Assessment of culvert passage of Yellowstone cutthroat trout in a Yellowstone River spawning tributary using a passive integrated transponder system

Abstract

Road crossing culverts can pose passage barriers during salmonid migration. However, existing models to predict passage have not been sufficiently tested, particularly for non-anadromous species. In addition, most existing tools for evaluating culvert passage status categorize culverts as passable or impassable whereas an estimate of the probability of passage under a range of flows and hydraulic conditions would be more useful for assessing barrier status. This study used Half-duplex (HDX) Passive Integrated Transponder (PIT) tags and PIT tag reading antennas to examine the ability of migrating fluvial-adfluvial Yellowstone cutthroat trout Oncorhynchus clarkii bouvieri to pass through three distinct culvert types over a variety of hydraulic conditions. PIT tag detecting antennas were installed on the upper and lower ends of each culvert and below the plunge pool downstream of the culvert. This design allowed determination of (1) passage success or failure; (2) the number of passage attempts made at each culvert; and (3) the time required for passage through each culvert; and (4) the number of tagged individuals that attempted passage at a particular culvert.

Factors shown to increase the probability of passage included shorter fish length, higher water temperature, lower water velocity, and small outlet drop height. Stepwise logistic regression was used to create a model that contained water velocity alone that correctly predicted the probability of passage in 88.9% of cases (r2 = 0.91, P < 0.001). Box culverts with baffles had significantly higher passage rates and lower velocities that paralleled that observed in a natural stream channel. Congruency of observed passage with that predicted by the FishXing model was generally low (50%) suggesting the need for further field validation of the model. Study results indicated that 1) PIT tag detection systems offer distinct advantages for assessing culvert fish passage but that further testing of antenna designs is needed to increase efficacy of their use; 2) addition of baffles can increase culvert passage rates; and 3) probability of passage curves developed in this study offer a useful tool for assessing barrier status of individual culverts and estimating the need for culvert removal or retrofitting.