Multi-scale analysis of spatially varying relationships between agricultural landscape patterns and urbanization using geographically weighted regression

Scientific interpretation of the relationships between agricultural landscape patterns and urbanization is important for ecological planning and management. Ordinary least squares (OLS) regression is the primary statistical method in previous studies. However, this global regression lacks the ability to uncover some local-specific relationships and spatial autocorrelation in model residuals. This study employed geographically weighted regression (GWR) to examine the spatially varying relationships between several urbanization indicators (urbanization intensity index, distance to urban centers and distance to road) and changes in metrics describing agricultural landscape patterns (total area, patch density, perimeter area ratio distribution and aggregation index) at two block scales (5 km and 10 km). Results denoted that GWR was more powerful than OLS in interpreting relationships between agricultural landscape patterns and urbanization, since GWR was characterized by higher adjust R², lower Akaike Information Criterion values and reduced spatial autocorrelations in model residuals. Character and strength of the relationships identified by GWR varied spatially. In addition, GWR results were scale-dependent and scale effects were particularly significant in three aspects: kernel bandwidth of weight determination, block scale of pattern analysis, and window size of local variance analysis. Homogeneity and heterogeneity in the relationships between agricultural landscape patterns and urbanization were subject to the coupled influences of the three scale effects. We argue that the spatially varying relationships between agricultural landscape patterns and urbanization are not accidental but nearly universal. This study demonstrated that GWR has the potential to provide references for ecological planners and managers to address agricultural landscapes issues at all scales.     

Exploring spatially varying and scale-dependent relationships between soil contamination and landscape patterns using geographically weighted regression

Landscape pattern is an important determinant of soil contamination at multiple scales, and a proper understanding of their relationship is essential for alleviating soil contamination and making decisions for land planners. Both soil contamination and landscape patterns are heterogeneous across spaces and scale-dependent, but most studies were carried out on a single scale and used the conventional multivariate analyses (e.g. correlation analysis, ordinary least squared regression-OLS) that ignored the issue of spatial autocorrelation. To move forward, this paper examined spatially varying relationships between agricultural soil trace metal contamination and landscape patterns at three block scales (i.e. 5 km × 5  km, 10 km × 10 km, 15 km × 15 km) in the Pearl River Delta (PRD), south China, using geographically weighted regression (GWR). This paper found that GWR performed better than OLS in terms of increasing R square of the model, lowering Akaike Information Criterion values and reducing spatial autocorrelation. GWR results revealed great spatial variations in the relationships across scales, with an increasing explanatory power of the model from small to large block scales. Despite a few negative correlations, more positive correlations were found between soil contamination and different aspects of landscape patterns of water, urban land and the whole landscape (i.e. the proportion, mean patch area, the degree of landscape fragmentation, landscape-level structural complexity, aggregation/connectivity, road density and river density). Similarly, more negative correlations were found between soil contamination and landscape patterns of forest and the distance to the river and industry land (p < 0.05). Furthermore, most significant correlations between soil contamination and landscape variables occurred in the western PRD across scales, which could be explained by the prevailing wind, the distribution of pollutant sources and the pathway of trace metal inputs.     

Environmental changes,remote sensing,and infrastructure development: The case of Egypt's East Port Said harbour

Using Landsat TM data, this article examines the environmental impact of the East Port Said harbour project on the surrounding landscape. The optimum three-band combination and the most appropriate multispectral bands were selected to enhance the images and monitor land cover changes for the periods of 1984–1991 and 1991–2003. The results indicate that wetland areas declined from 103 km² in 1984 to 30 km² in 2003. In addition, the surface area of El-Malha Lake has shrunk from 27 km² to 18 km² over the same period. In contrast, the area covered by salt crust has increased from 11 km² in 1984 to 19 km² in 2003. Urban land use and designed cultivated lands were also significant in 2003, covering 49 km² and 71 km², respectively. The rate of shoreline change between 1984 and 2003, the period when the East Port Said harbour was constructed, was calculated. Vector data indicate that the rate of shoreline loss was ?13 m/year from 1984 to 1991 and ?15 m/year from 1991 to 2003. Despite the fact that construction of the East Port Said harbour caused significant changes in the study area, there are several factors controlling coastline and land cover changes including industrial development and fish cultivation farms.     

