Effects of hurricane-induced hydrilla reduction on the largemouth bass fishery at two central Florida lakes

Abstract

Lakes Weohyakapka and St. Johns Water Management Area, Florida, experienced severe impacts from multiple hurricanes in August and September 2004, resulting in the loss of all submersed aquatic vegetation, primarily hydrilla (Hydrilla verticillata). We assessed at both lakes changes in largemouth bass (Micropterus salmoides) population size distribution, recreational fishing effort and success, angler expenditures, and catches of trophy fish in relation to disparate levels of hydrilla coverage for prehurricane (1999–2004) and posthurricane (2005–2009) periods. Tests revealed significant differences at both lakes in the population size distribution between prehurricane (high percentage coverage of hydrilla) and posthurricane (no hydrilla) periods. At both lakes, the population size distribution comprised more juvenile (age-1) largemouth bass before the hurricanes, indicating that a decline in recruitment strength coincided with the absence of hydrilla posthurricanes. Declines in directed fishing effort, angler expenditures, and angler catches of trophy-sized fish also occurred following the absence of hydrilla posthurricanes. These findings demonstrate an important link between radical changes in hydrilla coverage with recruitment of juvenile largemouth bass and the strength of the largemouth bass fishery. The Florida Fish and Wildlife Conservation Commission recently adopted a new agency position on hydrilla management that allows flexibility regarding waterbodies with limited or absent native submersed vegetation, recognizing that hydrilla at a low to moderate coverage can be beneficial to fish and wildlife.

Key words:

• angler expenditure
• creel survey
• hurricane
• hydrilla
• management
• recruitment
• submersed aquatic vegetation
• trophy fish

Estimates of recruitment, density, abundance, and survival of age-0 and age-1 largemouth bass (Micropterus salmoides) have been positively correlated with percentage coverage of submersed aquatic vegetation (SAV; Moxley and Langford 1982, Maceina et al. 1995, Hoyer and Canfield 1996, Tate et al. 2003, Havens et al. 2005). The nonnative macrophyte hydrilla (Hydrilla verticillata) is frequently the dominant submersed plant in Florida lakes (Tate et al. 2003, Bonvechio and Bonvechio 2006) and in southeastern reservoirs (Slipke et al. 1998, Kirk and Henderson 2006) and is present in systems throughout much of the United States (Michel et al. 2004). Hydrilla provides a complex habitat type that supports a high density and biomass of prey items, provides cover, and reduces predation on age-0 largemouth bass. In central Florida lakes, abundance (Tate et al. 2003) and survival (Moxley and Langford 1982) of age-0 largemouth bass are positively correlated with percentage cover by hydrilla. Thus, the presence and amount of hydrilla can play a critical role in a largemouth bass fishery.

It follows, then, that recreational fishers for largemouth bass also benefit from the presence of SAV in lakes and reservoirs, which has been borne out by a number of studies. Angler creel estimates including effort, catch, and success have been positively correlated with SAV coverage (Maceina and Reeves 1996, Wrenn et al. 1996, Bonvechio and Bonvechio 2006). At Lake Okeechobee, Florida, hydrilla is considered one of the most important aquatic macrophytes because of its benefit to the recreational fishery for largemouth bass (Furse and Fox 1994). Other researchers found that in several southeastern reservoirs, moderate to high levels (30–50%) of hydrilla coverage were preferred by anglers (Slipke et al. 1998, Henderson et al. 2003) because of the perception that hydrilla positively affected their fishing success. Lake managers must also consider other recreational users, however, and the cost associated with managing hydrilla for the benefit of one user group.

The economic impact of a largemouth bass recreational fishery and the submersed aquatic vegetation that affect the fishery is also important. For example, the economic value of the fishery at Guntersville Reservoir, Alabama, was estimated to be $1.65–$4.15 million (Wrenn et al. 1996). Modeling demonstrated increases in multiple facets of the local economies surrounding 2 reservoirs in South Carolina that experienced a large fluctuation in hydrilla coverage (between 33 and <0.1%), supported by increases in the number of trips by anglers following increases in hydrilla coverage (Henderson et al. 2003).

