Testing the Effects of Thematic Uncertainty on Spatial Decision-making

Dasymetric mapping techniques can be employed to estimate population characteristics of small areas that do not correspond to census enumeration areas. Land cover has been the most widely used source of ancillary data in dasymetric mapping. The current research examines the performance of alternative sources of ancillary data, including imperviousness, road networks, and nighttime lights. Nationally available datasets were used in the analysis to allow for replicability. The performance of the techniques used to examine these sources was compared to areal weighting and traditional land cover techniques. Four states were used in the analysis, representing a range of different geographic regions: Connecticut, New Mexico, Oregon, and South Carolina. Ancillary data sources were used to estimate census block group population counts using census tracts as source zones, and the results were compared to the known block group population counts. Results indicate that the performance of dasymetric methods varies substantially among study areas, and no single technique consistently outperforms all others. The three best techniques are imperviousness with values greater than 75 percent removed, imperviousness with values greater than 60 percent removed, and land cover. Total imperviousness and roads perform slightly worse, with nighttime lights performing the worst compared to all other ancillary data types. All techniques performed better than areal weighting.     

Data-enriched interpolation for temporally consistent population compositions

This research evaluates the performance of areal interpolation coupled with dasymetric refinement to estimate different demographic attributes, namely population sub-groups based on race, age structure and urban residence, within consistent census tract boundaries from 1990 to 2010 in Massachusetts. The creation of such consistent estimates facilitates the study of the nuanced micro-scale evolution of different aspects of population, which is impossible using temporally incompatible small-area census geographies from different points in time. Various unexplored ancillary variables, including the Global Human Settlement Layer (GHSL), the National Land-Cover Database (NLCD), parcels, building footprints and the proprietary ZTRAX® dataset are utilized for dasymetric refinement prior to areal interpolation to examine their effectiveness in improving the accuracy of multi-temporal population estimates. Different areal interpolation methods including Areal Weighting (AW), Target Density Weighting (TDW), Expectation Maximization (EM) and its data-extended approach are coupled with different dasymetric refinement scenarios based on these ancillary variables. The resulting consistent small area estimates of white and black subpopulations, people of age 18–65 and urban population show that dasymetrically refined areal interpolation is particularly effective when the analysis spans a longer time period (1990–2010 instead of 2000–2010) and the enumerated population is sufficiently large (e.g., counts of white vs. black). The results also demonstrate that current census-defined urban areas overestimate the spatial distribution of urban population and dasymetrically refined areal interpolation improves estimates of urban population. Refined TDW using building footprints or the ZTRAX® dataset outperforms all other methods. The implementation of areal interpolation enriched by dasymetric refinement represents a promising strategy to create more reliable multi-temporal and consistent estimates of different population subgroups and thus demographic compositions. This methodological foundation has the potential to advance micro-scale modeling of various subpopulations, particularly urban population to inform studies of urbanization and population change over time as well as future population projections.     

Comparisons of two global built area land cover datasets in methods to disaggregate human population in eleven countries from the global South

ABSTRACT

Mapping built land cover at unprecedented detail has been facilitated by increasing availability of global high-resolution imagery and image processing methods. These advances in urban feature extraction and built-area detection can refine the mapping of human population densities, especially in lower income countries where rapid urbanization and changing population is accompanied by frequently out-of-date or inaccurate census data. However, in these contexts it is unclear how best to use built-area data to disaggregate areal, count-based census data. Here we tested two methods using remotely sensed, built-area land cover data to disaggregate population data. These included simple, areal weighting and more complex statistical models with other ancillary information. Outcomes were assessed across eleven countries, representing different world regions varying in population densities, types of built infrastructure, and environmental characteristics. We found that for seven of 11 countries a Random Forest-based, machine learning approach outperforms simple, binary dasymetric disaggregation into remotely-sensed built areas. For these more complex models there was little evidence to support using any single built land cover input over the rest, and in most cases using more than one built-area data product resulted in higher predictive capacity. We discuss these results and implications for future population modeling approaches.     

