Hydrobiologia (2017) 792:121–135

DOI 10.1007/s10750-016-3049-3

PRIMARY RESEARCH PAPER

The functional trophic role of whale shark (Rhincodon

typus) in the northern Mexican Caribbean: network analysis
and ecosystem development
Esmeralda C. Ibarra-Garcı́a . Marco Ortiz . Eduardo Rı́os-Jara . Amı́lcar L. Cupul-Magaña .
Álvaro Hernández-Flores . Fabián A. Rodrı́guez-Zaragoza

Received: 26 August 2016 / Revised: 5 November 2016 / Accepted: 7 November 2016 / Published online: 18 November 2016
Ó Springer International Publishing Switzerland 2016

Abstract The whale shark (Rhincodon typus) has a assess the fishery effects on different target species;
high value to tourism, fishing, and pharmaceutical and (4) evaluate the whale shark systemic role in this
industry, but this species is considered as vulnerable system. The results showed that the macroscopic
and with potential extinction. The largest aggregation properties of the Holbox ecosystem were similar to
of whale sharks occurs in an upwelling system other upwelling systems with high rates of primary
(Holbox) at northeastern Mexican Caribbean, where productivity and accumulated biomass in lower
we analyzed the trophic role of whale shark. A mass- trophic levels. Even so, the food web was more like
balanced trophic model was built to: (1) analyze the the structure of coral reefs. On the other hand, Holbox
ecosystem emergent or macroscopic properties; (2) system had a lower maturity and development level
estimate the system recovery time (resilience); (3) than coral reefs, suggesting an increased resistance to
disturbances. The mixed trophic impacts and short-
term dynamic simulation showed that species like
Handling editor: Vasilis Valavanis
snook, sharks, groupers, benthic autotrophs, small
Electronic supplementary material The online version of benthic epifauna, and phytoplankton had the largest
this article (doi:10.1007/s10750-016-3049-3) contains supple- direct and indirect effects. The most significant
mentary material, which is available to authorized users.

E. C. Ibarra-Garcı́a E. C. Ibarra-Garcı́a  E. Rı́os-Jara 

Programa de Maestrı́a en Ciencias en Biosistemática y F. A. Rodrı́guez-Zaragoza (&)
Manejo de Recursos Naturales y Agrı́colas, CUCBA, Laboratorio de Ecosistemas Marinos y Acuicultura
Universidad de Guadalajara, Guadalajara, Mexico (LEMA), Departamento de Ecologı́a, CUCBA,
Universidad de Guadalajara, Camino Ramón Padilla
M. Ortiz Sánchez No. 2100 Nextipac, C.P. 45110 Zapopan, Jalisco,
Laboratorio de Modelamiento de Sistemas Ecológicos Mexico
Complejos (LAMSEC), Instituto de Antofagasta (IA), e-mail: [email protected]
Instituto de Ciencias Naturales AvH, Facultad de Ciencias
del Mar y Recursos Biológicos, Universidad de Á. Hernández-Flores
Antofagasta, P.O. Box 170, Antofagasta, Chile Periférico norte tablaje catastral 13941 Carretera Mérida,
Universidad Marista de Mérida, Progreso,
A. L. Cupul-Magaña C.P. 97300 Mérida, Yucatán, Mexico
Centro de Investigaciones Costeras, Departamento de
Ciencias Biológicas, CUCosta, Universidad de
Guadalajara, Av. Universidad de Guadalajara 203, Del.
Ixtapa, P.O. Box 48280, Puerto Vallarta, Jalisco, Mexico

123
122 Hydrobiologia (2017) 792:121–135

systemic role of whale sharks is to provide the Holbox is a small coastal village where some fishers
resilience of the system. have changed their traditional livelihood to eco-
tourism, as a result of conservation initiatives. Today,
Keywords Rhincodon typus  Mexican Caribbean  ecotourism is the activity that generates the highest
Trophic models  Ecosystem development  Fishery income for locals. Swimming with whale sharks is one
assessment of the major attractions for tourists; this activity
recorded five million visitors in 2000, an amount that
was doubled by 2008 (Ardisson et al., 2011). Although
ecotourism is not an extractive activity, it could be
Introduction causing negative impacts on the ecosystem due to
interactions of tourism with whale sharks, together
The largest aggregation of whale shark (Rhincodon with the effects of fishing on other species. Therefore,
typus Smith, 1828) in the world occurs in the north of it is very important to know the degree of impact that
the Mexican Caribbean from Holbox to Contoy human activities could be producing on the Whale
Island, and includes the Marine Protected Area Shark Biosphere Reserve, as interventions propagate
Yum-Balam and the Whale Shark Biosphere Reserve changes in species abundance and community com-
(de la Parra et al., 2011). The greater abundance of position, modifying the structure and dynamics of the
whale sharks occurs from May to September, near food webs (Coll, 2002).
Cabo Catoche, the northernmost point of the Yucatan Multispecies models of marine ecosystems (e.g.,
Peninsula. This aggregation coincides with an Ecopath) have the advantage of integrating ecological
increase in primary productivity caused by an and biological information, and represent the interac-
upwelling that occurs when the Yucatan stream hits tions between suitably chosen functional groups. The
the continental slope and emerges as a water mass, dynamical versions of these models can simulate the
producing a high saturation of nutrients (Merino, impact of certain interventions and, thus, provide
1997; Cárdenas et al., 2010; Ramı́rez et al., 2012). As relevant information about the potential development
a result, this part of the continental shelf becomes a of management strategies. Ecosystem models require
zone of high biological activity with an abundances inputs of biomass, diets, consumption and productivity
of diet items for R. typus, such as fish eggs, larvae of rates, fishing landings, as well as identification of
gastropods, sergestid, chaetognaths, appendiculari- energy flow (Olson & Waters, 2003; Ortiz et al.,
ans, and squid (Compagno, 1984, 1998; Cárdenas 2009). Once calibrated, an ecosystem model, such as
et al., 2010). Ecopath, is able to estimate holistic macrodescriptors
Whale shark is a pelagic-oceanic species, widely that assess the ecosystem structure and organization.
distributed in tropical and subtropical seas; it is The theoretical framework of Ecopath with Ecosim
considered an opportunistic species as it can change (EwE) analyzes the food web structure and its trophic
its preferred preys to alternative preys, according to functioning, considering the mass balance steady-state
availability (Hacohen, 2007; Stevens, 2007). More- assumption (Christensen & Walters, 2005). The
over, this is the largest species of sharks in the world, effects of resource-exploitation regimes and the
and has a slow growth and late maturity, characteris- resulting environmental changes can be assessed;
tics that make it a species with low resilience to therefore, the sustainability of different management
overfishing (Coleman, 1997). For this reason, its strategies can be explored (Christensen & Walters,
exploitation for human consumption and for drugs 2004). In addition, Ecopath has a subroutine that
production has reduced their populations. In 2002, R. undertakes changes in biomass and consumption of
typus was classified in Appendix II of CITES, which functional groups by means of seasonal dynamic
means that the species is not necessarily now threat- simulation (Walters & Christensen, 2007). Based on
ened with extinction but it may become so, unless its these calculations, it is possible to assess the resilience
trade is closely controlled (https://cites.org) and in of the system and the propagation of direct and indirect
2010, the species was listed as Vulnerable by IUCN effects through the food web generated by distur-
(www.iucn.org); both actions contribute to control of bances on a particular species or functional groups
its international trade (de la Parra et al., 2011). (Ortiz et al., 2010).

