and dynamic time-space alteration analysis. Remote sensing monitoring mainly relies on the distinguished characteristics of the spectrum of polluted water, which varies from clean water, and will cause great difference on absorption as well as reflection ratio in certain wavelength period. There are lots of parameters, among which chlorophyll a, suspending solids and transparency are mature. By constructing a WebGIS system, we can realize the same retrieval monitoring system on distributed computers, altogether with common features of GIS environment monitoring. For example, after adding water body distribution graph as well as monitoring and sampling points distribution graph, remote sensing monitoring will become more intuitionistic and convenient. Furthermore, it’ll be very portable when similar research is done on other areas, by which time we only need to modify certain parameters in retrieval equations.The paper discusses the traits of the water body in Lvshui county, Jiangsu province. After comparison on applicability I chose CBERS as the source of remote sensing images, and then designed a sampling route for water quality parameters to accomplish the experiment on remote sensing retrieval operations of it, and finally got the chlorophyll a, suspending solids and transparency. After researching on the water body quality parameters and the CBERS images of the same period of the agricultural areas in Lishui, I got the best retrieval equation (maximum R2 is considered):(1) y=-0.292x+8.908 (R2=0.869) for chlorophyll a, (2) y=-0.236x3+0.464x2-0.446x+2.068 (R2=0.857) for suspending solids, (3) y=-66.722x3+352.264x2-556.504x+325.807 (R2=0.907) for transparency. The former one is done well simply by linear equations, while the latter two will have its R2 increases as the equation power rises, which can be treated as the basis for further researches. After data fetching, I chose ArcGIS Server as the development platform and constructed a water body information remote sensing monitoring system in agricultural areas based on the retrieval equations of Lishui.The construction of retrieval model for water pollution management in agricultural areas of Lishui, Jiangsu Province by CBERS is efficient, whose advantages are:(1) the precision of retrieval equations are high, which shows the availability of computer processing algorithms; (2) the system developed on ArcGIS Server platform can manage water resources effectively, and integration analysis can be done through the combination of different layers by remote sensing retrieval operations, which shows the multi-functions of the platform; (3) the introduction of WebGIS system enables non-GIS or non-remote-sensing people to get involved to environment management using GIS; (4) the B/S pattern of WebGIS enables the separation of the data and the system, and also solves the security problem of the data, and meanwhile users don’t need to install a large system on client sides, which makes them convenient; (5) the mechanism of the background algorithm realization enables the probability of retrieval operations done to other areas only by modifying the retrieval model to fit them, which is quite portable.
|