ART‐P‐MAP Neural Networks Modeling of Land‐Use Change: Accounting for Spatial Heterogeneity and Uncertainty

Spatial land‐use models over large geographic areas and at fine spatial resolutions face the challenges of spatial heterogeneity, model predictability, data quality, and of the ensuing uncertainty. We propose an improved neural network model, ART‐Probability‐Map (ART‐P‐MAP), tailored to address these issues in the context of spatial modeling of land‐use change. First, it adaptively forms its own network structure to account for spatial heterogeneity. Second, it explicitly infers posterior probabilities of land conversion that facilitates the quantification of prediction uncertainty. Extensive calibration under various test settings is conducted on the proposed model to optimize its utility in seeking useful information within a spatially heterogeneous environment. The calibration strategy involves building a bagging ensemble for training and stratified sampling with varying category proportions for experimentation. Through a temporal validation approach, we examine models’ performance within a systematic assessment framework consisting of global metrics and cell‐level uncertainty measurement. Compared with two baselines, ART‐P‐MAP achieves consistently good and stable performance across experiments and exhibits superior capability to handle the spatial heterogeneity and uncertainty involved in the land‐use change problem. Finally, we conclude that, as a general probabilistic regression model, ART‐P‐MAP is applicable to a broad range of land‐use change modeling approaches, which deserves future research.     

Integrating Multi‐Sensor Remote Sensing Data for Land Use/Cover Mapping in a Tropical Mountainous Area in Northern Thailand

Accurate mapping of land use/cover conditions provides essential information for managing natural resources and is critical for further examination of land use/cover change and its subsequent impacts on the environment. Remote sensing offers a means of acquiring land use/cover data in a timely manner, with optical remote sensing images commonly being used in land use/cover related studies. The persistent cloud cover during the rainy season in Southeast Asia, however, presents a challenge for using optical images in land use/cover mapping. Integrating multi‐sensor images of different spectral domains is thus desirable because more information can be extracted to improve the mapping accuracy. The purpose of this study is to assess the potential of using multi‐sensor data sets for land use/cover mapping in a tropical mountainous area in northern Thailand. Optical data from Landsat Thematic Mapper, radar images from Advanced Land Observing Satellite/Phased Array type L‐band Synthetic Aperture Radar (PALSAR), and topographical data were used, providing complementary information on land use/cover. Classification and accuracy assessment were conducted for 12 different combinations of the data sets. The results suggested that short crop mapping using multi‐temporal Phased Array type L‐band Synthetic Aperture Radar images offered insights into the distributions of crop and paddy fields. Because of the mountainous environment of the study area, combining topographic data of elevation and slope into the classification greatly reduced the confusion between different land use/cover types. Improvement of classification accuracy was evident especially in separating evergreen and deciduous forests from other vegetation types and discriminating urban village and the fallow field classes.     

Using Linear Integer Programming for Multi‐Site Land‐Use Allocation

Research in the area of spatial decision support (SDS) and resource allocation has recently generated increased attention for integrating optimization techniques with GIS. In this paper we address the use of spatial optimization techniques for solving multi‐site land‐use allocation (MLUA) problems, where MLUA refers to the optimal allocation of multiple sites of different land uses to an area. We solve an MLUA problem using four different integer programs (IP), of which three are linear integer programs. The IPs are formulated for a raster‐based GIS environment and are designed to minimize development costs and to maximize compactness of the allocated land use. The preference for either minimizing costs or maximizing compactness has been made operational by including a weighting factor. The IPs are evaluated on their speed and their efficacy for handling large databases. All four IPs yielded the optimal solution within a reasonable amount of time, for an area of 8 × 8 cells. The fastest model was successfully applied to a case study involving an area of 30 × 30 cells. The case study demonstrates the practical use of linear IPs for spatial decision support issues.     

Monitoring Saline Discharge Areas: Farrell Flat,a Case Study

By monitoring saline groundwater discharge areas, an indication of the extent of dryland salinisation processes occurring in an area may be determined. Processed remotely sensed data (SPOT MS and aerial photography) has been analysed within a geographic information system database to identify the dynamic relationship between changing vegetation cover and increasing salt levels on a saline groundwater discharge area. Temporal analysis of this database was conducted for a saline discharge area on a wheat/sheep property in South Australia from the period 1966 to 1990. The results provide a directional and quantitative assessment of the change in salt affected areas during the study period. Information of this type benefits land managers at the regional scale by quantifying salinisation trends.     

