Can we locate shrimp aquaculture areas from space? - A case study for Thailand

Globally, shrimp aquaculture has undergone a rapid development in the last decades, as it can help to satisfy the increasing food demand of a growing population. However, shrimp production can be accompanied by environmental impacts, such as land cover changes associated with pond construction, or the degradation of coastal areas through pollution. Environmental footprinting, has proven to be a valuable tool for tracing environmental impacts from human consumption back to their location and sector of origin. Here, we focus on the land footprint, which quantifies the area of required land resources to satisfy human consumption (of shrimp production). However, today's footprinting tools often lack spatially explicit land cover information for land footprint assessments. In this study we developed a new method, which allows us to identify the land cover change caused by shrimp pond construction in Thailand without using sample shrimp pond shape polygons as input data. We use the global water surface explorer (using globally 3 million Landsat 5 TM, Landsat 7 ETM and Landsat 8 OLI images, acquired between 1984 and 2015), aerial photographs and land cover maps in combination with known aquaculture locations, to identify water areas in Thailand that have a high likelihood to be a shrimp pond and to assess the corresponding land cover change. We estimated that in 2015 an area of 377 km(2) had a high likelihood of being shrimp pond water area. Further, we show that the construction of shrimp ponds in Thailand was responsible for the transformation of 552 km(2) primary habitat, such as mangrove areas. Our results support the environmental footprint assessment of shrimp ponds in Thailand, while our proposed method allows identifying possible shrimp pond areas on a global scale.