Glover`s Reef Marine Reserve Long-term Atoll Monitoring Program

Glover’s Reef Marine Reserve Long‐term Atoll Monitoring Program (LAMP) Results for queen conch, spiny lobster and key finfish species for the 2004‐2012 survey period September 2013 Wildlife Conservation Society, Global Conservation Program, Belize Belize City, Belize ACKNOWLEDGEMENTS The results presented in this report are based on field surveys conducted by the staff of the Wildlife Conservation Society’s Belize, Global Conservation Program with the invaluable assistance of the Belize Fisheries Department’s Glover’s Reef Marine Reserve Staff and volunteers. A special thank you to Dr. Samantha Strindberg for providing technical expertise in data analyses for this report. Funding for the study was provided generously by the Oak Foundation and the United States Agency for International Development. Photo on cover page: Long‐term Atoll Monitoring Program surveyor, A. Miranda, running transect tape for conch surveys. (R.Coleman/WCS). This report is made possible by the generous support of the American people through the United States Agency for International Development (USAID). The contents are the responsibility of the Wildlife Conservation Society and do not necessarily reflect the views of USAID or the United States Government. ii
TABLE OF CONTENTS ACKNOWLEDGEMENTS ............................................................................................................ii LIST OF TABLES ........................................................................................................................iv LIST OF FIGURES.......................................................................................................................iv ACRONYMS..............................................................................................................................v EXECUTIVE SUMMARY .............................................................................................................1 INTRODUCTION .......................................................................................................................2 Long‐Term Atoll Monitoring Program ................................................................................ 2 Aim and Objectives of Study.................................................................................. 2 Glover’s Reef Marine Reserve ............................................................................................ 3 METHODOLOGY ......................................................................................................................4 Sampling Sites..................................................................................................................... 4 Species Survey .................................................................................................................... 6 Queen Conch ......................................................................................................... 6 Spiny Lobster ......................................................................................................... 6 Finfish .................................................................................................................... 6 Environmental Variables .................................................................................................... 6 RESULTS...................................................................................................................................7 Queen Conch ...................................................................................................................... 7 Spiny Lobster ...................................................................................................................... 12 Finfish ................................................................................................................................. 15 SUMMARY AND CONCLUSIONS................................................................................................24 LITERATURE CITED ...................................................................................................................24 APPENDICES ............................................................................................................................25 iii
