Closing the gap between statistical and scientific workflows for improved forecasts in ecology.

Area of Science:

• Ecology
• Ecological Modeling
• Statistical Workflow

Background:

• Increasing biodiversity loss and climate change necessitate robust ecological models.
• Ecological datasets are growing in complexity (geographical, temporal, phylogenetic).
• Current modeling workflows hinder evaluation and comparison of ecological models.

Purpose of the Study:

• To address challenges in ecological model fitting and evaluation.
• To propose an integrated workflow for trend estimation and forecasting.
• To improve the accuracy and applicability of ecological models for policy.

Main Methods:

• Proposing a workflow integrating statistical and scientific practices.
• Utilizing data simulation to inform modeling decisions.
• Advocating for open model sharing and common datasets.

Main Results:

• The proposed workflow connects trend estimation and forecasting.
• Data simulation aids in decision-making throughout the modeling process.
• The approach facilitates better-informed sustainable policies.

Conclusions:

• An integrated statistical and scientific workflow is crucial for modern ecological modeling.
• Harmonizing trend estimation and forecasting efforts enhances policy relevance.
• Universal training and open data sharing are key to advancing ecological modeling.