Abstract

Grass Carp Ctenopharyngodon idella have been intentionally stocked for aquatic vegetation control across the Midwestern United States for several decades. During the 1970s, escapement of Grass Carp into the Missouri River facilitated their naturalization into much of the Mississippi River basin, including the Upper Mississippi River. Lock and Dam 19 (LD19) in Keokuk, Iowa, is a high-head dam that represents a focal point for naturalized Grass Carp management where populations may differ between upstream and downstream pools as result of limited upstream migration, but potential differences between populations have yet to be evaluated to the best of our knowledge. The objective of this study was to compare the relative abundance, size structure, condition, growth, and recruitment variability of Grass Carp collected upstream and downstream of LD19. We sampled Grass Carp monthly (April–October) during 2014 and 2015 from four locations in the Des Moines River (downstream of LD19) and five locations throughout the Skunk, Iowa, and Cedar rivers (upstream of LD19) using boat electrofishing and trammel net sets. We captured 29 Grass Carp upstream of LD19 compared with 179 individuals captured downstream. Trammel nets only captured Grass Carp downstream of LD19; trammel net catch per unit effort upstream of LD19 was low and ranged from 0.0 to 8.0 fish/net lift (mean ± SE = 0.39 ± 0.13). Electrofishing catch per unit effort ranged from 0.0 to 22.7 fish/h (1.49 ± 0.30) and was higher downstream (2.42 ± 0.30) of LD19 than upstream (0.57 ± 0.07). Grass Carp downstream of LD19 tended to be smaller, younger, of lower body condition, had higher mortality rates, and were slower growing compared with those collected upstream and to populations documented in other systems. Understanding and monitoring adult Grass Carp population characteristics upstream and downstream of LD19 is necessary to determine how they may change in response to ongoing harvest efforts for invasive carps in these river reaches.

Introduction

Native to the region between northern Vietnam and the Amur River basin in southern Siberia, Grass Carp Ctenopharyngodon idella are now common throughout much of the world (Cudmore and Mandrak 2004). Grass Carp <200 mm in length typically feed on chironomidae larvae and larger zooplankton (Cladocera and Copepoda; Opuszynski 1968; Watkins et al. 1981) but Grass Carp >270 mm in length are primarily herbivores, where micro- and macroflora material comprise the majority of diets (Michewicz et al. 1972; Opuszynski 1972; Colle et al. 1978). Adult Grass Carp diets rarely deviate from plant material unless food resources become scarce (Bain 1993), which can significantly reduce aquatic plant biomass (Bettoli et al. 1993; Schramm and Brice 2000). Reductions of micro- and macroflora resulting from Grass Carp herbivory can result in limnological changes, including increased nutrient and phytoplankton concentrations and algal biomass, and decreased water clarity (Maceina et al. 1992). These limnological changes may also affect native fishes through indirect pathways (e.g., reduction or elimination of micro- and macroflora) rather than directly (e.g., competition or predation). For example, native fishes that use vegetated littoral areas as predator refuge, for foraging, or as nursery habitats have experienced decreased population biomass (Shireman and Smith 1983; Chilton and Muoneke 1992) as well as increased mortality via double-crested cormorant Phalacrocorax auritus predation (Hubert 1994) following Grass Carp introduction.

Grass Carp show strong preference for aquatic vegetation, and so were commonly stocked throughout the United States from the 1960s to the early 1990s for biological control of aquatic plants (Mitchell and Kelly 2006; Kelly et al. 2011). Early stocking efforts targeted lakes or reservoirs that were open to stream or river systems, particularly throughout Arkansas, USA, and by the 1970s there were numerous reports of Grass Carp in the Missouri River (Courtenay et al. 1984). The ability of Grass Carp to migrate long distances (e.g., Gorbach and Krykhtin 1988) and tolerate a wide range of environmental conditions (e.g., Opuszynski 1967; Bettoli et al. 1985; Trimm et al. 1989) has since facilitated their introduction and naturalization into many large Midwestern rivers (Pflieger 1978), including the Upper Mississippi River (UMR) watershed (Pflieger 1978; NAS 2019). Naturalized populations of Grass Carp occur throughout much of the UMR watershed, particularly in river sections bordering Missouri, Illinois, and Iowa (Camacho 2016; Larson et al. 2017; NAS 2019). However, Grass Carp populations have yet to reach high densities, unlike other invasive carps inhabiting the UMR (e.g., Silver Carp Hypophthalmichthys molitrix;Irons et al. 2007; Kolar et al. 2007; Sullivan et al. 2017). For example, fish community monitoring efforts from 1990 to 2017 in Mississippi River Pools 8, 13, and 26 using pulsed-DC boat electrofishing captured only 260 Grass Carp (<0.001% of total catch in number; LTRM 2019). In addition to low densities, Grass Carp are difficult to capture (Wanner and Klumb 2009a; Sullivan et al. 2019). As a result, limited information exists on naturalized Grass Carp populations despite their widespread distribution throughout the UMR watershed and negative effects (e.g., Maceina et al. 1992) on invaded ecosystems.

Throughout the UMR, a series of 29 lock and dams have been erected that regulate river discharge and form a series of slow-moving pools that are more lentic than the historical natural lotic discharge regime. Most notably, Lock and Dam 19 (LD19; Keokuk, Iowa) is a high-head dam that controls water levels at all flows, whereas most other UMR lock and dams (with the exception of Lock and Dam 1 near St. Paul, Minnesota) are tainter and roller gates or a series of tainter gates that create a more free-flowing river when their gates are open (Knights et al. 2002). Spillways at LD19 are unique in that they are elevated approximately 6 m above the downstream river surface (Wilcox et al. 2004), creating a semipermanent barrier, particularly when completely open (i.e., velocity barrier), to upstream migration except through the navigation lock chambers (Tripp et al. 2014). Lock and Dam 19 represents the farthest downstream UMR dam where fish movement upstream is restricted to the navigation locks, and invasive carp density is much higher downstream of LD19 than upstream as a result (Camacho et al. 2016; Maher 2016). In addition to limiting fish migration, the creation of slow-moving pools between each dam greatly reduces the amount of quality spawning habitat available for many riverine species, including Grass Carp. During spawning periods in the spring and summer when discharge is high, Grass Carp migrate upstream and often congregate in large numbers within turbulent, higher velocity river sections to spawn (Kolar et al. 2007). In areas such as the dammed UMR, these higher velocity river sections are more common immediately downstream of dams, particularly LD19 where spawning of Grass Carp and other invasive carp has been documented to occur (Camacho 2016). Thus, LD19 is considered a key choke point (i.e., limited movement upstream and quality spawning habitat downstream where fish congregate) for invasive carp population management in the main stem UMR (e.g., Larson et al. 2017; Whitledge et al. 2019).

