关键词: 景观格局, 杨树锈病, 相关性分析, 逐步回归分析, RDA分析

Abstract:

Objective: This study aims to explore the relationship between the landscape pattern and the epidemiology of poplar rust disease, and explain the incidence of poplar rust disease in Yanqing district of Beijing in terms of the landscape, in order to provide essential information for the control of the disease. Method: A total of 78 sites were set in Yanqing, Beijing to investigate the disease. According to the disease classification standards, the incidence of poplar rust disease was recorded, and the disease index was calculated. Based on the remote sensing image classification map of the research area, the sample site is set as the center and a buffer zone with a radius of 250, 500, 750 and 1 000 m was set up. Fragstats software was used to calculate the landscape metrics of landscape level and class level in each scale buffer zone.R software was used to implement statistical analysis, correlation analysis of landscape metrics under different scales, and stepwise regression analysis of disease index and landscape metrics, and Canoco, the ecological statistical analysis of ware, was used for RDA analysis of class level. Result: The result showed that the disease occurred frequently in the study area, with the disease index between 5.6 and 94.1. The severity of the disease was significantly affected by the landscape pattern, habitat environment and management status. At the landscape level, the landscape pattern in different buffer zones had different responses to the severity of poplar rust disease, and the factors with significant influence on the disease were also different. At the class level, cropland patches and grassland patches had significant impacts on the occurrence of poplar rust disease, which was mainly reflected in the common metrics such as patch density(PD), and intersperse and juxtaposition index IJI. Conclusion: In the landscape structure, the corridor of road and the foundation of grassland and cultivated land provide convenient conditions for the spread of disease, which can promote the occurrence of poplar rust disease. In this study, the landscape pattern of 500 m buffer is the optimal scale of disease management. Under this scale, the landscape metrics establish the most significant correlation with the disease index. Combined with the pathological characteristics of poplar rust disease, this study considers that the occurrence of the disease can be explained by the area-shape, landscape aggregation and diversity. The complexity of the shape of the patches inside the landscape and the level of diversity of the landscape will affect the occurrence of disease, and it will greatly promote the prevalence of disease with dispersed patches in the landscape, refined landscape structure as well as the uniform distribution between patches.

Key words: landscape pattern, poplar rust disease, correlation analysis, stepwise regression analysis, RDA analysis