Introduction

The importance of wetlands to fish populations can vary seasonally, and knowing how and when wetland ecosystem services benefit fishes may help inform decisions on the management of wetland resources. Wetlands support a diversity and abundance of taxa that provide food and habitat resources for many species (Cucherousset et al. 2008; Dahl 2011; DeAngelis et al. 2005). Wetland use and access is closely tied to fishery health (Meynecke et al. 2008; Modde 1997; Seilheimer and Chow-Fraser 2006) and can lead to increased food availability, consumption, and growth in fishes (Madon et al. 2001; Woo et al. 2018). High diversity and abundance of prey and habitat resources are essential to support ecosystem resilience (Chapin et al. 2000; Walker 1995) and sustainable fisheries in the face of changing habitat or climate.

Fish benefit from wetlands when they have access to wetland resources, undisrupted by dams or other barriers. The impacts of reduced habitat connectivity on fisheries are well-documented and often result in reduced fish diversity (Bouvier et al. 2009), abundance (Meynecke et al. 2008), recruitment (US ACE 2020), and individual growth (Craig et al. 2014). For wetland complexes already impacted by reduced or lost connectivity, restoration and reconnection can help restore diversity and abundance of fishes (Vilizzi et al. 2013) and maintain resilience. Since wetlands tend to have gradual slopes and shallow depths, even small changes in water level can have a large impact on the surface area of wetlands available to fishes, making them disproportionately vulnerable to fluctuating water levels.

In the late 1990s and early 2000s, the Prairie Pothole Region of North America experienced above average annual precipitation that led to increased water levels region-wide, connecting lakes to previously isolated wetlands and giving fish access to high-quality habitat (South Dakota GFP 2019; Vanderhoof and Alexander 2016). Many highly productive sportfisheries emerged throughout the region in the newly created and expanded waters (Dembkowski et al. 2014). The increased accessibility of wetlands during these high water periods led to greater relative weights and faster growth of Walleye (Blackwell et al. 2019, 2020; Dembkowski et al. 2014; Neumann et al. 2012).

Although the importance of wetlands in the Prairie Pothole Region is well documented, fine-scale seasonal and habitat-specific differences that could impact fish populations are unknown. Connected wetlands are common in Prairie Pothole glacial lakes, but wetland drainage and/or filling have become common practices in the region owing to agricultural expansion (Dahl 2011; Wright and Wimberly 2013). Walleye are one of the most important sportfish in the Midwestern United States (McClanahan and Hansen 2005; South Dakota GFP 2019; US Department of the Interior et al. 2016), and knowing when wetlands are most important for Walleye can help managers ensure wetland access during critical periods to potentially improve Walleye growth, abundance, recruitment, and condition. Our objective was to evaluate temporal differences in prey availability and Walleye diets in a large, glacial lake (Lake Kampeska, South Dakota, USA) and a connected wetland. Specifically, we examined prey fish catch per unit effort, alpha and beta diversity, and Walleye relative weight, percent empty stomachs, diet weight, stomach fullness, diet energy, and diet taxa importance (percent by weight and energy). The economic and social importance of high-priority sportfish like Walleye can be used to enhance ecosystem resilience and countless other ecological services provided by wetlands.

Methods

Study Site

Lake Kampeska, located in Watertown, South Dakota, USA, is a 2,125 ha glacial lake connected to the Big Sioux River via a 25 ha, shallow, open-water wetland (Fig. 1). Hereafter, we will use lake to refer to the lake habitat, wetland to refer to the wetland habitat, and Lake Kampeska system to refer to both habitats combined. The lake has a relatively simple basin with a maximum depth of 5 m, minimal submerged vegetation, and largely silt substrates with some sand, gravel, and rock primarily along the shorelines. Although many wetlands were historically connected to the lake, shoreline development and alterations around the lake have left the wetland that we studied as the only substantial wetland that fish can access. The wetland is classified as palustrine wetland with a semi-permanently flooded water regime (Cowardin et al. 1979). The wetland consists mostly of fine textured substrates, shallow waters (< 1.5 m), seasonally variable water temperatures (Online Resource 2), thick submerged vegetation during the growing season, and is surrounded by cattails (Typha spp.), rushes (Juncus spp.), and other emergent wetland vegetation. The Lake Kampeska system is managed for Walleye and Smallmouth Bass (Micropterus dolomieu), but also features important fisheries for Yellow Perch (Perca flavescens), Northern Pike (Esox lucius), crappie (Pomoxis nigromaculatus and P. annularis), Bluegill (Lepomis macrochirus), and White Bass (Morone chrysops).

