Publications

Zhang, L; Weng, QH; Shao, ZF (2017). An evaluation of monthly impervious surface dynamics by fusing Landsat and MODIS time series in the Pearl River Delta, China, from 2000 to 2015. REMOTE SENSING OF ENVIRONMENT, 201, 99-114.

Abstract
Researchers have been attending increasingly to impervious surface dynamics to better understand the urbanization process and its impacts on urban environments. Previously, numerous studies have only estimated and mapped impervious surface dynamics at annual or decadal time scales. It is challenging to estimate impervious surface dynamics at a finer time scale, such as on a monthly scale, while using a single source of medium spatial resolution satellite imagery. However, urban infrastructure construction could cause changes in impervious surfaces in a short period of time. This paper aimed at developing a new methodology for evaluating monthly impervious surface dynamics by fusing Landsat and MODIS time series. The Pearl River Delta in China, is located in a humid subtropical region and was selected as the study area due to its dramatic urbanization in the past three decades. Available Landsat images with cloud cover < 90%, 7-Day MODIS NDVI 250 m smooth time series, and daily MODIS LST 1000 m time series from 2000 to 2015 were downloaded. These data were used to develop temporal features of land covers (i.e., monthly Landsat NDVI and LST time series) and to monitor impervious surface dynamics using semi-supervised time series fuzzy clustering method. The results showed the effectiveness of temporal features in differentiating land covers. Additionally, the average overall classification accuracy yielded reasonable accuracies (up to 89.36%). The proposed methodology has illustrated numerous, considerable advantages over previous methods. It has offered consistent maps of impervious surfaces on a monthly time scale as well as enhanced distinguishability of land covers with similar spectral characteristics. This study can be utilized to establish relationships between urban expansion, climate change, urban environment, population, and other socio-economic variables on a monthly basis. The study is also crucial for predicting the timing, duration, and density of ecological change for increased impervious surfaces.

DOI:
10.1016/j.rse.2017.08.036

ISSN:
0034-4257