Spatial expansion and water requirements of urban agriculture in Khartoum,Sudan

Urban agriculture (UA) in Khartoum, Sudan, contributes to meeting the increasing food demand of the city's rapidly growing population, but its spatial extent, development over time and resource consumption are unknown. To fill this knowledge gap, we analyzed scenes captured by Landsat satellites in 1972, 1987 and 2000. For each dataset, we calculated the Normalized Difference Vegetation Index (NDVI) and classified all grid cells with NDVI values higher than one standard deviation above the mean as vegetated land. We also quantified agricultural vegetation on aerial photographs taken in 1958 and on a recent Google Earth image. The built-up area of the city was estimated for each point in time. Based on these spatial estimates, we computed water consumption by UA based on geographic, temporal and climatic datasets and official data on the cropping patterns of Khartoum State, using the Penman–Monteith equation to estimate crop water demands, and assuming a water use efficiency of 0.56.Urban agriculture increased from 8751 ha in 1972 to 13,249 ha today. In the municipal core zone, UA decreased from 4799 ha in 1958 to 2869 ha today. Along the Nile, UA persisted through time, whereas many urban irrigation schemes have been pushed into the periphery. The built-up area increased strongly and today exceeds the UA area 5.5-fold. Estimated crop water demand on UA land increased from 241 to 365 million m³ year^?1 between 1972 and today.     

环杭州湾城市群土地利用的空间冲突与复杂性

土地利用的空间冲突与复杂性研究可为土地利用空间管控提供理论依据和实践支撑。以环杭州湾城市群1990年、2000年、2008年和2013年四期遥感影像为例,分析土地利用系统复杂性、脆弱性及动态性,计算空间冲突综合指数。结果表明：土地利用空间冲突等级表现为一般冲突>中度冲突>轻度冲突>重度冲突;空间冲突的空间格局呈现团状、块状及带状聚集模式分布,具有明显的梯度特征;随着时间的推移,空间冲突的空间自相关性减弱,并具有明显的边缘效应。研究区土地利用空间冲突表现为均衡→打破原有均衡→出现空间自相关→出现空间趋势→形成空间自组织的演变规律。土地利用空间冲突从多核心向单核心模式发展,区域发展水平较高且发展趋于均衡。     

城市景观格局尺度效应的空间统计规律--以上海中心城区为例

以上海市外环以内的中心城区为区域背景,基于SPOT全色波段遥感影像数据和GIS技术,运用空间统计分析方法,研究城市景观格局的空间尺度效应。研究结果表明:(1)城市景观格局的空间自相关性具有显著的尺度依赖性。Moran指数I与Geary系数c对粒度变化的敏感点相同,均为50m。(2)城市景观形态具有分形特征,各类景观斑块的分维数对粒度变化的响应不同,它们随着粒度的变化呈非线性变化趋势;在粒度较小时,各类景观斑块之间的分维数差异较大,而随着粒度的增大,其分维数之间的差异逐渐缩小。(3)城市景观多样性格局,与空间区位及人类活动的空间格局息息相关。在城市中心,主导型景观是经济效益较高的商业文化景观,并且景观斑块聚集度、破碎度大;经济效益较低的农业景观分布在城市边缘区,而且景观类型单一,斑块面积较大,破碎度较小;在由城市中心向边缘过渡的中间地带,景观类型多样,空间格局复杂。(4)景观多样性具有尺度依赖性。随着幅度的增加,景观多样性指数逐渐增大,多样性的空间格局也显著变化。在0.5km幅度下,多样性指数的最大值出现在市中心,从市中心向外呈现高低起伏的环状模式扩展,随着幅度增加,多样性指数的高值区向景观类型变化最剧烈的城乡过渡地带转移。(5)景观多样性的空间变异也具有明显的尺度依赖性。在较     

Changes in arable land in response to township urbanization in a Chinese low hilly region: Scale effects and spatial interactions