Natural disturbances such as hurricanes can have deleterious effects on shallow lakes with aquatic macrophyte communities dominated by SAV. The dislocation of SAV following a hurricane can affect not only the largemouth bass fishery but also the whole lake ecosystem, especially if SAV is drastically reduced and fails to recolonize. Accounts of SAV loss in Florida lakes due to hurricane impacts were documented in 1947 at Lake Apopka following a Category 4 hurricane. The lake was reported to have changed from a clear lake dominated by Illinois pondweed (Potamogeton illinoensis) and noted for its largemouth bass fishery to a turbid, phytoplankton-dominated lake with a poor fishery (Clugston 1963, Bachmann et al. 1999).

The most notable hurricane impacts to aquatic macrophyte communities have been documented at Lake Okeechobee. Wind-driven wave action coupled with high water levels affected SAV communities by physically uprooting plants and resuspending sediments, which strongly increased turbidity and suppressed SAV growth (Havens et al. 2001, Havens 2005, Abtew and Iricanin 2008, Rogers and Allen 2008). Lake Okeechobee experienced multiple hurricanes in 2004 and 2005, resulting in a drastic loss of aquatic macrophytes (>75% lakewide). Rogers and Allen (2008) found that the composition of the littoral fish community shifted from a high abundance of centrarchids and fundulids/poeciliids (families that supported Lake Okeechobee's important fisheries) prehurricanes, to a high abundance of omnivorous and obligate planktivorous species posthurricanes. These authors also reported that juvenile largemouth bass were not found in littoral samples posthurricanes.

Between 13 August and 26 September 2004, hurricanes Charley, Frances, and Jeanne made landfall in central Florida as category 4, 2, and 3 storms, respectively (Fig. 1). Because the presence of SAV is important in maintaining a high-quality largemouth bass fishery, we sought to quantify changes in the fishery at 2 central Florida lakes after hydrilla was drastically reduced by uprooting from wind-driven wave action and failed to recolonize following the 2004 hurricanes. Our objectives were to evaluate changes posthurricanes (2005–2009, relative to prehurricanes 1999–2004) in (1) largemouth bass population size distribution; (2) largemouth bass angling effort and success and the resulting economic impact; and (3) the catch of trophy-sized largemouth bass.

Figure 1 Paths of hurricanes Charley, Frances, and Jeanne, which made landfall between 13 Aug and 26 Sep 2004 in central Florida as category 4, 2, and 3 storms, respectively. Hurricane Charley moved in a N–NE direction, while Frances and Jeanne moved in a W–NW direction across Florida. (WIW = Lake Walk-in-Water; SJWMA = St. Johns Water Management Area.)

Materials and methods

Study sites and background

Lake Weohyakapka (27°49′18^″N; 81°24′48^″W), commonly referred to as Lake Walk-in-Water (WIW), is a 2964 ha eutrophic natural lake located in Polk County, Florida (Florida LAKEWATCH 2009). St. Johns Water Management Area (SJWMA; 27°47′55^″N; 80°44′12^″W), known as Lake Farm 13/Stick Marsh, is a 2630 ha hypereutrophic reservoir on the upper St. Johns River basin located in Brevard and Indian River counties, Florida (Florida LAKEWATCH 2009). Both lakes are managed for trophy largemouth bass with the use of harvest regulations. WIW is regulated with a 381–610 mm total length (TL) protective slot limit (3 fish per person per day bag limit, of which only one fish may be>610 mm), and SJWMA is catch and release only. The protective slot limit at WIW prohibits the harvest of fish where the lengths fall within the protected interval. Water levels fluctuate in both lakes, but mean water depth is generally <3 m (Florida LAKEWATCH 2009).