As a result of limited movement through LD19 into upstream UMR reaches, Grass Carp populations are hypothesized to differ upstream and downstream of LD19. However, no studies have quantified these differences beyond interpreting commercial fisheries catch data (e.g., Maher 2016) to the best of our knowledge. Therefore, the primary objective of this study was to assess Grass Carp population characteristics (relative abundance, size structure, condition, growth, and recruitment variability) between populations upstream and downstream of LD19. Understanding Grass Carp population characteristics upstream and downstream of this key choke point is a necessary prerequisite for the development of effective control and preventative measures (e.g., MICRA 2017) and should provide the basis for a better understanding of an elusive nonnative species.

Study Area

We assessed Grass Carp populations in the four southernmost major tributaries of the UMR in Iowa located both upstream and downstream of LD19: Des Moines, Skunk, Iowa, and Cedar rivers (Figure 1). We limited our study to these rivers because they represent most of the farthest upstream UMR pools (Pools 17–20) where natural reproduction by Grass Carp has been documented (Larson et al. 2017), limiting differences attributable to recruitment. The Des Moines and Mississippi river confluence is approximately 6 river kilometers (km) downstream from LD19 (Pool 20). The Skunk and Iowa river confluences with the Mississippi River are approximately 32 and 69 river km upstream (Pools 18–19), respectively. These rivers have an array of wing dikes and levees and have been channelized to manipulate river discharge. Both the Ottumwa Dam (near Ottumwa, Iowa) on the Des Moines River and Oakland Mills Dam (near Oakland Mills, Iowa) on the Skunk River operate to mitigate flooding but fish are believed to pass upstream under most river conditions except during low-discharge periods (J. Euchner, Iowa Department of Natural Resources, personal communication). Starting in either southern Minnesota or central Iowa, these tributaries drain a substantial portion of north-central to southeastern Iowa. Catchment areas range between 11,222 km2 (Skunk River) and 37,296 km2 (Des Moines River; USGS 2019). In contrast, the Iowa River catchment is composed mainly (62%) of the Cedar River catchment (20,279 km2; USGS 2019).

Methods

Field and laboratory methods

We sampled Grass Carp monthly from April to October 2014 and 2015 at four sites downstream of LD19 in the Des Moines River and five sites upstream in the Skunk, Iowa, and Cedar rivers (Figure 1). We selected sampling sites within rivers based on the location of river access points, logistical constraints, and agency interests. Grass Carp have been effectively captured using both boat electrofishing (Cumming et al. 1975; Wanner and Klumb 2009a; Clemens et al. 2016) and stationary trammel nets (George and Chapman 2015). Consequently, we used both boat electrofishing surveys and trammel net sets concurrently. At each of the nine sites, we surveyed three fixed sampling locations separated by approximately 1.5 river km once per month. Sampling occurred in areas <4 m deep (DeGrandchamp et al. 2008) within areas of low velocity (<1.0 m/s; e.g., eddies, dike pools, inside river bends, etc.). Within each fixed sampling location, we first deployed a stationary, multifilament trammel net (2.4-m-deep inner wall, 1.8-m-deep outer wall, 38.1-m-long, 10-cm-bar inner mesh) by anchoring one end of the net on shore and stretching the remaining net toward deeper water or an opposite shore, restricting fish movement out of low-velocity areas. We only deployed trammel nets when current velocity was low enough to avoid hazardous conditions and decrease the likelihood of net entanglement. Next, we conducted a 15-min daytime boat electrofishing (DC; 4-13 A, 100–500 V, 25% duty cycle, 25 Hz frequency, 60 pulses/s with two netters) survey using a “standardizing by power” approach (Miranda 2009) parallel to the shoreline. We then collected the trammel net immediately after each electrofishing transect (net set duration ranged between 20 and 30 min). We measured thalweg water temperature (Yellow Springs Instruments 550A; Celsius) and conductivity (EC400 ExStik 2 Conductivity Meter; μs/cm) during each monthly sampling event at each site and obtained mean daily discharge values (cubic meters per second [m3/s]) on the day of sampling from U.S. Geological Survey gauging stations upstream from each sampling location (Table S1, Supplemental Material).

We weighed (nearest 1 g) and measured for total length (TL; nearest 1 mm) all Grass Carp captured during both years of the study. During 2015, we removed the first pectoral fin ray on each side for age and growth analysis. We used pectoral fin rays because the assigned age agreement ±1 y is relatively high between pectoral fin ray and otolith age estimates in other closely related carps (e.g., Silver Carp; Seibert and Phelps 2013) and have been previously used to age Grass Carp (Wieringa et al. 2017). Furthermore, otoliths collected from a subset of Grass Carp throughout this study revealed that annuli were not easily discernable (CJ Sullivan, unpublished data). We air-dried pectoral fin rays (fin ray hereafter) at room temperature for ≥4 wk following collection before we processed them. We cut a 1-mm-thick cross-section at the base of the fin ray using a Buehler Isomet low-speed saw (Isomet Corporation, Springfield, VA). We mounted each cross-section to a glass microscope slide using Crystalbond 509 (Electron Microscopy Sciences). We used wetted, 2,000-grit sandpaper to polish the surface of the fin ray cross-section to improve clarity. We wetted cross-sections with immersion oil to further improve clarity, and viewed annuli under a dissecting microscope with transmitted light. Each fin ray cross-section was independently aged by two experienced readers with no knowledge of fish length, estimated age of other structure, or location. If the readers disagreed, then they jointly decided a common age to ensure confidence in annulus identification.

