Key Platform Capabilities

MAGNET users can almost instantly create a big data-enabled model of groundwater flow (steady or unsteady) and contaminant transport anywhere in the world. Model inputs and representations (e.g., conceptual/structural, boundary and initial conditions, sources and sinks, number of layers, grid resolution, and display parameters) can be interactively modified and further customized based on study objectives and other available (local) data, as well as user experience and expertise.

The MAGNET platform offers the following additional unique capabilities:

Realtime IGW / MODFLOW Numerical Engine
The MAGNET model engine can utilize MODFLOW or Interactive Groundwater (IGW) solvers/algorithms and is real-time interactive, meaning you can visualize and computationally steer simulated flow and plume dynamics and water budget dynamically as the simulation proceeds (the output is provided dynamically as a "movie" or animation).

Realtime Cloud-powered Modeling
MAGNET is cloud-powered – accessible on demand anywhere anytime, through an internet browser on your PC, smart phone, or tablet. There is no software or data to download, there is are no operating system requirements, and simulation, analysis and visualization are handled on MAGNET’s server. Data, software, platform, as well as computational infrastructure are provided as a service, benefiting the entire environmental/water community – researchers, consultants, educators, students, and managers and planners.

Collaborative/Participatory Modeling
And because MAGNET is web-based, it enables an exciting, new type of participatory modeling experience where collaborators from around the world operate, control and work live on the same model(s). In other words, changes and updates made by one user in one location are displayed in realtime to other users in completely different locations. And when the user submits a model for simulation, all users in the group simultaneously view the results.

Realtime Data-enabled Modeling
MAGNET is data-enabled, meaning it is live-linked to a preprocessed global spatial database so that you can instantly create a unified flow and transport model that can be further refined / modified. Key global datasets that are available for modeling include: land topography and surface water networks (from high resolution Digital Elevation Models), surficial aquifer thickness, hydraulic conductivity of surficial aquifers/deposits, and mean long-term aquifer recharge. More datasets are available for specific regions (e.g., Michigan, USA), and we are continuously expanding the scope of our global spatial database.

Realtime Calibration with Bigdata
MAGNET allows flow model calibration at regional scale using water level measurements from densely populated, statewide/provincial water well databases. Calibrating regional flow models to high-density (but noisy) water level data provides an excellent starting point for sub-scale data collection, modeling, and analysis.

Realtime Hierarchical Modeling
MAGNET users can interactively and ‘on the fly’ insert models inside of models (inside of models, etc.) to characterize flow and transport processes at different spatial scales. This approach allows solving large complex problems encountered in real-world hydrogeology as a series of smaller, interconnected problems: multiple models across different scales or areas of interest are solved, instead of solving one large, high-resolution model. This approach significantly relaxes computational/numerical bottlenecks traditionally involved with tackling complexity.

Realtime Stochastic Modeling
MAGNET enables users to examine the impact of unmodeled small-scale heterogeneities and perform risk-based prediction through stochastic Monte Carlo simulations. Spatial parameters are represented random fields, and flow and transport simulations are automatically “recomputed” for the various realizations. Monte Carlo forward and reverse particle tracking applications allows realtime delineation of probabilistic capture zones and impact/influence zones, respectively. Monte Carlo solute transport modeling can be used to calculate the risk of exceeding MCL / threshold concentration values at locations of interest/concern.

Global Model Network and Instant Intelligent Reporting System
MAGNET is directly linked to a Global Model Network, allowing users to share selectively their model results (images, plots, animations) and/or the model and its data. MAGNET’s Intelligent Report System automatically and instantly converts “incomprehensible” low-level model inputs/options/settings into a high-level report summary of the model representation for easy assimilation by the general audience. Users are offered the flexibility to determine what and how much to make public.

Realtime Situational Help
MAGNET’s user reference material is situationally embedded inside the modeling platform interfaces and submenus; real-time help buttons provide instant access to sub-pages explaining how to use different options or features. The Realtime Help Pages are continuously improving and will include not only technical reference material, but also links to related scientific material (e.g., typical parameter values, numerical / computational aspects, etc.). Eventually, our goal is to allow users to ‘model without a manual’.

Action-oriented Curriculum
MAGNET is supported by a comprehensive set of web materials for groundwater education and learning how to use the various tools. The action-oriented groundwater curriculum consists of engaged-learning lessons, short problems and exercises, integrated capstone projects, video lectures and demos, and a Digital Library of model animations and visualizations. The problems / projects can be edited, customized, and refined by instructors based on their specific course learning objectives / curricular expectations, or new contents can be added directly to the Curriculum Network through MAGNET’s embedded editing system.

Free of Charge
It is FREE to use MAGNET. To ensure that we can offer this big data-enabled realtime interactive modeling platform as a service to the entire global community, we limit the grid cells to 150 in the x- (west-to-east) horizontal direction and the total number of grid cells in both the X and Y directions to be NxNy_max = 100 * 100 = 10000. The total number of vertical numerical layers is subLyr_max = 15, and the total number of model grid cells is NxNyNz_max = 100 * 100 * 10 = 100000. The maximum number of time steps is 300.

However, we do not really see this as a limitation; a vast majority of problems can be solved with smaller grid sizes, especially when taking advantage of MAGNET hierarchical modeling capabilities (using multiple models across different scales / areas of interest instead of one large, high-resolution model).

If the nature of your work requires that these restrictions be removed (at a cost), please contact us at and