Refining Population Surface Models: Experiments with Northern Ireland Census Data

This paper compares and contrasts alternative methods for the construction of discontinuous population surface models based on the census and remotely sensed data from Northern Ireland. Two main methods of population distribution are employed: (1) a method based on redistribution from enumeration district (ED) and postcode centroids, and (2) a method based on dasymetric redistribution of ED population counts to suitable land cover zones from classified remotely sensed imagery. Refinements have been made to the centroid redistribution algorithm to accommodate an empirical measure of dispersion, and to allow redistribution in an anisotropic form. These refinements are evaluated against each other and the dasymetric method. The results suggest that all of the methods perform best in urban areas, and that while the refinements may improve the statistical performance of the models, this is at the expense of reduced spatial detail. In general, the techniques are highly sensitive to the spatial and population resolution of the input data.     

Mapping Population Distribution in the Urban Environment: The Cadastral-based Expert Dasymetric System (CEDS)

This paper discusses the importance of determining an accurate depiction of total population and specific sub-population distribution for urban areas in order to develop an improved "denominator," which would enable the calculation of more correct rates in GIS analyses involving public health, crime, and urban environmental planning. Rather than using data aggregated by arbitrary administrative boundaries such as census tracts, we use dasymetric mapping, an areal interpolation method using ancillary information to delineate areas of homogeneous values. We review previous dasymetric mapping techniques (which often use remotely sensed land-cover data) and contrast them with our technique, Cadastral-based Expert Dasymetric System (CEDS), which is particularly suitable for urban areas. The CEDS method uses specific cadastral data, land-use filters, modeling by expert system routines, and validation against various census enumeration units and other data. The CEDS dasymetric mapping technique is presented through a case study of asthma hospitalizations in the Bronx, New York City, in relation to proximity buffers constructed around major sources of air pollution. The case study shows the impact that a more accurate estimation of population distribution has on a current environmental justice and health disparities research project, and the potential of CEDS for other GIS applications.     

Using Semi-variance Image Texture Statistics to Model Population Densities

This study presents a method to model population densities by using image texture statistics of semi-variance. In a case study of the City of Austin, Texas, we first selected sample census blocks of the same land use to build population models by land use. Regression analyses were conducted to infer the relationship between block population densities and image texture statistics of the semi-variance. We then applied the population models to an area of 251 blocks to estimate populations for within-blocks land-use areas while maintaining census block populations. To assess the proposed method, the same analysis was performed while census block-group populations were maintained, and the aggregated block populations were compared with original census block populations. We also tested a conventional land-use-based dasymetric mapping method with pre-calculated population densities for land uses. The results show that our approach, which is based on initial land-use stratification and further image-texture statistical modeling of population, has higher accuracy statistics than the conventional land-use-based dasymetric mapping method.     

Dasymetric Estimation of Population Density and Areal Interpolation of Census Data

This paper describes techniques to compute and map dasymetric population densities and to areally interpolate census data using dasymetrically derived population weights. These techniques are demonstrated with 1980-2000 census data from the 13-county Atlanta metropolitan area. Land-use/land-cover data derived from remotely sensed satellite imagery were used to determine the areal extent of populated areas, which in turn served as the denominator for dasymetric population density computations at the census tract level. The dasymetric method accounts for the spatial distribution of population within administrative areas, yielding more precise population density estimates than the choroplethic method, while graphically representing the geographic distribution of populations. In order to areally interpolate census data from one set of census tract boundaries to another, the percentages of populated areas affected by boundary changes in each affected tract were used as adjustment weights for census data at the census tract level, where census tract boundary shifts made temporal data comparisons difficult. This method of areal interpolation made it possible to represent three years of census data (1980, 1990, and 2000) in one set of common census tracts (1990). Accuracy assessment of the dasymetrically derived adjustment weights indicated a satisfactory level of accuracy. Dasymetrically derived areal interpolation weights can be applied to any type of geographic boundary re-aggregation, such as from census tracts to zip code tabulation areas, from census tracts to local school districts, from zip code areas to telephone exchange prefix areas, and for electoral redistricting.     

