Dean G. Fitzgerald
Department of Biology, University of Waterloo, Waterloo, ON
The physical alteration of Laurel Creek is representative of a general attitude that these changes are necessary to regulate natural processes which do not conform to our development plans (Rasmussen 1996). Stream regulation provides control over flood waters, abatement of low flow pollution, creation of lacustrine in place of riverine habitats, and the opportunity for development of hydroelectric potential. The propensity of stream modification has increased in recent decades without a concurrent improvement in our understanding of the effects on upstream or downstream biota, particularly small stream fish, which are not of commercial or sport value (Benke 1990; Hynes 1970; Smith 1971).
The ecological impacts of restricting a lotic system extend far beyond the facing of the new water control structure. Impoundments limit contributions of woody materials, modify nutrient dynamics, and increase surface area available for exposure to solar irradiation (Hynes 1970). Accordingly, primary productivity becomes dominated by pelagic algae which changes the quantity and composition of available resources downstream (Moss 1995). Impoundments control flows and also facilitate either an increase or decrease in downstream water temperatures (Gordon et a/. 1992); Laurel Creek Reservoir, and all other lakes on Laurel Creek, boast top-draw outlets, that only release warm surface waters. As such downstream zones become both biologically and physically dissimilar to the former stream habitat. Impoundment-modified lotic systems have been shown to demonstrate decreased invertebrate and fish diversity (Hansen and Ramm 1994; Pflieger and Grace 1987) with tolerant species dominating the biomass (Benke et a/. 1988; Ruhr 1956). Urbanization of watersheds inherently leads to stream channel modifications primarily involving bankside vegetation removal, armouring of banks, and ultimately the replacement of natural with concrete channels (Gordon et a/. 1992). Physical modification of the channel in association with altered water quality from upstream impoundments act synergistically against specialized taxa and juvenile fish, and select for habitat generalists (Benke 1990; Gorman and Karr 1978; Strange et a/. 1992).
Fish community characteristics can be used as inexpensive and objective tools for evaluating the status of aquatic resources (Hodson et a/. 1996; Karr 1981). Community interpretations can identify key environmental factors through a functional examination of the ecological tolerances and requirements of abundant species, species associations within specific zones, species relative abundances, and functional indices (Berkman and Rabeni 1987; Hansen and Ramm 1994; Rooke and Mackie 1982). This examination, in concert with the use of historical fish community records of species presence or absence, provides a temporal along with a spatial perspective on changes within the aquatic resources of interest.
This paper describes the current fish community of Laurel Creek based on sampling completed during 1995. Community composition concerning the ecological tolerances and requirements of a suite of key species is then considered in light of historical records. This comparison will help elucidate the primary causative factors responsible for the observed changes in the ichthyofauna.
Fish collections at each location involved sampling two visibly similar reaches (20-30 m) and always involved isolating the reach with seine nets (5 mm stretched mesh; Franklin Net and Twine, Wheatley, ON). During the spring and fall surveys, one reach was sampled using a standard minnow seine with bag. The second reach was sampled using a backpack electrofisher (Smithroot, Vancouver, WA, USA). Summer sampling used only electrofishing. Collection of fish with contrasting gears allowed for the minimization in sampling gear biases and increased the opportunity to collect all species which were present.
Surveys involved field identification, enumeration, and release of individuals alive. Species new to the collection, or with unknown identity were returned to the laboratory for identification confirmation (Hubbs and Lagler 1964; Scott and Crossman 1973). Information concerning historical community structure was collected from government documents (German 1967; Mason 1972; Sandilands 1971), a thesis (Kakonge 1970), scientific publications (Hamor and Fernando 1978; Molnar et a/. 1974) and conversations with known fish authorities (E. Holm, Centre for Biodiversity and Conservation Biology, Royal Ontario Museum, personal communication; E. Kott, Department of Biology, Wilfrid Laurier University, personal communication).
