Nation-wide soil sampling was conducted in 41 cities that were evenly distributed across China during 2013–2014 . The selected cities were either provincial capitals or representative cities at the regional scale. They covered all major climate regimes, vegetation types and soil types in China, except for the Tibet Plateau. Soil samples were taken from multiple sites in each city. To guaranty that the sampling sites in a city were evenly distributed, each sample site belonged to a different city district. In addition, observations from previous studies in the Chendu city and Yixin city were compiled and added to this dataset.

The sampled ISA types included pavements, highways, patios, driveways, parking lots etc. Field surveys and remote sensing analysis have confirmed that the sealing of impervious soils and the land cover/vegetation types (trees, shrubs, lawns, bare land, vegetable fields ect) of the adjacent pervious surface area (PSA) have been stable for over ten years at the sampling sites (detailed information of the sampling sites are found in the data file). At each site, three representative sampling plots, at least 10 m apart from each other, were set up under the ISA with three paired sampling plots in an adjacent PSA for comparison. A 100 cm depth profile pit was dug in each plot. Using 100 cm3 sample rings, soil profiles in the pits were sampled at 20 cm intervals to 100 cm depth. Technosols were thought to contain large amount of artefacts such as bricks, glass, etc. (FOA, 2015; Lorenz & Lal, 2009). However, our study across China found that most of the Ekranic (sealed) Technosols profiles has a clear boundary between the building material layers and the soil. These soils didn’t have extraordinary large amount of artefacts mixed in. Where the boundary is unclear, we treated the surface layers with large amount of hard building materials, where artefacts > 0.15 mm accounted for over half of the soil weight or volume, as the building material layers and only sampled the soils below them. Samples with notable additions of anthropogenic artefacts mixed in the soil were discarded, because we were unsure if they represented the building materials or the soils. These samples (ID# XJBIZC0001-XJBIZC4356) are currently stored in the Herbarium of the Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences, Urumqi, China. We measured the soil bulk density (BD) and SOC content in each soil sample and then calculated the SOC density by multiplying them together. All samples were air-dried, grounded and sieved at 0.15 mm. BD was measured using the volumetric ring method. To match the methodology of the Second China’s National Soil Surveys that provided information of the background SOC density, the SOC content was measured using the Mebius method involving Walkley-Black acid digestion. SOC density was calculated according to dry BD (g cm-3) and SOC content based on a soil depth of 20 cm. For each soil layer, the SOC densities of the three sampling plots in a site were averaged to derive the site-level mean SOC density.