Improved 1D Cover Modeling
A feature of SVFlux (part of SVOffice 2009)
March 20, 2009
In order to improve the stability and performance of our 1D coupled climate modeling within SVFlux, software a number of examples models have been set up and run in the software. It was desired that a number of represented models be selected from across North America. In this manner, represented climate conditions and material properties can be selected as a good basis for a basis for entering a wide range of models and trying them in the SVFlux software.
A series of models originally presented in the ACAP program were selected for the modeling exercise (Benson, 2007). These models represented a 1D covered design models from a series of mine sites from across the United States. Because of their diversity, they represented a wide range of possible modeling scenarios.
The geometry of these models was selected as presented by Benson (2007). The location of each of the numerical models was known and therefore the climate data for the nearest weather station was selected for the model from the National Climatic Data Centre. Material properties were approximated for each zone in the numerical model and only one-year of climate data was selected and run through the model for this particular example.
It should be noted that most models should ideally be run for more than one year in order to come to a steady-state conditions. However, the desire for this modeling exercise was to demonstrate stability and performance on a wide range of numerical models. Therefore, the models were not run for extended periods of time. It should also be noted that these models should not in any way be construed to represent real-world conditions at any of these actual mine sites. These illustrative models are different than their real-world counterparts in that the material properties in these models are estimated.
Average representative material properties were selected from the SoilVision database and were entered into the numerical models.
Historically such cover models are highly prone to convergence difficulties during the solution. This is because the cycles of wetting (precipitation) and drying (evaporation) are constantly happening at the upper boundary. The difficulties of solving this type of problem have been historically well documented and are primarily due a lack of mesh resolution at the upper portion of the profile.
Therefore, stability of software in modeling these types of situations is important. The purpose of this exercise was therefore to determine the stability of the software in solving average cases of 1D water flow over a wide variety of covers and conditions which are typical of mines sites around the United States. Most of the covers presented in this ACAP analysis are store and release covers, in that water is designed to be trapped by the cover and then later released to the atmosphere through evaporation.
The evaporation calculations in SVFlux software are made using the Penman-Wilson formulation (Wilson, 1990). This formulation allows calculations of the true actual evaporation (AE), based on suction levels on the upper zone of the soil. The reduction in the evaporation as the soil desaturates is handled through mathematical relationships.
The results of the library of model runs are as follows:
This small exercise demonstrates that if a numerical model is created with reusable conditions encountered in North America, then the model should be stable and complete the model run in a reasonable time.
These models can be downloaded from our software as part of our example models, and are found under the "Covers" project.
ACAP Website link Resources: