Navigation: Home ›  The model ›  Model development ›  Milestones

Milestones of model development

1994 - 1996 model version I with extended saturation area approach after TOPMODEL.
other: Provision of a complete set of help tools for pre- and post processing (e.g. Tanalys)
1997 - 1998 model version II with soil water dynamic in layered soils after RICHARDS.
other: implementing a 2D-groundwater model with exfiltration and re-infiltration; possibilities for irrigation control and tracer- und salt transport
2000 Extensions to the snow model
2001 Prototype of a graphic user interface for Windows (see also WaSiM-GUI)
2002 additional output options
2005 - 2006
  • WaSiM adapter for the Flood Early Warning System FEWS, a product made by Deltares (NL);
  • Implementing the dynamic glacier model (growing and shrinking of glaciers, consideration of firn and of the metamorphosis of snow to forn to ice);
  • dynamic time step control for groundwater model
  • Extensions to the silting-up module (expression parser for customizable parameter functions);
  • Extended model outputs for variables of the vegetation development;
  • Extensions to the interpolation methods (e.g. multiple lapse rates for method 12);
  • extended rules for reservoir abstractions (temporal flexibility for intraday and intraweek rules)
  • Implementing 1D vertical heat transfer model, coupled to the Richards-soil model, including thawing and freezing of the soil layers (heat conduction and heat advection)
  • Extension to the glacier model: consideration of debris-covered glaciers
  • Adding a new section to the control file for special outputs. Individual outputs of almost all internal entities can now be defined for several single grid cells and layers
  • internally, WaSiM now works with double precision, thus speeding up the model by about 20% and providing higher accuracy
  • Implementing MPI functionallity in some other model components (glacier model, groundwater model) to make the model ready for super computers (still under construction)
  • In snow model, eight new methods were added which are all possible combinations of the new options also with old methods: wind driven lateral snow redistribution, gravitational snow redistribution (small slides) und energy balance approach
  • WaSiM can now read and write stacks in ASCII format. So all inputs and all outputs can be provided in ASCII format, allowing an easier postprocessing (but requiring more time for reading and writing as well!)
  • A new mode is introduced which is especially usefull for spin-up runs for the heat transfer model but can also be used for general model purposes: The HRU mode (Hydrologic Response Units). Since it is used mainly for spin-up, the HRU generation is controlled in a new optional control file section ([spinUp]). HRUs can be defined as a combination of unique elements of any grid, e.g. elevations (in steps), land use, soil types, slope classes etc.
  • connected with Spin-Up: A pre-spin-up for estimating the soil temperature, especially for permafrost regions was implemented. Before the regular spin-up runs, the pre spin-up runs temperature interpolation only and estimates a mean annual permafrost table temperature (or, if no permafrost exists, the mean annual soil temperature). Thus, the following regular spin-up procedure will find an equilibrium much faster than from linearly interpolating an arbritrary air temperature as starting temperature for all soil layers down to the lower boundary condition.
  • Reading GIF Images (specifically: Radar precipitation measurements) as precipitation input, regardless of their grid domain and cell size. Values are then interpolated using IDW method to the model domain and model resolution
  • Soil depth and groundwater layer thickness are always identical: aquifer thickness is taken from soil tabel. ALso, easy scaling of soil thickness (and aquifer thickness) is possible per coild type by using a new optional parameter in the soil table
  • MPI memory optimization: until now, each instance used to allocate the complete memory that was required for a single instance. Now, each instance only allocates the stripe it is working on (for most grids and stacks - some grids are allocated completely in the master instance only and stripe-wise in the other ranks). The effect is dramatic: WaSiM can now run in much more instances per nodes (on big workstations and super computers) even for extremely large model domains with several GByte of allocated RAM for a single instance.
  • Some important bugs where fixed (interpolation and precipitation correction related and glacier runoff routing related bugs - see details in releasenotes).
  • Surface routing works with a linear set of equations and Gauss-Seidel solution algorithm now. This is much faster and lakes or other backwater playec have much more stable surfaces now. However, sionce this algorithm is a Finite Differences approach, some artifacts may show up (like water flowing sometimes over dry dams when using too les iteration steps)
