Abstract:

Mangroves are woody wetland communities that grow in the intertidal zones of tropical and subtropical coasts and are among the most productive ecosystems in the world. Mangroves provide important support for coastal ecological and environmental protection, social development, and economic progress. Those found in Southeast Asia are a key portion of global mangroves. The study of spatiotemporal changes of mangrove forests and their driving factors in Southeast Asia can provide a theoretical basis and supportive evidence for the scientific protection and effective management of mangrove ecosystems. Based on the cloud computing platform of the Google Earth Engine, Landsat satellite data, and mangrove distribution data sets, combined with the Theil-Sen median trend analysis and Mann-Kendall test methods, this study analyzed the spatial variation trend of mangroves in Southeast Asia from 1990 to 2020. A factor detector and an interaction detector in the geographic detector method were used to quantitatively analyze the driving factors. First, the results showed that, from 1990 to 2020, the overall mangrove area in Southeast Asia showed a downward trend, with an area reduction of 1,467,883.1 hm² and an annual average loss rate of 1.1%. According to the statistics of the mangrove area change rate based on cellular network data, the total loss of mangrove area was as high as 40.11%. Of the total mangrove area, the regions with a change rate between -99% and -50% accounted for 28.57%, regions with a change rate between -49% and -5% accounted for 8.79%, and regions with a change rate between -4% and -1% accounted for only 13.95%. Regions with a change rate between 1% and 99% accounted for 8.18%. Only 0.4% of the mangrove area remained unchanged. The regions with increased mangrove area were primarily distributed in the Philippines, western and eastern Indonesia, and northern Vietnam. Second, the improvement and degradation of mangroves in Southeast Asia showed a concomitant distribution from 1990 to 2020. The degradation area of mangroves (79.25%) was much larger than the improvement area (20.32%) in Southeast Asia. Specifically, the significantly degraded areas accounted for 31.58%, the slightly degraded areas accounted for 47.67%, the significantly improved areas accounted for 4.75%, the slightly improved areas accounted for 15.57%, and only 0.42% of the area remained stable. Finally, the increase in aquaculture pond area was the main driving factor for mangrove area decline, with q values above 30%. Additionally, the shortest distance of mangroves from roads and population changes also had significant effects on the decline of mangrove area. The average annual temperature, precipitation, and topography had a relatively low degree of influence on mangrove changes. The results of the interaction detector indicated that all factors showed interaction enhancement, and the interaction between the culture tank and other factors was significantly stronger than that between other factors. The single factors with relatively weak driving force, such as average annual rainfall, temperature, and topography, had a significantly stronger influence on the interaction with other factors and showed a nonlinear enhancement effect.

Key words: mangrove, Google Earth Engine, geographic detector, drivers, Southeast Asia