Passive integrated transponder tags — called PIT tags — are battery-free electronic identification implants that are ideal for studying the life cycle of an anadromous fish. A review of the PIT-tagging work performed to track salmon and steelhead fish populations in t he Pacific Northwest shows the many uses of this technology.
By John R. Skalski
The Columbia—Snake River Basin — in Oregon, Washington, Idaho, Montana, Wyoming, Utah, and Nevada in the U.S. and the British Columbia in Canada — contains 136 hydropower projects. This consists of 58 dams built specifically for hydro production (see Figure 1) and 78 multipurpose dams that include hydro generation. Total hydro capacity from these projects is about 33,000 MW, and hydropower produced by these projects supplies 60 to 70 percent of the electrical needs of the Pacific Northwest.
In addition to this hydroelectric resource, the Columbia Basin is renowned for its salmon and steelhead fish populations. These salmonid stocks are a living symbol of the Pacific Northwest. These anadromous fish hatch in the basin, migrate to the ocean as juveniles, and return to the basin as adults one to four years later. However, industrial, agricultural, and urban development in the region have contributed to the near century-long decline in these fish. A number of these fish species are listed as threatened or endangered under the Endangered Species Act (ESA). The area’s hydro projects have contributed to these declines by impeding juvenile and adult migration, by direct mortality to juveniles passing through turbines on their way to the ocean, and by changes in the river environment due to impounded river waters and altered flow regimes.
In 1980, the U.S. Congress passed the Pacific Northwest Electric Power Planning and Conservation Act (Northwest Power Act), which charges federal agencies to ensure both hydro production and the restoration and protection of fish populations. Thirty-one federal hydro projects in this basin are coordinated under the act. The “action agencies” responsible for implementing mitigation activities to help restore the salmonid populations are: the Bonneville Power Administration (BPA), which markets and distributes the power from these projects, and U.S. Department of the Interior’s Bureau of Reclamation and U.S. Army Corps of Engineers, which own and operate the dams.
Since the Northwest Power Act was passed, more fish have been tagged and released in the Columbia Basin than anywhere else in the world. Currently, more than 20 million coded wire tags, 2 million passive integrated transponder (PIT) tags, and 30,000 to 40,000 radio and acoustic tags are used to mark juvenile salmonids for research and monitoring annually. Federal, state, tribal, and utility managers use the results from this cumulative effort to evaluate mitigation efforts at hydro projects, to manage fishery harvests, and to monitor stock recoveries. But the primary tool in all these efforts is the tiny PIT tag.
Historical use of PIT tags in the Columbia Basin
The PIT tag is small — only 12 millimeters long, 2 millimeters in diameter, and one tenth of a gram in weight. This passive, glass-encapsulated microchip is injected in the abdominal cavity of a fish. The tag is only detected, and its unique identification code transmitted, when it passes through a magnetic field. Detection systems have been installed within the juvenile bypass systems at seven federally owned dams in the Columbia River Basin that impound water for hydro projects. These are: 810-MW Lower Granite, 810-MW Little Goose, 810-MW Lower Monumental, and 603-MW Ice Harbor on the Snake River, and 980-MW McNary, 2,160-MW John Day, and 1,090-MW Bonneville on the Columbia River. The fish bypass systems are intended to reroute salmonid smolts away from the turbines as they outmigrate to the ocean.
At four of these projects (Bonneville, Ice Harbor, Lower Granite, and McNary) and at four projects owned by public utility districts (907-MW Priest Rapids, 624-MW Rock Island, 1,236.6-MW Rocky Reach, and 774-MW Wells on the Columbia River), PIT-tag detection systems also have been installed in adult fish ladders and fish counting windows. These systems are used to detect returning adult salmonids migrating upstream that were tagged as juveniles. At all of these detection locations, redundant detectors are used to ensure virtually 100 percent detection, regardless of fish density. At hydro projects with smolt transportation facilities, PIT-tagged fish that enter the barges destined for re-release below Bonneville Dam also are detected and recorded.
