Foreword
Special Issue on Change Analysis
by Bo Huang
The ecosystem, one of the most important systems for the survival of human beings, is continuously changing as a result of human activities. Such changes may present in different forms, such as rural-urban land conversion, land degradation, deforestation, biodiversity loss, and desertification. Central to these is probably land use/land cover change, which has significant economic and environmental impacts with implications for a wide variety of policy issues, including maintenance of water quality, preservation of open space, and mitigation of global climate change.
Under the umbrella of sustainable development and stimulated by the joint international Land Use/Cover Change (LUCC) project of the International Geosphere-Biosphere Program (IBGP) and the International Human Dimensions Program on Global Environmental Change (IHDP), detection, monitoring, understanding, modeling, and projection of land use change from global to regional scale have aroused tremendous interest within the community of Remote Sensing and GIS. A significant success in the use of satellite images to detect and monitor ecosystem changes has been achieved in the past three decades.
Detecting and monitoring the change, however, is just the first step for environmental planning and management. A further step is to identify factors that drive the change (e.g., land use change) and explore their relative importance, analyze the change pattern, understand the dynamic process of change, and simulate “what-if” decision making based on a variety of scenarios; that is, change analysis (Huang et al., 2009). Change analysis is a prerequisite to understanding the complexity of spatial change and forecasting the future trends of change patterns and their ecological impacts. Only after an effective change analysis can decision makers generate solutions to maintain the balance between ecology and the needs of ever-demanding population.
This special issue collects eight papers on the subject of spatial change analysis. Although change analysis is primarily about the deducing and analyzing of the causal effects behind the change, it also entails the support of change detection and management, a phase before change analysis, and change optimization, a phase after change analysis. The papers collected mainly address three topics under the theme of change analysis.
Change detection aims to identify the temporal differences between two images of the same area. Although many change detection algorithms have been developed, most mapping agencies still rely on photo/image interpretation to undertake the map updating tasks. Currently there is also no automated system to detect and map the changes to a satisfied quality as expected by a skilled analyst. Castilla et al. introduce the Land-cover Change Mapper (LCM), an automated change detection and delineation tool for remote sensing images. LCM can efficiently generate a polygon vector layer (shapefile) of regions deemed to have undergone significant change in land-cover and allow further editing of this polygon layer by the users, if necessary.
Jungho et al. introduce a new concept, Moving Threshold Window (MTW), for binary change detection. The MTW-based model is free from the assumption of symmetry for difference and ratio types of changeenhanced images. Their experiment results demonstrate that the MTWbased calibration model outperforms the traditional STM-based (Symmetric Threshold Window) binary change detection methods for both single and multiple change-enhanced images of the study area.
Sudden oak death (SOD) has caused widespread mortality in a number of tree and shrub species in coastal California. To quantify changes in horizontal canopy structure for the oak woodlands in the study area, De Chant and Kelly developed a novel change detection technique to track canopy changes of individual objects. It is probably the first to track changes in forest canopy gap structure on an individual basis.
Creating a consistent set of spatio-temporal maps is of importance to track the land use change. Julia et al. present a framework for generating temporally and categorically dynamic land-cover maps which provide a reliable and adaptable foundation for change analysis. The centerpiece is a spatio-temporal disturbance-inventory database, created through semiautomated change detection and conditioned with boundary-matching procedures. It can be used to backdate and update both continuous and categorical reference maps.
The next three papers deal with change analysis. Pontius and Connors present a method to address the mixed pixel problem – the proportion of each category in a pixel is known, but the spatial allocation of the categories within the pixel is unknown. This method defines a range of possibilities for the cross-tabulation matrix to quantify the association between any two categories. The technique is appropriate to compare two different maps or to assess the patterns within a single map.
While the comparison of two maps with mixed pixels is considered as a special type of change analysis, the identification of the factors underlying the changes and the analysis of their relative importance is a significant part of change analysis. Wu et al. devised an improved regression method, kernel logistic regression (KLR), to model the nonlinear relationship between land use changes and various causal factors such as population, distance to road and facilities, surrounding land use and others. Their method can also be used for future land use pattern prediction.
Impact analysis is also significant to change analysis as the effects of change usually need to be elucidated. In order to identify the impacts of human activities on local and regional ecological processes, Chen and Fraser examine the spatial and temporal dynamics of Normalized Difference Vegetation Index (NDVI) at three adjacent areas with different proportions of private land in the Bankhead National Forest of Alabama, USA from 1998 to 2004. They suggest that properly managed land ownership plays an important role in maintaining the integrity of regional ecological processes.
Change analysis generally establishes the relationship between change and various causal factors. However, change optimization is to channel the change to a right direction (i.e., sustainable development in our context). Magesh et al. develop a genetic algorithm for creating sustainable land use plans with the use of visualization technology for visual assessment.
Collectively, these papers cover a wide spectrum of related topics and provide a valuable insight into the emerging area of spatial change analysis. Clearly, change analysis is a natural step after change detection and its integration with change optimization will play an increasingly important role in global change and sustainable development studies.
I wish to thank the previous PE&RS Editor-in-Chief, James Merchant, for offering me an opportunity to edit this special issue and the current Editor-in-chief, Russ Congalton, for his careful scrutiny of all the papers accepted for publication in this issue. I would also like to thank the reviewers for their constructive comments, which have helped improve the quality of the papers. Special thanks are given to the authors who have made efforts to produce and revise their papers.
Bo Huang
Department of Geography and Resource Management
The Chinese University of Hong Kong
Reference
Huang, B., L. Zhang, and B. Wu, 2009. Spatio-temporal analysis of
rural-urban land conversion, International Journal of Geographical
Information Science, 23(3), 379-398.