123
Hydrobiologia (2017) 792:121–135 123

The overall objective of this article is to assess the snappers (Lutjanidae), hoghfish (Lachnolaimus max-
trophic role of the whale shark, and determine the imus (Walbaum, 1792)), and other pelagic and reef
trophic functioning and development of the marine fish.
ecosystem where this species aggregates. Thus, a
trophic model was constructed, which included the Ecopath with Ecosim
whale shark as a specific group in the food web. The
systemic descriptors obtained with the model were as Ecopath with Ecosim (EwE) analyzes the flow of
follows: (i) distribution of biomass and structure of the matter and energy by means of system network
flow in the ecosystem; (ii) the most important predator analysis and calculates the macrodescriptors: total
consumption rates and dietary factors matrix; (iii) system throughput (T), ascendency (A), development
maturity of the ecosystems and its flow organization; capacity (C), overhead (Ov), and relative ascendency
and (iv) the ecological resilience measured as its (A/C) (Ulanowicz, 1986, 1997). T measures the system
degree of resistance against perturbations (different size and its metabolism; A describes the system flows,
regimes of fishing exploitation), taking into account their structures, and magnitude depending on the
three different flow control mechanisms (top-down, number and diversity of interactions between compo-
mixed, and bottom-up). The results of this type of nents, and C defines the upper limit of ascendency.
analysis are appropriate for the design of ecosystem- The A/C ratio is the maturity index that describes the
based management strategies. highest degree of specialization achieved by the
system. Ov represents the energy required by the
ecosystem to maintain homeostasis and its ability to
Materials and methods resist perturbations (Supplementary Appendix A).
Finally, Finn cycling index (FCI) was calculated to
Area of study determine the system’s maturity; this index is based on
the fact that mature ecosystems have more energy and
In this article, we refer to the area of the whale shark matter cycled, so their trophic pathways are extended
aggregation as ‘‘Holbox,’’ located in the northern (Odum, 1969).
Mexican Caribbean Sea (Fig. 1), which covers both
the Whale Shark Biosphere Reserve and the Marine Model design and data sources
Protected Area ‘‘Yum Balam,’’ a total surface of *2
577 km2. This site is characterized by high produc- The trophic model was built with 28 functional groups
tivity as a result of extreme variations of temperatures and eight individual species (n = 36). The functional
(from 17 to 30°C) and the convergence of the Gulf of groups of fish were based on their habits (pelagic,
Mexico and Caribbean Sea currents, which produce a benthic or associated to coral reefs), trophic groups,
stationary upwelling at this point, from May to diet, caudal fin shape and size, maximum-minimum
September. This upwelling enhances the concentra- length and weight (Randall, 1963; Opitz, 1991).
tion of nutrients that produce a high biological activity Species of commercial and tourist interest were
(Merino, 1983, 1997). considered separately to analyze their functional role
In Holbox, the continental shelf is wide and shallow in the ecosystem. Biomass (B) for fish, corals, sea
(10–20 m); rocks cover most of the seabed and there grass, and macroalgae were obtained from field
are no widespread hard coral reef formations, as seen observations. Field surveys were carried out from
in other parts of the Mesoamerican Reef System. This April of 2006 in five locations identified with local
is the result of high primary production, turbidity, and names as Corsario, Punta Mosquito, Piedra Corrida,
temperature variations, all produced by upwelling. Boca Nueva, and Cuevones. At these sites, 60 visual
Instead, the soft corals, fleshy macroalgae, and census were performed for the quantification of fish
seagrass beds are the most common components of (50 9 2 m) and invertebrates (20 9 5 m), and 30
this ecosystem, providing most of the structures for video-transects to record benthos (50 9 0.4 m).
benthic habitat. The target species are lobster (Pan- Biomass of dolphins, birds, snook, shark, some
ulirus argus (Latreille, 1804)), octopus (Octopus maya carnivorous fish, octopus, whale shark, spotted seatr-
(Voss y Solis 1966)), groupers (Epinephelus spp.), out, Manta birostris, schooling fish, lobster, turtles,

123
124 Hydrobiologia (2017) 792:121–135

Fig. 1 The study area includes Whale Shark Biosphere Reserve and Yum-Balam Flora and Fauna Protected Area at Holbox, Quintana
Roo, Mexico. Codes of sampling sites: Corsario (Co), Punta Mosquito (Pm), Cuevones (Cu), Piedra Corrida (Pc), and Boca Nueva (Bn)

and manatee were taken from published literature, or et al., 1993; Criales-Hernández et al., 2006). In order
well, estimated from fishing mortality and catches. to reduce the predation mortality of some preys, the
Other input parameters as production (P), consump- ratio of food elements of some functional groups was
tion (Q) and diet items were taken from the scientific also adjusted (Supplementary Appendix C). Those
literature, while catches (Y) were obtained from groups whose intake ratios were modified are as
official statistics published by the federal agency follows: small benthic epifauna, soft corals, sea
SAGARPA (Secretariat of Agriculture, Livestock, urchins, medium carnivorous reef fish 1 and 2, large
Rural Development, Fisheries and Food). Supplemen- carnivorous reef fish, snappers, large pelagic fish,
tary Appendix B shows the functional groups and their grunts, shark, and R. typus.
corresponding species (Supplementary Appendix
Table B1), as well as data sources used to build the Propagation of direct and indirect effects through
Ecopath model (Supplementary Appendix Table B2). the network

Model balance The mixed trophic impacts (MTI) is an Ecopath

subroutine that analyzes the direct and indirect trophic
In order to have a balanced model, the parameters of relationships between functional groups, and assesses
various functional groups were adjusted, considering a the impact of increased biomass of one group over the
range of one standard deviation. The new parameters other functional groups and the potential cascading
reached values within the ranges reported by other effects within the ecosystem (Ulanowicz & Puccia,
authors for similar ecosystems (Arreguı́n-Sánchez 1990). Short-term dynamic simulations of Ecosim