Fig. 1
figure 1

Map of Lake Kampeska, South Dakota (a) and zoomed in view of the wetland (b). Aquatic habitats designated as open water (white), wetland complex (hatched), or riverine (gray)

Full size image

The Lake Kampeska system has faced a range of challenges including eutrophication, sedimentation, shoreline development, variable water levels, and most recently the establishment of a Zebra Mussel (Dreissena polymorpha) population. A primary concern is high sediment load from the lake’s inlet–outlet to the Big Sioux River, which has contributed to undesirable water quality and declines in tourism, fish populations, and recreational opportunities (City of Watertown 2019). To mitigate sediment deposition, a single-channel weir (concrete wall across channel with V-shaped opening) was installed in November of 2001. As a result, access for fishes to the only remaining connected wetland was limited to a single access point approximately 1 m in width. Despite the narrow opening, a concurrent acoustic telemetry study indicated that fish were able to access both habitats throughout the study period (Cutler 2023).

Prey Fish

Prey fish, defined as fish < 150 mm in total length, were used to assess adult Walleye food availability in both the lake and wetland. Prey fish were captured using mini-fyke nets set overnight perpendicular to shore in water < 2 m. Mini-fyke nets (2 m long by 0.6 m tall, 3 mm bar mesh, 4.3 m long by 0.6 m tall lead, unrestricted throat) were made of two rectangular frames (1.3 m wide by 0.6 m tall) followed by two circular hoops (0.6 m diameter). Six nets were set in each habitat (i.e., lake and wetland), each season (i.e., spring [late April to early June], summer [mid-July to mid-August], and fall [late September to early November]). Sampling occurred in summer and fall of 2021 and spring, summer, and fall of 2022. Collected prey fish were weighed (0.1 g) and measured (total length; mm) for up to 30 individuals per species, season, and habitat, and all remaining individuals were counted. Prey fish biomass was calculated using these weights and extrapolated to non-weighed individuals using the mean weight for each species, season, and habitat combination. Prey fish were grouped into the following categories due to limited sample sizes and our inability to consistently distinguish them in Walleye diets: sunfish (i.e., Bluegill, Orangespotted Sunfish [Lepomis humilis], and Green Sunfish [L. cyanellus]), crappie (i.e., Black Crappie and White Crappie), shiners (i.e., Spottail Shiner [Notropis hudsonius], Emerald Shiner [Notropis atherinoides], and Golden Shiner [Notemigonus crysoleucas]), darters (i.e., Johnny Darter [Etheostoma nigrum], Iowa Darter [E. exile], and Logperch [Percina caprodes]), bullheads (i.e., Black Bullhead [Ameiurus melas] and Yellow Bullhead [A. natalis]), and large-bodied Cypriniformes (hereafter Cypriniformes; i.e., Common Carp [Cyprinus carpio] and Bigmouth Buffalo [Ictiobus cyprinellus]). Sunfish were separated into two size classes (small, < 60 mm; large, ≥ 60 mm) due to distinct differences in size and energy density.

