Michael F. Fodale

Techniques and methods for estimating abundance of larval and metamorphosed sea lampreys in Great Lakes tributaries, 1995 to 2001

Abstract Before 1995, Great Lakes streams were selected for lampricide treatment based primarily on qualitative measures of the relative abundance of larval sea lampreys, Petromyzon marinus. New integrated pest management approaches required standardized quantitative measures of sea lamprey. This paper evaluates historical larval assessment techniques and data and describes how new standardized methods for estimating abundance of larval and metamorphosed sea lampreys were developed and implemented. These new methods have been used to estimate larval and metamorphosed sea lamprey abundance in about 100 Great Lakes streams annually and to rank them for lampricide treatment since 1995. Implementation of these methods has provided a quantitative means of selecting streams for treatment based on treatment cost and estimated production of metamorphosed sea lampreys, provided managers with a tool to estimate potential recruitment of sea lampreys to the Great Lakes and the ability to measure the potential consequences of not treating streams, resulting in a more justifiable allocation of resources. The empirical data produced can also be used to simulate the impacts of various control scenarios.

Selecting Great Lakes Streams for Lampricide Treatment Based On Larval Sea Lamprey Surveys

The Empiric Stream Treatment Ranking (ESTR) system is a data-driven, model-based, decision tool for selecting Great Lakes streams for treatment with lampricide, based on estimates from larval sea lamprey (Petromyzon marinus) surveys conducted throughout the basin. The 2000 ESTR system was described and applied to larval assessment surveys conducted from 1996 to 1999. A comparative analysis of stream survey and selection data was conducted and improvements to the stream selection process were recommended. Streams were selected for treatment based on treatment cost, predicted treatment effectiveness, and the projected number of juvenile sea lampreys produced. On average, lampricide treatments were applied annually to 49 streams with 1,075 ha of larval habitat, killing 15 million larval and 514,000 juvenile sea lampreys at a total cost of

Compensatory Mechanisms in Great Lakes Sea Lamprey Populations: Implications for Alternative Control Strategies

5.3 million, and marginal and mean costs of

Estimating Lake-wide Abundance of Spawning-phase Sea Lampreys (Petromyzon marinus) in the Great Lakes: Extrapolating from Sampled Streams Using Regression Models

85 and

The Sea Lamprey in Lake Erie: a Case History

10 per juvenile killed. The numbers of juvenile sea lampreys killed for given treatment costs showed a pattern of diminishing returns with increasing investment. Of the streams selected for treatment, those with > 14 ha of larval habitat targeted 73% of the juvenile sea lampreys for 60% of the treatment cost. Suggested improvements to the ESTR system were to improve accuracy and precision of model estimates, account for uncertainty in estimates, include all potentially productive streams in the process (not just those surveyed in the current year), consider the value of all larvae killed during treatment (not just those predicted to metamorphose the following year), use lake-specific estimates of damage, and establish formal suppression targets.

Optimizing Larval Assessment to Support Sea Lamprey Control in the Great Lakes

Abstract Compensatory mechanisms are demographic processes that tend to increase population growth rates at lower population density. These processes will tend to reduce the effectiveness of actions that use controls on reproductive success to suppress sea lamprey ( Petromyzon marinus ), an economically important pest in the Great Lakes. Historical evidence for compensatory mechanisms in sea lamprey populations was reviewed, and revealed: (1) strong evidence for shifts in sex ratios as sea lamprey abundance was reduced in the early years of the control program; (2) weak and equivocal evidence for increased growth rates of sea lamprey cohorts re-colonizing streams following a lampricide treatment; and (3) suggestions of other compensatory processes, such as earlier ages at metamorphosis, but with little empirical evidence. Larval size distribution data for cohorts in the first and second years following a lampricide treatment (26 pairs of cohorts in 20 streams) was analyzed and did not indicate a consistent pattern of more rapid growth of the first colonizing cohort (only 11 of 33 cases). To test for compensation between spawning and age-1 in sea lamprey populations, data were analyzed for 49 stream-years for which spawning female abundance was known and age-1 abundance was estimated in the following year. A fit of these data to a Ricker stock-recruitment function showed evidence for compensation, measured as reduced survival to age 1 at higher abundance of spawning females. More obvious, however, was a large amount of density-independent variation in survival, which tends to mask evidence for compensatory survival. The results were applied to a simple model that simulates sea lamprey populations and their control in a hypothetical lake. Control strategies that targeted reproductive success performed far less well than comparable strategies that targeted larval populations, because density-independent recruitment variation leads to occasional strong year classes even when spawner abundance is reduced to low levels through alternative control. It is concluded that further study of recruitment variation in lamprey populations is critical to rationalizing alternative controls that target reproductive success, and that recruitment variation needs to be incorporated into models used to evaluate sea lamprey control options.

Planning and Executing a Lampricide Treatment of the St. Marys River Using Georeferenced Data

Lake-wide abundance of spawning-phase sea lampreys (Petromyzon marinus) can be used as one means to evaluate sea lamprey control efforts in the Great Lakes. Lake-wide abundance in each Great Lake was the sum of estimates for all streams thought to contribute substantial numbers of sea lampreys. A subset of these streams was sampled with traps and mark-recapture studies were conducted. When sea lampreys were captured in traps, but no mark-recapture study was conducted, abundance was estimated from a relation between trap catch and mark-recapture estimates observed in other years. In non-sampled streams, a regression model that used stream drainage area, geographic region, larval sea lamprey, production potential, the number of years since the last lampricide treatment, and spawning year was used to predict abundance of spawning-phase sea lampreys. The combination of estimates from sampled and non-sampled streams provided a 20-year time series of spawning-phase sea lamprey abundance estimates in the Great Lakes.

