Completed Research

 

Evaluating the Effects of Environment and Stressors on Thiaminase Expression in Alewife (Former Graduate Student: Jesse Lepak)

alewifeFor the past 30 years Great Lakes scientists have been trying to put together an extraordinary puzzle resulting in the reproductive failure of lake trout and Atlantic salmon, two valuable native Great Lakes fishery resources. "Early mortality syndrome" (EMS) is another name for a thiamin deficiency that results in the complete reproductive failure of certain predatory fish - such as lake trout and Atlantic salmon - that feed on forage fish containing thiaminase, a thiamin destroying enzyme. Members of the herring family such as alewife are particularly prone to having high thiaminase levels, and alewife are the most abundant Great Lakes forage fish fed upon by large predators. Unfortunately, EMS is not limited to the Great Lakes and has also been observed to limit reproduction by Baltic Sea Atlantic salmon that feed upon Baltic herring. Although EMS has been observed for over 30 years, the cause of this problem has resisted understanding as scientists have explored a series of explanations regarding factors that influence thiamin and thiaminase in fish. We conducted a series of pond and laboratory experiments to evaluate the influence of stressful environmental conditions (e.g. low food consumption, temperature and physiological stress) upon thiaminase levels in alewife and gizzard shad. In our experiments, thiaminase levels in these Clupeid fishes always increased in response to increasing fish condition (assessed by measuring the proportion of body weight comprised of dry matter), suggesting that high food consumption was the key factor responsible for high thiaminase levels. The observation that high levels of food consumption produce high thiaminase levels is believed to result from the fact that bacteria in alewife and gizzard shad stomachs produce thiaminase, and these bacteria likely thrive when provided with steady food inputs and environmental conditions associated with high levels of food consumption. Results from these experiments, new assay techniques, and the use of molecular tools that can evaluate changes in microbial communities in fish stomachs will hopefully lead to a better understanding of EMS in Great Lakes fishes. The ultimate goal of this work is to develop management actions that will reduce thiaminase levels in Great Lakes fishes, leading to the re-establishment of self-sustaining populations of lake trout and Atlantic salmon.

For further information:

Lepak, J.M., C.E. Kraft, D.C. Honeyfield and S.B. Brown. 2008. Evaluating the effect of stressors and immunosuppression on thiaminase in alewife. Journal of Aquatic Animal Health 20:63-71. [Link to Full article]

Dynamics of wood recruitment and accumulation within northeastern US streams (Former Graduate Student: Dana Warren)

Large wood and woody debris dams are important habitat features in forested stream (riparian) ecosystems. Given the importance of wood to stream ecosystems, the volume, abundance, and characteristics of large wood in streams is of interest to researchers and managers alike. However, the standing stocks, or standing timber, and input rates for wood in streams is regionally specific and influenced by the composition of riparian forests, the geology of a region, and the types of regional disturbance. Prior to this study, information about the dynamics of wood in northeastern streams was scarce, therefore limited predictive information was available to individuals involved in stream research and management efforts.

As part of this effort we surveyed 28 forested streams across the northeastern United States, with a range of wood, stream, and riparian forest characteristics. Overall, the volume and frequency of wood in streams was most closely associated with the age of the dominant canopy trees in the riparian forest. Results indicate that the biomass of wood in streams will continue to increase as forests mature; that is, until dominant trees in the riparian forest reach up to 250 years of age or more.

The linear models developed in this study allow researchers to estimate current and future wood loading to forested streams in the Northeast from a few easily collected stream and riparian forest measurements. This information can be used by managers to estimate the amount of large wood that would be expected to occur within a given stream, as well as to evaluate the amount of in-stream wood required to meet restoration goals.

In addition to predicting wood loading to streams, we have been evaluating factors influencing the spatial distribution of large woody debris (LWD) in streams. In our initial efforts investigating this subject, we used a one-dimensional version of Ripley’s K, a second-order statistic that evaluates the spatial pattern of points within a landscape. This neighbor-K statistic evaluates the number of points within a series of distances centered at each point (e.g. large wood) within a stream. We subsequently employed a test statistic evaluating the distribution of points within specific distance "bins" rather than throughout an entire distance interval. The latter statistic has been effective at describing both aggregated and regularly spaced distributions of wood within two dozen study streams in the northeastern U.S. Previous studies of wood distribution in streams have not observed regularly spaced distribution patterns. Patterns in the longitudinal spatial distribution of large woody debris in streams are useful to understand processes responsible for the accumulation of in-stream wood.

