| Aquaculture plays an important role in ensuring global food security and nutrient supply.In recent decades,the depletion of wild fishery resources,the rising global population and the growing demand for food and protein have promoted the rapid development of aquaculture.The low-lying terrain and abundant brackish water resources make coastal zone a major area for aquaculture development.However,the rapid development of coastal aquaculture has also led to a series of environmental and ecological problems.For example,large-scale conversion of coastal land use types,destruction of natural habitats in wetlands,eutrophication of nearshore water bodies,etc.Chinese aquaculture accounts for the largest share of the world,with aquaculture activities spread throughout coastal zone.Aquaclture ponds is the main main aquaculture model,whose distribution directly reflects the spatial layout of aquaculture in China.Rapid and accurate assessment for the spatial-temporal distribution of aquaculture ponds in the coastal zone provides support for the management and sustainable development of aquaculture industry.Some studies have focused on the remote sensing extraction of coastal aquaculture.However,most of these sdudies are based on conventional water indices and object-oriented identification methods.Due to the similarity of the spatial morphology among aquaculture ponds and salt pans,lakes,reservoirs and other natural water bodies,aquaculture ponds are easily confused with these water bodies.Up to now,the characteristics of high trophic status in aquaculture ponds have not been uesd as the remote sensing discrimination rules for aquaculture ponds.Our study brings the water environmental parameters into the identification system of aquaculture ponds,and based on Google Earth Engine(GEE)cloud computing platform and high-resolution Sentinel-2 time series remote sensing data,we proposed a remote sensing intelligent identification algorithm for coastal aquaculture ponds,which combines the spatial morphology and water environment parameters(i.e.chlorophyll-a concentration,nutrient status index and floating algae index).Using this algorithm,we completed the extraction of Chinese coastal aquaculture ponds in 2017,2019 and 2021,and then analyzed the spatial distribution,temporal variation characteristics and driving forces of aquaculture ponds in Chinese coastal zone,meanwhile the water environment status and dynamics of aquaculture ponds.The main conclusions are as follows:(1)This paper proposed a remote sensing intelligent identification algorithm for coastal aquaculture ponds based on spatial morphology and water environment parameters.The algorithm uses water index and threshold methods,and combines with time-series remote sensing data to eliminate the influence of rainstorm inundation and water bodies with seasonal characteristic such as paddy fields on the extraction results.In addition,and then combined three water environment parameters retrieved from the bio-optical models,namely Chl-a concentration,TSI and FAI.and solves the problem of misclassification among aquaculture ponds and water bodies with similar morphological and spectral characteristics such as salt pans,rivers,lakes and reservoirs.It has been verified that the overall accuracy of our algorithm can reach 91% in Chinese coastal zone with obvious spatial heterogeneity.(2)In 2021,the total area of aquaculture ponds in Chinese coastal zone is 1,023,034 ha.Shandong Province,Guangdong Province and Jiangsu Province account for21.89%,19.06% and 14.07% of the total area of aquaculture ponds in Chinese coastal zone,respectively.The distribution of aquaculture ponds in Chinese coastal zone has obvious characteristics of aggregation.On the one hand,it is manifested in geographical location: aquaculture ponds are densely distributed in low-lying plains and estuarine areas such as Bohai Bay,Northern Jiangsu Plain and Pearl River Estuary.On the other hand,it is manifested in the distance from the sea: aquaculture ponds are most densely distributed within 5 km of the coastline,accounting for about 58% of the total area of aquaculture ponds in the study area.With the increase of the distance from the sea,the area and intensity of aquaculture ponds in the unit buffer zone gradually decreased,and84.30% of the aquaculture ponds were concentrated in the 15 km buffer zone along the shoreline.(3)From 2017 to 2021,the area of aquaculture ponds in Chinese coastal zone show an overall decreasing trend,with a total decrease of 31,961 ha in the four years,showing a characteristic of “increasing in the north and decreasing in the south” in the geospatial pattern.Specific to each province,there are mainly four different trends 1)first increase,and then decrease.For example,Jiangsu Province and Taiwan Province 2)keep stable.For example,Tianjin,Shanghai and Hong Kong 3)continuously increase.For example,Shandong Province;4)continuously decrease.For example,Hebei Province and Guangdong Province.(4)From 2017 to 2021,the mean water environment parameters of aquaculture ponds in Chinese coastal zone show a general trend of decline.In 2021,the Chl-a concentration,TSI and FAI of aquaculture ponds in Chinese coastal zone are 25.70μg/L,59.34 and 0.015,respectively.Which decreased by 4.85%,4.14% and 11.76%respectively compared with 27.01 μg/L,61.90 and 0.017 in 2017.By analyzing the relationship between the water environmental parameters of aquaculture ponds and aquaculture species,the results show that the water environment parameters of aquaculture ponds in Chinese coastal zone have obvious correlations with their aquaculture species.Specically,in fish and shrimp ponds,the mean Chl-a concentration,TSI and FAI are about 35.05 μg/L,65.33,0.038.Whereas in the sea cucumber ponds,the mean Chl-a concentration,TSI and FAI are 21.20 μg/L,55.33,0.Therefore,the water environment parameters can not only distinguish aquaculture ponds from other water types,but also have great potential in distinguishing different aquaculture species. |
PDF Full Text Request