Estimating population size and observation bias for spring Chinook Salmon

Abstract

The primary conservation prioritization tool for spring Chinook Salmon (Oncorhynchus tshawytscha), population viability analysis, is often conducted with biased spawner abundance data with no associated statistical uncertainty or error. This study estimated observation error of surveyors counting redds in two spring Chinook Salmon populations where hatchery supplementation is implemented as a conservation tool. Habitat complexity, redd density and the amount of observer experience were important in estimating error rates. Increases in both habitat complexity and experience reduced net error rates. Conversely, net error rates increased as redd density increased. Unbiased estimates of redd abundance were generated and converted to spawner abundance using population specific redd expansion factors. The precision (i.e., coefficient of variation [CV]) of spawner abundance estimates were similar in the Wenatchee (natural CV = 5%; hatchery CV = 6%) and Methow (natural CV = 5%; hatchery CV = 2%) watersheds because average net error rates were similar (Wenatchee = −0.1512; Methow = −0.1748). This study addresses a criticism of population viability analysis (i.e., parameter uncertainty) that should result in more scientifically defensible conservation priorities and recommendations that can be implemented with greater certainty.

Identification of salmon redds using RPV‐based imagery produces comparable estimates to ground counts with high inter‐observer variability

“…Our data provides little evidence as to which survey method is better able to identify redds in poor viewing conditions. redd identification (Dunham et al, 2001;Howell & Sankovich, 2012;Muhlfeld et al, 2006;Murdoch et al, 2019), and was found prevalent within our study (CV of 37-50%). Inconsistency among aerial observers poses a significant challenge for redd counting.…”

“…Enumerating the spawning population size of salmonids is critical for informing environmental management decisions and complying with policy requirements. Redd counts remain a common and cost‐effective method for assessing population status and trends despite evidence of bias and uncertainty in visual cues of a redd (Gallagher & Gallagher, 2005; Murdoch et al, 2019). Our research demonstrates that RPV‐based aerial imagery can be used to both identify and enumerate salmonid redds under certain stream and environmental conditions, namely water clarity, redd size, and stream width.…”

“…In scenarios where the river is bifurcated around an island or side channel, one technician will float one channel while the other technician walks the other channel. It is important to note that WDFW conducted previous studies to quantify the bias and develop efficient monitoring protocols (Murdoch et al, 2019). Therefore, for this study, there is only a single estimate of redd counts per survey.…”

“…Redds are identified and counted by observers on field surveys from watercraft or on foot throughout the spawning season. Ground observer variability has shown to be large and is influenced by multiple factors, including observer experience, river flow, habitat complexity, and redd density (Dauphin, Prévost, Adams, & Boylan, 2010; Dunham, Davis, & Rieman, 2001; Muhlfeld, Taper, Staples, & Shepard, 2006; Murdoch, Frady, Hughes, & See, 2019). In a similar system, Murdoch et al (2019) report a 51% coefficient of variation among ground observers.…”

“…Ground observer variability has shown to be large and is influenced by multiple factors, including observer experience, river flow, habitat complexity, and redd density (Dauphin, Prévost, Adams, & Boylan, 2010; Dunham, Davis, & Rieman, 2001; Muhlfeld, Taper, Staples, & Shepard, 2006; Murdoch, Frady, Hughes, & See, 2019). In a similar system, Murdoch et al (2019) report a 51% coefficient of variation among ground observers. Muhlfeld et al (2006), Howell and Sankovich (2012), and Dunham et al (2001) report a wide range of CV values across streams in bull trout redd surveys (11%, 20%, and 25%, 49%, and 37%, 78%).…”