Detecting spatiotemporal changes of large-scale aquaculture ponds regions over 1988–2018 in Jiangsu Province, China using Google Earth Engine

Abstract

Although the aquaculture industry is a major supplier of aquatic products used as protein sources for humans, the rapid expansion of aquaculture in recent decades has led to numerous environmental problems in coastal and lakeside areas. Nonetheless, the spatiotemporal understanding of aquaculture ponds is limited, despite playing a crucial role in the management, planning, and development within the aquaculture industry. Remote sensing images can be used to acquire information pertaining to aquaculture; however, the automatic identification of aquaculture ponds remains an important challenge due to their complex spectral and spatial characteristics. This study utilized a decision-tree classifier using Landsat data derived from the Google Earth Engine (GEE) cloud platform to automatically identify large-scale aquaculture ponds in Jiangsu Province through seven time slices between 1988 and 2018. The results showed that the area allocated to aquaculture ponds in Jiangsu Provinces has continually increased from 660.29 km2 in 1988 to 4097.95 km2 in 2018, and three regions with dense aquaculture pond distributions have developed in the coastal zone, as well as in the south and center of Jiangsu Province. The area devoted to aquaculture ponds grew exponentially from 1988 to 2003 and later experienced a slower growth rate. Moreover, inland aquaculture pond areas began to decrease since 2013; however, a rapid expansion endures in the coastal area. Socio-economic development and industry policies are the main drivers of aquaculture growth; however, the strengthening of national environmental protection policies in the 21st century has contributed to a slowdown in these trends.