期刊
LANDSCAPE AND URBAN PLANNING
卷 214, 期 -, 页码 -出版社
ELSEVIER
DOI: 10.1016/j.landurbplan.2021.104187
关键词
Floor area map; Random forest; Mainland China
类别
资金
- National Natural Science Foundation of China [32071580, 41730647, 41871192]
- Youth Innovation Promotion Association of CAS [2021194]
This study established a model to predict the urban floor area in mainland China using remote sensing data, resulting in the generation of the FPFA map. The distribution of floor area was found to be uneven, with higher intensity in the southern urban areas along the coastline compared to the northern regions, and significant disparities among provinces.
Urbanization studies are of global interest and mainly focus on mapping urban areas and areas of expansion using remote sensing data. However, information about the 3-dimensional characteristics or expansion of urban buildings is absent due to difficulties in data acquisition. Quantifying the urban floor area is crucial for assessing urban 3-D morphology. We used a random forest regression model to predict the first urban floor area of mainland China at a 130-m spatial resolution based on high spatial resolution nighttime light LUOJIA 1-01 images (130-m), a population map (100-m), and a single building dataset encompassing 71 cities. The predicted floor area (PFA) map for mainland China was estimated from the single building dataset of 50 cities, and data from the other 21 cities were used to estimate the accuracy. The results showed that the total accuracy of the PFA map is strong (R2 = 0.68, RMSE = 7277.46 m2/ha). The PFA map overestimated the values in low value areas and underestimated the values in high value areas. The accuracy was also acceptable at the single city scale based on the results from six cities (R2 > 0.6). The calculated floor area map for 71 cities was merged with the PFA map by replacing the values in the corresponding locations to generate the final predicted floor area (FPFA) map, which enabled higher accuracy. The total floor area is 76038.39 km2, which is 0.79% of the total area of China. The general distribution of the floor area amount and intensity showed that the coastline had a higher intensity than the inner region of the country and the southern region had a higher intensity than the northern urban area along the coastline. The floor area distribution was extremely uneven among the provinces. The top six provinces represent 50.01% of the total floor area; however, the last six provinces represent only 3.31%. The high spatial resolution FPFA map of mainland China calculated by us has great potential application in urban ecology research, such as the impact of FPFA on heat island and haze.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据