Fish density estimates in aquatic plants are inherently variable due to sampling techniques, plant density, and patchy fish distribution. Dibble et al. (1996) made a frequency list of the methods and tools used to quantify fish in or near vegetated areas. The numbers correspond to individual papers listing a particular method: electro-shocker (11); divers (11); seine (10); rotenone (8); popnets (7); gillnets (7); drop or throw nets (6); light trap or minnow trap (6); explosives (6); belt transect (5); rotenone and block net (5); modified traps (3); push net (3); underwater camera (3); electro-shocker and block net (3); modified nets (2); strip counts (2); ten other methods, each used once.
Popnets, drop nets and throw nets seem to be most effective in sampling fish in aquatic macrophytes. Popnets and drop nets measure distribution, diversity and abundance of adult and juvenile fish in densely vegetated areas where traditional methods (i.e. seining and electrofishing gear) are ineffective. Killgore et al. (1988, 1991) and Morgan II et al. (1988) used popnets (Fig. 44) in their study of fish in Myriophyllum in Lake Guntersville in Alabama, and in Hydrilla beds in the Potomac River (USA), respectively. Variability between individual popnet collections were high for each site in the Potomac River indicating patchy distributions within the Hydrilla beds. In contrast to techniques that require the collection of stunned (electrofishing) or dead (rotenone) fish with a dip net in dense plant beds, the popnets reduce the potential bias of underestimating fish abundance that is associated with the above methods. Larson et al. (1986) reported that the sampling efficiency of the popnets was near 100 percent. Freeman et al. (1984) who used a drop trap (1 m2) recovered 90 percent of tagged fish. Enclosure traps, such as drop nets and popnets, may have the least source of bias in estimating the density of fish residing in dense plant beds because all fish within the net can be collected or accurately estimated (Killgore et al., 1988). This technique can also be easily replicated. However, fish biomass estimates may be underestimated because larger fish can escape more easily than smaller fish. Killgore et al. (1988) also commented on the use of electrofishing in plant beds: the efficiency of electrofishing generally decreases as plant density increases because of the difficulty in locating and collecting stunned fish in dense vegetation.
a. Popnet set on bottom-float and lead line attached with pin/key system
b. Popnet fully extended subsequent to release of floatline
Fig. 44. Popnet for sampling fish among submersed aquatic vegetation. (From Killgore et al., 1991).
Chick and McIvor (1994) in their study of fish in macrophytes of Lake Okeechobee used for sampling the fish a circular drop trap and square throw trap. Both of them were found to be effective for sampling fish in shallow vegetated habitats.
In two freshwater streams in Costa Rica to obtain estimates for fish biomass per unit area with floating aquatic plants (Eichhornia, Hydrocotyl, Salvinia, Azolla) a small seine net was hauled moving the leadline under the floating vegetation mass, then quickly raising the trapped vegetation and fish above the water and removing them from the net at the shore or in a boat. The area of the haul was the area of the seine net (Winemiller and Ponwith, 1998).
In a study of fish in the varzea lake Mamiraua, situated between the rivers Solimoes and Japura, the fish within floating vegetation mats were sampled quantitatively using a 23 m sand eel seine net. The net was made up of five panels of differing mesh size. The vegetation was sampled by encircling an area of 2 to 12 m2 with the net deployed from a boat. The lead line was then pulled under the root mat and the mass of plants pulled towards the boat (Henderson and Hamilton, 1995).
When rotenone sampling in vegetated areas the efficiency of fish recovery decreases with the increasing plant density (Shireman et al., 1981). Seines were commonly used near vegetation, but were difficult to use in dense plant beds.
Electrofishing has been used in numerous studies. However, electrofishing is known to be selective, with smaller fish being less vulnerable to electric shock than larger fish (Zalewski and Cowx, 1990). The presence of vegetation may also hide fish fleeing from the electrical current, and contribute to a lower catchability (Randall et al., 1996). Rider et al. (1994) compared rotenone and electrofishing for largemouth bass in vegetated (Myriophyllum spicatum) and non-vegetated areas in Guntersville Reservoir in Alabama (Fig. 45). The results from both methods used were not significantly different, but using rotenone was more costly than electrofishing.
 |  |
 |  |
Fig. 45. Length-frequency distributions of age-0 largemouth bass collected from vegetated and non-vegetated areas using electrofishing and rotenone. (From Rider et al., 1994).
Using grass carp in aquatic macrophyte control raises sometimes a problem of its recapture, after the job has been done. In a small water body of only 0.2 ha in Florida, USA, Mallison et al. (1994) used rotenone bait, which succeeded in removing up to 79% of the triploid grass carp present. Using Fish Management Bait (FMB), as the authors call it, may be feasible in lakes if moving feeders consistently attract naïve fish and if the majority of the triploid grass carp population can be attracted to feeders. FMB was selective to triploid grass carp and resulted in minimal kill of the native fish species. Public angling has also been used as a method of triploid grass carp removal (Mallison et al., 1994a). Trammel and gill nets were successful in recapturing grass carp in Luisiana (Richardson, 1994). Bonar et al. (1993) evaluated the effectiveness of various capture methods of grass carp in a vegetated lake. Herding of fish was the most effective of the seven techniques, followed by angling in baited areas. Herding removed 0.8% to 8.2% of the carp stock in one sweep in lakes containing dense vegetation, submersed logs and other underwater obstructions. The authors recommended this method for reducing the numbers of fish in overstocked water bodies smaller than 10 ha, or for capturing fish to monitor growth. Electroshocking has a potential as a herding device.
