Evaluating sampling techniques for low-density populations of arctic grayling (Thymallus arcticus).
Key words: Alberta foothills, angling, Arctic Grayling, backpack electrofishing, low-density, mark-recapture, probability of detection, sampling techniques, three-pass removal, Thymallus arcticus
Increased land use from expanding energy and forestry development, habitat fragmentation due to improperly installed and maintained watercourse crossing structures, and increased angler access and angling pressure have resulted in extreme declines of Arctic Grayling (Thymallus arcticus) populations across the province of Alberta (Berry 1998; Blackburn and Johnson 2004; ASRD 2005). Although such declines are relatively new to Alberta, similar trends have been previously observed over much of the species' North American range. Native Arctic Grayling populations have been extirpated in eastern North America (for example, in Michigan), and populations have severely declined in the southern parts of their western range (for example, in Montana, Wyoming, and British Columbia; Northcote 1995). Since the 1950s in Alberta, the species has experienced high population reductions, with declines estimated at 90% for approximately 50% of subpopulations (ASRD 2005). Currently, Alberta Sustainable Resource Development lists Arctic Grayling as 'sensitive' (ASRD 2001).
The ability to monitor abundance and assess the status of a species allows fisheries biologists to formulate management strategies and facilitate the protection and recovery of populations. Reliable data on distribution and abundance is especially important for species at risk, for which a failure to obtain credible ecological information could result in inappropriately assigned conservation designations or management decisions (Mace 1994). Thus, reliable information on Arctic Grayling distributions and abundances are essential to their future persistence. However, the extent of previous declines, current (baseline) ranges, and the overall status of Arctic Grayling are unclear. Monitoring fish populations requires both financial and human resources, as well as a cost-effective framework upon which programs can rely (Kennard and others 2006).
Fish populations are assessed by numerous methods. Electrofishing remains one of the most widespread fisheries sampling tools and is regarded as the most effective for sampling stream fish assemblages (Bohlin and others 1989; Reynolds 1996). Electrofishing alone, however, may not be suitable for all situations. Angling is a commonly used fisheries technique, and has been successfully applied when sampling Arctic Grayling in Alberta, as well as other salmonids (Paul and others 2003; Fitzsimmons and Blackburn 2009). Another less frequently used technique to identify Arctic Grayling spawning habitat is egg-kick surveys (RL&L 1982, 1995a, 1995b). Egg-kick surveys for Arctic Grayling have the potential to be a monitoring tool (see redd counts, Gallagher and others 2007). However, difficulties often arise in conducting fisheries assessments because capture efficiencies and the vulnerability of fish life-history stages may vary with sampling gear, size, behavior, and time of sampling (Bayley and Austen 2002; Peterson and others 2004; Kennard and others 2006). Because Arctic Grayling are highly mobile, the influence of temporal variability on the success of sampling techniques is of particular importance. Adults move long distances seasonally between reaches to spawn in the spring; and young-of-the-year, juveniles, and adults will migrate to locate feeding grounds and find appropriate overwintering habitat (Ward 1951; Lucko 1992; Nelson and Paetz 1992; Stanislawski 1997). Many sites may contain a range of fish from young-of-the-year to the largest adults.
Three-pass removal and mark-recapture are commonly employed fisheries methods for estimating population abundance (Pine and others 2003); however, in Alberta the likelihood of detecting and accurately estimating Arctic Grayling abundances using these methods is largely unknown. Many authors caution against the use of the removal model because it can overestimate sampling efficiency and underestimate abundances due to declining capture efficiencies with successive removal passes (Zippin 1958; Peterson and Cederholm 1984; Peterson and others 2004; Rosenberger and Dunham 2005; Dauwalter and Fisher 2007). Although the mark-recapture method may be less susceptible to sampler and environmentally-induced biases that can alter abundance estimates (Zippin 1958; Peterson and Cederholm 1984), its success hinges on a sufficiently large amount of the population being marked and recaptured to avoid statistical bias (Ricker 1975; Jensen 1992). Adequate sample sizes, however, are often difficult to achieve for rare species (Thompson 2004).
We initiated this study given the paucity of peer-reviewed literature on Arctic Grayling sampling methods and the lack of a standardized stream sampling protocol for the species in Alberta. Our primary objective was to evaluate the common sampling techniques of backpack electrofishing and angling, and to detect Arctic Grayling and determine population estimates in wadeable streams. As part of our evaluation, we also report results from a less commonly employed technique for detecting spawning habitat through the use of egg-kick surveys; recommend the best methods as those, for example, that had the highest detection probabilities or results that met the assumptions of the analyses; and describe biases of gear type and season (early versus later summer) by comparing the structure and size catches of Arctic Grayling. We then summarize the findings of our results in a sampling decision flow diagram.
We conducted our study in wadeable tributaries of the Athabasca River in the Foothills ecoregion, near the cities of Whitecourt, Edson, and Hinton in west-central Alberta (Fig. 1) (NRC 2006). Basins with high historical Arctic Grayling abundances were chosen for our study by referring to the province-wide Fish and Wildlife Management Information System (FWMIS) database (2), reviewing reports from consultants and non-profit agencies (for example, Golder Associates Ltd, Alberta Conservation Association), and discussions with provincial fisheries biologists and private consultants with experience in sampling Arctic Grayling in Alberta. Located in the Lower Foothills and Upper Foothills subregions (NRC 2006), study sites extended south to the Embarras River (UTM: Zone 11U, 500018E, 5903604N, NAD83) and north to the Freeman River drainage (UTM: Zone 11U, 560779E, 6060116N, NAD83). The Alberta foothills have high annual precipitation, cold winters and warm summers (NRC 2006). The majority of study sites were in the Freeman River (n = 27) and Sakwatamau River (n = 36) drainages near Whitecourt. These basins had high historical Arctic Grayling populations and continue to support some of the highest Arctic Grayling densities in the area. Average discharge of the Athabasca River near study sites peaked at 600 m3/s during the spring freshet, ranging from 150 m3/s to 600 m3/s from May to August (Alberta Environment 2010). Study sites had an average bankfull channel width of 6.52 m, ranging from 0.71 m to 29.8 m. Land use in this area is dominated by forestry, oil, and gas exploration and extraction. Recreational activities include fishing, hunting, camping, and ATV use.
We sampled from May through August (2008 and 2009), and sampling consisted of a single pass of angling (July through September, 2008; June through August, 2009) and a single pass of electrofishing (May through August 2008 and 2009) to detect Arctic Grayling (Table 1). We performed a mark-recapture and three-pass removal study during late summer (July and August, 2008 and 2009) on a sub-sample of sites supporting Arctic Grayling to estimate population abundance.
[FIGURE 1 OMITTED]
In 2009 from 12 to 27 May, we performed egg-kick surveys at 21 sites in 13 sub-basins in an attempt to locate Arctic Grayling spawning areas. The sites were also electrofished and angled for Arctic Grayling (Table 1). Potential spawning sites were identified in 2nd- to 4thorder streams, described as riffle-run transitions with water depths of 0.15 to 0.5 m, water velocities of 0.35 to 0.55 m/s, and gravel-cobble substrates (Stewart and others 1982; RL&L 1995a, 1995b; Berry 1998). For ease of access, surveyed sites were located upstream and downstream within 1 km of road-stream crossing intersections. Each site consisted of 3 riffle-run transitions within 500 m of one another.
In Alberta boreal streams, Arctic Grayling spawning often coincides with the spring freshet where the water is often turbid from high rainfall, and has been observed during the first 2 weeks of May and sometimes into early June (Nelson and Paetz 1992; Berry 1998; Joynt and Sullivan 2003). Females lay 5000 to 6000 eggs/0.5 kg of body weight (Behnke 2002). The amber-colored eggs measure 2.4 to 2.7 mm dia, are easily distinguished from all other cooccurring eggs (Berry 1998; Benhke 2002), and hatch in 11 to 22 d in water temperatures 7 to 11[degrees]C (Berry 1998; Joynt and Sullivan 2003). Our surveys followed the spring freshet and turbid conditions such that we could potentially observe males occupying spawning territory prior to hatch (12 to 27 May).
We sampled for eggs in suitable spawning habitat using a D-net placed downstream of 1-m2 sampling plots (RL&L 1995b), where we disturbed the substrate for 1 min. This was repeated at 3 to 9 locations within each riffle-run transition area. We surveyed the next site upstream if no eggs were located after sampling 3 riffle-run transitions. There were up to 18 sample plots in total for a site. If eggs were located at a plot, sampling stopped immediately, and the tributary was deemed to support spawning Arctic Grayling.
We also electrofished a total of 84 sites in 15 sub-basins, of which 33 sites were also sampled by angling with dry flies during the 2008 and 2009 field seasons. An experienced 3-person crew electrofished sites for 600 m (300 m upstream and downstream of the road-stream crossing), using a model 12 or LR24 Smith-Root backpack electrofishing unit. Electrofishing proceeded from downstream to upstream to avoid silting the stream and maximize catchability. We made an effort to sample all stream habitat types (under cover, riffle, run, pools). No block nets were used. One person operated the electrofisher, while 2 people netted fish. Voltage, frequency, and duty cycle were adjusted to maximize capture efficiency without injuring any fish species (settings range: voltage = 400500 V, frequency = 40-60 Hz, duty cycle = 2025%, water conductivity = 109-478 [micro]S/cm, temperature = 7.0-16.9[degrees]C).
Sites with a large enough bankfull channel width were sampled by angling with a 3-person crew: 1 angler with >5-y experience and 2 anglers with 2- to 5-y experience. We sampled by angling with fly rods and dry flies. All 3 anglers thoroughly fished all stream habitats, and fished for the same amount of time (1 to 3 h depending on the size of the stream). Fly sizes ranged from 8 to 14. Arctic Grayling are opportunistic feeders and highly susceptible to angling (Nelson and Paetz 1992; ASRD 2005), and typically do not show a preference for specific flies (Sullivan and Johnson 1994). We used stimulators, humpies, and elk-hair caddis fly types, and 5X tippets (diameter 0.152 mm) with a leader length of 2.1-2.7 m. Dry flies were presented by dead drift and dragged through seams located in pools and deeper runs. When fish were captured, they were held live in buckets with ambient stream water and then processed every 50 m. We measured fish fork length (mm), identified the species, and returned the fish to a downstream location if they appeared healthy and uninjured. We euthanized fish that, in rare cases (2.5%), were severely injured and unable to recover.
