Home » Public Domain Software My Account  |  Cart Contents  |  Checkout   
   
Search
Free Catalog!
Register to receive Free Catalog
Featured more
MODRET
MODRET
Information
Order & Payment
Shipping & Returns
Privacy Statement
Price Guarantee
SSG Home
Contact Us
HST3D

Detailed Description

HST3D

HST3D





HST3D GUI - A Windows Preprocessor and Postprocessor for Simulation of Heat and Solute Transport in Three-Dimensional Ground-Water Flow Systems

Save Money - Work with all the models you wish to interface yourself.

Save Time Learning - Use the same modeling environment with all models.

Reuse Your Data - Data for one model can be used for all other models.

Introduction

HST3D, "A Computer Code for Simulation of Heat and Solute Transport in Three-Dimensional Ground-Water Flow Systems" is by Kenneth L. Kipp, Jr. of the USGS. HST3D enables you to:

  • Assess well performance including the type of well bore.
  • Analyze pressure flow, heat and solute transport in the saturated zone with variable or constant density and viscosity.
  • Model ground-water flow separately.
  • Model heat and/or solute transport coupled with ground-water flow.
  • Predict chemical species transport including landfill contaminant movement.
  • Predict waste injection into saline aquifers.
  • Analyze freshwater storage in saline aquifers and saltwater intrusion in coastal aquifers.
  • Analyze liquid-phase geothermal systems and heat storage in aquifers.
  • Model brine disposal and movement of connate water.
  • Model contaminant (single species) transport in complex 3D aquifer systems.
  • Model hydraulic barriers, liners and water-quality protection systems.

The heat and solute transport program, HST3D, simulates ground-water flow and associated heat and solute transport in three dimensions. The HST3D model may be used for analysis of problems such as those related to subsurface waste injection, landfill leaching, saltwater intrusion, freshwater recharge and recovery, radioactive waste disposal, water geothermal systems, and subsurface energy storage. The three governing equations in HST3D are coupled through the interstitial pore velocity, the dependence of the fluid density on pressure, temperature, and solute-mass fraction. The solute-transport equation in HST3D is for only a single, solute species with possible linear-equilibrium sorption and linear decay. Finite-difference techniques in HST3D are used to discretize the governing equations using a point-distributed grid. The flow, heat, and solute-transport equations in HST3D are solved, in turn, after a partial Gauss-reduction scheme is used to modify them. The modified equations are more tightly coupled and have better stability for the numerical solutions.

The basic source-sink term represents wells. A complex well flow model may be used to simulate specified flow rate and pressure conditions at the land surface or within the aquifer, with or without pressure and flow-rate constraints. Boundary condition types offered in HST3D include specified value, specified flux, leakage, heat conduction, an approximate free surface, evapotranspiration, and two types of aquifer influence functions. All boundary conditions can be functions of time.

Four techniques are available in HST3D for resolution of the finite-difference matrix equations. There are two variations of a generalized conjugate gradient iterative solver, a triangular-factorization direct solver and a two-line successive over-relaxation solver.

A restart option is available in HST3D for storing intermediate results and restarting the simulation at an intermediate time with modified boundary conditions. This feature can also be used as protection against computer failure. HST3D is a descendant of the Survey Waste Injection Program (SWIP) written for the USGS under contract.

HST3D is now connected with the general Pre and Postprocessor Argus Open Numerical Environments (Argus ONE ) that permits it to work in Windows with easy database management. Data input and output may be in metric (IS) units or inch-pounds units. Input and Output may be represented as tables of dependent variables and parameters, zone contour maps, and plots of dependent variables versus time.

HST3D Key Features
  • Specified-value and specific-flux boundary conditions are independent on each portion of the boundary and may vary with time in HST3D.
  • Specified heat and solute flux boundary conditions are available.
  • Leakage, aquifer-influence function river leakage, and evapotranspiration boundary conditions are available in HST3D.
  • Porous media thermal properties, dispersivity and compressibility may have spatial variation defined by zones.
  • A point-distributed, finite-difference grid is used rather than a cell or block-centered grid. It allows a better truncation error and an easy incorporation of boundary conditions.
  • The heat conduction boundary condition is generalized to apply to any cell face.
  • Global flow, heat and solute balance calculations are performed including flux calculations through specified pressure, temperature and mass fraction boundaries.
  • A robust algorithm for the computation of the optimum over-relaxation factor for the two-line successive over-relaxation matrix solution method is employed with a convergence criterion that includes the matrix spectral radius estimate.
  • A read-echo file is written. It assists the user in locating errors in the data-input file.
  • Although the internal calculations of the program are performed in metric units, the input and output can be chosen to be in inch-pound units.
  • Error tests are included to locate mistakes in data input.
  • Error messages are printed explicitly rather than as code numbers.
  • The solute concentration can be chosen to be the mass fraction or a scaled mass fraction that ranges from 0 to 1.
  • Map contours of any output or input data may be created directly on the screen.
  • Initial pressure conditions can be specified to be other than hydrostatic. For example, an initial water-table configuration can be used.
  • Precipitation and replenishment can be specified using the distributed flux-boundary conditions.
  • The conductive heat loss to overburden and underburden is a general heat transfer calculation, applicable to any cell face in the region.
  • The well-riser calculation has been formulated to solve the total energy and momentum balance equations simultaneously using the Bulirsh-Stoer algorithms for integration of the ordinary differential equations.
  • The well bore equations in HST3D are implicitly coupled to the system equations for cases of cylindrical geometry.
  • The well-datum pressure and the well flow rate calculations may be performed explicitly or iteratively in conjunction with the solution of the flow equation.
  • The full nine components or an approximate three-component dispersion coefficient tensor may be used for cross-dispersive flux calculations in HST3D.

