Research Of Soil Fertility Evaluation Based On GIS And Data Mining Technology

Soil fertility is the basis of the sustainable development of soil. The level of soil fertility directly affects the quality and yield of agricultural products. Soil fertility evaluation is very important to agricultural production. The rules of soil spatial variability in aspect of soil fertility are important indicator to value the soil fertility and they are also important basis to guide precision fertilization. Aiming at the problems existing in the evaluation of soil fertility, in order to promote the development of agricultural information and modernization. The main contents of this paper are as follows:(1)Using Natural Neighbor algorithm based on the Soil Nutrient Spatial Database, this paper has researched into the changing of soil nutrients’ spatial variability for three consecutive years of precision fertilization 2008, 2009 and 2010. The spatial distribution pictures of alkali-hydrolyzable nitrogen and soil available potassium are produced. The pictures of alkali-hydrolyzable nitrogen content’s area proportion between 110mg/kg to 140mg/kg in three consecutive years and the pictures of soil available potassium content’s area proportion between 100mg/kg to 160mg/kg in three consecutive years are produced. The conclusion can be drawn that the pictures of soil Nitrogen spatial variability have obviously changed after precision fertilization. Alkali-hydrolyzable nitrogen and soil available potassium content’s area proportion have increased by year by year, which indicates that soil fertility’s diversity has become smaller. This result is an effective way to improve the scientific and objective evaluation of soil fertility based on the spatial analysis technology.(2) Association rule mining is one of the important applications of data mining technology in agriculture. According to Association rules we can find the relationship among attributes in large agricultural data. In this paper, the data mining method of association rules are used to combine the soil nutrient content and the yield of maize. Using association rule of data mining methods, this paper researches on Nong’an County Kai’an town soil nutrient indexes of confidence, support degree of relevancy attribute analysis. The correlation between the alkali solution nitrogen, available potassium, available phosphorus and the yield of maize and the correlation degree of each attribute were found, and the knowledge rules were established to lay a foundation for the evaluation of soil fertility.(3) This paper designed and developed the evaluation of soil fertility based on GIS and data mining according to the theory and method of GIS software engineering. Take Kai’an town as the study area, design and develop a system to achieve user management, spatial data display and query, thematic data editing and modification, spatial thematic data analysis, data mining, association rules, soil fertility evaluation and other functions. Application of the evaluation system provides guidance for agricultural production and technical support for decision-making.According to spatial database technology, geographic information system technology and association rules mining technology, this paper carries on the study of soil fertility evaluation, and then realize the evaluation system of soil fertility. The level of modernization of agricultural technology extension can be improved on the basis of the study of soil fertility evaluation system...