GWL_FCS30: a global 30 m wetland map with a fine classification system using multi-sourced and time-series remote sensing imagery in 2020

Abstract

Abstract. Wetlands, often called the “kidneys of the earth”, play an important role in maintaining ecological balance, conserving water resources, replenishing groundwater and controlling soil erosion. Wetland mapping is very challenging because of its complicated temporal dynamics and large spatial and spectral heterogeneity. An accurate global 30 m wetland dataset that can simultaneously cover inland and coastal zones is lacking. This study proposes a novel method for wetland mapping by combining an automatic sample extraction method, existing multi-sourced products, satellite time-series images and a stratified classification strategy. This approach allowed for the generation of the first global 30 m wetland map with a fine classification system (GWL_FCS30), including five inland wetland sub-categories (permanent water, swamp, marsh, flooded flat and saline) and three coastal tidal wetland sub-categories (mangrove, salt marsh and tidal flats), which was developed using Google Earth Engine platform. We first combined existing multi-sourced global wetland products, expert knowledge, training sample refinement rules and visual interpretation to generate large and geographically distributed wetland training samples. Second, we integrated the Landsat reflectance time-series products and Sentinel-1 synthetic aperture radar (SAR) imagery to generate various water-level and phenological information to capture the complicated temporal dynamics and spectral heterogeneity of wetlands. Third, we applied a stratified classification strategy and the local adaptive random forest classification models to produce the wetland dataset with a fine classification system at each 5∘×5∘geographical tile in 2020. Lastly, GWL_FCS30, mosaicked by 961 5∘×5∘ regional wetland maps, was validated using 25 708 validation samples, which achieved an overall accuracy of 86.44 % and a kappa coefficient of 0.822. The cross-comparisons with other global wetland products demonstrated that the GWL_FCS30 dataset performed better in capturing the spatial patterns of wetlands and had significant advantages over the diversity of wetland sub-categories. The statistical analysis showed that the global wetland area reached 6.38 million km2, including 6.03 million km2 of inland wetlands and 0.35 million km2 of coastal tidal wetlands, approximately 72.96 % of which were distributed poleward of 40∘ N. Therefore, we can conclude that the proposed method is suitable for large-area wetland mapping and that the GWL_FCS30 dataset is an accurate wetland mapping product that has the potential to provide vital support for wetland management. The GWL_FCS30 dataset in 2020 is freely available at https://doi.org/10.5281/zenodo.7340516 (Liu et al., 2022).

The distribution of global tidal marshes from earth observation data

“…Moving north, beyond the reach of our maps, it is likely that tidal marshes will be less extensive owing to the impact of ice abrasion and the harsh climate (Adam, 2002;Scott et al, 2014a). Zhang et al (2023) map 4,800 km 2 in the high Arctic area, however given the differences in our mapping approaches, we would predict a much smaller extent than this in these high latitudes. It is important to note that while the tidal marshes in these high northern latitudes tend to be limited in distribution with low diversity, they are extremely vulnerable to environmental change (Adam, 1990(Adam, , 2002Chapman, 1974;Martini et al, 2019;Sergienko, 2013).…”

“…The larger extent predicted is somewhat explained by the inclusion of 4,800 km 2 along Arctic coastlines which were precluded from our study (see below). In addition, Zhang et al (2023) study predicts the presence of "coastal saltmarsh" over a far greater proportion of the global coastline, however this is partly explained by the fact that it does not apply an area filter to remove the noise (predictions of wetland distribution confined to single pixels) in their maps. No recent maps have come close to the highest previous global estimates of 400,000 km 2 (Duarte et al, 2005).…”

“…The former was a more broadly based study of intertidal wetland dynamics and thus not directly comparable. Zhang et al (2023) is a lower resolution (30m) global assessment of all wetlands, which includes "coastal saltmarsh". The larger extent predicted is somewhat explained by the inclusion of 4,800 km 2 along Arctic coastlines which were precluded from our study (see below).…”

“…In total, the authors identified almost 55,000 km 2 of tidal marshes across 43 countries and territories, although they also identified several large regions -Canada, Northern Russia, South America and Africa -for which data was lacking (Mcowen et al, 2017). A global assessment of all wetlands, Zhang et al (2023) also included a category of tidal salt marsh, mapping approximately 75,000km 2 , with other estimates from the literature ranging from 22,000 -400,000 km 2 (McLeod et al, 2011;Pendleton et al, 2012). Other efforts to characterise tidal marsh dynamics globally have typically focused on change analyses (Campbell et al, 2022; or simulation (Schuerch et al, 2018); however, a high-resolution map baseline has yet to be established.…”

An improved global land cover mapping in 2015 with 30 m resolution (GLC-2015) based on a multisource product-fusion approach

“…Globeland30, FROM_GLC, and GLC_FCS30 are excellent and indispensable GLC products which have contributed much to various research, such as biodiversity conservation (Wu et al, 2020;Meng et al, 2023), climate change (Kim et al, 2016;Xue et al, 2021;Zheng et al, 2022), and land management (Shafizadeh-Moghadam et al, 2019). In addition to these multiple-class GLC products, GLC products for individual LC classes, such as cropland (Yu et al, 2013;Lu et al, 2020), forest (Hansen et al, 2013;Shimada et al, 2014;Zhang et al, 2020), wetland Zhang et al, 2023), water (Liao et al, 2014;Pekel et al, 2016;Pickens et al, 2020), and impervious surfaces (Gong et al, 2020;Huang et al, 2021Huang et al, , 2022, have been successfully generated.…”