Hydrilla was first observed by biologists at WIW in 1990 and at SJWMA in 1987, shortly after initial flooding of the reservoir. Before the 2004 hurricanes, hydrilla was the dominant SAV species in both lakes (Florida LAKEWATCH 2009). Percentage area covered (PAC) and percentage volume infested (PVI) of SAV were determined for both lakes every other year between 1999 and 2007 during the summer (May–Aug) period of maximum aquatic macrophyte growth. At WIW between 1999 and 2003 (before the 2004 hurricanes), PAC and PVI averaged 83 and 52%, respectively. In 2005, PAC and PVI were only 6% and 2%, respectively, and by 2007 no hydrilla was detected. At SJWMA in 2003, PAC and PVI were 40 and 23%, respectively. Sampling posthurricanes at SJWMA in 2005, PAC and PVI were only 4% and 0.5%, respectively, and by 2007 no hydrilla was detected.

Secchi depth also decreased at both lakes posthurricanes, by an average of 0.6 m at WIW (from 1.4 m to 0.8 m) and 0.5 m (from 1.0 m to 0.5 m) at SJWMA (Florida LAKEWATCH 2009). This decrease in Secchi depth reflected an increase in turbidity from resuspension of sediments or increased chlorophyll concentrations (a measure of planktonic algal biomass). An increase in turbidity at both lakes likely contributed to the suppression of hydrilla growth due to light limitation posthurricanes.

Data collection and analysis

Largemouth bass were collected at WIW and SJWMA with a double-boom electrofishing boat equipped with a 5.0, 7.5, or 9.0 GPP Smith-Root electrofisher (680–1000 volts, 4–9 amps, and 60 pulses/s direct current); all fish were measured to the nearest mm TL. Sampling was conducted annually from 1999 to 2009 in the late winter to early spring (Feb–Apr), but no electrofishing samples were collected at SJWMA in 2000 or 2003. Sampling along electrofishing transects was conducted in the littoral habitat until a minimum of 300 largemouth bass had been collected and measured. Transect locations and sampling effort varied prior to adopting the Florida Fish and Wildlife Conservation Commission's (FWC) standardized sampling protocol in 2006 (Bonvechio 2009), but available littoral habitat was routinely sampled until a minimum of 300 fish had been collected and measured.

Concerns about the comparability of historical data on largemouth bass (including size structure) gathered along long-standing fixed electrofishing transects with data collected using the newly adopted standardized sampling protocols were addressed by Bonvechio et al. (2008). These authors reported that the proportional size structure of quality-size and preferred-size largemouth bass did not differ significantly between the 2 sampling methods at 6 central Florida lakes. Thus, for analysis, size data were pooled for prehurricane period (1999–2004) and posthurricane period (2005–2009), regardless of sampling method. We used a Kolmogorov-Smirnov (KS) test to determine whether population size distribution pre- and posthurricane periods differed significantly, using p ≤ 0.05 as our level of significance. The KS test compares the shape of the curves as a whole rather than individual size classes. Fish within the first strong mode of each length frequency were considered to be age-1 (DeVries and Frie 1996, Allen et al. 2003) in descriptions of trends in juvenile recruitment for pre- and posthurricane periods.

Estimates derived from angler creel surveys were used to evaluate largemouth bass-directed fishing effort (angler-h) and success (fish/angler-h). Data were obtained from a roving survey at WIW (1999–2006, 2009) and an access-point creel survey at SJWMA (1999, 2001, 2003, 2005, 2007–2009). An access-point creel survey was selected at SJWMA because there is only one access point for the lake, and all anglers could be intercepted there. Both surveys were designed using nonuniform probability sampling, described in Malvestuto et al. (1978) and Malvestuto (1996). All surveys were conducted during the peak largemouth bass fishing season (Jan–May at WIW and Jan–Apr atSJWMA).