New in HST3D 2.0
  1. There is a new iterative solder for the matrix equations based on a generalized conjugate-gradient method.
  2. An evapotranspiration boundary condition has been added.
  3. The simulator output has been divided into many files.
  4. There is a new set of output files designed for use by post-processing programs for graphical visualization and for flow totalization.
  5. There is a pre-processor for evaluating dimensioning requirements.
  6. There is a post-processor for totaling boundary flow rates and cumulative amounts.
  7. The free surface boundary condition algorithm has been revised to allow the free surface to move to any elevation in the simulation region and cells to dry and rewet.

One way to objectively assess the impact of existing or proposed activities on ground-water quantity and quality is through the use of ground-water flow, heat and solute transport models. HST3D allows a quantitative understanding of how the sources and sinks, the boundary conditions and the aquifer parameters interact to cause ground-water flow patterns and consequent thermal and solute concentration movement in a studied area. The magnitude of concentrations and discharges at the boundary of the studied area are of particular interest in the study of a contaminated aquifer. The degree of realism and the accuracy of a given simulation is strongly dependent on the quantity and quality of the parameter distribution, boundary conditions and source sink data.

HST3D Limitations

HST3D is suitable for simulating ground-water flow and the associated solute transport in saturated, three-dimensional flow systems with variable density and viscosity. As such, the code is applicable to the study of waste injection into saline aquifers, landfill contaminant movement, sea-water intrusion in coastal regions, brine disposal, freshwater storage in saline aquifers, heat storage in aquifers, liquid-phase geothermal systems, and similar transport situations. If needed, only the ground-water flow may be solved. Also, after the computation of the ground-water flow, only the heat or the solute transport equation may be solved. Three-dimensional Cartesian or axisymmetric, cylindrical coordinate systems are available. (The GUI only supports cartesian coordinates.)

The finite-difference techniques used for spatial and temporal derivative approximations have some limitations:
  1. Where longitudinal and transverse dispersivities may be small in HST3D, cell sizes will need to be small to minimize numerical dispersion or oscillations. Furthermore, if the region of solute is somewhat convoluted and three dimensional, the projection of nodal lines from regions of high nodal density will result with an excessive node number in other regions. These two factors may cause an excessive number of nodes for a given simulation, thus making the simulation prohibitively expensive because of computer storage and computation time requirements. In such cases, a simple model of the system, useful for investigating mechanisms and testing hypothesis, may be the only practical solution.
  2. Another limitation of HST3D results from the so-called grid-orientation effect (Aziz and Settary, 1979, p.332). Numerical simulations of miscible displacement converge to two separate solutions as the mesh size is refined depending on whether the major velocity vectors are parallel to one of the coordinate directions or are diagonally oriented. This effect is more pronounced for conditions of little dispersion or piston-like displacement of the solute, and for conditions of small viscosity of the displacing fluid. The effect is almost absent when the two viscosities are nearly equal, or if the dispersion coefficient is large. One of the causes of the grid orientation effect appears to be the use of a seven-point difference formula for the three-dimensional flow and solute transport equations because this formula restricts transport in the diagonal directions. Use of a grid where the major velocity vectors are oriented parallel to one of the coordinate directions has been found to give more realistic results (Aziz and Settari 1979, p. 336). To completely eliminate this problem, a higher-order differencing scheme or curvilinear coordinates need to be used, but these modifications are not implemented in the present version of HST3D.
  3. The boundary conditions in HST3D can be used with a tilted coordinate system. The free surface and leakage boundary conditions require that the Z-axis be oriented in the vertical direction.
  4. HST3D has difficulty in representing quantitatively viscous fingering instabilities and an abrupt change of fluid density that may occur when a fluid of greater density overlies one of lesser density. For most of ground-water flow and transport modeling these physical phenomena are secondary. Viscous fingering instabilities may occur during the displacement of a resident fluid by an injected fluid with significantly less viscosity. The injected fluid forms channels or fingers through the resident fluid, as described by Saffman and Taylor (1958). When a fluid of greater density overlies one of lesser density, Raleigh-Taylor convective cells are formed. In these cells, the two fluids mix. Numerical simulation tends to predict these transport instabilities later than they occur in laboratory scale experiments. However, laboratory-scale viscous fingering and convective cell formation may be much more unstable than the corresponding field scale. Therefore, at the field scale, numerical simulation may be more valid than at a laboratory scale. Nevertheless, these limitations need to be kept in mind when simulating fluid flow with large viscosity or density contrasts.

HST3D Documentation and Verification

A user guide in the form of a series of linked web pages comes with the GUI. The source code is available on request for registered users. In addition, a tutorial in Windows permits the user to easily understand the menus and commands. For more general explanations on the mathematical and physical aspects of the HST3D code and the exact definition of each parameter, the HST3D report/doc is available separately. Two example problems are described in detail with input and output files. The Help Menu and the different documents are intended to be sufficiently complete and easy to use. The user may now easily obtain successful simulations, diagnose computational problems and develop remedies.

HST3D-GUI System Requirements

To use the HST3D GUI, you must have Windows 95/98/2000/NT. You also need at least 16 MB of memory. Full installation requires approximately 7 MB of free disk space. Of this, approximately 6 MB comprises HST3D and related executable programs (HST3D: 5.4 MB, DIMEN: 340 Kb, and BCFLOW: 145 Kb.) To view the online Help for the HST3D GUI, you need a frames-capable web browser. (The more popular web browsers are frames-capable.) The memory requirements of HST3D vary depending on the size of the arrays used in the program. Registered users may have HST3D recompiled to change the memory requirements up to three times per year without charge.



Copyright © 2014 Scientific Software Group