Fish community survey data collected for each sampling season allowed for the construction of species summary presence/absence data matrices for each location. This information was combined with historical fish community data collected before or just after reservoir construction (1967), and again between 1973 and 1975. Hierarchial cluster analysis using SYSTAT 5.0 with a Euclidean distance metric and an average linkage method (Wilkinson 1990) examined the temporal and spatial relationships between species associations and sampling locations. Specific evaluation of the current status of the following suite of species was also completed: rosyface shiner Notropis rubellus, creek chub Semotilus atromaculatus, blacknose shiner Notropis heterolepis, common carp Cyprinus carpio, Cyprinidae; rainbow darter, Etheostoma caeruleum, yellow perch Perca flavescens, Percidae; pumpkinseed sunfish Lepomis gibbosus, and largemouth bass Micropterus salmoides, Centrarchidae.
The fish community survey data from 1995 allowed for the determination of diversity and evenness indices to describe the fish community structure based on electrofishing results for each stream section and location. One measure of diversity was species richness (S.), defined as the number of fish species collected at each location during sampling. The second diversity measure was the Shannon-Weaver index (H'; Pielou 1975) while the third measure was Simpson's Diversity (Ds; Poole 1974). A measure of evenness (Ji) was also determined (Pielou 1975). For each indice, it is assumed the higher the value, the higher the evenness or diversity of the community. The use of diversity indices has two important assumptions. It assumes that H' is derived from population surveys that have enumerated all species, and relative abundances during surveys is representative of actual community structure.
Community surveys completed during 1995 collected 25 species representing seven families. Numbers and presence of fish varied between survey seasons. Some species were collected during all seasons, others were either collected during the spring or fail, but never only during the summer. For example, consider three members of the Cyprinidae: creek chub were collected during all seasons, rosyface shiner only during the spring, and northern redbelly dace Phoxinus eos only during the fall. Generally, more fish were encountered during the spring than the summer or fall. Spring collections were dominated by adult members of the Cyprinidae (common carp and brassy minnow Hybognathus hankinsoni) and Catostomidae (common white sucker Catostomus commersoni); fall collections had numerous juvenile Centrarchidae (pumpkinseed sunfish and smallmouth bass Micropterus dolomieu/) and Cyprinidae (creek chub and blacknose dace Rhinchthys atratulus). No one species or group dominated summer surveys.
Examination of historical and current fish community data by cluster analysis provided obvious spatial and temporal separation of stream sections (Fig. 2). The dendrogram clearly isolates locations downstream of Silver Lake with direct association with the Grand River proper from those associated with urban and reservoir habitat and water quality modifications. This second branch separates community associations from 1967 and 1975 from 1995 results except results from downstream of Columbia Lake for 1975. Consideration of the aforementioned suite of eight species demonstrate the general patterns evident that have occurred in this fish community over the last 30 or so years. Some species are still common throughout despite habitat modifications (e.g. creek chub), some species formerly common are now restricted to specific sections of the stream (e.g. rainbow darter), and some species formerly common were not collected during 1995 (e.g. blacknose shiner). Species formerly rare or absent now dominate areas downstream of the reservoir (e.g. common carp), one introduced species is now abundant throughout the stream (pumpkinseed sunfish), another introduced species occurs throughout the stream, but is rare (largemouth bass) and the third introduced species was not collected during 1995 (yellow perch).
Diversity and evenness measures changed relative to the distance downstream from the reservoir (Table 1). Greater estimates of diversity and evenness were recorded above the reservoir, on the University of Waterloo campus, and below Silver Lake. Locations spatially close to the reservoir had lower estimates. Large populations of one or two species during a survey resulted in reduced estimates of diversity and evenness, despite high species richness. This was evident for location LJ during the spring and LA during the fall.