A Comparative Analysis of Areal Interpolation Methods

Over the years many approaches to areal interpolation have been developed and utilized. They range from the simple 2-D areal weighing method which uses only the spatial Z variable being processed, to more sophisticated approaches which use auxiliary variable(s) to provide more specificity to the results. In the research reported here, four promising approaches are implemented and comparatively tested. These approaches have widely varying conceptual foundations, and different auxiliary variables, if used. The areal weighing reflects many earlier methods which assumes uniform distributions of the spatial Z variable, and does not use any auxiliary variable. Tobler's pycnophylactic method uses a volumetric preservation approach, which assumes spatial Z variable is heterogeneously distributed, but does not use any auxiliary variable. The traditional dasymetric method of Wright is used with remote sensing spectral data of land use. Xie's road network hierarchically weighted interpolation uses the road network as the auxiliary variable, and assumes that population density is related to the class of the road, and to the density of the road network. The research design implemented here uses Census population distributions at different levels in the hierarchy as the source and target variables analyzed. The source zone population is taken at the Census Tract level, and the target zones are specified at the Census Block Group level in the hierarchy. The first two tests use only the Census population Z data, and no auxiliary variables, whereas the next uses remotely sensed land use data as the auxiliary data variable, and the fourth test utilizes the road network hierarchy as the auxiliary variable. The paper reports on the findings from these tests, and then interprets them in a spatial setting in terms of accuracy and effectiveness. This research points to the network method as the most accurate of the areal interpolation methods tested in this research.     

Comparative assessment of CORINE2000 and GLC2000: Spatial analysis of land cover data for Europe

Given the current lack of interoperability between global and regional land cover products, efforts are underway to link the new European global land cover map (GLOBCOVER) with the existing global land cover 2000 map (GLC2000) and European CORINE mapping initiative. Since both datasets apply different mapping standards, key for a successful implementation is a thorough understanding of the heterogeneities among both datasets. Thus, this paper provides an assessment of compatibilities and differences between the CORINE2000 and GLC2000 datasets. The comparative assessment considers inconsistencies between the thematic legends (using the UN land cover classification system-LCCS), class specific accuracies, and the spatial resolution and heterogeneity of the datasets. The results are summarized with implications for the development of the new GLOBCOVER datasets.     

Instructions for Contributors to Cartography and Geographic Information Science

A number of areal interpolation methods have been developed to estimate population for overlapping, discontinuous, or fragmented areas. Previous studies examined the relative accuracy of various methods; this research advances those endeavors by comparing the effectiveness of seven different methods using a national random sample of census block groups and blocks. As the results show, the areal interpolation methods produce good population estimates for nested census blocks except in areas of heterogeneous land use or unusual contexts. In addition, estimation conducted in areas with small populations or low population density was vulnerable to high percentage error. Amongst the different methods, road network allocation and statistical regression (with area and roads as predictors) produced the best population estimates for the sample blocks.     

Classifiers vs. input variables—The drivers in image classification for land cover mapping

The study investigates the performance of image classifiers for landscape-scale land cover mapping and the relevance of ancillary data for the classification success in order to assess and to quantify the importance of these components in image classification. Specifically tested are the performance of maximum likelihood classification (MLC), artificial neural networks (ANN) and discriminant analysis (DA) based on Landsat7 ETM+ spectral data in combination with topographic measures and NDVI. ANN produced high accuracies of more than 75% also with limited input information, while MLC and DA produced comparable results only by incorporating ancillary data into the classification process. The superiority of ANN classification was less pronounced on the level of the single land cover classes. The use of ancillary data generally increased classification accuracy and showed a similar potential for increasing classification accuracy than the selection of the classifier. Therefore, a stronger focus on the development of appropriate and optimised sets of input variables is suggested. Also the definition and selection of land cover classes has shown to be crucial and not to be simply adaptable from existing land cover class schemes. A stronger research focus towards discriminating land cover classes by their typical spectral, topographic or seasonal properties is therefore suggested to advance image classification.     

