摘要: 目的 探索网格化的蚊媒监测模式，为蚊虫及蚊媒传染病监测和防制提供科学依据。方法 2019年7－10月在上海市静安区北、中、南3个街道结合自然道路，按照长、宽分别为400 m划分为相邻的监测块，以监测块为单位开展外环境布雷图指数（BI）法和人诱停落法的现场监测，通过比较不同地理位置、不同监测块间的监测结果来评估网格化监测的有效性。采用Excel 2016和SPSS 16.0软件处理数据，进行Kruskal-Wallis秩和检验、Spearman相关等统计学分析。结果 共划定监测块21个，完成监测8次；白纹伊蚊停落指数为4.06只/（人·h），其中最高为7月下旬9.58只/（人·h），其次为7月中旬7.62只/（人·h）和8月中旬5.62只/（人·h），北、中、南部停落指数差异无统计学意义（Z=0.587，P=0.746）；平均BI为27.21，7月中旬最高为40.64，其次是7月下旬为37.90，再次是8月中旬为35.56，时间变化趋势与停落指数基本一致，北、中、南部差异有统计学意义（Z=47.161，P < 0.001）；停落指数和BI以块为单位统计，在不同监测块间差别较大，7月中旬各块间差异最大，停落指数均值为7.62只/（人·h），最高监测块达29.00只/（人·h），最低为0，BI均值为44.07，最高达178.57，最低为0，随密度下降监测块间差异逐渐减小；停落指数和BI在不同场所间差异有统计学意义（Z=18.747，P=0.001；Z=18.722，P=0.001），其中学校最高，其次为居民区。结论 蚊虫密度受环境及控制措施实施等因素影响，不同场所、不同区域蚊虫孳生也存在差异，因此建议在蚊媒监测中采用网格化的监测模式，扩大监测覆盖范围，使监测结果更具代表性。

Abstract: Objectve To explore a grid-based model for vector mosquito surveillance, and to provide a scientific basis for surveillance and control of mosquito-borne diseases. Methods From July to October 2019, three subdistricts (including the natural paths within the subdistricts) in the north, middle, and south of Jing'an district, Shanghai, China were divided into adjacent surveillance blocks with an area of 400 m×400 m. The Breteau index (BI) and human-baited landing method were used to conduct on-site surveillance in blocks. The effectiveness of grid-based surveillance was evaluated by comparing the results of different geographical locations and surveillance blocks. Excel 2016 and SPSS 16.0 softwares were used for data processing, and statistical analyses such as Kruskal-Wallis rank sum test and Spearman correlation analysis were performed. Results A total of 21 surveillance blocks were demarcated and 8 rounds of surveillance were conducted. The langding index (LI) of Aedes albopictus was 4.06 mosquitoes/person·hour, with the highest value of 9.58 mosquitoes/person·hour in late July, followed by 7.62 mosquitoes/person·hour in mid-July and 5.62 mosquitoes/person·hour in mid-August; there was no significant difference in the LI between north, middle, and south areas (Z=0.587, P=0.746). The mean BI was 27.21, with the highest value of 40.64 in mid-July, followed by 37.90 in late July and 35.56 in mid-August; the changes over time were basically the same as those of LI, with a significant difference between north, middle, and south areas (Z=47.161, P < 0.001). The LI and BI were counted in blocks, with significant differences between surveillance blocks; the most significant inter-block difference was observed in mid-July; the mean LI was 7.62 mosquitoes/person·hour, with the highest value of 29.00 mosquitoes/person·hour and the lowest value of 0 mosquito/person·hour; the mean BI was 44.07, with the highest value of 178.57 and the lowest value of 0; the difference between the surveillance blocks decreased gradually with decreasing density. There were significant differences in both LI and BI between different places (Z=18.747 and 18.722, respectively, P=0.001 and 0.001, respectively), with the highest value in schools, followed by residential areas. Conclusion Mosquito density is affected by multiple factors such as the environment and the implementation of control measures. There are also differences in mosquito breeding in different places and regions. Therefore, it is recommended to expand the surveillance coverage by adopting grid-based surveillance model in vector mosquito surveillance, which will make the results more representative.