Online Program Home
  My Program

Abstract Details

Activity Number: 456 - Statistical Challenges and Opportunities for Supporting National Ecological Monitoring Programs
Type: Topic Contributed
Date/Time: Wednesday, August 2, 2017 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistics and the Environment
Abstract #324596 View Presentation
Title: Designing a Robust Monitoring Scheme for Natural Resources at a Continental Scale
Author(s): Emily L Weiser* and James E. Diffendorfer and Laura López-Hoffman and Darius J Semmens and Wayne E Thogmartin
Companies: U.S. Geological Survey and U.S. Geological Survey and The University of Arizona and U.S. Geological Survey and U.S. Geological Survey
Keywords: Study design ; Power analysis ; Spatial variation ; Temporal variation ; Population trend ; Sample size determination
Abstract:

To comprehensively measure population trends or other characteristics, a species must be monitored across its geographic range. Sampling locations must be spatially representative and sufficiently numerous to provide adequate statistical power. However, designing an adequate sampling strategy across a large spatial area is challenging. We develop a statistically robust framework for detecting population trends, using a case study of milkweed, which is a crucial resource for monarch butterflies. We illustrate two tools that address the challenges facing large-scale monitoring schemes: 1) a simulation-based power analysis to estimate the number of sampling locations needed for robust inference, and 2) Generalized Random Tessellation Stratified (GRTS) sampling to identify a list of spatially balanced sampling locations. Both tools are sufficiently flexible to be applied to a wide variety of organisms and study designs, and can be easily updated as additional information becomes available. Together, these tools provide a broadly applicable framework for designing a statistically robust, spatially balanced, and comprehensive sampling scheme at a continental scale.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2017 program

 
 
Copyright © American Statistical Association