2021-09-30, 08:00–08:30, Academic
Socio-economic and demographic data is often released at the level of census administrative units. However, there is often a need for data available at a higher spatial resolution. Dasymetric mapping is an approach that has, in recent decades, increasingly received attention in order to exploit socio-economic data for spatially detailed analysis and/or to explore spatial phenomena that do not follow existing administrative units (the modifiable areal unit problem). This approach can be used to disaggregate such data into finer levels of detail, relying on the assumption that proxies available at a higher spatial resolution, typically as land-cover (LC) / land-use (LU) maps, can be used to produce weights in order to spatially reallocate the data to a finer scale layer.
For a long time, these weights have been subjectively determined based on expert knowledge, where higher weights are attributed to urban areas, slightly lower weights to suburban or rural areas, and a weight of zero for forest areas or water bodies. Recent research, however, has shifted this paradigm by taking advantage of the power and the efficiency of machine learning (ML) algorithms to create weighting layers for dasymetric mapping without any a priori knowledge.
This session will present a tool that has been developed to provide non-programmer users with a ready-to-use tool to create weighting layers using ML for dasymetric mapping. This tool is implemented as a GRASS GIS add-on and will be accessible on the official repository. It will facilitate the implementation of a machine learning-based approach to produce weighting layers for dasymetric mapping of socio-economic variables. We will demonstrate its ability to perform dasymetric mapping of population data based on earth observation (EO) derived products such as LC and LU maps.
Socio-economic and demographic data is usually collected at the individual or household level, and numbers are then aggregated and released at the level of administrative units. The spatial extent of many phenomena, however, do not correspond to any existing administrative limits, making them difficult to exploit. Additionally, geospatial information has started to be available at more and more detailed spatial resolutions, thanks to progress made using high-resolution EO data. Consequently, scientists often aim to perform spatial analyses at a fine resolution, but face issues related to the fact that the spatial resolution of administrative units, on which socio-economic and demographic data are aggregated, is too coarse and does not fit their needs. Dasymetric mapping can be used to create a more meaningful gridded layer of disaggregated socio-economic data, but the major challenge resides in determining the spatial distribution of a variable within aggregated spatial units.
The dasymetric mapping approach has been made more accessible with an existing GRASS GIS addon “v.area.weigh" (Metz, Grass Development Team, 2013), available on the official GRASS GIS add-on repository. It provides a tool for dasymetric mapping, however requires that the user provide their own weighted layer. Grippa et al. (2019) published a replicable approach that implements the random forest algorithm for the creation of a weighting layer for dasymetric mapping with the related computer code. While this code allows replicating the method, it is very specific to the experiments presented in the paper and may not fit the needs of other scientists. Moreover, since it is computer code, potential users not skilled enough in Python and R programming could be reluctant to use it.
An important step of the approach has already been implemented in a GRASS GIS add-on, “r.zonal.classes” (Grippa, Grass Development Team, 2019), which consists of the zonal extraction of class proportions from categorical raster data. The tool presented today completes the implementation of this approach, in a more generic and user-friendly manner. To our knowledge, there is no other existing open-source and ready-to-use tool, with a Graphical User Interface (GUI) for creation of dasymetric mapping weighting layers, using a ML approach.
Metz, M., GRASS Development Team, 2013. Addon v.area.weigh, Geographic Resources Analysis Support System (GRASS) Software, Version 7.8, Open Source Geospatial Foundation. https://grass.osgeo.org/grass78/manuals/addons/v.area.weigh.html (7 June 2021)
Grippa, T., Linard, C., Lennert, M., Georganos, S., Mboga, N., Vanhuysse, S., Gadiaga, A., Wolff, E. 2019. Improving urban population distribution models with very-high resolution satellite information, Data, 4(1), 1–17. doi.org/10.3390/data4010013.
Grippa, T., GRASS Development Team, 2019. Addon r.zonal.classes. Geographic Resources Analysis Support System (GRASS) Software, Version 7.8. Open Source Geospatial Foundation. https://grass.osgeo.org/grass78/manuals/addons/r.zonal.classes.html (7 June 2021)
Flasse, Charlotte (1)
Grippa, Taïs (1)
Fennia, Safa (1)
(1) Université Libre de Bruxelles, BelgiumTrack –
FOSS4G implementations in strategic application domains: land management, crisis/disaster response, smart cities, population mapping, climate change, ocean and marine monitoring, etc.Level –
2 - Basic. General basic knowledge is required.Language of the Presentation –
Charlotte is a researcher with a background in geography, specialising in remote sensing and GIS.