Assessing Impacts of Urban Form on Landscape Structure of Urban Green Spaces in China Using Landsat Images Based on Google Earth Engine

The remote sensing based mapping of land cover at extensive scales, e.g., of whole continents, is still a challenging task because of the need for sophisticated pipelines that combine every step from data acquisition to land cover classification. Utilizing the Google Earth Engine (GEE), which provides a catalog of multi-source data and a cloud-based environment, this research generates a land cover map of the whole African continent at 10 m resolution. This land cover map could provide a large-scale base layer for a more detailed local climate zone mapping of urban areas, which lie in the focus of interest of many studies. In this regard, we provide a free download link for our land cover maps of African cities at the end of this paper. It is shown that our product has achieved an overall accuracy of 81% for five classes, which is superior to the existing 10 m land cover product FROM-GLC10 in detecting urban class in city areas and identifying the boundaries between trees and low plants in rural areas. The best data input configurations are carefully selected based on a comparison of results from different input sources, which include Sentinel-2, Landsat-8, Global Human Settlement Layer (GHSL), Night Time Light (NTL) Data, Shuttle Radar Topography Mission (SRTM), and MODIS Land Surface Temperature (LST). We provide a further investigation of the importance of individual features derived from a Random Forest (RF) classifier. In order to study the influence of sampling strategies on the land cover mapping performance, we have designed a transferability analysis experiment, which has not been adequately addressed in the current literature. In this experiment, we test whether trained models from several cities contain valuable information to classify a different city. It was found that samples of the urban class have better reusability than those of other natural land cover classes, i.e., trees, low plants, bare soil or sand, and water. After experimental evaluation of different land cover classes across different cities, we conclude that continental land cover mapping results can be considerably improved when training samples of natural land cover classes are collected and combined from areas covering each Köppen climate zone.

Section: Implication Of Our African Land Cover Mapmentioning

confidence: 99%

Mapping the Land Cover of Africa at 10 m Resolution from Multi-Source Remote Sensing Data with Google Earth Engine

Qiu

et al. 2020

“…In this Special Issue (see Table 1), most studies used medium-resolution satellite images, such as Landsat Satellite series and Advanced Land-Observing Satellite (ALOS), to understand how the urbanization process influences urban ecology and urban sustainability. Particularly, the urban heat island [16][17][18] and urban green space development [19] receive the most attention from the Special Issue contributors. The study area of most studies focuses on a single city due to the heavily satellite image analysis efforts, but the existing cloud-based satellite image analysis platforms, such as Google Earth Engine, can significantly improve the spatial coverage from city level to country level [19].…”

Section: Papers In the Special Issuementioning

confidence: 99%

“…Particularly, the urban heat island [16][17][18] and urban green space development [19] receive the most attention from the Special Issue contributors. The study area of most studies focuses on a single city due to the heavily satellite image analysis efforts, but the existing cloud-based satellite image analysis platforms, such as Google Earth Engine, can significantly improve the spatial coverage from city level to country level [19]. Due to the limitation of the 3D satellite images, most of the studies explore the urban landscape patterns in 2D.…”

Section: Papers In the Special Issuementioning

confidence: 99%

“…A further extension of 3D urban landscape analysis is significant and necessary, similar to the multidimensional urban form concept proposed by Wentz et al (2018) [20] and the recent attempt by Handayani et al (2018) [21] in this Special Issue. Most of the studies incorporated a temporal dimension into the consideration (2-4 time stages) to understand the urban expansion process, except the country scale analysis by Huang et al (2018) [19]. From the methodology perspective, image analysis such as land use/land cover classification, land surface temperature derivation, and biophysical indicators calculation have been commonly used to extract useful information from the satellite images.…”

Section: Papers In the Special Issuementioning

confidence: 99%

See 1 more Smart Citation

Editorial for the Special Issue: “Remote Sensing of Urban Ecology and Sustainability”

Zhao

Wentz

2020

The remote sensing of urban ecology and sustainability is an emerging topic to understand the human living environment in urban areas from outer space, airplanes, and unmanned aerial vehicles. In this editorial, we provide an overview of the five papers published in this Special Issue and offer suggestions for future research directions in this field, both with respect to the remote sensing platforms and algorithms and the development of new applications.

“…We are, of course, not the first to face these challenges-the authors of references [16,[40][41][42] have all suggested solutions.…”

Section: Dealing With the Inevitable Instabilities Arising From Diffementioning

confidence: 99%

Combining Measurements of Built-up Area, Nighttime Light, and Travel Time Distance for Detecting Changes in Urban Boundaries: Introducing the BUNTUS Algorithm

Luqman

Gurney

2019

This paper introduces a new algorithm (BUNTUS-Built-up, Nighttime Light, and Travel time for Urban Size) using remote sensing techniques to delineate urban boundaries. The paper is part of a larger study of the role of urbanisation in changing fossil fuel emissions. The method combines estimates of land cover, nighttime lights, and travel times to classify contiguous urban areas. The method is automatic, global and uses data sets with enough duration to establish trends. Validation using ground truth from Landsat-8 OLI images revealed an overall accuracy ranging from 60% to 95%. Thus, this approach is capable of describing spatial distributions and giving detailed information of urban extents. We demonstrate the method with examples from Brisbane, Australia, Melbourne, Australia, and Beijing, China. The new method meets the criteria for studying overall trends in urban emissions.In social science, the areas with high population densities are regarded as urban or city areas [8]. Economics defines the city by its political, cultural, and economic characteristics [9].This work forms the first part of a study of the global distribution of trends in urban fossil fuel emissions and related quantities. That study requires a definition of urban extent so we can attribute emissions as urban or not. The needs of the overall study provide some important requirements for the algorithm we use for the urban extent. We will use these so often throughout the paper that we label them here: R1The study is global, so we may only choose globally consistent and available datasets. R2 We wish to study enough cities to establish patterns, so the algorithm must be computationally efficient enough for large-scale use. R3 Changes in urban extent can be small, so the algorithm must work at high resolution, no more than 1 km. R4 The study must be long enough to establish trends. We estimate this requires two decades of data.In this paper, we propose a novel and multiple-step approach which satisfies these requirements. We define an urban area as a contiguous (i.e., simply connected) and compact region including a pre-defined urban center and which satisfies several criteria relating to density and surface properties. The urban boundary is the bounding polygon for this region. The compactness requirement means that gaps such as green belts (but not water) should be included in the urban area. The criteria which define the urban area must be deducible from datasets satisfying R1, R3, and R4.The outline of the paper is as follows. In Section 2, we review existing methods for determining urban extent, focusing on their utility for our task. In Section 3, we describe our methodology and its underlying data sets. In Section 3, we also present available validation for the method and three case studies of different cities. In Section 4, we validate and discuss the results with a particular focus on the uncertainties of the method and summary of the main results. Section 5 comprises of discussion and Section 6 comprises of conclusion.