Fine particulate matter (PM_2.5) pollution is a serious environmental problem in the process of urbanization in modern society, and it is one of the most important research hotspots in the study of urban atmospheric environment. As a world-class urban agglomeration, Guangdong-Hong Kong-Macao Greater Bay Area is not only an important embodiment of the economic, social and cultural development of urban areas, but also an important component of the national regional development strategy and the foothold of policy implementation. Therefore, the ecological environment of this region is particularly important. The temporal and spatial distribution characteristics of PM_2.5 concentration data in Guangdong-Hong Kong-Macao Greater Bay Area from 1999 to 2016 were analyzed. Moran'I index was selected as the measurement in spatial autocorrelation analysis. The relationship between PM_2.5 concentration and meteorological factors in the study area was discussed by multiple linear regression model. The results indicated that the PM_2.5 concentration in Guangdong-Hong Kong-Macao Greater Bay Area increased firstly and then decreased from 1999 to 2016. The air quality improved significantly after the highest concentration in 2008, and the similar values of annual average PM_2.5 concentration tended to aggregate distribution in 1999, 2009 and 2016. The hot spots were concentrated in the administrative core area of the bay area; the cold spots were concentrated in the core edge area, where the air quality was better. Pearson correlation analysis (actual evapotranspiration (aet), solar radiation (srad), minimum temperature (tmmn), vapor pressure (vap), saturated vapor pressure difference (vpd), wind speed (ws)) was used to select six meteorological factors for regression analysis to evaluate the significant factors affecting the temporal and spatial distribution of PM_2.5 concentration. The results showed that there was a negative correlation between srad and PM_2.5 in this study area, which is quite different from other related studies in other cities. The tmmn and vpd were positively correlated with PM_2.5 concentration, while ws was negatively correlated with PM_2.5 concentration.