This response pattern follows the framework of Rooke and Mackie (1982), designed to characterize lotic environments based on the ecological tolerances and requirements of resident organisms. All species which have declined or have been extirpated can't tolerate one or more factors acting to reduce water or habitat quality. The common carp, formerly rare before reservoir construction, is tolerant of warm waters with high sedimentation. More importantly, this species spawns more effectively in lentic over lotic habits (Scott and Crossman 1973). The success of pumpkinseed sunfish is also partially due to high tolerances for variable water quality, preference for lentic over lotic spawning, and the key ecological strategy of producing multiple cohorts of young during a season, dependent on high summer water temperatures (Scott and Crossman 1973).1 observed at least three distinct size classes of young sunfish during 1995 in areas downstream of Laurel Creek Reservoir and the other urban impoundments, but only one in the cooler headwater habitats. The decline of yellow perch and perhaps other indigenous fish species was likely due to the introduction of large numbers of the piscivorous largemouth bass. This species is known to have a voracious feeding behaviour (Scott and Crossman 1973) and following introduction, immediately decimated the yellow perch population of Laurel Creek Reservoir (Kott, personal communication). Largemouth bass also prefer lentic over lotic spawning, but I observed spawning outside of the impoundments. Perhaps lotic spawning behaviour proves less successful for this bass species and contributes to the low numbers observed in my surveys. Another possibility is this species' susceptibility to the numerous recreational anglers that visit Laurel Creek.
Community modification within up- and downstream sections is also dependent on the degree of fish movement from the newly created impoundment over the former riverine environment which provides a continual source of out-migrants into adjacent waters (Martinez et a/. 1994; Ruhr 1956). During fall draw-down of Laurel Creek Reservoir and other urban lakes, numerous species including common carp, pumpkinseed sunfish and largemouth bass are displaced due to the reduction in available habitat. A similar process of active movement out due to density-dependent factors (e.g. food availability) also helps make the impoundments a source of species which are not adapted for successful reproduction in stream environments. These movements on mass from the reservoir can also exacerbate interspecific competition with indigenous species (e.g. brassy minnow).
Of particular interest is the observation that the fish community changed minimally between 1967 and 1975. Perhaps this time lapse between substantial habitat change (reservoir construction) and observed community composition modification is due to the use of community binary data (presence or absence). Data in this form does not identify whether a species is successful in an environment. This is an obvious limitation of using historical binary data to assess community change over time (Poole 1974). Upstream areas could have acted as source populations transporting individuals to downstream sink populations, thus allowing species to be represented in survey results. Spence and Hynes (1971a, 1971b) clearly demonstrated modified fish and invertebrate community structure downstream of Belwood Reservoir and attributed the observed changes primarily to decreased water temperatures associated with the bottom-draw dam. Modified thermal regimes and sedimentation of substrate may have allowed for minimal reproductive success of indigenous species like rainbow darter, rosyface shiner and blacknose shiner, each known to require variable flows, clean gravel and low turbidity for their success (Scott and Crossman 1973). Alternatively, survey results could have been including old individuals and not documenting the lack of spawning success. The potential for transport of individuals and delayed life-cycle responses suggests that any form of environmental impact assessment of habitat modification would not have detected immediate community changes. Detection of change would have required waiting until, at the minimum, for one complete life-history to be completed for the organisms or community of interest (Fitzgerald et al. 1997a).
Of further interest is the inability of species like rosyface shiner to migrate beyond Silver Lake. This species and others only occur downstream of Silver Lake to the Grand River, and appear to successfully spawn in the stream (personal observation). Clearly this barrier acts to interrupt expected species longitudinal succession in this stream (Fitzgerald 1997a). Given the opportunity these species would likely re-colonize upstream areas from the Grand River.
Collection of creek chub throughout Laurel Creek from headwater sections to the confluence with the Grand River suggests this species is able to successfully inhabit and reproduce within a highly modified stream environment. However, examination of a suite of biological endpoints in these fish suggests the contrary is true. Examination of separate populations within Laurel Creek during 1995 identified reduced performance (fecundity, growth) of individuals downstream of the reservoir, urban lakes and urban core compared to the more pristine headwater section despite similar fish condition relationships (weightxlength). Synchronous with these responses was the observation of reproductive anomalies (egg atresia, ovarian masses and tumours) in downstream fish only. The overall response pattern identified was consistent with fish being forced to forage outside of their preferred niche on food resources smaller in size and of less energetic value, and also suggestive of exposure to some form of sub-lethal stressor(s) causing the reproductive anomalies. With the identification of this response pattern was the absence of young-of-the-year fish within downstream collections, and the question whether or not creek chub are effectively reproducing in these habitats, or if their continued existence is due to transport of individuals from headwater source populations (Fitzgerald et a/. 1997b).
