Sara A. Adlerstein

20 years of the German Small-Scale Bottom Trawl Survey (GSBTS): A review

AbstractThe German Small-scale Bottom Trawl Survey (GSBTS) was initiated in 1987 in order to provide complementary investigations to the International Bottom Trawl Survey (IBTS) in the North Sea, using the same methodology but focussing high-intensity sampling on selected survey areas. Over the last 20 years, the initial number of 4 survey areas (10 × 10 nautical miles; “Boxes”) has been increased to 12, which are distributed over the entire North Sea. This paper describes the survey methods of the GSBTS, summarizes the scientific outcome of the first 20 years, and suggests that international fisheries research institutions would join the GSBTS.The major outcomes of the survey include to date:— Documentation changes in the distribution of fish species and in species assemblages (e.g. changes in species richness, shifts in the southern species component).— Geostatistical evaluation of GSBTS data.— Analysis of spatial scale effects: the relevance of GSBTS survey results for interpreting large-scaled abundance and distribution data from the IBTS.— Description of benthic habitats, composition of invertebrate fauna and its variability.— Process studies, especially investigation of predator-prey interactions between fish through analyses of stomach contents.— Characterization of the typical hydrographic conditions in the survey areas and their variability, and description of the nutrient supply.— Observations of seabirds and their feeding habits.— Analysis of the effects of different parameters on catch rates for bottom fish and on the estimates of abundance indices (e.g. vessel and gear effects, towing time, hydrographic conditions, time of day, number of hauls per area). In continuing this interdisciplinary survey with simultaneous sampling of all faunal and environmental compartments and especially in making it an international effort, we see the possibility of contributing data for the implementation of the ecosystems approach to fisheries management. Particularly, the following aspects can be addressed and would further increase the scientific value of the GSBTS:— Combining the survey data with highly resolved data from the commercial fishery to separate the effects of fishing from natural variability.— Further interdisciplinary analyses of the entire data set. Main aspects include benthos-fish-bird-community changes over time and their relation to historic fisheries impacts, and the coupling of biological and physical habitat characterisation.— Collection of accompanying data (phyto-, zoo- and ichthyoplankton data) in order to make the GSBTS a true ecosystem survey in detecting temporal changes in nearly all major levels of the food web.

Estimating Seasonal Movements of Chinook Salmon in Lake Huron from Efficiency Analysis of Coded Wire Tag Recoveries in Recreational Fisheries

Abstract The decline of hatchery-reared Chinook salmon Oncorhynchus tshawytscha stocks in Lakes Huron and Michigan during the 1980s prompted mass-tagging programs to investigate reproduction, poststocking survival, and movements. In Lake Huron, millions of smolts implanted with coded wire tags (CWTs) were released in Michigan waters and recovered from charter and noncharter fisheries, surveys, and weirs. Using generalized linear models (GLMs), we investigated Chinook salmon seasonal movements based on the spatial and temporal distributions of recoveries by fishing trips in U.S. recreational fisheries and recovery efficiency. We used models incorporating area, month, year, and recovery source; creel-clerk and “headhunter” (CWT collection specialist) samples; and charter captain reports. We implemented models for recoveries regardless of release area and from one particular area. All model predictors and interactions between month and area were significant. The variation in recovery levels among recovery so...

Causes of Phytoplankton Changes in Saginaw Bay, Lake Huron, During the Zebra Mussel Invasion

ABSTRACT Colonization of the Laurentian Great Lakes by the invasive mussel Dreissena polymorpha was a significant ecological disturbance. The invasion reached Saginaw Bay, Lake Huron, in 1991 and initially cleared the waters and lowered algal biomass. However, an unexpected result occurred 3 years after the initial invasion with the return of nuisance summer blooms of cyanobacteria, a problem that had been successfully addressed with the implementation of phosphorus controls in the late 1970s. A multi-class phytoplankton model was developed and tested against field observations and then used to explore the causes of these temporal changes. Model scenarios suggest that changes in the phytoplankton community can be linked to three zebra mussel-mediated effects: (1) removal of particles resulting in clearer water, (2) increased recycle of available phosphorus throughout the summer, and (3) selective rejection of certain Microcystis strains. Light inhibition of certain phytoplankton assemblages and the subsequent alteration of competitive dynamics is a novel result of this model. These results enhance our understanding of the significant role of zebra mussels in altering lower trophic level dynamics of Saginaw Bay and suggest that their physical reengineering of the aquatic environment was the major force driving changes in the phytoplankton community composition.

Phytoplankton community composition of Saginaw Bay, Lake Huron, during the zebra mussel (Dreissena polymorpha) invasion: a multivariate analysis.

ABSTRACT The colonization of the zebra mussel (Dreissena polymorpha) in Saginaw Bay dramatically altered the phytoplankton community composition resulting in exclusion of light sensitive species and dominance of species with oligotrophic preferences and light resistance. In 1990, the NOAA Great Lakes Environmental Research Laboratory initiated a 7-year survey program to monitor changes in the lower food web of Saginaw Bay, where zebra mussels became established in the fall of 1991. To investigate shifts in the phytoplankton community composition over the 7-year period from 1990 to 1996 we searched for clusters of similar composition using multivariate principal component analysis (PCA) on proportions of 22 taxonomic groupings of the total phytoplankton density (cells per milliliter). We then used an agglomerative hierarchical clustering analysis of the PCA scores. We identified five characteristic phytoplankton communities in configurations that allowed recognizing four distinct periods in Saginaw Bay linked to the zebra mussel invasion. Significant changes were indicative of increased water clarity and eutrophic conditions being replaced by more oligotrophic conditions as clusters dominated by light sensitive species, such as the cyanobacteria Oscillatoria redekii, became immediately rare and clusters dominated by diatoms such as Cyclotella spp. became common. Microcystis spp., a light tolerant cyanobacteria not grazed by zebra mussel, dominated assemblages after 1994. The shifts in phytoplankton composition confirm that zebra mussels effects on phytoplankton communities are mediated by both direct (filtration) and indirect (nutrient cycling) mechanisms and also suggests that increased light penetration is an important mechanism behind some changes.