Together, the juvenile and adult detections provide valuable information on migratory success and passage through the hydro system. Initial PIT-tag loss in salmonid smolts can be as small as 0.2 percent, but recent evidence indicates tag loss through adulthood may be as high as 8 percent.1 The effect of these tag losses depends on what performance measures are estimated and how the tagging studies are designed.
What binds these individual detection sites into a cohesive information network is an automated detection system and database managed by the Pacific States Marine Fisheries Commission. Release and detection information on PIT-tagged fish is made available in near-real-time on a publicly available website, Pit Tag Information System (PTAGIS) at www.ptagis.org. Consequently, the tagging data are readily accessible to investigators interested in their study results, as well as managers making decisions on how best to operate the hydro system for the benefit of both power production and fish protection.
With both wild and hatchery raised salmonid smolts being tagged and released from more than 600 locations in the Columbia Basin, the PTAGIS database provides a wealth of information for real-time and retrospective inquiries. Three of the many uses of PIT tags in monitoring and evaluating salmonid stocks in the basin are described in the following paragraphs.
Predicting migration timing
During the spring and summer out-migration of juvenile salmon and steelhead, water is diverted from turbines and flow through surface spillbays increased to help move fish through the hydro system quicker and safer. One use of the PIT-tag database is to predict in real time the downstream migration timing. This allows hydro project owners/operators to optimize the benefits of spill programs. For example, certain habitat conservation plans specify spill levels during the middle 95 percent of the fish outmigration. This specification requires spill between the times when 2.5 percent and 97.5 percent of the total run pass a dam.
Program RealTime at the University of Washington, for instance, provides in-season predictions of the percent passage of index stocks on a daily basis via the Internet at www.cbr.washington/ edu/rt. Under these circumstances, Program RealTime predicts the migration timing based on pattern recognition software that uses both historical and in-season PIT-tag information. The migration timing of more than 50 PIT-tagged stocks of wild chinook salmon, steelhead, and sockeye salmon are tracked as they pass through as many as seven federally owned dams on their way to the ocean. Dam managers can use this information to time additional spill to facilitate fish passage.
Estimating passage survival
Recovery of ESA-listed Columbia Basin salmonid stocks depends on increasing the productivity and survival of these species. To this end, a variety of PIT-tag release and re-detection studies are performed annually by federal, state, and tribal agencies and local utility districts to estimate survival during the juvenile outmigration of salmonids. At individual hydro projects, paired upstream-downstream releases of tens of thousands of PIT-tagged smolts are used to isolate and estimate passage survival. These release and re-detection investigations are offshoots of tagging methods initially developed to study bird populations and subsequently adapted to fish investigations in the 1980s.2 The studies consist of concurrent releases of PIT-tagged fish above and below the river reach of interest, with at least two downstream detection sites. The common statistical standard for such project survival estimates is to be within ±0.05 of the true probability of survival, 95 percent of the time (i.e., standard error of 0.025).
Special applications of these paired release and re-detection designs include estimating smolt passage survival through a turbine unit or spillbay at a dam, comparing adult returns of transported and nontransported smolts, or comparing alternative hatchery rearing practices. In most applications, the survival studies are replicated in-season and often across years to ensure reliable conclusions over a meaningful range of environmental conditions. Investigators can use publicly available software developed by the University of Washington, called Program SampleSize, to help determine required release sizes and numbers of replicates.3
Life history monitoring
Among the earliest PIT-tag investigations were single releases of tagged fish to estimate smolt survival between dams. The first formal investigation was at Lower Granite Dam in 1993. By the late 1990s, investigators throughout the Snake River Basin were taking advantage of the fixed-location PIT-tag detection facilities to evaluate survival of both wild and hatchery stocks.
