Models
A model is more than the software it runs on. Four things have to be there, and only the first of them is code.
Scripts. The GeoDMS configuration states what every result is: which sources it comes from, how it is derived, and under which assumptions. It is plain text, so it can be versioned, compared and reviewed like any other code, and it is the model's real definition rather than a description of one.
Data. Assembling, validating and harmonising the sources is usually the larger part of the work. National registers have to be made usable for spatial modelling; domain data has to be collected from many places and checked against local knowledge.
Documentation. A result that cannot be traced is not much use in a policy discussion. Each model carries its own documentation, so that the assumptions and the limitations can be read, checked and argued with by someone who did not build it.
Partners. The domain knowledge rarely sits in one organisation. Most of these models are developed and maintained together with the agency, institute or company that owns the question, and several of them are theirs rather than ours.
The models
| Model | What it does | Documentation |
|---|---|---|
| RSopen | Open-source RuimteScanner 2.0: land use, actors and objects for the Netherlands | geodms.nl/rsopen |
| CRISP | Global downscaling of population and built-up area on a 1 km grid, behind the UN urbanization outlook | geodms.nl/crisp |
| NetworkModel PBL | Car, bicycle, walking and public transport accessibility for the Netherlands | geodms.nl/networkmodel_pbl |
| NetworkModel World | A routable car network for the entire planet, with catchments and functional rural areas | Wiki |
| NetworkModel EU | The European equivalent, for the Joint Research Centre | GitHub |
| LUISA / EU-ClueScanner | Land-use component of the Commission's territorial assessment platform | JRC LUISA |
| Vesta | Energy demand and supply matching in the built environment | github.com/pbl-nl |
| Land Use Scanner Germany | Land-use allocation for Germany, with the BBR | GitHub |
| IGOR | Individuals Grouped Over Ratios: splits gridded population totals into five-year sex and age groups | Wiki |
What these models are used for, per client and with the year, is in the project list. The domains they serve are described under what we do, and the environment they are built in under software.