Study On Afforestation Decision Based On Data Mining

The main task of forestry production is the quantity enlargement of afforestation and imrpovement of its quality. Especially, afforestation quality has become the priority which currently influences the effects of forestry production. Species selection, based on the principle-Matching Tree Species with Site, is the basis of the improvement of afforestation quality from the perspective of technique. The principle-Matching Tree Species with Site means that taking the advantage of different characteristics of tree species, mainly the ecological characteristics, to match the site conditions, in order to reach the as high production as possible under existing economic conditions. Mass data has been accumulated in the longtime forest resources investigation and statistics in China. Therefore, how to find out the relationship between the site factors (eg. Terrain, Soil condition, Climate and Vegetation) and the matching species from these mass data to guide the silviculture decision has become a key, which influences the afforestation quality. As one of the important development directions in modern computer technologies, Data Mining can dig out the knowledege process that is implicit, unknown but useful to decisions from a lot of data.In the paper, Data Mining was applicated in the afforestation planning and design. The purpose was to raise the accuracy and automation of silviculture decision, gather the knowledge and provide the new theory , method and technology for silviculture decision.Based on the data aquired by the forest resources management survey and afforestation survey, on the method of Data Mining and on the tool of SPSS Clementine Client software, the paper did the research and analysis of the application of Data Mining to silviculture decision.The main aspects done in this paper are as followings:(1)General steps of Data Mining of silviculture decision were studied and analyzed, combining the specific flow of Data Mining and afforestation planning and design;(2)The specific method of Data Mining applied in silviculture decision was studied, based on the analysis of silviculture decision demanding;(3)The analysis of the relationship between chinese pine growth and relevant site factors in Fangshan District was done, using Data Mining. Growth suitable of chinese pine was predicted and results generated by different data preprocessing were compared;(4)Relationship of afforestation constrction operation technologies were studied by the application of Data Mining Association Rules technology. The research innovations are as followings:(1)Data Mining was used in afforestation planning and design. Combining the specific flow of afforestation planning and design and features of Data Mining, silviculture decision knowledge was devided in to Matching Tree Species with Site and afforestation design;(2)General steps of silviculture decision data mining was proposed;(3)The tree growth suitable under certain site condition was predicted according to the Decision Tree Classification technology of Data Mining;(4)Data Mining Association Rules technology was studied applied in afforestation design, and relationship between different measures was found out.Forestry information has stepped into the knowledge stage, in which decision is made by knowledge. This study was a beneficial attempt of Data Mining and silviculture decision. Decision of afforestation planning and design was provided by the knowledge discovery of Matching Tree Species with Site and afforestation design.