- Email: [email protected]
Fish as indicators of diving and fishing pressure on high-latitude coral reefs
Ocean & Coastal Management 84 (2013) 130e139
Contents lists available at ScienceDirect
Ocean & Coastal Management journal homepage: www.elsevier.com/locate/ocecoaman
Fish as indicators of diving and fishing pressure on high-latitude coral reefs C. Floros*, M.H. Schleyer, J.Q. Maggs Oceanographic Research Institute, P.O. Box 10712, Marine Parade, 4056 Durban, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history: Available online
Despite the proclamation of South Africa’s coral reef marine protected areas (MPAs) more than 20 years ago, the effects of human activities on the fish communities have not been investigated. This study used a multi-species Fish-index to compare ecological indicators such as biomass, abundance, trophic structure and reproductive potential between multiple-use and no-take sanctuary zones. Seven study reefs were surveyed; six in South Africa and a non-MPA reef in southern Mozambique. Randomly stratified underwater visual censuses (UVC) using the point count technique were used to survey fish communities. Environmental variables and habitat characteristics were also recorded. Nonmetric multidimensional scaling ordinations were similar for abundance and biomass trends and revealed a high degree of overlap between all zones, except for the no-take Sanctuaries. The latter formed discrete clusters and were significantly different (Analysis of Similarity) to the other zones. Total abundance and biomass were highest in the Sanctuary zones and lowest in the Open zone. Differences in trophic composition between zones were largely due to predatory species. This was supported by similarity percentages analysis (SIMPER) which identified six discriminating species. Length-frequency analysis of these species revealed consistent trends with higher numbers of large individuals in the Sanctuary zones and reduced numbers of small individuals in zones open to human activity. These results along with those of the Generalised Linear Models (GLM) demonstrate that human activities are affecting the southern African coral reef fish communities. Marginal differences between the multiple-use MPA zones on the South African reefs and the non-MPA reef in southern Mozambique suggest that MPA management objectives require reevaluation. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Coral reefs provide worldwide benefits and services to the value of US$ 29.8 billion every year (Cesar et al., 2003). The biggest proportion is generated by tourism (US$ 9.6 billion) and fisheries (US$ 5.7 billion). Yet, these activities have increasingly caused degradation on many reefs throughout the world (Halpern et al., 2008; Wilkinson, 2008). Various measures to manage and mitigate coral reef damage have been proposed and, among those, marine protected areas (MPAs) have been proposed as an ideal management solution because they were perceived to simultaneously address issues of overfishing, habitat degradation, and tourism development (Kelleher and Kenchington, 1992; IUCN, 2004). MPAs are specifically intended to limit human activities in designated locations (Sale et al., 2005; Mora et al., 2006) and the
* Corresponding author. Tel.: þ27 31 3288229. E-mail address: cfl[email protected] (C. Floros). 0964-5691/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ocecoaman.2013.08.005
degree to which human activities are limited determines the type of MPA. In most instances, MPAs can be classified into two broad types; areas that are open to resource use and areas closed to resource use. No-take MPAs are areas closed to exploitation (here termed sanctuaries). The second type of MPA allows harvesting of resources, but under protective regulations that pertain to each species being harvested. In addition, the types of fishing or harvesting gear may be restricted. Such MPAs are multiple-use zones and most often permit recreational activities such as SCUBA diving, snorkeling, whale watching and fishing. Nevertheless, MPAs are not ‘cure alls’ (Alder, 1996; Agardy et al., 2003; Mora et al., 2006) and many face difficulty in implementation and enforcement due to poor governance, and lack of management guidance and evaluation (White et al., 2006). Among the major challenges restricting effective MPA management is a lack of scientific information about the status and nature of activities operating therein (Kelleher et al., 1995; Pomeroy et al., 2005; Wells et al., 2007). Assessment of MPA effectiveness is a matter of great urgency and importance given the multitude of stressors threatening the
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139
future of coral reefs (Bellwood et al., 2004; Knowlton and Jackson, 2008). Furthermore, many MPAs are promulgated as multiple-use zones and complex decision-making processes are often required to balance conservation and socio-economic objectives. Thus, it is imperative to assess the impacts of human activities on the biological components of coral reefs and whether these impacts are consistent with the management objectives of the MPA. Obtaining such information requires regular monitoring and assessment of ecological integrity within MPA boundaries. Ecological integrity refers to system wholeness, including the presence of appropriate species, populations, and communities and the occurrence of ecological processes at appropriate rates and scales (Karr, 1981; Angermeier and Karr, 1994). Evaluating changes in ecological integrity is often a major obstacle facing MPA managers due to a scarcity of technical skills, experience and funding (Pomeroy et al., 2005). Indicator-based monitoring programmes may provide an effective solution to these challenges because
131
ecological indicators may represent inexpensive means of gathering scientific data that does not require skilled personnel. For this reason, they have become important tools in coral reefs research, reporting and management (see Jameson and Kelty, 2004 for review). The challenge is determining which biological criteria are effective measures of ecological integrity but are also simple enough to monitor (Dale and Beyeler, 2001). Fish are conspicuous biological components of coral reefs that have good indicator potential owing to their importance as a valuable protein source (Pauly et al., 2002; Bell et al., 2009) and functional ecological roles on reefs (Bellwood et al., 2004). Accordingly, this study selected a multi-species index (Fish-index) to assess the effects of human activities on high-latitude coral reefs in South Africa. The aim was to use comparisons of ecological parameters such as biomass, abundance, trophic structure and reproductive potential as measures of ecological integrity. Changes observed in functional
Fig. 1. Studied reefs on the north east coast of South Africa and southern Mozambique. Sites included three Sanctuary zones (Leadsman Shoal, Red Sands Reef, and Rabbit Rock), three multiple-use zones (Two-mile, Seven-mile and Nine-mile reef) and a non-MPA zone (Shallow Malongane).
132
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139
Table 1 Zonation of the reefs based on SCUBA diving and fishing intensities. SCUBA diving statistics are averages for the period 2007e2008 (Pieters, 2009). Zone
Open Diving-Fishing High-Diving Sanctuary
Activity
Reef name
SCUBA diving intensity
Fishing intensity
Low (w4500 dives/year) Low (w2100 dives/year) High (w54,000 dives/year) Nil
High (unregulated angling and spearfishing) Restricted (angling and spearfishing; gamefish onlya). Nil Nil
Shallow Malongane Seven-mile Reef, Nine-mile Reef Two-mile Reef Leadsman Shoal, Red Sands, Rabbit Rock
a Pelagic bony fish of the families Scombridae, Carangidae, Pomatomidae, Coryphaenidae, Rachycentridae, Xiphiidae, Ostiophoridae and Sphyraenidae, the species Aprion virescens, as well as pelagic cartilaginous fish of the families Carcharinidae, Isuridae, Sphyrnidae, Alopiidae and Odontaspididae (Marine Living Resources Act, Section 3.1 (G) Regulation R1429).
processes such as growth, reproduction and trophic functioning may provide MPA managers with insight into current conditions and aid in predicting future trends. South Africa’s coral reefs are located within the boundaries of two longstanding, contiguous MPAs. Two types of conservation strategies are recognized in the MPAs: no-take sanctuary zones and multiple-use zones. Recreational fishing and SCUBA diving are the most common activities in the MPAs (Schleyer, 2000). Fishing is known to have a direct effect on fish communities via the harvest of target and bait species, and the removal of functional groups, and has the potential to cause significant changes in the structure of reef fish communities (Cooke and Cowx, 2004; Dulvy et al., 2004). SCUBA diving, has been shown to affect coral communities (Tratalos and Austin, 2001; Zakai and Chadwick-Furman, 2002; Barker and Roberts, 2004; Hawkins et al., 2005), however, the paucity of empirical studies investigating the effects of high diving intensity on fish communities in the literature represents a ‘knowledge-gap’. It is thus both timely and pertinent that the effects of human activities on the South African coral reef fish assemblages are assessed for efficient MPA management.
respectively. No changes have been made to the MPA zonation since their proclamation. For this study, the reefs were zoned according to the intensity and type of human activity (Table 1). These were Sanctuary, High-Diving, Diving-Fishing and Open zones. Sanctuary zones theoretically prohibited all forms of human activity. The High-Diving zone permitted SCUBA diving only. The Diving-Fishing zone permitted SCUBA diving and restricted fishing (gamefish only). Lastly, the Open zone had no active law enforcement at the time of data collection which meant that activities on the reef were unregulated. No data for fishing intensity were available for any of the zones.
2. Materials and methods
2.4. Surveys of reef fish communities
2.1. Study area
Surveys of fish communities were conducted from August 2007 to February 2009 to include four summer and four winter sampling periods. The point count underwater visual census technique, adapted from Samoilys and Carlos (2000), was used to estimate the abundance and biomass of the Fish-index species. A total of 60 point counts were conducted per reef. Each census (point count) consisted of a 5 min count within a circle 10 m in diameter. All Fish-index species observed on the reef and within the water column during each point count were recorded and their sizes were estimated to the nearest 10 cm. Estimates of fish length were used to generate biomass using known length-weight regression coefficients from Fishbase (Froese and Pauly, 2011). Only two divers entered the water at any one time, a surveyor and a buddy diver. The same diver conducted all point counts to minimize variation and error incurred by diver bias. Point counts within each reef were separated by at least 50 m. The time of each point count was conducted between 0800 and 1400. All counts were conducted at a depth range of 12e15 m.
The South African coral reefs are located along the north-east coast of South Africa. Six study reefs were selected to represent different types of MPA zones (Fig. 1). The study area was extended to include non-MPA reefs in southern Mozambique for comparative purposes. However, only one reef, located at Ponta Malongane, could be included as a study site due to political and logistical constraints. The southern Mozambican and South African coral reefs are the southernmost reefs in the Western Indian Ocean (Riegl et al., 1995). Due to their close proximity they share numerous similarities. They are exposed to similar oceanographic conditions, which are dominated by the southwards flowing Agulhas Current (Lutjeharms, 2006). They are confined to the narrow continental shelf (Ramsay, 1996) and can be classified as patch reefs (Ramsay and Mason, 1990). They are atypical coral reefs in that the coral communities are non-accretive and form a veneer on latePleistocene beach rock (Ramsay, 1996). Their benthic communities are similar with corals contributing 50e70% to the total living cover (Jordan and Samways, 2001; Pereira, 2003; Celliers and Schleyer, 2008). The fish assemblages in both regions are very similar, with approximately 300 reef-associated species; 80% of which are Indo-Pacific (Pereira, 2003; Celliers and Schleyer, 2008; Floros et al., 2012). 2.2. Reef protection status and human activity The South African coral reefs are situated within the Maputaland MPA and St Lucia MPA, proclaimed in 1986 and 1979,
2.3. Fish-index selection Species were included as indicators based on literature reviews and field observations. The criteria included vulnerability to exploitation by fishing or hobbyists, trophic status and ease of identification. The Fish-index species are listed in Appendix C along with the justification for their selection.
2.5. Trophic levels Species were allocated to one of eight trophic guilds; top-level predators, medium-level predators, planktivores, omnivores, herbivores, benthivores, corallivores and invertivores (Appendix C). Trophic allocation was based on diet information retrieved from the literature and supplemented by field observations (see Floros et al., 2012 for details).
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139
a
2D Stress: 0.26
Human activity High-Diving Diving-Fishing Sanctuary Open
133
b
2D Stress: 0.25
Human activity High-Diving Diving-Fishing Sanctuary Open
Fig. 2. Non-metric multidimensional scaling ordination of samples based on a) fourth-root transformed abundance data and b) log(xþ1) transformed biomass data.
2.6. Habitat characteristics and environmental variables On completion of each point count, aspects of the reef habitat were described within the point count area; topography and coral cover. Coral cover was estimated using a rapid visual technique adapted from English et al. (1994), where estimates where divided into three categories (low, medium and high) according to the percentage of coral (hard and soft). The topographic complexity of the substratum was visually estimated using a six point scale system adapted from Polunin and Roberts (1993). The adaptations were specific to southern African coral reefs and included the following categories: 1 e flat reef; 2 e low undulating spur and groove; 3 e medium slopes or ridges with no grooves or gullies; 4 e medium spur and groove or pinnacles; 5 e high slopes or ridges; and 6 e high spur and groove, overhangs or pinnacles. Depth was measured using a dive computer and sea temperatures were obtained from an underwater temperature probe stationed at 18 m. 2.7. Statistical analysis Univariate analyses were carried out on abundance and biomass data using One-Way analysis of variance (ANOVA). Data were not normally distributed and thus non-parametric KruskaleWallas One-Way ANOVA on ranks was used. Dunn’s pairwise multiple comparisons procedure was used to detect differences between fish abundance and biomass parameters. Univariate analyses were conducted using the statistical package Sigma Plot 11.0. The criterion for significance for all tests was p 0.05. Multivariate analyses were undertaken using PRIMER v.6 to further explore difference between zones. Abundance data were fourth-root transformed, while the log(xþ1) transformation was used for biomass data due to large differences in the values for toplevel predators. Non-metric multidimensional scaling (MDS) was used to examine differences in spatial distribution of Fish-index assemblages across the different zones. Analysis of Similarity (ANOSIM) was also used to confirm or refute trends observed in the MDS ordinations. R-statistics >0.45 were considered to signify significant differences with a limited degree of overlap between fish community structure and values <0.45 indicated large degrees of overlap between community structures (Clarke and Gorley, 2006). SIMPER analysis was used to identify those Fish-index species responsible for the BrayeCurtis dissimilarity between MPA zones. Length-frequency graphs were generated for the six discriminating species identified by SIMPER. Size at sexual maturity for each of the six species was determined from the literature and
used as an approximation for reproductive potential. Generalised Linear Models (GLM) using R (R Development Core Team, 2011) were used to examine the influence of environmental variables (temperature, depth), habitat characteristics (coral cover and topography) and activity on fish community parameters (abundance and biomass). Activity was classified according to the intensity of diving and fishing in the different MPA zones (Table 1). The GLM for both abundance and biomass data followed the form:
ðcountsÞ ¼ b0 þ b1 ðactivityÞ þ b2 ðtemperatureÞ þ b3 ðcoral coverÞ þ b4 ðtopographyÞ þ b5 ðdepthÞ þε Overdispersion in the abundance data invalidated the use of the Poisson distribution and the negative binomial distribution with log-link function was used (Maunder and Punt, 2004). Biomass data were continuous and were thus modelled using gamma distribution with inverse-link function.
3. Results 3.1. Spatial distribution of species The spatial orientation of the abundance and biomass data suggested a high degree of overlap between the High-Diving, Diving-Fishing and Open zones (Fig. 2). In contrast, the Sanctuary zones formed distinct clusters in both data sets. ANOSIM tests confirmed the trends observed in both MDS plots (Tables 2 and 3). In terms of abundance, the Fish-index community differed significantly between Sanctuary zones and the Open zone, and between Sanctuary and Diving-Fishing zones (Global R ¼ 0.423; p < 0.001).
Table 2 Results of ANOSIM run on fourth-root transformed species abundance data for differences between zones. Global R ¼ 0.423. Significance of Global R < 0.001. Significant differences are in bold. Pairwise tests
R statistic
Significance level %
High-Diving, Fishing-Diving High-Diving, Sanctuary High-Diving, Open Diving-Fishing, Sanctuary Fishing-Diving, Open Sanctuary, Open
0.302 0.429 0.087 0.461 0.235 0.6
0.006 0.005 9.2 <0.0001 0.6 0.0005
134
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139
Table 3 Results of ANOSIM run on log(xþ1) transformed species biomass data for differences between zones. Global R ¼ 0.423. Significance of Global R < 0.001. Significant differences are in bold. Pairwise tests
R Statistic
Significance level %
High-Diving, Diving-Fishing Sanctuary, High-Diving High-Diving, Open Sanctuary, Diving-Fishing Diving-Fishing, Open Sanctuary, Open
0.202 0.516 0.162 0.468 0.236 0.713
0.2 0.0005 1.7 <0.0001 0.4 0.0005
The biomass data revealed that the Fish-index structure in Sanctuary zones was significantly different to the other three zones (Global R ¼ 0.423; p < 0.0001). 3.2. Abundance, biomass and trophic guild composition The total mean abundance and total mean biomass manifested similar trends across zones with the highest values recorded in Sanctuary zones and the lowest in the Open zone (Fig. 3). Differences in mean abundances were significant between Sanctuary (18.01 fish 78 m2) and High-Diving (13.86 fish 78 m2), and Sanctuary and Open zones (9.81 fish 78 m2) (p < 0.001). In addition, mean abundances between Diving-Fishing (15.93 kg 78 m2) and Open reefs were also significantly different. Total biomass on Sanctuary reefs (23.71 kg 78 m2) was significantly different and at least three times greater than in all other zones (p < 0.001). Invertivores were the most abundant trophic guild in all MPA zones and did not differ significantly between zones. Planktivores and herbivores were the next most abundant guilds. Abundances for herbivores were significantly different between High-Diving and the Open zone (P < 0.01). Top- and medium-level predators were significantly more abundant in Sanctuary zones (P < 0.001). Differences in biomass between MPA zones were most significant for predator guilds. In the Sanctuary zones, the total biomass of toplevel predators (8.32 kg 78 m2) was seven times great than that recorded in the High-Diving zones and four times greater than in the Diving-Fishing zone. No top-level predators were recorded in the Open zone. Medium-level predators in the Sanctuary zones (10.6 kg 78 m2) were at least six times greater than in the other three zones. 3.3. Discriminating species Comparisons between Sanctuary zones and the other zones revealed that six species were the top contributors to the dissimilarity in each pairwise comparison (Table 4). These species were Epinephelus tukula, Lutjanus bohar, Aprion virescens, Caranx melampygus, Variola louti and Oplegnathus robinsoni and their cumulative contribution to the dissimilarity between each comparative group was almost 50%. Excluding O. robinsoni, all the above species are important reef predators. All these species are potentially targeted in the Open zone; however, only C. melampygus and A. virescens may be targeted on the South African coral reefs. 3.4. Length-frequency analysis Length-frequency graphs were generated for the six discriminating species identified by SIMPER analysis. All six species were recorded to have greater abundance and mean body size in the Sanctuary zones (Fig. 4; Appendix A & B). There was a clear trend for reduced abundances and mean size in zones open to fishing (Fishing-Diving and Open) such that only low numbers of sexually
Fig. 3. Total mean abundance (a), biomass (b) and trophic structure of the Fish-index species in the four zones on the southern African study reefs.
mature individuals were recorded in these two zones. The Open zone had the lowest number and smallest mean body size of the six species, and two species, A. virescens (Fig. 4a) and E. tukula (Fig. 4e), were absent from this zone. In the High-Diving zone, no sexually mature individuals of A. virescens, Oplegnathus robinsoni and E. tukula were recorded, while sexually mature individuals of the three other species were present in low numbers. 3.5. Environmental variables, habitat characteristics and human activity The recorded environmental variables and habitat characteristics appeared to explain little of the deviance observed in the fish abundance and biomass between MPA zones. All variables were
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139 Table 4 Results of SIMPER analysis. Only species providing the highest percent contribution towards the average dissimilarity in abundance and biomass data between zones have been included. Species highlighted in bold are those contributing >40% to the overall dissimilarity.
Epinephelus tukula Lutjanus bohar Aprion virescens Caranx melampygus Variola louti Oplegnathus robinsoni Odonus niger Pomacanthus imperator Acanthurus leucosternon Balistoides conspicillum Chaetodon meyeri Scarus rubroviolaceus Bodianus diana Siganus sutor Forcipiger flavissimus Chaetodon trifascialis Amphiprion allardi Average dissimilarity
High-Diving vs sanctuary
Diving-Fishing vs sanctuary
Open vs sanctuary
10.32 9.84 8.37 6.97 7.19 6.31 5.88 5.44 4.34 3.67 3.65 3.26 3.74 2.57 3.44 2.12 0 45.01
9.18 8.65 9.42 6.9 6.96 6.63 7.1 5.45 4.31 4.71 4.61 3.12 3.99 2.77 3.25 0 3.28 44.11
9.85 9.58 9.51 7.73 7.18 6.68 6.55 4.94 4.06 3.86 3.77 3.63 3.35 3.34 3.25 2.47 2.36 52.49
initially modelled in the GLMs; however, only activity, topography and coral cover were found to be significant and were thus included in the final models (Tables 5 and 6). Activity was most accountable for the explained deviance in both the abundance (8.7%) and biomass (18.8%) between zones. Topography explained 0.9% of the deviance in abundance, while coral cover explained 2% of the deviance in biomass between zones. 4. Discussion This study used a pre-determined list of species to investigate differences in fish community parameters between zones of varying protection levels. Most of the significant differences in abundance, biomass, trophic levels and reproductive potential were recorded for predatory and target reef fish species. For all parameters, values were highest in the Sanctuary zones and lowest in the Open zone. These results are consistent with other studies comparing reef fish communities across gradients of protection in the Western Indian Ocean (McClanahan & Graham et al., 2005; McClanahan et al., 2007; Floros et al., 2012) and Indo-Pacific (Graham et al., 2003; Westera et al., 2003; Evans and Russ, 2004; Lester and Halpern. 2008) and strongly suggests that the differences are a result of human activities. These findings were confirmed by the GLMs which demonstrated that the deviance between MPA zones was explained more by human activity than by habitat characteristics or environmental variables. 4.1. Fishing Efforts have been made to regulate fishing on the South African coral reefs by restricting the extraction of species to gamefish. In addition to the species restrictions, there are daily fishing quotas for each target species (C. melampygus ¼ 5 day1, A. virescens ¼ 10 day1), but no minimum size limits. Despite the daily quotas, the results suggest that fishing is having an effect on populations of the aforementioned target fish species. Low densities and small mean body sizes of target species are detectable effects of overfishing on coral reefs (Russ and Alcala, 1989; Westera et al., 2003). In the Diving-Fishing zones, the reduced mean size of target species corresponded closely to theoretical predictions (see
135
Dulvy et al., 2004; Graham et al., 2005), with all target species showing significantly smaller mean sizes. A similar, but more severe trend was evident in the Open zone where target species such as L. bohar and C. melampygus were represented by very low abundances of small individuals. In addition A. virescens and E. tukula were not recorded. At the time of data collection, there were no fishing restrictions for C. melampygus, A. virescens, and L. bohar, while E. tukula was fully protected from recreational fishing only (Decree 51/99 of 31 August). The absence of the latter species from the study records may due to the inclusion of only one reef in southern Mozambique, due to intensive fishing pressure (recreational, subsistence, or semi-commercial) or a combination of both factors. However, similar sampling effort on other reefs recorded high abundances of E. tukula and the results are most likely indicative of high levels of exploitation. A reduction in the mean size of targeted species in fished areas is termed size selective fishing or ‘age truncation’ and is due to fishers targeting larger individuals (Berkeley et al., 2004). Age truncation can have an important effect on fish assemblage structure and function, potentially affecting the productivity and resilience of fish populations (Baskett et al., 2005). The low abundance and absence of sexually mature individuals of A. virescens, C. melampygus, O. robinsoni, E. tukula and L. Bohar recorded in the non-Sanctuary zones suggests reduced reproductive potential and could have significant ramifications for future generations. These populations may be reliant on juveniles from surrounding no-take zones to replenish stocks. Sanctuary or no-take zones have been advocated as areas of high reproductive output because there are greater densities of larger, sexually mature fish present. It has been further suggested that the increased reproductive output, whether in the form of eggs, larvae or juvenile fish, may repopulate areas open to fishing (Berkeley et al., 2004; Francini-Filho and Moura, 2008). Whether this occurs, depends on the location of Sanctuary zones, the oceanographic conditions in the region, larval dispersal and larval life history characteristics (Watson et al., 2009). The South African Sanctuary zones are situated north and south of the multiple-use zones. Given the greater biomass and abundance of fish within the Sanctuary zones, the potential for connectivity through adult spillover or larval dispersal between the zones is thus high (e.g. Russ, 2002; Tupper, 2007). It is most likely that dispersal or movement would take place along the northsouth gradient due to the predominantly south-flowing Agulhas current, making the reefs in the Northern sanctuary zones the most important reefs due to their strategic position. On a larger geographic scale, it is also likely that some degree of connectivity exists between coral reef fish communities along the east African coast and the fish communities on the southern African coral reefs. The prevailing southward flowing currents in the Mozambique Channel (Lutjeharms, 2006) have been suggested as a mechanism linking populations of certain coral species (Ridgway et al., 2008; Macdonald et al., 2011). If the northern reefs are acting as sources of propagules for fish communities in the south, any threats or disturbances to these reefs will have significant consequences for the southern African coral reef communities. 4.2. Scuba diving Due to increased mortality of target species associated with fishing, it was anticipated that the densities and biomass of these species would be lowest in the Diving-Fishing zones and similar in the Sanctuary and High-Diving zones. Differences between Sanctuary and Diving-Fishing zones were significant; however, so too were the differences between Sanctuary and High-Diving zones. Lack of compliance by the fishers in the High-Diving zone may provide an explanation for the low abundances of the
136
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139
Fig. 4. Length-frequency distribution plots of discriminating species in the four different zones. a) Aprion virescens, b) Caranx melampygus, c) Oplegnathus robinstoni, d) Variola louti, e) Epinephelus tukula and f) Lutjanus bohar. Circles indicate size at sexual maturity.
six discriminating species. However, the high volume of divers on this reef would make it likely that suspicious fishing activity would be reported to the authorities. While poaching cannot be completely excluded, the high diving intensity in this zone may
Table 5 Results of abundance GLM with negative binomial distribution and log-link function where d.f. ¼ degrees of freedom, AIC ¼ Akaike’s information criterion, P ¼ significance level. Abundance model structure
d.f.
AIC
Residual deviance
Explained deviance %
p
Null þActivity þTopog Full model
3 1 4
2951.6 2948.2 2918.3 2916.3
473.32 432.05 428 428
8.7 0.9 9.6
<0.001 <0.05 <0.001
be having a negative effect on the abundance of these large species. The low densities of the top-level predator, E. tukula on the High-Diving reef, were of particular concern as this species is one of the most significant large-bodied predators on South African
Table 6 Results of biomass GLM with gamma distribution and inverse function where d.f ¼ degrees of freedom, AIC ¼ Akaike’s information criterion, P ¼ significance level. Biomass model structure Null þCoral cover þActivity Full model
d.f.
AIC
Residual deviance.
2 3 5
9048.2 9041.6 8930.8 8921.8
768.52 753.53 608.01 592.98
Explained deviance %
p
2.0 18.9 20.9
<0.01 <0.001 <0.001
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139
coral reefs. Historic records and anecdotal accounts of higher abundances of E. tukula on the High-Diving reef during the 1980s to early 1990s (Koornhof, 1991; Chater et al., 1995) indicate that a reduction in population numbers has occurred during the last three last decades. The results and observations of this study suggest that this decline may be linked to high diving intensity. E. tukula is a large resident predator known to show aggressive territorial behaviour towards divers on reefs isolated from human activities (Debelius, 2001; Peschak, 2010). All E. tukula encountered on Sanctuary reefs displayed aggressive or curious behaviour towards divers which included open mouth displays, bumping of divers, biting of the buoy-reel and stalking of divers throughout the dive. In contrast, E. tukula behaviour on the Diving-Fishing reef was cautious and the divers were seldom approached. Furthermore, E. tukula were most commonly observed at the edge of diver visibility where they moved from one overhang to another. Thus the persistent presence of SCUBA divers on the High-diving reef may be causing a competitive disturbance to E. tukula. 4.3. Implications for management No-take MPAs carry a high socio-economic cost because of lost fishing grounds (Jones, 2008), but show great benefits and yield significantly great biomass and densities (Gell and Roberts, 2003; Lester and Halpern, 2008). Multiple-use MPAs are typically seen as more feasible and are often implemented as a compromise because they confer some benefits over open access areas (Lester and Halpern, 2008). This study also provided evidence that notake MPA have significantly greater benefits to coral reef fish communities in South Africa; however, the differences between the multiple-use and open zones were considerably smaller than anticipated. The implications for this are significant in terms of the objectives and goals of the different MPA zones. The open zone in southern Mozambique had little legislation and no enforcement
137
protecting the reefs at the time of data collection. In contrast, the South African MPAs have clear objectives outlined in the MPA management plan (Ezemvelo KwaZulu-Natal Management Plan 2007). However, the objectives are identical for all MPA zones, despite differences in the type and intensity of human activity in each zone. Thirteen objectives are relevant to the coral reefs and fall within three functional categories; protection, fisheries management and utilisation (see Floros, 2010 for details). When assessed using the indicator parameters (biomass, densities, trophic composition and reproductive potential), eight objectives were questionable with regard to achieving their management goals. This indicates that the multiple-use zones are not providing protection to the reef fish communities under the current levels of human activity. It also suggests that these zones require a more complex management approach which includes a reassessment of their objectives to balance the trade-off between sustainable resource use and conservation. In contrast, the no-take Sanctuary zones appear to have achieved all management objectives. The prohibition of all human activity in these areas has promoted greater biomass, densities and reproductive potential of the reef fish communities. Considering their long history of closure to human activities, the Sanctuary zones potentially represent undisturbed ecosystems on which future management plans should be based. Acknowledgements We are grateful for the financial support provided by the National Research Foundation and the South African Association for Marine Biological Research. We also thank the staff of the Oceanographic Research Institute for their technical support in the field. The 4x4 vehicle used in this project was sponsored by the Mazda Wildlife Fund. Ezemvelo KwaZulu-Natal Wildlife and the iSimangaliso Wetland Authority are acknowledged for their logistical support.
Appendix A Mean abundance (fish 78 m2) of the Fish-index species and differences between zones shown by one-way ANOVA.
Acanthurus leucosternon Amphiprion allardi Aprion virescens Balistoides conspicillum Bodianus diana Caranx melampygus Chaetodon madagaskariensis Chaetodon meyeri Chaetodon trifascialis Chaetodon trifasciatus Diplodus cervinus Epinephelus tukula Forcipiger flavissimus Labroides dimidiatus Lutjanus bohar Odonus niger Oplegnathus robinsoni Plectroglyphidodon johnstonianus Pomacanthus imperator Pygoplites diacanthus Scarus rubroviolceus Siganus sutor Thalassoma hebraicum Variola louti Zebrasoma desjardini a
Sanctuary
High-Diving
Diving-Fishing
Open
p
0.904 0.220 0.617 0.072 0.502 0.919 0.804 0.885 0.038 0.024 0.019 0.292 0.512 1.507 1.512 2.129 0.502 0.464 0.187 0.005 1.292 0.072 3.024 0.531 0.014
1.200 0.077 0.062 0.046 0.554 0.615 0.754 0.400 0.169 0 0.385 0.077 0.292 2.338 0.138 1.754 0.108 0.369 0.246 0.015 1.569 0.215 2.462 0.185 0.092
1.296 0.576 0.024 0.464 0.464 0.288 0.984 0.144 0 0 0 0.080 0.352 2.704 0.184 4.024 0.104 0.032 0.144 0.016 1.200 0.080 2.576 0.184 0.024
0.353 0.333 0 0.059 0.333 0.098 0.706 0.549 0.137 0.078 0 0 0.294 2.235 0.118 1.137 0.039 0.235 0.039 0 0.725 0.137 2.294 0.098 0
NS <0.001 <0.001 0.007a NS 0.003 NS <0.001 0.007a 0.007a NS <0.001 NS 0.005 <0.001 NS 0.005 <0.001 NS NSa NS NSa NS <0.001 NSa
Indicates species that were inconclusive in the ANOVA or Multiple comparisons procedure due to their uncommon status.
138
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139
Appendix B Mean biomass (kg 78 m2) of the Fish-index species and differences between zones shown by one-way ANOVA.
Acanthurus leucosternon Amphiprion allardi Aprion virescens Balistoides conspicillum Bodianus diana Caranx melampygus Chaetodon madagaskariensis Chaetodon meyeri Chaetodon trifascialis Chaetodon trifasciatus Diplodus cervinus Epinephelus tukula Forcipiger flavissimus Labroides dimidiatus Lutjanus bohar Odonus niger Oplegnathus robinsoni Plectroglyphidodon johnstonianus Pomacanthus imperator Pygoplites diacanthus Scarus rubroviolaceus Siganus sutor Thalassoma hebraicum Variola louti Zebrasoma desjardinii a
Sanctuary
High-Diving
Diving-Fishing
Open
p
0.125 0.004 1.914 0.102 0.045 2.393 0.020 0.058 0.001 0.001 0.005 8.126 0.025 0.010 4.361 1.079 1.197 0.002 0.194 0.000 1.340 0.050 0.242 1.170 0.005
0.145 0.001 0.062 0.029 0.056 0.828 0.046 0.031 0.019 0 0.029 1.075 0.015 0.016 0.407 0.731 0.134 0.002 0.329 0.001 2.571 0.133 0.258 0.490 0.021
0.193 0.016 0.017 0.069 0.038 0.402 0.031 0.017 0 0 0 2.077 0.018 0.025 0.215 1.244 0.190 0.001 0.209 0.001 2.508 0.066 0.314 0.151 0.009
0.036 0.012 0 0.119 0.025 0.092 0.031 0.046 0.019 0.010 0 0 0.018 0.015 0.048 1.000 0.114 0.003 0.058 0 1.243 0.125 0.239 0.215 0
0.047 <0.001 <0.001 NS NS <0.001 NS <0.001 0.011a NSa NSa <0.001 NS 0.004 <0.001 NS 0.04 <0.001 NS NSa <0.001 NSa NS <0.001 NSa
Indicates species that were inconclusive in the ANOVA or Post Hoc comparison due to their uncommon status.
References
Appendix C Trophic guild allocation of the Fish-index species and the rationale for their inclusion in the study. B ¼ benthivore, C ¼ corallivore, H ¼ herbivore, I ¼ invertivore, M-L P ¼ medium-level predator, O ¼ omnivore, P ¼ planktivore, T-L P ¼ top-level predator. Trophic guild
Rationale
Acanthurus leucosternon Amphiprion allardi Aprion virescens
H O M-L P
Balistoides conspicillum Bodianus diana Caranx melampygus
I I T-L P
Chaetodon madagaskariensis Chaetodon meyeri Chaetodon trifascialis Chaetodon trifasciatus Diplodus cervinus Epinephelus tukula Forcipiger flavissimus Labroides dimidiatus Lutjanus bohar Odonus niger Oplegnathus robinsoni Plectroglyphidodon johnstonianus Pomacanthus imperator Pygoplites diacanthus
O C C C I T-L P I I M-L P P B C B B
Scarus rubroviolceus Siganus sutor Thalassoma hebraicum Variola louti Zebrasoma desjardinii
H H I M-L P H
Common herbivore Iconic species Target species and important predator Sensitive to diver presence Common prey species Target species and important predator Generalist feeder Specialist feeder Specialist feeder Specialist feeder Regional endemic Largest reef predator Prey species Specialist feeder Important predator Trophic representative Regional endemic Coral specific species Specialist feeder Specialist feeder and uncommon status Largest reef herbivore Important regional herbivore Common prey species Important reef predator Conspicuous grazer
Agardy, T., Bridgewater, P., Crosby, M.P., Day, J., Dayton, P.K., Kenchington, R., Laffoley, D., McConney, P., Murray, P.A., Parks, J.E., Peau, L., 2003. Dangerous targets? Unresolved issues and ideological clashes around marine protected areas. Aquat. Conserv. 13, 353e367. Alder, J., 1996. Have tropical marine protected areas worked? An initial analysis of their success. Coast Manage. 24, 97e114. Angermeier, P., Karr, J., 1994. Biological integrity versus biological diversity as policy directives. BioScience 44, 690e697. Barker, N.H.L., Roberts, C.M., 2004. Scuba diver behaviour and the management of diving impacts on coral reefs. Biol. Conserv. 120, 481e489. Baskett, M.L., Levin, S.A., Gaines, S.D., Dushoff, J., 2005. Marine reserve design and the evolution of size at maturation in harvested fish. Ecol. Appl. 15, 882e907. Bell, J.D., Kronen, M., Vunisea, A., Nash, W., Keeble, G., Demmkea, A., Pontifex, S., Andréfouët, S., 2009. Planning the use of fish for food security in the Pacific. Mar. Policy 33, 64e76. Bellwood, D.R., Hughes, T.P., Folke, C., Nyström, M., 2004. Confronting the coral reef crisis. Nature 429, 827e833. Berkeley, S.A., Hixon, M.A., Larson, R.J., Love, M.S., 2004. Fisheries sustainability via protection of age structure and spatial distribution of fish populations. Fisheries 29, 23e31. Celliers, L., Schleyer, M.H., 2008. Coral community structure and risk assessment of high-latitude reefs at Sodwana Bay, South Africa. Biodivers. Conserv. 17, 3097e 3117. Cesar, H.S.J., Burke, L., et al., 2003. The Economics of Worldwide Coral Reef Degradation. Cesar Environmental Economics Consulting (CEEC), 6828GH Arnhem, The Netherlands. Chater, S.A., Beckley, L.E., van der Elst, R.P., Garrat, P.A.,1995. Underwater visual census of fishes in the St Lucia Marine Reserve, South Africa. Lammergeyer 43, 15e23. Clarke, K.R., Gorley, R.N., 2006. PRIMER v6: User Manual/Tutorial. PRIMER-E, Plymouth. Cooke, S.J., Cowx, I.G., 2004. The role of recreational fishing in the global fish crisis. BioScience 54, 857e859. Dale, V.H., Beyeler, S.C., 2001. Challenges in the development and use of ecological indicators. Ecol. Indic. 1, 3e10. Debelius, H., 2001. Indian Ocean Reef Guide. IKAN-Unterwasserarchiv, Frankfurt. Dulvy, N.K., Polunin, N.V.C., Mill, A.C., Graham, N.A.J., 2004. Size structural changes in lightly exploited coral reef fish communities: evidence for weak indirect effects. Can J. Fish Aquat. Sci. 61, 466e475. English, S., Wilkinson, C., Baker, V. (Eds.), 1994. Survey Manual for Tropical Marine Resources. Australian Institute of Marine Science, Townsville. Evans, R.D., Russ, G.R., 2004. Larger biomass of targeted reef fish in no-take marine reserves on the Great Barrier Reef, Australia. Aquat. Conserv. 14, 505e519. Floros, C., 2010. An Evaluation of Coral Reef Fish Communities in South African Marine Protected Areas (PhD dissertation). Oceanographic Research Institute, University of KwaZulu-Natal.
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139 Floros, C., Schleyer, M.H., Maggs, J.Q., Celliers, L., 2012. Baseline assessment of highlatitude coral reef fish communities in southern Africa. Afri J. Mar Sci. 34, 55e69. Froese, R., Pauly, D., 2011. Fishbase. www.fishbase.org (accessed July 2011). Francini-Filho, R.B., Moura, R.L., 2008. Evidence for spillover of reef fishes from a notake marine reserve: an evaluation using the before-after control-impact (BACI) approach. Fish Res. 93, 346e356. Gell, F.R., Roberts, C.M., 2003. Benefits beyond boundaries: the fishery effects of marine reserves. Trends Ecol. Evol. 18, 448e455. Graham, N.A.J., Evans, R.D., Russ, G.R., 2003. The effects of marine reserve protection on the trophic relationships of reef fishes on the Great Barrier Reef. Environ. Conserv. 30, 200e208. Graham, N.A.J., Dulvy, N.K., Jennings, S., Polunin, N.V.C., 2005. Size-spectra as indicators of the effects of fishing on coral reef fish assemblages. Coral Reefs 24, 118e124. Halpern, B.S., Walbridge, S., Selkoe, K.A., Kappe, C.V., Micheli, F., D’Agrosa, C., Bruno, J.F., Casey, K.S., Ebert, C., Fox, H.E., Fujita, R., Heinemann, D., Lenihan, H.S., Madin, E.M.P., Perry, M.T., Selig, E.R., Spalding, M., Steneck, R.S., Watson, R., 2008. A global map of human impact on marine ecosystems. Science 319, 948e952. Hawkins, J.P., Roberts, C.M., Kooistra, D., Buchan, K., White, S., 2005. Sustainability of SCUBA diving tourism on coral reefs of Saba. Coastal Manage. 33, 373e387. IUCN, 2004. Managing Marine Protected Areas: a Toolkit for the Western Indian Ocean. In: IUCN Eastern African Regional Programme. Nairobi, Kenya. Jameson, S.C., Kelty, R.A., 2004. A review of indicators of land-based pollution stress on coral reefs. In: Joint EPA/NOAA/USGA/DOI Workshop: Assessing Pollution Stress on Coral Reefs Aug 31eSept 2 Honolulu, Hawaii. Jones, P.J.S., 2008. Fishing industry and related perspectives on the issues raised by no-take marine protected area proposals. Mar. Policy 32, 749e758. Jordan, I.E., Samways, M., 2001. Recent changes in the coral composition of a South African coral reef, with recommendations for long-term monitoring. Biodivers. Conserv. 10, 1027e1037. Karr, J.R., 1981. Assessment of biotic integrity using fish communities. Fisheries 6, 21e27. Kelleher, G., Kenchington, R., 1992. Guidelines for Marine Protected Areas. IUCN, Gland, Switzerland. Kelleher, G., Bleakley, C., Wells, S., 1995. A Global Representative System of Marine Protected Areas. The World Bank, Washington. Knowlton, N., Jackson, J.B.C., 2008. Shifting baselines, local impacts, and global change on coral reefs. PLoS ONE 6, 215e220. Koornhof, A., 1991. The Dive Sites of South Africa. Comprehensive Coverage of Diving and Snorkelling. New Holland, Cape Town. Lester, S.E., Halpern, B.S., 2008. Biological responses in marine no-take reserves versus partially protected areas. Mar. Ecol. Prog. Ser. 367, 49e56. Lutjeharms, J.R.E., 2006. The Agulhas Current. Springer, Berlin. Macdonald, A.H.H., Lamb, J., Schleyer, M.H., 2011. Acropora austera connectivity in the south western Indian Ocean assessed using nuclear intron sequence data. Mar Biol. 158, 613e621. Maunder, M.N., Punt, A.E., 2004. Standardizing catch and effort data: a review of recent approaches. Fish Res. 70, 141e159. McClanahan, T.R., Graham, N.A.J., Calnan, J.M., MacNeil, M.A., 2007. Toward pristine biomass: reef fish recovery in coral reef marine protected areas in Kenya. Ecol. Appl. 17, 1055e1067. Mora, C., Andrefouet, S., Costello, M.J., Kranenburg, C., Rollo, A., Vernon, J., Gaston, K.J., Muyers, R., 2006. Coral reefs and the global network of marine protected areas. Science 312, 1750. Pauly, D., Christensen, V., Guénette, S., et al., 2002. Towards sustainability in world fisheries. Nature 418, 689e695. Pereira, H.M., 2003. Recreational Scuba Diving and Reef Conservation in Southern Mozambique (MSc thesis). Oceanographic Research Institute, University of KwaZulu-Natal.
139
Peschak, T.P., 2010. Lost World: the Marine Realm of Aldabra and the Seychelles. JB Publishing. Pieters, M., 2009. Dive Statistics for Sodwana Bay. Ezemvelo KZN Wildlife Authorities. Polunin, N.V.C., Roberts, C.M., 1993. Great biomass and value of target coralreef fishes in two small Caribbean marine reserves. Coral Reefs 100, 167e 176. Pomeroy, R.S., Watson, L.M., Parks, J.E., Cid, G.A., 2005. How is your MPA doing? A methodology for evaluating the management effectiveness of marine protected areas. Ocean. Coast. Manage. 48, 485e502. R Development Core Team, 2011. R: a Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria, ISBN 3900051-07-0. http://www.R-project.org/. Ramsay, P.J., 1996. Quaternary marine geology of the Sodwana Bay continental shelf, Northern KwaZulu-Natal. Bull. Geo. Surv. S. Afr. 117, 1e85. Ridgway, T., Riginos, C., Davis, J., Hoegh-Guldberg, O., 2008. Genetic connectivity patterns of Pocillopora verrucosa in southern African marine protected areas. Mar. Ecol. Prog. Ser. 354, 161e168. Riegl, B., Schleyer, M.H., Cook, P.J., Branch, G.M., 1995. Structure of Africa’s southernmost coral communities. Bull. Mar. Sci. 5, 676e691. Russ, G.R., Alcala, A.C., 1989. Effects of intense fishing pressure on an assemblage of coral reef fishes. Mar. Ecol. Prog. Ser. 56, 13e27. Russ, G.R., 2002. Yet another review of marine reserves as reef fishery management tools. In: Sale, P. (Ed.), Coral Reef Fishes. Academic Press, San Diego, CA, pp. 421e443. Sale, P.F., Cowen, R.K., Danilowicz, B.S., Jones, G.P., Kritzer, J.P., Lindeman, K.C., Planes, S., Polunin, N.V.C., Russ, G.R., Sadovy, Y.J., Steneck, R.S., 2005. Critical science gaps impede use of no-take fishery reserves. Trends Ecol. Evol. 20, 74e80. Samoilys, M.A., Carlos, G., 2000. Determining methods of underwater visual census for estimating the abundance of coral reef fishes. Environ. Biol. Fish 57, 289e304. Schleyer, M.H., 2000. South African coral communities. In: McClanahan, T.R., Sheppard, C., Obura, D. (Eds.), Coral Reefs of the Indian Ocean. Oxford University Press, New York, pp. 83e105. Tupper, M.H., 2007. Spillover of commercially valuable reef fishes from marine protected areas in Guam, Micronesia. Fish. Bull. 105, 527e537. Tratalos, J.A., Austin, T.J., 2001. Impacts of recreational SCUBA diving on coral communities of the Caribbean Island of Grand Cayman. Biol. Conserv. 102, 67e75. Watson, D., Anderson, M.J., Kendrick, G.A., Nardi, K., Harvey, E.S., 2009. Effects of protection from fishing on the lengths of targeted and non-targeted fish species at the Houtman Abrolhos Islands, Western Australia. Mar. Ecol. Prog. Ser. 384, 241e249. Wells, A., Burgess, N., Ngusaru, A., 2007. Towards the 2012 marine protected area target in Eastern Africa. Ocean. Coast. Manage 50, 67e83. Westera, M., Lavery, P., Hyndes, G., 2003. Differences in recreationally targeted fishes between protected and fished areas of a coral reef marine park. J. Exp. Mar. Biol. Ecol. 294, 145e168. White, A.T., Meneses, A.B.T., Ovenden, M., Tesch, S., 2006. Sustaining marine protected areas through continued monitoring and evaluation: the MPA report guide and management rating system. In: Proc 10th Int Coral Reef Symp, Okinawa, Japan, pp. 1466e1470. Wilkinson, C.R., 2008. Status of Coral Reefs of the World: 2008. Global Coral Reef Monitoring Network and Reef and Rainforest Research Centre, Townville, Australia. Zakai, D., Chadwick-Furman, N.E., 2002. Impacts of intensive recreational diving on reef corals at Eilat, northern Red Sea. Biol. Conserv. 105, 179e187.
Contents lists available at ScienceDirect
Ocean & Coastal Management journal homepage: www.elsevier.com/locate/ocecoaman
Fish as indicators of diving and fishing pressure on high-latitude coral reefs C. Floros*, M.H. Schleyer, J.Q. Maggs Oceanographic Research Institute, P.O. Box 10712, Marine Parade, 4056 Durban, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history: Available online
Despite the proclamation of South Africa’s coral reef marine protected areas (MPAs) more than 20 years ago, the effects of human activities on the fish communities have not been investigated. This study used a multi-species Fish-index to compare ecological indicators such as biomass, abundance, trophic structure and reproductive potential between multiple-use and no-take sanctuary zones. Seven study reefs were surveyed; six in South Africa and a non-MPA reef in southern Mozambique. Randomly stratified underwater visual censuses (UVC) using the point count technique were used to survey fish communities. Environmental variables and habitat characteristics were also recorded. Nonmetric multidimensional scaling ordinations were similar for abundance and biomass trends and revealed a high degree of overlap between all zones, except for the no-take Sanctuaries. The latter formed discrete clusters and were significantly different (Analysis of Similarity) to the other zones. Total abundance and biomass were highest in the Sanctuary zones and lowest in the Open zone. Differences in trophic composition between zones were largely due to predatory species. This was supported by similarity percentages analysis (SIMPER) which identified six discriminating species. Length-frequency analysis of these species revealed consistent trends with higher numbers of large individuals in the Sanctuary zones and reduced numbers of small individuals in zones open to human activity. These results along with those of the Generalised Linear Models (GLM) demonstrate that human activities are affecting the southern African coral reef fish communities. Marginal differences between the multiple-use MPA zones on the South African reefs and the non-MPA reef in southern Mozambique suggest that MPA management objectives require reevaluation. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Coral reefs provide worldwide benefits and services to the value of US$ 29.8 billion every year (Cesar et al., 2003). The biggest proportion is generated by tourism (US$ 9.6 billion) and fisheries (US$ 5.7 billion). Yet, these activities have increasingly caused degradation on many reefs throughout the world (Halpern et al., 2008; Wilkinson, 2008). Various measures to manage and mitigate coral reef damage have been proposed and, among those, marine protected areas (MPAs) have been proposed as an ideal management solution because they were perceived to simultaneously address issues of overfishing, habitat degradation, and tourism development (Kelleher and Kenchington, 1992; IUCN, 2004). MPAs are specifically intended to limit human activities in designated locations (Sale et al., 2005; Mora et al., 2006) and the
* Corresponding author. Tel.: þ27 31 3288229. E-mail address: cfl[email protected] (C. Floros). 0964-5691/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ocecoaman.2013.08.005
degree to which human activities are limited determines the type of MPA. In most instances, MPAs can be classified into two broad types; areas that are open to resource use and areas closed to resource use. No-take MPAs are areas closed to exploitation (here termed sanctuaries). The second type of MPA allows harvesting of resources, but under protective regulations that pertain to each species being harvested. In addition, the types of fishing or harvesting gear may be restricted. Such MPAs are multiple-use zones and most often permit recreational activities such as SCUBA diving, snorkeling, whale watching and fishing. Nevertheless, MPAs are not ‘cure alls’ (Alder, 1996; Agardy et al., 2003; Mora et al., 2006) and many face difficulty in implementation and enforcement due to poor governance, and lack of management guidance and evaluation (White et al., 2006). Among the major challenges restricting effective MPA management is a lack of scientific information about the status and nature of activities operating therein (Kelleher et al., 1995; Pomeroy et al., 2005; Wells et al., 2007). Assessment of MPA effectiveness is a matter of great urgency and importance given the multitude of stressors threatening the
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139
future of coral reefs (Bellwood et al., 2004; Knowlton and Jackson, 2008). Furthermore, many MPAs are promulgated as multiple-use zones and complex decision-making processes are often required to balance conservation and socio-economic objectives. Thus, it is imperative to assess the impacts of human activities on the biological components of coral reefs and whether these impacts are consistent with the management objectives of the MPA. Obtaining such information requires regular monitoring and assessment of ecological integrity within MPA boundaries. Ecological integrity refers to system wholeness, including the presence of appropriate species, populations, and communities and the occurrence of ecological processes at appropriate rates and scales (Karr, 1981; Angermeier and Karr, 1994). Evaluating changes in ecological integrity is often a major obstacle facing MPA managers due to a scarcity of technical skills, experience and funding (Pomeroy et al., 2005). Indicator-based monitoring programmes may provide an effective solution to these challenges because
131
ecological indicators may represent inexpensive means of gathering scientific data that does not require skilled personnel. For this reason, they have become important tools in coral reefs research, reporting and management (see Jameson and Kelty, 2004 for review). The challenge is determining which biological criteria are effective measures of ecological integrity but are also simple enough to monitor (Dale and Beyeler, 2001). Fish are conspicuous biological components of coral reefs that have good indicator potential owing to their importance as a valuable protein source (Pauly et al., 2002; Bell et al., 2009) and functional ecological roles on reefs (Bellwood et al., 2004). Accordingly, this study selected a multi-species index (Fish-index) to assess the effects of human activities on high-latitude coral reefs in South Africa. The aim was to use comparisons of ecological parameters such as biomass, abundance, trophic structure and reproductive potential as measures of ecological integrity. Changes observed in functional
Fig. 1. Studied reefs on the north east coast of South Africa and southern Mozambique. Sites included three Sanctuary zones (Leadsman Shoal, Red Sands Reef, and Rabbit Rock), three multiple-use zones (Two-mile, Seven-mile and Nine-mile reef) and a non-MPA zone (Shallow Malongane).
132
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139
Table 1 Zonation of the reefs based on SCUBA diving and fishing intensities. SCUBA diving statistics are averages for the period 2007e2008 (Pieters, 2009). Zone
Open Diving-Fishing High-Diving Sanctuary
Activity
Reef name
SCUBA diving intensity
Fishing intensity
Low (w4500 dives/year) Low (w2100 dives/year) High (w54,000 dives/year) Nil
High (unregulated angling and spearfishing) Restricted (angling and spearfishing; gamefish onlya). Nil Nil
Shallow Malongane Seven-mile Reef, Nine-mile Reef Two-mile Reef Leadsman Shoal, Red Sands, Rabbit Rock
a Pelagic bony fish of the families Scombridae, Carangidae, Pomatomidae, Coryphaenidae, Rachycentridae, Xiphiidae, Ostiophoridae and Sphyraenidae, the species Aprion virescens, as well as pelagic cartilaginous fish of the families Carcharinidae, Isuridae, Sphyrnidae, Alopiidae and Odontaspididae (Marine Living Resources Act, Section 3.1 (G) Regulation R1429).
processes such as growth, reproduction and trophic functioning may provide MPA managers with insight into current conditions and aid in predicting future trends. South Africa’s coral reefs are located within the boundaries of two longstanding, contiguous MPAs. Two types of conservation strategies are recognized in the MPAs: no-take sanctuary zones and multiple-use zones. Recreational fishing and SCUBA diving are the most common activities in the MPAs (Schleyer, 2000). Fishing is known to have a direct effect on fish communities via the harvest of target and bait species, and the removal of functional groups, and has the potential to cause significant changes in the structure of reef fish communities (Cooke and Cowx, 2004; Dulvy et al., 2004). SCUBA diving, has been shown to affect coral communities (Tratalos and Austin, 2001; Zakai and Chadwick-Furman, 2002; Barker and Roberts, 2004; Hawkins et al., 2005), however, the paucity of empirical studies investigating the effects of high diving intensity on fish communities in the literature represents a ‘knowledge-gap’. It is thus both timely and pertinent that the effects of human activities on the South African coral reef fish assemblages are assessed for efficient MPA management.
respectively. No changes have been made to the MPA zonation since their proclamation. For this study, the reefs were zoned according to the intensity and type of human activity (Table 1). These were Sanctuary, High-Diving, Diving-Fishing and Open zones. Sanctuary zones theoretically prohibited all forms of human activity. The High-Diving zone permitted SCUBA diving only. The Diving-Fishing zone permitted SCUBA diving and restricted fishing (gamefish only). Lastly, the Open zone had no active law enforcement at the time of data collection which meant that activities on the reef were unregulated. No data for fishing intensity were available for any of the zones.
2. Materials and methods
2.4. Surveys of reef fish communities
2.1. Study area
Surveys of fish communities were conducted from August 2007 to February 2009 to include four summer and four winter sampling periods. The point count underwater visual census technique, adapted from Samoilys and Carlos (2000), was used to estimate the abundance and biomass of the Fish-index species. A total of 60 point counts were conducted per reef. Each census (point count) consisted of a 5 min count within a circle 10 m in diameter. All Fish-index species observed on the reef and within the water column during each point count were recorded and their sizes were estimated to the nearest 10 cm. Estimates of fish length were used to generate biomass using known length-weight regression coefficients from Fishbase (Froese and Pauly, 2011). Only two divers entered the water at any one time, a surveyor and a buddy diver. The same diver conducted all point counts to minimize variation and error incurred by diver bias. Point counts within each reef were separated by at least 50 m. The time of each point count was conducted between 0800 and 1400. All counts were conducted at a depth range of 12e15 m.
The South African coral reefs are located along the north-east coast of South Africa. Six study reefs were selected to represent different types of MPA zones (Fig. 1). The study area was extended to include non-MPA reefs in southern Mozambique for comparative purposes. However, only one reef, located at Ponta Malongane, could be included as a study site due to political and logistical constraints. The southern Mozambican and South African coral reefs are the southernmost reefs in the Western Indian Ocean (Riegl et al., 1995). Due to their close proximity they share numerous similarities. They are exposed to similar oceanographic conditions, which are dominated by the southwards flowing Agulhas Current (Lutjeharms, 2006). They are confined to the narrow continental shelf (Ramsay, 1996) and can be classified as patch reefs (Ramsay and Mason, 1990). They are atypical coral reefs in that the coral communities are non-accretive and form a veneer on latePleistocene beach rock (Ramsay, 1996). Their benthic communities are similar with corals contributing 50e70% to the total living cover (Jordan and Samways, 2001; Pereira, 2003; Celliers and Schleyer, 2008). The fish assemblages in both regions are very similar, with approximately 300 reef-associated species; 80% of which are Indo-Pacific (Pereira, 2003; Celliers and Schleyer, 2008; Floros et al., 2012). 2.2. Reef protection status and human activity The South African coral reefs are situated within the Maputaland MPA and St Lucia MPA, proclaimed in 1986 and 1979,
2.3. Fish-index selection Species were included as indicators based on literature reviews and field observations. The criteria included vulnerability to exploitation by fishing or hobbyists, trophic status and ease of identification. The Fish-index species are listed in Appendix C along with the justification for their selection.
2.5. Trophic levels Species were allocated to one of eight trophic guilds; top-level predators, medium-level predators, planktivores, omnivores, herbivores, benthivores, corallivores and invertivores (Appendix C). Trophic allocation was based on diet information retrieved from the literature and supplemented by field observations (see Floros et al., 2012 for details).
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139
a
2D Stress: 0.26
Human activity High-Diving Diving-Fishing Sanctuary Open
133
b
2D Stress: 0.25
Human activity High-Diving Diving-Fishing Sanctuary Open
Fig. 2. Non-metric multidimensional scaling ordination of samples based on a) fourth-root transformed abundance data and b) log(xþ1) transformed biomass data.
2.6. Habitat characteristics and environmental variables On completion of each point count, aspects of the reef habitat were described within the point count area; topography and coral cover. Coral cover was estimated using a rapid visual technique adapted from English et al. (1994), where estimates where divided into three categories (low, medium and high) according to the percentage of coral (hard and soft). The topographic complexity of the substratum was visually estimated using a six point scale system adapted from Polunin and Roberts (1993). The adaptations were specific to southern African coral reefs and included the following categories: 1 e flat reef; 2 e low undulating spur and groove; 3 e medium slopes or ridges with no grooves or gullies; 4 e medium spur and groove or pinnacles; 5 e high slopes or ridges; and 6 e high spur and groove, overhangs or pinnacles. Depth was measured using a dive computer and sea temperatures were obtained from an underwater temperature probe stationed at 18 m. 2.7. Statistical analysis Univariate analyses were carried out on abundance and biomass data using One-Way analysis of variance (ANOVA). Data were not normally distributed and thus non-parametric KruskaleWallas One-Way ANOVA on ranks was used. Dunn’s pairwise multiple comparisons procedure was used to detect differences between fish abundance and biomass parameters. Univariate analyses were conducted using the statistical package Sigma Plot 11.0. The criterion for significance for all tests was p 0.05. Multivariate analyses were undertaken using PRIMER v.6 to further explore difference between zones. Abundance data were fourth-root transformed, while the log(xþ1) transformation was used for biomass data due to large differences in the values for toplevel predators. Non-metric multidimensional scaling (MDS) was used to examine differences in spatial distribution of Fish-index assemblages across the different zones. Analysis of Similarity (ANOSIM) was also used to confirm or refute trends observed in the MDS ordinations. R-statistics >0.45 were considered to signify significant differences with a limited degree of overlap between fish community structure and values <0.45 indicated large degrees of overlap between community structures (Clarke and Gorley, 2006). SIMPER analysis was used to identify those Fish-index species responsible for the BrayeCurtis dissimilarity between MPA zones. Length-frequency graphs were generated for the six discriminating species identified by SIMPER. Size at sexual maturity for each of the six species was determined from the literature and
used as an approximation for reproductive potential. Generalised Linear Models (GLM) using R (R Development Core Team, 2011) were used to examine the influence of environmental variables (temperature, depth), habitat characteristics (coral cover and topography) and activity on fish community parameters (abundance and biomass). Activity was classified according to the intensity of diving and fishing in the different MPA zones (Table 1). The GLM for both abundance and biomass data followed the form:
ðcountsÞ ¼ b0 þ b1 ðactivityÞ þ b2 ðtemperatureÞ þ b3 ðcoral coverÞ þ b4 ðtopographyÞ þ b5 ðdepthÞ þε Overdispersion in the abundance data invalidated the use of the Poisson distribution and the negative binomial distribution with log-link function was used (Maunder and Punt, 2004). Biomass data were continuous and were thus modelled using gamma distribution with inverse-link function.
3. Results 3.1. Spatial distribution of species The spatial orientation of the abundance and biomass data suggested a high degree of overlap between the High-Diving, Diving-Fishing and Open zones (Fig. 2). In contrast, the Sanctuary zones formed distinct clusters in both data sets. ANOSIM tests confirmed the trends observed in both MDS plots (Tables 2 and 3). In terms of abundance, the Fish-index community differed significantly between Sanctuary zones and the Open zone, and between Sanctuary and Diving-Fishing zones (Global R ¼ 0.423; p < 0.001).
Table 2 Results of ANOSIM run on fourth-root transformed species abundance data for differences between zones. Global R ¼ 0.423. Significance of Global R < 0.001. Significant differences are in bold. Pairwise tests
R statistic
Significance level %
High-Diving, Fishing-Diving High-Diving, Sanctuary High-Diving, Open Diving-Fishing, Sanctuary Fishing-Diving, Open Sanctuary, Open
0.302 0.429 0.087 0.461 0.235 0.6
0.006 0.005 9.2 <0.0001 0.6 0.0005
134
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139
Table 3 Results of ANOSIM run on log(xþ1) transformed species biomass data for differences between zones. Global R ¼ 0.423. Significance of Global R < 0.001. Significant differences are in bold. Pairwise tests
R Statistic
Significance level %
High-Diving, Diving-Fishing Sanctuary, High-Diving High-Diving, Open Sanctuary, Diving-Fishing Diving-Fishing, Open Sanctuary, Open
0.202 0.516 0.162 0.468 0.236 0.713
0.2 0.0005 1.7 <0.0001 0.4 0.0005
The biomass data revealed that the Fish-index structure in Sanctuary zones was significantly different to the other three zones (Global R ¼ 0.423; p < 0.0001). 3.2. Abundance, biomass and trophic guild composition The total mean abundance and total mean biomass manifested similar trends across zones with the highest values recorded in Sanctuary zones and the lowest in the Open zone (Fig. 3). Differences in mean abundances were significant between Sanctuary (18.01 fish 78 m2) and High-Diving (13.86 fish 78 m2), and Sanctuary and Open zones (9.81 fish 78 m2) (p < 0.001). In addition, mean abundances between Diving-Fishing (15.93 kg 78 m2) and Open reefs were also significantly different. Total biomass on Sanctuary reefs (23.71 kg 78 m2) was significantly different and at least three times greater than in all other zones (p < 0.001). Invertivores were the most abundant trophic guild in all MPA zones and did not differ significantly between zones. Planktivores and herbivores were the next most abundant guilds. Abundances for herbivores were significantly different between High-Diving and the Open zone (P < 0.01). Top- and medium-level predators were significantly more abundant in Sanctuary zones (P < 0.001). Differences in biomass between MPA zones were most significant for predator guilds. In the Sanctuary zones, the total biomass of toplevel predators (8.32 kg 78 m2) was seven times great than that recorded in the High-Diving zones and four times greater than in the Diving-Fishing zone. No top-level predators were recorded in the Open zone. Medium-level predators in the Sanctuary zones (10.6 kg 78 m2) were at least six times greater than in the other three zones. 3.3. Discriminating species Comparisons between Sanctuary zones and the other zones revealed that six species were the top contributors to the dissimilarity in each pairwise comparison (Table 4). These species were Epinephelus tukula, Lutjanus bohar, Aprion virescens, Caranx melampygus, Variola louti and Oplegnathus robinsoni and their cumulative contribution to the dissimilarity between each comparative group was almost 50%. Excluding O. robinsoni, all the above species are important reef predators. All these species are potentially targeted in the Open zone; however, only C. melampygus and A. virescens may be targeted on the South African coral reefs. 3.4. Length-frequency analysis Length-frequency graphs were generated for the six discriminating species identified by SIMPER analysis. All six species were recorded to have greater abundance and mean body size in the Sanctuary zones (Fig. 4; Appendix A & B). There was a clear trend for reduced abundances and mean size in zones open to fishing (Fishing-Diving and Open) such that only low numbers of sexually
Fig. 3. Total mean abundance (a), biomass (b) and trophic structure of the Fish-index species in the four zones on the southern African study reefs.
mature individuals were recorded in these two zones. The Open zone had the lowest number and smallest mean body size of the six species, and two species, A. virescens (Fig. 4a) and E. tukula (Fig. 4e), were absent from this zone. In the High-Diving zone, no sexually mature individuals of A. virescens, Oplegnathus robinsoni and E. tukula were recorded, while sexually mature individuals of the three other species were present in low numbers. 3.5. Environmental variables, habitat characteristics and human activity The recorded environmental variables and habitat characteristics appeared to explain little of the deviance observed in the fish abundance and biomass between MPA zones. All variables were
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139 Table 4 Results of SIMPER analysis. Only species providing the highest percent contribution towards the average dissimilarity in abundance and biomass data between zones have been included. Species highlighted in bold are those contributing >40% to the overall dissimilarity.
Epinephelus tukula Lutjanus bohar Aprion virescens Caranx melampygus Variola louti Oplegnathus robinsoni Odonus niger Pomacanthus imperator Acanthurus leucosternon Balistoides conspicillum Chaetodon meyeri Scarus rubroviolaceus Bodianus diana Siganus sutor Forcipiger flavissimus Chaetodon trifascialis Amphiprion allardi Average dissimilarity
High-Diving vs sanctuary
Diving-Fishing vs sanctuary
Open vs sanctuary
10.32 9.84 8.37 6.97 7.19 6.31 5.88 5.44 4.34 3.67 3.65 3.26 3.74 2.57 3.44 2.12 0 45.01
9.18 8.65 9.42 6.9 6.96 6.63 7.1 5.45 4.31 4.71 4.61 3.12 3.99 2.77 3.25 0 3.28 44.11
9.85 9.58 9.51 7.73 7.18 6.68 6.55 4.94 4.06 3.86 3.77 3.63 3.35 3.34 3.25 2.47 2.36 52.49
initially modelled in the GLMs; however, only activity, topography and coral cover were found to be significant and were thus included in the final models (Tables 5 and 6). Activity was most accountable for the explained deviance in both the abundance (8.7%) and biomass (18.8%) between zones. Topography explained 0.9% of the deviance in abundance, while coral cover explained 2% of the deviance in biomass between zones. 4. Discussion This study used a pre-determined list of species to investigate differences in fish community parameters between zones of varying protection levels. Most of the significant differences in abundance, biomass, trophic levels and reproductive potential were recorded for predatory and target reef fish species. For all parameters, values were highest in the Sanctuary zones and lowest in the Open zone. These results are consistent with other studies comparing reef fish communities across gradients of protection in the Western Indian Ocean (McClanahan & Graham et al., 2005; McClanahan et al., 2007; Floros et al., 2012) and Indo-Pacific (Graham et al., 2003; Westera et al., 2003; Evans and Russ, 2004; Lester and Halpern. 2008) and strongly suggests that the differences are a result of human activities. These findings were confirmed by the GLMs which demonstrated that the deviance between MPA zones was explained more by human activity than by habitat characteristics or environmental variables. 4.1. Fishing Efforts have been made to regulate fishing on the South African coral reefs by restricting the extraction of species to gamefish. In addition to the species restrictions, there are daily fishing quotas for each target species (C. melampygus ¼ 5 day1, A. virescens ¼ 10 day1), but no minimum size limits. Despite the daily quotas, the results suggest that fishing is having an effect on populations of the aforementioned target fish species. Low densities and small mean body sizes of target species are detectable effects of overfishing on coral reefs (Russ and Alcala, 1989; Westera et al., 2003). In the Diving-Fishing zones, the reduced mean size of target species corresponded closely to theoretical predictions (see
135
Dulvy et al., 2004; Graham et al., 2005), with all target species showing significantly smaller mean sizes. A similar, but more severe trend was evident in the Open zone where target species such as L. bohar and C. melampygus were represented by very low abundances of small individuals. In addition A. virescens and E. tukula were not recorded. At the time of data collection, there were no fishing restrictions for C. melampygus, A. virescens, and L. bohar, while E. tukula was fully protected from recreational fishing only (Decree 51/99 of 31 August). The absence of the latter species from the study records may due to the inclusion of only one reef in southern Mozambique, due to intensive fishing pressure (recreational, subsistence, or semi-commercial) or a combination of both factors. However, similar sampling effort on other reefs recorded high abundances of E. tukula and the results are most likely indicative of high levels of exploitation. A reduction in the mean size of targeted species in fished areas is termed size selective fishing or ‘age truncation’ and is due to fishers targeting larger individuals (Berkeley et al., 2004). Age truncation can have an important effect on fish assemblage structure and function, potentially affecting the productivity and resilience of fish populations (Baskett et al., 2005). The low abundance and absence of sexually mature individuals of A. virescens, C. melampygus, O. robinsoni, E. tukula and L. Bohar recorded in the non-Sanctuary zones suggests reduced reproductive potential and could have significant ramifications for future generations. These populations may be reliant on juveniles from surrounding no-take zones to replenish stocks. Sanctuary or no-take zones have been advocated as areas of high reproductive output because there are greater densities of larger, sexually mature fish present. It has been further suggested that the increased reproductive output, whether in the form of eggs, larvae or juvenile fish, may repopulate areas open to fishing (Berkeley et al., 2004; Francini-Filho and Moura, 2008). Whether this occurs, depends on the location of Sanctuary zones, the oceanographic conditions in the region, larval dispersal and larval life history characteristics (Watson et al., 2009). The South African Sanctuary zones are situated north and south of the multiple-use zones. Given the greater biomass and abundance of fish within the Sanctuary zones, the potential for connectivity through adult spillover or larval dispersal between the zones is thus high (e.g. Russ, 2002; Tupper, 2007). It is most likely that dispersal or movement would take place along the northsouth gradient due to the predominantly south-flowing Agulhas current, making the reefs in the Northern sanctuary zones the most important reefs due to their strategic position. On a larger geographic scale, it is also likely that some degree of connectivity exists between coral reef fish communities along the east African coast and the fish communities on the southern African coral reefs. The prevailing southward flowing currents in the Mozambique Channel (Lutjeharms, 2006) have been suggested as a mechanism linking populations of certain coral species (Ridgway et al., 2008; Macdonald et al., 2011). If the northern reefs are acting as sources of propagules for fish communities in the south, any threats or disturbances to these reefs will have significant consequences for the southern African coral reef communities. 4.2. Scuba diving Due to increased mortality of target species associated with fishing, it was anticipated that the densities and biomass of these species would be lowest in the Diving-Fishing zones and similar in the Sanctuary and High-Diving zones. Differences between Sanctuary and Diving-Fishing zones were significant; however, so too were the differences between Sanctuary and High-Diving zones. Lack of compliance by the fishers in the High-Diving zone may provide an explanation for the low abundances of the
136
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139
Fig. 4. Length-frequency distribution plots of discriminating species in the four different zones. a) Aprion virescens, b) Caranx melampygus, c) Oplegnathus robinstoni, d) Variola louti, e) Epinephelus tukula and f) Lutjanus bohar. Circles indicate size at sexual maturity.
six discriminating species. However, the high volume of divers on this reef would make it likely that suspicious fishing activity would be reported to the authorities. While poaching cannot be completely excluded, the high diving intensity in this zone may
Table 5 Results of abundance GLM with negative binomial distribution and log-link function where d.f. ¼ degrees of freedom, AIC ¼ Akaike’s information criterion, P ¼ significance level. Abundance model structure
d.f.
AIC
Residual deviance
Explained deviance %
p
Null þActivity þTopog Full model
3 1 4
2951.6 2948.2 2918.3 2916.3
473.32 432.05 428 428
8.7 0.9 9.6
<0.001 <0.05 <0.001
be having a negative effect on the abundance of these large species. The low densities of the top-level predator, E. tukula on the High-Diving reef, were of particular concern as this species is one of the most significant large-bodied predators on South African
Table 6 Results of biomass GLM with gamma distribution and inverse function where d.f ¼ degrees of freedom, AIC ¼ Akaike’s information criterion, P ¼ significance level. Biomass model structure Null þCoral cover þActivity Full model
d.f.
AIC
Residual deviance.
2 3 5
9048.2 9041.6 8930.8 8921.8
768.52 753.53 608.01 592.98
Explained deviance %
p
2.0 18.9 20.9
<0.01 <0.001 <0.001
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139
coral reefs. Historic records and anecdotal accounts of higher abundances of E. tukula on the High-Diving reef during the 1980s to early 1990s (Koornhof, 1991; Chater et al., 1995) indicate that a reduction in population numbers has occurred during the last three last decades. The results and observations of this study suggest that this decline may be linked to high diving intensity. E. tukula is a large resident predator known to show aggressive territorial behaviour towards divers on reefs isolated from human activities (Debelius, 2001; Peschak, 2010). All E. tukula encountered on Sanctuary reefs displayed aggressive or curious behaviour towards divers which included open mouth displays, bumping of divers, biting of the buoy-reel and stalking of divers throughout the dive. In contrast, E. tukula behaviour on the Diving-Fishing reef was cautious and the divers were seldom approached. Furthermore, E. tukula were most commonly observed at the edge of diver visibility where they moved from one overhang to another. Thus the persistent presence of SCUBA divers on the High-diving reef may be causing a competitive disturbance to E. tukula. 4.3. Implications for management No-take MPAs carry a high socio-economic cost because of lost fishing grounds (Jones, 2008), but show great benefits and yield significantly great biomass and densities (Gell and Roberts, 2003; Lester and Halpern, 2008). Multiple-use MPAs are typically seen as more feasible and are often implemented as a compromise because they confer some benefits over open access areas (Lester and Halpern, 2008). This study also provided evidence that notake MPA have significantly greater benefits to coral reef fish communities in South Africa; however, the differences between the multiple-use and open zones were considerably smaller than anticipated. The implications for this are significant in terms of the objectives and goals of the different MPA zones. The open zone in southern Mozambique had little legislation and no enforcement
137
protecting the reefs at the time of data collection. In contrast, the South African MPAs have clear objectives outlined in the MPA management plan (Ezemvelo KwaZulu-Natal Management Plan 2007). However, the objectives are identical for all MPA zones, despite differences in the type and intensity of human activity in each zone. Thirteen objectives are relevant to the coral reefs and fall within three functional categories; protection, fisheries management and utilisation (see Floros, 2010 for details). When assessed using the indicator parameters (biomass, densities, trophic composition and reproductive potential), eight objectives were questionable with regard to achieving their management goals. This indicates that the multiple-use zones are not providing protection to the reef fish communities under the current levels of human activity. It also suggests that these zones require a more complex management approach which includes a reassessment of their objectives to balance the trade-off between sustainable resource use and conservation. In contrast, the no-take Sanctuary zones appear to have achieved all management objectives. The prohibition of all human activity in these areas has promoted greater biomass, densities and reproductive potential of the reef fish communities. Considering their long history of closure to human activities, the Sanctuary zones potentially represent undisturbed ecosystems on which future management plans should be based. Acknowledgements We are grateful for the financial support provided by the National Research Foundation and the South African Association for Marine Biological Research. We also thank the staff of the Oceanographic Research Institute for their technical support in the field. The 4x4 vehicle used in this project was sponsored by the Mazda Wildlife Fund. Ezemvelo KwaZulu-Natal Wildlife and the iSimangaliso Wetland Authority are acknowledged for their logistical support.
Appendix A Mean abundance (fish 78 m2) of the Fish-index species and differences between zones shown by one-way ANOVA.
Acanthurus leucosternon Amphiprion allardi Aprion virescens Balistoides conspicillum Bodianus diana Caranx melampygus Chaetodon madagaskariensis Chaetodon meyeri Chaetodon trifascialis Chaetodon trifasciatus Diplodus cervinus Epinephelus tukula Forcipiger flavissimus Labroides dimidiatus Lutjanus bohar Odonus niger Oplegnathus robinsoni Plectroglyphidodon johnstonianus Pomacanthus imperator Pygoplites diacanthus Scarus rubroviolceus Siganus sutor Thalassoma hebraicum Variola louti Zebrasoma desjardini a
Sanctuary
High-Diving
Diving-Fishing
Open
p
0.904 0.220 0.617 0.072 0.502 0.919 0.804 0.885 0.038 0.024 0.019 0.292 0.512 1.507 1.512 2.129 0.502 0.464 0.187 0.005 1.292 0.072 3.024 0.531 0.014
1.200 0.077 0.062 0.046 0.554 0.615 0.754 0.400 0.169 0 0.385 0.077 0.292 2.338 0.138 1.754 0.108 0.369 0.246 0.015 1.569 0.215 2.462 0.185 0.092
1.296 0.576 0.024 0.464 0.464 0.288 0.984 0.144 0 0 0 0.080 0.352 2.704 0.184 4.024 0.104 0.032 0.144 0.016 1.200 0.080 2.576 0.184 0.024
0.353 0.333 0 0.059 0.333 0.098 0.706 0.549 0.137 0.078 0 0 0.294 2.235 0.118 1.137 0.039 0.235 0.039 0 0.725 0.137 2.294 0.098 0
NS <0.001 <0.001 0.007a NS 0.003 NS <0.001 0.007a 0.007a NS <0.001 NS 0.005 <0.001 NS 0.005 <0.001 NS NSa NS NSa NS <0.001 NSa
Indicates species that were inconclusive in the ANOVA or Multiple comparisons procedure due to their uncommon status.
138
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139
Appendix B Mean biomass (kg 78 m2) of the Fish-index species and differences between zones shown by one-way ANOVA.
Acanthurus leucosternon Amphiprion allardi Aprion virescens Balistoides conspicillum Bodianus diana Caranx melampygus Chaetodon madagaskariensis Chaetodon meyeri Chaetodon trifascialis Chaetodon trifasciatus Diplodus cervinus Epinephelus tukula Forcipiger flavissimus Labroides dimidiatus Lutjanus bohar Odonus niger Oplegnathus robinsoni Plectroglyphidodon johnstonianus Pomacanthus imperator Pygoplites diacanthus Scarus rubroviolaceus Siganus sutor Thalassoma hebraicum Variola louti Zebrasoma desjardinii a
Sanctuary
High-Diving
Diving-Fishing
Open
p
0.125 0.004 1.914 0.102 0.045 2.393 0.020 0.058 0.001 0.001 0.005 8.126 0.025 0.010 4.361 1.079 1.197 0.002 0.194 0.000 1.340 0.050 0.242 1.170 0.005
0.145 0.001 0.062 0.029 0.056 0.828 0.046 0.031 0.019 0 0.029 1.075 0.015 0.016 0.407 0.731 0.134 0.002 0.329 0.001 2.571 0.133 0.258 0.490 0.021
0.193 0.016 0.017 0.069 0.038 0.402 0.031 0.017 0 0 0 2.077 0.018 0.025 0.215 1.244 0.190 0.001 0.209 0.001 2.508 0.066 0.314 0.151 0.009
0.036 0.012 0 0.119 0.025 0.092 0.031 0.046 0.019 0.010 0 0 0.018 0.015 0.048 1.000 0.114 0.003 0.058 0 1.243 0.125 0.239 0.215 0
0.047 <0.001 <0.001 NS NS <0.001 NS <0.001 0.011a NSa NSa <0.001 NS 0.004 <0.001 NS 0.04 <0.001 NS NSa <0.001 NSa NS <0.001 NSa
Indicates species that were inconclusive in the ANOVA or Post Hoc comparison due to their uncommon status.
References
Appendix C Trophic guild allocation of the Fish-index species and the rationale for their inclusion in the study. B ¼ benthivore, C ¼ corallivore, H ¼ herbivore, I ¼ invertivore, M-L P ¼ medium-level predator, O ¼ omnivore, P ¼ planktivore, T-L P ¼ top-level predator. Trophic guild
Rationale
Acanthurus leucosternon Amphiprion allardi Aprion virescens
H O M-L P
Balistoides conspicillum Bodianus diana Caranx melampygus
I I T-L P
Chaetodon madagaskariensis Chaetodon meyeri Chaetodon trifascialis Chaetodon trifasciatus Diplodus cervinus Epinephelus tukula Forcipiger flavissimus Labroides dimidiatus Lutjanus bohar Odonus niger Oplegnathus robinsoni Plectroglyphidodon johnstonianus Pomacanthus imperator Pygoplites diacanthus
O C C C I T-L P I I M-L P P B C B B
Scarus rubroviolceus Siganus sutor Thalassoma hebraicum Variola louti Zebrasoma desjardinii
H H I M-L P H
Common herbivore Iconic species Target species and important predator Sensitive to diver presence Common prey species Target species and important predator Generalist feeder Specialist feeder Specialist feeder Specialist feeder Regional endemic Largest reef predator Prey species Specialist feeder Important predator Trophic representative Regional endemic Coral specific species Specialist feeder Specialist feeder and uncommon status Largest reef herbivore Important regional herbivore Common prey species Important reef predator Conspicuous grazer
Agardy, T., Bridgewater, P., Crosby, M.P., Day, J., Dayton, P.K., Kenchington, R., Laffoley, D., McConney, P., Murray, P.A., Parks, J.E., Peau, L., 2003. Dangerous targets? Unresolved issues and ideological clashes around marine protected areas. Aquat. Conserv. 13, 353e367. Alder, J., 1996. Have tropical marine protected areas worked? An initial analysis of their success. Coast Manage. 24, 97e114. Angermeier, P., Karr, J., 1994. Biological integrity versus biological diversity as policy directives. BioScience 44, 690e697. Barker, N.H.L., Roberts, C.M., 2004. Scuba diver behaviour and the management of diving impacts on coral reefs. Biol. Conserv. 120, 481e489. Baskett, M.L., Levin, S.A., Gaines, S.D., Dushoff, J., 2005. Marine reserve design and the evolution of size at maturation in harvested fish. Ecol. Appl. 15, 882e907. Bell, J.D., Kronen, M., Vunisea, A., Nash, W., Keeble, G., Demmkea, A., Pontifex, S., Andréfouët, S., 2009. Planning the use of fish for food security in the Pacific. Mar. Policy 33, 64e76. Bellwood, D.R., Hughes, T.P., Folke, C., Nyström, M., 2004. Confronting the coral reef crisis. Nature 429, 827e833. Berkeley, S.A., Hixon, M.A., Larson, R.J., Love, M.S., 2004. Fisheries sustainability via protection of age structure and spatial distribution of fish populations. Fisheries 29, 23e31. Celliers, L., Schleyer, M.H., 2008. Coral community structure and risk assessment of high-latitude reefs at Sodwana Bay, South Africa. Biodivers. Conserv. 17, 3097e 3117. Cesar, H.S.J., Burke, L., et al., 2003. The Economics of Worldwide Coral Reef Degradation. Cesar Environmental Economics Consulting (CEEC), 6828GH Arnhem, The Netherlands. Chater, S.A., Beckley, L.E., van der Elst, R.P., Garrat, P.A.,1995. Underwater visual census of fishes in the St Lucia Marine Reserve, South Africa. Lammergeyer 43, 15e23. Clarke, K.R., Gorley, R.N., 2006. PRIMER v6: User Manual/Tutorial. PRIMER-E, Plymouth. Cooke, S.J., Cowx, I.G., 2004. The role of recreational fishing in the global fish crisis. BioScience 54, 857e859. Dale, V.H., Beyeler, S.C., 2001. Challenges in the development and use of ecological indicators. Ecol. Indic. 1, 3e10. Debelius, H., 2001. Indian Ocean Reef Guide. IKAN-Unterwasserarchiv, Frankfurt. Dulvy, N.K., Polunin, N.V.C., Mill, A.C., Graham, N.A.J., 2004. Size structural changes in lightly exploited coral reef fish communities: evidence for weak indirect effects. Can J. Fish Aquat. Sci. 61, 466e475. English, S., Wilkinson, C., Baker, V. (Eds.), 1994. Survey Manual for Tropical Marine Resources. Australian Institute of Marine Science, Townsville. Evans, R.D., Russ, G.R., 2004. Larger biomass of targeted reef fish in no-take marine reserves on the Great Barrier Reef, Australia. Aquat. Conserv. 14, 505e519. Floros, C., 2010. An Evaluation of Coral Reef Fish Communities in South African Marine Protected Areas (PhD dissertation). Oceanographic Research Institute, University of KwaZulu-Natal.
C. Floros et al. / Ocean & Coastal Management 84 (2013) 130e139 Floros, C., Schleyer, M.H., Maggs, J.Q., Celliers, L., 2012. Baseline assessment of highlatitude coral reef fish communities in southern Africa. Afri J. Mar Sci. 34, 55e69. Froese, R., Pauly, D., 2011. Fishbase. www.fishbase.org (accessed July 2011). Francini-Filho, R.B., Moura, R.L., 2008. Evidence for spillover of reef fishes from a notake marine reserve: an evaluation using the before-after control-impact (BACI) approach. Fish Res. 93, 346e356. Gell, F.R., Roberts, C.M., 2003. Benefits beyond boundaries: the fishery effects of marine reserves. Trends Ecol. Evol. 18, 448e455. Graham, N.A.J., Evans, R.D., Russ, G.R., 2003. The effects of marine reserve protection on the trophic relationships of reef fishes on the Great Barrier Reef. Environ. Conserv. 30, 200e208. Graham, N.A.J., Dulvy, N.K., Jennings, S., Polunin, N.V.C., 2005. Size-spectra as indicators of the effects of fishing on coral reef fish assemblages. Coral Reefs 24, 118e124. Halpern, B.S., Walbridge, S., Selkoe, K.A., Kappe, C.V., Micheli, F., D’Agrosa, C., Bruno, J.F., Casey, K.S., Ebert, C., Fox, H.E., Fujita, R., Heinemann, D., Lenihan, H.S., Madin, E.M.P., Perry, M.T., Selig, E.R., Spalding, M., Steneck, R.S., Watson, R., 2008. A global map of human impact on marine ecosystems. Science 319, 948e952. Hawkins, J.P., Roberts, C.M., Kooistra, D., Buchan, K., White, S., 2005. Sustainability of SCUBA diving tourism on coral reefs of Saba. Coastal Manage. 33, 373e387. IUCN, 2004. Managing Marine Protected Areas: a Toolkit for the Western Indian Ocean. In: IUCN Eastern African Regional Programme. Nairobi, Kenya. Jameson, S.C., Kelty, R.A., 2004. A review of indicators of land-based pollution stress on coral reefs. In: Joint EPA/NOAA/USGA/DOI Workshop: Assessing Pollution Stress on Coral Reefs Aug 31eSept 2 Honolulu, Hawaii. Jones, P.J.S., 2008. Fishing industry and related perspectives on the issues raised by no-take marine protected area proposals. Mar. Policy 32, 749e758. Jordan, I.E., Samways, M., 2001. Recent changes in the coral composition of a South African coral reef, with recommendations for long-term monitoring. Biodivers. Conserv. 10, 1027e1037. Karr, J.R., 1981. Assessment of biotic integrity using fish communities. Fisheries 6, 21e27. Kelleher, G., Kenchington, R., 1992. Guidelines for Marine Protected Areas. IUCN, Gland, Switzerland. Kelleher, G., Bleakley, C., Wells, S., 1995. A Global Representative System of Marine Protected Areas. The World Bank, Washington. Knowlton, N., Jackson, J.B.C., 2008. Shifting baselines, local impacts, and global change on coral reefs. PLoS ONE 6, 215e220. Koornhof, A., 1991. The Dive Sites of South Africa. Comprehensive Coverage of Diving and Snorkelling. New Holland, Cape Town. Lester, S.E., Halpern, B.S., 2008. Biological responses in marine no-take reserves versus partially protected areas. Mar. Ecol. Prog. Ser. 367, 49e56. Lutjeharms, J.R.E., 2006. The Agulhas Current. Springer, Berlin. Macdonald, A.H.H., Lamb, J., Schleyer, M.H., 2011. Acropora austera connectivity in the south western Indian Ocean assessed using nuclear intron sequence data. Mar Biol. 158, 613e621. Maunder, M.N., Punt, A.E., 2004. Standardizing catch and effort data: a review of recent approaches. Fish Res. 70, 141e159. McClanahan, T.R., Graham, N.A.J., Calnan, J.M., MacNeil, M.A., 2007. Toward pristine biomass: reef fish recovery in coral reef marine protected areas in Kenya. Ecol. Appl. 17, 1055e1067. Mora, C., Andrefouet, S., Costello, M.J., Kranenburg, C., Rollo, A., Vernon, J., Gaston, K.J., Muyers, R., 2006. Coral reefs and the global network of marine protected areas. Science 312, 1750. Pauly, D., Christensen, V., Guénette, S., et al., 2002. Towards sustainability in world fisheries. Nature 418, 689e695. Pereira, H.M., 2003. Recreational Scuba Diving and Reef Conservation in Southern Mozambique (MSc thesis). Oceanographic Research Institute, University of KwaZulu-Natal.
139
Peschak, T.P., 2010. Lost World: the Marine Realm of Aldabra and the Seychelles. JB Publishing. Pieters, M., 2009. Dive Statistics for Sodwana Bay. Ezemvelo KZN Wildlife Authorities. Polunin, N.V.C., Roberts, C.M., 1993. Great biomass and value of target coralreef fishes in two small Caribbean marine reserves. Coral Reefs 100, 167e 176. Pomeroy, R.S., Watson, L.M., Parks, J.E., Cid, G.A., 2005. How is your MPA doing? A methodology for evaluating the management effectiveness of marine protected areas. Ocean. Coast. Manage. 48, 485e502. R Development Core Team, 2011. R: a Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria, ISBN 3900051-07-0. http://www.R-project.org/. Ramsay, P.J., 1996. Quaternary marine geology of the Sodwana Bay continental shelf, Northern KwaZulu-Natal. Bull. Geo. Surv. S. Afr. 117, 1e85. Ridgway, T., Riginos, C., Davis, J., Hoegh-Guldberg, O., 2008. Genetic connectivity patterns of Pocillopora verrucosa in southern African marine protected areas. Mar. Ecol. Prog. Ser. 354, 161e168. Riegl, B., Schleyer, M.H., Cook, P.J., Branch, G.M., 1995. Structure of Africa’s southernmost coral communities. Bull. Mar. Sci. 5, 676e691. Russ, G.R., Alcala, A.C., 1989. Effects of intense fishing pressure on an assemblage of coral reef fishes. Mar. Ecol. Prog. Ser. 56, 13e27. Russ, G.R., 2002. Yet another review of marine reserves as reef fishery management tools. In: Sale, P. (Ed.), Coral Reef Fishes. Academic Press, San Diego, CA, pp. 421e443. Sale, P.F., Cowen, R.K., Danilowicz, B.S., Jones, G.P., Kritzer, J.P., Lindeman, K.C., Planes, S., Polunin, N.V.C., Russ, G.R., Sadovy, Y.J., Steneck, R.S., 2005. Critical science gaps impede use of no-take fishery reserves. Trends Ecol. Evol. 20, 74e80. Samoilys, M.A., Carlos, G., 2000. Determining methods of underwater visual census for estimating the abundance of coral reef fishes. Environ. Biol. Fish 57, 289e304. Schleyer, M.H., 2000. South African coral communities. In: McClanahan, T.R., Sheppard, C., Obura, D. (Eds.), Coral Reefs of the Indian Ocean. Oxford University Press, New York, pp. 83e105. Tupper, M.H., 2007. Spillover of commercially valuable reef fishes from marine protected areas in Guam, Micronesia. Fish. Bull. 105, 527e537. Tratalos, J.A., Austin, T.J., 2001. Impacts of recreational SCUBA diving on coral communities of the Caribbean Island of Grand Cayman. Biol. Conserv. 102, 67e75. Watson, D., Anderson, M.J., Kendrick, G.A., Nardi, K., Harvey, E.S., 2009. Effects of protection from fishing on the lengths of targeted and non-targeted fish species at the Houtman Abrolhos Islands, Western Australia. Mar. Ecol. Prog. Ser. 384, 241e249. Wells, A., Burgess, N., Ngusaru, A., 2007. Towards the 2012 marine protected area target in Eastern Africa. Ocean. Coast. Manage 50, 67e83. Westera, M., Lavery, P., Hyndes, G., 2003. Differences in recreationally targeted fishes between protected and fished areas of a coral reef marine park. J. Exp. Mar. Biol. Ecol. 294, 145e168. White, A.T., Meneses, A.B.T., Ovenden, M., Tesch, S., 2006. Sustaining marine protected areas through continued monitoring and evaluation: the MPA report guide and management rating system. In: Proc 10th Int Coral Reef Symp, Okinawa, Japan, pp. 1466e1470. Wilkinson, C.R., 2008. Status of Coral Reefs of the World: 2008. Global Coral Reef Monitoring Network and Reef and Rainforest Research Centre, Townville, Australia. Zakai, D., Chadwick-Furman, N.E., 2002. Impacts of intensive recreational diving on reef corals at Eilat, northern Red Sea. Biol. Conserv. 105, 179e187.