At 9 sites confirmed to support Arctic Grayling, we attempted to calculate population estimates using mark-recapture and three-pass removal methods. Each site was approximately 300 m in length. Prior to sampling for these estimates, the upstream and downstream ends of each transect were blocked using beach seine nets with a 3.18 mm mesh, ensuring that block nets were completely secured to the streambed. After block nets were in place, we conducted an initial pass of angling with dry flies. Arctic Grayling caught during angling were marked with a clip to their adipose fin. Immediately after a run in which we marked fish collected by angling, we completed a single electrofishing pass. Fish captured from electrofishing were marked with a small clip to a pelvic fin. After the first angling and electrofishing passes, marked fish were returned to the closed section of the stream where they were captured. Following the initial marking passes and an overnight recovery period (Rosenberger and Dunham 2005), we conducted three-pass removal sampling with an electrofisher the following day. Before proceeding from 1 removal pass to another, we waited for a minimum of 1 h to allow fish to recover from electrofishing activity (Poos and others 2007). Every 50 m during electrofishing passes we recorded the number of marked captures, measured fork length (mm), and immediately released all fish (marked and unmarked) to the stream outside of the block nets. Unmarked fish captured during the three-pass removal effort were not marked. All block nets remained in position for 2 d until the depletion study was completed.
Data from 33 sites sampled using both angling and electrofishing were used to estimate occupancy ([psi]) and detection probabilities (p) for Arctic Grayling using the program PRESENCE (version 3.1; Hines 2006). Egg-kick surveys were performed at 21 of these sites. Probabilities of detection were estimated from encounter histories over all sites using a maximum likelihood function (MacKenzie and others 2002). Bankfull channel width (m) was used as a covariate. For our analysis we assumed that species presence and detection probabilities were constant across time and sites, [psi](.) p(.). We assumed sites were closed spatially and temporally to changes in occupancy. Backpack electrofishing and angling occurred in the same day. In addition, although the timing of egg-kick surveys differed, the goal was to detect non-mobile Arctic Grayling eggs. We explored how p, the probability that a species will be detected at a site (given its presence) independent of local abundance and density, differed between backpack electrofishing, angling, and egg-kick surveys.
We then explored possible biases of gear type and sampling season by comparing the size structure of Arctic Grayling catches using analysis of variance (ANOVA, [alpha] = 0.05). Average fork length at a site (mm) was the dependent variable; independent variables were categorical and included early summer (May-June) versus late summer (July-August), year 2008 versus 2009, and gear type (electrofishing versus angling). We also investigated a variety of potential interactions.
To evaluate sampling methods for monitoring the abundance of low-density Arctic Grayling populations, we considered minimum thresholds for the successful application of each technique. We deemed the mark-recapture method unsuccessful and susceptible to systematic statistical bias if <4 fish were recaptured during successive passes at a site (Ricker 1975). Further, a three-pass removal site was not considered for analysis if the efficiency of capture (p) was [less than or equal to] 0.20 because abundance estimates become biased and unreliable (Lockwood and Schneider 2000). Lockwood and Schneider (2000) describe capture efficiency as: p = T / (kN-X), where T = the total number of fish caught from all passes, k - the number of removal passes, N = the absolute abundance of fish, and X = the intermediate statistic as determined by the number of fish captured in successive passes (Lockwood and Schneider 2000).
Arctic Grayling Detection
Of the sites electrofished (n = 84) and angled (n = 33), Arctic Grayling were located at only 25 different sites. Using electrofishing and angling, we captured a total of 277 Arctic Grayling in the Freeman River, Sakwatamau River, Wolf Creek, Sundance Creek, Pinto Creek, Chickadee Creek, Two Creek, and Marsh Head Creek drainages (Fig. 1). At sites where both angling and electrofishing were performed, we detected Arctic Grayling at 19 of 33 sites for a naive occupancy estimate of 0.56; the naive estimate is occupancy given perfect detectability. The estimated probability of occupancy was 0.75 ([S.sub.[x.bar]] [+ or -] 9.77). Because the estimated probability of occupancy is >25% larger than the naive occupancy estimate, this suggests that Arctic Grayling were never detected at 1 in every 4 occupied sites. The probability of detecting Arctic Grayling using backpack electrofishing was 0.68 ([S.sub.[x.bar]] [+ or -] 0.11). The probability of detecting Arctic Grayling where angling was performed was 1.00 ([S.sub.[x.bar]] [+ or -] 0.00). Because angling always detected fish at sites where they were captured, the probability of detection was 1.00. Egg-kick surveys had the lowest probability of detecting Arctic Grayling eggs (p = 0.13, [S.sub.[x.bar]] [+ or -] 0.12). At the 21 sites surveyed using the egg-kick survey method, we located a total of 20 Arctic Grayling eggs at I site only, Sundance Creek. Of the 3 Arctic Grayling detection techniques evaluated, backpack electrofishing and angling had the highest Arctic Grayling detection probabilities and were retained for further analysis in this study.
Arctic Grayling Size Selectivity: Influence of Gear Type and Temporal Variation
At sites confirmed to support Arctic Grayling using at least 1 sampling method, we captured on average 0.37 fish/100 s of backpack electrofishing (range: 0 to 4.63 fish/100 s). Using angling we caught an average of 1.36 fish/h of angling (range: 0 to 4.31 fish/h). Size structures of catches differed significantly by gear type (ANOVA, [F.sub.(1,33)] = 10.14, P = 0.004) and between summer seasons (ANOVA, [F.sub.(1,33)] = 5.99, P = 0.021; Fig. 2). The year of survey did not significantly influence the size of fish captured (ANOVA, [F.sub.(1,33)] = 0.04, P = 0.852). Further, no interactions between year, season, and gear type were significant. Although angling caught larger fish ([bar.x] = 173 mm, s = 39 mm, range = 100 to 285 mm), angling frequently failed to capture young-of-the-year (Fig. 2). In contrast, electrofishing captured a broader size range ([bar.x] = 92 mm, s = 42 ram, range = 23 to 282 mm), including young-of-the-year (Fig. 2). For instance, the number of larger (>110 ram) Arctic Grayling captured/km by angling was 3.1 times greater than by backpack electrofishing. The average fork length decreased in late summer (July to August; [bar.x] = 119 mm, s = 56 mm, range = 23 to 285 mm) compared to early summer ([bar.x] = 144 mm, s = 55 mm, range = 73 to 282 mm).
