Consistent, accurate, high resolution, long time-series mapping of built-up land in the North China Plain

Jinzhu Wang,Michalis Hadjikakou,Brett A Bryan

doi:10.1080/15481603.2021.1948275

Abstract

ABSTRACT Accurate, long time-series, high-resolution mapping of built-up land dynamics is essential for understanding urbanization and its environmental impacts. Despite advances in remote sensing and classification algorithms, built-up land mapping which only uses spectral data and derived indices remains prone to uncertainty. We mapped the extent of built-up land in the North China Plain, one of China’s most important agricultural regions, from 1990 to 2019 at three-yearly intervals and 30 m spatial resolution. We applied Discrete Fourier Transformation to dense time-stack Landsat data to create Fourier predictors to reduce mapping uncertainty. As a result, we improved the overall accuracy of built-up land mapping by 8% compared to using spectral data and derived indices. In addition, a temporal correction algorithm applied to remove misclassified pixels further improved mapping accuracy to a consistently high level (>94%) over the time periods. A cross-product comparison showed that our maps achieved the highest accuracies across all years. The built-up land area in the North China Plain increased from 37,941 km2 in 1990–1992 to 131,578 km2 in 2017–2019. Consistent, high-accuracy, long time-series built-up land mapping provides a reliable basis for formulating policy and planning in one of the most rapidly urbanizing regions on this planet.

Highlights

Economic development and population growth have led to drastic changes of the Earth’s terrestrial surface, not least through the expansion of built-up lands (Elmore et al, 2012), with urbanization continuing to accelerate in developing countries (United Nations, 2014)
Tracking built-up land dynamics with high accuracy over time is a significant challenge because mapping for earlier years is often less accurate than for more recent times (e.g., 2010 onwards)
This study aims to accurately track built-up land dynamics at three-year intervals from 1990 to 2019 using dense-stack Landsat data and temporal correction

Summary

Introduction

Economic development and population growth have led to drastic changes of the Earth’s terrestrial surface, not least through the expansion of built-up lands (Elmore et al, 2012), with urbanization continuing to accelerate in developing countries (United Nations, 2014). Tracking the dynamics of built-up land is essential to linking human activities with ecological, environmental, and climatic impacts and supporting policy and planning for sustainable development (Vannier et al, 2019). Tracking built-up land dynamics with high accuracy over time is a significant challenge because mapping for earlier years (e.g., the 1990s) is often less accurate than for more recent times (e.g., 2010 onwards). Random noise such as cloud and cloud shadows can lead to inconsistencies in the built-up land mapping (Foga et al, 2017). Unfavorable atmospheric conditions such as cloud cover can compromise satellite observation, builtdoi:10.20944/preprints202012.0105.v1 up land mapping may require additional images from a broader temporal range (Liu et al, 2018). Increasing classification accuracy in earlier years, removing inconsistent classification, and using an appropriate time frame is crucial for mapping built-up land with high accuracy through time

Objectives

Methods

Results

Discussion

Conclusion