Estimates of directed fishing effort for largemouth bass were used to determine changes in economic impact of fisheries at both lakes following the hurricanes. On both lakes, anglers were identified as resident or nonresident (i.e., out of state), and estimates of total effort were distributed as a proportion of each user group for each year. Angler expenditures were calculated from the sum of hours of fishing effort at an hourly expenditure rate of $18.20 and $43.89 for resident and nonresident largemouth bass anglers, respectively, based on economic data reported in the 1996 National Survey of Fishing, Hunting and Wildlife-Associated Recreation (US Department of the Interior, Fish and Wildlife Service, and US Department of Commerce, Bureau of the Census 1997). Because these data were not reported after 1996, a consumer price index inflation calculator (US Department of Labor, Bureau of Labor Statistics 2011) was used to estimate resident and nonresident angler expenditures for each year of the study to adjust for inflation, according to the average consumer price index for a given calendar year. These data represent a measure of the average change over time in the prices of all goods and services purchased for consumption by urban households.

Angler recognition programs are a common method for collecting information on the catch of trophy fish (Quinn 1987, Wilson and Dicenzo 2002, Wilde and Pope 2004, Myers and Allen 2005). The FWC Big Catch angler recognition program was established to award certificates to anglers who voluntarily reported catches of largemouth bass that exceeded the minimum size criteria (i.e., 610 mm or 3.6 kg). We documented catches of trophy largemouth bass using data collected from the FWC's Big Catch angler recognition program for each entire calendar year from 1999 to 2009 and evaluated the percentage of the total number of Big Catch entries submitted statewide caught at lakes WIW and SJWMA.

Results

We found significant differences (KS, p < 0.001) between the pre- and posthurricane population size distribution of largemouth bass in lakes WIW and SJWMA. For both lakes, a greater proportion of the population size distribution consisted of age-1 juvenile largemouth bass (e.g., first modal peak of size distribution) before the 2004 hurricanes, indicating that a decline in recruitment strength coincided with the absence of hydrilla posthurricanes (Fig. 2 and 3). For both lakes a greater proportion of the population size distribution consisted of intermediate-sized fish (300–450 mm TL) following the 2004 hurricanes, coinciding with the absence of hydrilla, but there was little change in the proportion of the population consisting of trophy-sized largemouth bass (>610 mm TL).

Figure 2 Pooled largemouth bass length frequencies prehurricanes (1999–2004) and posthurricanes (2005–2009) used to evaluate changes in the largemouth bass population size distribution at Lake Walk-in-Water. Pooled length frequencies consist of 3796 largemouth bass prehurricanes and 4074 posthurricanes.

Figure 3 Pooled largemouth bass length frequencies prehurricanes (1999–2004; excluding 2000 and 2003) and posthurricanes (2005–2009) used to evaluate changes in the largemouth bass population size distribution at St. Johns Water Management Area. Pooled length frequencies consist of 2641 largemouth bass prehurricanes and 2179 posthurricanes.

Estimates of total fishing effort directed at largemouth bass ranged from 15,411 to 78,708 angler-h per year (mean = 53,858 angler-h) at WIW and from 18,677 to 65,943 angler-h per year (mean = 35,430 angler-h) at SJWMA (Table 1). Estimates of fishing success for largemouth bass ranged from 0.40 to 0.89 fish/angler-h (mean = 0.66 fish/angler-h) at WIW and from 0.43 to 0.87 fish/angler-h (mean = 0.73 fish/angler-h) at SJWMA (Table 1). Fishing effort directed at largemouth bass was greatest for both lakes when hydrilla was abundant, prior to the hurricanes.

Table 1 Largemouth bass-directed fishing effort (angler-h), success (fish/angler-h), and estimated annual expenditures (US$), with averages, pre- and posthurricanes, at Lake Walk-in-Water (WIW) and St. Johns Water Management Area (SJWMA) between 1999 and 2009. Data used to evaluate changes at both lakes in angler effort, success, and expenditures pre- and posthurricanes. Asterisks denote missing data for the years in which no creel data were collected.