LIST OF TABLES AND FIGURES Table 1 Figure 1 Figure 2 Figure 3 Figure 4 Distribution of sampling sites by habitat and management zone in the Glover’s Reef Marine Reserve Location of Glover’s Reef Atoll, Belize Bathymetry of Glover’s Reef Atoll Glover’s Reef Marine Reserve: Management Zones Glover’s Reef Atoll with the sampling sites used for the Long‐term Atoll Monitoring Program (LAMP). Figure 5 Glover’s Reef Atoll showing habitats of sampling sites used for the Long‐term Atoll Monitoring Program (LAMP). Figure 6 The estimated overall queen conch density by (a) fishing season and (b) management zone. Figure 7 Estimated queen conch density stratified by (a) all age classes and (b) without small juveniles. Figure 8 Estimated queen conch density stratified by (a) legal size limit (176mm), (b) habitat type with (c) the adjusted y‐axis scale showing the variability at a smaller scale. Figure 9 (a) Percent queen conch by age class over time and (b) Percent of legal and sub‐legal queen conch over time. Figure 10 The estimated overall density for spiny lobster by (a) fishing season, (b) management type and (c) legal size limit (76.2mm). Figure 11 (a) Percent spiny lobster by sex class over time and (b) Percent of legal and sub‐legal spiny lobster over time. Figure 12 Estimated conditional dependence of spiny lobster density over time using Generalized Additive Mixed Models Analyses. Figure 13 The estimated biomass (a) overall and by (b) management type for key finfish. Figure 14 The estimated biomass by management type for the key finfish species (a) black grouper, (b) hogfish, (c) mutton snapper, (d) Nassau grouper, (e) princess parrotfish, (f) redband parrotfish, (g) redtail parrotfish, (h) stoplight parrotfish, (i) striped parrotfish, and (j) yellowtail parrotfish. Figure 15 Percent by fork length class over time for the key finfish species (a) black grouper, (b) hogfish, (c) mutton snapper, (d) Nassau grouper, (e) combined parrotfish species (princess, redband, redtail, stoplight, striped, yellowtail). Appendix 1 Generalized Additive Mixed Models (GAMMs) results for queen conch (Strombus gigas) Appendix 2 Generalized Additive Mixed Models (GAMMs) results for Finfish 5 3 3 3 5 5 8 9 10 11 13 14 14 15 19 21 25 27 iv
Acronyms CARICOMP Cl CZ GAMM GRMR GUZ LAMP MPA WCS WZ YSI Caribbean Coastal Marine Productivity Carapace length Conservation Zone Generalized Additive Mixed Models Glover’s Reef Marine Reserve General Use Zone Long‐term Atoll Monitoring Program Marine Protected Area Wildlife Conservation Society Wilderness Zone Yellow Springs Instrument v
EXECUTIVE SUMMARY Fishing is one of the most important economic activities in Belize, and the Marine Protected Area (MPA) system is expected to contribute to sustainable fisheries by providing refuge areas that allow for species reproduction and ultimately, the replenishment of adjacent fished areas. Long‐term monitoring is required to determine the status of these commercially‐exploited stocks within MPAs. In 2004, a long term fishery‐independent monitoring program was introduced at Glover’s Reef Atoll. The aim of the Glover’s Reef Long‐term Atoll Monitoring Program (LAMP) is to collect baseline information and data over time that will be used to determine the current status and monitor trends of commercial fish species (distribution, density, size class structure, reproduction) and habitat quality. The information on the population dynamics of target species will also be used to develop recommendations to guide management decisions on fishing quotas, length of fishing season, size limits and other regulations to ensure profitability and sustainability of the fishery. This report presents the results of 24 sampling occasions of the Long‐term Atoll Monitoring Program (LAMP) for Glover’s Reef Atoll covering the period August 2004 till May 2012. The data were collected at sample sites located within the Atoll’s three main management areas: the Wilderness, Conservation and General Use Zones. The report focuses on data collected on the spiny lobster (Panulirus argus); queen conch (Strombus gigas); five commercial finfish species (Nassau grouper Epinephelus striatus, Black grouper Mycteroperca bonaci, Hogfish Lachnolaimus maximus, Mutton snapper Lutjanus analis and Queen triggerfish Balistes vetula); and six species of parrotfish (Stoplight Sparisoma viride, Redtail Sparisoma chrysopterum, Yellowtail Sparisoma rubripinne, Princess Scarus taeniopterus, Striped Scarus croicensis and Redband Sparisoma aurofrenatum). For queen conch there was no indication of a significant change in adult mean density over time, whereas for large juveniles there was a significant increase in mean density. However, for both of these age classes the estimated mean density for the last sampling period for the LAMP at Glover’s Reef atoll are well below the minimum density required for high mating frequency. The sand algal flats and seagrass beds continue to show high densities of all age classes of queen conch, particularly medium and small juveniles. Overall mean density across all age classes was lower in the General Use Zone, however, medium and small juvenile conch had significantly higher densities in the General Use Zone compared to the Conservation Zone. The spiny lobster showed an increase in mean density over time with the highest of 39.3 individuals per hectare recorded during the May 2012 sampling period. This overall increase in mean density, however, was not consistently higher for either the Conservation or General Use Zone. The LAMP study has also showed that in general, over time the relative proportion of legal size (>76.2mm) to sub‐legal size spiny lobster is higher. Overall, for all finfish, the mean biomass was higher in the Conservation Zone than the General Use Zone. Over time, there was an initial decline in biomass in the CZ, followed by an increase and then a decline for the last few sampling occasions. For hogfish and mutton snapper there was a statistically significant decline in biomass in both management zones, and for black grouper the decline was not statistically significant. For Nassau grouper the biomass remained stable in the CZ over time, but there was a significant decline in biomass in the GUZ. Overall, for parrotfish species even though the biomass was consistently higher in the CZ, the estimated values are low even for the parrotfish which are legally protected since 2009. 