[FIGURE 2 OMITTED]
Arctic Grayling Abundance Estimates
We recaptured very few fish in the markrecapture and three-pass removal study (n = 20; Table 2). Our results demonstrated that abundance estimates of Arctic Grayling using the mark-recapture techniques are likely unreliable as >4 fish were recaptured at only 2 of the 9 sites (Ricker 1975; Table 3). We therefore rejected the mark-recapture technique as a potential method to estimate low-density Arctic Grayling abundances in our study streams. In addition, at only 2 of the 9 three-pass depletion sites was a capture efficiency (p) of Arctic Grayling greater than our minimum threshold of p = 0.20 (Table 3). As a result, we considered abundance estimates from this method generally biased and unreliable.
Overall Evaluation of Sampling Techniques
Given our findings, we created a decision flow diagram to guide the selection of sampling techniques to detect and determine the catch-per-unit-effort and abundance estimates of Arctic Grayling (Fig. 3). Additionally, we addressed the detection of spawning habitat. To detect low-density Alberta stream populations, managers must decide if they need to confirm occupancy of young-of-the-year ([less than or equal to]110 mm) or juvenile and adult (>110 mm) fish, because the gear type and timing of sampling will vary. Backpack electrofishing more reliably detected small young-of-the year fish; angling with dry flies most reliably detected juvenile and adult fish. Backpack electrofishing did detect some juveniles and adults, but only with moderate efficiency. Young-of-the-year were only detected in late summer (July and August). Using egg-kick surveys, we rarely detected spawning habitat despite a large amount of sampling effort for the low-density grayling populations in the region.
Despite declining Arctic Grayling populations in Alberta, sampling and monitoring has been inconsistent throughout the province and has only recently been under evaluation. A fundamental step in Alberta's management and conservation depends on the availability of accurate and consistent detection and population estimate monitoring protocols. The results presented in our study will assist in providing resource managers with the information required to efficiently assess Arctic Grayling distributions in wadeable tributary streams of the Athabasca River.
In our comparison of 3 Arctic Grayling sampling techniques, we recommend using angling and backpack electrofishing because they had the highest probability of detecting occupancy. Although electrofishing is often the preferred sampling method in wadeable streams (Bohlin and others 1989; Bonar and others 2009), our results indicate that angling using dry flies could be an equally effective or complimentary sampling method, in contrast, egg-kick surveys appeared to be the least successful of the 3 methods, despite the fact that the method successfully detected high egg densities in House River and Hartley Creek in northeastern Alberta (RL&L 1982, 1995a, 1995b). In our study we may have failed to sample appropriate spawning sites, identify precise spawning times, or sample larger portions of the study basins, all of which may be required to enhance the detectability of Arctic Grayling using the egg-kick survey method. Given these uncertainties and the potential damage to spawning grounds, we recommend that egg-kick surveys not be used to identify Arctic Grayling spawning habitat in low-density populations unless absolutely necessary. We also suggest that managers not use mark-recapture and three-pass depletion methods for assessing abundance estimates of low-density Arctic Grayling populations unless the minimum data thresholds can be met for unbiased and reliable estimates. This is especially critical for low-density fish populations like Arctic Grayling on the fringes of their southern range, because the conservation and management of fish species typically requires knowledge of the distributions of existing populations (Peterson and Bayley 2004).
[FIGURE 3 OMITTED]
In our investigation of the biases of the sampling methods, we found that the effectiveness of electrofishing and angling differed temporally and by the size of Arctic Grayling. Electrofishing was the most efficient method to capture small young-of-the-year, while juvenile and adult fish were more susceptible to capture by angling with dry flies. Contrary to our findings, numerous electrofishing studies have found that larger fish are easier to capture (Bayley and Dowling 1993; Peterson and others 2004; Buckmeier and Schlechte 2009). Theoretically, electrofishing efficiency should increase with fish length (Bohlin and others 1989). For example, Peterson and others (2004) found that electrofishing capture efficiency was greatest for the largest size classes of Bull Trout (Salvelinus confluentus) and Westslope Cutthroat Trout (Oncorhynchus clarkii lewisi). In practice, however, differences in fish behavior and the surrounding habitat are more important for capture efficiency (Bohlin and others 1989). The differences of our findings may be a result of behavioral avoidances of the electrical field by Arctic Grayling. Upon an electrofishing disturbance, juvenile and adult Arctic Grayling were observed rapidly swimming away either upstream or downstream without stopping to conceal themselves (L. MacPherson pers. obs.). Similarly, Ernst and Nielsen (1981) found that European Arctic Grayling (Thymallus thymallus) avoided the electrical field, resulting in very few captures. In contrast to grayling, we often observed Brook Trout (Salvelinus fontinalis) and Rainbow Trout (Oncorhynchus mykiss) avoiding capture by concealing themselves under larger boulders or undercut banks, making them easier to capture (L MacPherson, pers. obs.). In our study we suspect that angling was a more successful method for detecting and capturing juvenile and adult Arctic Grayling because field crews could approach potential holding areas while not disturbing the water and fish in the vicinity. Further, Fitzsimmons and Blackburn (2009) found that in the Little Smoky River in west-central Alberta, Arctic Grayling capture efficiencies were extremely low using a boat electrofisher. Consequently, they relied solely on angling to determine abundances. Conceivably, increasing the voltage, frequency, and pulse width while backpack electrofishing may improve the capture efficiencies of larger Arctic Grayling, but it would be at the risk of seriously injuring or killing smaller Arctic Grayling and other stream fish species.
