Sampling bias exaggerates a textbook example of a trophic cascade

• Main Authors: Brice, Elaine, Larsen, Eric, MacNulty, Daniel
• Format: Software
• Language: unknown
• Published: 2022
• Subjects: Aspen; Populus tremuloides; elk; Cervus canadensis; wolf; Canis lupus; carnivore; predator indirect effects; nonrandom sampling; sampling bias; preferred browsing height; Ungulate
• Online Access: https://zenodo.org/record/5637263; https://doi.org/10.5281/zenodo.5637263

Understanding trophic cascades in terrestrial wildlife communities is a major challenge because these systems are difficult to sample properly. We show how a tradition of nonrandom sampling has confounded this understanding in a textbook system (Yellowstone National Park) where carnivore [Canis lupus (wolf)] recovery is associated with a trophic cascade involving changes in herbivore [Cervus canadensis (elk)] behavior and density that promote plant regeneration. Long-term data indicate a practice of sampling only the tallest young plants overestimated regeneration of overstory aspen (Populus tremuloides) by a factor of 3-8 compared to random sampling because it favored plants taller than the preferred browsing height of elk and overlooked non-regenerating aspen stands. Random sampling described a trophic cascade, but it was weaker than the one that nonrandom sampling described. Our findings highlight the critical importance of basic sampling principles (e.g., randomization) for achieving an accurate understanding of trophic cascades in terrestrial wildlife systems. This dataset has 18,792 records, including 18,623 records of individual young aspen (plants > 1 year-old & < 600 cm) and 169 records of plots with no young aspen ("zero plots"). Records of individual young aspen (N = 18,623) were used in the majority of analyses (Fig. 2-5a,b), including generalized linear mixed models (GLMMs) that tested how the effect of year on browsing, height, and recruitment of stems differed by sampling method. This dataset was also used to model the effect of stem height on browsing to estimate the preferred browse height (PBH) and browse escape height (BEH). The full dataset that includes plots with no young aspen was only used to calculate the percentage of plots and stands each year with median heights greater than 200 (Fig. 5c) or 300 cm (Fig. 5d). The dataframe has the following 6 columns: Plot: individual identifier for each of 113 plots distributed randomly across the study area. Each plot was a 1 × 20 m belt ...