Abstract
Accurately mapping impervious surface dynamics has great scientific significance and application value for urban sustainable development research, anthropogenic carbon emission assessment and global ecological environment modeling. In this study, a novel and accurate global 30 m impervious surface dynamic dataset (GISD30) for 1985 to 2020 was produced using the spectral generalization method and time-series Landsat imagery, on the Google Earth Engine cloud-computing platform. Firstly, the global training samples and corresponding reflectance spectra were automatically derived from prior global 30 m land-cover products after employing the multitemporal compositing method and relative radiometric normalization. Then, spatiotemporal adaptive classification models, trained with the migrated reflectance spectra of impervious surfaces from 2020 and pervious surface samples in the same epoch for each 5° × 5° geographical tile, were applied to map the impervious surface in each period. Furthermore, a spatiotemporal consistency correction method was presented to minimize the effects of independent classification errors and improve the spatiotemporal consistency of impervious surface dynamics. Our global 30 m impervious surface dynamic model achieved an overall accuracy of 91.5 % and a kappa coefficient of 0.866 using 18,540 global time-series validation samples. Cross-comparisons with four existing global 30 m impervious surface products further indicated that our GISD30 dynamic product achieved the best performance in capturing the spatial distributions and spatiotemporal dynamics of impervious surfaces in various impervious landscapes. The statistical results indicated that the global impervious surface has doubled in the past 35 years, from 5.116 × 105 km2 in 1985 to 10.871 × 105 km2 in 2020, and Asia saw the largest increase in impervious surface area compared to other continents, with a total increase of 2.946 × 105 km2. Therefore, it was concluded that our global 30 m impervious surface dynamic dataset is an accurate and promising product, and could provide vital support in monitoring regional or global urbanization as well as in related applications. The global 30 m impervious surface dynamic dataset from 1985 to 2020 generated in this paper is free to access at http://doi.org/10.5281/zenodo.5220816 (Liu et al., 2021b).
Highlights
Impervious surfaces are usually defined as surfaces “preventing the surface water from penetrating into 35 the ground” and are composed of anthropogenic materials, such as steel, cement, asphalt, bricks and stone (Chen et al, 2016; Weng, 2012; Zhang et al, 2020)
As it was difficult to assess the accuracy of all the training samples, we randomly selected 10,000 impervious surface samples from the global sample pool, and found that these impervious 600 training samples achieved an accuracy of 95.52% in 2020
The spatiotemporal consistency checking method was applied to independent impervious surface products in order to minimize the effects of classification errors and ensure the reliability and spatiotemporal consistency of the final impervious surface dynamic dataset
Summary
Impervious surfaces are usually defined as surfaces “preventing the surface water from penetrating into 35 the ground” and are composed of anthropogenic materials, such as steel, cement, asphalt, bricks and stone (Chen et al, 2016; Weng, 2012; Zhang et al, 2020). Over the past few decades, with the rapid growth of the population and the economy, impervious surfaces have been undergoing dramatic expansion, especially in developing countries (Gong et al, 2019; Kuang, 2020). As an indicator of the intensity of human activities and economic development, the dynamic information of impervious surfaces plays a significant role in urban planning (Li et al, 2015), biogeochemical cycles (Zhang and Weng, 2016), greenhouse gas emissions and urban heat island effects (Gao et al, 2012; Zhou et al, 2018), and urban sustainable development pathways (Liu et al, 2020b). With the continuous improvement of remote sensing techniques as well as computer storage and computing capabilities, global impervious surface monitoring has been undergoing a transition from the coarse spatial resolution of 1 km to the fine resolution of 30/10 m (Corbane et al, 2020; Gong et al., 2020; Liu et al, 2018; Liu et al, 2020b; Schneider et al, 2009; Zhao et al, 2020; Zhou et al, 2018)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