Exploring nationally and regionally defined models for large area population mapping

Interactions between humans, diseases, and the environment take place across a range of temporal and spatial scales, making accurate, contemporary data on human population distributions critical for a variety of disciplines. Methods for disaggregating census data to finer-scale, gridded population density estimates continue to be refined as computational power increases and more detailed census, input, and validation datasets become available. However, the availability of spatially detailed census data still varies widely by country. In this study, we develop quantitative guidelines for choosing regionally-parameterized census count disaggregation models over country-specific models. We examine underlying methodological considerations for improving gridded population datasets for countries with coarser scale census data by investigating regional versus country-specific models used to estimate density surfaces for redistributing census counts. Consideration is given to the spatial resolution of input census data using examples from East Africa and Southeast Asia. Results suggest that for many countries more accurate population maps can be produced by using regionally-parameterized models where more spatially refined data exists than that which is available for the focal country. This study highlights the advancement of statistical toolsets and considerations for underlying data used in generating widely used gridded population data.     

Urban land cover thematic disaggregation,employing datasets from multiple sources and RandomForests modeling

Urban land cover mapping has lately attracted a vast amount of attention as it closely relates to a broad scope of scientific and management applications. Late methodological and technological advancements facilitate the development of datasets with improved accuracy. However, thematic resolution of urban land cover has received much less attention so far, a fact that hampers the produced datasets utility. This paper seeks to provide insights towards the improvement of thematic resolution of urban land cover classification. We integrate existing, readily available and with acceptable accuracies datasets from multiple sources, with remote sensing techniques. The study site is Greece and the urban land cover is classified nationwide into five classes, using the RandomForests algorithm. Results allowed us to quantify, for the first time with a good accuracy, the proportion that is occupied by each different urban land cover class. The total area covered by urban land cover is 2280 km² (1.76% of total terrestrial area), the dominant class is discontinuous dense urban fabric (50.71% of urban land cover) and the least occurring class is discontinuous very low density urban fabric (2.06% of urban land cover).     

Using control data to determine the reliability of volunteered geographic information about land cover

There is much interest in using volunteered geographic information (VGI) in formal scientific analyses. This analysis uses VGI describing land cover that was captured using a web-based interface, linked to Google Earth. A number of control points, for which the land cover had been determined by experts allowed measures of the reliability of each volunteer in relation to each land cover class to be calculated. Geographically weighted kernels were used to estimate surfaces of volunteered land cover information accuracy and then to develop spatially distributed correspondences between the volunteer land cover class and land cover from 3 contemporary global datasets (GLC-2000, GlobCover and MODIS v.5). Specifically, a geographically weighted approach calculated local confusion matrices (correspondences) at each location in a central African study area and generated spatial distributions of user's, producer's, portmanteau, and partial portmanteau accuracies. These were used to evaluate the global datasets and to infer which of them was 'best’ at describing Tree cover at each location in the study area. The resulting maps show where specific global datasets are recommended for analyses requiring Tree cover information. The methods presented in this research suggest that some of the concerns about the quality of VGI can be addressed through careful data collection, the use of control points to evaluate volunteer performance and spatially explicit analyses. A research agenda for the use and analysis of VGI about land cover is outlined.     

Building population mapping with aerial imagery and GIS data

Geospatial distribution of population at a scale of individual buildings is needed for analysis of people's interaction with their local socio-economic and physical environments. High resolution aerial images are capable of capturing urban complexities and considered as a potential source for mapping urban features at this fine scale. This paper studies population mapping for individual buildings by using aerial imagery and other geographic data. Building footprints and heights are first determined from aerial images, digital terrain and surface models. City zoning maps allow the classification of the buildings as residential and non-residential. The use of additional ancillary geographic data further filters residential utility buildings out of the residential area and identifies houses and apartments. In the final step, census block population, which is publicly available from the U.S. Census, is disaggregated and mapped to individual residential buildings. This paper proposes a modified building population mapping model that takes into account the effects of different types of residential buildings. Detailed steps are described that lead to the identification of residential buildings from imagery and other GIS data layers. Estimated building populations are evaluated per census block with reference to the known census records. This paper presents and evaluates the results of building population mapping in areas of West Lafayette, Lafayette, and Wea Township, all in the state of Indiana, USA.     

Comparison and relative quality assessment of the GLC2000, GLOBCOVER,MODIS and ECOCLIMAP land cover data sets at the African continental scale

Land cover dynamics at the African continental scale is of great importance for global change studies. Actually, four satellite-derived land cover maps of Africa now available, e.g. ECOCLIMAP, GLC2000, MODIS and GLOBCOVER, are based on images acquired in the 2000s. This study aims at stressing the compliances and the discrepancies between these four land cover classifications systems. Each of them used different mapping initiatives and relies on different mapping standards, which supports the present investigation. In order to do a relative comparison of the four maps, a preamble was to reconcile their thematic legends into more aggregated categories after a projection into the same spatial resolution. Results show that the agreement between the four land cover products is between 56 and 69%. While all these land cover datasets show a reasonable agreement in terms of surface types and spatial distribution patterns, mapping of heterogeneous landscapes in the four products is not very successful. Land cover products based on remote sensing imagery can indeed significantly be improved by using smarter algorithms, better timing of image acquisition, improved class definitions. Either will help to improve the accuracy of future land cover maps at the African continental scale. Data producers may use the areas of spatial agreement for training area selection while users might need to verify the information in the areas of disagreement using additional data sources.     

Improving Large Area Population Mapping Using Geotweet Densities

Many different methods are used to disaggregate census data and predict population densities to construct finer scale, gridded population data sets. These methods often involve a range of high resolution geospatial covariate datasets on aspects such as urban areas, infrastructure, land cover and topography; such covariates, however, are not directly indicative of the presence of people. Here we tested the potential of geo‐located tweets from the social media application, Twitter, as a covariate in the production of population maps. The density of geo‐located tweets in 1x1 km grid cells over a 2‐month period across Indonesia, a country with one of the highest Twitter usage rates in the world, was input as a covariate into a previously published random forests‐based census disaggregation method. Comparison of internal measures of accuracy and external assessments between models built with and without the geotweets showed that increases in population mapping accuracy could be obtained using the geotweet densities as a covariate layer. The work highlights the potential for such social media‐derived data in improving our understanding of population distributions and offers promise for more dynamic mapping with such data being continually produced and freely available.     

Modelling changing population distributions: an example of the Kenyan Coast, 1979–2009

ABSTRACT

Large-scale gridded population datasets are usually produced for the year of input census data using a top-down approach and projected backward and forward in time using national growth rates. Such temporal projections do not include any subnational variation in population distribution trends and ignore changes in geographical covariates such as urban land cover changes. Improved predictions of population distribution changes over time require the use of a limited number of covariates that are time-invariant or temporally explicit. Here we make use of recently released multi-temporal high-resolution global settlement layers, historical census data and latest developments in population distribution modelling methods to reconstruct population distribution changes over 30 years across the Kenyan Coast. We explore the methodological challenges associated with the production of gridded population distribution time-series in data-scarce countries and show that trade-offs have to be found between spatial and temporal resolutions when selecting the best modelling approach. Strategies used to fill data gaps may vary according to the local context and the objective of the study. This work will hopefully serve as a benchmark for future developments of population distribution time-series that are increasingly required for population-at-risk estimations and spatial modelling in various fields.