			WIW					SJWMA
			Expenditures					Expenditures
Year	Effort	Success	Resident	Nonresident	Total	Effort	Success	Resident	Nonresident	Total
Prehurricane period
1999	68,094	0.40	895,055	1,015,418	1,910,473	23,502	0.72	299,834	372,366	672,200
2000	65,901	0.64	815,947	1,206,291	2,022,238	*	*	*	*	*
2001	78,708	0.83	808,331	1,949,597	2,757,928	65,943	0.67	731,413	1,502,735	2,234,149
2002	58,326	0.58	450,388	1,849,028	2,299,415	*	*	*	*	*
2003	53,758	0.63	619,486	1,272,785	1,892,271	52,555	0.87	672,914	1,082,002	1,754,917
2004	46,053	0.54	423,789	1,411,396	1,835,184	*	*	*	*	*
Average	58,725	0.60	668,832	1,450,752	2,119,585	47,333	0.75	568,054	985,701	1,553,755
Posthurricane period
2005	49,954	0.89	735,448	955,145	1,690,593	42,891	0.86	563,459	984,117	1,547,576
2006	48,604	0.87	886,741	602,972	1,489,713	*	*	*	*	*
2007	*	*	*	*	*	18,677	0.43	278,493	411,641	690,134
2008	*	*	*	*	*	21,962	0.69	350,970	476,198	827,168
2009	15,411	0.59	233,984	360,678	594,661	22,484	0.85	397,335	391,287	788,622
Average	37,990	0.78	618,724	639,598	1,258,322	26,504	0.71	397,564	565,811	963,375

Fishing effort remained stable in the 2 years following the hurricanes (2005–2006) at WIW but declined dramatically in 2009. At SJWMA, fishing effort declined slightly in the year following the hurricanes (2005) but declined dramatically from 2007 to 2009. No creel data were available for 2006 at SJWMA to evaluate whether fishing effort remained stable, as was found in WIW, and no data were available in 2007–2008 for WIW to evaluate whether fishing effort declined other than in one year (2009), as was found in SJWMA; however, average annual angler effort declined 35% at WIW and 44% at SJWMA following the hurricanes (2005–2009). Despite decreased fishing effort at both lakes posthurricanes, estimates of success of fishing directed at largemouth bass were variable among the pre- and posthurricane periods at both lakes and did not decline posthurricanes in relation to the fishing effort.

Annual expenditures by anglers fishing for largemouth bass ranged from $594,661 to $2,757,928 (mean = $1,832,497) at WIW and from $672,200 to $2,234,149 (mean = $1,216,395) at SJWMA (Table 1). Average annual expenditure declined 41% at WIW and 38% at SJWMA following the hurricanes (2005–2009). Nonresident anglers accounted for 68% and 63% of the average annual expenditures on WIW and SJWMA, respectively, before the hurricanes, but these percentages declined to 51% and 59% after 2004. The decline of nonresident anglers accounted for 94% and 71%, respectively, of the reduction in average annual total expenditures for the posthurricane period. Thus, the reduced effort of nonresident anglers was a major contributor to the reduced total angler expenditures after the storms.

From 1999 to 2009, 2675 Big Catch certificates for largemouth bass were awarded to recreational anglers in Florida. Lake WIW produced the most certificates for largemouth bass of any lake in Florida (N = 358) during the prehurricane period (1999–2004), and SJWMA produced the second greatest number (N = 199). From 1999 to 2004 WIW and SJWMA together accounted for 31% of all largemouth bass Big Catch certificates statewide. Following the hurricanes (2005–2009), however, Big Catch certificates dropped 88% at WIW (N = 43) and 66% (N = 68) at SJWMA, together accounting for only 12% of all certificates issued statewide during that period. This dramatic decline in the number of trophy bass caught by recreational anglers coincided with the absence of hydrilla at lakes WIW and SJWMA following the 2004 hurricanes (Fig. 4). The total number of Big Catch certificates issued statewide dropped 49% from the prehurricane period (1772 certificates) to the posthurricane period (903 certificates).