Prey fish catch per unit effort (CPUE) was calculated for each net as biomass (g) per hour of set time. Prey fish CPUE was log(x + 0.01) transformed due to high variability (Aarestrup et al. 2015; Schrandt et al. 2021) and compared between habitats and seasons using linear models. All statistical analyses were conducted using R Statistical Software 4.2.0 (R Core Team 2022) and an alpha of 0.05. Seasons sampled in consecutive years were treated separately. Interactive, additive, and single factor (i.e., only season and only habitat) models were compared using likelihood ratio tests with the “lmtest” package (Zeileis and Hothorn 2002). Tukey’s post-hoc test was then run on covariates in the best model using package “multcomp” (Hothorn et al. 2008). Prey fish alpha diversity was estimated using taxa richness and Shannon-Weiner diversity for each habitat and season with the “vegan” package (Oksanen et al. 2022). To assess which seasons the lake and wetland prey communities were most similar (i.e., beta diversity), we used the Bray–Curtis community dissimilarity index with catch data (biomass per hour as an index of relative abundance; Oksanen et al. 2022; as in Bray and Curtis 1957, David et al. 2014, and Kornis et al. 2015). Species groupings mentioned earlier were not used when calculating any diversity metrics.

Walleye Diets

Up to 25 Walleye (> 275 mm) per season and habitat were captured using pulsed DC boat electrofishing and short-term (1–3 h) gill net sets. Sampling occurred in summer and fall of 2021 and spring, summer, and fall of 2022. Multiple gears were used due to differences in physical habitat that led to limited success with gill nets in the wetland and electrofishing in the lake. Differences in gear selectivity are not expected to substantially impact diet sampling (Glade et al. 2023; Pothoven et al. 2017; Quist et al. 2002; Selch et al. 2019). Collected Walleye were euthanized via carbon dioxide-induced hypoxia and measured for total length (mm) and weight (g). Stomachs were removed by cutting at the connections to the esophagus and small intestine, placed in a bag with distilled water, immediately stored in a cooler with ice to slow digestion, and frozen upon arrival to the lab.

We identified invertebrate diet items to family and fish diet items to the previously described taxa groups. However, invertebrates were an extremely small portion of diets (< 5%) and had little influence on results. Cleithrum morphology was used in addition to external characteristics to assist in the identification of digested prey fish (Traynor et al. 2010). Empty stomachs and stomachs with only inorganic material, detritus, and/or vegetation were recorded as empty. We blotted fish diet items dry with a paper towel and individually weighed them to the nearest 0.1 mg. Invertebrates were enumerated, blotted dry, and batch weighed by taxon within individual diets (0.1 mg; Bertoli et al. 2021).

We estimated the energy content of diets using wet weights of diet items and energy density estimates from our prey fish specific to the species group, season, and habitat. Up to 30 individuals from each prey fish species, habitat, and season from mini-fyke net samples were euthanized via carbon dioxide-induced hypoxia, measured for total length and weight, and frozen for later energy density estimation. If less than five individuals were captured for a given species, season, and habitat combination, boat electrofishing was used to supplement mini-fyke net collections. Within 6 months of capture, prey fish were thawed, weighed for wet weight (0.1 mg), dried at 60 °C until reaching a constant weight, and re-weighed for dry weight. We estimated energy density (ED, J/g wet weight) as,

$$ED=45.29*{\left(\frac{DW}{WW}\right)}^{1.507}$$

where DW is dry weight (g), and WW is wet weight (g; Hartman and Brandt 1995b). Energy densities were averaged by season, habitat, and species grouping. Two small sunfish from fall 2021 (1 from each habitat) were removed from analyses due to having energy density estimates over 6 standard deviations above the mean while all other individuals were within 2 standard deviations of the mean.

The most complete prey fish energy density data (> 10 individuals per season and habitat) were available for small sunfish. Therefore, relative differences among habitats and seasons observed in small sunfish were used to estimate missing energy densities in some seasons/habitats for other species groups. For example, we only captured bullheads from the wetland in fall, but sunfish energy density was 15% higher in the lake in fall; therefore, we estimated lake bullhead energy density as 15% higher than wetland energy density in fall. Shiners were used instead of small sunfish for all proportions including spring in the lake, as small sunfish were not collected in the lake during the spring. The average energy density of all prey fish in a season and habitat was used for unidentified fish in Walleye diets. Cypriniformes and White Sucker (Catostomus commersonii) collection was limited, so energy densities could not be estimated. Energy density for Common Carp < 10 g was obtained from Breck (2008) for Cypriniformes, and White Sucker energy density was obtained from Johnson et al. (2017) using the mean weight of White Suckers in Walleye diets. For energy densities of insects and Amphipoda, we used literature values for Ephemeroptera (Pizzul et al. 2009) and Gammaridae (James et al. 2012), as those were the dominant taxa in Walleye diets. Crayfish energy density was obtained from Eggleton and Schramm (2004).

We compared prey fish energy density between habitats and seasons using a linear mixed effects model (fixed effects of season and habitat; random effect of species group) using package “lme4” (Bates et al. 2015). Seasons sampled in consecutive years were combined due to small sample size of some species, season, and habitat groups. The best model was selected by likelihood ratio testing and significance was determined using Tukey’s test.

Walleye relative weight was calculated as

$$\frac{100 * Weight}{{10}^{-5.329 + (3.2 * {log}_{10}Length)}}$$

(Murphy et al. 1990) with the diet weight removed from the total weight of the fish. A stomach fullness indicator was calculated as

$$\frac{100 * Diet Weight}{Predator Weight - Diet Weight}$$

(Pothoven et al. 2017). We compared Walleye relative weight, diet weight, stomach fullness, and diet energy among seasons and between habitats using linear models. Percent empty stomachs was calculated and compared using binomial family generalized linear models. The best model was selected by likelihood ratio testing and significance was determined using Tukey’s tests. Proportions of diets by weight and by energy were used to compare the relative importance of taxa in Walleye diets across habitats and seasons (Garvey and Chipps 2012).

Results

Prey Fish

We captured a total of 3,470 prey fish from the lake and 2,021 prey fish from the wetland, but biomass catch per unit effort (biomass per hour) was significantly lower in the lake and during spring 2022 (additive model; Table 1). Prey fish taxa richness ranged from 4 to 10 species (\(\overline{x}\) = 6.5 ± 0.6) but showed no consistent differences between seasons or habitats (Online Resource 1). Shannon diversity was higher in the wetland in all seasons except spring 2022 (Fig. 2). Shannon evenness, Simpson diversity, and Simpson evenness showed similar patterns to Shannon diversity (Online Resource 1). Bray–Curtis dissimilarity indices did not show any consistent patterns, but values were between 0.60 and 0.97 (\(\overline{x}=\) 0.82 ± 0.06) indicating that prey communities were different between habitats regardless of season.

Table 1 Number of net nights and mean prey fish catch per unit effort (CPUE; g biomass / hr) for mini-fyke nets in Lake Kampeska (lake), South Dakota, and a connected wetland by season and year. Values in parentheses represent 1 standard error. Lower case letters indicate significant differences among seasons from Tukey’s test. Habitats were significantly different overall (p = 0.011)
Full size table
Fig. 2
figure 2

Shannon diversity index of prey fish communities in Lake Kampeska (lake), South Dakota, and a connected wetland in summer (SU) and fall (FA) of 2021 (21SU; 21FA) and spring (SP), summer, and fall of 2022 (22SP; 22SU; 22FA)

Full size image

Mean energy densities of prey fish were 2,982 ± 78 S.E. J / g wet weight or 3,276 ± 73 J / g wet weight in the lake and wetland, respectively (Table 2). Prey fish energy density was best described by the interactive model, so grouping habitats or seasons to increase sample size of species groups would not be appropriate (Table 2). Prey fish energy density was higher in the wetland during summer and higher in the lake during fall (interactive model; p < 0.02). In the lake, prey energy density was lower in summer than in spring or fall, and in the wetland, prey energy density was lower in summer than in spring (all p < 0.02).

