Analysis and Prediction of Expansion of Central Cities Based on Nighttime Light Data in Hunan Province, China

Quantifying the characteristics of urban expansion as well as influencing factors is essential for the simulation and prediction of urban expansion. In this study, we extracted the built-up regions of 14 central cities in the Hunan province using the DMSP-OLS night light remote sensing datasets from...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Yuxin Liu, Tian He, Yi Wang, Changhui Peng, Hui Du, Shuai Yuan, Peng Li
Formato: article
Lenguaje:EN
Publicado: MDPI AG 2021
Materias:
Acceso en línea:https://doaj.org/article/ebb5bdee33b048dfabc2498b21a2e0ca
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Descripción
Sumario:Quantifying the characteristics of urban expansion as well as influencing factors is essential for the simulation and prediction of urban expansion. In this study, we extracted the built-up regions of 14 central cities in the Hunan province using the DMSP-OLS night light remote sensing datasets from 1992 to 2018, and evaluated the spatial and temporal characteristics of the built-up regions in terms of the area, expansion speed, and main expansion direction. The backpropagation (BP) neural network and autoregressive integrated moving average (ARIMA) model were used to predict the area of the built-up regions from 2019 to 2026. The model predictions were based on the GDP, ratio of the secondary industry output to the GDP, ratio of the tertiary industry output to the GDP, year-end urban population, and urban road area. The results demonstrated that the built-up area and expansion speed of the central cities in the eastern part of the Hunan province were significantly higher than those in the western part. The main expansion directions of the 14 central cities were east and south. The urban road area, year-end urban population, and GDP were the main driving factors of the expansion. The urban expansion model based on the BP neural network provided a high prediction accuracy (R = 0.966). It was estimated that the total area of urban built-up regions in the Hunan province will reach 2463.80 km<sup>2</sup> by 2026. These findings provide a new perspective for predicting urban areas rapidly and simply, and it also provides a useful reference for studying the spatial expansion characteristics of central cities and formulating a sustainable urban development strategy during the 14th Five-Year Plan of China.