Data analysis

We grouped Grass Carp population characteristics as either downstream (Des Moines River sites) or upstream (Skunk, Iowa, and Cedar river sites) of LD19 for analyses. We limited comparisons of Grass Carp population characteristics on account of small sample sizes; therefore, we combined data collected from both 2014 and 2015 to evaluate location (upstream or downstream of LD19) differences. Sampling gears represented a combination of a passive and an active technique and thus catch per unit effort (CPUE) is not directly comparable; therefore, we expressed Grass Carp relative abundance as mean CPUE separately for boat electrofishing and trammel net sets and analyzed separately. We calculated boat electrofishing CPUE as average number of Grass Carp captured per hour (fish/hour), whereas we calculated trammel net CPUE as average number of Grass Carp captured per net lift (fish/net lift). Preliminary analysis indicated that the CPUE data were zero-inflated (i.e., many surveys where Grass Carp were not captured), overdispersed, and the number of fish captured per survey was generally low. In these situations, the mean and variance are often correlated and modeling using a negative-binomial distribution can rectify these issues (Gardner et al. 1995). Thus, we used a negative-binomial generalized linear model to assess differences in monthly mean Grass Carp CPUE upstream and downstream of LD19. We modeled Grass Carp CPUE data as a count (i.e., number of Grass Carp captured) for a given amount of effort  
formula
where i represents the month, j represents the location (upstream or downstream of LD19), and k represents the individual trammel net or electrofishing survey. The λij term is the Grass Carp CPUE for month i and location j, while ϕijk is the effort (offset) for month i, location j, and trammel net or electrofishing survey k. We used pairwise comparisons of monthly Grass Carp CPUE ratios (proportional changes between months) to compare CPUE among months upstream and downstream of LD19. We used a post hoc Tukey's honestly significant difference test using Bonferroni corrections to determine which location (upstream or downstream of LD19) and months differed at a significance level α of 0.05.

We constructed total length (TL) frequency histograms and used a nonparametric Kolmogorov–Smirnov (K–S) two-sample test to test the null hypothesis that cumulative frequency of Grass Carp TL distributions did not differ between gears. Using Grass Carp captured downstream of LD19, TL frequency distributions were not significantly different between gears (D = 0.16, P = 0.23), suggesting that both gears captured similar sizes of Grass Carp. Therefore, we used a K–S two-sample test to test the null hypothesis that cumulative frequency of Grass Carp TL distributions did not differ upstream and downstream of LD19 using data combined across gears and years. We set statistical significance at α of 0.05, which we Bonferroni-corrected to maintain family-wise error rates of 0.05. We evaluated Grass Carp condition using a length–weight relationship (Ricker 1975) and used an analysis of covariance (ANCOVA) of weight to assess potential differences in condition between populations located both upstream and downstream of LD19 using log10length as a covariate.

Using age-frequency distributions, we evaluated location-specific (upstream or downstream of LD19) interannual recruitment variability using the recruitment variability index (RVI; Guy and Willis 1995) calculated as  
formula
where SN is the sum of the cumulative relative frequencies across year-classes in the sample, NM is the number of missing year-classes from the sample (year-classes beyond the oldest year-class in the sample are excluded), NP is the total number of year-classes present in the sample, and only included age-5 and older Grass Carp. We estimated instantaneous total mortality rates (Z) for Grass Carp populations upstream and downstream of LD19 using a weighted, age-based catch-curve analysis with age as the independent variable and ln(frequency of catch) as the dependent variable, where each data point was weighted by the ln(total catch) of that age class. The descending limb of age-frequency histograms suggests a full recruitment to the sampling gears at ages 5 to 6 for Grass Carp. Then, we estimated total annual mortality rates (A),  
formula
where Z is the instantaneous total mortality rate. Estimates of A were considered significantly different among sites if 95% confidence intervals did not overlap. Differences in mortality rates could be attributed to both natural and fishing-induced mortality because variable amounts of commercial harvest occurs across sites (e.g., Maher 2016) and rates herein likely reflect both sources of mortality.
We used a weighted, two-way ANOVA to test for site and age differences in Grass Carp length. If we detected significant differences, we used pairwise t-tests with Bonferroni corrections to determine which groups differed. In addition, we estimated growth trajectories of Grass Carp upstream and downstream of LD19 using von Bertalanffy models fit to individual length at estimated age data using nonlinear least-squares regression (von Bertalanffy 1938):  
formula
where Lt is the TL at time t, L is the average asymptotic maximum TL, K is the Brody–Bertalanffy growth coefficient, t0 is the x-intercept, and εi is an additive error term. Preliminary growth analysis yield nonsensical results for t0 due to low numbers of Grass Carp <3 y of age captured. The t0 parameter was fixed at 0 when deriving K and L from the von Bertalanffy model, which can help reduce biases when using sampling gears that are selective for large individuals (Gwinn et al. 2010). We conducted all statistical analyses in Program R Version 3.2.0 (R Core Team 2016) or SAS Version 9.2 (SAS Institute, Cary, NC) with a significance level of α = 0.05.

Results

Data collected for this project are available in electronic format (Data S1, Supplemental Material). We conducted 95.7 electrofishing hours and 198 trammel net sets throughout rivers in southeastern Iowa from April to October 2014 and 2015. We captured 179 Grass Carp at the Des Moines River (downstream of LD19) and captured 29 Grass Carp upstream of LD19 at the Skunk, Iowa, and Cedar rivers (Tables 1, 2). Boat electrofishing captured 144 Grass Carp (Table 1) while trammel nets captured 64 Grass Carp (Table 2). Electrofishing CPUE ranged from 0 to 22.7 fish/h (mean ± SE = 1.49 ± 0.3) while trammel net CPUE ranged from 0 to 8.0 fish/net lift (0.46 ± 0.15). Mean electrofishing CPUE varied between locations upstream and downstream of LD19 (P < 0.001); electrofishing CPUE was higher downstream of LD19 (2.42 fish/h ± 0.30) than upstream (0.57 fish/h ± 0.07). Downstream of LD19, mean electrofishing CPUE varied among months (P < 0.001), where the pairwise comparisons of monthly CPUE ratios revealed that catch rates were highest during May and October and lowest in July (pairwise comparisons P < 0.001; Figure 2). Grass Carp mean electrofishing CPUE did not vary among months upstream of LD19 (all pairwise comparisons P > 0.05). We did not capture Grass Carp upstream of LD19 using trammel nets, precluding comparisons upstream and downstream of LD19. Mean trammel net CPUE was similar among months (P = 0.06; all pairwise comparisons P> 0.14), with consistently lower capture rates from May through September (Figure 2).