We found that mark-recapture and three-pass removal methods often failed to meet the literature-derived minimum requirements for generating reliable population estimates (Ricker 1975; Lockwood and Schneider 2000). Often, we recaptured very few if any Arctic Grayling during the mark-recapture study (Table 2). Capture efficiencies (p) were extremely low. As a result, we were unable to estimate abundances, or abundance estimates were likely subject to substantial statistical bias (Ricker 1975; Lockwood and Schneider 2000). Increasing the sampling length of the study sites and the number of electrofishing units and samplers may have helped increase detectability and the reliability of our three-pass removal abundance estimates. This, however, would have been more costly and time consuming. Perhaps abundance estimates derived from the mark-recapture method would have been more reliable if only angling with dry flies was used to sample stream reaches, as done by Fitzsimmons and Blackburn (2009). Such sampling techniques warrant further investigation. Information on capture efficiency (p) by gear type and fish size would be useful, and allow fisheries managers to correct catch data and provide a more reliable index of population sizes of Arctic Grayling in Athabasca River tributary streams.
Although not explored in our study due to timeline and budgetary constraints, snorkel studies have been used successfully to determine salmonid population estimates (Christie and others 2010; Hagen and others 2010). Underwater observations could be especially advantageous as a cost-effective sampling method to determine spawning, distribution, and abundance of Arctic Grayling. For instance, Christie and others (2010) found that snorkeling surveys in mid-sized Alberta foothill streams proved to be an effective technique for estimating Arctic Grayling population sizes. In addition, snorkeling may prove to be less intrusive methods than backpack electrofishing, which risks injuring fish.
Our results have established that researchers must not only be conscientious when choosing Arctic Grayling sampling gear, but they also need to consider the timing of sampling. Often researchers are hindered by economic constraints, which results in the use of a single-gear sampling method without ensuring its effectiveness (Poos and others 2007). Our study, however, demonstrated that sampling Arctic Grayling using a single sampling gear would be less effective than using multiple types of gear because the use of a single gear would likely exclude sampling entire life-history stages. Fisheries managers should therefore choose the sampling method that will best achieve the objectives of their study or monitoring effort. Based on the results of our study, we were able to develop a decision flow diagram (Fig. 3) to guide future sampling efforts for efficiently and reliably detecting and assessing Arctic Grayling abundances; allocation of sampling effort toward the most efficient gears with the highest detection probabilities can minimize the variance in fish collection and lead to a more effective monitoring program (Peterson and Rabeni 1995).
Undoubtedly, there will be natural variations in Arctic Grayling densities among our Athabasca River tributary study sites and throughout Alberta. However, sampled areas in this study have undergone high forest and energy sector development and have extensive culvert networks that have likely resulted in low Arctic Grayling densities (MacPherson and others 2012). Sampled streams were generally on the southern limit of Alberta's Arctic Grayling range where population declines have been most pronounced. The difficulties we experienced in detecting and capturing Arctic Grayling are likely due in part to this decline. Therefore, our outlined sampling protocols should be used as a guide until similar region-specific studies have also been undertaken. In areas with higher densities of Arctic Grayling, sampling to determine population estimates using mark-recapture and three-pass removal techniques may be feasible. Our research, however, emphasizes some of the challenges in formulating effective stream sampling protocols for monitoring declining species characterized by low densities and patchy distributions.
This project represents a portion of the MSc thesis work of Laura MacPherson in collaboration with the Alberta Conservation Association (ACA), Alberta Sustainable Resource Development (ASRD) and the University of Alberta, Department of Renewable Resources (U of A). All fieldwork affecting fish was conducted under peer reviewed animal care protocols approved by the Canadian Council on Animal Care (#585806 2007-2009). The results of this study are also presented as a report in the ACA report series. Thank you to S Spencer, D Hildebrandt, O Watkins, S Stambaugh (ASRD), T Furukawa (ACA), and S Sullivan (U of A) for all their hard work during field sampling. A special thank you to S Spencer, D Park (ASRD), and P Aku (ACA) for guidance in project design and sampling.
* ALBERTA ENVIRONMENT. 2010. Alberta river basins. Government of Alberta. Available from: http:// www.environment.alberta.ca/forecasting/reports/ index.html [accessed 2 February 2011].
[ASRD] ALBERTA SUSTAINABLE RESOURCE DEVELOPMENT.2001. The general status of Alberta wild species 2000. Edmonton, AB: Alberta Sustainable Resource Development, Fish and Wildlife Service. 46 p.
[ASRD] ALBERTA SUSTAINABLE RESOURCE DEVELOPMENT. 2005. Status of the Arctic Grayling (Thymallus arcticus) in Alberta. Wildlife Status Report No. 57. Edmonton, AB: Alberta Sustainable Resource Development, Fish and Wildlife Service. 41 p.
BAYLEY PB, AUSTEN DJ. 2002. Capture efficiency of a boat electrofisher. Transactions of the American Fisheries Society 131:435-451.