Rapid urbanization leads to losses in arable land; quantitatively analyzing the impact of urbanization on arable land is significant for arable land management. However, changes in arable land due to urbanization with scale and neighborhood effects remains poorly understood at the town scale. In this study, high-resolution historical land use data, landscape metric analysis and spatial regression were integrated to quantify the impacts of urbanization on arable land use change (abandonment and conversion) at spatial scales of 300 m–3300 m using a block size increment of 200 m and at the catchment scale in the town of Jinjing in subtropical central China. Arable land abandonment was the predominant type of arable land change and presented strong spatial autocorrelations at each spatial scale. Arable land was converted to tea fields because agricultural structure transformations were occurring around the urban cores, and the amount of arable land converted to residential land accounted for only a small proportion of the total arable land loss and had no spatial autocorrelation. The significance and robustness of the arable land changes impacted by urbanization had obvious scale effects and strong neighborhood effects in nearby regions. Compared with block scales, the catchment scale is an optimal scale for assessing the influence of urbanization and applying planning policy. Our results highlight the significance of incorporating spatial interactions in urbanization research, which can generate less biased estimations and consequently lead to proper policy implication and recommendations. In addition, multi-scale comparisons are helpful for better understanding the relationships between arable land changes and urbanization and provide further insights into the harmonious development of rural settlements and urban cores to preserve arable land.     

Ecological benefits of urban forestry: The case of Kerwa Forest Area (KFA), Bhopal,India

In developing countries like India, migration of people from rural to urban areas is responsible for ever expanding urban boundaries. This trend is exerting significant pressure on unprotected natural forests located near urban centers. This paper highlights the case of Kerwa Forest Area (KFA), located at about 10 km from the city of Bhopal, capital of Madhya Pradesh state. The objectives of this study are to quantify the extent of disturbance faced and ecosystem services provided by the KFA. Suitable spatial technologies and forest sampling techniques have been used to achieve the objectives of the study. It was found that the KFA is currently facing severe anthropogenic pressure. Parts of the KFA, located close to the settlements, were found more disturbed than the parts which were located far from the settlements. In spite of disturbances, KFA is a habitat for many threatened and endangered plant, animal, and bird species. KFA also plays a critical role of a carbon sink with a total storage of about 19.5 thousand tons of aboveground carbon. Immediate precautionary measures are required to prevent further degradation of the KFA for ensuring better environmental quality for the residents of Bhopal city in the future.     

Unmanned magnetometer network observation in the 44th Japanese Antarctic Research Expedition: Initial results and an event study on auroral substorm evolution

In the 44th Japanese Antarctic Research Expedition (JARE-44) during 2003 to 2004, four unmanned magnetometers were deployed in Antarctica to establish a observation network for studying ionospheric and magnetospheric phenomena. Three of them were set about 80 km from Syowa Station, and the fourth was set at Dome Fuji Camp, about 800 km from Syowa. Observations were carried out continuously with a maximum sampling rate of 1 Hz. The purpose of the close network around Syowa was to observe the two-dimensional distribution of the ionospheric equivalent current in a localized area within the field-of-view of the all-sky auroral imager operated at Syowa. Dome Fuji is located at the higher-latitude edge of the auroral zone, while Syowa is located in the middle of it. Auroral observations using the all-sky imager were also carried out at Dome Fuji in 2003. The simultaneous auroral and magnetic observations both at Dome Fuji and Syowa enabled us to study auroral activities in a larger area. This paper describes, the deployment and performance of the unmanned magnetometers in the JARE-44. One auroral substorm event is analyzed in detail to show the usefulness the unmanned magnetometer network.     

Forest dynamics and the importance of place in western Honduras

Analyses of landscape change using remotely sensed satellite imagery constitute a large component of forest transition research, allowing for assessments of large areas. In the western highlands of Honduras is an area of complex forest dynamics (～45,000 ha) that has seen significant forest regeneration in recent years. However, analysis of the larger region (～500,000 ha) shows net forest loss. The comparative aspects highlight the importance of site selection and scale in forest transition analysis, a process often ignored in the land-use and land-cover change (LULCC) and forest transition literature. Results also highlight the importance of analyzing human-induced fragmentation at a variety of selected sites and a range of spatial scales, and producing quality, accurate forest cover and change maps.     