Using data combined across gears, Grass Carp TL frequency distributions differed between populations located at sites upstream and downstream of LD19 (D = 0.56, P < 0.001; Figure 3). Grass Carp TL ranged from 421 to 964 mm (787 mm ± 29) and 0.9 to 11.6 kg (6.4 kg ± 0.5) upstream of LD19 and 574 to 996 mm (725 mm ± 6) and 2.0 to 9.4 kg (4.2 kg ± 0.1) downstream (Tables 1, 2). Grass Carp <900 mm in length comprised 76% of total catches upstream of LD19 and 95% of total catches downstream; however, Grass Carp <500 mm were only captured upstream of LD19 (n = 4 fish). Grass Carp weight-at-length relationships were significantly different between populations located upstream and downstream of LD19 (F2, 197 = 21.3, P < 0.001; Figure 4) with fish captured downstream weighing less at a given length compared with those captured upstream.

Age-frequency distributions and recruitment variability index scores suggest that Grass Carp recruitment was moderately consistent downstream of LD19 (RVI = 0.53; Figure 5) despite missing age classes. Upstream of LD19, limited samples (n = 27) precluded the comparison of age-frequency distributions between sites; however, the recruitment variability index indicated relatively consistent recruitment (RVI = 0.41; Figure 5). The Grass Carp annual mortality rate (A) for populations captured downstream was 0.40 (95% confidence interval: 0.31, 0.49) whereas the mortality rate upstream of LD19, based on a small sample of 27 fish, was 0.10 (95% confidence interval: 0.04, 0.16). Lastly, Grass Carp length-at-age varied by location and age (F1, 107 = 4.98, P = 0.027); fish were consistently longer at a given age upstream compared with downstream of LD19, with the exception of age 8 and 9 Grass Carp (Figure 6). The von Bertalanffy growth-parameter estimates for Grass Carp captured downstream of LD19 were L = 810 mm and K = 0.42 and L = 916 mm and K = 0.38 for Grass Carp captured upstream.

Discussion

Difficulties related to capture and low Grass Carp densities offer limited opportunities to formally evaluate populations unless substantial effort is allocated toward the collection of Grass Carp (Sullivan et al. 2019). Assessments of adult Grass Carp populations have often focused on stocked or introduced populations in the southeastern or eastern United States (Shireman et al. 1980; Morrow et al. 1997; Stich et al. 2013) with few assessments conducted within the Mississippi River watershed (but see Wanner and Klumb 2009a). Lock and Dam 19, a key focal point for naturalized Grass Carp management because limited upstream migration occurs, creates a semipermanent barrier between Grass Carp populations that exhibited different population characteristics throughout our study. Grass Carp downstream of LD19 tended to be smaller, younger, of lower body condition, had a higher mortality rate, and were slower growing compared with fish collected upstream. For an open system like the UMR, it may be possible to reduce propagule pressure by targeting key management areas for control like LD19. Commercial fishing efforts already occur upstream and downstream of LD19 but focus on decreasing abundance and understanding trends in Silver and Bighead Carp demographics; these efforts also capture and remove Grass Carp but do not focus on determining population demographics and monitoring changes through time (MICRA 2017). For removal efforts to work for Grass Carp, it is important to understand population status upstream and downstream of these areas to determine appropriate management strategies.

Little information is available documenting Grass Carp capture rates throughout their native or introduced range. Methods such as bow fishing (Morrow et al. 1997; Stich et al. 2013) and commercial harvests (Pflieger 1978) have been used to obtain individuals but cannot be used to infer patterns in relative abundance estimates. Within the Missouri River basin, standardized sampling surveys (e.g., trammel, gill, mini-fyke, and hoop nets) have been used to estimate Grass Carp relative abundance. Trammel net CPUE in that system was relatively low (mean CPUE <0.05 fish/100 m net lift; Wanner and Klumb 2009a) compared with catch rates in the Des Moines River downstream of LD19, but was higher than catch rates in the Skunk, Iowa, and Cedar rivers upstream of LD19 (zero captures). However, those sampling methodologies were designed for the capture of Pallid Sturgeon Scaphirhynchus albus (Wanner and Klumb 2009a), potentially excluding habitats commonly used by Grass Carp (e.g., backwater; Shireman and Smith 1983). In this study, Grass Carp electrofishing CPUE was higher (up to 22.7 fish/h) than values reported throughout literature and at least one Grass Carp was captured in 17.8% of electrofishing surveys (382 total surveys), whereas trammel net CPUE was lower (up to 8.0 fish/net lift) and Grass Carp were captured in 13.1% of trammel net sets (198 total net sets). Differences in capture rates may be due in part to the active sampling in shallow-water habitats that would not be effectively sampled using trammel nets or the short duration of net sets (20–30 min/set used herein versus overnight sets) that may have limited Grass Carp captures. Additionally, Grass Carp electrofishing CPUE varied intra-annually where catches were highest during May and October compared with other months, especially July. Grass Carp are a highly migratory species (Gorbach and Krykhtin 1988; Bain et al. 1990) and spring spawning migrations or autumn movement patterns could result in seasonal changes in local abundances that would result in higher catch rates (e.g., congregating below a dam).

Our results indicate that Grass Carp recruitment was moderately consistent (indicated by RVI values) both upstream and downstream of LD19. Grass Carp are broadcast spawners that release semibuoyant eggs into turbulent flowing water (Verigin et al. 1978; Kolar et al. 2007). Eggs must remain suspended in the water column for 24 to 48 h for successful hatching, which equates to a drift distance of roughly 15 to 80 river km (Krykhtin and Gorbach 1981; Garcia et al. 2015). Previous studies have found that invasive carp reproduction is highest during high discharge events (DeGrandchamp et al. 2007; Lohmeyer and Garvey 2009; Sullivan et al. 2018) and have speculated that the impounded UMR may only offer high-quality spawning conditions during those periods. However, large tributaries of the UMR, such as the Des Moines, Skunk, Iowa, and Cedar rivers, offer long stretches of free-flowing river where Grass Carp reproduction has occurred (Camacho 2016), likely leading to the more consistent recruitment patterns in our study. In addition, the migratory abilities of Grass Carp (Gorbach and Krykhtin 1988) indicate that migrants from more consistently recruiting populations downstream of LD19 (e.g., Larson et al. 2017) could supply a few recruits periodically, similar to other invasive carp (e.g., Whitledge et al. 2019). Placement of deterrents within the navigation locks at LD19 are being considered to reduce migration of invasive carps and propagule pressures upstream (e.g., Donaldson et al. 2016). We did not determine the natal origin of Grass Carp captured upstream of LD19; however, further research could be conducted to determine if adult Grass Carp upstream of LD19 are born upstream of LD19, migrated through LD19, or are of hatchery origin (i.e., escapement) to help inform management decisions.

