AQUATIC CONSERVATION: MARINE AND FRESHWATER ECOSYSTEMS

Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/aqc.1047

Trophic mass balanced models and dynamic simulations of benthic communities from La Rinconada Marine Reserve off northern Chile: network properties and multispecies harvest scenario assessments MARCO ORTIZ, MIGUEL AVENDAN˜O, MARCELA CANTILLAN˜EZ, FERNANDO BERRIOS and LEONARDO CAMPOS Modelacio´n de Sistemas Ecolo´gicos Complejos, Instituto Antofagasta de Recursos Naturales Renovables (IARnR), Instituto de Investigaciones Oceanolo´gicas, Facultad de Recursos del Mar, Universidad de Antofagasta, P.O. Box 170, Antofagasta, Chile ABSTRACT 1. Mass balanced trophic models and dynamic simulations of two benthic ecological systems from La Rinconada Marine Reserve (Antofagasta Bay, SE Pacific) were constructed. 2. The scallop Argopecten purpuratus is the most important benthic resource in La Rinconada Marine Reserve, followed by the carnivorous snail Thais chocolata, and the filter-feeder bivalves Tagelus dombeii and Transennella pannosa. 3. Information on biomass, P/B ratios, catches, food spectrum, consumption, and dynamics of commercial and non-commercial species was obtained and examined using Ecopath with an Ecosim software package. 4. The bivalves A. purpuratus and T. dombeii represented the most abundant compartments in the studied subsystems. Of the carnivores, the snail T. chocolata was dominant, followed by the crabs Cancer spp. and the functional group of large epifauna. 5. The two subsystems presented similar values of system throughput. The mean trophic level of their fisheries also reached similar magnitudes (2.0), showing that the harvests in each system concentrated on secondary producers. Likewise, both subsystems presented similar A/C ratios (29.9 and 30.3), suggesting that they were immature. 6. The results obtained using mixed trophic impact (MTI) and Ecosim (increasing the fishing mortality Fi by four times) showed that only four species propagated the highest direct and indirect effects. Coincidentally, these species are the most economically important and the changes produced by the scallop A. purpuratus are noteworthy. 7. With regard to the system recovery time (SRT) estimates, only three species or functional groups presented the highest magnitudes, from highest to lowest: the sea star Luidia magallanica, the scallop A. purpuratus, and the crabs Cancer spp. 8. The topological keystone indexes of Jorda´n and Libralato had divergent results. According to Jorda´n’s index, the keystone species were L. magallanica, Cancer spp., and detritus; whereas Libralato’s index showed phytoplankton to be the keystone species. 9. Based on the results obtained, it is concluded that trophic mass balanced models and simulated management scenarios have considerable value for planning interventions and manipulations or for planning more sustainable management strategies in La Rinconada Marine Reserve. Copyright r 2009 John Wiley & Sons, Ltd. Received 1 August 2008; Revised 4 March 2009; Accepted 18 March 2009 KEY WORDS:

La Rinconada Marine Reserve; Ecopath with Ecosim; benthic communities; trophic dynamics

*Correspondence to: M. Ortiz, Modelacio´n de Sistemas Ecolo´gicos Complejos, Instituto Antofagasta de Recursos Naturales Renovables (IARnR), Instituto de Investigaciones Oceanolo´gicas, Facultad de Recursos del Mar, Universidad de Antofagasta, P.O. Box 170, Antofagasta, Chile. E-mail: [email protected]

Copyright r 2009 John Wiley & Sons, Ltd.

