Most regional model codes cover only part of the processes within a region. For coming to grips with many issues of integrated water management it is necessary to have a model that covers the whole (regional) system, including plant-atmosphere interactions, soil water, groundwater and surface water. SIMGRO (a dated acronym of SIMulation of GROundwater) was developed for that purpose.
The name ‘SIMGRO’ was formerly used for referring to an integrated model code, including submodels for the compartments and processes as shown in Fig. 1. Now it is used in the meaning of a modelling framework. This framework has been connected to a number of ‘inhouse’ components, but also has possibilities for coupling to other codes. The in-house components are:
1. a SVAT-model that is commonly referred to as ‘MetaSWAP’, covering the plantatmosphere interactions and soil water;
2. a simplified surface water metamodel;
3. a drainage package, for simulating groundwater drainage with fast feedback from surface water.
The current possibilities for coupling to other codes are:
1. MODFLOW for groundwater (operational);
2. SOBEK-CF for surface water (under development).
The used schematisation assumes that the separate SVAT columns only interact with each other via their connections to groundwater and surface water.
Figure A.1: Overview of the processes modelled in SIMGRO. MetaSWAP [Van Walsum and Groenendijk(2008)] is used for the SVAT (Soil Vegetation Atmosphere Transfer) processes that are modelled within vertical columns. These column models are integrated with the groundwater model (MODFLOW) and a surface water model; for the latter there are several options, including a simplified metamodel that can be linked to form a basin network.
The SIMGRO framework is intended for regions with an undulating topography and unconsolidated sediments in the (shallow) subsoil. Both shallow and deep groundwater levels can be modelled by MetaSWAP. This model is based on a simplification of ‘straight Richards’, meaning that no special processes like hysteresis, preferential flow and bypass flow are modelled. Snow is not modelled, and neither the influence of frost on the soil water conductivity. A perched watertable can be present in the SVAT column model, but interflow is not modelled. There are plans for including the mentioned special processes in MetaSWAP Inundation water can be modelled as belonging to both groundwater and surface water at the same time. Processes that are typical for steep slopes are not included.
The code contains several parameterized water management schemes, including irrigation and water level management.
The spatial scale of the model is typically for a unit of 1 km\({}^{2}\) and less. Model applications have involved up to 500 000 units (National Hydrological Instrument for the Netherlands). A prototype model of a small basin involved 800 000 cells of 5x5 m. A resolution finer than 5x5 m is considered to be beyond the scope of the model, because then the one-dimensional flow schematization is not adequate.
It is possible to couple several MetaSWAP columns to a single ground water cell. In that way it is possible to use ‘tiles’ for representing fractions of the soil surface, for e.g. the vegetated part and the built-up part.
The model uses two nested time scales:
1. a fast cycle for the plant-atmosphere interactions and the interactions with surface water;
2. a slow cycle for the unsaturated zone and the coupling to the groundwater model.
Typically an interval of 1 hour is used for the fast processes, and 0.5 or 1 day for the slow processes. The time step of the slow processes and that of the groundwater model should be equal.
The required input data are described in Alterra Report 913.2. The main categories of input data are:
1. temporal scales;
2. schematization and coupling to other models;
3. soil elevation and soil physical data;
4. land and water use parameters, including irrigation demand;
5. irrigation water supply capacities;
6. meteorological data, including the option of grid files;
7. output control parameters
The main categories of output data are:
1. databases involving up to 129 (optional) data simulated items per SVAT unit, at the time scale the groundwater model, and also for user-defined output periods;
2. files in csv-format that are accessible while the model is still running (‘monitoring files’)
The model is run from a DOS-prompt and communicates with the user via files, both binary and ASCII ones. By using the facility of the csv-files, the user can follow a simulation as it progresses.
The model available for Windows platforms. The used language (Intel Fortran) makes it possible to migrate to a Linux platform, however, this is not supported by Wageningen Environmental Research (WEnR). The model has been coded with ‘dynamic’ memory allocation, meaning that the amount of used RAM memory is exactly tuned to the needs of the application. Large models will require a 64-bit environment due to RAM memory requirements.
The model is freely available, including the source code.
The model and documentation is available at the SIMGRO ftp-site
ftp://ftp.wur.nl/simgro/ and contains the following folders:
The doc folder contains (among others) the following documents:
More information about the model can be obtained from: