Record Details
Comparison of methods to estimate population density of Pacific giant salamanders in small streams of the southern Oregon Cascades
ScholarsArchive at Oregon State University
| Field | Value |
|---|---|
| Title | Comparison of methods to estimate population density of Pacific giant salamanders in small streams of the southern Oregon Cascades |
| Names |
Leuthold, Niels C.
(creator) Gregory, Stanley V. (advisor) |
| Date Issued | 2003-08-11 (iso8601) |
| Note | Graduation date: 2004 |
| Abstract | I compared hand sampling, two-pass multiple removal sampling, mark-recapture and catchability-based population estimates for the Pacific giant salamander (Dicamptodon tenebrosus) at 22 sites in 11 streams of Cascade Mountains of Oregon. Mark-recapture and catchability-based population estimates were not significantly different (p 0.86). Hand sampling and two-pass multiple removal population estimates were not significantly different (p = 0.57). However, mark-recapture and catchability-based estimates were significantly greater than two-pass multiple removal and hand sampling estimates. Hand sampling and multiple removal population estimates were frequently lower than the number of individual Pacific giant salamanders captured by all methods at a site. The catchability of each method differed between streams and within consecutive sites on the same stream. Unacknowledged differences in catchabilities among sites are a potential source of error when using hand-sampling index counts to estimate and compare populations without correcting the estimates for the differences in Pacific giant salamander catchability. Catchability decreased significantly between the first pass of electroshocking and the subsequent recapture pass of electroshocking. This decrease was observed even though the recapture passes were performed the day after the multiple removal electroshocking. A decrease in catchability violates the assumptions of constant catchability for multiple removal population estimates and potentially explains the weaker performance of removal estimates. Hand sampling estimates did not adjust for the proportion of the population captured at each site, and this is probably the cause of their poor performance. The presumption of constant catchability in hand sampling produced biases in the population estimates, because the catchabilities were not constant. In future studies of stream amphibian abundance, catchability needs to be included in population estimation procedures to produce accurate estimates and to allow valid comparisons of population sizes between sites. Catchability models can be used to calibrate less intensive survey methods, such as hand sampling or a single pass of electroshocking, with the results from more intensive mark-recapture methods. Intensive work would be needed to do the calibrations, but afterwards a standard, more convenient method, such as electroshocking or hand sampling, can be used within the ranges of habitat values for which the calibration model is valid. |
| Genre | Thesis/Dissertation |
| Topic | Pacific giant salamander -- Counting -- Oregon -- Little River (Douglas County) |
| Identifier | http://hdl.handle.net/1957/20822 |