For further information:

Kraft, C. E. and D. R. Warren. 2003. Development of spatial pattern in large woody debris and debris dams in streams. Geomorphology 51:127-139.

Stable Isotope Measurements as Indicators of Diet Shifts in a Lake Trout (Salvelinus namaycush) Population in an Oligotrophic Adirondack Lake (Former Graduate Student: Jesse Lepak)

This study addressed the following objectives: 1) using naturally occurring stable isotopes to assess the recovery of lake trout and other native fish species in an aquatic ecosystem after the removal of a non-native predator: smallmouth bass; 2) evaluated changes in lake trout in relation to changes in stable isotope signatures and dietary shifts.

Stable isotopes have become increasingly useful in evaluations of aquatic ecosystems. Specifically, carbon and nitrogen isotopes have enhanced our ability to evaluate food sources and relative trophic position of fishes. In this study, shifting carbon sources and the food web structure of an Adirondack lake trout (Salvelinus namaycush) population were assessed using naturally occurring stable isotopes. Smallmouth bass (Micropterus dolomieu) abundance has declined precipitously in a 271 hectare oligotrophic lake in the Adirondack mountains (New York, U.S.A.) due to a removal effort initiated in May 2000. Prior to the bass removal, lake trout in the study lake relied heavily on zooplankton and chironomids as a food source. Lake trout then took advantage of increased numbers of native prey fishes resulting from diminished smallmouth bass predation.

In this study, δ13C and δ15N measurements from lake trout tissue samples were examined prior to and following removal of smallmouth bass from a 260 ha lake to identify the impact of this dominant non-native predator (i.e. bass are not native in Adirondack lakes) on lake trout trophic position. Throughout the study, lake trout diets were examined as littoral zone prey fish abundance increased in response to the bass removal. Following initiation of bass removal, lake trout diets increasingly consisted of a larger proportion of littoral prey fish. Stable isotope measurements showed that lake trout tissue responded with higher δ13C, indicating that lake trout are now assimilating more carbon from the littoral zone relative to those lake trout collected prior to the smallmouth bass removal. Post-removal lake trout exhibited elevated δ15N values and increasing numbers of prey-fish in their diets, implying utilization of higher quality food sources. Results from this work indicate that removal of non-native predators can restore food web trophic linkages typical of pre-disturbance lake conditions.

For further information:

Lepak, J.M., C.E. Kraft and B.C. Weidel. 2006. Rapid food web recovery in response to removal of an introduced apex predator. Canadian Journal of Fisheries and Aquatic Sciences 63:569-575. [Link to Full article]

 

Gastric Evacuation and Daily Ration of Naturally Produced Age-0 Chinook Salmon (Oncorhynchus tshawytscha) in Lake Ontario (Former M.S. Student: Nick Principe)

This study assessed consumption by naturally produced juvenile Chinook salmon as they entered the nearshore region of Lake Ontario after emigrating from the Salmon River, a major U.S. tributary of Lake Ontario. Evacuation rate was estimated as a function of temperature via controlled laboratory experiments, and diel field sampling for daily ration estimates occurred on three dates in late spring (June 6, 12, and 18, 2001). Although no consistent diel feeding pattern was observed, juvenile Chinook maintained an elevated gut fullness value throughout the day and did not feed after dark. Moreover, daily ration values were high in comparison to similar studies in other waters. Comparisons of daily ration estimates to bioenergetics model estimates showed that juvenile Chinook salmon were consuming prey at or above their maximum estimated daily ration. Daily ring counts of sagittal otoliths confirmed that Chinook salmon in this investigation were naturally produced. Our results indicate that current food conditions in the nearshore waters of Lake Ontario are suitable for sustaining maximum consumption rates by naturally produced juvenile Chinook salmon that have recently migrated into the lake from spawning tributaries.

For further information:

Principe, N.D., C.E. Kraft, and E.L. Mills. 2007. Gastric evacuation and daily ration of naturally produced age-0 Chinook salmon Oncorhynchus tshawytscha in Lake Ontario. Transactions of the American Fisheries Society 136:1206-1215. [Link to Full article]

Stream fish communities in relation to landscape-scale and local habitat variables in the Beaverkill/Willowemoc watershed, NY (Former M.S. Student: Tamara Smith)

This study was designed to examine fish abundance and species assemblages in Beaverkill/Willowemoc watershed (Sullivan/Delaware Co., NY) tributaries, and to determine how these tributaries contribute to the main stem fishery. We evaluated 69 randomly-selected stream reaches within the watershed and compared relative fish abundance over local and landscape-level scales. Of particular interest was the influence of stream position within the drainage network on fish community structure and abundance.

