Model Inventory and Selection Tool (MIST) Codebook
Overview
The Mid-Atlantic Regional Integrated Sciences and Assessment (MARISA) program is committed to supporting communities with climate resiliency planning and preparedness through improved data, decision support, and public engagement. Models are an important part of understanding climate change impacts and implementing adaption strategies. In the Mid-Atlantic region and specifically the Chesapeake Bay watershed, many models and decisionmaking tools have been created. These tools are intended to support various components of watershed improvement for social, economic, and ecologic benefits. However, stakeholders and decisionmakers with and without expertise in modeling and simulation might have difficulty keeping track of and selecting from support tools because of the sheer number available and the often high-level of model complexity. To support the use of these models and tools for climate adaptation planning efforts, MARISA developed the Model Inventory and Selection Tool (MIST)—a database categorizing models relevant to water- and climate-related decisionmaking in the Chesapeake Bay watershed. The tool was first released in 2021. An updated, revised version in 2023 added three new models and provided minor corrections to the database.
The MIST database is a collection of empirical, statistical, and conceptual models relevant to characterizing physical systems in the Chesapeake Bay watershed. These include both natural and human-made systems, such as watershed, estuary, meteorological, and coastal models and urban development, urban runoff, transportation and infrastructure models.
The model list was compiled from a literature review and a search of relevant resources from agencies creating or supporting modeling efforts. From this list, each model was investigated in more detail. Model characteristics (e.g., model type, intended user) were defined to assist decisionmakers in selecting a model best suited to their needs and to provide sufficient granularity to differentiate between many similar models. Once categories were created, the database was populated with model-specific descriptor fields for each category (e.g., the model type category includes coastal or groundwater). In instances where model information for a specific category was not available, the field was left blank.
This page contains a summary of each category and the related descriptor fields. Each field is defined. Definitions are intended to provide additional information about what a field implies about model functionality and the context in which fields apply to a model.
Categories and Descriptor Fields
Categories are described, and each field option is defined.
Model type: This describes the most important goal or task that can be addressed by a specific model. This includes what system the model is best suited to describe and the types of questions it can answer. These categories are intentionally broad. Sorting by this category will provide a broad indication of the models that could be used for a specific environmental system.
| Fields | Definition |
|---|---|
| parameter calibration | This model uses data and statistical techniques to determine parameter values. This is intended for use during the calibration process. |
| groundwater | The primary model purpose includes analysis of groundwater systems. This can include subterranean water flow, mechanics and implications of water withdrawals, and fate and transport processes. |
| coastal | The primary model purpose includes analysis of coastal processes. This includes estuary water quality, sediment, and interactions with upstream riverine systems and downstream ocean systems. Analysis includes effects of sea level rise, nearshore wave interactions, flooding extents, and water quality constituent fate and transport processes. |
| watershed | The primary model purpose includes evaluation of watershed systems. This includes movement of water and nutrients from land surfaces to riverine or groundwater systems and how those processes change under changing physical conditions. Physical conditions include land use change, changes in precipitation, urbanization, and changes in infrastructure located in watersheds. |
| infrastructure | The primary model purpose includes evaluation of interactions between natural systems and infrastructure systems. This includes hydropower and geothermal power generation systems, how storms affect coastal infrastructure, and urbanization effects on stormwater flows and nutrient transport. |
| decision making | The primary model purpose includes models that are structured to help inform decisionmakers. This is through visualization add-ons, spatial analysis, or statistical analyses intended to inform or clarify decision choices. |
| ecosystem | The primary model purpose includes analysis of ecosystems. Models evaluate biological system response to different stressors. Biological systems in this context means the flora and fauna found in the Chesapeake Bay region. Most models are focused on estuary, ocean, and riverine communities, but terrestrial systems are also included. |
| transportation | The primary model purpose is to evaluate transportation systems. Models include planning and design models of road and public transit systems. Capabilities include project planning, cost assessments, and travel demand and traffic simulation. |
| climatology and meteorology | The primary model purpose for climatology and meteorology models include models that predict near-term weather and storms and long-term climate models that approximate long-run trends in temperature, precipitation, etc. |
| air quality | The primary model purpose for air quality models is to evaluate fate and transport of pollutants in air sheds. Deposition to ground and volatilization from water and land systems capture the interactions between water and air sheds. |
| risk communication | The primary purpose of risk communication models is to use results from other models and add visualization or integration of results from many models to show uncertainty or variability. These models are intended for communication of results to nontechnical audiences. |
| urban development | Urban development models evaluate urban land use change and how urban structure affects temperature, stormwater flow, and quality and more generally the interactions between human-made and natural systems. |
Model dimensions: This category identifies the physical spatial dimensions captured by the model. The highest dimensionality of the model is listed. Models in 3D typically allow for simplifications to lower dimensions if needed.