Figure 4 The percentage of the total number of Big Catch entries from the Florida Big Catch angler recognition program submitted statewide each year for Lake Walk-in-Water (WIW) and St. Johns Water Management Area (SJWMA) between 1999 and 2009 used to evaluate changes in the catch of trophy largemouth bass prehurricanes (1999–2004) and posthurricanes (2005–2009) at both lakes.

Discussion

This study demonstrates that high percentage coverage (40–80%) of hydrilla was more strongly correlated with a greater proportion of age-1 largemouth bass than low percentage coverage of hydrilla (<5%). Hydrilla coverage values we report were collected during the summer period of peak aquatic macrophyte growth; however, hydrilla coverage at both lakes likely fluctuated during these years due to herbicide treatments and/or environmental conditions. Fluctuating hydrilla coverage may have also beneficially impacted recruitment strength at both lakes.

Other studies have documented a positive relationship between SAV coverage and recruitment of juvenile largemouth bass in Florida (Havens et al. 2005) and southeastern water bodies (Miranda and Pugh 1997). Similar to the results of our study, Maceina et al. (1995) and Tate et al. (2003) documented increases in juvenile largemouth bass recruitment and abundance at high SAV coverage levels (∼60–90%). Like other species of SAV, hydrilla provides a complex habitat type that protects largemouth bass nests from wind-driven waves, allows greater production and availability of prey items, and affords safety from predators. Our study and similar studies suggest a direct link between SAV coverage (including hydrilla) up to high coverage levels and largemouth bass recruitment strength.

Directed fishing effort for largemouth bass declined at both WIW and SJWMA following the absence of hydrilla posthurricanes, indicating that recreational anglers may have responded to the absence of hydrilla by fishing less or shifting fishing effort to other systems. Wrenn et al. (1996) reported similar results at Guntersville Reservoir, Alabama, where largemouth bass-directed effort declined 46% in the years following a reduction in SAV coverage from approximately 30% to 10%. Conversely, at Lake Seminole, Georgia, Slipke et al. (1998) found that largemouth bass angler effort declined approximately 40% over a 17-year period as hydrilla coverage increased from 8% in 1979 to 50% in 1996 (with a peak coverage of 65% in 1992). Despite the overall decline in fishing effort, Lake Seminole largemouth bass anglers expended the greatest amount of effort fishing the reservoir arm that contained the greatest percentage coverage (85%) of hydrilla (Slipke et al. 1998). The researchers speculated that reduced access around the lake due to extensive coverage by hydrilla may have contributed to the overall decline in fishing effort. Thus, angler effort seems to be positively associated with hydrilla coverage to an extent, but a negative correlation may result when percentage coverage becomes too high (Colle et al. 1987).

We used fishing effort as an index of the quality of the largemouth bass fishery, assuming that if angling satisfaction declined there would be a corresponding decline in fishing effort (Cox et al. 2003). Other studies have found a similar relationship between angling satisfaction and fishing effort and their relationship with the presence of SAV or hydrilla. Henderson (1996) estimated that at Guntersville Reservoir, Alabama, the number of angler visits would increase as much as 53% if SAV coverage returned to the peak coverage of 30%. Henderson et al. (2003) showed that 62% of creel survey respondents believed that hydrilla helped their fishing success at Lake Murray, South Carolina, while at Lake Moultrie, South Carolina, 58% believed that removal of hydrilla hurt their fishing. Wilde et al. (1992) reported similar results from a mail survey study in which largemouth bass anglers expressed opposition to aquatic vegetation control in bodies of water in Texas. Thus, we believe that many anglers directly associate the presence of largemouth bass with the presence of SAV (e.g., hydrilla), and that anglers believe that SAV is directly linked to their ability to locate and catch largemouth bass.