BAYLEY PB, DOWLING DC. 1993. The effect of habitat in biasing fish abundance and species richness estimates when using various sampling methods in streams. Polskie Archiwum Hydrobiologii 40:5-14.
BEHNKE RJ. 2002. Trout and salmon of North America. New York, NY: The Free Press. 359 p. BERRY DK. 1998. Alberta's Arctic Grayling management and recovery plan. Edmonton, AB: Alberta Environmental Protection, Natural Resources Service. 27 p.
BLACKBURN M, JOHNSON CF. 2004. Status and distribution of Arctic Grayling (Thymallus arcticus) in the Pembina River. Edson, AB: Technical Report, T2004-003, Alberta Conservation Association. 85 p.
BOHLIN T, HAMRIN S, HEGGBERGET TG, RASMUSSEN G, SALTVEIT SJ. 1989. Electrofishing--theory and practice with special emphasis on salmonids. Hydrobiologia 173:9-43.
BONAR SA, HUBERT WA, WILLIS DW. 2009. Standard methods for sampling North American freshwater fishes. Bethesda, MD: America Fisheries Society. 335 p.
BUCKMEIER DL, SCHLECHTE JW. 2009. Capture efficiency and size selectivity of Channel Catfish and Blue Catfish sampling gears. North American Journal of Fisheries Management 29:404-416.
CHRISTIE K, MCCLEARY R, OUELLET S. 2009. Estimating Arctic Grayling in mid-size streams with night snorkeling. Fish and Watershed Program, Foothills Research Institute. 32 p.
DAUWALTER DC, FISHER WL. 2007. Electrofishing capture probability of Smallmouth Bass in streams. North American Journal of Fisheries Management 27:162-171.
ERNST ME, NIELSEN J. 1981. Populationsdynamiske undersogelser Over Stalling (Thymallus thymallus (L.) i ovre Gudena [thesis]. Denmark: University of Aarhus. 159 p.
FITZSIMMONS K, BLACKBURN M. 2009. Abundance and distribution of Arctic Grayling in the upper Little Smoky River, Alberta, 2007. Sherwood Park, AB: Alberta Conservation Association, Conservation Series Report 1/347. 22 p.
GALLAGHER SP, HAHN PKJ, JOHNSON DH. 2007. Redd counts. In: Johnson DH, Shier BM, O'Neal JS, Knutzen JA, Augerot X, O'Neil TA, Pearsons TN, editors. Salmonid field protocols handbook: Techniques for assessing status and trends in salmon and trout populations. Bethesda, MD: American Fisheries Society. p 197-234.
HAGEN J, DECKER S, KORMAN J, BISON RG. 2010. Effectiveness of night snorkeling for estimating steelhead parr abundance in a large river basin. North American Journal of Fisheries Management 30:1303-1314.
* HINES JE. 2006. PRESENCE2-Software to estimate patch occupancy and related parameters. USGSPWRC. Available from: www.mbr-pwrc.usgs.gov/ software/presence.html [accessed 2 February 2011].
JENSEN AL. 1992. Integrated area sampling and markrecapture experiments for sampling fish populations. Biometrics 48:1201-1205.
JOYNT A, SULLIVAN MG. 2003. Fish of Alberta. Edmonton, AB: Lone Pine Publishing. 176 p.
KENNARD MJ, PUSEY BJ, HARCH BD, DORE E, ARTHINGTON AH. 2006. Estimating local stream fish assemblage attributes: Sampling effort and efficiency at two spatial scales. Marine & Freshwater Research 57:635-653.
LOCKWOOD RN, SCHNEIDER JC. 2000. Stream fish population estimates by mark-and-recapture and depletion methods. In: Schneider JC, editor. Chapter 7: Manual of fisheries survey methods II: with periodic updates. Ann Arbor, MI: Michigan Department of Natural Resources, Fisheries Special Report 25. p 1-14.
LUCKO B. 1992. Distribution of young of the year Arctic Grayling in the Swan River drainage. Peace River, AB: Alberta Fish and Wildlife Division. 19 p.
MACE GM. 1994. Classifying threatened species: Means and ends. Philosophical Proceedings of the Royal Society of London Series B 344:91-97.
MACKENZIE DI, NICHOLS JD, LACHMAN GB, DROEGE S, ROYLE JA, LANGTIMM CA. 2002. Estimating site occupancy rates when detection probabilities are less than one. Ecology 83:2248-2255.
MACPHERSON LM, SULLIVAN MG, FOOTE AL, STEVENS CE. 2012. Effects of culverts on stream fish assemblages in the Alberta foothills. North American Journal of Fisheries Management 32:480-490.
NELSON JS, PAETZ MJ. 1992. The fishes of Alberta. Edmonton, AB: The University of Alberta Press. 437 p.
NORTHCOTE TG. 1995. Comparative biology and management of Arctic and European Grayling (Salmonidae, Thymallus). Reviews in Fish Biology and Fisheries 5:141-194.
[NRC] NATURAL REGIONS COMMITTEE. 2006. Natural regions and subregions of Alberta. Report No. T/ 852. Edmonton, AB: Government of Alberta. 254 p.
PAUL AJ, POST JR, STELFOX JD. 2003. Can anglers influence the abundance of native and nonnative salmonids in a stream from the Canadian Rocky Mountains? North American Journal of Fisheries Management 23:109-119.
PETERSON JP, BAYLEY PB. 2004. A Bayesian approach to estimating presence when a species is undetected. In: Thompson WL, editor. Sampling rare or elusive species. Concepts, designs, and techniques for estimating population parameters. Washington, DC: Island Press. p 173-188.
PETERSON JT, RABENI CF. 1995. Optimizing sampling effort for sampling warmwater stream fish communities. North American Journal of Fisheries Management 15:528-541.
PETERSON JT, THUROW RF, GUZEVICH JW. 2004. An evaluation of multipass electrofishing for estimating the abundance of stream-dwelling salmonids. Transactions of the American Fisheries Society 133: 462-475.
PETERSON NP, CEDERHOLM CJ. 1984. A comparison of the removal and mark-recapture methods of population estimation for juvenile Coho Salmon in small streams. North American Journal of Fisheries Management 4:99-102.
PINE WW, WIEMELS JL, JAYABOSE S, SANDOVAL C. 2003. A review of tagging methods for estimating fish population size and components of mortality. Fisheries 28:10-23.
POOS MS, MANDRAK NE, MCLAUGHLIN RL. 2007. The effectiveness of two common sampling methods for assessing imperiled freshwater fishes. Journal of Fish Biology 70:691-708.
REYNOLDS JB. 1996. Electrofishing. In: Murphy BR, Willis DW, editors. Fisheries Techniques. Bethesda, MD: American Fisheries Society. p 221-253.
RICKER WE. 1975. Computation and interpretation of biological statistics of fish populations. Fisheries Research Board of Canada, Bulletin 191. 400 p.
* [RL&L] RL&L ENVIRONMENTAL SERVICES LTD. 1982. Aquatic investigations in the Hartley Creek Area, 1981. Report to Sandalta Project, Mineable Oil Sands Division, and Gulf Canada Resources Inc., Calgary, AB. Available from: Golder Associates, 300-10525 170 Street, Edmonton, AB T5P 4W2. 160 p.
* [RL&L] RL&L ENVIRONMENTAL SERVICES LTD. 1995a. Upstream fish movements and population densities in Sundance Creek, Alberta, 1993. Report to Trout Unlimited Canada and Alberta Fish and
Wildlife Services, Alberta. Available from: Golder Associates, 300-10525 170 Street, Edmonton, AB T5P 4W2. 48 p.
* [RL&L] RL&L ENVIRONMENTAL SERVICES LTD. 1995b. Arctic Grayling spawning study in he House River drainage of northeastern Alberta, 1994. Report to Alberta Environmental Protection, Fish and Wildlife Services, Lac La Biche, Alberta. Available from: Golder Associates, 300-10525 170 Street, Edmonton, AB T5P 4W2.34 p.
ROSENBERGER AE, DUNHAM JB. 2005. Validation of abundance estimates from mark-recapture and removal techniques for Rainbow Trout captured by electrofishing in small streams. North American Journal of Fisheries Management 25:1395-1410.
STANISLAWSK1 S. 1997. Fall and winter movements of Arctic Grayling (Thymallus arcticus) in the Little Smoky River [thesis]. Edmonton, AB: University of Alberta. 91 p.
* STEWART RJ, MCLENEHAN RE, MORGAN JD, OLMSTED WR. 1982. Ecological studies of Arctic Grayling (Thymallus arcticus), Dolly Varden Char (Salvelinus malma), and Mountain Whitefish (Prosopium williamsoni) in the Liard River drainage, BC. Report of EVS Consultants Ltd for Westcoast Transmission Company Ltd and Foothills Pipe Line Ltd. Vancouver, BC. Available from: Golder Associates, 300-10525 170 Street, Edmonton, AB T5P 4W2. 99 p.
* SULLIVAN MG, JOHNSON C. 1994. Arctic Grayling sampled from Little Smoky River, August 1994. Alberta Sustainable Resource Development, Fish and Wildlife. Memorandum to Carl Hunt, Regional Fisheries Biologist, Alberta. Available from: Fish and Wildlife, 111-54 Street, Edson, AB T7E 1T2. 8 p.
THOMPSON WL, editor. 2004. Sampling rare or elusive species. Concepts, designs, and techniques for estimating population parameters. Washington, DC: Island Press. 429 p.
WARD JC. 1951. The biology of Arctic Grayling in the southern Athabasca drainage [thesis]. Edmonton, AB: University of Alberta. 64 p.
ZIPPIN C. 1958. Removal method of population estimation. The Journal of Wildlife Management 22:82-90.
Submitted 16 February 2011, accepted 9 December 2011. Corresponding Editor: James Orr.