1 INTRODUCTION Long Term Atoll Monitoring Program (LAMP) The Long Term Atoll Monitoring Program (LAMP) is a fishery‐independent monitoring program designed specifically for the long‐term monitoring of physical and biological parameters at the Glover’s Reef Atoll and for generating data comparable to the existing Caribbean Coastal Marine Productivity (CARICOMP) dataset. The CARICOMP program is a comprehensive, long‐term plan for research and monitoring in the Caribbean basin. Fishery‐independent monitoring involves sampling of habitats of the target species to get direct estimates of the population in its natural habitat. In Marine Protected Areas, this type of monitoring in conjunction with an appropriate study design also allows for the comparison of the impact in the different management zones and for detecting changes in a fished population (e.g. changes due to over fishing). Aim and Objectives of Study Aim: To monitor and analyze the viability of a fished population in order to determine trends showing increase, decrease or stability of the population. Objectives: 1) To gather data on the number of animals in each size class of the population. 2) To gather data on the number of adults that are reproducing. 3) To determine any major changes in habitat quality from that required by the species. 4) To compare the effectiveness of the different management zones in the reserve. 5) Based on the results of the data gathered make recommendations for management decisions on fishing quotas, length of season, size limits, and other regulations that can be modified to make the fishery both profitable and sustainable. This report presents data collected for the fishery‐independent component of the protocol from July 2004 to May 2012 during 24 sampling occasions with between one and four surveys conducted per year. 2 Glover’s Reef Marine Reserve The Glover’s Reef Atoll (16º44’N, 87º48’W) is about 32 km long and 12 km wide with an area of 35,876 ha. The atoll lies approximately 45 km east of the Belizean mainland and 25 km to the east of the Belize Barrier Reef (Figure 1). The depth ranges from 300 to 400 m to the north and west of the atoll, while the east side drops to over 1000 m. Water depth in the inner lagoon averages 6‐8 m deep with depths up to 18 m (Figure 2). There are three main channels that connect the ocean, reef and lagoon habitats, with the latter containing approximately 850 patch reefs. The entire Glover’s Reef Atoll was established as a Marine Protected Area in 1993 (Statutory Instrument 38 of 1993 under the Fisheries Act Chapter 210) and is managed by the Belize Fisheries Department. The Glover’s Reef Marine Reserve includes five management zones: General Use Zone, Conservation Zone, Wilderness Zone, Seasonal Closure Zone and Spawning Aggregation Site (Figure 3). NASA Figure 2: Bathymetry of Glover’s Reef Atoll
Glover’s Reef Atoll Figure 1: Location of Glover’s Reef Atoll, Belize. 3 Figure 3: Glover’s Reef Marine Reserve: Management Zones
METHODOLOGY The LAMP protocol was developed in 1996 to monitor the spiny lobster and queen conch fisheries in the Glover’s Reef Marine Reserve. In 2000, it was expanded to include the monitoring of five commercially important finfish species: (Nassau grouper Epinephelus striatus, Black grouper Mycteroperca bonaci, Hogfish Lachnolaimus maximus, Mutton snapper Lutjanus analis and Queen triggerfish Balistes vetula). The protocol is described in Field protocol for monitoring coral reef fisheries resources in Belize (Acosta, 2003) and conforms to the methodology described in the CARICOMP Methods Manual Levels 1 and 2 (CARICOMP, March 2001 edition). In March 2006, parrotfish species were also included as part of the monitoring program given their importance as herbivores and their increasing importance as a commercially fished species1. Sampling Sites The surveys followed a design stratified by management type, namely Conservation Zone (CZ, which also includes the Wilderness Zone) and General Use Zone (GUZ). The sampling units were randomly located according to a systematic stratified design and comprised either patch reefs or belt transects (50 m long and 2 m wide) with the latter falling in a combination of seagrass and sandflat habitats (Figure 4). The original design comprised 33 sampling units with 11 patch reef (5 in the CZ and the remainder in the GUZ) sites and the remaining 22 belt transects