kuenm
kuenm facilitates development and calibration of ecological niche models (ENMs) using Maxent for species distribution modeling and extrapolation risk assessment.
Key Features:
- Detailed Model Calibration: Implements calibration workflows that select optimal parameters based on statistical significance, predictive power, and model complexity.
- Model Creation and Evaluation: Generates final models from multiple parameter sets and enables model transfers for projection to other scenarios.
- Extrapolation Risk Analysis: Evaluates extrapolation risk in model transfers using the mobility-oriented parity (MOP) metric and the MESS (Minimum Estimated Sample Size) metric to assess strict-extrapolation risks.
- Reproducibility and Efficiency: Integrates with R and Maxent to enable reproducible model calibration and evaluation and to accelerate calibration and transfers across scenarios.
Scientific Applications:
- Biodiversity Conservation: Predicts species distributions under various environmental conditions to inform conservation planning.
- Climate Change Impact Studies: Assesses potential impacts of climate change on species habitats by evaluating extrapolation risks during model transfers.
- Ecological Research: Supports development of niche models for studying ecological dynamics and species interactions.
Methodology:
Performs systematic model calibration, selection, and evaluation with parameter optimization based on statistical significance, predictive power, and model complexity; creates final models from multiple parameter sets and conducts model transfers; evaluates extrapolation risk using MOP and MESS; implemented in R and interfacing with Maxent.
Topics
Details
- Maturity:
- Mature
- Cost:
- Free of charge
- Tool Type:
- library
- Operating Systems:
- Linux, Windows, Mac
- Programming Languages:
- R
- Added:
- 5/19/2019
- Last Updated:
- 6/16/2020
Operations
Publications
Cobos ME, Peterson AT, Barve N, Osorio-Olvera L. kuenm: an R package for detailed development of ecological niche models using Maxent. PeerJ. 2019;7:e6281. doi:10.7717/peerj.6281. PMID:30755826. PMCID:PMC6368831.