TY - JOUR

T1 - A New Approach to High-Resolution Urban Land Use Classification Using Open Access Software and True Color Satellite Images

AU - Chapa Zumba, José Fernando

AU - Hariharan, Srividya

AU - Hack, Jochen

N1 - Funding Information: Acknowledgments: We acknowledge support from the German Research Foundation and the Open Access Publishing Fund of Technische Universität Darmstadt. Funding Information: Funding: This research was funded by the Working Internships in Science and Engineering (WISE) program of the German Academic Exchange Service (DAAD), and by the German Fedral Ministry of Eduaction and research (BMBF), grant number 01UU1704.

PY - 2019/9/25

Y1 - 2019/9/25

N2 - Urbanization nowadays results in the most dynamic and drastic changes in land use/land cover, with a significant impact on the environment. A detailed analysis and assessment of this process is necessary to take informed actions to reduce its impact on the environment and human well-being. In most parts of the world, detailed information on the composition, structure, extent, and temporal changes of urban areas is lacking. The purpose of this study is to present a methodology to produce high-resolution land use/land cover maps by the use of free software and satellite imagery. These maps can help to understand dynamic urbanizations processes to plan, design, and coordinate sustainable urban development plans, especially in areas with limited resources and advancing environmental degradation. A series of high-resolution true color images provided by Google Earth Pro were used to do initial classifications with the Semi-Automatic Classification Plug-in in QGIS. Afterwards, a new methodology to improve the classification by the elimination of shadows and clouds, and a reduction of misclassifications through superimposition was applied. The classification was carried out for three urban areas in León, Nicaragua, with different degrees of urbanization for the years 2009, 2015, and 2018. Finally, the accuracy of the classification was analyzed using randomly defined validation polygons. The results are three sets of high-resolution land use/land cover maps of the initial and the improved classification, showing the detailed structures and temporal dynamics of urbanization. The average accuracy of classification reaches 74%, but up to 85% for the best classification. The results clearly identify advancing urbanization, the loss of vegetation and riparian zones, and threats to urban ecosystems. In general, the level of detail and simplicity of our methodology is a valuable tool to support sustainable urban management, although its application is not limited to these areas and can also be employed to track changes over time, providing therefore, relevant information to a wide range of decision-makers.

AB - Urbanization nowadays results in the most dynamic and drastic changes in land use/land cover, with a significant impact on the environment. A detailed analysis and assessment of this process is necessary to take informed actions to reduce its impact on the environment and human well-being. In most parts of the world, detailed information on the composition, structure, extent, and temporal changes of urban areas is lacking. The purpose of this study is to present a methodology to produce high-resolution land use/land cover maps by the use of free software and satellite imagery. These maps can help to understand dynamic urbanizations processes to plan, design, and coordinate sustainable urban development plans, especially in areas with limited resources and advancing environmental degradation. A series of high-resolution true color images provided by Google Earth Pro were used to do initial classifications with the Semi-Automatic Classification Plug-in in QGIS. Afterwards, a new methodology to improve the classification by the elimination of shadows and clouds, and a reduction of misclassifications through superimposition was applied. The classification was carried out for three urban areas in León, Nicaragua, with different degrees of urbanization for the years 2009, 2015, and 2018. Finally, the accuracy of the classification was analyzed using randomly defined validation polygons. The results are three sets of high-resolution land use/land cover maps of the initial and the improved classification, showing the detailed structures and temporal dynamics of urbanization. The average accuracy of classification reaches 74%, but up to 85% for the best classification. The results clearly identify advancing urbanization, the loss of vegetation and riparian zones, and threats to urban ecosystems. In general, the level of detail and simplicity of our methodology is a valuable tool to support sustainable urban management, although its application is not limited to these areas and can also be employed to track changes over time, providing therefore, relevant information to a wide range of decision-makers.

KW - Google Earth

KW - Green infrastructure

KW - High resolution

KW - Land use/land cover classification

KW - León

KW - Nicaragua

KW - True color satellite image

KW - Urban ecology

KW - Urban planning

UR - http://www.scopus.com/inward/record.url?scp=85073591218&partnerID=8YFLogxK

U2 - 10.25534/tuprints-00009656

DO - 10.25534/tuprints-00009656

M3 - Article

VL - 11

JO - Sustainability

JF - Sustainability

SN - 2071-1050

IS - 19

M1 - 5266

ER -