Laurel Creek has been modified extensively over the last 200 years. The most dramatic of these changes appears to have been the removal of a wetlands complex just downstream of the headwater zone for the construction of Laurel Creek Reservoir. This event reduced stream flow diversity, altered sedimentation processes, thermal regimes, and replaced a heterogeneous stream habitat (wetlands) with a homogenous one (reservoir) (Kakonge 1970). This key event in the history of Laurel Creek was followed by a decline and shift in the species assemblage of both invertebrates and fish. Laurel Creek will never be the same again. We have to live with this fact. Numerous species have been completely lost from the stream, others are on the decline. One option for the future to improve stream habitats and water quality is the transformation of the heavily silted and antiquated Laurel Creek Reservoir back to a wetlands complex (Fitzgerald 1996). This exercise would likely be unique in North America, and promote a rubric over appropriate forms of stream rehabilitation and management of water quality in a primarily urban stream. This form of action is in accordance with the limited recent practice in some watersheds involving the replacement of concrete channel back to natural forms (Rasmussen 1996). Further, with proposed urban development within the west side of Waterloo, the transformation of agriculture to urban land-use will also have a dramatic impact on the headwater section of this stream. This zone seems to be integral part of the stream, and maybe the source of many species not able to effectively reproduce in highly modified downstream habitats. If this process of headwater urbanization occurs in the same fashion as other development processes in this stream, most indigenous fish species will likely disappear soon after. Care must be taken to preserve adequate stream riparian buffer zones, manage storm and urban drainage more effectively, and avoid at all costs any further channel modifications or re-directions. The stakeholders of Laurel Creek need to consider the implications of their actions and development prior to implementation. The ichthyofauna of Laurel Creek would surely appreciate this.
Benke, A.C. and six co-authors. 1988. Bioenergetic considerations in the analysis of stream ecosystems. J. N. A. Benthol. Soc. 7:480-502.
Berkman, H.E., and Rabeni, C.F. 1987. Effect of siltation on stream fish communities. Env. Biol. Fish.18:285-294.
Cox, J., Hendrickson, C, Skelton, l., and Suffling, R.1996. Watershed planning for urbanization to avoid undesirable stream outcomes. Can. Water Res. J. 21:237-251.
Fitzgerald, D.G. 1996. Evaluation of fish in anthropogenically-modified streams within the Grand River watershed, Ontario. University of Waterloo, Department of Biology, unpublished M.Sc. thesis.
Fitzgerald, D.G., Lanno, R.P. and Dixon, D.G.1997a. A quarter century of change in three streams modified by reservoirs and urbanization in the Grand River watershed. submitted J. Ecosyt. Dist. Rec.
Fitzgerald, D.G., Lanno, R.P. and Dixon, D.G. 1997b. A comparison of a sentinel species evaluation using creek chub (Semotilus atromaculatus Mitchill) to a fish community approach for the identification of environmental stressors in small streams. submitted Can. J. Fish. Aquat. Sci.
German, M.J.1967. Biological survey of the Grand River and its tributaries. Department of Lands and Forests, Toronto, Ontario. Available for review at the Grand River Conservation Authority, 400 Clyde Rd. Cambridge, ON, Canada.
Gordon, N.D., T.A. McMahon, and B.L. Finalyson. 1992. Stream Hydrology: An introduction for ecologists. John Wiley and Sons, Ltd. New York.
Gorman, O.T., and Karr. J.R. 1978. Habitat structure and stream fish communities. Ecology 59:507-515.
GRCA (Grand River Conservation Authority). 1993. Laurel Creek Watershed Study Technical Appendix 7: Aquatic Resources. Available for review at the Grand River Conservation Authority, 400 Clyde Rd. Cambridge, ON, Canada.