Table 2 Seasonal energy density (ED; J/g wet weight) estimates for prey fish species in Lake Kampeska (lake), South Dakota, and a connected wetland in 2021 and 2022. Values from this study were obtained from dry weight to wet weight ratios (Hartman and Brandt 1995a). Bullheads (Ameiurus spp.; BHD), (Crayfish (Orconectes spp.; CRF), darters (Etheostoma spp. and Percina caprodes; DRT), shiners (Notropis spp. and Notemigonus crysoleucas; SNR), unidentified fish (UID Fish), Yellow Perch (Perca flavescens; YEP), White Bass (Morone chrysops; WHB), White Sucker (Catostomus commersonii; WHS), Fathead Minnow (Pimephales promelas; FHM). We defined small sunfish (Lepomis spp.) as < 60 mm, while large sunfish were ≥ 60 mm
Full size table

Walleye Diets

We collected a total of 100 Walleye from the lake and 79 Walleye from the wetland (Table 3). Relative weight was always highest among Walleye collected from the wetland (additive model; Fig. 3). Walleye collected in fall 2021 had higher relative weights than summer and fall 2022 (Fig. 3). Percent of empty stomachs was always highest among Walleye collected from the lake (additive model; Fig. 4a). Walleye collected in summer 2022 had a greater proportion of empty stomachs than during fall 2021 and 2022 and spring 2022 (Fig. 4a). Diet weights were greater for Walleye collected in the wetland, and no pairwise differences among seasons were found (additive model; Fig. 4b). Stomach fullness and diet energy both showed temporal variation where fall 2021 values were high and summers 2021 and 2022 were low (additive models; Fig. 4c-d). Stomach fullness was greater in the wetland (p = 0.004; Fig. 4c), but diet energy was not significantly different between habitats (p = 0.053; Fig. 4d).

Table 3 Number and mean total length (mm ± 1 standard error) of Walleye (Sander vitreus) collected for diet analysis from Lake Kampeska (lake), South Dakota, and a connected wetland by season and year. Values in parentheses represent the range of total lengths included
Full size table
Fig. 3
figure 3

Mean relative weight with diet weight removed of Walleye (Sander vitreus) captured in Lake Kampeska (lake), South Dakota, and a connected wetland in summer (SU) and fall (FA) of 2021 (21SU; 21FA) and spring (SP), summer, and fall of 2022 (22SP; 22SU; 22FA). Error bars represent 1 standard error. Lower case letters indicate significant differences among seasons from Tukey’s test. Significant differences between habitats from Tukey’s tests indicated in the right margin

Full size image
Fig. 4
figure 4

Percent empty stomachs (a), mean diet weight (b), mean stomach fullness (c), and mean diet energy (d) for Walleye (Sander vitreus) in Lake Kampeska (lake), South Dakota, and a connected wetland in summer (SU) and fall (FA) of 2021 (21SU; 21FA) and spring (SP), summer, and fall of 2022 (22SP; 22SU; 22FA). Error bars represent 1 standard error. Lower case letters indicate significant differences between seasons from Tukey’s test. Significant differences between habitats from Tukey’s tests indicated in the right margin

Full size image

We examined 195 and 233 prey items in Walleye diets from the lake and wetland, respectively. Proportions by weight and energy showed similar patterns each season (Fig. 5). Most common taxa were different between habitats in all seasons except for spring when bullhead was the primary Walleye prey taxon in both habitats. In general, Walleye in the lake relied heavily on Yellow Perch, White Bass, and bullheads (Fig. 5). Walleye in the wetland consumed mostly large-bodied Cypriniformes (i.e., Common Carp and Bigmouth Buffalo) in 2021 and bullheads, White Suckers, and sunfish in 2022 (Fig. 5).