If SAV is removed or lost from a system, largemouth bass anglers may fish there less, believing that fewer largemouth bass are present or are more difficult to catch with SAV now lacking. A decrease in angler effort was observed at both WIW and SJWMA after hydrilla was absent; however, no definable trend in largemouth bass-directed fishing success (largemouth bass caught per hour) was evident between prehurricane (high percentage coverage of hydrilla) and posthurricane (no hydrilla) periods, likely the result of knowledgeable local anglers adapting to the change in habitat with alternate fishing techniques.

Adaptive fishing methods were evident at both lakes. Local anglers successfully learned to fish other types of aquatic structures, including giant bulrush (Scirpus californicus) at WIW and large woody debris at SJWMA, which enabled them to catch large numbers of largemouth bass, despite the overall decrease in fishing effort. The continued fishing success is contrary to findings by Bonvechio and Bonvechio (2006), who reported a positive correlation between hydrilla coverage and angler success for largemouth bass at West Lake Tohopekaliga, Florida, over a 20-year period. Wrenn et al. (1996) also reported a positive correlation between SAV coverage and angler success for largemouth bass, and Slipke et al. (1998) found that largemouth bass anglers had the highest success fishing the reservoir arm that contained the greatest percentage coverage of hydrilla. Colle et al. (1987) reported that during an 8-year study at Orange Lake, Florida, angler success for largemouth was greatest during the year when hydrilla exceeded 80% coverage. Similar results were seen for largemouth bass tournaments, with a positive correlation between angler success and SAV coverage (Maceina and Reeves 1996, Wrenn et al. 1996).

Our results also indicate the economic importance of recreational fishing for largemouth bass and suggest a relationship between the presence of hydrilla and recreational expenditures at both WIW and SJWMA. Angler expenditures were greatest on both lakes when hydrilla was abundant before the hurricanes and declined strongly as fishing effort decreased following the absence of hydrilla posthurricanes. Over a 4-year period at Guntersville Reservoir, Alabama, Wrenn et al. (1996) found that angler trip expenditures declined by $1.4 million, concurrent with a 46% reduction in angler effort, in the years following the reduction in SAV coverage. Similar economic results were presented by Henderson et al. (2003) for lakes Murray and Moultrie in South Carolina following reductions in hydrilla coverage. Resident anglers were always the primary users of both lakes during the study period; however, the reduction in average annual expenditures by nonresident anglers made up a major portion of the economic impact on these fisheries. Nonresident anglers may have responded to the absence of hydrilla by traveling less to fish either lake, or by shifting their travel plans to fish other systems.

We documented that the catch of trophy-sized largemouth bass dramatically declined following the absence of hydrilla posthurricanes at both WIW and SJWMA. The Florida Big Catch program, which we used to assess changes in the catch of trophy-sized largemouth bass, does have biases and limitations, including angler reporting rates and a lack of validation of reported catches (Quinn 1987). Length–frequency analyses, however, indicated little change in the proportion of the population made up of trophy-sized largemouth bass following the hurricanes at both lakes. Nevertheless, largemouth bass-directed fishing effort and catches of trophy fish were greatest at both lakes when hydrilla was abundant.

We speculate that the presence of hydrilla (or other SAV species) at moderate to high percentage coverage is important to the ability of anglers to catch trophy-sized largemouth bass. The number of guide services and tackle/bait shops in the communities surrounding WIW and SJWMA declined following the absence of hydrilla posthurricanes. This decline may have affected participation in the Florida Big Catch program due to a lack of program promotion from these businesses that were dependent on the popular largemouth bass fishery at both lakes.

Total participation in the Florida Big Catch program also declined after 2004, which may suggest that the decrease in the number of certificates at WIW and SJWMA was not solely due to the absence of hydrilla. Lack of sustained interest and participation are problems for angler recognition programs (Quinn 1987) and could explain some of the decline in participation in the Florida Big Catch program at WIW and SJWMA and statewide, which hindered our analysis of these data; however, the rate of decline in the number of Big Catch entries posthurricanes at WIW and SJWMA were greater than the statewide decline during the study period, which may point to a legitimate decline in the catch occurrence of trophy largemouth bass following the absence of hydrilla.

