Documentation"

In glossa: User-Friendly 'shiny' App for Bayesian Species Distribution Models

Quick reference guide

For more detailed guidance, visit the full documentation.

Key Tabs Overview

• Home: Start here to access tutorials, documentation, and the quickstart guide.
• New analysis: Set up your species distribution model. Upload data, select options, and run the analysis.
• Reports: View and explore analysis results with visualizations and metrics.
• Exports: Download your analysis results, including predictions and model data.

Data Requirements

1. Occurrences: Upload a CSV file with columns decimalLongitude, decimalLatitude, pa, and timestamp.
2. Environmental data: Upload a ZIP file containing raster files, organized by variable and time period.
3. Optional files:
4. Projection layers: ZIP files matching the structure of your environmental data for predictions.
5. Study area: A polygon file to define your study area and filter data.

Analysis options

• Model types:
• Native range: Includes environmental variables + spatial smoother.
• Suitable habitat: Includes only environmental variables.
• Functional responses: Computes response curves.
• Variable importance: Computed using a permutation-based approach.
• Cross-validation: Perform 5-fold cross-validation.

Advanced options

• Occurrences thinning: Apply spatial thinning based on coordinate precision (number of decimal digits).
• Standardize covariates: Z-score standardization is applied by subtracting the mean and dividing by the standard deviation.
• Enlarge polygon: Apply a buffer to expand the polygon if needed.
• Model selection: Currently, only BART is available.
• Set a seed: For reproducibility.

Viewing results

• Species selector: Choose the species to view results for.
• Key metrics: View suitable habitat area, mean probability, and number of points used.
• Prediction maps: Toggle between fitting layers and projection layers.
• Response curves and variable importance: Analyze the relationship between variables and species presence.

Exporting results

• Exports tab: Download prediction maps, model data, variable importance, and cross-validation results.

Need more help?

Visit the full documentation.

glossa documentation built on Oct. 15, 2024, 5:08 p.m.