9710 - Laserfiche WebLink

Home Browse Search

406 CONNOLLY ET AL. four comers directly to the stream substrate, and thus they were horizontal to stream flow. This orientation differs from Zydlewski et aI.'s (2006) "swim-through" antennas and from Armstrong et aI.'s (1996) and Greenberg and Giller's (2000) "flat plate" design. We refer to this antenna orientation as "pass-by." We prefer this generic term, rather than "swim-by," because the use of these antennas are applicable to other PIT-tagged animals and objects that mayor may not swim (e.g., tagged rocks for streambed movement studies). While a tagged fish could pass over or under a pass-by antenna, it could also weave through the opening within the rectangular frame of the antenna. In array B of our interrogation systems, we used two so- called "hybrid" antennas capable of pivoting in the water column as depth increased. These hybrid antennas had only the upstream side of the antenna attached to the substrate at two or more pivot points, thus enabling the downstream side of the antenna to float in the water column. As water depth changed, the antenna changed its angle in the water column in reference to its attached upstream side. This hybrid antenna was often in a "pass-through" orientation (i.e., vertical to the flow), but would be in a pass-by position during extremes of low flow because of lack of water to float the downstream edge, and during extremes of high flow because high velocity forced the floating edge downward. Fiber optic cables were installed for data transfer from the trdllsceiver to a computer housed in an existing building on site. TIle computer recorded detection data using the MiniMon program available through the Pacific States Marine Fisheries Commis- sion (PSMFC, Portland, Oregon). MiniMon configured the data to a format for loading into a regional database (Pr AGlS) maintained by PSMFC. We queried the PrAGIS database for detection data that were to be used in subsequent analyses. In September 2004, we installed a similar PIT tag interrogation system in Beaver Creek. A site was selected where three antenna arrays could be placed within 30.5 m of the transceiver and where two antennas would span the wetted width at most flows. The antennas were placed in the tail-out of pools and in shallow riffle areas. We tied the antennas to metal stakes driven vertically into the streambed, which consisted of cobble and gravel. At the upper most array (array A), we insta1led a 1.8-m X 0.9-m antenna (number I) on river left and a 3.1-m X 0.9-m antenna (number 2) on river right (Figure 2). At the middle array (array B) we installed two 3.l-m X 0.9-m antennas (numbers 3 and 4), and for the downstream array (array C) we installed two 1.8-m X 0.9-m antennas (numbers 5 and 6). As described previously for Rattlesnake Creek, arrays A andC were anchored to the stream on all four comers in a pass-by configuration while array B was installed in a hybrid configuration. A Digital Angel FSlOOIM transceiver was used to operate the six antennas. This transceiver was attached to a bank of four 12- V batteries to provide 24- V DC power to the transceiver. The batteries were exchanged on a regular basis (about every 5-7 d depending on factors such as ambient weather and transceiver settings). In addition, a lland-held computer was used to record the data from the transceiver using a Mobile Monitor program (available through PSMFC). All equipment was installed in a 1.2-m X l.2-m box placed underground to decrease exposure to high heat and excessive cold. Detection efficiency calculations.-We evaluated the interrogators' detection efficiencies over the biologically impoTh11lt increments of low- and high- tlow periods while differentiating between upstream and downstream movement. Depending on the tuning and power setting of the transceiver and the particular antenna, the read distance above the pass-by antennas for a 12.5-mm, 134.2-kHz ST PIT tag ranged up to 45 cm. Under normal operating conditions, any tag passing through the rectangular openings ofthe hybrid antennas had the potential to be read by the interrogator, but factors such as tag orientation (Zydlewski et aI. 2006) and the presence of another tag (Greenberg and Giller 2000) could decrease this potential. When tag-reading ability dipped below 10 cm from a pass-by antenna or when a tag was not read pa~sing through a hybrid antenna, we modified or replaced the antenna. The incremental change in the interrogation system's efficiency when an individual antenna's tag-reading ability changed was not evalu- ated. To do so would not likely mimic a practical field practice for most applications. The PIT tag interrogation system operated almost continuously from 4 November 2003 to 6 October 2005 in Rattlesnake Creek and from 27 September 2004 to 15 May 2006 in Beaver Creek. The detection data were used to calculate detection efficiencies of the individual interrogation systems. The data were sorted into upstream- and downstream-moving fish based on time of detection at two or more arrays. If a fish was detected at a single array, it was often possible to detennine direction of movement based on the location of its last detection (i.e., upstream or downstream of the interrogation system). To distinguish between low and high flow, we used stage-discharge relationships and information abont the read ranges of the PIT tags. In each stream, stage- discharge data were available from gauges just upstream of the PIT tag interrogator. In Rattlesnake