Consistent yet adaptive global geospatial identification of urban–rural patterns: The iURBAN model

Abstract

The main motivation of this paper is to shed new light on the problem of spatial identification of urban and rural areas globally, and to provide a compatible disaggregation framework for linking associated country-specific, non-spatial data compilations, such as building type inventories. Existing homogeneously setup global urban extent models commonly ignore local-level specifics. While global consistency and regional comparability of urban characteristics are much strived-for goals in the global development and remote sensing communities, non-conformity at the national level often renders such models inapplicable for effective decision-making. Furthermore, the focus on identifying 'urban' leads to an illdefined 'rural', which is simply defined by contrast as 'everything else'; a questionable definition when referring to strongly spatially localized residential patterns. In this paper we introduce the novel iURBAN geospatial modeling approach, identifying Urban-Rural patterns in Built-up-Adjusted and Nationally-adaptive manner. The model operates at global scale, but at the same time conforms to country specifics. In this model, high-resolution, satellitederived, built-up data is used to consistently detect global human settlements at unprecedented spatial detail. In combination with global gridded population data, and with reference to national level statistical information on urban population ratios globally compiled in the annually-released UN World Urbanization Prospects, iURBAN identifies matching urban extents. Additionally, a novel reallocation algorithm is introduced which addresses the poor representation of rural areas that is inherent in existing global population grids. Associating all of the population with inhabitable, built-up area and conforming to national urban-rural ratios, iURBAN sets a new standard by enabling careful consideration of both urban and rural as opposed to traditional urban-biased approaches.

Mapping Urban Extent at Large Spatial Scales Using Machine Learning Methods with VIIRS Nighttime Light and MODIS Daytime NDVI Data

“…The definition of an urban area varies from different research perspectives [33,58,59]. For example, census-related urban studies refer mainly to population distributions while those using nighttime lights or multi-spectral data may be related to economic conditions or "built-up areas" (physical attributes of land surface) [9,10,23,31,57,60,61]. As the characteristics of all these definitions are correlated, most of the urban areas identified by relevant methods are consistent.…”

“…Studies have shown that the global urbanization rate hit about 54% in 2014 [1], and urban land areas will grow 1.2 million km 2 by 2030 if the current trend continues [8]. To better understand the patterns, dynamics, drivers, and impacts of urban expansion and effectively support decision makings regarding sustainable urban development, it is fundamental to obtain timely, accurate, and consistent measurements of urban extent at large spatial scales [9][10][11].…”

Exposure and Vulnerability Geospatial Analysis Using Earth Observation Data in the City of Liege, Belgium

“…Firstly, remote sensing information is used for mapping danger occurrence (Nayak and Zlanatova, 2008) and specifically danger of flood (Chormanski et al, 2011), earthquake (Tralli et al, 2005), air pollution (Martin, 2008;Prud'homme et al, 2013) or urban heat . Secondly, assets and built-up area are estimated according to various urbanization definition and settlements maps using EO data (Aubrecht et al, 2016, Pesaresi et al, 2016. The link to population and then risk exposure is established by dasymetric mapping methods integrating Land Use (LU) data derived from EO data.…”

“…Urban or built-up areas are then an element at risk for specific infrastructure and assets but are also considered as a proxy of the presence of people. There are several approaches addressing spatial delineation of urban areas but remote sensing data and derived products are an important category (Aubrecht et al, 2016). Satellite-observed night-time lights data was the first potential proxy in the early 1970s (Sutton, 1997).…”

Central America Urbanization Review

“…This is particularly true for urban areas where most of the exposure value is located. CDRP results at the country level quantifying earthquake and hurricane risk, and differentiating them between urban and rural areas (Aubrecht et al 2016) show that catastrophic risk is highly concentrated in urban areas in five out of the six countries (see figure 4.4). Map 4.2 provides an example of the concentration of earthquake risk in Managua represented by the annual average loss (AAL).…”