This file includes R code to select annual sampling units for PRISM surveys on the Arctic National Wildlife Refuge using a stratified random sampling approach.
We used the Generalized Random Tessellation Stratified (GRTS) procedure (Stevens and Olsen 2004) to obtain a spatially balanced sample of 108 plots within each of the four strata to be surveyed. To determine how to allocate the 108 plots among strata, we initially used shorebird densities and variances (averaged across species) presented in Brown et al. (2007), but adjusted these values to include additional upland and riparian plots. Final allocation included 45 wetland plots, 35 moist plots, 12 upland plots, and 16 riparian plots. We also selected an additional 55 “oversample” plots (in the same proportion as the primary allocation) to replace plots that were found to be inaccessible during field operations.
As an objective of this survey was to obtain an estimate of population change since original surveys were conducted in 2002 and 2004, a portion of plots sampled in 2002 and 2004 were selected to be re-surveyed. As plots were surveyed in triplicates in 2002 and 2004, we first selected one plot per triplicate to re-survey in each year, removing plots that were not completely surveyed, exact plot boundaries were unknown, or were on non-federal lands. This resulted in the selection of 54 plots to be re-surveyed in 2019 and 52 plots in 2022. To maintain a spatially-balanced design, for each re-surveyed plot, we removed the closest plot from the GRTS selection with the same habitat stratum. Final plot allocation included: wetland (2019: 30 new, 15 repeat plots; 2022: 28 new, 17 repeat), moist (2019: 20 new, 15 repeat plots; 2022:19 new, 16 repeat), upland (2019: 0 new, 12 repeat plots; 2022: 2 new, 10 repeat), and riparian (2019: 4 new, 12 repeat, 2022: 7 new, 9 repeat).
The data providers have invested considerable effort in QA/QC, but it is possible that undetected errors remain. It is strongly recommended that careful attention be paid to the contents of the metadata file associated with these data to evaluate data set limitations, restrictions or intended use. The originators of this dataset shall not be held liable for improper or incorrect use of the data described and/or contained herein.
Potential users of these data should first contact the data authors listed below, as potential biases may occur within the data depending on the intended use. Therefore, potential users should contact the data authors to discuss intended use and how these data may or may not be applicable.
see 1) Saalfeld, S. T., H. R. Gates, S. C. Brown, K. M. Sowl, B. A. Andres, J. A. Johnson, J. E. Lyons, B. J. McCaffery, and R. B. Lanctot. 2022. Survey Protocol for the Arctic Program for Regional and International Shorebird Monitoring (PRISM). Version 1.0. U.S. Fish and Wildlife Service, Division of Migratory Bird Management, Anchorage, AK and 2) annual field protocol documents for specific methodologies.
