The ability to rapidly produce accurate land use and land cover maps regularly and consistently has been a growing initiative as they have increasingly become an important tool in the efforts to evaluate, monitor, and conserve Earth’s natural resources. Algorithms for supervised classification of satellite images constitute a necessary tool for the building of these maps and they have made it possible to establish remote sensing as the most reliable means of map generation. In this paper, we compare three machine learning techniques: Random Forest, Support Vector Machines, and Light Gradient Boosted Machine, using a 70/30 training/testing evaluation model. Our research evaluates the accuracy of Light Gradient Boosted Machine models against the more classic and trusted Random Forest and Support Vector Machines when it comes to classifying land use and land cover over large geographic areas. We found that the Light Gradient Booted model is marginally more accurate with a 0.01 and 0.059 increase in the overall accuracy compared to Support Vector and Random Forests, respectively, but also performed around 25% quicker on average.

As green infrastructure gets its attention in hazard mitigation, it is necessary to improve general understanding on what makes green infrastructure important for hazard and resiliency research. To better understand how green infrastructure fits with more traditional notions of structural and nonstructural mitigation, this study examines the relationship between green infrastructure and ‘structural and nonstructural’ mitigation approaches for hazard mitigation. Also, this study discusses a new measurement of locational aspects and spatial patterns of green infrastructure by utilizing high-resolution data in urban areas, and its potential implementation in hazard mitigation. Compared to previous research using land-use land-cover datasets, the normalized difference vegetation index (NDVI) utilizing National Agriculture Imagery Program dataset provides an ability to capture green infrastructure in greater detail. A visual comparison suggests that the NDVI data are able to capture and identify more types of ‘green’ land uses in Harris County. The total green infrastructure percentages for Harris County, Texas, based on 1-m high resolution were found to be 61.5% of the area, compared to the 51.5% based on the National Land Cover Database. This study provides support for utilizing high-resolution data to establish guidelines for green infrastructure’s spatial characteristics and sustainable hazard mitigation. The outcomes of this study will be helpful in the strategic planning and implementation of green infrastructure in urban areas with hazard issues.

Sprawl has been named as one of the critical reasons for the latest social problems in many parts of the world. This is particularly true for developing countries, as their national status largely depends on economic stability and interacts with the rise and decline of major cities. This study focuses on a detailed notion on environmental impact of physical expansion and answers how to specifically estimate the ecological impact of sprawl using the GIS and ecological valuation method. Especially, South Korean cities are examined to identify how development-oriented growth would affect natural stock and the ecology as a whole. By implementing land cover datasets and an estimation method, value transfer, the authors examine the economic losses of Korean ecological stock between 1980 and 2000. Since 1980, the society has gained a significant amount of growth in its national economics. Specifically, GDP has increased from about $40 billion to $640 billion. However, due to its rapid growth, the entire natural stock has lost about 5% of its total features, using the median economic values. If calculated with the maximum values, it is about a 7% decrease. The results indicate that $2076/person for environmental opportunity costs is estimated as a consequence of rapid urbanization. If we had estimated the ecological consumptions of rapid growth from the beginning and considered $2076/person for environmental opportunity costs, then the development patterns and other associated urban planning agendas would have shifted accordingly to increase the overall sustainability. Like most developing cities in the world, major cities in South Korea and the central government concentrated its main strategy on economic growth. Doing so stimulated national economy and made it possible to level up the quality of life. If this quality of life needs to be sustained for a long term, then we should focus on our usage of ecological features, as their characteristics are completely different from man-made resources.