Peatlands play a crucial role in the global carbon cycle. Sphagnum mosses (peat mosses) are considered to be the peatland ecosystem engineers and contribute to the carbon accumulation in the peatland ecosystems. As cold-adapted species, the dominance of Sphagnum mosses in peatlands will be threatened by climate warming. The response of Sphagnum mosses to climate change is closely related to the future trajectory of carbon fluxes in peatlands. However, the impact of climate change on the habitat suitability of Sphagnum mosses on a global scale is poorly understood. To predict the potential impact of climate change on the global distribution of Sphagnum mosses, we used the MaxEnt model to predict the potential geographic distribution of six Sphagnum species that dominate peatlands in the future (2050 and 2070) under two greenhouse gas emission scenarios (SSP1-2.6 and SSP5-8.5). The results show that the mean temperature of the coldest quarter, precipitation of the driest month, and topsoil calcium carbonate are the main factors affecting the habitat availability of Sphagnum mosses. As the climate warms, Sphagnum mosses tend to migrate northward. The suitable habitat and abundance of Sphagnum mosses increase extensively in the high-latitude boreal peatland (north of 50° N) and decrease on a large scale beyond the high-latitude boreal peatland. The southern edge of boreal peatlands would experience the greatest decline in the suitable habitat and richness of Sphagnum mosses with the temperature rising, and would be a risk area for the transition from carbon sink to carbon source. The spatial-temporal pattern changes of Sphagnum mosses simulated in this study provide a reference for the development of management and conservation strategies for Sphagnum bogs.

This dataset contains six Sphagnum species occurrence record maps, the correlation between 36 environmental variables, and the potential distribution of six Sphagnum species under current climatic scenario. Sphagnum occurrence records were obtained from the Global Biodiversity Information Facility (https://www.gbif.org/), the Consortium of North American Bryophyte Herbaria (www.bryophyteportal.org), the Smithsonian National Museum of Natural History (https://naturalhistory.si.edu/), the National Specimen Information Infrastructure database (http://www.nsii.org.cn), the Chinese Virtual Herbarium (https://www.cvh.ac.cn), and field surveys. Thirty-six environmental factors were obtained to model the current species distribution patterns, including elevation, 19 bioclimatic variables, and 16 soil variables. The elevation and 19 bioclimatic variables with 5 min (approximately 10 km²) spatial resolution were obtained from the World Climate Database (version 2.0, http://worldclim.org/) (Fick & Hijmans, 2017), and 16 soil variables were obtained from Harmonized World Soil Database (version 1.2, https://www.fao.org/soils-portal/). All of the environmental variables were resampled at a 5 min spatial resolution. To avoid overfitting the model owing to the multi-collinearity of environmental variables, we examined the correlation between environmental variables using ArcGIS10.2. 

The potential distribution of Sphagnum species was simulated using the maximum entropy model (MaxEnt version 3.4.0; http://www.cs.princeton.edu/; Phillips et al., 2006). In this study, 75% of occurrence data were used for model training, and the remaining 25% was randomly selected for model testing. The maximum number of background points was set to 10000 and the algorithm was run with 3000 iterations, while other values were kept as default. Ten replications of each scenario were performed to assess the average results.