| Fields | Definition |
|---|---|
| 1d | A one-dimensional model that is typically applied to river models in which a single flow direction is dominant. |
| 2d | A two-dimensional model that is typically applied to complex river interactions with bank conditions or as a simplification of watershed flow models. |
| 3d | A three-dimensional model that is most complex model form, typically used for coastal and estuary models for which interactions at the shore are complex. |
| not applicable | Model does not address dimensionality or function does not include a spatial evaluation. |
| not available | Model does not specify dimensions. |
Geographic region: This category identifies the geographic region that a model is designed to represent.
| Fields | Definition |
|---|---|
| generalized | Model is not specific to any specific location; the equations and mathematics are adaptable through parameters and input data to address the behavior of any geographic region |
| USA | Model is intended only for modeling regions in the United States. This is typically because of preloaded data sets or data availability that is restricted only to the United States. |
| international | Model only addresses modeling questions for locations outside the United States. Typically, this is because data sets used in the model or efforts of model developers have been in locations outside the United States. |
| coastal regions: east | Model is specific to the East Coast of the United States. |
| coastal regions: west | Model is specific to the West Coast of the United States. |
| Maryland | Model is specific to the state of Maryland. |
| Chesapeake Bay | Model is specific to the bay and/or watershed of the Chesapeake Bay. |
| Delaware River Basin | Model is specific to the Delaware River Basin. |
Minimum spatial resolution: This category identifies the minimum order of magnitude for a model’s spatial resolution. This is the scale at which a model completes calculations and is typically a range. The spatial resolution used for a specific model is determined by the user and is often based on available data.
| Fields | Definition |
|---|---|
| km | The model is not accurate or intended to model systems at a scale less than on the order of kilometers. |
| m | The model is not accurate or intended to model systems at a scale less than on the order of meters. |
| not applicable | The model does not include a defined spatial representation. |
Temporal resolution: This category identifies the time step used in the model. This is a measure of the resolution of the model. A model that does not perform computations at the second or minutes level will not have the accuracy to model short-duration events or storms. These short times steps are unnecessarily computationally intensive for some longer-run models evaluating change at the decadal scale.
| Fields | Definition |
|---|---|
| seconds | Computational time step is on the order of seconds. |
| minutes | Computational time step is on the order of minutes. |
| hours | Computational time step is on the order of hours. |
| days | Computational time step is on the order of days. |
| months | Computational time step is on the order of months. |
| years | Computational time step is on the order of years. |
| steady state | The model is intended for evaluation of steady-state systems; time step used is less important. |
Source code availability: This category describes how the model is made available to users.
| Fields | Definition |
|---|---|
| open source | Source code is freely available to see and edit. |
| limited | Source code is available to see and edit but behind a paywall or special request. |
| closed source | Source code is not available to see and edit, even if model can be downloaded for free, or source code availability is unknown based on available information. |
| not applicable | Model is not based on code. |
Developed by: This category is formatted as a string and is the organization that developed the model. This is typically an institution or government agency. Individual authors are not indicated unless the model was developed independently.
Maintained: This category indicates whether the model is currently maintained, or, if not, it describes if the code is still available. Models that are discontinued were listed if they were found referenced in other sources or as a part of other models.
| Fields | Definition |
|---|---|
| not supported | Model code is available but not regularly updated, and coding bugs are not corrected by the developer. |
| discontinued | Model is not actively being maintained, has not been updated in several years, and, in many instances, the code is no longer available. |
| maintained | Model is actively being updated to fix bugs and add functionality. Often these models will have user forums for assistance in model use and applications. |
Last update: This category indicates the most recent update of the code and is formatted in mm/dd/yyyy. If the code date was not found, the date on the user manual was used. If only a month and year are provided, the first of the of month was used in the day field to maintain consistent format. If only year is provided, January 1 is used as the month and day field to maintain a consistent format.
Computational burden: This category functions as a qualitative measure of the time it takes to run the model and the computer power required to run the model.
| Fields | Definition |
|---|---|
| high | The model is described as being recommended for use on a supercomputer or model scope and resolution indicate extensive computational times. |
| medium | The model includes some simplifications intended to reduce computational time, but the model scope is still large. |
| low | Model simplifications are extensive and the model resolution is reduced; time step or spatial scale are both larger. |
| unknown | Model computational burden is unknown based on available information. |
Replicability and scenarios: This category addresses if and how the model is structured to run in batch mode for automatic replicability, if scenarios must be run by hand, or if no replicability function is described. Replicability is necessary when multiple parameter values need to be run or different future scenarios or alternatives need to be evaluated systematically.