深圳市近20年城市景观格局演变及其驱动因素

基于深圳市1996—2015年土地利用数据,利用景观指数、景观转移矩阵和景观扩张指数等方法探究了深圳市近20年景观格局时空变化、主要景观类型转移和建筑用地扩张模式,最后使用Binary Logit模型考察了市级和区级建筑用地景观扩张的主要驱动因素。结果表明：① 1996—2015年,深圳市建筑用地景观优势性逐步增强,面积增加15.81%,以蔓延式（61.9%）和填充式（36.27%）扩张为主;② 1996—2006年为城市化快速扩张期,建筑用地扩张呈集中开发形态,景观多样性和均匀性增加,城市扩张中心略微向北部和东部移动,2006—2015为城市化低速过渡期,景观破碎化加剧,城市扩张重心向北部和西部偏移;③ 在市级尺度上,GDP密度和人口密度对建筑用地景观扩张有显著正影响,生态控制线、高程、坡度和至道路的距离有着显著负影响。每单位生态控制线范围、坡度的增加分别将使建筑用地景观扩张的机会比率将平均减少到原来0.8168倍、0.8841倍。各驱动因素表现出区域和尺度差异性,GDP对宝安区、南山区和坪山区,人口增长对宝安区、龙华区,以及交通可达性对大鹏新区、龙岗区驱动分别最为突出。研究结果可以为中国城市快速扩张过程中的景观格局变化提供科学实践。     

Spatial and temporal patterns across an ecological boundary: Allochthonous effects of a young saltwater lake on a desert ecosystem

We documented changes in the abundance and composition of terrestrial flora and fauna with respect to distance from the sea edge and timing of large allochthonous inputs from the Salton Sea, California. We found significant effects that were most pronounced within 300 m of the shore, but extended 3 km inland via coyote scat deposition. The zone within 300 m of the sea had a higher density of vegetation with a distinctly different plant composition. The denser vegetation supported higher abundances of birds and reptiles. Coyotes exhibited spatial and temporal responses to marine subsidies of fish, while birds were likely subsidized by aquatic aerial insects. Top-down control, as well as dietary and habitat preferences, may have resulted in reduced number of ants, beetles, and small mammals near the sea. Species responses to the habitat edge appeared to be associated with life history, as the near shore habitat favored habitat generalists and shore specialists, while inland desert habitat favored many sand and open desert specialists. Ecosystem responses support current theories of allochthonous spatial subsidies and consumer-resource dynamics but were limited in scope, magnitude, and distance.     

Aboveground biomass in Tibetan grasslands

This study investigated spatial patterns and environmental controls of aboveground biomass (AGB) in alpine grasslands on the Tibetan Plateau by integrating AGB data collected from 135 sites during 2001–2004 and concurrent enhanced vegetation index derived from MODIS data sets. The AGB was estimated at 68.8 g m^?2, with a larger value (90.8 g m^?2) in alpine meadow than in alpine steppe (50.1 g m^?2). It increased with growing season precipitation (GSP), but did not show a significant overall trend with growing season temperature (GST) although it was negatively correlated with GST at dry environments (<200 mm of GSP). Soil texture also influenced AGB, but the effect was coupled with precipitation; increased silt content caused a decrease of AGB at small GSP, and generated a meaningful increase under humid conditions. The correlation between AGB and sand content indicated an opposite trend with that between AGB and silt content. An analysis of general linear model depicted that precipitation, temperature, and soil texture together explained 54.2% of total variance in AGB. Our results suggest that moisture availability is a critical control of plant production, but temperature and soil texture also affect vegetation growth in high-altitude regions.     

Spatiotemporal heterogeneity of soil fertility in the Central Monte desert (Argentina)

In arid environments, soil fertility exhibits a high degree of spatial and temporal heterogeneity, which results from high climatic variability seasonally and heterogeneous plant distribution. However, because most desert areas have been altered by human activities, heterogeneous fertility would originate from grazing or logging activities. We evaluated spatial and temporal heterogeneity of soil fertility in cattle-excluded sites under and outside woody plant cover (Prosopis flexuosa and Larrea divaricata), and in sites disturbed by tree removal during wet and dry season in Ñacuñán Biosphere Reserve (Central Monte desert of Argentina). Soil organic matter, fulvic acids, bioavailable organic matter, and nitrate were lower outside plant canopy (8.9 mg g^?1, 0.03 mg g^?1, 8.2 mg g^?1, and 4.17 mg kg^?1, respectively). Total N, humic acids, and abundance of microbial functional groups did not show differences among sites. Most parameters differed between seasons, tending to be higher in the wet season. Overall soils of Ñacuñán Reserve are characterized by: a) more homogenous spatial pattern than expected from woody plant presence; b) very heterogeneous temporal pattern; and c) after two years, tree removal does not seem to induce infertile soil formation.     