Fig. 5
figure 5

Proportion by weight (a) and proportion by energy (b) of Walleye (Sander vitreus) prey items in Lake Kampeska (lake), South Dakota, and a connected wetland in summer (SU) and fall (FA) of 2021 (21SU; 21FA) and spring (SP), summer, and fall of 2022 (22SP; 22SU; 22FA). Prey items that made up < 1% by weight or energy were omitted for clarity. Bullheads (Ameiurus spp.), (Crayfish (Orconectes spp.), darters (Etheostoma spp. and Percina caprodes), shiners (Notropis spp. and Notemigonus crysoleucas), unidentified fish (UID Fish), White Bass (Morone chrysops; WHB), White Sucker (Catostomus commersonii; WHS), Yellow Perch (Perca flavescens; YEP). We defined small sunfish (Lepomis spp.) as < 60 mm, while large sunfish were ≥ 60 mm

Full size image

Discussion

Our study highlights the importance of lake-wetland connections for fishes that has also been documented in river floodplains (Modde 1997; Sarkar et al. 2021), Great Lakes coastal wetlands (Jude and Pappas 1992; Seilheimer and Chow-Fraser 2006), and marine estuaries (Madon et al. 2001; Meynecke et al. 2008; Woo et al. 2018). Walleye accessed more abundant and diverse prey in the wetland than in the lake, leading to improved condition of individuals collected from the wetland. This finding supports that of Dembkowski et al. (2014) that higher Walleye relative weight in high water years can be attributed to increased access to flooded terrestrial vegetation and other attributes associated with wetlands. Increased Walleye feeding in the wetland may be related to the higher relative abundance of prey fish in this habitat, and Walleye have been found to move between these habitats on a regular basis (Blackwell 2001; Cutler 2023). Walleye are also known to feed more effectively in low light conditions (Lester et al. 2004; Raabe et al. 2020), which the wetland provided due to high turbidity and macrophyte cover. However, water temperature in the wetland was also greater during part of the year (Online Resource 2), which increases the energetic requirements of Walleye. Walleye are known to thermoregulate by moving to more desirable habitat (Raby et al. 2018), and macrophyte cover may help mitigate higher temperatures by providing refuge from direct sunlight. Bioenergetic modeling could be used in the future to better explain the influence of each habitat on Walleye growth in this system.

The wetland provided a more diverse prey community than the lake which is likely more resilient to ecological change (Bernhardt and Leslie 2013; McCann 2000; Petchey 2000), including changes in climate, land use, habitat connectivity, disturbance regimes, and species introductions. Additionally, the prey fish communities were consistently different between habitats, further increasing the diversity and resilience of the system as a whole. However, small sample sizes (4–6 net nights per season and habitat) led to large seasonal and annual variation, making temporal trends difficult to detect. Alternatively, prey fish production may be sporadic and highly variable between years, further demonstrating the value of a diversity of habitats to consistently provide sufficient prey. More broadly, heterogeneity and connectivity are cited as some of the most important factors for maintaining ecosystem resilience (Dahl 2011; Grantham et al. 2019; Van Looy et al. 2019). The more diverse prey community of the wetland may also lead to better food resource partitioning among predators, decreasing the risk of competition for food resources (Sánchez-Hernández et al. 2017) and benefitting valuable sportfish.

Walleye feeding can also vary seasonally as different prey resources become available and as somatic and gonadal growth requirements change. We observed greater food consumption in fall and the lowest consumption during summer. This matched the findings of Quist et al. (2002), where Walleye in a large Kansas reservoir exhibited greater feeding rates during fall and had a higher proportion of empty stomachs and lower total consumption during summer. Many spring-spawning predators, including Walleye, feed heavily in fall in order to support gonadal growth and prepare for declining food resources over the winter (Quist et al. 2002). Although variability in prey CPUE was high, some of the highest CPUEs of prey fish also occurred in fall. Fish could be targeting these high-density resources (Lyons 1987) resulting in higher fall consumption. Low consumption during the summer can be attributed to increased metabolic and digestion rates (Kawaguchi et al. 2007; Madon and Culver 1993) and decreases in maximum consumption due to high, stressful temperatures (Kitchell et al. 1977).