| Fields | Definition |
|---|---|
| batch mode described | User manual specifically states batch mode capabilities. |
| batch mode likely | Code language and structure are such that configuring the model to run in batches is likely straightforward. |
| scenarios by hand only | Scenarios are described, but the user manual does not indicate automatic processing of batch mode; model structure does not appear to be compatible with batch capability. |
| not applicable | Batch mode is not possible with a specific model or not necessary for the intended model function. |
| unknown | Batch mode is unknown based on available information. |
Data requirements: This category describes how data are entered or integrated into the model. Note that many models now include preloaded databases for many regions or “typical” values for certain parameters. Not having to generate all data and parameter information for a model with heavy data requirements can make a model easier to use. It also makes it easier to identify which parameters a region should investigate more closely without investing in site-specific information for everything.
| Fields | Definition |
|---|---|
| data provided by the user | All data required by the model are provided by the user. |
| some data sets built in | A portion of the data is already loaded in the model. Only some site-specific data sets and parameters are entered by the user. |
| all data sets built in | All input data required to run the model are preloaded. Many of these models still allow for adjustment of parameters, but a model can be run without any external data. |
Climate change: This category is a measure of how climate change and climate change–induced changes are captured by the model.
| Fields | Definition |
|---|---|
| explicit | Model explicitly addresses climate change through input data sets, parameter adjustment, linking to climate model output etc. |
| possible | Manipulation of existing input data fields allows for climate change–induced impacts to be modeled; however, the model was not structured specifically to address climate change. |
| not applicable | The model is not intended to evaluate climate change. For example, a model intended to evaluate event-based storms does not provide results over a temporal duration sufficiently long enough to capture climate change–induced effects. |
| unknown | Insufficient information is provided to determine if the model can be modified to evaluate climate change–induced effects. |
| not possible | The model is not structured to evaluate climate change effects on a system. |
User manual availability: This category records whether a user manual is available. It also documents whether it's readily available or must be requested from the maintaining organization.
| Fields | Definition |
|---|---|
| web | A user manual is available on the website in the form of a downloadable PDF or as a detailed webpage. |
| by request | A user manual exists but must be requested through the model website. |
| not available | A user manual is not available; overall limited information is provided on model functionality and operation. |
Intended user: This category determines who is the intended user of a model.
| Fields | Definition |
|---|---|
| decision maker | This model is intended for use by decisionmakers. Output and post processing functionality are tailored toward decisionmakers. This includes models that are easier to use and are often project-specific. |
| practitioner | This model is intended for use by practioners or researchers. These models are often more complex and require additional modeling or coding experience. Many models have a high computational burden and are highly scientific, requiring detailed background in model knowledge area. These models are also often flexible enough to be tailored to specific research questions. |
| public | This model is intended for public edification. These models are very easy to use—often web-based tools and mainly focused on visualization and communication of results. Many of these models are post-processors that provide visualization and interpretation capabilities to other models. |
Primary citation: URL link to the model main webpage.
Math approach: This category describes the mathematical approaches and techniques used by a model to generate results.
| Fields | Definition |
|---|---|
| spatial analysis | The model provides output that is spatially referenced. |
| expert elicitation | Expert input is incorporated into the model through the use of expert elicitation techniques. |
| modular modelling system | The model integrates multiple systems and has a modular approach. This includes models that evaluate interactions of large complex systems. |
| deterministic | The model uses any of the following: (1) mass balance to estimate flows or constituents at points in the system; (2) deterministic physics-based equations to represent system behavior; (3) control volumes and defined flows in and out of model compartments (boxes) to represent system behavior; (4) economic analysis of system behavior; (5) infiltration equations to estimate water flow from the surface through to groundwater; (6) curve number method to estimate runoff and infiltration based on land use; (7) cost-benefit analysis to compare alternatives; and (8) project-ranking analysis that ranks alternatives based on criteria. |
| statistical | The model uses any of the following: (1) regression to predict systems performance based on input data; (2) descriptive statistics that are used to elucidate trends in the data; (3) statistical frequency analysis to determine the frequency of model output or events; (4) probabilistic output; (5) spatially distributed parameter model with model statistics applied and parameters estimated based on a spatially resolved approach; (6) parameter estimation based on input data to estimate the parameters required for another model or another model component; (7) bias correction based on statistics to determine the bias associated with a model approach or parameter selected; (8) least squares estimation of model fit; (9) singular value decomposition solves a system of equations based on matrix decomposition; and (10) parameter correlation evaluates model parameters for correlation. |
