State of the Greater Fundy Ecosystem - Hayward Brook Fish populations

GREATER FUNDY ECOSYSTEM RESEARCH PROJECT

UNB Faculty of Forestry and Environmental Management

State of the Greater Fundy Ecosystem

Hayward Brook Watershed Study:
Fish Populations

Alyre Chaisson
U de Moncton, Dept de Biologie
Morton Avenue, Moncton, NB
E1A 3E9

This study examined the effects of riparian zone widths on fish and their habitat in the Hayward and Holmes Brook watersheds (Figure 1). The most commonly found fish in these brooks is a species of salmonid; the Brook Trout.

GOALS

Critical elements of importance to salmonids that were examined in this study included: 1) stream channel characteristics (width, depth, and length of pools, runs and riffles), 2) composition of substrates in pools, and 3) presence of large woody debris. Fish abundance was determined through electrofishing and live trapping. Pre-harvest baseline data were collected in 1994 and 1995 and post-harvest data from 1996 are currently being analyzed. As well as contributing to our knowledge of the population dynamics of brook trout and illustrating the importance of small streams as refugia, the results of this study will be incorporated into the Fundy Model Forest (FMF) Management Plan.

METHODS

Brook Trout tend to prefer pools as habitat and so the frequency of pools, runs and riffles were recorded for each stream. Wet length, width and depth of pools, runs and riffles were also measured in each year of the study. Figure 2 shows star graphs for all sites. Similar form (shape and size) on the graph indicates similar site characteristics. Changes in these profiles may be used to visually assess the effects brought about by harvesting when post-harvest data collection is completed. Table 1 shows the number of pools, runs and riffles in each study site.

Substrate composition was evaluated by placing a 1 m2 frame subdivided into 625 cm2 grids over the first ten pools in each study site. A colour code was used to identify different substrate types found in each grid.

Woody debris was assessed in each year by sketching the first 50 m in the lower, middle and upper reaches of site 5 (30 m buffer) and site 3 (control). Resources were unavailable to do the 60 m buffer sites, the other 30 m buffer nor the other control site. Angle and length were recorded for woody pieces larger than 0.5 m as well as their position in the stream.

Abundance and distribution of fish were assessed using minnow traps and electrofishing. Tributaries were divided into a downstream (100 m), middle (100 m) and upstream (100 m) section. Six minnow traps were distributed within each 100 m section, with two traps in each of the following habitat types: pools, runs and riffles. Sites were flagged with tape and sampled twice during June-August and once during October in each year. Traps were set for two consecutive 24-hour periods and verified each morning. Fish were identified, measured and released.

In 1994, fish from the traps were marked with a Panjet dye-innoculator and had their adipose fin clipped. This allowed for the monitoring of the movement of the fish both within and among sites for each watershed. In 1995, fish were fin clipped and in 1996 all fish were uniquely marked with external Floy fingerling tags which remain readable throughout the life of the fish.

Electrofishing was conducted at all sites in 1994 except site 10, which proved difficult to access with heavy equipment. Site 10 was electrofished in 1995, following the construction of an access road. A 50 m section of each stream, away from the fish traps, was selected for sampling. Barrier nets were erected both upstream and downstream. Fish were identified, measured, adipose fin-clipped and released.

RESULTS

Sand tended to be a significant substrate component at all the sites in both years surveyed. Sand became a major component in 1995. Sand tends to limit spawning to gravel areas of sufficient depth and current or to areas of sand that contain upwellings of ground water. Although the sand was fine, it did not contain silt which would make many of the sites unsuitable for spawning.

Woody debris consists of logs and large branches that have fallen into the river or stream. Such debris is important in creating pools and in moderating water flow. Woody debris was removed in the past under the mistaken assumption that fish passage was being improved. Logs lying between 61degrees and 90 degrees from the bank of the stream are most likely to contribute to the formation of pools because they are more likely to become wedged in the stream rather than being swept downstream. As expected, most logs observed during the study were oriented in this way.