  • Layered snow model introduced. This model component actually only works with heat transfer model switched on, since the snow layers are handled as soil and water/ice at the same time. So the temperature is quite important for all computations of heat and water fluxes
  • A new abstraction rule for reservoirs allows the physically based routing between hydraulically connected lakes (abstraction rule of type HydraulicConnection). Thus, fluxes can accure in both directions. Sample application: three peripheral Jura Lakes in Switzerland.
  • External coupling reactivated. Works now also with MPI. In addition to the classic functionallity, the semaphorfile optionally contains a command that can be interpreted by WaSiM or an external application (continue, cancel, save, full stop).
  • reading and writing of netCDF files implemented. netCDF files must follow the classic data format with one default group. Compression is supported for both writing and readin. Grid time series can be stored in a single, compressed netCDF file.
  • Re-infiltration is now available in MPI version by re-ordering river link cells in a separate grid row and processing all cells of a single link in a single MPI rank from highest to lowest cell, thus following the natural flow paths. This feature is also used in non-MPI-versions, so the re-infiltration will show slightly different results if compared to older versions. The main advantage of the re-ordering of river link cells and processing them separately from the normal cells is the fact that this can be used in MPI without too big performance losses (and that it is possible now at all to use re-infiltration in MPI).
  • Extended netCDF handling implemented: until now, the variable to read must have had the dimensions in order var(time,x,y), whereas the y variable was considered top-down. New behavior: var(time,x,y) and var(time,y,x) are possible. x-dimension is still always considered to be read from west to east, while y variable can be from south to north or vice versa. var(time,y,x) means, that the grid was written in rows to the netCDF file, starting either from the top or from the bottom of the grid. WaSiM will read the y-dimension variable and analyze the first two entries. If the second entry is larger than the first entry, the rows are considered to start at the lower left hand corner, while when the second y value is smaller than the first y value, the rows are considered to start in the upper left hand corner.
  • Extension of the simple interception model by a physically based snow canopy interception model (implemented by Kristian Förster, Uni Hannover, Germany) and by a canopy energy balance model for a snow filled canopy interception storage (implemented by Matthias Kopp, TU Munich, Germany). Both modules allow a much better snow pack modelling in forested areas, where the ground snow cover varies often quite remarkably from open field conditions.
  • Data assimilation as new optional sub model available. Grids can be read in at arbitrary times, e.g. snow water equivalent grids that are available on a weekly basis but only in winter months. SWE values can be split into liquid and solid components based on the present SWE at the actual or neighboring cells. Also, reservoir water table values can be read in to adjust for accumulated errors in inflow/outflow computation, thus bringing the model back to realistic lake water tables.
  • sub time steps of down to one minute can explicitely be used by adding a fifth time header column "mm" (for minutes) in each input file and in each output (graphlines and all time series processing tools are compatible with this change)
  • time variabel melt parameters for all T-index and T-u-index melt methods can be defined in the snow model section
  • command line option -mode:routing invokes the "routing-only" mode of WaSiM
  • Seamless coupling of model runs with different time steps and/or differing hydraulic parameters for channel routing possible by computing initialization values from the original structure instead of simple reading
  • full support for netCDF (with same restricted in netCDF funktionality as before): stacks can be read in netCDF format and all standard grids, variable grids and StateIni-grids and stacks can now be provided in netCDF format
  • I was not lazy, but the ongoing development was a bis slowed down...
  • Ice wedge degradtion can be used to model the landscape development for arctic tundra regions. DEM is dynamically changed when ice wedges thaw during the melting period in the arctic summer, thus laying the ground for fututre model extensions that model the land use change in the arctic (vegetation patterns) in a changing climate
  • Implementing MPI support in C instead of the no longer supported C++ framework, making the MPI version the new default version,
    Dicontinuing 32 bit versions.
    Discontinuing Topmodel versions.
    implementing additional standard grids for usable field capacity, ETR2ETP and relative soil moisture (compared to tooal pore volume)
back top home



new release availabe:
WaSiM 10.08.00 -> More »