123
Hydrobiologia (2017) 792:121–135 125

were used to assess the propagation of direct and functional groups. On the other hand, the assessment
indirect effects through the network (Supplementary of fishing exploitation showed that the mean trophic
Appendix A). Ecosim took into consideration the level of catch was 3.3, with secondary and tertiary
criteria of Ortiz et al. (2009) with the following flow consumers as the species with the highest exploitation,
controls: bottom-up (v = 1.0), mixed (v = 2.0), and such as snook, octopus, and hogfish. However, the
top-down (v = 6.0). The purpose of these criteria was species with highest fishing mortality (F) were as
(i) to test scenarios with different flow control follows: snook (0.84), spotted seatrout (0.56), and
perturbations; (ii) to assess the propagation of instan- groupers (0.10).
taneous direct and indirect effects; and (iii) to Network Analysis results showed an ascendency
calculate the system recovery time (SRT), in response (A) equal to 102,343.9 t km-2 year-1, overhead (Ov)
to increased total mortality (Z) (i.e., Z = M ? F, 143,437.5 t km-2 year-1, development capacity
where Z is total mortality, M is natural mortality and (C) 245,781.3 t km-2 year-1, ascendency/capacity
F is fishing mortality) equivalent to 30% of total (A/C) 41.6%, overhead/capacity (Ov/C) 58.4%. The
production (P), where P = B * Z. The impact was average mutual information (AMI) was 1.29 bits and
caused between the second and the third year of the the pathway redundancy (of overhead) was 42.7%.
simulation. Consequently, the instant effects propa- Table 2 shows the most important trophic macrode-
gation was calculated as the change in biomass of all scriptors and flow indexes.
functional groups in the third year, i.e., one or two From the MTI results, the groups and species that
years after increased fishing mortality occurs. This propagate the major number of direct and indirect
was useful to estimate and analyze the resistance of the effects with highest magnitude through the food web
system and the SRT and, thus, providing an internal are as follows: sea turtles, groupers, spotted seatrout,
measure of stability of the system (Ortiz et al., 2009). octopus, snook and shark. In contrast, the manatee
For details see Supplementary Appendix A. (Trichechus manatus Linnaeus, 1758) and M. birostris
(Walbaum, 1792) had less important effects on the
other elements of the system (Fig. 3; Supplementary
Results Appendix D). With respect to the three types of flow
controls, Ecosim simulations showed that benthic
Trophic structure autotrophs, small benthic epifauna, phytoplankton,
large pelagic fish, snappers, medium carnivorous reef
Benthic autotrophs and soft corals were the most fish 1, and grunts produced the highest impacts over
important groups in terms of biomass and ascendency the other species and functional groups (Fig. 4).
in the ecosystem. Higher trophic levels were top Table 3 shows estimates of system recovery time
predators such as large pelagic fish (4.80), dolphins (SRT) only for two flow controls (bottom-up and
(4.02), and seabirds (4.02) (Table 1). Primary produc- mixed) since the system was not able to achieve the
ers were symbiotic algae, phytoplankton and benthic steady state under the top-down flow control. From
autotrophs (fleshy macroalgae and seagrass bed) this analysis, we observe that benthic autotrophs,
(Fig. 2). Throughput (T) was 79,363.9 t km-2 year-1, small benthic epifauna, zooplankton, and phytoplank-
of which 18.03% was for consumption, 29.8% for ton generated the highest magnitudes of SRT, affect-
exports, 11.94% respiration flows, and 40.16% went ing ecosystem resilience. The SRT average value for
into detritus. Total net primary production (P) was bottom-up scenario was 8.7 years, and for mixed
35,142.9 t km-2 year-1, respiration (R) 9,478.1 t km-2 control flow was 31.5 years.
year-1, and the sum of all consumption (Q) 14,315.2 t
km-2 year-1. The total primary production/total Functional role of R. typus
respiration (PP/R) was 3.5, which is evidence that
the system has a eutrophic behavior, typical of an Rhincodon typus reached a trophic level of 3.13, close
upwelling system. Regarding the food web structure, to that obtained by M. birostris (3.03) which feeds also
the connectance index (CI) reached 0.19, omnivore on zooplankton. MTI results showed that R. typus
index (OI) 0.15, the number of trophic pathways was propagated negative effects on hogfish, large pelagic
20,881 with an average path length (APL) was 11.5 fish, and seabirds, and positive effects on snappers,

123
126 Hydrobiologia (2017) 792:121–135

Table 1 Input parameters Group code TL HA BHA (t/km2) B (t/km2) P/B Q/B EE P/Q F (t/km2)
(in bold) and values
estimated by Ecopath for LPF 4.08 1 3.5 3.5 1.95 9.82 0.97 0.2 0.001
Holbox
Do 4.02 1 0.02 0.02 0.1 33.5 0.66 0
SB 4.02 0.43 0.02 0.01 5.4 80 0.29 0.07
Sk 3.97 1 0.02 0.02 2.4 8.2 0.48 0.29 0.0192
Sh 3.64 1 1.3 1.3 1.86 8.17 0.94 0.23 0.0032
LPisF 3.62 1 2.78 2.78 0.9 7.65 0.96 0.12 0.0013
Sn 3.39 0.43 55.72 24.15 0.58 7.34 0.95 0.08 0.0023
Gn 3.28 0.43 0.72 0.31 0.53 5.09 0.94 0.1 0.0314
LCRF 3.21 0.43 90.01 39.02 0.32 3.04 0.93 0.11 0.0033
Gr 3.15 0.43 23.1 10.01 1.87 10.79 0.95 0.17 0.0054
Oc 3.14 1 0.46 0.46 2.3 8 0.99 0.29 0.0288
Ho 3.13 0.43 3.4 1.47 0.56 4.78 0.95 0.12 0.0136
Rt 3.13 1 6.33 6.33 0.07 9.8 0 0.01
SCRF 3.07 0.43 0.07 0.03 1.85 19.55 0.93 0.09
SS 3.07 1 0.01 0.01 1.4 5.9 0.92 0.24 0.0057
1MCRF 3.06 0.43 32.5 14.09 1.84 11.14 0.99 0.16 0.0003
Mb 3.04 1 0.01 0.01 0.41 8.6 0 0.05
SF 3 0.43 6.5 2.82 19.44 65.1 0.94 0.3
2MCRF 2.8 0.43 42.5 18.42 0.79 7.23 0.96 0.11
Lo 2.78 1 5.21 5.21 0.96 7.4 1 0.13 0.0024
ST 2.69 1 0.04 0.04 0.3 3.5 0.99 0.09
OF 2.44 0.43 0.25 0.11 1.65 31.48 0.81 0.05
LBE 2.38 1 55.11 55.11 1.1 3.7 0.98 0.3
SBE 2.13 1 391.5 391.5 1.9 15.02 0.94 0.13
SU 2.11 0.43 26.3 11.4 1.1 3.8 0.98 0.29
StC 2.1 0.43 4.7 2.04 1.09 9 0.98 0.12
Ma 2 0.43 0 0 0.11 39.71 0 0
Functional group codes are Spo 2 0.43 19.76 8.57 1.4 5.2 0.99 0.27
shown in Fig. 2 and
Supplementary Appendix Zoo 2 1 9.68 9.68 40 133.8 0.91 0.3
A, Table A1 HFSM 2.02 1 46.5 46.5 1.05 22.58 0.98 0.05
TL trophic level, HA habitat SoC 2.01 0.43 1105.35 479.19 1.09 9 0.95 0.12
area, BHA biomass at LHF 2 0.43 49.12 21.29 1.15 18.92 0.98 0.06
habitat, B biomass, P/B
BA 1 0.43 5421.24 2350.19 13.25 0 0.08
production, Q/B
consumption, EE ecotrophic Phy 1 1 28.81 28.81 70 0 0.91
efficiency, P/Q SA 1 0.43 2.75 1.19 10.2 0 0.98
production/consumption De 1 1 100 100 0.26
rate and F catches

groupers, and omnivore fish. On the other hand, the scenario, and 9 years under a bottom-up control
groups that produced higher negative direct and scenario (Table 3; Supplementary Appendix D).
indirect effects on R. typus were soft corals, small
benthic epifauna, and medium carnivorous reef fish 1.
In contrast, the groups that enhanced positive effects Discussion
on whale shark were zooplankton, phytoplankton and
detritus. Ecosim’s simulations showed that when Regarding to biomass and energy flow, the most
whale sharks are disturbed, resilience of the system important group consisted of benthic autotrophs,
is equivalent to 55 years of SRT under a mixed control besides the great impact that this group has on the

123
Hydrobiologia (2017) 792:121–135 127

Fig. 2 Food web topology for Holbox ecosystem. Vertical reef fish 1 (MCRF1), (17) Manta birostris (Mb), (18) schooling
position approximates trophic level (numbers on the left). The fish (SF), (19) medium carnivorous reef fish 2 (MCRF2), (20)
circle size is proportional to the compartment (populations and/ lobster (Lo), (21) sea turtles (ST), (22) omnivorous fish (OF),
or functional groups) biomass. The code in the circle (23) large benthic epifauna (LBE), (24) small benthic epifauna
corresponds to the species or functional groups: (1) large (SBE), (25) sea urchins (SU), (26) stony corals (StC), (27)
pelagic fish (LPF), (2) dolphin (Do), (3) seabirds (SB), (4) snook manatee (Ma), (28) sponges (Spo), (29) zooplankton (Zoo), (30)
(Sk), (5) shark (Sh), (6) large piscivorous fish (LPisF), (7) herbivorous fish small-medium (HFSM), (31) soft corals (SoC),
snappers (Sn), (8) groupers (Gp), (9) large carnivorous reef fish (32) large herbivorous fish (LHF), (33) benthic autotrophs (BA),
(LCRF), (10) grunta (Gr), (11) octopus (Oc), (12) hogfish (Ho), (34) phytoplankton (Phy), (35) symbiotic algae (SA), (36)
(13) Rhincodon typus (Rt), (14) small carnivorous reef fish detritus (De)
(SCRF), (15) spotted seatrout (SS), (16) medium carnivorous

resilience of the system. The high biomass of this producers, the soft corals find suitable conditions to
group is explained by the effect of the stationary settle down in the area. Because soft corals are sessile
upwelling from May to September which provides creatures that require planktonic organisms and
large amounts of nutrients, favoring the growth of organic matter in the water column, the upwelling
algae and seagrasses (Merino, 1997). In addition, a makes large amounts of food available for them.
large area covered by hard substrate allows the Therefore, the large area of hard substrate in Holbox
establishment of different algae (e.g., fleshy macroal- allows the establishment of these extensive gardens of
gae, incrusting calcareous algae, filament algae, and soft corals. Soft coral colonies are autogenous engi-
other). On the other hand, the wide sandy substrate neers and they play an important role in the ecosystem
surface favors the establishment of extensive seagrass by providing shelters to other fish and invertebrates
beds (i.e., Thalassia testudinum König, Syringodium (Arias-González et al., 2008; Rodrı́guez-Zaragoza
filiforme Kützing). For these reasons, this functional et al., 2011).
group plays a key role in primary productivity, The total biomass calculated for Holbox was lower
becoming the major food supplier for the primary than that reported for other coral reef systems in the
consumers and herbivores. Other important features of Caribbean Sea, like Cayos Cochinos, Honduras
this group are that several fish and invertebrates use (Cáceres et al., 2016) and Chinchorro Bank, Mexico
them as substrate, spawning and nursing sites (Rodrı́guez-Zaragoza et al., 2016). However, it was
(Ramı́rez, 1995). higher than that reported for Boca Paila, Talampam,
The second group that contributed with biomass and Mahahual in the Mexican Caribbean Sea (Arias-
and energy to the system was the soft corals. This González et al., 2004), Media Luna, Honduras
group provided significant amounts of biomass and (Cáceres et al., 2016) and the upwelling systems of
energy through the food web. Similar to primary Bahı́a Tongoy, Chile (Ortiz & Wolff, 2002 and La

123
128 Hydrobiologia (2017) 792:121–135

Table 2 Summary Parameter Value Units

statistics after mass balance
process by Ecopath and Sum of all consumption (Q) 14,315.2 t km-2 year-1
Network Analysis flow
Sum of all exports (E) 23,690.7 t km-2 year-1
indices
Sum of all respiratory flows (R) 9478.1 t km-2 year-1
Sum of all flows into detritus 31,879.7 t km-2 year-1
Total system throughput (T) 79,363.9 t km-2 year-1
Sum of all production (P) 35,142.9 t km-2 year-1
Mean trophic level of the catch 3.3 t km-2 year-1
Gross efficiency (catch/net p. p.) 3.55E-06 t km-2 year-1
Calculated total net primary production (PPN) 33,168.98 t km-2 year-1
Total primary production/total respiration (PP/R) 3.5 t km-2 year-1
Net system production 23,690.79 t km-2 year-1
Total primary production/total biomass 9.3
Total biomass/total throughput 0.04
Total biomass (excluding detritus) (B) 3535.59 t/km2
Total catch 0.11 t km-2 year-1
Connectance index (CI) 0.19
Omnivory index (OI) 0.15
Ascendency (A) 102,343.9 Flowbites
Overhead (Ov) 143,437.5 Flowbites
Development capacity (C) 245,781.3 Flowbites
Average mutual information (AMI) 1.29 Bits
Ascendency/capacity (A/C) 41.6 %
Pathway redundancy (of overhead) 42.7 %
Overhead/capacity (Ov/C) 58.4 %

Guajira, Colombia (Criales-Hernández et al., 2006). (e.g., Cáceres et al., 2016, Rodrı́guez-Zaragoza et al.,
According to Pauly and Christensen (1995), the 2016). Holbox model also has a higher total produc-
energy transfer of primary producers in upwelling is tion than those estimated in other upwelling systems.
*10%, which coincides with our study since Holbox For example, La Guajira, Colombia, is a tropical
has a transfer of 10.7%. The main reason that the flow upwelling with large areas of seagrass beds (Criales-
control type ‘‘bottom-up’’ prevails in Holbox is due to Hernández et al., 2006). On the other hand, total
the high productivity of phytoplankton and benthic production and PP/R ratio are similar to those
autotrophs. In general, primary producers provide estimated for temperate ecosystems with upwelling
41.9% of the flow toward higher trophic levels, a in Chile, such as the Mejillones Peninsula that has kelp
phenomenon that intensifies during the upwelling forests (Ortiz, 2008, 2010). However, in Chile there
from May to September, and is compatible with the are other marine ecosystems with lower production
large biomass of R. typus for this time of the year (de la than Holbox, as the Marine Reserve of La Rinconada
Parra et al., 2011; Ramı́rez et al., 2012). (Ortiz et al., 2010). But when comparing the results of
The trophic functioning in Holbox system is similar this work with coral reefs from the Caribbean Sea,
to others located in areas of upwelling, which are Holbox has higher production and PP/R than: (i) Tam-
characterized by high rates of primary productivity, palam and Mahahual, Mexico (Arias-González et al.,
accumulation of biomass at lower trophic levels and a 2004); and (ii) Media Luna reef system, Honduras
total primary production/total respiration ratio greater (Cáceres et al., 2016). In contrast, other coral reef
than 1.0. In contrast, the structure of the food web in systems have higher values than Holbox: (i) Chin-
Holbox is more like the structure of coral reef systems chorro Bank, Mexico (Rodrı́guez-Zaragoza et al.,

123
Hydrobiologia (2017) 792:121–135 129

Fig. 3 Mixed trophic impacts (MTI) (direct and indirect) as response to the main impacting functional groups for Holbox ecosystem.
Numbers on the x-axis correspond to the species or functional groups of Fig. 2

2016); and (ii) Cayos Cochinos, Honduras (Cáceres Caribbean Sea, considered more mature and well-
et al., 2016). developed ecosystems (Alvarez-Hernandez, 2003;
Given the FCI, Holbox has an intermediate value Cáceres et al., 2016; Rodrı́guez-Zaragoza et al.,
when compared with upwelling systems in northern 2016). These comparisons suggest that the food web
Chile (Medina et al., 2007), La Guajira, Colombia and structure in Holbox is less linear than observed in coral
coral reefs from the Mexican Caribbean coast (Arias- reefs and similar to food webs present in upwelling
González & Morand, 2006), Chinchorro Bank car- ecosystems. This coincides with Odum’s observation
bonate platform coral reef (Rodrı́guez-Zaragoza et al., (1971), who mentions that as ecosystems reach
2016) and coral reefs of the Honduran Caribbean, as maturity, their trophic structures change from food
Media Luna and Cayos Cochinos (Cáceres et al., chain to food web. Nevertheless, Holbox has a large
2016). Regarding the omnivory index, Holbox had a number of trophic pathways and a larger average path
lower value than the upwelling of La Guajira (Criales- length, as in coral reef ecosystems. Therefore, it is
Hernández et al., 2006) and other coral reefs from the considered that Holbox’s trophic functioning is similar
Mesoamerican Barrier Reef System (MBRS) (Cáceres to temperate upwelling systems, but with food webs
et al., 2016; Rodrı́guez-Zaragoza et al., 2016). This similar to those of coral reef ecosystems.
means that the functional groups and species that make The results of Network Analysis show that
up the Holbox system have more specialized diets. Throughput (T) in Holbox is higher than in many
After analyzing the structure of the food web in other coral reef systems, like Boca Paila, Tampalam,
Holbox, it is appreciated that the connectance index is Mahahual (Arias-González et al., 2004), and Media
similar to those estimated in temperate climates, as in Luna (Cáceres et al., 2016). Although two coral reef
Tongoy Bay, Chile (Ortiz & Wolff, 2002), the marine ecosystems had T values higher than Holbox: Chin-
reserve of La Rinconada (Ortiz et al., 2010) and the chorro Bank (Rodrı́guez-Zaragoza et al., 2016) and
Santa Maria Island in the Mejillones Peninsula (Ortiz, Cayos Cochinos (Cáceres et al., 2016). High T values
2010). By contrast, the value of connectance index in (e.g., Chinchorro Bank and Cayos Cochinos) are the
Holbox was lower than other coral reefs in the result of great total biomass provided by benthic

123
130 Hydrobiologia (2017) 792:121–135

123
 Hydrobiologia (2017) 792:121–135 131

b Fig. 4 Dynamical responses of the main impacting functional autotrophs, while lower T values are because the food
group subject to 1 year of increased fishing mortality (F) (be- webs have less biomass throughout their components.
tween year 1 and 2 of the simulation) under three flow-control
mechanisms using Ecosim. The biomass responses were On the other hand, Holbox had higher T values than
obtained for the third year of the simulation. Numbers on the other upwelling systems, like Tongoy Bay (Ortiz &
x-axis correspond to the species or functional groups of Fig. 2 Wolff, 2002) and the northern of Chile (Medina et al.,

Table 3 Summary of the Harvest on Bottom-up (v = 1.0) Mixed (v = 2.0) Top-down (v = 6.0)
system recovery time (SRT)
(years) for Holbox, Flow control mechanisms
ecosystem using bottom-up,
LPF 9 54 Oscillatory
mixed, and top-down flow-
control mechanisms (v) Do 7 51 Oscillatory
SB 3 10 Oscillatory
Sk 5 6 Oscillatory
Sh 6 18 Oscillatory
LPisF 9 44 Oscillatory
Sn 9 38 Oscillatory
Gp 7 20 Oscillatory
LCRF 10 53 Oscillatory
Gr 8 25 Oscillatory
Oc 5 8 Oscillatory
Ho 7 14 Oscillatory
Rt 9 55 Oscillatory
SCRF 5 6 Oscillatory
SS 6 10 Oscillatory
1MCRF 9 39 Oscillatory
Mb 8 15 Oscillatory
SF 10 33 Oscillatory
2MCRF 10 43 Oscillatory
Lo 8 19 Oscillatory
ST 8 24 Oscillatory
OF 5 8 Oscillatory
LBE 11 44 Oscillatory
SBE 13 62 Oscillatory
SU 8 13 Oscillatory
StC 6 23 Oscillatory
Ma 7 43 Oscillatory
Spo 10 29 Oscillatory
Zoo 13 58 Oscillatory
HFSM 13 38 Oscillatory
SoC 11 48 Oscillatory
LHF 13 18 Oscillatory
BA 16 62 Oscillatory
Phy 14 57 Oscillatory
SA 9 17 Oscillatory
Highest values are in bold

123
132 Hydrobiologia (2017) 792:121–135

2007). These differences may be to Holbox’s location decreased, fishing effort targets the lower trophic
in an ecotone between the MBRS (tropical coral reefs levels (Pauly et al., 1998).
systems) and the Gulf of Mexico, with a strong Fishing mortality estimated in Holbox is lower than
influence of the upwelling. It is important to mention natural mortality due to predation, indicating a
that in the Tongoy Bay and northern Chile models, reduced probability of fisheries depletion (Jarre-
benthic and pelagic communities were analyzed Teichman & Christensen, 1998). Nevertheless, fishing
separately, but not the flows of energy between the mortality values of snook, spotted seatrout, and
two components, while the Holbox model took into groupers suggest high exploitation rates. These fish
account the flows of energy between different com- species are functionally important, since they are
ponents like benthic groups, coral reef, pelagic, and tertiary consumers that have a top-down effect in the
others, such as birds and marine mammals. food web, contributing to a steady flow of energy from
Based on the A/C ratio, this study shows that the lower to the higher trophic levels in the system
Holbox is a less mature system than coral reefs of (Branco, 1999). For that reason, commercial and
Chinchorro Bank in the Mexican Caribbean (Rodrı́- recreational fishing in Holbox should be subject to a
guez-Zaragoza et al., 2016), and the coral reef of the management plan that promotes sustainability, since it
Honduran Caribbean Cayos Cochinos (Cáceres et al., has been reported that fishing in Holbox has exceeded
2016). By contrast, Holbox is more developed than the limits of maximum yield capacity of populations of
other upwelling systems, being 19.1% higher than that some targeted species. This has led to a decline of
of La Guajira, Colombia (Criales-Hernández et al., fishery resources, contributing to an increase in tourist
2006), 21.3% higher than the Tongoy Bay (Ortiz & activity by fishermen and other users in the area (Salas
Wolff, 2002) and 6.09% higher than Santa Maria et al., 2006).
Island in the Mejillones Peninsula, Chile (Ortiz, 2010). Moreover, the MTI values show that functional
These are interesting findings, since Holbox is located groups, such as sea turtles, groupers, spotted seatrout,
in the upwelling zone of the Mesoamerican Barrier octopus, snook, and shark, are secondary and tertiary
Reef System (Signoret et al., 1998; Kramer & Kramer, consumers those contribute to the major propagation
2002). of direct and indirect effects through the food web.
On the other hand, Holbox has lower ascendency Some of these species are subject to heavy fishing
than all other coral reef systems, but higher than the exploitation, particularly the grouper and the spotted
upwelling systems. This is due to the high efficiency of sea trout that have the highest fishing mortalities.
energy transfer in Holbox, enhanced by the high These results do not match to those reported by
presence of primary producers and consumers. Based Arreguı́n-Sánchez et al. (1993) for the north of
on indicators Ov/C, pathway redundancy and AMI, it Yucatan Peninsula, where groups that produce higher
is possible to predict that the system of Holbox is still propagation effects are producers and primary con-
in development, with an increased resistance to sumers. The difference could be due to the fact that the
disturbances as a result of high energy storage capacity upwelling has less influence on distant ecosystems to
due to the upwelling, which other coral reefs do not the northwest.
have. Regarding to Ecosim simulations and system
The mean trophic level of the catch indicates that recovery time, both analyses agree that the most
fishing Holbox are targeting predators such as snook, important functional groups are benthic autotrophs,
groupers, octopus, and hogfish. This type of fishing small benthic epifauna, phytoplankton and zooplank-
inflicts a high cost of energy in the ecosystem, because ton. This is due to these groups’ contribution to total
the energy normally transferred from intermediate biomass, throughput, and the great connectivity that
trophic levels to higher levels decreases due to the they have with the rest of the functional groups. The
respiration rate and a decrease in the efficiency of predominance of these groups is a result of the
energy transfer (Christensen & Pauly, 1993). This stationary upwelling and the subsequent high primary
suggests that perhaps Holbox does not have an productivity.
important effect of fishing. It has been observed that Concerning the whale shark, its trophic level was
when the higher trophic levels are over-exploited and very similar to M. birostris, which exploits same
the average trophic level of the food web has resources and contributes with 0.07% of the

123
Hydrobiologia (2017) 792:121–135 133

throughput. MTI results show that the zooplankton diversity in the region, increasing the complexity of
and phytoplankton were the groups that caused the food webs.
greatest positive impact on the whale shark, as a result Holbox is a system that has trophic characteristics
of its feeding habits. In Holbox ecosystem, the model from both types of ecosystems: coral reefs and
indicates that a disturbance in the whale shark upwelling ecosystems. In terms of ascendancy, Hol-
population would affect the system resilience, this box is more mature than upwelling systems, but less
being a significant result, since this species has direct developed than coral reef ecosystems. This suggests
and indirect connections with the functional groups that due to its energy storage, Holbox is more resistant
with economic importance and therefore, this would to disturbances than other coral reef systems from the
propagate negative effects especially on the large region.
piscivorous fish, snappers, grunts, benthic autotrophs, Functional groups exploited in Holbox by fisher-
and zooplankton. men have an important trophic role in the ecosystem;
This work shows that it is important to complement so the fisheries management should be based more on
population dynamical studies by an ecosystem comprehensive analysis with ecosystem approach to
approach, the assessment of multispecies fisheries, ensure their preservation. Adaptive management
and its direct and indirect effects on the different strategies in a transient dynamics are very important
components of the system. After observing the results to exploit marine resources under more appropriate
of Ecosim simulations, it is clear that overfishing schemes to maintain the resilience and resistance of
affects the population of whale sharks and other the natural ecosystem. In order to complete this
species important to the local economy, reducing their quantitative analysis, we suggest develop models that
populations, causing structural changes into the food include other types of ecological interactions, such as
web, especially in the ecosystem functioning. Finally, competition, commensalism, mutualism, and human
a detailed assessment of the impact of tourism interactions like tourism and economic variables of the
activities on the whale shark is needed to complement fishery resources. This would translate into the basis
the results of this study, considering that according to for designing management strategies for sustainable
the dynamic simulations and the SRT, an impact on use of resources in Holbox.
the biomass of R. typus has strong consequences on the
structure of the ecosystem in Holbox. Acknowledgements The authors thank the support of World
Wildlife Fund (WWF) and the Mexican Government through
the Comisión Nacional de Áreas Naturales Protegidas
(CONANP). The authors thank also Maren Headley for proof
Conclusions reading; Francisco Remolina, Director of the National Park Isla
Contoy and the Whale Shark Biosphere Reserve, the Secretariat
of Agriculture, Livestock, Rural Development, Fisheries and
Although the present study shows that R. typus does
Food (SAGARPA), and particularly Cristian Dı́az from the
not play a key role in the trophic functioning and National Commission for Aquaculture and Fisheries
development of Holbox ecosystem, but when whale (CONAPESCA) for information of landings; and Bernardo
sharks are affected, there is a significant impact on the Caamal Madrigal, Alberto Coral Santana, and José Betancourt
Sabatini for the fisheries and tourism information from Holbox.
SRT. On the other hand, according to the MTI, Ecosim The authors also thank fishing cooperatives from Holbox y
simulations, and SRT, the most important group in Chiquilá, who helped them during field sampling. The first
terms of indirect and direct effects propagation author (Esmeralda C. Ibarra-Garcı́a) thanks Mexican National
consisted of benthic autotrophs, additionally to its Council for Science and Technology (CONACYT) for the
scholarship at the Universidad de Guadalajara for Master
significant contributions to biomass, production,
Degree (BIMARENA).
throughput and ascendency in the system.
Because of its fishery resources and tourist attrac-
tions, Holbox has become one of the most important References
ecosystems in the Mexican Caribbean. Furthermore,
its ecological importance exceeds its uses; the envi- Alvarez-Hernandez, J. H., 2003. Trophic model of fringing coral
reef in the Southern Mexican Caribbean. Fisheries Centre
ronmental service provided by the upwelling is one of Research Reports 11(6): 227:235.
the principal sources of nutrients that contribute to the Ardisson, P. L., M. A. May-kú, D. M. T. Herrera & G.
suitable conditions that maintain the high biological A. Arellano, 2011. El Sistema Arrecifal Mesoamericano-

123
134 Hydrobiologia (2017) 792:121–135

México: consideraciones para su designación como Zona Compagno, L. J. V., 1998. Rhincodontidae. Whale sharks. In
Marı́tima Especialmente Sensible. Hidrobiológica 21(3): Carpenter, K. E., Niem, V. H. (eds), FAO identification
261–280. guide for fishery purposes. The Living Marine Resources
Arias-González, J. E. & S. Morand, 2006. Trofic functioning of the Western Central Pacific. FAO, Rome: 163.
with parasites: a new insight for ecosystem analysis. Criales-Hernández, M. I., C. B. Garcı́a & M. Wolff, 2006. Flujos
Marine Ecology Progress Series 320: 43–53. de biomasa y estructura de un ecosistema de surgencia
Arias-González, J. E., E. Nuñez-Lara, C. González-Salas & R. tropical en La Guajira, Caribe colombiano. Revista de
Galzin, 2004. Trophic models for investigation of fishing Biologı́a Tropical 54(4): 1257–1282.
effect on coral reef ecosystems. Ecological Modelling 172: de la Parra, V. R., R. Hunter, C. J. González, J. Tyminski, J.
197–212. G. Remolina, M. Maslanka, A. Ormos, L. Weigt, B. Carl-
Arias-González, J. E., P. Legendre & F. A. Rodrı́guez-Zaragoza, son & A. Dove, 2011. An unprecedented aggregation of
2008. Scaling up beta diversity on Caribbean coral reef. Whale Sharks, Rhincodon typus, in Mexican Coastal
Journal of Experimental Marine Biology and Ecology 366: waters of the Caribbean sea. PLoS ONE 6(4): 1–8.
28–36. Hacohen, A. D., 2007. Ecologı́a alimentaria del tiburón ballena
Arreguı́n-Sánchez, F., J. C. Seijo & E. Valero-Pacheco, 1993. (Rhincodon typus, Smith 1828) en la costa occidental del
An application of ECOPATH II to the north continental golfo de california, Mexico. Tesis para obtener el grado de
shelf ecosystem of Yucatan, Mexico, 269–278. In Chris- maestrı́a. CICIMAR-IPN. La Paz, B.C.S. 57.
tensen, V., D. Pauly (eds.) Trophic models of aquatic Jarre-Teichman, A. & V. Christensen, 1998. Comparative
ecosystems. ICLARM Conference Proceedings 1992: 26: modelling of trophic flows in four large upwelling
390. ecosystems vs local effects, 423–443. In Durand, M. H.,
Branco, S. M., 1999. Ecossistêmica. Uma Abordagem Integrada Cury, P., Medelsossohn, R., Roy, C., Bakun, A., Pauly, D.
dos Problemas do Meio Ambiente., Segunda edição ed. (eds). Global vs local changes in upwelling ecosystems.
Editora Edgard Blücher, Sao Paulo. Proceedings of the First International CEOS Meeting,
Cáceres, I., M. Ortiz, A. L. Cupul-Magaña & F. A. Rodrı́quez- September 6–8, 1994, Monterrey, C.A. EEUU. ORSTOM,
Zaragoza, 2016. Trophic models and short-term simulations Parı́s.
for the coral reefs of Cayos Cochinos and Media Luna Kramer, P. A. & P. R. Kramer, 2002. In McField, M. (ed),
(Honduras): a comparative network analysis, ecosystem Ecoregional conservation planning for the Mesoamerican
development, resilience, and fishery. Hydrobiologia 770(1): Caribbean Reef. World Wildlife Fund, Washington, D.C.
209–224. Medina, M., H. Arancibia & S. Neira, 2007. Un modelo trófico
Cárdenas, N. P., J. S. Herrera & O. Reyes, 2010. Distribución preliminar del ecosistema pelágico del norte de Chile
espacio-temporal de variables fisicoquı́micas y biológicas (18°200 S-24°000 S). Investigaciones Marinas 35(1): 25–38.
en el hábitat del tiburón ballena Rhincodon typus (Orec- Merino, I. M., 1983. Estudios ambientales del Arrecife de Puerto
tolobiformes: Rhincodontidae) al norte del Caribe Mexi- Morelos, Quintana Roo. Informe del proyecto de Investi-
cano. Revista de Biologı́a Tropical 58(1): 399–412. gación PAMOR. UACP y P-CCH. Maestrı́a en Ciencias del
Christensen, V. & C. J. Walters, 2005. Using ecosystem mod- Mar, Universidad Nacional Autónoma de México, México
eling for fisheries management: Where are we? CM D.F.
2005/M:19 ICES Annual Science Conference, Aberdeen, Merino, I. M., 1997. Upwelling on the Yucatan Shelf: hydro-
Scotland 20–24. graphic evidence. Journal of Marine Systems 13: 101–121.
Christensen, V. & D. Pauly, 1993. Flow characteristics of Odum, E. P., 1969. The strategy of ecosystem development: an
aquatic ecosystem. In: Christensen V., Pauly, D. (eds), understanding of ecological succession provides a basis for
Trophic models of aquatic ecosystems. ICLARM Confer- resolving man’s conflict with nature. Science 164: 262–270.
ence Proceedings 26: 338–352. Odum, E. P., 1971. Fundamentals of Ecology. Saunders Co,
Christensen, V. & C. Walters, 2004. Ecopath with Ecosim: Philadelphia, W.B. 574.
methods, capabilities and limitations. Ecological Model- Olson, R. J. & G. M. Waters, 2003. Un modelo del ecosistema
ling 172: 109–139. pelágico en el océano Pacı́fico Oriental Tropical. Comisión
Coleman, J., 1997. Whale shark interaction management, with Interamericana del Atún Tropical 22(3): 135–211.
particular reference to Ningaloo Reef Marine Park, Opitz, S., 1991. Quantitative models of trophic interactions in
1997–2007. Western Australia Wildlife Management Caribbean Coral Reefs. Dissertation for doctor degree of
Program No. 27. Mathematics and Natural Science Faculty of the University
Coll, M. M., 2002. Construcción de un modelo de equilibrio de Christian-Albrechts, 470.
masas Ecopath en el ecosistema marino explotado asociado Ortiz, M., 2008. Mass balanced and dynamic simulations of
al Delta del Ebro, Mediterráneo noroccidental. Universi- trophic models of kelp ecosystems near the Mejillones
dad Autónoma de Barcelona, Instituto de Ciencias del Mar/ Peninsula of northern Chile (SE Pacific): Comparative
CSIC, Barcelona. network structure and assessment of harvest strategies.
Compagno, L. J. V., 1984. FAO Species Catalogue. Vol. 4. Ecological Modelling 216: 31–46.
Sharks of the world. An annotated and illustrated catalogue Ortiz, M., 2010. Dynamic and spatial models of kelp forest of
of shark species known to date. Part 1 – Hexanchiformes to Macrosystis integrifolia and Lessonia trabeculata (SE
Lamniformes. FAO Fish. Synop. 125(4/1):1–249. Rome: Pacific) for assessment harvest scenarios: short-term
FAO. Froese, R., Pauly, D. (eds), 2014. FishBase. www. responses. Aquatic Conservation: Marine and Freshwater
fishbase.org, version (02/2014). Ecosystems 20: 494–506.

123
Hydrobiologia (2017) 792:121–135 135

Ortiz, M. & M. Wolff, 2002. Trophic models of four benthic Rodrı́guez-Zaragoza, F. A., A. L. Cupul-Magaña, C. Galván-
communities in Tongoy Bay (Chile): comparative analysis Villa, E. Rı́os-Jara, M. Ortiz, E. G. Robles-Jarero, E.
and preliminary assessment of management strategies. López-Uriarte & J. E. Arias-González, 2011. Additive
Journal of Experimental Marine Biology and Ecology 268: partitioning of reef fish diversity variation: a promising
205–235. marine biodiversity management tool. International Jour-
Ortiz, M., M. Avendaño, L. Campos & F. Berrios, 2009. Spatial nal of Biodiversity and Conservation 20(8): 1655–1675.
and mass balanced trophic models of La Rinconada Marine Rodrı́guez-Zaragoza, F. A., M. Ortiz, F. Berrios, L. Campos, A.
Reserve (SE Pacific coast), a propected benthic ecosystem: de Jesús-Navarrete, J. Castro-Pérez, A. Hernández-Flores,
management strategy assessment. Ecological Modelling M. Garcı́a-Rivas, F. Fonseca-Peralta, & E. Gallegos-
220: 3413–3423. Aguilar, 2016. Trophic models and short-term dynamic
Ortiz, M., M. Avedaño, M. Cantillañez, F. Berrios & L. Campos, simulations for benthic-pelagic communities at Banco
2010. Trophic mass balanced models and dynamic simu- Chinchorro Biosphere Reserve (Mexican Caribbean): A
lations of benthic communities from La Rinconada Marine conservation case. Community Ecology 17: 48–60.
Reserve off northern Chile: network properties and multi- Salas, S., G. Mexicano-Cı́ntora & M. A. Cabrera, 2006. >Hacia
species harvest scenario assessments. Aquatic Conserva- dónde van las pesquerı́as en Yucatán? Tendencias, Retos y
tion: Marine and Freshwater Ecosystems 20: 58–73. Perspectivas. Cinvestav, Ciudad de México, D.F. 97.
Pauly, D. & V. Christensen, 1995. Primary production required Signoret, M., C. Bulit & R. Pérez, 1998. Patrones de distribución
to sustain global fisheries. Nature 374: 255–257. de clorofila a y producción primaria en aguas del Golfo de
Pauly, D., V. Christensen, J. Dalsgaard, R. Froese & F. Torres México y Mar Caribe. Hidrobiológica 8(2): 81–88.
Jr., 1998. Fishing down marine food web. Science 279: Stevens, J. D., 2007. Whale Shark (Rhincodon typus) biology
860–863. and ecology: a review of the primary literature. Fisheries
Ramı́rez, M. D., M. Meekan, V. R. Parra, S. F. Remolina, M. Research 84: 5–9.
M. Trigo & J. R. Vázquez, 2012. Patterns in composition, Ulanowicz, R., 1986. Growth and Development: Ecosystems
abundance and scarring of whale sharks Rhincodon typus Phenomenology. Springer, New York.
near Holbox Island, Mexico. Journal of Fish Biology 80: Ulanowicz, R., 1997. Ecology, the Ascendant Perspective.
1401–1416. Complexity in Ecological Systems Series. Columbia
Ramı́rez, M., 1995. ‘‘Algas Marinas Bentónicas’’. In Simonetti, University Press, New York.
J., Arroyo, M. T. K., Spotorno, A., Lozada, E., (eds), Ulanowicz, R. E. & C. J. Puccia, 1990. Mixed trophic impacts in
Diversidad Biológica de Chile. Comisión Nacional de ecosystems. Ceonoces 5: 7–16.
Investigación Cientı́fica y Tecnológica, Santiago: 38–47. Walters, C. & V. Christensen, 2007. Adding realism to foraging
Randall, J. E., 1963. An analysis of the fish populations of arena predictions of trophic flow. York Academic of Sci-
artificial and natural reefs in the Virgin Islands. Caribbean ence 231: 123–138.
Journal of Science 3(1): 31–47.

123