The absence of hydrilla seemed to be the driving force behind the apparent changes in largemouth bass recruitment and may have been related to changes in fishing effort and catch occurrence of trophy fish posthurricanes; however, other factors must be considered. Sampling vulnerability of age-1 largemouth bass to electrofishing or the distribution of age-1 largemouth bass could have changed following loss of hydrilla at both lakes, regardless of any real decline in recruitment strength. If the catchability of age-1 fish decreased following the loss of hydrilla, then their percentage frequency in the pooled length frequencies would be underestimated. Similarly, if the spatial distribution of age-1 fish changed following the loss of hydrilla, then their vulnerability to the littoral zone electrofishing would also underestimate their percentage frequency in the pooled length frequencies. The decline in fishing effort and catches of trophy largemouth bass at WIW and SJWMA could have resulted from an increase in angler success at other nearby lakes in central Florida, causing a natural shift in angler effort and trophy catches to other bodies of water that were not impacted by the hurricanes. Other factors, such as changes in lake levels, water quality, and/or the economic recession, may have also contributed to the decline.

Before the hurricanes, hydrilla coverage was high at WIW and SJWMA because of herbicide costs and efficiency (i.e., resistance), not because of recommendations of managers to manage it at some coverage level. Although moderate to high coverage of hydrilla seems beneficial to largemouth bass recruitment, negative impacts can occur at the expense of the fishery and nonangling recreational users. Colle and Shireman (1980) found that high coverage of hydrilla (>30%) at 2 central Florida lakes was associated with decreases in growth and condition of adult largemouth bass. At the multipurpose Guntersville Reservoir, Alabama, Henderson (1996) reported that nonangling user groups perceived the nonnative SAV species present in the lake as bothersome and preferred less vegetation coverage or none at all. Lake managers must therefore consider all user groups and the potential negative impacts when formulating aquatic plant management objectives.

Hydrilla was first reported in Florida in 1960 and spread rapidly throughout the state, becoming the dominant submersed plant in many Florida lakes and reservoirs (Langeland 1996). Our results suggest that hydrilla can serve an important role in largemouth bass fisheries when native submersed vegetation is limited or absent. The management strategy for hydrilla in Florida lakes before 2008 focused on control at the lowest feasible coverage due to the threat of expansion, which can create competition with desirable native vegetation, impede navigation, and reduce water quality.

The FWC recently adopted a new agency position on hydrilla management (FWC 2011) that allows greater flexibility regarding bodies of water for which native SAV is absent or limited. The FWC still maintains the position that native aquatic vegetation communities in Florida water bodies provide the ecological functions needed to support diverse fish and wildlife communities, and still considers hydrilla to be an invasive nonnative macrophyte that, once established, has proven extremely difficult, if not impossible, to eradicate and expensive to manage. Although FWC prefers managing for native aquatic macrophytes and opposed the deliberate introduction of hydrilla, it recognizes that in water bodies where native submersed vegetation are absent or limited, hydrilla can be beneficial to fish and wildlife when maintained at low to moderate percentage coverage. In water bodies where hydrilla is already well established, FWC will manage hydrilla at levels that correspond with the primary uses and functions of that water body (FWC 2011). Hydrilla management actions are considered separately for each body of water using a risk-based analysis approach that takes into account input from all stakeholders. A risk-based approach to managing nonnative macrophytes should be considered by other management agencies.

Acknowledgments

The authors thank Travis Tuten, Bob Wattendorf, Richard Krause, and the 3 anonymous reviewers for their comments, which led to improvements in this paper.

Funding

This project was partially supported by the USFWS Sport Fish Restoration Program.