falling in the seagrass/sand flats habitat (9 in the CZ and 13 in the GUZ). In May 2010, the survey area was extended further into the GUZ with 17 new sampling sites ‐ 12 patch reef and 5 in the seagrass/sand flats habitat ‐ added in the GUZ bringing the total number of sampling units to 50 (Table1). The average depth of the sampling sites ranged from 1.2 m to 9.0 m. In addition, the survey effort for the belt transect was increased with each original transect replaced by a set of four 50 m by 2 m transects (treated as a single sampling unit) and complemented by half‐hour timed swims. Only conch were counted on all sampling units with lobster and finfish only recorded on the patch reef sampling units. The area of each patch reef was estimated by taking GPS points along the perimeter of the patch reef and inputting the information into a Geographic Information System (GIS) to calculate the area. The patch reefs ranged in size from 0.04 ha to 1.43 ha. Table 1: Distribution of sampling sites by habitat and management zone in the Glover’s Reef Marine Reserve Habitat Management Area Wilderness Conservation General Use Sand Algal Flats and Seagrass Beds 1 8 18 Patch Reefs 0 5 18 Total No. of Sites 1 13 36 1
In May 2009, the Government of Belize passed new legislation which now bans the harvesting of all species of parrotfish. 4 Densities for each sampling unit were obtained for queen conch and spiny lobster by dividing the counts made for the target species by the size of the patch reef or belt transect and then averaged to obtain mean densities D̂. The 95% confidence intervals were calculated as Dˆ  1.96 se ( Dˆ ), where se (Dˆ ) is the standard error of the mean density. Either overall density estimates were obtained or the estimates were stratified by some factor of interest, such as management type (CS vs. GUZ), legal size limit, age class, etc. For finfish, the biomass per hectare was instead obtained by converting the fork length estimate s recorded for each fish during the survey into biomass B̂ in grams using the formula: Bˆ  10
a b
log F
log 10
, where a and b are species specific conversion coefficients and F is the fork length in millimeters. Data collected during the half‐hour timed surveys for conch were excluded from the density or biomass calculations, but included for summaries of size classes or legal size limit. Generalized Additive Mixed Models (GAMMs) were used to examine changes in species density over time, as well as whether there was a significant difference in density by age class, management type, fishing season or habitat type. Figure 4: Glover’s Reef Atoll with the sampling sites used for the Long‐term Atoll Monitoring Program (LAMP). Sampling units include original sites (red) and sites added in May 2010 (yellow) with stars indicating patch reef habitat and squares seagrass/sand flat habitat. The Conservation Zone (CZ) covers the approximately triangular area with the Wilderness Zone corresponding to the circular area. The General Use Zone (GUZ) buffers the CZ. Figure 5: Glover’s Reef Atoll showing habitats of sampling sites used for the Long‐term Atoll Monitoring Program (LAMP). 5 Species Surveys Conch Survey The queen conch was surveyed on sand algal flats and seagrass beds (i.e. the sand flats areas) as well as near shallow patch reef habitats. In sand‐algal and seagrass habitats, density surveys were conducted along four straight line belt transects measuring 50 m long by 2 m wide. A 50 m measurement tape was laid along the substrate and conch were counted along 2 m on either side of the tape. Half‐hour timed swims were also conducted. Conch were measured for size and checked for egg‐laying activity or presence of egg masses. To measure size, shell length was measured in mm from the tip of the spire to the notch opening. Mature conch stop increasing in shell length, but the shell lip starts to thicken, therefore, the lip thickness was also measured to estimate the age of mature conch. Lobster Survey Spiny lobsters were surveyed on patch reef sites only since large juveniles and adults use patch reef habitat for shelter and feeding and smaller size classes of lobster, which use seagrass and macroalgal habitats, cannot be visually surveyed accurately by standard methods. Patch reefs were surveyed by swimming crossing patterns across the entire reef structure. Survey duration was recorded. Lobsters were measured for size by estimating the carapace length to the nearest cm with a marked tickle stick placed over the dorsal side from the posterior end of the carapace to the space between the eyes. The sex was determined by observing external dimorphic characteristics and the lobster was checked for the presence of egg masses. Finfish Survey Finfish species were surveyed on patch reef sites, as in the spiny lobster sampling protocol described above. Data for the following species were collected during the survey: Nassau grouper Epinephelus striatus; Black grouper Mycteroperca bonaci; Hogfish Lachnolaimus maximus; Mutton Snapper Lutjanus analis and Queen