Grass Carp population characteristics described herein varied from those reported for both stocked populations in lakes and naturalized populations in other Mississippi River tributaries. Grass Carp captured both downstream and upstream of LD19 were smaller compared with populations captured using similar gears from the Missouri River (mean TL ± SE = 803 mm ± 6.8; Wanner and Klumb 2009a) and age structure was younger and represented a more restricted age structure than populations in a Virginia lake (Stich et al. 2013). Additionally, the mean length-at-age and von Bertalanffy parameter estimates for Grass Carp captured both upstream and downstream of LD19 suggests that Grass Carp are smaller, reach the asymptotic length quicker, and reach a smaller maximum size compared with Grass Carp captured in South Carolina, USA (Morrow et al. 1997), Florida, USA (Shireman et al. 1980), and Virginia, USA (Stich et al. 2013). Grass Carp collected downstream of LD19 were of lesser body condition than those collected from the Missouri River (Wanner and Klumb 2009b) and a Virginia lake (Stich et al. 2013), but Grass Carp upstream of LD19 were of greater body condition. Grass Carp populations upstream and downstream of LD19 are generally smaller, composed younger fish, and in lower condition than Grass Carp captured from other lotic and lentic systems, potentially in response to highly variable river environments (e.g., Gutreuter et al. 1999) and the greater energetic demands of lotic systems (e.g., Glebe and Leggett 1981) or due to commercial harvest of adult Grass Carp through time (Klein et al. 2018).

Grass Carp <420 mm TL were not captured upstream or downstream of LD19. Currently, Grass Carp reproduction is known to occur as far upstream as Pool 12 (Larson et al. 2017) and did occur within the tributaries sampled herein during 2014 and 2015 (Camacho 2016). Thus, smaller sized Grass Carp may be present throughout our sites but went undetected, suggesting that boat electrofishing and trammel nets select for larger sized Grass Carp. Wanner and Klumb (2009a) employed boat electrofishing, trammel nets, hoop nets, gill nets, and mini-fyke nets throughout the Missouri River to collect invasive carp, and mini-fyke nets were the only gear to capture multiple Grass Carp ≤300 mm TL, although captures were relatively low (<20 captures over 5 y). Most commonly used sampling gears (e.g., electrofishing and trammel nets) require depths of approximately 1.0–2.0 m and shorelines that are generally free of obstructions. Smaller sized Grass Carp tend to use areas with a higher density of submerged vegetation (Bain et al. 1990) in currents <0.05 m/s (Raibley et al. 1995). In the impounded UMR, fine sediments are trapped in shallow, low-flow areas, and water turbidity is high, resulting in absent or sparse patches of aquatic vegetation that may be only available seasonally in shallow habitats (Moore et al. 2010). Smaller sized Grass Carp may be locally abundant in these vegetation patches that are generally difficult to sample with most sampling gears, but this remains unknown. Since 2013, Western Illinois University has routinely conducted surveys targeting juvenile invasive carps in these habitat types throughout the UMR, but with few captures reported (JT Lamer, Illinois Natural History Survey, personal communication). Identifying effective gears and capturing smaller sized Grass Carp will allow managers to describe the dynamics of early life stages across the UMR, knowledge of which is currently lacking.

The Grass Carp is a widely dispersed nonnative species throughout the UMR and our study provides important insight for management. Lock and Dam 19 represents a focal point for management where contract commercial fishing efforts immediately downstream and upstream aim to reduce population density and potential upstream migrants to decrease propagule pressures in upstream reaches of the UMR (MICRA 2017). Generally, commercial fishers employ trammel nets or gill nets to capture invasive carp (Maher 2016) that select for larger sized Grass Carp (e.g., Wanner and Klumb 2009a). Grass Carp populations upstream and downstream of LD19 were generally smaller and slower growing compared with many other populations; therefore, shifting commercial harvest efforts to use gears that capture smaller sized Grass Carp may be required, particularly downstream of LD19. Targeting smaller sized invasive carp for removal has been suggested as an effective management strategy in other UMR rivers (Tsehaye et al. 2013) and could lead to a more successful reduction in Grass Carp density through time. Further, the lack of historical demographic information about Grass Carp populations limits our capacity to determine how populations have changed since their first detection near LD19 in the 1970s (NAS 2019). However, our study represents the first evaluation of Grass Carp population characteristics within the UMR, establishing a baseline for future evaluations. In the future, managers will be able to track population characteristics through time and can set benchmarks for the success of various management efforts (e.g., harvest, barriers, etc.). We suggest that improved monitoring (e.g., expansion of sampling locations, use of gears that are effective in shallow vegetated habitats) across several generations, robust population abundance estimates, and a better understanding of movement patterns (e.g., frequency of Grass Carp movement through LD19) enable refining of the management goals for Grass Carp in the UMR and at LD19.

Supplemental Material

Please note: The Journal of Fish and Wildlife Management is not responsible for the content or functionality of any supplemental material. Queries should be directed to the corresponding author for the article.

Data S1. Grass Carp Ctenopharyngodon idella captures (e.g., site and date) and individual data (e.g., total length, weight, sex, age) upstream and downstream of Lock and Dam 19.

Found at DOI: https://doi.org/10.3996/062019-JFWM-046.S1 (51 KB XLSX).

Table S1. Mean, maximum (Max), and minimum (Min) daily river discharge (m3/s; estimates obtained from U.S. Geological Survey [USGS] gauging stations), water temperature (°C), and conductivity (μS/cm) measured during sampling occasions from April to October 2014 and 2015 at the nine sampling sites upstream and downstream of Lock and Dam 19 used to assess differences in Grass Carp Ctenopharyngodon idella population characteristics.