Visual assessment of the Australian Land Erodibility Model

There is a growing requirement for techniques to assess land susceptibility to wind erosion, i.e. land erodibility, over large geographic areas (>10⁴ km²). This requirement stems from a lack of wind erosion research between the field (10¹ km²) and regional scales, and a need to evaluate the performance of spatially explicit wind erosion models across these scales. This paper addresses this issue by presenting a methodology for monitoring land erodibility at the landscape scale (10³ km²). First, we define criteria suitable for evaluating land erodibility based on empirical relationships between soil texture, vegetation cover, geomorphology, and wind erosion. The criteria were used to visually assess land erodibility over long distances (10³ km) using vehicle-based transects run through the rangelands of western Queensland, Australia. Application of the data for testing the performance of a spatially explicit land erodibility model (AUSLEM) is then demonstrated by comparing the visual assessments of land erodibility with the model output. The model performed best in the west of the study area in the open rangelands. In regions with higher woody shrub and tree cover the model performance decreases. This highlights the need for research to better parameterise controls on erodibility in semi-arid landscapes consisting of forested and rangeland mosaics.     

Fluctuations of the Última Esperanza ice lobe (52°S), Chilean Patagonia,during the last glacial maximum and termination 1

We present a new record from the Última Esperanza region (51°25’-52°25'S), southwestern Patagonia, to unravel the timing and structure of glacial fluctuations during the Last Glacial Termination (T1). This sector of southern South America represents the only windward-facing continental landmass in the Southern Hemisphere that intersects the core of the Southern Westerly Wind belt.Geomorphic, stratigraphic and geochronological evidence indicate the following stages during and since the Last Glacial Maximum (LGM): (i) deposition of prominent moraine complexes during at least two advances dated between ~ 39 and > 17.5 ka; (ii) development of an ice-dammed proglacial lake (glacial lake Puerto Consuelo) accompanying ice recession; (iii) active deposition of moraine complexes at intermediate positions followed by recession at ≥ 15.2 ka; (iv) lake level drop and subsequent stabilization between 15.2-12.8 ka; (v) a glacial readvance in glacial lake Puerto Consuelo between 14.8-12.8 ka; (vi) ice recession, stabilization, and lake level lowering between 12.8-10.3 ka; and (vii) glacial withdrawal and disappearance of glacial lake Puerto Consuelo prior to 10.3 ka. By comparing our results with the chronologies from neighboring regions we explore whether there was a consistent temporal/geographic pattern of glacial fluctuations during the LGM and T1, and examine their implications at regional, hemispheric, and global scales. The correspondence of these variations with key paleoclimate events recorded in the Southern and the Northern Hemispheres suggest a common forcing that, most likely, propagated through the atmosphere. Regional heterogeneities at millennial timescales probably reflect the influence of processes related to deep ocean circulation, and changes in the position/intensity of the Antarctic Polar Front and Southern Westerly Winds.     

Modelling spatial patterns of urban growth in Africa

The population of Africa is predicted to double over the next 40 years, driving exceptionally high urban expansion rates that will induce significant socio-economic, environmental and health changes. In order to prepare for these changes, it is important to better understand urban growth dynamics in Africa and better predict the spatial pattern of rural-urban conversions. Previous work on urban expansion has been carried out at the city level or at the global level with a relatively coarse 5–10 km resolution. The main objective of the present paper was to develop a modelling approach at an intermediate scale in order to identify factors that influence spatial patterns of urban expansion in Africa. Boosted Regression Tree models were developed to predict the spatial pattern of rural-urban conversions in every large African city. Urban change data between circa 1990 and circa 2000 available for 20 large cities across Africa were used as training data. Results showed that the urban land in a 1 km neighbourhood and the accessibility to the city centre were the most influential variables. Results obtained were generally more accurate than results obtained using a distance-based urban expansion model and showed that the spatial pattern of small, compact and fast growing cities were easier to simulate than cities with lower population densities and a lower growth rate. The simulation method developed here will allow the production of spatially detailed urban expansion forecasts for 2020 and 2025 for Africa, data that are increasingly required by global change modellers.     