Spatial pattern of land cover changes across Northeast China over the past 300 years

To accurately model the regional climatic effects of land use/cover change processes, precise reconstructions of historical land cover are essential. Over the past 300 years, the northeastern region has undergone more significant land cover change than any other region in China owing to large numbers of farmer migrants from North China, reflecting the tremendous effects human activities can have on such changes. In this paper, based on official statistical data, documentary records, survey data, and modern remote-sensing data and vegetation and soil maps, land cover changes across Northeast China over the past 300 years have been reconstructed through cropland data calibrations using correlation analysis, potential vegetation reconstructions, urban area estimations, spatial analyses of land cover type changes by K-mean clustering analysis, and consideration of land use driving forces. The results indicate that cropland expanded gradually from south to north through the cities of Shenyang, Jilin, and Ha’erbin as well as out to the margins of the central plain of Northeast China, with a maximum expansion occurring in the 1950s. Pristine forests and grasslands diminished significantly, especially in the early twentieth century, due to this expansion of cropland. The conclusions drawn from these results are, first, that these results provide a scientific understanding of the process by which the original land cover was replaced by cropland in Northeast China over the past 300 years. Second, the results enabled the development of a series of land use/cover changes maps with 50–100 year time resolutions and county-level spatial resolutions. Based on the reconstructed results of these land cover type changes, the dynamics and causes of land cover changes are discussed in their social–historical contexts.     

Dynamics of Land Surface Temperature in Response to Land‐Use/Cover Change

In this study, we employed Geographical Information Systems and remote sensing techniques to investigate the impact of land‐use/cover change on land surface temperature (LST) in a rapidly urbanisation city, Kunming in south‐west China. Spatial patterns of LST and land use for 1992 and 2006 were derived from Landsat images to examine how LST responded to urban growth. Remote sensing indices were used to quantify land‐use types and employed as explanatory variables in LST modelling. The geographically weighted regression (GWR), a location dependent model, was performed to explore the influences of the spatially varied land‐use conditions on the LST patterns. Results revealed that rapid urbanisation in Kunming altered the local thermal environment, particularly in increasing the LST in the zone surrounding the urban core. Remote sensing indices demonstrated that water and vegetation played an important role in mitigating the urban heat island effect, while built‐up and barren land accounted for the increase in LST. The GWR improved the goodness‐of‐fit for LST modelling and provided insights into the spatially varied relationship between LST and land‐use conditions.     

Application of Landsat satellite imagery to monitor land‐cover changes at the Athabasca Oil Sands,Alberta, Canada

A major advantage of satellite remote sensing is that the imagery acquired provides a synoptic view of the landscape. Thus, repeat coverage by the satellite on a regular basis permits the detection of changes in land‐cover over time. This study demonstrates the application of remote sensing technology to the monitoring of mining activities at the Athabasca Oil Sands region of Alberta, Canada. First, we describe the techniques used to match a time sequence of Landsat imagery, both spatially and spectrally, to ensure that the spectral changes through time are due to land‐cover variations. A series of spectral trajectories were then extracted to assess changes in land‐cover through time. Secondly, a land‐cover classification was produced from the baseline 1984 imagery and, using historic and future mine extents, the classification was analyzed to determine the proportion of each land‐cover type affected through development. Results of the analysis indicate that since 1984 there has been a larger reduction in mixedwood dense and broadleaf vegetation classes than mixedwood sparse or dense conifer stands in the area. Based on the delineations of mine‐site activity, the area of woodland and wetland habitat subject to development has increased from approximately 2,520 hectare (ha) in 1984 to 32,930 ha in 2005.     

MODIS‐derived NDVI Characterisation of Drought‐Induced Evergreen Dieoff in Western North America

Vegetation stress or mortality can be the result of many factors including drought‐induced water deficit, insect infestations and failures of, or fluctuations in, precipitation sources typical to an area. Reduction of cover and reduced health are identifiable in remotely‐sensed multispectral satellite images. A suite of images from NASA's MODIS sensor was used to calculate the Normalised Difference Vegetation Index (NDVI) during the 2000–2006 North American growing seasons. Fluctuations in NDVI over this period show a significant decline in vegetative health in the region with specific areas showing changes linked to moisture sources, prevailing wind patterns, slope aspect and solar radiation receipt. Ground‐truthing of these areas has confirmed the extent and magnitude of the dieoff signal. Historically, dieoff has been reversed through regeneration as climate conditions return to a normal regime. However, quantification of recent vegetative change in western North America suggests that the degree of change may be too severe for regrowth to occur and may have far‐reaching impacts on a scale unseen in modern times. The loss of vegetative habitat and native species in semi‐arid regions and lack of regeneration in these marginal ecosystems due to prolonged drought are growing global problems. Similar drought stress impacts on marginal ecotypes have also been observed in semi‐arid regions of Australia, South America, Asia and Africa. Observations of the spatial pattern of temperate forest vegetation globally can be used to develop a precise picture of vegetative health in these regions and how they are reacting to global climate change.     

