The Research On The Appropriate Grain Of Forest Landscape Patterns Based On Gis

Based on landscape ecology and cartography for the theories foundation and area forest resources distribute diagram in West Dongting Lake Region for main data source, the forest landscape types diagram in west Dongting is drawn and the attribute and spatial database of forest landscape types is established by means of geographic information system software ArcGIS. Based on these, forest landscape patterns are analysis to go deep into understanding inset structure of whole structure and form of forest landscape by adopting the landscape indices; Spatial grain conversion With vector data of West Dongting Lake Region based on maximum area rule were used, so as to discuss the scale effete of landscape Pattern in the sealing from the accuracy loss, and the landscape indices variations, then found the optimal research grain to provide science reference and theories basis in order for integrated evaluation comparison of the forest landscape patterns; According to the principle composition analysis, we set up the forest landscape evaluation index system, build up the landscape pattern integrated evaluation model, complete the one-way ANOVA of the integrated evaluation index under different grains, inquire into the influence of the grain change to the integrated evaluation index of the landscape spatial pattern, provide basis to the optimization of the forest ecosystem and the exploitation and management of forest resources. The main contents and conclusion are as follows:(1) Analysis of the Forest Landscape PatternThe analysis of the forest landscape pattern Analyze the forest landscape pattern in West Dongting Lake Region. Except for Non- forestland landscape, broad-leaved, economic, and pine forest landscape possess a large proportion, and make up the main components of the forest landscape pattern in West Dongting Lake Region. Non- forestland landscape patch area is largest on average, far surpass the others, so does its variation coefficient. The Bush forest landscape area's variation coefficient is the smallest. Broad-leaved, Bamboo, and bush's forest landscape area is comparatively large. The complication of the economic forest landscape patch shape ranks the third. Kashiwagi forest landscape patch area is smallest on average, so is its variation coefficient, its shape index is large. Kashiwagi is in disadvantage of the landscape diversity and dominance, and is easiest influenced by other landscapes. Economic, pine and lanceolata forest landscape landscape's dominance, diversity index and uniformity is high, the general control of these landscapes are high, the research area is affected by the distribution of these landscapes and express high heterogeneity and diversity.(2) Selection of the Appropriate Grain of Landscape PatternIn the research of grain transform precision, with the increase of grid, uncertain factors increase. The loss precision of the total area of landscape, the area of each landscape style, and the proportion of landscape style increase, and appear different varying trend. Actually original information loss increase, and grain research threshold was 100m.Landscape indices can indicate the grain effect with the scale changing, the different responds of different landscape indices show the diversity of information derived from the original data with RAM and the first landscape indices scale domain were 10m-90m. The appropriate grain of landscape pattern based on all research results was 60m-90m.(3) Establishment of the Model of Integrated EvaluationBased on the analysis of the forest landscape pattern in West Dongting Lake Region, set up evaluation index system and the model of integrated evaluation is establish by using the principle composition analysis. The 4 main compositions and model of integrated evaluation are as follows. Z1=0.544X01+0.963X02-0.812X03+0.035X04+0.513X05+0.706X06+0.665X07+0.959X08 +0.970X09+0.839X10+0.871X11+0.362X12 Z2=-0.293X01-0.160X02-0.281X03+0.837X04+0.782X05+0.447X06-0.134X07-0.193X08-0.054X09+0.290X10-0.387X11-0.674X12 Z3=0.719X01-0.080X02+0.163X03+0.412X04+0.166X05+0.492X06-0.369X07-0.160X08-0.098X09-0.427X10+0.004X11+0.609X12 Z4=0.216X01-0.135X02-0.143X03+0.283X04-0.221X05-0.115X06+0.560X07-0.065X08-0.080X09-0.003X10-0.189X11+0.037X12In the formula, Fi means integrated evaluation index of the ith unit, Zj means the ith component value,λj means the jth Cumulative % of component. m means number of component.(4) Comparison of the Landscape Integrated EvaluationThe model of regional landscape integrated evaluation was established by using the main composition analysis principle. The theory is clear and the calculation is simple. The model reflects the uniformity and diversity of the landscape style compositely. Set granularity as the element of the comprehensive evaluation index to carry out the one-way ANOVA, choose 5% as the significant level, p>0.05, indicate that there's no significant difference between the index value of different granularity level and landscape structure integrated level of vector data. When choose data, the analysis of each landscape unit under different format and grain won't affect the result. Use vector data as the data source of the landscape spatial pattern comprehensive level evaluation. The results. show that:the higher level of the forest landscape Patterns is in Taoyuan County, Li county, Dingcheng District, Linli county, Hanshou County, among which the highest is in Taoyuan County; The middle level of the forest landscape Patterns is in Hefu Forest Farm, Changde Forest Farm, Jinshi county, Anxiang County, Liuyehu Tourist Resort; Xidongting farm, Wuling District, Cendan Forest Farm, Hejiashan farm is the lower, and Xidongting farm is the lowest.