(1) Present address: Alberta Sustainable Resource Development. Box 1500, 5226-53rd Ave., High Prairie, AB T0G 1E0; email: firstname.lastname@example.org
(2) The FWMIS database is available from [http:// www.srd.alberta.ca/FishWildlife/FWMIS/Default. aspx]
LAURA M MACPHERSON (1)
Department of Renewable Resources, University of Alberta, Edmonton, AB T6G 2H1
MICHAEL G SULLIVAN
Alberta Sustainable Resource Development, Fish and Wildlife Division, Edmonton, AB T6H 4P2
Department of Renewable Resources, University of Alberta, Edmonton, AB T6G 2H1
CAMERON E STEVENS
Golder Associates Ltd., 10525-170 Street, Edmonton, AB T5P 4W2
TABLE 1. Number of sites angled, electrofished, and surveyed for eggs in Athabasca River tributary streams of Alberta during the summers of 2008 and 2009. Tributaries where Arctic Grayling (ARGR) and Arctic Grayling eggs were detected are denoted with an 'X'. Numbers of sites in parentheses indicate the number of sites re-sampled for the mark-recapture and three-pass removal assessment. # of sites # of sites ARGR Tributary Name Closest city angled detected? Baseline Creek Hinton 1 1 Canyon Creek Hinton 1 1 Chickadee Creek Whitecourt 3 (2) 3 (2) X Embarras River Edson 3 2 Fox Creek Hinton 1 1 Freeman River Whitecourt 27 (5) 7 (5) X Marsh Head Creek Whitecourt 2 1 X Oldman Creek Hinton 2 2 Pinto Creek Hinton 2 2 X Prest Creek Edson 1 0 Sakwatamau River Whitecourt 36 (2) 4 (2) X Sundance Creek Edson 4 3 X Swartz Creek Edson 1 0 Two Creek Whitecourt 1 1 X Unnamed Tributary Hinton 1 1 Unnamed Tributary Hinton 1 1 Windfall Creek Whitecourt 1 0 Wolf Creek Edson 3 3 X # of sites ARGR # of egg Eggs Tributary Name electrofished detected? kick sites detected? Baseline Creek 1 1 Canyon Creek 1 1 Chickadee Creek 3 (2) X 3 Embarras River 2 3 Fox Creek 1 1 Freeman River 27 (5) X 0 Marsh Head Creek 1 2 Oldman Creek 2 1 Pinto Creek 1 0 Prest Creek 0 1 Sakwatamau River 36 (2) X 0 Sundance Creek 3 3 X Swartz Creek 0 1 Two Creek 1 X 0 Unnamed Tributary 1 0 Unnamed Tributary 1 1 Windfall Creek 0 1 Wolf Creek 3 2 TABLE 2. Number of Arctic Grayling captured during the mark-recapture (recap = recapture) and three-pass removal study in wadeable tributaries in the Athabasca River Basin, Alberta. Study sites: FM1 = Freeman River; JC1, JC2 = Judy Creek; SK1, SK2 = Sakwatamau River; CC1, CC2 = Chickadee Creek; TC1 = Two Creek. Electrofishing Angle Marking Site Start Date End Date Marking Run Run FM1 08/21/08 08/22/08 3 1 JC1 08/22/08 08/23/08 59 10 SK1 08/24/08 08/25/08 8 9 SK2 08/26/08 08/27/08 10 3 CC1 07/24/09 07/25/09 0 3 TC1 07/26/09 07/27/09 2 4 JC2 07/28/09 07/29/09 28 7 CC2 08/05/09 08/06/09 5 17 Electrofishing Angle Recap Site Pass #2 Recap Pass #2 Pass #2 Pass #3 FM1 6 1 0 6 JC1 36 2 0 14 SK1 5 0 0 5 SK2 7 2 0 2 CCl 0 0 0 0 TC1 0 0 0 0 JC2 31 3 1 19 CC2 3 0 0 5 Angle Electrofishing Recap Site Pass #1 Recap Pass #1 Pass #1 FM1 9 1 0 JC1 44 2 1 SK1 5 0 0 SK2 10 1 0 CC1 2 0 0 TC1 0 0 0 JC2 30 1 1 CC2 2 0 0 Electrofishing Angle Recap Total Total Total Site Recap Pass #3 Pass #3 Marked Captured Recap FM1 0 0 4 21 2 JC1 3 0 69 94 8 SK1 0 0 17 15 0 SK2 1 0 13 19 4 CC1 0 0 3 2 0 TC1 0 0 6 0 0 JC2 0 0 35 80 6 CC2 0 0 22 10 0 TABLE 3. Total number of Arctic Grayling recaptured from mark-recapture, and the efficiency of capture (p) of Arctic Grayling captured in the three-pass depletion study in wadeable tributaries in the Athabasca River basin. Sites which met the minimum threshold for reliable and unbiased abundance estimates for mark recapture (>4 recaptures; Ricker 1975), and three-pass removal (p> 0.20; Lockwood and Schneider 2000) are denoted with an 'X'. Study sites: FM1, FM3 = Freeman River; JC1, JC2 = Judy Creek; SK1, SK2 = Sakwatamau River; CC1, CC2 = Chickadee Creek; TC1 = Two Creek. Site Start date End date Total # Threshold Probability of recaptures met capture (p) FM1 08/22/08 08/23/08 2 0.00 JC1 08/23/08 08/24/08 8 X 0.41 SKI 08/25/08 08/26/08 0 0.04 SK2 08/27/08 08/28/08 4 0.03 FM3 09/06/08 09/07/08 1 0.01 JC2 07/29/09 07/30/09 6 X 0.25 CC2 08/06/09 08/07/09 0 0.02 CC1 07/25/09 07/26/09 0 0.00 TC1 07/27/09 07/28/09 0 0.01 Site Threshold met FM1 JC1 X SKI SK2 FM3 JC2 X CC2 CC1 TC1
|Printer friendly Cite/link Email Feedback|
|Author:||MacPherson, Laura M.; Sullivan, Michael G.; Foote, Lee; Stevens, Cameron E.|
|Publication:||Northwestern Naturalist: A Journal of Vertebrate Biology|
|Date:||Sep 22, 2012|
|Previous Article:||Nest-site characteristics of three woodpecker species in managed Ponderosa Pine forests of the eastern Cascade Range.|
|Next Article:||Recent trends in Western Screech-Owl and Barred Owl abundances on Bainbridge Island, Washington.|