Cities in are growing and increasing in numbers at extraordinary rates facing a number of challenges related to their sustainable development. The urban population in developing countries is projected to increase from 2 billion in 2000 to 5.5 billion by 2050. It is estimated that with the population increase, especially in developing countries, their land use and corresponding ground sealing levels with triple.
While the urbanization advances there is an overwhelming need to further collect data to better understand the extent and the rate of change in urban land cover on the global and regional scale, as well as patterns of urban land use on the level of individual cities.
Earth Observation-based mapping represents efficient method to deliver high quality harmonized and up-to-date information on land use and land cover for the assessment and analysis of spatial, and temporal patterns of urban growth, and with a retrospective mapping potential. It offers a sound basis for statistical calculations, while the collection of measurements in a harmonized and standardized manner allows globally consistent comparisons.
Collaboration with the European Space Agency
Within the framework of the South Asia Megacities Improvement Program, European Space Agency supported the World Bank in mapping of the 20 years of urban expansion in the metropolitan areas of Delhi (the Delhi National Capital Region, southeast part), Mumbai (the Metropolitan Region of Mumbai) and Dhaka (the Statistical Metropolitan Area). The analysis of high-resolution satellite data provided spatially-explicit information about current state and previous development of land use and land cover resources related to urbanization, including:
Structure of the land cover and land use stocks
The analysis conducted for 2010/11 revealed that the land cover and land use composition in three megacities is quite distinct. In Delhi National Capital Region (southeast part) the land use is dominated by agriculture (73%). Urbanized area represent there only 20% of land cover. In the Metropolitan Region of Mumbai only 15% area is urbanized, the share of non-urbanized area is balanced between the agriculture land (34%), (semi-)natural vegetation (28%) and forest (17%). In Dhaka Statistical Metropolitan Area urbanized area stand for 48%, but large part of that is occupied by rural settlements and scattered built-up (18%) surrounded by specific class of groves of trees. Non-urbanized area is dominated by agriculture land (43%).
Figure1: Example of the land use maps developed for the year 2010/2011 and statistical information that can be derived from spatial data analysis. Credit: Eoworld project, GISAT for European Space Agency/World Bank
Structure and intensity of change
Satellite-based monitoring is an extremely effective method of identifying the rate, location and the characteristics of urban sprawl which takes place outside of administrative boundaries of cities as well as other types of urban expansion. As a monitoring method it can offset any missing or inaccurate information that urban planners, city administrators and other public stakeholders may have available with regard to the spontaneous growth that happens outside the city official boundaries, hence outside of control or the supervision of specific authorities. The ability to collect this type of information is particularly important in view of the lack of the consensus on the borders of the 100 largest urban areas. In case of the three South Asia megacities, with the new generation of EO datasets it was possible to assess the structure of a massive urban sprawl taking place since 1991/1992, including the changes between residential areas, commercial and industrial zones. The study found that urban sprawl in Dhaka is clearly dominated by residential built-up, while in Delhi and Mumbai, the urban sprawl is accelerated by industrial development (in 2003-2010 period predominantly for Delhi). Large increase of construction sites also indicates that this trend will presumably continue in the future.
Figure2: Urban sprawl formation in Dhaka, Delhi and Mumbai. Credit: Eoworld project/GISAT for European Space Agency/World Bank
Impact on the surrounding environment
Urbanization trends clearly change the functions of the surrounding environment which is giving in to the new constructions projects either for new urban fabric residential quarters or new industrial and commercial facilities or transport infrastructure. If occurring in uncontrolled manner, rising percentages of the ground sealing levels may lead to a number of challenges to sustainable urban development such as uptake of the arable land, reduction in soil permeability and drainage capacity, among others. For example, while agriculture land is taken almost exclusively in Delhi area (>95%), in Mumbai, large share of land turned to urban is represented by (semi-)natural land (~38%). In Dhaka, land consumed by urban sprawl is dominated by rural settlements. The latter is a sign of both gradual densification of rural settlements leading to disappearance of surrounding high vegetation and small gardens.
Figure 3: Urban sprawl consumption in Dhaka, Delhi and Mumbai. Credit: Eoworld project/GISAT for European Space Agency/World Bank
Working with data: Better Visualization, Comparison and Analysis
In addition to the core mapping outputs, the project provided the state-of-the-art data exploitation Web Tool which facilitates spatial data integration with other in-situ datasets to support collaborative work, data analysis, and the use of data in reporting. The Web Tool can be accessed at the dedicated Urban Portal.
The advantage of working with digitalized spatial data lays in the possibility to analyze datasets at the level of different administrative units: metropolitan level, city level, sub-district level, as well as other city catchment (non-administrative) units depending on information needs. Such datasets allow flexible aggregation, for instance they can provide information concerning the ratio of population growth correlated with urban fabric growth rate, the proportion of sprawl by district, the distribution and density of urban sprawl, urban area classes evolution, the drivers of urban change, and so on. In combination with available environmental or socio-economic data it is further possible to measure different city indicators such as compactness (as a function of city density), or the ratio of green urban areas per inhabitant, or the proximity and accessibility of green urban areas, among others.