Monitoring to Learn,Learning to Monitor: A Critical Analysis of Opportunities for Indigenous Community‐Based Monitoring of Environmental Change in Australian Rangelands

Indigenous community‐based monitoring has been a central feature in many international attempts to improve monitoring of and local adaptation to environmental change. Despite offering much promise, Indigenous community‐based monitoring has been underutilised in natural resource management in Australia, particularly within the remote, semi‐arid rangelands. This paper discusses contextual social and environmental factors that may help to explain this apparent deficiency, before critically analysing key stakeholder perceptions of the roles for, and challenges of monitoring in the Alinytjara Wilur ara Natural Resources Management region in the north‐west of South Australia. The analysis guides a discussion of responses to better integrate monitoring in general, and Indigenous community‐based monitoring in particular, into regional environmental management approaches. We argue that community‐based monitoring offers a range of benefits, including: better coordination between stakeholders; a heightened ability to detect and respond to climatic trends and impacts; the effective utilisation of Indigenous knowledge; employment opportunities for managing and monitoring natural resources; and improved learning and understanding of rangeland socio‐ecological systems. Identified opportunities for spatial and temporal community monitoring designed for the Alinytjara Wilur ara region could be of value to other remote rangeland and Indigenous institutions charged with the difficult task of monitoring, learning from, and responding to environmental change.     

LOCAL PUBLIC SERVICE PROVISION AND SPATIAL INEQUALITY IN CHINESE CITIES: THE ROLE OF RESIDENTIAL INCOME SORTING AND LAND‐USE CONDITIONS

Spatial inequality refers to unequal access to local public services between high‐ and low‐income households in relation to their residential locations. We examine two hypotheses regarding the role of income sorting and land‐use conditions in shaping spatial inequality in Chinese cities, where residents have little direct influence on local public service provision. First, in the presence of resource indivisibility, travel cost, and location‐based rationing, scarcity of public‐service resources in a city makes access to public services more uneven across neighborhoods, thus exacerbating income sorting and spatial inequality in the city. Second, the exacerbating effect of resource scarcity is mitigated by land‐use conditions that limit income sorting. Estimates of willingness to pay by households of different income levels for public‐service resources across cities corroborate both the exacerbating effect of resource scarcity and the mitigating effect of inclusive land‐use conditions.     

基于GWR模型的合肥居住地价影响因素研究

以合肥市主城区为例,基于2010-2014年居住用地的出让数据,运用地统计法、GWR模型等方法,对合肥市居住地价的空间异质性及其影响因素进行研究。研究表明：①合肥市居住地价的空间分布呈现出显著的多中心的空间结构,地价的峰值区分别以老城区、政务区天鹅湖及滨湖新区塘西河公园为中心呈现圈层式分布;②不同的地价影响因素表现出不同的空间分布特征,其中容积率对居住地价的贡献度空间差异最大,其次是宗地面积,主干路次之,交通站点对居住地价的贡献度最小;③厘清各影响因素对地价的作用机制,建立动态的数字地价模型,不仅能促进土地资源的集约利用,重塑城市的空间结构,而且能为城市整体价值的发挥提供重要的理论支撑。     

长春市居住与工业空间演进的耦合性测度及影响因素

以长春市为研究对象,基于用地现状图等数据,借助ArcGIS软件,采用斑块聚合指数、标准差椭圆、核密度估计等研究方法,对1995-2013年长春市居住与工业空间演进的耦合性进行了测度,发现： ①长春市居住、工业与城市用地分布的“离散化”趋势明显,其中居住用地的斑块聚合指数波动幅度较大;②从居住与工业用地“中心-外围”分布特征来看,三环以内居住与工业用地呈现明显的“逆协同”演变规律,二者的用地密度变化表现出明显的差异性;③从居住与工业用地格局演进来看,二者布局的耦合度呈下降趋势。从成长路径依赖、政府调控引导、市场经济驱动、科技创新带动、空间主体的决策等六个方面分析了影响居住与工业空间演进耦合性的影响因素。     

Spatial Modelling in Decision Support for Land‐Use Planning: a Demonstration from the Lal Lal Catchment,Victoria, Australia

The outcomes of selected, contested development permit appraisals in a Victorian water catchment show that implications regarding the development status of other land parcels were not examined during negotiation and/or appeal. Given access to an adequate spatial data base, Geographical Information System (GIS) technology can be used to model such implications rapidly enough to be useful in decision support, thereby significantly widening the common ground that can be sought while seeking resolution to land‐use planning conflicts.     

Integrating Landsat and Land Systems for Cover Maps in Southern Inland Queensland

Digital analysis of Landsat data is applied to up-dating land cover maps for the Goondiwindi district of southern, inland Queensland. The imagery was paired with climax cover maps obtained from land systems reports with analysis conducted separately for each of three climax cover types (open forest, woodland and open woodland). Computer classification using this approach produced readily interpreted and reliable results. This work demonstrates that Landsat can be used in conjunction with existing, land systems derived maps to produce data that combine the advantages of two approaches to environmental survey, the former being based on current land cover/land use and the latter on the original environment and its land use potential.     

