Extraction and Spatio-Temporal Analysis of Impervious Surfaces over Dongying Based on Landsat Data

Abstract

It is necessary to understand the relationship between the impervious surface area (ISA) distribution, variation trends and potential driving forces over Dongying, Shandong Province. We extracted ISA information from Landsat images with 3–5 year intervals during 1995 to 2018 using Minimum Noise Fraction (MNF) transform, Pixel Purity Index (PPI), and Linear Spectral Mixture Analysis (LSMA), followed by the analysis on three driving forces of ISA expansion (physical geography, socioeconomic factors, and urban cultural features). Our results show the retrieved ISA thematic map fit the limited requirement of root mean square error (RMSE). The correct classification accuracy of ISA is greater than 83.08%. Further, the cross–comparison exhibits the general consistent with the ISA distribution of the land use classification map published by the National Basic Geographic Information Center. The gradual increasing trend can be captured on the expansion of ISA from 1995 to 2018. Despite of the central region always shown as the high ISA density, it still keeps increasing annually and radiating the surrounding region, especially in the southward which has formed into a new large–scale and high intensity of ISA in 2015–2018. Though the ISA patches scattered in the west region or along the northern and eastern part of the ocean coastline are still small, the expansion trend of ISA can be detected. The expansion intensity index (EII) of ISA measuring the situation of its expansion changes from the lowest value 0.12% between 1995 and 2000 up to the highest 0.73% between 2000 and 2005. Richly endowed by nature, the city’s natural geographical environment provides an elevated chance of further urbanization. The rapid increase of regional economy provides a fundamental driving force for expanding ISAs. The development of urban culture promotes the sustainable development of ISAs. Our results provide a scientific basis for future urban land use management, construction planning, and environmental protection in Dongying.

Comparing Machine Learning Algorithms for Soil Salinity Mapping Using Topographic Factors and Sentinel-1/2 Data: A Case Study in the Yellow River Delta of China

“…The saline soil accounts for 36% of the whole content of urban soil. The natural vegetation in Dongying City mainly includes reed, Suaeda salsa, thatch and artemisia [34]. is 555.9 mm.…”

Impervious Surface Mapping Based on Remote Sensing and an Optimized Coupled Model: The Dianchi Basin as an Example

“…Traditional research methodologies typically take into account a limited number of features present in remote sensing images, thereby failing to comprehensively capture the distinctions between IS and other land types. There are studies that attempt to leverage a diverse set of features in the process of IS identification, but they often overlook the significance of feature selection [25,26], leading to feature redundancy. In this research, it is hypothesized that the classification accuracy may be influenced by both the category and the number of combined features, and distinct categories of features might be optimally suited to varying machine learning algorithms.…”