Open AccessClassification and Risk Identification of Forest Ecology Based on Statistical Learning
Author(s) -
Xuan Sun,
Ting Lei
Publication year - 2020
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.179
H-Index - 26
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/514/3/032020
Subject(s) - forest ecology , afforestation , forest restoration , forest management , ecology , geography , identification (biology) , understory , ecosystem , environmental resource management , agroforestry , environmental science , forestry , biology , canopy
In this paper, the forest ecosystem classification and risk identification are studied based on statistical learning methods such as cluster analysis and discriminant analysis, taking Daxinganling Forest in Inner Mongolia, Beijing Forest, Jilin Changbai Mountain Forest, Hubei Huitong Forest, Guangdong Heshan Forest, and Linzhi Forest as the research objects. The results show that: (1) The classification of forest ecosystem activity is mainly determined by the forest species. At the same time, the composition and growth of forest species is an important basis for risk classification. (2) Climate factors and soil factors affect forest ecosystem stability and vitality, which is an important basis for testing models after risk identification. Therefore, according to the results obtained in this article, the key to study forest ecology is to plant the most suitable forest species and tree species on various site types and implement corresponding afforestation measures so that the entire region can achieve “suitable trees” and “rational management”, so that the potential of land production can be brought into full play and the goal of “use the land as much as possible” is achieved.