Abstract:

[Objectives] In the context of global climate change, studies on coastal wetlands and their carbon sink capacity face both major opportunities and challenges. Therefore, investigating their spatiotemporal distribution is crucial for achieving the “dual carbon” goals. [Methods] Taking the coastal wetlands of Chongming Island, Shanghai, as the study area, Sentinel-2 remote sensing images in 2015, 2017, 2019, and 2021 were used. Based on corrected carbon density and land use derived from supervised classification, the spatiotemporal distribution characteristics of carbon storage were obtained. The influencing factors of carbon storage were quantitatively analyzed using the geodetector method. [Results] The periphery of Chongming Island is dominated by wetlands, with natural wetlands (mainly river-lake water bodies, grasslands, reed beds, and tidal flats) primarily distributed along the shoreline, while the inner area is non-wetland. The area of both artificial and natural wetlands increased significantly, by approximately 20 000 hm². The carbon storage of Chongming Island first increased and then decreased, but wetland carbon storage remained high, showing an overall positive trend of annual increase (approximately 600 000 tons). Conversions from non-wetland to both natural and artificial wetlands led to increases in carbon storage, indicating the high carbon sequestration potential of coastal wetlands. Natural factors had a weak influence on wetland carbon storage in Chongming Island, whereas socioeconomic development had a stronger impact. The geodetector q-values for economic added value and land use intensity reached 0.79 and 0.82, respectively. The interactive effects of natural and human factors, such as GPP combined with economic added value and population, yielded a q-value of up to 0.99, highlighting the importance of human-nature harmony in enhancing carbon sequestration in wetlands. [Conclusion] Using meteorological data from Shanghai and Chongming Island, together with a carbon density correction model, the local carbon density of Chongming District was derived. This method has low data acquisition difficulty, as most meteorological data required for local carbon density calculations can be obtained from the study area’s statistical yearbooks, and pre-correction carbon density can be retrieved from other literature. The method is applicable to coastal wetlands and other “dual carbon” focus areas, enabling accurate acquisition of localized parameters and improving the accuracy of carbon storage estimation to some extent. Additionally, directly applying geodetector to carbon storage simplifies the analysis process compared to indirect detection via land cover types and improves accuracy. The results show that wetland areas are generally increasing, with a significant growth in the proportion of natural wetlands. Carbon storage in Chongming’s coastal wetlands has increased annually, indicating initial success in wetland conservation. Dual-factor interactive effects have a greater impact on coastal wetland carbon storage than single-factor effects, and carbon storage is greatly influenced by socioeconomic factors.

Key words: carbon storage, spatiotemporal distribution, driving factors, geodetector, Chongming Island