Adaptive Management of Sea Lamprey

Using Science to Improve Assessment Methods for Fisheries

Mar 14, 2009

Using adaptive management, two scientists evaluated alternative larval lamprey assessment methods and improved the Great Lakes Sea Lamprey management program.

A primary objective of the Great Lakes Sea Lamprey Control program is to kill as many larval lampreys as possible before they migrate to the lakes and become fish parasites. Tributary streams are prioritized for treatment with lampricides based on estimates of the number of larval lamprey ready to enter the parasitic life stage.

Streams are prioritized for treatment using a complex two-step process of stream sampling and computer modeling. The program operates under a fixed budget and stream sampling costs detract from the budget available for stream treatment. A key management question, then, is how effective is the current assessment method and could an alternative, cheaper method work just as well, allowing more streams to be treated?

An Experiment to Compare Sea Lamprey Assessment Methods in the Great Lakes

Quantitative Assessment Sampling (QAS) Estimates the Absolute Abundance of Larval Lamprey

The current Quantitative Assessment Sampling (QAS) field program was intensive and statistically sound, yet uncertainties remain that impaired the overall effectiveness of the treatment program. One way to address this would be to increasing the number of samples collected during assessment, but because the lamprey control program has a fixed budget more samples would mean increased sampling costs and fewer streams treated.

Rapid Assessment (RA) Sampling Estimates the Relative Abundance of Larval Lamprey

A Rapid Assessment (RA) sampling program was the alternative approach. It is less desirable from a statistical perspective because it produces a less accurate index of larval abundance, rather than an absolute estimate of abundance. However, it is about 70% cheaper than QAS, thus the loss of accuracy might be compensated for by allowing more streams to be sampled during assessment and subsequently treated with lampricide for the same fixed budget.

Using Adaptive Management to Compare Larval Lamprey Assessment Methods

To test this hypothesis, Hansen and Jones implemented a large-scale adaptive management experiment over three years (2005 to 2007). The QAS and RA methods were applied to the same streams during the annual assessments of larval lamprey abundance. The data collected under each method were then used to estimate stream larval abundance and to rank streams for treatment based on the cost per larvae killed.

The Rapid Assessment Method was Cheaper, Produced Similar Stream Rankings, and Killed more Larval Lamprey

Hansen and Jones had expected a loss of accuracy using the RA method, but the results showed that the stream rankings of the two methods were similar. Additionally, if the RA method had been the real method it would have saved enough money to allow the treatment of 8-10 additional streams per year. Finally, RA resulted in at least as many and likely more sea lampreys killed for equal program costs.

Thus, though the RA approach ran counter to the demand for better science in lamprey population assessment it performed better than the existing assessment method. The high cost of natural resource management often requires considering non-intuitive solutions to achieve management objectives; better science may not always be the best approach. However, managers must first consider the value of information and adaptive management provides the rigorous framework within which to explore such questions.

The copyright of the article Adaptive Management of Sea Lamprey in Interdisciplinary Science is owned by Ian Parnell. Permission to republish Adaptive Management of Sea Lamprey in print or online must be granted by the author in writing.