The continuous flow of tags and the daily detections throughout the outmigration season soon allowed collection of season-wide estimates of survival, as well as comparison of within- and between-year patterns in survival. Sophisticated tagging models were developed — at the University of Washington and elsewhere — to analyze this vast information and relate survival to environmental and hydro project conditions. Fisheries scientists also took their cues from health sciences and epidemiology and began relating the downstream fate of individual smolts to their unique characteristics, such as size and rearing history.4 When it comes to smolt survival, it appears that size matters. In one study, for example, survival increased more than twofold for the largest smolt in a release group.
Perhaps the greatest advance in PIT-tag technology occurred when the hydro system was equipped with both juvenile and adult PIT-tag detectors. The first adult PIT-tag detectors were installed at Lower Granite in 1988. Adult detectors followed at seven other dams. This system of juvenile and adult PIT-tag detections permits near complete life-cycle information, allowing researchers to partition total mortality into in-river juvenile, ocean (i.e., from juvenile passage at Bonneville Dam to adult return back to Bonneville), and upriver adult components.
Because some PIT-tagged smolts are barged down river as part of the smolt transportation program, the efficacy of this program also can be evaluated. As expected, the benefits of smolt transportation increase as the number of hydro projects traversed becomes greater. From 1997 to 2003 (excluding 2001), the transport system increased Snake River spring chinook salmon adult returns by an average of 15.7 percent. In 2001, when river flows were greatly reduced because of drought, the relative benefit of the transport system was a 515 percent increase in adult returns.5 Publicly available software such as Program ROSTER allows individuals and agencies to make their own assessments of juvenile and adult survivals and transportation effects from the tens of millions of PIT-tag records currently available at PTAGIS.6
Future work planned
The recent miniaturization and extended battery life of active acoustic tags are changing the face of tagging studies in the Columbia—Snake River Basin by allowing fisheries biologists to examine in fine detail the movement of fish in and around hydro projects. However, these new technologies cannot provide the multi-year, population-wide information extracted from the innocuous and inexpensive PIT tag.
The federal government and the private sector are developing innovative detection equipment to extend the applications of PIT tags. For example, in the near future, new detection equipment at the mouth of the Columbia River tributaries will be capable of recording adult PIT-tagged salmon as they enter spawning streams. Researchers will be able to use these detections to estimate both escapement and straying rates of adult salmon.
Detection equipment also is being modified to work at sluiceways and overspill weirs at dams. These advances will increase downstream detection rates and permit detection through non-bypass routes at a dam. Installation of detection equipment at weirs also is allowing investigations to examine movement and residence time of salmonid smolts in estuary environments near the mouth of the Columbia River. Such information is helping researchers to determine the benefits of estuary restoration efforts in tidal freshwater environments.
Dr. Skalski may be reached at University of Washington, 1325 Fourth Avenue, Suite 1820, Seattle, WA 98101; (1) 206-616-4851; E-mail: email@example.com. edu.
- Buchanan, Rebecca B., Personal Communication, 2008.
- Burnham, K.P., et al, “Design and Analysis Methods for Fish Survival Experiments based on Release-Recapture,” American Fisheries Society Monographs No. 5, 1987.
- www.cbr.washington.edu/paramest/ samplesize
- Program SURPH, www.cbr.washington. edu/paramest/surph
- Buchanan, Rebecca B., and John R. Skalski, “A Migratory Life-Cycle Release-Recapture Model for Salmonid PIT-Tag Investigations,” Journal of Agricultural, Biological, and Environmental Statistics, Volume 12, No. 3, September 2007, pages 325-345.
- www.cbr.washington.edu/paramest/ roster
John Skalski, PhD, is professor of biological statistics in the School of Aquatic & Fishery Sciences at the University of Washington. He has 30 years of experience in fish and wildlife tagging studies.
µ Peer Reviewed
This article has been evaluated and edited in accordance with reviews conducted by two or more professionals who have relevant expertise. These peer reviewers judge manuscripts for technical accuracy, usefulness, and overall importance within the hydroelectric industry.