Next Article in Journal
Ensemble Neural Networks for Modeling DEM Error
Previous Article in Journal
Automatic Detection of Objects in 3D Point Clouds Based on Exclusively Semantic Guided Processes
Open AccessArticle

Estimating 2009–2017 Impervious Surface Change in Gwadar, Pakistan Using the HJ-1A/B Constellation, GF-1/2 Data, and the Random Forest Algorithm

Research Center for Digital Mountain and Remote Sensing Application, Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, Chengdu 610041, China
*
Author to whom correspondence should be addressed.
ISPRS Int. J. Geo-Inf. 2019, 8(10), 443; https://doi.org/10.3390/ijgi8100443
Received: 28 August 2019 / Revised: 27 September 2019 / Accepted: 2 October 2019 / Published: 8 October 2019
The China–Pakistan Economic Corridor (CPEC) is the flagship project of the Belt and Road Initiative. At the end of the CPEC, the Gwadar port on the Arabian Sea is being built quickly, providing an important economical route for the flow of Central Asia’s natural resources to the world. Gwadar city is in a rapid urbanization process and will be developed as a modern, world-class port city in the near future. Therefore, monitoring the urbanization process of Gwadar at both high spatial and temporal resolution is vital for its urban planning, city ecosystem management, and the sustainable development of CPEC. The impervious surface percentage (ISP) is an essential quantitative indicator for the assessment of urban development. Through the integration of remote sensing images and ISP estimation models, ISP can be routinely and periodically estimated. However, due to clouds’ influence and spatial–temporal resolution trade-offs in sensor design, it is difficult to estimate the ISP with both high spatial resolution and dense temporal frequency from only one satellite sensor. In recent years, China has launched a series of Earth resource satellites, such as the HJ (Huangjing, which means environment in Chinese)-1A/B constellation, showing great application potential for rapid Earth surface mapping. This study employs the Random Forest (RF) method for a long-term and fine-scale ISP estimation and analysis of the city of Gwadar, based on the density in temporal and multi-source Chinese satellite images. In the method, high spatial resolution ISP reference data partially covering Gwadar city was first extracted from the 1–2 meter (m) GF (GaoFen, which means high spatial resolution in Chinese)-1/2 fused images. An RF retrieval model was then built based on the training samples extracted from ISP reference data and multi-temporal 30-m HJ-1A/B satellite images. Lastly, the model was used to generate the 30-m time series ISP from 2009 to 2017 for the whole city area based on the HJ-1A/B images. Results showed that the mean absolute error of the estimated ISP was 6.1–8.1% and that the root mean square error (RMSE) of the estimation results was 12.82–15.03%, indicating the consistently high performance of the model. This study highlights the feasibility and potential of using multi-source Chinese satellite images and an RF model to generate long-term ISP estimations for monitoring the urbanization process of the key node city in the CPEC. View Full-Text
Keywords: impervious surface; HJ-1A/B constellation; GF-1/2; time series; random forest; China-Pakistan Economic Corridor (CPEC) impervious surface; HJ-1A/B constellation; GF-1/2; time series; random forest; China-Pakistan Economic Corridor (CPEC)
Show Figures

Graphical abstract

MDPI and ACS Style

Bian, J.; Li, A.; Zuo, J.; Lei, G.; Zhang, Z.; Nan, X. Estimating 2009–2017 Impervious Surface Change in Gwadar, Pakistan Using the HJ-1A/B Constellation, GF-1/2 Data, and the Random Forest Algorithm. ISPRS Int. J. Geo-Inf. 2019, 8, 443.

Show more citation formats Show less citations formats
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop