In this demonstration project we develop a service solution, providing location-based information on the current and future distribution of the population by age group. It can be shown that information derived from EO data significantly improves existing census data through a modelling approach based on land cover and land use information. Thereby, AgeSpot links EO derived population density with state-of-the art demographic models. The unique approach not only provides spatially explicit demographic data along with information on health and income, but also allows long-term forecasting at different levels of granularity.
AgeSpot creates a new market and value network by offering a “one-stop-shop” for spatially explicit population and demographic data available for the past and the future. Customers do not need to arduously assemble and analyse different sources of external data, but AgeSpot enables them to access standardized key information with already integrated external data information and improve the existing data situation in many innovative ways.
Users and their needs
AgeSpot offers a unique service in providing a technical solution covering all sectors, including stakeholders such as from health and care, independent living, safety and security, social, leisure and education. AgeSpot enters markets and meets the information needs of both the private and public sectors in fields such as market research, business location analysis, risk assessment and urban and regional planning.
Private sector companies like, telecommunication and insurance companies, as well as companies offering consumer goods, can benefit strongly from the information provided by AgeSpot, helping them find out where to offer which specific services and to adapt their supply to the local demand depending on the demographic structures of different regions. For the public sector, AgeSpot data makes it much easier for governments to better meet the needs of regional and district populations. Furthermore, it is an important signpost for taking the right steps to promote local economies.
User’s needs provided by AgeSpot:
- Supporting decisions based on granular population data: AgeSpot provides exact information on where how many people belong to each target group at different levels of granularity.
- Customized service: AgeSpot offers a unique service in which the choice of products for every location is geared to the needs of the customers.
- Data driven decision making: AgeSpot allows to make cost-effective decisions on where to expand and from which markets to withdraw based on high-quality (future) projections of spatially explicit population information.
- Reliable projections of future demands: the AgeSpot service allows for investment decisions to be based on granular population data and realistic projections of future demand.
- Easy access to accurate and up-to-date spatial data on potential customers
An overview of the AgeSpot products and feature options are shown in Figure 1:
Figure 1: AgeSpot product range
Potential users belong to nearly all economic sectors operating world-wide on regional to global scales.
Service/ system concept
AgeSpot is tailored to the global demand for business intelligence data, a driving force for economic growth. It enables customers to easily access granular, standardized and up-to-date key information about their target groups providing a new information service for spatially locating and forecasting population demographics.
This information is derived through a system that implements the following key innovative technical components (Figure 2):
- Satellite Observation Segment: EO derived information enables a service that significantly improves existing census data through a modelling approach based on land use information.
- AgeSpot Engine:
- Population Disaggregation Model: This model provides statistical information about the distribution of the population in their residences. With the help of additional spatial data such as land use and degree of sealing of the surfaces, as well as statistical population data over a larger region, a precise, high resolution disaggregation of the population takes place.
- Demographic Model: In a Bayesian Model Averaging (BMA) approach a large number of spatial and demographic variables are identified to determine the population number per age cohort on up to a 50m grid level.
- Statistical forecasting model: Regressions are set up using the change in the share of the age group from one point in time to the next as dependent variable in the equation. The forecasts are refined by the use of an urban growth model. This method is applied to forecast the trends in urbanization and project, which parts of a city or country will emerge to urban built-up areas in the turn of the next years.
- Provision interface: Data provisioning interface used by both companies linking the production systems with the AgeSpot Customer Access.
Figure 2: High level illustration of the AgeSpot system architecture
Space Added Value
The baseline for this project is the Satellite Earth Observation (EO) component. The EO derived information significantly improves existing census data through a modelling approach based on land cover and land use information derived from EO satellite data (e.g. Sentinel). A basis for determining the spatial population distribution is formed by the generation of imperviousness densities. Satellite EO data can extract such land use information to a sufficient degree of accuracy to supply the reliable base information necessary for this service. The linking of the data with the statistical information on demography is sufficiently accurate thanks to the spatial population model developed by GeoVille and its continued improvement since first application. This EO modelling tool ingests high resolution EO data providing the basis for products on different granular levels (50m to 150m spatial resolution).
Figure 3: EO data (LS 8 2015; left) and Imperviousness densities (right) derived therefrom for part of the Greater London Area, UK.
The Demonstration Project kicked off in February 2019. The initial phase of the project has been focusing on the engagement with the various champion users with the goal of determining their needs and specific user requirements. Resulting from these, the system requirements on AgeSpot have been defined and form the basis for designing and setting up the system and service architecture. In parallel to the user engagement activities, Earth Observation and census data have been acquired and technological developments have taken place. These comprise improvements on the population, demographic and forecasting models, including the new urban growth model component, showing future urban expansion. Modelling activities have begun on the Greater London Area, which constitutes the first AOI for demonstration purposes in the project (Figure 4).
Figure 4: Spatial distribution of the population in the Greater London area for 2015 (Precision Product: 50m resolution).