Improving Distribution Prediction by Integrating Expert Range Maps and Opportunistic Occurrences: Evidence From...

Area of Science:

• Marine ecology
• Biodiversity assessment
• Conservation science

Background:

• Accurate biodiversity assessments are crucial for conservation, especially during the current biodiversity crisis.
• Species distribution models (SDMs) are vital tools for biodiversity assessment but suffer from prediction inaccuracies due to incomplete or biased species distribution data.
• Integrating diverse data types to enhance SDM predictions is a proposed solution, yet underexplored in marine environments.

Purpose of the Study:

• To investigate the effectiveness of integrating different data types for improving marine species distribution models (SDMs).
• To assess the utility of expert range maps in conjunction with opportunistic occurrence data for enhancing SDM predictions.
• To evaluate the impact of data integration on the reliability of spatial distribution predictions for the Japanese sea cucumber.

Main Methods:

• Fitted species distribution models (SDMs) for the Japanese sea cucumber using opportunistic occurrence records and four distinct modeling algorithms.
• Developed two ensemble models via stacked generalization: one solely based on the four SDM predictions, and another incorporating an expert-informed range map (IUCN).
• Compared the predictive performance of the integrated models against the baseline SDMs and the expert-informed ensemble model.

Main Results:

• The expert-informed ensemble model demonstrated improved distribution prediction accuracy compared to models using only opportunistic data.
• Integration of the expert range map effectively prevented overprediction in areas south of the Yangtze River's freshwater discharge, a known dispersal barrier.
• The study confirmed that combining expert knowledge with opportunistic data significantly enhances the reliability of marine species' spatial distribution predictions.

Conclusions:

• Integrating expert range maps into species distribution models (SDMs) significantly improves the accuracy and reliability of marine biodiversity assessments.
• This data integration approach offers a valuable strategy for overcoming limitations of opportunistic data in SDMs, leading to better-informed conservation planning.
• The findings underscore the importance of leveraging expert knowledge to refine ecological models and enhance predictions in data-scarce marine ecosystems.