Found at DOI: https://doi.org/10.3996/062019-JFWM-046.S2 (24 KB DOCX).

Reference S1. Camacho CA, Sullivan CJ, Weber MJ, Pierce CL. 2016. Distribution and population dynamics of Asian Carp in Iowa rivers. Des Moines: Iowa Department of Natural Resources. Annual Progress Report.

Found at DOI: https://doi.org/10.3996/062019-JFWM-046.S3 (5.22 MB PDF).

Acknowledgments

We thank the numerous undergraduate research technicians from Iowa State University that helped complete field work, and J. Euchner and K. Bogenschutz (Iowa Department of Natural Resources) for providing helpful insights about Grass Carp. We thank D. Stich, the journal's anonymous reviewers, and the Associate Editor for their time and comments on earlier drafts of this manuscript. This study was funded by the Iowa Department of Natural Resources through contract 14CRDFBGSCHO-0001. This study was performed under the auspices of Iowa State University Institutional Animal Care and Use Committee (IACUC) protocol permit 7-13-7599-I, and animals were collected under state permit SC1037.

Any use of trade, product, website, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

References

References
Bain
MB.
1993
.
Assessing impacts of introduced aquatic species: Grass Carp in large systems
.
Environmental Management
17
:
211
224
.
Bain
MB,
Webb
DH,
Tangedal
MD,
Mangum
LN.
1990
.
Movements and habitat use by Grass Carp in a large mainstream reservoir
.
Transactions of the American Fisheries Society
119
:
553
561
.
Bettoli
PW,
Maceina
MJ,
Noble
RL,
Betsill
RK.
1993
.
Response of a reservoir fish community to aquatic vegetation removal
.
North American Journal of Fisheries Management
13
:
110
124
.
Bettoli
PW,
Neill
WH,
Kelsch
WL.
1985
.
Temperature preferences and heat resistance of Grass Carp, Ctenopharyngodon idella (Valenciennes), Bighead Carp, Hypophthalmichthys nobilis (Gray), and their F1 hybrid
.
Journal of Fish Biology
27
:
239
247
.
Camacho
CA.
2016
.
Asian Carp reproductive ecology along the Upper Mississippi River invasion front. Master's thesis
.
Ames
:
Iowa State University
.
Camacho
CA,
Sullivan
CJ,
Weber
MJ,
Pierce
CL.
2016
.
Distribution and population dynamics of Asian Carp in Iowa rivers
.
Des Moines
:
Iowa
Department of Natural Resources. Annual Progress Report (see Supplemental Material, Reference S1).
Chilton
II EW,
Muoneke
MI.
1992
.
Biology and management of Grass Carp (Ctenopharyngodon idella, Cyprinidae) for vegetation control: a North American perspective
.
Reviews in Fish Biology and Fisheries
2
:
283
320
.
Clemens
BJ,
Spangler
JJ,
Robertson
PL,
Galovich
GM,
Banner
CR,
Gunckel
SL,
Ketchum
LL,
Bowden
RK,
Spagnoli
ST,
Kent
ML,
Kirk
JP.
2016
.
Persistence of triploid Grass Carp in Devils Lake, Oregon
.
Journal of Fish and Wildlife Management
7
:
153
161
.
Colle
DE,
Shireman
JV,
Rottman
RW.
1978
.
Food selection by Grass Carp fingerlings in a vegetated pond
.
Transactions of the American Fisheries Society
107
:
149
152
.
Courtenay
WR,
Hensley
DA,
Taylor
JN,
McCann
JA.
1984
.
Distribution of exotic fishes in the continental United States
.
Pages
41
77
in
Courtenay
Jr,
WR
Stauffer
JR
Jr,
editors.
Distribution, biology, and management of exotic fishes
.
Baltimore, Maryland
:
John Hopkins University Press
.
Cudmore
B,
Mandrak
NE.
2004
.
Biological synopsis of Grass Carp (Ctenopharyngodon idella). Canadian MS Report Fisheries and Aquatic Sciences No. 2705
.
Burlington, Ontario
:
Fisheries and Oceans Canada Great Lakes Laboratory for Fisheries and Aquatic Sciences
.
Cumming
KB,
Burress
RM,
Gilderhus
PA.
1975
.
Controlling Grass Carp (Ctenopharyngodon idella) with antimycin, rotenone, and thanite and by electrofishing
.
Progressive Fish Culturist
3737
:
81
84
.
DeGrandchamp
KL,
Garvey
JE,
Colombo
RE.
2008
.
Movement and habitat selection by invasive Asian Carps in a large river
.
Transactions of the American Fisheries Society
137
:
45
56
.
DeGrandchamp
KL,
Garvey
JE,
Csoboth
LA.
2007
.
Linking adult reproduction and larval density of invasive Carp in a large river
.
Transactions of the American Fisheries Society
136
:
1327
1334
.
Donaldson
MR,
Amberg
J,
Adhikari
S,
Cupp
A,
Jensen
N,
Romine
J,
Wright
A,
Gaikowski
M,
Suski
CD.
2016
.
Carbon dioxide as a tool to deter the movement of invasive Bigheaded Carps
.
Transactions of the American Fisheries Society
145
:
657
670
.
Garcia
T,
Murphy
EA,
Jackson
PR,
Garvia
MH.
2015
.
Application of the FluEgg model to predict transport of Asian Carp eggs in the Saint Joseph River (Great Lakes tributary)
.
Journal of Great Lakes Research
41
:
374
386
.
Gardner
W,
Mulvey
EP,
Shaw
EC.
1995
.
Regression analysis of counts and rates: Poisson, overdispersed Poisson, and negative binomial models
.
Psychological Bulletin
118
:
392
404
.
George
AE,
Chapman
DC.
2015
.
Embryonic and larval development and early behavior in Grass Carp, Ctenopharyngodon idella: implications for recruitment in rivers
.
PLoS One
10
:
e0119023
.
Glebe
BD,
Legget
WC.
1981
.
Latitudinal differences in energy allocation and use during the freshwater migrations of American shad (Alosa sapidissima) and their life history consequences
.
