SSNdesign-An R package for pseudo-Bayesian optimal and adaptive sampling designs on stream networks

Area of Science:

• Ecology
• Environmental Science
• Computational Statistics

Background:

• Freshwater ecosystems provide vital services and require monitoring for health assessment.
• Current monitoring relies on costly field sampling and in-situ sensors, necessitating efficient survey designs.
• Geostatistics and optimal design theory offer potential for optimizing sampling, but face challenges in stream networks.

Purpose of the Study:

• To develop and present SSNdesign, an open-source R package for optimal and adaptive sampling design on stream networks.
• To address the lack of accessible software for applying advanced geostatistical and design theories in aquatic monitoring.
• To demonstrate the effectiveness of SSNdesign in improving the efficiency of freshwater monitoring.

Main Methods:

• Developed SSNdesign, an R package integrating geostatistical modeling and optimal/adaptive design theory for stream networks.
• Utilized mathematical foundations for design optimization on branching, directional stream networks.
• Applied SSNdesign to two case studies with real-world data from Queensland, Australia.

Main Results:

• SSNdesign successfully implemented optimal and adaptive sampling designs on stream networks.
• Case studies demonstrated that SSNdesign's optimal/adaptive designs significantly outperformed random and spatially balanced designs.
• The package integrates seamlessly with existing open-source software for broader accessibility.

Conclusions:

• SSNdesign provides a powerful, open-source tool for optimizing freshwater monitoring efforts.
• The package enhances the efficiency and cost-effectiveness of aquatic biodiversity and condition assessments.
• SSNdesign has the potential to significantly improve freshwater conservation and management strategies.