Comparison of the Species Composition, Catch Rate, and Length Distribution of the Catch from Trap Nets with Three Different Mesh and Throat Size Combinations

—Trap nets of varying design are commonly used to assess fish populations, but the effect of the design on gear selectivity is not well known. In particular, it may be advantageous to use multiple net designs with different mesh and throat sizes to maximize the catch of specific length-classes and to minimize the risk of predation on small fish by larger fish. We compared the species composition, catch rate, and length distribution of fishes caught by three trap net designs with dimensions differing only in mesh size and throat size (0.6-cm delta mesh and 3.8-cm 3 3.8cm square throats, 1.3-cm square mesh and 7.6-cm 3 7.6-cm square throats, or 2.5-cm square mesh and 12.7-cm 3 12.7-cm square throats). A total of 3,473 fish of 18 species were captured from Sandy Lake, Portage County, Ohio, during 24 sample dates from June to August 1999. The large net design had a significantly higher average number of species captured (mean 5 11) than the medium or small net design (means 5 9 and 8, respectively). Whereas nets with larger mesh and throat size combinations typically caught larger fish for the six most commonly captured species, only rarely did nets capture fish as large or as small as possible based on their physical dimensions. Specific length-classes of some species were not captured in the nets or were very net design specific, indicating a possible difference in trap net vulnerability of different ontogenetic stages. We conclude that data from trap nets with different mesh and throat sizes should not be directly compared with each other, and that multiple net mesh and throat sizes (or even multiple gear types) should be used when a more complete picture of fish length and abundance is desired. Trap nets are commonly used to assess fish populations (Hubert 1996). While many trap-net designs have been described (e.g., using different mesh sizes, throat sizes and shapes, and heart dimensions), few studies have compared the catch among these designs. All commonly used fish sampling methods are biased with respect to species, size, or sex of the fish caught (Hubert 1996). Therefore, it is important to know the catch bias of sampling equipment so gear can be selected that will target the species, size, or sex of interest, or a combination of gear can be used to minimize the bias of samples of the overall fishery. The minimum size of fish that can be captured by a trap net is ultimately set by mesh size. Accordingly, several studies have found that nets with smaller mesh tend to catch either smaller fish or more fish in small length-classes for several spe* Corresponding author: [email protected] 1 Present address: Kaskaskia Biological Station, Center for Aquatic Ecology, Illinois Natural History Survey, Rural Route 1, Box 157, Sullivan, Illinois 61951, USA. Received April 23, 2001; accepted July 22, 2002 cies: yellow perch Perca flavescens (Kraft 1990; Kraft and Johnson 1992), channel catfish Ictalurus punctatus (Hesse et al. 1982; Holland and Peters 1992), white crappie Pomoxis annularis (Willis et al. 1984; Jackson and Bauer 2000), Black crappie P. nigromaculatus (McInerny 1988; Jackson and Bauer 2000), and bluegill Lepomis macrochirus (Jackson and Bauer 2000). However, some studies have also found that differences in trap-net mesh size did not affect the length distribution of captured fish (white crappie, Willis et al. 1984; northern pike Esox lucius, Clark and Willis 1989). Studies have also found that trap-net mesh size affects catch per unit effort (CPUE) for channel catfish (Hesse et al. 1982; Holland and Peters 1992) and some length-classes of black crappie (McInerny 1988). However, no consistent relationship between trap-net mesh size and CPUE has been found for yellow perch (Kraft 1990) or northern pike (Clark and Willis 1989). We know of no study that has investigated the relationship between trapnet mesh size and fish size or CPUE for many fishes that are commonly captured in trap nets (e.g., redear sunfish Lepomis microlophus, pump463 CATCH VARIATION AMONG TRAP NETS FIGURE 1.—Trap-net design used to test the effect of different mesh and throat size combinations on catch bias. kinseed Lepomis gibbosus, brown bullhead Ameiurus nebulosus, and largemouth bass Micropterus salmoides). Additionally, we know of no study that has compared the minimum size of fish actually captured by a trap net with the minimum size a given mesh size could physically retain. The maximum size of fish that can be captured by a trap net is ultimately set by the throat size of the net. However, we are unaware of any study that has investigated the relationship between throat size and fish size or CPUE. Theoretically, building trap nets with fine mesh size and large throats would maximize the range of fish sizes the net could catch. However, large throats may not be efficient at retaining small fish, and larger fish allowed into nets with large throats may prey on smaller individuals inside the net. Additionally, small mesh netting is heavy and expensive. Thus, it would seem most efficient to use multiple gear sizes and to vary mesh size and throat size together when using trap nets to sample fish of different sizes. However, with the limited information on the effects of mesh and throat size, this idea remains untested. Therefore, we compared the species composition, CPUE, and size distribution of fishes caught by three different trap-net designs of differing mesh and throat size. Methods Trap nets were designed so that only mesh and throat size varied among net types (Figure 1). The nets were constructed by sewing pieces of tartreated nylon netting together with tar-treated #21 nylon twine. Square frames were constructed from 2.54and 0.64-cm-diameter, schedule 20 polyvinyl chloride (PVC) pipes for the one heart and four pot frames, respectively. The first three pot frames contained square throats (Figure 1). Two nets were constructed for each of the three mesh and throat size designs. Each pair of nets with the same mesh and throat size combination was attached together with a single 30-m lead net fastened to one side of each net’s heart with 10.2cm cable ties. The lead net was set perpendicular to the opening of each net. A wing net was then attached to the other side of the each net’s heart with 10.2-cm cable ties and set at a 45o angle to the lead net. Leads and wings for each pair of nets were constructed from the same mesh as the nets. Floats (number 125 hard plastic, 12.7 cm long, 3.8cm diameter) were hung on the top of the leads and wings every 91 cm. Lead weights (23.3 g) were hung on the bottom of the leads and wings every 37 cm, and the wings and leads were attached to the trap nets with 10.2-cm cable ties.