Land cover and plant diversity in tropical coastal urban Haikou, China

Urban development proceeds rapidly, so understanding the mechanisms underpinning this process is critical to urban planning. One way to accomplish this is by measuring and comparing mosaics in urban ecosystems across space and time, and relating these to ecosystem processes. In this study, we examined 190 Urban Functional Units (UFUs) within urbanized areas in the tropical city of Haikou in China, and we assigned each to a UFU type. We assessed land cover as well as plant taxonomic and phylogenetic diversity within each UFU using field investigations and by interpreting imagery from remote sensing. Additionally, we quantified several relevant factors that may influence plant diversity. Using these data, we built general linear models to WA how UFU types, as well as relevant biophysical, socioeconomic, and management factors, predict variation in urban plant diversity. We found that the 190 UFUs examined in Haikou belong to 6 primary and 16 secondary UFUs. We observed that land cover varied strongly among secondary UFUs and that species richness and diversity differed significantly across the 6 primary UFUs. Most land cover and plant diversity variables were significantly positively correlated with total land area, the age of the UFU, and average housing prices. Tree species richness, phylogenetic distance, and the mean nearest taxon distance were positively correlated with maintenance times and watering frequency per year (green management variables). Our study indicated that socioeconomic variables help predict the percentage of green space and plant taxonomic and phylogenetic diversity differences among Haikou urban ecosystems.