Canadian Journal of Fisheries and Aquatic Sciences
38
:
806
820
.
Gorbach
EI,
Krykhtin
ML.
1988
.
Migration of the White Amur, Ctenopharyngodon idella, and Silver Carp, Hypophthalmichthys molitrix, in the Amur River Basin
.
Journal of Ichthyology
28
:
47
53
.
Gutreuter
S,
Bartels
AD,
Irons
K,
Sandheinrich
MB.
1999
.
Evaluation of the flood-pulse concept based on statistical models of growth of selected fishes of the Upper Mississippi River system
.
Canadian Journal of Fisheries and Aquatic Sciences
56
:
2282
2291
.
Guy
CS,
Willis
DW.
1995
.
Population characteristics of Black Crappie in South Dakota waters: a case for ecosystem-specific management
.
North American Journal of Fisheries Management
15
:
754
765
.
Gwinn
DC,
Allen
MS,
Rogers
MW.
2010
.
Evaluation of procedures to reduce bias in fish growth parameter estimates resulting from size-selective sampling
.
Fisheries Research
105
:
75
79
.
Hubert
W.
1994
.
Exotic fish
.
Pages
158
174
in
Parrish
TL,
Anderson
SH,
editors.
Exotic species manual
.
Laramie
:
Wyoming Game and Fish Department
.
Irons
KS,
Sass
GG,
McClelland
MA,
Stafford
JD.
2007
.
Reduced condition factor of two native fish species coincident with invasion of nonnative Asian Carps in the Illinois River, U.S.A. Is this evidence for competition and reduced fitness?
Journal of Fish Biology
71
:
258
273
.
Kelly
A,
Engle
C,
Armstrong
M,
Freeze
M,
Mitchell
AJ.
2011
.
History of introductions and governmental involvement in promoting the use of Grass, Silver, and Bighead Carps
.
Pages
163
174
in
Chapman
DC,
Hoff
MH,
editors.
Invasive Asian carps in North America
.
Bethesda, Maryland
:
American Fisheries Society
.
Symposium 74.
Klein
ZB,
Quist
MC,
Miranda
LE,
Marron
MM,
Steuck
MJ,
Hansen
KA.
2018
.
Commercial fisheries of the Upper Mississippi River: a century of sustained harvest
.
Fisheries
43
:
563
574
.
Knights
BC,
Vallazza
JM,
Zigler
SJ,
Dewey
MR.
2002
.
Habitat and movement of Lake Sturgeon in the upper Mississippi River system, USA
.
Transactions of the American Fisheries Society
131
:
507
522
.
Kolar
CS,
Chapman
DC,
Courtenay
WR
Jr,
Housel
CM,
Williams
JD,
Jennings
DP.
2007
.
Bigheaded Carps: a biological synopsis and environmental risk assessment
.
Bethesda, Maryland
:
American Fisheries Society
.
Special Publication 33.
Krykhtin
M,
Gorbach
E.
1981
.
Reproductive ecology of the Grass Carp, Ctenopharyngodon idella, and the Silver Carp, Hypophthalmichthys molitrix, in the Amur Basin
.
Journal of Ichthyology
21
:
109
123
.
Larson
JH,
Knights
BC,
McCalla
S,
Monroe
E,
Tuttle-Lau
MT,
Chapman
DC,
George
AE,
Vallazza
J,
Amberg
J.
2017
.
Evidence of Asian Carp spawning upstream of a key choke point in the Mississippi River
.
North American Journal of Fisheries Management
37
:
903
919
.
Lohmeyer
AM,
Garvey
JE.
2009
.
Placing the North American invasion of Asian Carp in a spatially explicit context
.
Biological Invasions
11
:
905
916
.
LTRM
.
2019
.
Fisheries
.
Upper Mississippi River Restoration Program: Long Term Resource Monitoring. Available: https://umesc.usgs.gov/data_library/fisheries/fish_page.html (December 2019).
Maceina
MJ,
Cichra
MF,
Betsill
RK,
Bettoli
PW.
1992
.
Limnological changes in a large reservoir following vegetation removal by Grass Carp
.
Journal of Freshwater Ecology
7
:
81
95
.
Maher
R.
2016
.
2015 commercial catch report, exclusive of Lake Michigan
.
Springfield
:
Illinois Department of Natural Resources
.
Michewicz
JE,
Sutton
DL,
Blackburn
RD.
1972
.
The White Amur for aquatic weed control
.
Weed Science
20
:
106
110
.
MICRA
.
2017
.
Monitoring and response plan for Asian carp in the Mississippi River Basin
.
Mississippi Interstate Cooperative Resource Association
.
Miranda
LES.
2009
.
Standardizing electrofishing power for boat electrofishing
.
Pages
223
230
in
Bonar
S,
WA,
Hubert
Willis
D,
editors.
Standard methods for sampling North American freshwater fishes
.
Bethesda, Maryland
:
American Fisheries Society
.
Mitchell
AJ,
Kelly
AM.
2006
.
The public sector role in the establishment of Grass Carp in the United States
.
Fisheries
31
:
113
121
.
Moore
M,
Romano
SP,
Cook
T.
2010
.
Synthesis of Upper Mississippi River System submersed and emergent aquatic vegetation: past, present, and future
.
Hydrobiologia
640
:
103
114
.
Morrow
JV
Jr,
Kirk
JP,
Killgore
KJ.
1997
.
Collection, age, growth, and population attributes of triploid Grass Carp stocked into the Santee–Cooper Reservoirs, South Carolina
.
North American Journal of Fisheries Management
17
:
38
43
.
NAS
.
2019
.
Ctenopharyngodon idella (Grass Carp). Nonindigenous aquatic species
.
U.S. Geological Survey. Available: https://nas.er.usgs.gov/queries/FactSheet.aspx?SpeciesID=514 (December 2019).
Opuszynski
K.
1967
.
Comparison of temperature and oxygen tolerance in Grass Carp (Ctenopharyngodon idella Val.), Silver Carp (Hypophthalmichthys molitrix Val.), and Mirror Carp (Cyprinus Carpio Val.)
.
Ekologia Polska
15
:
385
400
.
Opuszynski
K.
1968
.
Carp polyculture with plant-feeding fish: Grass Carp (Ctenopharyngodon idella Val.) and Silver Carp (Hypophthalmichthys molitrix Val.)
.
Bulletin De L Academie Polonaise Des Sciences-Serie Des Sciences Biologiques