Many trends and differences between habitats that we observed in summer and fall were reversed in spring. For example, spring was the only season in which prey diversity was higher in the lake than in the wetland, the only season that Walleye consumed more in the lake than the wetland, and the only season that the most important taxa in diets were similar between habitats. Additionally, we struggled to collect Walleye from the wetland during the spring, with only 10 captured in 24 h of electrofishing time. Seasonal life history events (i.e., spawning) contributed to the shift in habitat use in addition to a lack of differences in prey resources. Preyfish also appeared to use the wetland seasonally, resulting in differences in community composition throughout the year and changing benefits to feeding in the wetland. The dominance of fine sediments in the wetland provides less than ideal habitat for Walleye spawning (Johnson 1961; Raabe and Bozek 2012) which occurs in the spring. However, the abundance of submergent vegetation in the wetland is used by sunfish, crappies, and other taxa, creating high densities of young-of-year prey fish for Walleye and other piscivores to feed on. Alternatively, the lake has abundant shoreline gravel habitat to potentially be used by Walleye for spawning (Raabe and Bozek 2012), but may not be as beneficial for many of the prey species eaten by Walleye in the wetland. As a result, it was not surprising that the wetland was less important to Walleye during spring as compared to summer and fall. Therefore, for fish to move between habitats and optimize resource use, connectivity is necessary.

Currently, fish in the lake have access to only one main wetland despite historically abundant wetlands that were once connected to the lake. Similar trends have been observed throughout the Prairie Pothole Region and worldwide as human development has encroached upon shoreline habitats (Hartig and Bennion 2017; Johnston 2013; Sundar et al. 2015; Wright and Wimberly 2013). Wetland reconnection may be used as a tool for fisheries managers to increase growth, condition, and prey availability and to meet management objectives for sportfish populations. These reconnections around the lake may not only give Walleye and other sportfish access to more area of wetland, but also decrease the distance they may need to travel between thermal refugia or spawning habitat in the lake and high food availability of wetlands. Reconnections may also help create redundancy and resilience in the face of ecological change (e.g., changes in climate, water levels, human development, etc.). However, reconnection can also lead to unintended consequences. For example, reconnected wetlands can act as additional spawning, nursery, or feeding habitat for nuisance or invasive species including Common Carp (Bajer et al. 2009) and Silver Carp (Hypophthalmichthys nobilis; Coulter et al. 2017; DeGrandchamp et al. 2008). The large number of young carp that we found in Walleye diets supports the notion that Common Carp use these habitats for reproduction. As a result, some agencies have intentionally limited fish access to wetlands to minimize reproduction of these species (Iowa DNR 2022). Alternatively, seasonal differences in wetland importance for priority species may be leveraged to maximize benefits. For example, temporary disconnection of wetlands in spring may be effective at reducing Common Carp or Silver Carp recruitment while having minimal impacts on Walleye that could benefit more from summer and fall access. However, understanding how and when species of interest use wetland habitats may be needed to avoid negative impacts on non-target species that may rely heavily on wetlands as spawning habitat during spring (e.g., Bluegill and Northern Pike; Miller et al. 2001).

Access to wetlands has been severely limited in Lake Kampeska, throughout the Prairie Pothole Region, and globally. Yet many fish species, including Walleye, may use wetlands seasonally or throughout the year for feeding or as a thermal refuge. Through reconnection and restoration of wetlands, managers may be able to improve water quality and benefit fisheries. High connectivity and habitat heterogeneity is known to increase ecological resilience in a wide variety of ecosystems (Chapin et al. 2000; Levine et al. 2016; Walker 1995) including aquatic environments (Pander et al. 2018); therefore, continued access to wetlands may be necessary for fishery resilience now and into the future (Bernhardt and Leslie 2013; McCann 2000; Petchey 2000).