Redistribution population data across a regular spatial grid according to buildings characteristics

Abstract

Population data are generally provided by state census organisations at the predefi ned census enumeration units. However, these datasets very are often required at userdefi ned spatial units that differ from the census output levels. A number of population estimation techniques have been developed to address these problems. This article is one of those attempts aimed at improving county level population estimates by using spatial disaggregation models with support of buildings characteristic, derived from national topographic database, and average area of a fl at. The experimental gridded population surface was created for Opatów county, sparsely populated rural region located in Central Poland. The method relies on geolocation of population counts in buildings, taking into account the building volume and structural building type and then aggregation the people total in 1 km quadrilateral grid. The overall quality of population distribution surface expressed by the mean of RMSE equals 9 persons, and the MAE equals 0.01. We also discovered that nearly 20% of total county area is unpopulated and 80% of people lived on 33% of the county territory.

Multi-Criteria Analysis for Solar Farm Location Suitability

“…The main source of spatial data for the study was BDOT10k (Database of Topographic Objects), provided by the Central Geodetic and Cartographic Documentation Centre. BDOT10k is a country wide topographic data with the level of detail and thematic scope corresponding to the civilian maps at a 1:10000 scale (Bielecka, 2015;Calka et al, 2016). The geometric accuracy of object size estimated by Ławniczak and Kubiak (2016) is about 2% different than field measurement.…”

Analysis of land use and land cover maps suitability for estimating population density in urban areas - exclusion of the equifinality phenomenon on population maps developed from CLC, UA and OBIA data

“…Another approach validating the method is the cancellation of recalculated population maps, after their re-aggregation, to the original census units. Detection of differences at this level is the basis for the evaluation of the method -for example, Tomás et al (2016), Całka et al (2016). The lists presented by the authors indicate average errors in units below 10%.…”

Analysis of the usefulness land use and land cover maps for estimating the population of urban areas – valorisation of multi-variant population maps based on the GUS kilometre network

“…The first, basic problem is to find a correct way to convert the population from census units to units related to buildings (so-called limiting variables), including the diversity of buildings (so-called variable relationships) resulting from their nature (density, height, location, function, etc.). A more extensive overview of the conversion methods used can be found, for example, in works by Eicher and Brewer (2001), Wu et al (2005), Wang and Wu (2010), Całka et al (2016).…”

“…The second problem is the correct determination of limiting variables and variable relationships. The implementation at the national level of detailed databases on buildings and monitoring of the coverage of overland areas (Corine Land Cover, Urban Atlas) were quickly reflected in population mapping (Gallego and Peedell, 2001;Bielecka, 2005;Kunze and Hecht, 2015;Cockx and Canters 2015;Batista and Poelman, 2016;Całka et al, 2016;Pirowski and Pomietłowska, 2017;Pirowski and Bartos, 2018). These databases are in many cases insufficient -they do not cover a given area, they have a limited time range and too low spatial accuracy.…”