Abstract
Modellers simulating the state of the physical and biological environment using agents and fields, face the challenge of storing the state variables, like the distribution of biomass and the location and properties of animals. These variables differ from each other, depending on how exactly they represent their state. In this paper we describe an approach for storing multiple kinds of representations of model state variables using a unified physical data model. We identified a set of aspects in which these representations differ from each other and which have an implication for the data model. Based on these aspects we defined a physical data model for storing spatio-temporal objects, which we implemented in an open source software library. We illustrate how the resulting data model can be used to store multiple kinds of model state variables and explain how this data model can be useful in environmental modelling software.
Original language | English |
---|---|
Article number | 104553 |
Journal | Environmental Modelling and Software |
Volume | 122 |
DOIs | |
Publication status | Published - 1 Dec 2019 |
Funding
This work was supported by the Research IT innovation programme (Utrecht University, The Netherlands) and the Global Geo Health Data Center (Utrecht University and University Medical Center Utrecht, The Netherlands) .
Keywords
- Field
- HDF5
- LUE
- Modelling
- Object
- Simulation