MODFLOW Groundwater Modeling — Modular Finite-Difference Groundwater Flow
MODFLOW Modular Three-Dimensional Finite-Difference Groundwater Flow Model · Also known as: MODFLOW-2005, MODFLOW-6, modular groundwater flow model, USGS groundwater model
MODFLOW is the U.S. Geological Survey's open-source, modular finite-difference model for simulating three-dimensional groundwater flow through porous media. First released in 1984 and continuously updated — most recently as MODFLOW-6 — it is the global standard for quantitative hydrogeological analysis, widely used in civil engineering, environmental consulting, water-resource management, and groundwater contamination studies.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
MODFLOW is appropriate whenever a quantitative, spatially distributed estimate of groundwater flow or head distribution is required and sufficient field data exist to characterise the aquifer. Typical applications include municipal well-field design, dewatering analysis for excavations, groundwater-surface-water interaction studies, contaminant plume prediction, and aquifer recharge assessment. The method is not suitable when data are too sparse to parameterize a grid reliably — in such cases simpler analytical solutions or statistical water-balance approaches may be more defensible. MODFLOW also assumes flow through porous media and is not appropriate for karst conduit systems without specialised extensions.
Strengths & limitations
- Open-source and freely distributed by the USGS, with decades of peer review and a large global user community.
- Modular architecture allows optional packages (transport, variably saturated flow, surface-water coupling) to be activated as needed without rewriting the core code.
- Capable of simulating complex, heterogeneous, multi-layer aquifer systems under both steady-state and transient conditions.
- Extensive calibration tooling (PEST, PEST++) enables automated parameter estimation and formal uncertainty analysis.
- Output is fully spatial and temporal, enabling detailed water-budget accounting and scenario comparison.
- Finite-difference grids cannot conform to irregular geological boundaries or curved features without significant cell refinement, increasing computational cost.
- Standard MODFLOW assumes saturated flow through porous media; karst, fractured rock, and variably saturated (vadose-zone) flow require specialised extensions or alternative codes.
- Data requirements are high: reliable hydraulic conductivity fields, boundary conditions, and calibration targets (head measurements, streamflow records) are needed for defensible results.
- Model results are only as good as the conceptual model; structural errors in the conceptualization propagate through all subsequent steps and cannot be corrected by calibration alone.
Frequently asked
What is the difference between MODFLOW-2005 and MODFLOW-6?
MODFLOW-2005 is the last release in the original Fortran-77/90 codebase and remains widely used. MODFLOW-6 is a complete rewrite using modern Fortran with an object-oriented structure; it supports simultaneous simulation of multiple models (e.g., coupled regional and local grids), introduces the Groundwater Transport (GWT) model natively, and uses a new input-file format. New projects are generally encouraged to use MODFLOW-6, but MODFLOW-2005 results remain valid and its packages are extensively documented.
How fine does my grid need to be?
Grid resolution is a balance between accuracy and computational cost. Near features with high hydraulic gradients — pumping wells, rivers, or sharp geological contacts — finer cells reduce discretisation error. A commonly applied rule of thumb is that cells adjacent to wells should have a width no greater than one to two times the well screen length. Away from stress features, coarser cells are acceptable. Sensitivity analysis of grid resolution is good practice before relying on model outputs.
Do I need field data to use MODFLOW?
Yes — MODFLOW requires spatially distributed hydraulic parameters (at minimum hydraulic conductivity and storativity per layer), boundary conditions (e.g., river stage, well pumping rates), and calibration targets such as measured water-table elevations or spring discharge records. Without adequate field data, parameter values are essentially guessed and model credibility is very low. The quantity and spatial coverage of data needed scales with model complexity and the decisions the model must support.
Can MODFLOW simulate contaminant transport?
The core MODFLOW code simulates only groundwater flow (hydraulic head and fluxes). Contaminant transport requires a coupled transport model. Options include the native GWT model introduced in MODFLOW-6, the standalone MT3D-USGS package (advection, dispersion, sorption, decay), or SEAWAT for variable-density flow and transport in coastal aquifers. The flow model is run first to produce velocity fields, which the transport model then uses to track solute concentrations through time.
Sources
- Harbaugh, A. W. (2005). MODFLOW-2005, the U.S. Geological Survey modular ground-water model — the Ground-Water Flow Process. U.S. Geological Survey Techniques and Methods 6-A16. link ↗
- Langevin, C. D., Hughes, J. D., Banta, E. R., Niswonger, R. G., Panday, S., & Provost, A. M. (2017). Documentation for the MODFLOW 6 Groundwater Flow Model. U.S. Geological Survey Techniques and Methods 6-A55. link ↗
How to cite this page
ScholarGate. (2026, June 3). MODFLOW Modular Three-Dimensional Finite-Difference Groundwater Flow Model. ScholarGate. https://scholargate.app/en/civil-engineering/modflow