16
:
677
681
.
Opuszynski
K.
1972
.
Use of phytophagous fish to control aquatic plants
.
Aquaculture
1
:
61
74
.
Pflieger
WL.
1978
.
Distribution and status of the Grass Carp (Ctenopharyngodon idella) in Missouri streams
.
Transactions of the American Fisheries Society
107
:
113
118
.
Raibley
PT,
Blodgett
D,
Sparks
RE.
1995
.
Evidence of Grass Carp (Ctenopharyngodon idella) reproduction in the Illinois and Upper Mississippi Rivers
.
Journal of Freshwater Ecology
10
:
65
74
.
R Core Team
.
2016
.
R: a language and environment for statistical computing
.
Vienna
:
R Foundation for Statistical Computing
.
Ricker
WE.
1975
.
Computation and interpretation of biological statistics of fish populations
.
Ottawa, Ontario
:
Fisheries Research Board of Canada Bulletin 191
.
Schramm
HL
Jr,
Brice
MW.
2000
.
Use of triploid Grass Carp to reduce aquatic macrophyte abundance in recreational fishing ponds
.
Proceedings of the Annual Conference Southeastern Association of Fish and Wildlife Agencies
52
:
93
103
.
Seibert
JR,
Phelps
QE.
2013
.
Evaluation of aging structures for Silver Carp from Midwestern U.S. rivers
.
North American Journal of Fisheries Management
33
:
839
844
.
Shireman
JV,
Colle
DE,
Maceina
MJ.
1980
.
Grass Carp growth rates in Lake Wales, Florida
.
Aquaculture
19
:
379
382
.
Shireman
JV,
Smith
CR.
1983
.
Synopsis of biological data on the Grass Carp, Ctenopharyngodon idella (Cuvier and Valenciennes, 1844)
.
Food and Aquaculture Organization Synopsis. No. 135. Food and Agriculture Organization of the United Nations.
Available: http://www.fao.org/3/a-ap938e.pdf (January 2020).
Stich
DS,
Dicenzo
V,
Frimpong
EA,
Jiao
Y,
Murphy
BR.
2013
.
Growth and population size of Grass Carp incrementally stocked for hydrilla control
.
North American Journal of Fisheries Management
33
:
14
25
.
Sullivan
CJ,
Camacho
CA,
Weber
MJ,
Pierce
CL.
2017
.
Intra-annual variability of Silver Carp populations in the Des Moines River, USA
.
North American Journal of Fisheries Management
37
:
836
849
.
Sullivan
CJ,
Weber
MJ,
Pierce
CL,
Camacho
CA.
2019
.
Influence of river discharge on Grass Carp occupancy dynamics in south-eastern Iowa rivers
.
River Research and Applications
35
:
60
67
.
Sullivan
CJ,
Weber
MJ,
Pierce
CL,
Wahl
DH,
Phelps
QE,
Camacho
CA,
Colombo
RE.
2018
.
Factors regulating year-class strength of Silver Carp throughout the Mississippi River basin
.
Transactions of the American Fisheries Society
147
:
541
553
.
Trimm
DL,
Guillen
G,
Menn
CT,
Matlock
GC.
1989
.
The occurrence of Grass Carp in Texas waters
.
Texas Journal of Science
41
:
413
417
.
Tripp
SE,
Brooks
R,
Herzog
D,
Garvey
J.
2014
.
Patterns of fish passage in the upper Mississippi River
.
River Research and Applications
30
:
1056
1064
.
Tsehaye
I,
Catalano
M,
Sass
G,
Glover
D,
Roth
B.
2013
.
Prospects for fishery-induced collapse of invasive Asian carp in the Illinois River
.
Fisheries
38
:
445
454
.
USGS
.
2019
.
USGS water resources: of the Unites States: boundary descriptions and names of regions, subregions, accounting units and cataloging units
.
U.S. Geological Survey
.
Verigin
BV,
Makeyeva
AP,
Zaki Mokhamed
MI.
1978
.
Natural spawning of the Silver Carp (Hypophthalmichthys nobilis), the Bighead Carp (Aristichthys nobilis), and the Grass Carp (Ctenopharyngodon idella) in the Syr-Dar'ya River
.
Journal of Ichthyology
18
:
143
146
.
von Bertalanffy
L.
1938
.
A quantitative theory of organic growth
.
Human Biology
10
:
181
213
.
Wanner
GA,
Klumb
RA.
2009
a.
Asian Carp in the Missouri River: analysis from multiple Missouri River habitat and fisheries programs
.
National Invasive Species Council materials. Paper 10. Lincoln: University of Nebraska - Lincoln Digital Commons.
Wanner
GA,
Klumb
RA.
2009
b.
Length–weight relationships for three Asian Carp species in the Missouri River
.
Journal of Freshwater Ecology
24
:
489
495
.
Watkins
CE,
Shireman
JV,
Rottman
RW,
Colle
DE.
1981
.
Food habits of fingerling Grass Carp
.
Progressive Fish-Culturist
43
:
95
97
.
Whitledge
GW,
Knights
B,
Vallazza
J,
Larson
J,
Weber
MJ,
Lamer
JT,
Phelps
QE,
Norman
JD.
2019
.
Identification of Bighead Carp and Silver Carp early-life environments and inferring Lock and Dam 19 passage in the Upper Mississippi River: insights from otolith chemistry
.
Biological Invasions
21
:
1007
1020
.
Wieringa
JG,
Herbst
SJ,
Mahon
AR.
2018
.
The reproductive viability of Grass Carp (Ctenopharyngodon idella) in the western basin of Lake Erie
.
Journal of Great Lakes Research
43
:
405
409
.
Wilcox
D,
Stefanik
E,
Kelner
D,
Cornish
M,
Hodgins
I,
Zigler
S,
Johnson
B.
2004
.
Improving fish passage through navigation dams in the Upper Mississippi River System
.
St. Paul, MN: U.S. Army Corps of Engineers: Upper Mississippi River–Illinois Waterway Navigation Study, ENV Report 54.

Author notes

Citation: Sullivan CJ, Weber MJ, Pierce CL, Camacho CA. 2020. A comparison of Grass Carp population characteristics upstream and downstream of Lock and Dam 19 of the Upper Mississippi River. Journal of Fish and Wildlife Management 11(1):99–111; e1944-687X. https://doi.org/10.3996/062019-JFWM-046

Supplementary data