Seasonal Movements of Chinook Salmon in Lake Michigan Based on Tag Recoveries from Recreational Fisheries and Catch Rates in Gill-Net Assessments

Abstract There are no specific studies of the movements of introduced Chinook salmon Oncorhynchus tshawytscha in Lake Michigan, despite the need for such information for population assessments and stocking allocations. We investigated the seasonal distribution of hatchery-reared Chinook salmon between May and September based on fishery-dependent (recoveries from recreational fisheries of fish marked with coded wire tags [CWTs]) and fishery-independent sources (catches in assessment gill-net surveys). We modeled recoveries by fishing trips in Michigan waters of Lake Michigan to estimate spatially and temporally explicit abundance indices using generalized linear models (GLMs) and accounted for the efficiency among recovery sources (charter boat captain reports, creel clerk interviews, and headhunter collections of CWT samples from charter boat and non-charter boat catches). Recovery levels varied among areas, months, years, and recovery sources, and distribution among areas also varied by month. We used CW...

Lake Trout Movements in U.S. Waters of Lake Huron Interpreted from Coded Wire Tag Recoveries in Recreational Fisheries

ABSTRACT Lake trout management and restoration make use of age-structured population models that incorporate parameters to represent movement among management areas, and harvest quotas are based on proportions of stocked fish remaining in and moving out of areas of release. We investigated movements of lake trout in U.S. waters of Lake Huron based on spatial and temporal distributions of coded-wire-tag (CWT) recoveries by trip in recreational fisheries using Generalized Linear Models (GLMs). For the analysis, we incorporated as model predictors the area, month, year, and source of CWT recovery, which included reports by charterboat captains, creel-clerk interviews of non-charter anglers, and “headhunter” (CWT collection specialist) samples from charter and non-charter catch. Results indicated that CWT recoveries by trip were lowest from charterboat operators, followed by recoveries from creel clerks (2× captain reports), headhunter non-charter (3×), and headhunter charter (9×). Standardized recovery levels were highest in the management area of release and one area immediately adjacent, with remaining percentages decreasing with distance from release. CWT recovery levels decreased from May to September and suggest seasonal movement among areas that have implications for stock assessment. From standardized recoveries, we estimated that 40% of the CWT lake trout were recaptured in areas where released and others moved north, south, and southeast. Our results indicate that higher proportions of lake trout move out of release areas fish than previously shown and suggest that prior studies may be biased, in part due to lack of standardization among tag recovery sources and ignoring seasonal movements.

Assessment of Top-Down and Bottom-Up Controls on the Collapse of Alewives (Alosa pseudoharengus) in Lake Huron

Food web models are powerful tools to inform management of lake ecosystems, where top-down (predation) and bottom-up (resource) controls likely propagate through multiple trophic levels because of strong predator–prey links. We used the Ecopath with Ecosim modeling approach to assess these controls on the Lake Huron main basin food web and the 2003 collapse of an invasive pelagic prey fish, alewife (Alosa pseudoharengus). We parameterized two Ecopath models to characterize food web changes occurring between two study periods of 1981–1985 and 1998–2002. We also built an Ecosim model and simulated food web time-dynamics under scenarios representing different levels of top-down control by Chinook salmon (Oncorhynchus tshawytscha) and of bottom-up control by quagga mussels (Dreissena rostriformis bugensis) and nutrients. Ecopath results showed an increase in the relative importance of bottom-up controls between the two periods, as production decreased across all trophic levels. The production of non-dreissenid benthos decreased most, which could cause decreases in production of pelagic prey fishes feeding on them. Ecosim simulation results indicated that the alewife collapse was caused by a combination of top-down and bottom-up controls. Results showed that while controls by Chinook salmon were relatively constant before alewife collapse, controls by quagga mussels and nutrients increased jointly to unsustainable levels. Under current conditions of low nutrients and high quagga mussel biomass, simulation results showed that recovery of alewives is unlikely regardless of Chinook salmon biomass in Lake Huron, which implies that the shrinking prey base cannot support the same level of salmonine predators as that prevailed during the 1980s.

Towards a process domain-sensitive substrate habitat model for sea lampreys in Michigan rivers.

Abstract Habitat mapping is a common and often useful tool in the ecological management of rivers. The complex nature of fluvial processes, however, makes it difficult to predict the reach-scale distribution of substrate habitat from landscape-scale covariates. An option is to identify and partition a data set on boundaries of geomorphic process domains, within which the globally complex relationships between landscape, climate, and instream habitat may potentially be approximated by a simpler model. In this study, we used regression trees as a machine learning method for partitioning and identifying useful strata in a geographically extensive substrate habitat model for larvae of the sea lamprey Petromyzon marinus, an invasive and economically harmful species in the Laurentian Great Lakes. We used field survey data from over 5,000 substrate habitat transects collected in 43 watersheds of the Lower Peninsula of Michigan, and we created a geographic database of geographical information systems-derived cova...