| numerical | The model uses any of the following: (1) finite difference to approximate differential equations used to define the system; (2) finite volume to approximate solutions to partial differential equations used to define a system; (3) head balance to estimate water flow by equating the system pressure at points open to the atmosphere; (4) search space algorithm to search the decision space for optimal or near-optimal solutions; (5) analytical element to approximate solutions to partial differential equations used to define a system; (6) finite element to approximate solutions to physics-based equations that represent a system that is broken into small elements for which a system of algebraic equations is solved; (7) Lagrangian processes to solve process equations with Lagrange approximation; (8) fourth- and fifth-order Runge-Kutta to solve temporal discretization of a system of differential equations describing system behavior; (9) numerical integration to solve a system of integral equations with a series of numerical approaches; (10) nonlinear kernel approximates a nonlinear system of equations that is based on a user-specified kernel function; (11) Newton-Raphson method to approximate differential equations with a root-finding algorithm; and (12) Eulerian processes, a first-order approximation approach for a system of differential equations. |
| simulation | The model uses any of the following: (1) extended period simulation that runs over long durations to reach steady-state conditions or to evaluate long-term effects; (2) event-based simulation to look at the effect of short-term peak events on a system's performance; (3) Monte Carlo analysis models that input parameters with statistical distributions and that solves the model repeatedly to generate distributions representative of model behavior; (4) transportation microsimulation that includes unit-by-unit-level simulation of transportation processes; and (5) agent-based model in which units are represented as agents and their interactions are represented by defined equations. |
| optimization | The model uses any of the following: (1) linear programming—a linear system of constraint equations and linear objective functions that can be solved to determine a global solution to an optimization problem; (2) nonlinear programming—a nonlinear system of constraint equations or objective functions; (3) trade-off analysis that applies optimization models with multiple objectives to evaluate different optimal solutions; and (4) heuristics that approximate the objective space and that identify local optima. |
Outputs: This category describes the output datasets generated by the model. Specific parameters and data sets for an individual model can be found in the associated user manuals.
| Fields | Definition |
|---|---|
| water quality | Nutrient and pollutant load or concentration data are generated by the model for the defined system. |
| coastal water system dynamics | This includes wave-to-shore interactions and wave-to-built infrastructure interactions. Flooding and inundation levels are calculated by the model. |
| physical system parameters | The model generates physical system parameters, such as flow routing or land use–impervious parameters that can then be used as inputs for other models. |
| climate/meteorological parameters | This model generates climate data, rainfall data, temperature data, etc. |
| flow | This model generates flow data, including pressurized or open-pipe flow, overland flow, stream flow, etc. |
| damages | Typically associated with flood models, this model includes infrastructure and building data and computes the monetary damages associated with the storm or flood event. |
| risk | This model estimates the risk of an event or specific outcomes. This is typically associated with probabilistic and decision-oriented models. |
| storm characteristics | This model generates storm event characteristics, such as hydrograph data or rainfall-duration estimates. |
| flood extents | Flood extents are estimated. This is associated with spatially resolved models and can include both spatial inundation extents and flood depth estimates. |
| fish population dynamics | Many ecology models produce fish population data, which can then be used in fisheries models to estimate sustainable levels of fishing or the productivity that can be expected from specific water systems. |
| vegetative interactions | This model includes flow and nutrient interactions with vegetation. This can be either riverbed vegetation or near-shore vegetation. This is typically useful for models that design natural shore or erosion protection systems. |
| not applicable | This model does not have usable outputs and/or is intended only for visualizations. |
| other ecological outcomes | Ecological system data that do not include fishery information. This could be other population dynamics, energy or food system flows, etc. |
| decision support | Model outputs are structured specifically to assist decisionmakers. This could include visualizations, flow charts, or other outputs that help decisionmakers navigate the differences between alternatives. |
| spatial output | Output data sets are spatially resolved. |
OS platform: This category specifies the computer platform on which the model was designed to run. Systems are listed only if they were mentioned in model documentation. Some models may be compatible on systems that are not listed.
| Fields | Definition |
|---|---|
| Windows | Models are compatible with operation on Windows machines. |
| Mac | Models are compatible with operation on Mac machines. |
| Linux/Unix | Models are compatible with operation on Linux or Unix systems. |
| DOS | Models are formatted for operation on DOS systems. |
| not applicable | Model is web-based or compatible with any operating system. |
Coding language: This category specifies a coding language required to run or manipulate different versions of the model. Many models have different requirements, depending on the level of customization desired.
| Fields | Definition |
|---|---|
| FORTRAN | Model is written in or run through use of this language. |
| C/C++ | Model is written in or run through use of this language. |
| GUI | Graphical user interface exists for the model. |
| python | Model is written in or run through use of this language. |
| MATLAB | Model is written in or run through use of this language. |
| HTML | Model is written in or run through use of this language. |
| visual basic | Model is written in or run through use of this language. |
| macros | Model is written in or run through use of this language. |
| excel | Model is written in or run through use of this language. |
| java | Model is written in or run through use of this language. |
| R | Model is written in or run through use of this language. |
| not applicable | Model is written in or run through use of this language. |
| other: Delphi, Perl, Mathcad | Model is written in or run through use of this language. |