triggerfish Balistes vetula. The six species of parrotfish surveyed were: Stoplight Sparisoma viride, Redtail Sparisoma chrysopterum, Yellowtail Sparisoma rubripinne, Princess Scarus taeniopterus, Striped Scarus croicensis and Redband Sparisoma aurofrenatum. The fork length of the target species was estimated in cm from the tip of the snout to the fork of the tail. Environmental Variables At each site, the following variables were recorded using a Yellow Springs Instrument (YSI) meter: water temperature, conductivity, salinity and depth. Visibility was measured using a secchi disk. 6 RESULTS QUEEN CONCH – Strombus gigas For queen conch, the overall results can be seen in Figure 6a with the mean densities by sampling period by management type shown in 6b. Queen conch density is not consistently higher in the closed season for conch. For the majority of the surveys the mean queen conch density is higher in the Conservation Zone. When estimating density stratified by age class 1 there is much more variability in mean density for the small and medium juveniles between sampling occasions with large juvenile and adult densities being more similar to each other and less variable over time (Figure 7). A stratification by legal size limit (176mm) also highlights the variability in density over time in the smaller individuals that correspond to the sub‐legal sized individuals (Figure 8a). When stratifying by habitat type it is evident that there is considerable variation associated with the seagrass/sand flat habitat over time, whereas the patch reef habitat density remains fairly stable (Figures 8b & c). Bar chart summaries of the relative proportions of the different age classes (Figure 9a) shows the same variation as the density estimates; the medium and in particular small juveniles have experienced the greatest relative changes over the duration of the surveys from 2004‐2012, dropping to their lowest level during the August 2008 survey. The relative proportion of legal size (>176mm) to sub‐legal size conch over time again shows the relatively lower numbers of smaller conch during the middle of the 2004‐2012 period of the surveys (Figure 9b). GENERALIZED ADDITIVE MIXED MODELS (GAMMs) The design‐based estimates of density were also reinforced by model‐based results obtained by means of Generalized Additive Mixed Models (GAMMs; Zuur et al. 2009). GAMMs were used to examine changes in queen conch density over time, as well as whether there was a significant difference in density by age class, management type, fishing season or habitat type. The dependent variable was the number of queen conch observed per transect or patch reef. By including the surface area sampled as an offset term in the models, density was in effect being modeled. A negative binomial error distribution was assumed. Analyses were completed using the R software (RDCT 2013). The models were fitted using the gamm function from the mgcv library (Wood 2006) The trend in queen conch density over time differed significantly by habitat type, management type, and age class (Appendix 1), whereas fishing season (open vs. closed) played a more minor role in explaining the variability in conch density. The modeling results indicated a fairly steady increase in density associated with the patch reef habitat over the 2004‐2012 period, whereas for the seagrass/sand flat habitat there was an initial decline until the August 2008 survey with an increase in density after that point in time. With many of the seagrass/sand flat sampling units located in the GUZ, this same pattern was also reflected for this management type. There were even stronger trends in density in the CZ with density peaks associated with the May 2007 and April 2011 survey occasions, and troughs associated with the March 2006 and May 2008 survey occasions. Small Juvenile: < 150mm; Medium Juvenile: 150‐200mm; Large Juvenile > 200mm; Adult: thick lip or eggs 7 1
There was no significant trend in density for adult queen conch, but large juveniles showed an increase over time. For medium juveniles and in particular small juveniles there were statistically significant changes during the 2004‐2012 period with a decline until the August 2008 survey and an increase in density after that point in time. Top ranking models by Akaike’s Information Criterion (AIC) value included an interaction between age class and management type and age class and habitat type. Overall densities were lower in the General Use Zone (p = 0.007) with medium and small juvenile density significantly higher in the GUZ (P < 0.0001). Overall densities were higher in the seagrass habitat (p < 0.0001) with a significantly higher density for medium and small juveniles (p < 0.0001). Queen conch density was significantly higher during the closed season (p = 0.044). Overall large juvenile density was not significantly different to adult densities, but medium juvenile densities were significantly lower at the 10% level, and small juveniles densities were very much lower (with p < 0.001). a. b. Figure 6: (a) The estimated overall queen conch density including all age classes with the 95% confidence limits indicated by the dashed blue lines. Whether the survey was conducted during the open or closed season for conch is also shown. (b) Estimated density stratified by management type with the 95% confidence limits indicated by the dashed light blue lines. 8 a. b. Figure 7: (a) Estimated density stratified by age class 2 with the 95% confidence limits indicated by the dashed light blue lines shows that the large juvenile and adult age classes are relatively stable over time with most of the variability in density occurring in the small juvenile age class. (b) Adjusting the y‐axis scale to show more detail, omitting the small juveniles for clarity. 2
Small Juvenile: < 150mm; Medium Juvenile: 150‐200mm; Large Juvenile > 200mm; Adult: thick lip or eggs 9 a. b. c. Figure 8: Estimated density stratified by (a) legal size limit (176mm), and (b) habitat type with (c) the adjusted y‐axis scale showing the variability at a smaller scale. The 95% confidence limits indicated by the dashed light blue lines. 10 a. b. Figure 9: (a) Percent queen conch by age class over time. The age class categories are defined as Small juvenile: < 150mm; Medium Juvenile: 150‐200mm; Large Juvenile > 200mm; Adult thick lip or eggs. (b) Percent of legal and sub‐legal queen conch over time. The sample sizes per sampling period are shown above the bars. SPINY LOBSTER – Panulirus argus 11 For spiny lobster, the overall results can be seen in Figure 10a with the mean densities by sampling period by management type shown in 10b. Although there seems to be an overall increase in mean density over time, density is not consistently higher in the closed season nor for any one of the management types for spiny lobster. A stratification by legal size limit (76.2 mm) also highlights the variability in density over time in the smaller individuals that correspond to the sub‐legal sized individuals (Figure 10c). a. b. Figure 10: (a) The estimated overall density for spiny lobster with the 95% confidence limits indicated by the dashed blue lines. Whether the survey was conducted during the open or closed season for lobster is also shown. Estimated density stratified by (b) management type with the 95% confidence limits indicated by the dashed light blue lines. 12 c. Figure 10: (c) Estimated density stratified by legal size limit (76.2 mm), with the 95% confidence limits indicated by the dashed light blue lines. Bar chart summaries of the relative proportions of male vs. female spiny lobster (Figure 11a) show some variation over time with the proportions being approximately equal on average. The relative proportion of legal size (>76.2mm) to sub‐legal size spiny lobster over time (equivalent to adult vs. juveniles) shows the relatively lower numbers of smaller individuals (Figure 11b). These summaries exclude individuals without a sex classification or for which length measurements were not obtained. a. Figure 11: (a) Percent spiny lobster by sex class over time. The sample sizes per sampling period are shown above the bars.
13 b. Figure 11: (b) Percent of legal and sub‐legal spiny lobster over time. The sample sizes per sampling period are shown above the bars. The GAMM analysis indicated a statistically significant upward trend in spiny lobster density over time (Figure 12) with other factors, such as management type or fishing season, not improving models in terms of AIC ranking. Figure 12: Estimated conditional dependence of spiny lobster density over time. Shown are estimates on the scale of the linear predictor (solid lines) and confidence intervals (dashed lines). 14 FINFISH For finfish, the surveys focused on a subset of key species, namely black grouper, hogfish, mutton snapper, Nassau grouper, queen trigger fish, princess parrotfish, redband parrotfish, redtail parrotfish, stoplight parrotfish, striped parrotfish, and yellowtail parrotfish. The overall results can be seen in Figure 13a with the mean biomass in grams per hectare for each sampling period by management type shown in 13b. For the majority of the surveys the mean biomass is higher in the Conservation Zone and the biomass for individual species stratified by management type can be seen in Figures 14a‐j. a. b. Figure 13: The estimated biomass (a) overall and by (b) management type for key finfish species with the 95% confidence limits indicated by the dashed blue lines. 15 a. b. c. Figure 14: The estimated biomass by management type for the key finfish species (a) black grouper, (b) hogfish and (c) mutton snapper with the 95% confidence limits indicated by the dashed blue lines. 16 d. e. f. Figure 14: The estimated biomass by management type for the key finfish species (d) Nassau grouper, (e) princess parrotfish and (f) redband parrotfish with the 95% confidence limits indicated by the dashed blue lines. 17 g. h. i. Figure 14: The estimated biomass by management type for the key finfish species (g) redtail parrotfish, (h) stoplight parrotfish and (i) striped parrotfish with the 95% confidence limits indicated by the dashed blue lines. 