Glossary

WORKING DEFINITIONS FOR THE MODEL EVALUTION TOOL

Evaluators

Evaluators come from multiple disciplines and expertises and provide actionable feedback to models’ inputs, outputs and metadata. For instance, evaluators can be taxonomic (e.g., birds) and survey experts that possess a deep knowledge and understanding of the ecology, biology and the geographic/environmental context of the species. Their knowledge allows them to review part of and/or the entire model, providing feedback to modelers. Evaluators can also review specific parts of the model. For instance, statisticians can review how modelers treated data sets (e.g., aggregating observations) and also assess the appropriateness of statistical analysis and the rigor in meeting statistical assumptions. Other modelers, unrelated to the model’s design, can play a significant role in providing feedback on specific aspects of the models (e.g., alternative algorithms, model tuning, etc.). Resource managers can evaluate information needs for multiple applications (e.g., forest management, habitat conservation, co-benefits) when confronting model outputs. They also can assess the risk of implementing a specific scenario due to a certain level of uncertainty.

Model evaluation tool (MET)

MET is an interactive Shiny app tool in R that integrates the functionality, attributes, infrastructure of all modules to evaluate SDMs, synthesize expert evaluation with quantitative evaluation, and inform model improvement and multiple uses and applications.

Model materials

It refers to a collection of datasets and information in multiple formats (Tables, text, figures and layers such as rasters and vectors) for model design, development and implementation. These materials will inform users and evaluators during the model evaluation and information extraction process. For instance,

Inputs

refers to datasets used for model implementation species observations, covariates, area of interest, etc. These inputs are usually raster and vector layers. It can also be a table with species observations coordinates.

Outputs

It refers to model predictions and uncertainty (usually raster layers). It also includes a matrix or table with information about model predictions, covariates relative importance, model performance metrics, etc. They are useful to create figures or summarizing tables.

Metadata

It refers to all information used in model design, development and implementation. It includes describing all aspects in the model cycle (utility of the model, datasets used, data preparation, spatial and temporal scales, algorithm used, model tuning, model evaluation, etc.). This metadata is usually text organized in a series of sections (model cycle) and items/tasks/questions that modelers have to populate.

Modelers

Modelers are usually quantitative/spatial analysts coders experts that design, develop, implement, test and report SDMs. They deliver spatially explicit predictions and their associated uncertainty of single or multiple species within a geographic area of interest. The predictions of SDMs are usually aimed to identify species potential distribution, habitat suitability, relative density, among others.

Species distribution model (SDMs)

SDMs are numerical tools that combine species occurrences (e.g., presence, presence/absence and counts) with covariates (environmental data sets related to species) to make spatially explicit predictions of habitat suitability or species density, using a variety of algorithms.