Since 1994, only 4 fish species have been captured using minnow traps. These include Brook Trout, Slimy Sculpin, American Eel and Threespine Stickleback. Brook Trout heavily dominated all the sites in all years. Sizes and numbers of fish captured in all the tributaries was fairly similar except for site 4 (see below). Some recaptures of tagged fish were made during the summer, but overall returns were low. Since no within season declines in fish numbers were noted, the data suggest significant movement of untagged fish into the tributaries. Electrofishing sampling in 1994 indicated that large numbers of fish appeared to have moved into the tributaries in the fall.

Control Plots (No treatment):

To maintain uniformity, the nomenclature used to identify the terrestrial study sites was maintained for the fish study. Each site was between 650 m and 1 km in length. Sites 3 and 4 represent the control (no cutting) sites. Site 4 had relatively small runs and riffles as this tributary ran nearly dry in each year of the study leaving small scattered pools in its lower reaches. Site 4 also had the lowest overall number of fish captured during the course of the study. This may be because the upper reaches of the stream ran dry after a summer of little rainfall and also beaver dam construction downstream which may have hindered fish passage.

30 m Buffer Plots:

Sites 5 and 6 represent sites having a 30 m buffer. Site 5 had a very similar pools, runs and riffles profile as site 9 (60 m buffer).

60 m Buffer Plots:

Sites 9 and 10 represent sites having a 60 m buffer. Site 10 contained the greatest number of pools, runs and riffles and thus showed the greatest diversity of habitat of all sites for Brook Trout.

IMPLICATIONS FOR MANAGEMENT

No correlation was found between any of the physical habitat measurements in this study and fish abundance and distribution with the exception of site 4, where lack of fish was associated with lack of water. Other than site 4, this may indicate 1) a similarity among tributaries, 2) that fish are not limited by habitat in these streams, or 3) differences are small enough to be masked by variability from site to site. The cold waters of Hayward and Holmes Brooks may limit food abundance necessary to support large populations of fish. Correlation between fish and habitat are more likely to exist following forest harvest if habitat quality is significantly changed. The post-harvest data of 1996 may reflect the conditions in the first year after harvest.

Examining the effects of forestry practices on fish requires an ecosystem approach. This entails understanding the biological communities of the stream, availability of resources, and physical and chemical composition of the environment (Murphy and Meehan, 1991). On small to mid-sized streams, riparian zones have been found to reduce temperatures and sediment input, provide important sources of organic matter and stabilize stream banks (Osborne and Kovacic, 1993). Stream bank vegetation may also minimize damage to aquatic habitat and maintain the integrity of fish populations (Hicks et al., 1991). However, questions about the width requirements of riparian zones have not been answered.

This study suggests that further research is needed on the following questions:

i) What are the effects of selective cutting within riparian zones? This extends not only to volume of wood removed but also frequency of removal, and size and species selection.

ii) To what extent is the capacity for riparian zones to act as nutrient sinks reduced or remaining vegetation saturated by the removal of vegetation bordering streams and rivers?

iii) How may selective removal within buffer zones affect the supply of woody debris and the dynamics of woody debris input from season to season and year to year? Managers must ensure that a sufficient number of trees of the proper age class and species remain standing to eventually fall into the stream or river.

iv) What is the relationship of the riparian zone to groundwater? Groundwater stabilizes temperatures in small streams and ensures that fish eggs deposited in the gravel do not freeze during the winter. Riparian zones may have to be extended to areas known to be important sources of groundwater. Conversely, riparian zones may be smaller in watersheds where groundwater sources are mainly responsible for the flow and adequately protected.

Watershed analysis, which may be described as a systematic procedure for characterizing watershed and ecological processes to meet specific management and social objectives, would serve to address many of the above needs. Monitoring of key species such as Brook Trout and Atlantic Salmon, can provide useful information on whether biodiversity goals are being met and should be included in future plans to evaluate the effectiveness of new forestry management practices on watersheds.

Further Reading:

Parker, G., J. Pomeroy and A. Chaisson. 1997. The Hayward Brook Watershed Study ( a research project of the Fundy Model Forest): Interim report (1993-1995). Fundy Model Forest, Sussex, N.B.

- Table of Contents

- View other case studies

- Bibliography