Statistical description of these data using canonical correspondence analysis (CCA) revealed that stream order, percent fine substrate, number of downstream confluences (C-link), percent canopy cover, water temperature, and percent in-stream vegetation were the six primary environmental factors influencing distributional patterns of fish species. Two additional CCAs were conducted on four repeatedly-sampled sites: one using the same four environmental factors selected through analysis of all 69 sites, in which little seasonal variability was found within sites, and another identifying six environmental variables that explained the most temporal variance among sites. Species assemblages were similar in the seasonal comparison as was evidenced in the CCA results for all sites, and the greatest variance was explained by several landscape-scale variables in combination with maximum water depth and percent in-stream vegetation. An associated study evaluated populations of the three salmonine species (brook, rainbow and brown trout) present in first-, second-, and third-order tributaries throughout the watershed. From 79 surveys during the 2000 field season, we found differences in lengths, weights and habitat preferences for three year- classes of these trout species. The only native trout, brook trout, were found throughout the watershed and were often associated with small first- order cool-water tributaries with low percent pools and low in-stream vegetation. Rainbow trout also preferred cool stream temperatures and low percent pools, but were often found in larger stream orders. In contrast, brown trout were most often associated with warmer second- and third-order streams with high percent pools and high amounts of in-stream vegetation.

Smith, T.A. and C.E. Kraft. 2005. Stream fish assemblages in relation to landscape position and local habitat variables. Transactions of the American Fisheries Society 134:430-440. [Link to Full article]

Ice Storm Impacts and Woody Debris Removal in Adirondack Tributary Streams: Biotic Responses and Changes in Physical Habitat (Former Graduate Student: Dana Warren)

In January 1998, an ice storm in northern New York damaged forest canopies, resulting in increased debris dam frequency and LWD loading in streams (Kraft et al. 2002). While most wood addition studies have evaluated the effects of wood placed and anchored in streams, the 1998 ice storm provided an opportunity to assess naturally-deposited wood structures and determine their role in creating biotic habitat in a boulder-dominated system. In this study, we removed wood from streams within the region of ice storm damage to determine the influence of naturally-deposited woody debris and debris dams on brook trout (Salvelinus fontinalis) and stream invertebrate abundance.

A Before-After, Control-Impact (BACI) study was employed to evaluate the response of biota to debris dams and woody debris removal. In contrast with expectations that we would see reductions in trout abundance after wood removal, we found that the brook trout response to the experimental manipulation was complex and variable. Several debris dams in our study clearly impacted local trout abundance, though many appeared to have little influence. Whereas some of the pools formed in association with large accumulations of in-stream wood - and several debris dams formed in this fashion were removed as part of our experimental manipulation - other pools were created by large boulders that retained pool-forming characteristics even after wood removal. Consequently, the role of boulders in pool formation likely contributed to the variable biotic responses that we observed.

In the evaluation of the response of stream invertebrates to wood removal, the relative proportion of grazers increased upstream and downstream from removed dams in all streams; however, comparisons of other metrics indicated no significant response to removal. Invertebrate responses to wood removal were lower than expected, which again was likely due to the presence of abundant boulder-formed pools in this high gradient system.

The lack of consistent negative biotic responses to wood removal in many study reaches suggests that boulders can provide sufficient habitat complexity for brook trout and stream insect populations in sections of high-gradient streams. Boulders were the dominant structural elements at our study sites, many of which were larger in diameter than in-stream LWD or even streamside trees. The negative response to wood removal observed at a few study sites does suggest, however, that wood addition can increase fish habitat at locations in streams without pre-existing pools.

This project was funded by the U.S. Forest Service, with assistance from the New York State Dept. of Environmental Conservation, Division of Lands and Forests.

For further information:

Warren, D. R. and C. E. Kraft. 2003. Brook trout (Salvelinus fontinalis) response to wood removal from high gradient streams of the Adirondack Mountains (NY, USA). Canadian Journal of Fisheries and Aquatic Sciences 60:379-389.[Link to Full article]

Warren, D.R. and C.E. Kraft. 2006. Invertebrate community and stream substrate responses to woody debris removal from an ice-storm-impacted stream system, NY USA. Hydrobiologia 568:477-488. [Link to Full article]

 

Back to Research Interests