Effectiveness of Using Backpack Electrofishing Gear for Collecting Sea Lamprey (Petromyzon marinus) Larvae in Great Lakes Tributaries

Abstract Sea lampreys ( Petromyzon marinus ), first reported in Lake Erie in 1921, emigrated from Lake Ontario via the Welland Canal. It was not until the advent of pollution abatement, stream rehabilitation, and salmonid enhancement programs that sea lampreys proliferated. The Great Lakes Fishery Commission (GLFC), in co-operation with state, provincial, and federal fisheries agencies, implemented an integrated sea lamprey management (IMSL) plan for Lake Erie in 1986. Suppression of sea lampreys was nearly immediate, as indicated by declining larval-, parasitic-, and spawning-phase abundance, while survival of lake trout ( Salvelinus namaycush ) was markedly improved. Consistent with their vision statement, the GLFC began reducing lampricide use by the mid-1990s, while increasing reliance on alternative control methodologies. Reduction of treatment effort coincided with the development of new lampricide application techniques and treatment selection criteria, in addition to heightened regional concern for the impact of lampricide on non-target species. Subsequently, Lake Eries sea lamprey numbers have rebounded, and marking rates on lake trout have approached pre-control levels. It is hypothesized that Lake Eries rising abundance is primarily fuelled by untreated and residual larval populations, although some migration of parasitic-phase sea lampreys from Lake Huron is suspected. Model simulations infer that treatment effort on Lake Erie was sub-optimal from 1995 to 1998. Beginning in 1999, the GLFC enhanced measures to identify and control sources of sea lampreys. Based on historical abundance patterns and model results, it is anticipated that intensified management in Lake Erie will reduce sea lamprey numbers and provide an opportunity for lake trout restoration.

Assessing Assessment: Can the Expected Effects of the St. Marys River Sea Lamprey Control Strategy Be Detected?

Elements of the larval sea lamprey (Petromyzon marinus) assessment program that most strongly influence the chemical treatment program were analyzed, including selection of streams for larval surveys, allocation of sampling effort among stream reaches, allocation of sampling effort among habitat types, estimation of daily growth rates, and estimation of metamorphosis rates, to determine how uncertainty in each element influenced the stream selection program. First, the stream selection model based on current larval assessment sampling protocol significantly underestimated transforming sea lam-prey abundance, transforming sea lampreys killed, and marginal costs per sea lamprey killed, compared to a protocol that included more years of data (especially for large streams). Second, larval density in streams varied significantly with Type-I habitat area, but not with total area or reach length. Third, the ratio of larval density between Type-I and Type-II habitat varied significantly among streams, and that the optimal allocation of sampling effort varied with the proportion of habitat types and variability of larval density within each habitat. Fourth, mean length varied significantly among streams and years. Last, size at metamorphosis varied more among years than within or among regions and that metamorphosis varied significantly among streams within regions. Study results indicate that: (1) the stream selection model should be used to identify streams with potentially high residual populations of larval sea lampreys; (2) larval sampling in Type-II habitat should be initiated in all streams by increasing sampling in Type-II habitat to 50% of the sampling effort in Type-I habitat; and (3) methods should be investigated to reduce uncertainty in estimates of sea lamprey production, with emphasis on those that reduce the uncertainty associated with larval length at the end of the growing season and those used to predict metamorphosis.

Classification of Lentic Habitat for Sea Lamprey (Petromyzon marinus) Larvae Using a Remote Seabed Classification Device

The St. Marys River is believed to be the primary source of sea lampreys (Petromyzon marinus) in Lake Huron. Planning or evaluating lampricide treatments required knowing where lampricides could effectively be placed and where larvae were located. Accurate maps of larval density were therefore critical to formulating or evaluating management strategies using lampricides. Larval abundance was systematically assessed with a deepwater electrofishing device at 12,000 georeferenced locations during 1993 to 1996. Maps were produced from catches at those locations, providing georeferenced detail previously unavailable. Catches were processed with a geographic information system (GIS), to create a map of larval density. Whole-river treatment scenarios using TFM (3-trifluoromethyl-4-nitrophenol) were evaluated by combining the map with one of lethal conditions predicted by a lampricide-transport model. The map was also used to evaluate spot treatment scenarios with a granular, bottom-release formulation of another lampricide, Bayluscide (2’,5-dichloro-4’ -nitro-salicylanilide). Potential high-density plots for Bayluscide treatment were selected from the map and estimates of area, cost, and larval population were developed using the GIS. Plots were ranked by the cost per larva killed. Spot treatments were found to be more cost effective than a conventional TFM treatment and Bayluscide was applied to 82 ha in 1998 and 759 ha in 1999. Effectiveness was estimated with stratified-random sampling before and after treatment in 1999 at 35%. Ten percent already had been removed in 1998, for a total reduction of 45% percent. This marked a change in how research and planning were combined in sea lamprey management to minimize treatment costs and evaluate success.