基于GIS的盐城城市内部空间结构演变分析

根据1984、1997、2003、2006年4个时相数据,运用GIS的空间分析方法,对盐城城市工业、居住、公共服务用地的空间分异特征进行分析,并利用多种景观格局指数,分析了三种功能用地的完整性、集聚性和异质性特征。结果表明:1984-2006年盐城城市内部空间结构演变过程可划分为集聚、分散、再集聚三个阶段,同时演变中存在"工业围城"、城市中心区的工业和居住用地比重偏大等问题。     

Google Earth, GIS and stone-walled structures in southern Gauteng, South Africa

In South Africa, air photos were used in the 1960s and 1970s to plot distribution maps of pre-colonial stone-walled structures in order to study the peopling of this landscape. Different architectural styles of stone-walled structures were attributed to different cultures, who shared a mixed agricultural and pastoralist economic base and a cattle centered world-view. New technologies such as Google Earth satellite imagery as well as Geographic Information System software justify revisiting these structures as they facilitate more complex analyses of larger databases. The spatial analysis of remotely sensed settlement data from the Suikerbosrand Nature Reserve near Johannesburg shows significant changes in settlement patterns from dispersed homesteads to nucleated towns during the last two or three centuries before colonial times. These changes echo similar patterns reported in the neighboring North West Province, where they have been interpreted as a sequence of evolution in social, political and economic complexity. In the Suikerbosrand reserve climate change, conflict and other factors may have helped bring about the observed changes in settlement patterns.     

Exploring the land development process and its impact on urban form in Hamilton, Ontario

The form of many Canadian cities has dramatically evolved over the past six decades due to urban sprawl. Several patterns can characterize this evolution including unlimited horizontal expansion of the city, leapfrog, and low‐density residential development at the outskirts, and widespread strip commercial and power centre retail development. Hamilton, Ontario is an example of a Canadian city that has experienced suburbanization and sprawled development for several decades. However, the nature of this sprawled development is unclear and its impact on urban form is not entirely understood. In this article, several hypotheses pertaining to sprawled land development and urban form are postulated and tested. The tests rely on point source data of the developed land parcels in Hamilton during the period 1950–2003. A number of spatial statistics techniques, including kernel estimation and complete spatial randomness (CSR) K‐function tests, are employed to examine the emerging nature of urban form. We hypothesize that while the city has been sprawling, the ongoing land development process is leading urban form into multinucleation. To support this assumption, we further hypothesize the existence of an interdependent spatial relationship between residential and commercial land uses at the emerging nuclei. Accordingly, we examine the strength of co‐clustering among these land use activities over time. The findings indicate that while the city has been sprawling, several consequent urban nuclei with mixed land use activities have been emerging and become more visible in recent years. This is an indication that the city's form is progressively becoming multinuclear. Furthermore, the estimates for the 1990s indicate interdependence between the locational patterns of residential and commercial land development. Co‐clustering between these two types of land uses is bi‐directional and occurs at a time lag of three to seven years. These findings affirm the existence of interdependence between land use activities at the observed nuclei, which support the emergence of a multinucleation.     

Pastoralism and Land‐Tenure Change in Kenya: The Failure of Customary Institutions

Until recently, the Pokot in the highlands of the Baringo area in Kenya have practised semi‐nomadic pastoralism. Today they are rapidly sedentarizing and in many areas suitable for farming, they are adopting rain‐fed agriculture. As a result of these dynamics, claims to individual property on de facto communal rangelands have arisen, and to such an extent that they seriously threaten the peace of the community. This article explores the conflicts that emerge in the transition from common property to private tenure. Using locally prominent land disputes as exemplary cases, it focuses on the role of traditional gerontocratic authorities in the attempt to resolve a growing number of land disputes; on the emerging power of patrilineal clans and local elites in the enforcement of access to land; and on the incompetence of government agencies to intervene. The failure of customary institutions to ensure land tenure security leads to a situation in which women and marginalized actors in particular are threatened with displacement, and in which most local actors want the state to intervene and establish formal property rights.     

长株潭城市群都市区各类城市建设用地的时空变化研究

以长株潭城市群的都市区为研究对象,通过收集1990-2019年中7个年份的8类城市建设用地的空间分布数据,建立空间数据库,并依托各类城市建设用地的增量、对城市群及其片区的贡献率、在城市群中的扩张程度分布的测算,归纳各类城市建设用地时空变化的特征与规律.结果表明:1990-2019年,长株潭城市群都市区的城市建设用地扩张...