目的 了解蚊虫与地物特征的对应关系，为利用“3S”技术〔地理信息系统（GIS）、全球定位系统（GPS）、遥感（RS）的统称）〕开展蚊虫预警监测提供参考。方法 2010－2011年3－12月调查无锡市建设用地、农村用地、湿地、林地和耕地5种环境类型中的蚊虫群落，并应用3S技术分析每个研究点50、100、150、200、250、300、350、400 m缓冲区内建筑用地、耕地、绿地、湿地、静止水体、牲畜的地物构成，在Canoco 4.53软件中对蚊虫群落与地物特征的对应关系做冗余分析（RDA）。结果 蚊虫与环境变量的RDA经蒙特卡罗置换检验，轴ⅠF=42.198，P=0.002，总轴F=17.616，P=0.002，分析结果非常可靠。研究点蚊虫群落与250 m缓冲区内地物特征的典范特征值总和最高，为0.889。向前引入法分析，蚊虫群落与建筑用地、牲畜、耕地、绿地、湿地的相关性差异有统计学意义（P<0.01），与静止水体的相关性无统计学意义（P>0.05）。RDA经二维排序显示，与淡色库蚊显著正相关的地物特征为湿地，与三带喙库蚊、中华按蚊、骚扰阿蚊显著正相关的是牲畜>耕地，与骚扰阿蚊显著正相关的是耕地>牲畜，与白纹伊蚊显著正相关的是建筑用地。结论 优势蚊种与地物特征的对应关系显著相关，可以应用3S技术对目标点位的优势蚊种进行预警监测。建筑用地、耕地、牲畜、绿地、湿地与蚊虫群落演替显著相关，是蚊虫防制的重点地物类型。

Objective To establish the relationship between mosquito fauna as well as population densities and land use for providing reference for the early warning and surveillance of mosquitoes using 3S technology. Methods From March to December in 2010-2011, a sampling was conducted on the mosquito communities in the construction land, rural land, wetland, forest land, and farmland in Wuxi, Jiangsu province, China. The surface objects of construction land, farm land, green land, wetland, static water body, and livestock in the 50, 100, 150, 200, 250, 300, 350, and 400 m buffer areas of each study site were analyzed by 3S technology (Remote Sensing, Geography Information Systems, and Global Positioning Systems). The relationship between mosquito communities and land use was determined by redundancy analysis (RDA) using Canoco 4.53. Results The Monte Carlo permutation test confirmed the statistical significance of the relationship between mosquitoes and environment variables determined by RDA (axisⅠ: F=42.198, P=0.002; total axis: F=17.616, P=0.002). The sum of all canonical eigenvalues was the highest (0.889) for the mosquito communities and the land use in the 250 m buffer area of study site. The forward selection analysis showed that mosquito communities were significantly correlated with the construction land, livestock, farm land, green land, and wetland (P<0.01), but had no significant correlation with the static water body (P>0.05). The RDA biplot showed that Culex pipiens pallens was significantly positively correlated with the wetland, Cx. tritaeniorhynchus, Anopheles sinensis, and Armigeres subalbatus were significantly positively correlated with the livestock (>farm land), Ar. subalbatus was significantly positively correlated with the farm land (>livestock), and Aedes albopictus was significantly positively correlated with the construction land. Conclusion Predominant mosquito species have significant correlation with land use features, and 3S technology can be used for the early warning and surveillance of dominant mosquito species at the target areas. In Wuxi, the land use is significantly associated with specific mosquito communities, which should be taken into consideration in formulating strategies and practices for mosquito management.