18 j. Figure 14: The estimated biomass by management type for (j) yellowtail parrotfish with the 95% confidence limits indicated by the dashed blue lines. Note that the parrotfish species were only recorded during surveys conducted from the sixth sampling occasion onwards. Bar chart summaries of the relative proportions of fish species by length class based on five equal interval fork length categories show the variation over time (Figure 15a‐e). These summaries exclude individuals without a fork length measurement. a. Figure 15: Percent by fork length class over time for the key finfish species (a) black grouper. The sample sizes per sampling period are shown above the bars and the length classes are species specific equal interval classes based on the maximum fork length recorded. 19 b. c. d. Figure 15: Percent by fork length class over time for (b) hogfish, (c) mutton snapper and (d) Nassau grouper. Sample sizes per sampling period are shown above the bars and the length classes are species specific equal interval classes. 20 e. Figure 15: Percent by fork length class over time for combined parrotfish species (princess, redband, redtail, stoplight, striped, yellowtail). The sample sizes per sampling period are shown above the bars and the length classes are species specific equal interval classes based on the maximum fork length recorded. The GAMM analysis for all finfish species combined indicated a somewhat different trend in biomass by management type over time (Appendix 2). In the CZ there was an initial decline followed by an increase and then a decline for the last few sampling occasions. For the GUZ, the analysis indicated an initial increase with two peaks in biomass. Overall, for all finfish, the biomass was higher in the CZ (P < 0.097), which was also the case for black grouper (P < 0.07), and hogfish (P < 0.094). For mutton snapper, biomass was lower in the GUZ, but the difference was not statistically significant. For black grouper there was some indication of a decline in biomass in both management zones over time, but the trend was not significant, whereas for hogfish and mutton snapper there was a statistically significant decline in biomass in both management zones. For Nassau grouper the biomass remained stable in the CZ over time, but there was a significant decline in biomass in the GUZ. For each of the parrotfish species the biomass was consistently higher in the CZ compared to the GUZ, although this difference was only statistically significant for stoplight parrot fish. There was a statistically significant change in biomass over time by management type for the parrotfish species with the exception of stoplight parrotfish where the increasing trend was not statistically significant in either management type, and yellowtail parrotfish where there was an increasing biomass trend in both management types, but only in the CZ was it statistically significant. For princess parrotfish, redband parrotfish, redtail parrotfish, and striped parrotfish there was in general an initial decrease in biomass with a turning point some time during the 11‐13 sampling occasion (end of 2007 / beginning of 2008) reaching a peak around the 18‐19 sampling period in 2009 followed by another period of decline in biomass. 21 SUMMARY AND CONCLUSIONS Mating behavior and egg‐laying in queen conch are both directly related to the density of mature adults (Stoner and Ray‐Culp, 2000). The minimum density where mating can occur is ~50 mature adults ha‐1 however, ~100 mature adults ha‐1 is required for high mating frequency (McField, M. and Kramer, P., 2007; Stoner et al., 2012). The adult and large juvenile mean densities recorded during the last sampling period for the LAMP at Glover’s Reef atoll are well below this minimum density with 19.0 and 24.9 individuals per ha, respectively. These densities represent both sand algal flats and seagrass beds and patch reef habitats. The mean densities of smaller sized individuals were 176.6 medium juveniles per ha and 403.54 small juveniles per ha. Overall, the trend in queen conch density over time differed significantly by habitat type, management type and age class. During the May 2012 sampling period, mean densities of 1116.7 conch per ha were recorded in the sand algal flats and seagrass beds and 45.8 conch per ha on patch reefs. There was a steady increase in density associated with the patch reef habitat over all sampling periods, whereas for the seagrass/sand flat habitat there was an initial decline until the August 2008 survey with an increase in density after that point in time. Overall densities were lower in the General Use Zone, however, the medium and small juvenile density was significantly higher in the GUZ. There was a statistically significant increasing trend in spiny lobster density over time. The overall mean density during the May 2012 sampling period was the highest recorded with 39.3 individuals per hectare. This overall increase in mean density, however, was not consistently higher for either the Conservation or General Use Zone. Acosta and Robertson (2003) in their study of spiny lobster density on coral patch reef, forereef and reef wall at Glover’s Reef Atoll showed that in 2001, the mean densities of lobsters on coral patch reefs were significantly higher in the Conservation Zone (146.7 individuals per hectare) compared to the fished areas. The Acosta study also showed that the lobsters occupying the deeper forereef and wall reef were mostly large adults (range 70‐