Hamor, T. and Fernando, C.H. 1978. Differences in the fish fauna in a reservoir and upstream and downstream. Unpublished manuscript presented at Environmental Effects of Man-made Lakes; Belgrade, Hungary, 5-9 June.
Hansen, M.J., and Ramm, C.W. 1994. Persistence and stability of fish community structure in a southwest New York stream. Am. Mid. Nat. 132:52-67.
Hodson, P.V., Munkittrick, K.R., Stevens, R. and Colodey, A. 1996. A tier-testing strategy for managing programs of environmental effects monitoring. Wat. Qual. Res. J. Can. 31:215-224.
Hubbs, C.L., and Lagler, K.F.1964. Fishes of the Great Lakes region. The University of Michigan Press, Ann Arbor, Ml.
Hynes, H.B.N.1970. The ecology of running waters. University of Toronto Press, Toronto.
Kakonge, S.1970. Ecology of small stream fishes. University of Waterloo, Department of Biology, unpublished M.Sc. thesis.
Karr, J.R.1981. Assessment of biotic integrity using fish communities. Fisheries 6:21-27.
Martinez, P., Chart, T., Trammell, M., Wullschleger, J. and Bergersen, E. 1994. Fish species composition before and after construction of a main stream reservoir on the White River, Colorado. Env. Biol. Fish. 40:227-239.
Mason, P.1972. The 1972 benthic and fish analysis report. Grand River Conservation Authority, Land Management Division, Biological Services Branch, Cambridge, ON. Available for review at the Grand River Conservation Authority, 400 Clyde Rd. Cambridge, ON, Canada.
Molnar, K., Hanek, G. and Fernando, C.H. 1974. Parasites of fish from Laurel Creek, Ontario. J. Fish. Biol. 6:717-728.
Moss, B. 1995. The microwaterscape-A four dimensional view of interactions among water chemistry, phytoplankton, periphyton, macrophytes, animals and ourselves. University of Liverpool, Department of Environmental Biology, Liverpool, U.K. unpublished manuscript.
Pielou, E.C.1975. Ecological Diversity. John Wiley and Sons, New York.
Pflieger, W.L. and Grace, T.B.1987. Changes in the fish fauna of the Lower Missouri River,19401983. In: W.J. Matthews and D.C. Heins, eds., The ecology and evolution of North American stream fish communities. University of Oklahoma Press, Norman, OK.
Poole, R.W.1974. An introduction to Quantitative Ecology. McGraw Hill Inc. Scientific Publications, New York.
Rasmussen, J.L. 1996. American Fisheries Society (AFS) position statement: floodplain management. Fisher. 21:4:6-10.
Rooke, J.B., and Mackie, G.L. 1982. An ecological analysis of lotic environments: I. Design and testing. J. Fresh. Ecology 1:42-431.1
Ruhr, C.E.1956. Effect of stream impoundment in Tennessee on the fish populations of tributary streams. Trans. Am. Fish. Soc. 75:144-157.
Sandilands, A. 1971. Fish report for Grand River Valley. Grand River Conservation Authority, 400 Clyde Road, Cambridge,. ON.
Scott, W.B., and Crossman, E.J. 1973. Freshwater fishes of Canada. Bull. Fish. Res. Board Can. No. 184.
Smith, P.W. 1971. Illinois streams: A classification based on their fish and an analysis of factors responsible for the disappearance of native species. III. Nat. Hist. Sur. Biol. Note. No. 76.
Spence, J.A., and H.B.N. Hynes. 1971a. Differences in Fish Populations Upstream and Downstream of a Mainstream Impoundment. J. Fish. Res. Board Can. 28:45-46.
Spence, J.A., and H.B.N. Hynes. 1971b. Differences in benthos upstream and downstream of an impoundment. J. Fish. Res. Board Can. 28:35-43.
Strange, E.M., Moyle, P.B., and Foin, T.C. 1992. Interactions between stochastic and deterministic processes in stream fish community assembly. Env. Biol. Fish. 36:1-15.
Wilkinson, L. 1990. SYSTAT: the system for statistics. SYSTAT Inc., Evanson, III.