Allometric relationships and peak-season community biomass stocks of native shrubs in Senegal's Peanut Basin

Allometric equations and community biomass stocks are presented for Guiera senegalensis J.F. Gmel (Gs) and Piliostigma reticulatum (DC.) Hochst (Pr) – two native shrub species in the Sahel. These shrubs are of interest because they dominate semi-arid sub-Sahalien Africa but have been largely overlooked as a key biomass component and regulator of ecosystem composition and function in this landscape. In Year 1, best predictors of aboveground biomass were height and number of stems (Gs) and crown diameter (Pr); and for belowground biomass were height and basal diameter (Gs) and basal diameter (Pr). In Year 2, height and crown diameter were the best predictors of aboveground biomass (R² = 0.90 for Gs and 0.87 for Pr), whereas basal diameter and number of stems (Gs) and basal diameter (Pr) were best predictors of belowground biomass. Peak-season biomass estimates ranged from 0.44 to 4.58 ton ha^?1 (mean = 2.38 ton ha^?1) in the Gs sites and from 0.33 to 7.38 ton ha^?1 (mean = 3.71 ton ha^?1) in the Pr communities. Both species exhibited unusually large root:shoot ratios (4.5:1 for Gs and 10.2:1 for Pr). Although models differ between years, allometric relationships provide reasonable biomass estimates for Gs and Pr.     

中国城市群城市用地扩张时空动态特征

认识城市群城市用地扩张特征、扩张质量及景观生态效应等综合信息将有助于城市群空间高质量发展规划与决策。选取10个不同发育等级的城市群作为研究对象,综合运用城市用地扩张强度指数、城市用地扩张强度差异指数、景观扩张指数、景观格局指数,分析1990—2015年期间城市群城市用地扩张时空特征及景观生态效应,并构建人口、经济与城市用地协同发展指数,据此分析城市用地扩张中“人—地”关系协同程度。结果表明：① 1990—2015年,各个城市群城市用地面积持续增加,建成区的范围不断扩大,扩张强度呈现“上升—下降”的趋势;在2010年之前成熟型城市群的扩张强度最高,2010年之后培育型城市群成为扩张强度最高的类型。② 城市群城市用地扩张总体上呈现不均衡布局的趋势,高速扩张区域主要分布在沿海地区。从全国视角来看,东部高于中、西、东北部,整体城市群快速扩张的中心由东向西移动。③ 城市群城市用地空间扩张模式表现出边缘式和飞地式两种空间扩张模式并存,不同发育等级之间具有明显差异性,成熟型城市群的空间结构较为稳定,其他两种类型的城市群空间结构初具雏形或尚未形成。④ 城市群城市用地扩张“吞没”了郊区和农村地区,破碎化程度、形状的复杂程度较高,景观格局表现出不稳定。⑤ 城市群“人口—土地”协同程度逐步增强,“经济—土地”协同程度整体较高,整体上表现出城市用地扩张质量有所提升。新时期城市群城市用地应严格控制扩张规模,加快中小城市的发展,促进城市群城市用地扩张均衡发展,为城市群国土空间高质量发展提供重要参考。     

厦门岛城市空间扩张特征及其影响因素

Most of the world’s cities are concentrated in coastal areas. As a special geographical component of the coastal system, island urban spatial expansion is the outcome of interactions between city development and the physical environment. This paper takes Xiamen Island, located in Southeastern China, as an example to analyze island urban spatial expansion and its determinants by combining an analysis of the literature on urban development policies, urban overall plans, population growth and industrial development, with geographical information analysis using historical maps and remote sensing photographs. Firstly, we reviewed the history of the Xiamen City development during the last 100 years, which can be divided into four periods: the embryonic modern city and early development from 1908 to 1949; administrative boundary expansion and infrastructure development from 1950 to 1979; special economic zone construction and rapid urbanization from 1980 to 2003; and Bay City construction since 2003. The dynamic changes to the coastline, island shape, built-up area, transportation, administrative division, and major land use type conversion which occurred during approximately the past 100 years were analyzed and the characteristics of the island urban spatial expansion were concluded: early expansion from a central point, followed by expansion along a section of coastline, and expansion from the coastline inland. Secondly, we discussed the potential determinants of island urban spatial expansion including administrative division adjustment, urban master planning revision, industrial development, topographical factors, coastal area reclamation, transportation expansion, and population growth. Finally, the effects of each potential determinant on island urban spatial expansion were concluded. Island urban spatial expansion is the result of a synthesis of natural and socio-economic factors which are not independent but interacting. Built-up area expansion is the major driver of island land cover and land use changes. By this paper, we hope to provide a scientific reference contributing to the rational understanding of island and coastal sustainable urbanization in China, and the world beyond.