180 mm Cl). The LAMP study has showed that over time the relative proportion of legal size (>76.2mm) to sub‐
legal size spiny lobster is higher. One caveat of the LAMP surveys is that the forereef or wall reef areas are not sampled, so the number of adult lobsters in these habitats is not being monitored. In addition, due to the cryptic nature of the species it is possible that the LAMP densities are underestimated. Overall, for all finfish, the mean biomass was higher in the Conservation Zone than the General Use Zone. Over time, there was an initial decline in biomass in the CZ, followed by an increase and then a decline for the last few sampling occasions. For hogfish and mutton snapper there was a statistically significant decline in biomass in both management zones over time. For Nassau grouper the biomass remained stable in the CZ over time, but there was a significant decline in biomass in the GUZ. For each of the parrotfish species the biomass was consistently higher in the CZ compared to the GUZ, although this difference was only statistically significant for stoplight parrot fish. For the parrotfishes: princess, redband, redtail and striped there was an initial decrease in biomass, an increase beginning of 2008 with a peak in 2009 and then a period of decline in biomass. Overall, the finfish biomasses recorded are low even for the parrotfish which are legally protected since 2009. Again, without any sampling sites on the forereef areas, the coral patch reef sites in the lagoon may be an underestimate of finfish biomass on the atoll. 22 LITERATURE CITED Acosta, C.A. 2003. Field protocol for monitoring coral reef fisheries resources in Belize. Wildlife Conservation Society, Belize. Acosta, C.A. and Robertson, D.N. 2003. Comparative spatial ecology of fished spiny lobsters Panulirus argus and an unfished congener P. guttatus in an isolated marine reserve at Glover’s Reef atoll, Belize. Coral reef 22: 1‐9. Caribbean Coastal Marine Productivity Program (CARICOMP). 2001. A Cooperative Research and Monitoring Network of Marine Laboratories, Parks, and Reserves, CARICOMP Methods Manual Levels 1 and 2, CARICOMP Data Management Centre, Jamaica and the Florida Institute of Oceanography, U.S.A. McField, M. and Kramer, P. 2007. Health Reefs for Healthy People: A Guide to Indicators of Reef Health and Social Well Being in the Mesoamerican Reef Region. With contributions by M. Gorrez and M. McPherson. 208 pp. R Development Core Team. 2013. R: A language and environment for statistical computing. X64 3.0.1 ed. Vienna, Austria: R Foundation for Statistical Computing, ISBN 3‐900051‐07‐0, URL http://www.R‐
project.org. Accessed 2012 October 15. Stoner, A.W. and Ray‐Culp, M. 2000. Evidence for Allee effects in an over‐harvested marine gastropod: density‐
dependent mating and egg production. Marine Ecology Press Series Vol. 202: 297‐302. Stoner, A.W., Davis, M. and Booker, C. 2012. Abundance and population structure of queen conch inside and outside a marine protected area: repeat surveys show significant declines. Marine Ecology Press Series Vol. 460: 101‐114. Wood, S.N. 2006. Generalized Additive Models: An Introduction with R. Chapman and Hall, Boca Raton, Florida,.USA. 391 pp. Zuur A.F., E.N. Ieno, N. Walker, A.A. Saveliev, and G.M. Smith. 2009. Mixed Effects Models and Extensions in Ecology with R. Springer, New York, USA. 574 pp. 23 APPENDIX 1: GENERALIZED ADDITIVE MIXED MODELS (GAMMs) RESULTS FOR CONCH (Strombus gigas) Estimated conditional dependence of queen conch density (a) habitat types patch reef (left) and seagrass/sand flat (right), (b) management types the Conservation Zone (left) and General Use Zone (right), (c) by age class for adults (top left), large juveniles (top right), medium juveniles (bottom left) and small juveniles (bottom right). Shown are estimates on the scale of the linear predictor (solid lines) and confidence intervals (dashed lines). a. b. 24 c. 25 APPENDIX 2: GENERALIZED ADDITIVE MIXED MODELS (GAMMs) RESULTS FOR FINFISH Estimated conditional dependence by management zone (CZ on the left and GUZ on the right) for the key finfish species (a) overall, (b) black grouper, (c) hogfish, (d) mutton snapper, (e) Nassau grouper, (f) princess parrotfish, (g) redband parrotfish, (h) redtail parrotfish, (i) stoplight parrotfish, (j) striped parrotfish, and (k) yellowtail parrotfish. Shown are estimates on the scale of the linear predictor (solid lines) and confidence intervals (dashed lines). a. b. 26 c. d. e. 27 f. g. h. 28 i. j. k. 29