M. ORTIZ ET AL.

INTRODUCTION La Rinconada Marine Reserve, located at the northern end of Antofagasta Bay (Mejillones Peninsula) is a coastal area of ca 270 ha where the main natural populations of the scallop Argopecten purpuratus (Lamarck, 1819) are distributed off northern Chile. This marine reserve was established by the Chilean Government in 1997 in order to protect the scallops from overfishing. Simultaneously, this measure has also reduced the exploitation of other species of commercial interest living in the reserve such as the bivalves Aulacomya ater (Molina, 1782) and Tagelus dombeii (Lamarck, 1818), and the carnivorous snail Thais chocolata (Duclos, 1832). Although the initial motivation for this protective measure was a noble bio-ecological reason (the preservation of the scallop), it has resulted in increased interest (pressure) on the part of various artisanal fishing organizations to exploit the reserve’s different resources. Unfortunately, this social pressure has resulted in an illegal scallop fishery, significantly reducing the scallop abundance and modifying its size structure (Avendan˜o and Cantilla´nez, 1996). Many research projects have been carried out in the reserve’s coastal area, most notably, studies on oceanographic conditions (Rodrı´ guez et al., 1991; Marı´ n et al., 1993; Escribano et al., 1995), primary productivity (Rodrı´ guez et al., 1991), and different aspects of A. purpuratus population dynamics (Avendan˜o and Le Pennec, 1998; Avendan˜o et al., 2001, 2007; Avendan˜o and Cantilla´n˜ez, 2005; Cantilla´n˜ez et al., 2005). In this respect, a recent study by Avendan˜o et al. (2007) has shown that the reserve contains areas that favour the artificial collection of A. purpuratus spat, awakening the interest of scallop farmers for use of the reserve as a natural spat supply that can then be incorporated into their own culture systems. Although much has been learned regarding some properties of the reserve’s coastal area, this knowledge is not enough if we want to respond to the multiple expectations of the different groups involved (artisanal fishermen, farmers, authorities) regarding the design, implementation, and execution of ecologically sustainable, multispecies co-management. Given our limited knowledge, the development of an ecological and systemic (holistic) research programme that permits the collection of biological and ecological information on the more relevant species and/or functional groups inhabiting the benthic communities of the Marine Reserve is needed urgently. In recent years, an important interest in evaluating, quantifying, and predicting the changes produced in the ecosystem properties by the fishery has been observed (Hall, 1999a,b; Hawkins, 2004; Francis et al., 2007; Scotti et al., 2007). This has led to a change in the way studies are conducted, focusing on the different types of ecological interrelationships occurring in the communities and ecosystems rather than on the protection of isolated populations (Scotti et al., 2007). Multispecies or ecosystem models provide alternative mechanisms to the classic population-reductionist models for those cases in which the goals are to approach the holistic properties of the ecosystems and assess the propagation of higher-order effects within complex subsystems (Levins, 1974, 1998a; Hawkins, 2004) and to estimate the topological keystone species in the networks (Jorda´n et al., 1999; Jorda´n, 2001; Libralato et al., 2006; Vasas et al., 2007). Supported by the ecological network theory of Ulanovicz (1986, 1997), some explorations have been performed with regard to the properties, Copyright r 2009 John Wiley & Sons, Ltd.

dynamics, and global health of ecosystems (Costanza and Mageau, 1999). It is important to indicate that, along with the estimation of ecosystem macro-descriptors, estimates can be made with respect to the system recovery time in response to human disturbances such as fisheries and aquaculture (Jørgensen, 1992, 2000; Wolff, 1994; Gaedke, 1995; Monaco and Ulanowicz, 1997; Ortiz and Wolff, 2002a,b; AriasGonza´lez et al., 2004; Pinneger and Polunin, 2004; Patrı´ cio and Marques, 2006). Multispecies trophic models based on Ecopath II (Christensen and Pauly, 1992) and Ecosim (Walters et al., 1997) have become quite popular, especially for describing trophic webs and predicting the effects produced by the application of different exploitation scenarios in marine ecosystems (Christensen and Pauly, 2004; Pikitch et al., 2004). The main objective of this work was to build trophic models that represent interspecific relationships (prey–predator) taking place in the benthic subsystems of La Rinconada Marine Reserve, including the effects produced by illegal A. purpuratus fisheries and the eventual implementation of multispecies management (exploitation of other species), using the Ecopath with Ecosim theory and software package (Walters et al., 1997). In turn, and based on the trophic models created, two methods designed to identify the topological keystone species in networks were used, namely the Libralato index (Libralato et al., 2006) and the Jorda´n index (Jorda´n et al., 1999). Supported by the aforementioned theoretical frameworks the following were investigated: (1) biomass distribution and flow structures in the existing ecological subsystems; (2) main benthic predators, consumption rates, and prey items; (3) redundancy (i.e. the existence of species with similar trophic functions in the ecological systems (sensu Lawton, 1994); (4) recognition of topological keystone species (i.e. species or functional groups with low biomass but with important effects on other species, sensu Paine, 1966, 1969; Mills et al., 1993); (5) degree of resistance to disturbances and the resilience time of ecological subsystems in response to different multispecies exploitation strategies; and (6) the most affected species or functional groups given the development of different management scenarios. All the simulations were executed using a mixed type flow control (both top-down and bottom-up), since this control mechanism is considered to be more realistic (Levins, 1998b); its use was recently demonstrated to result in a high degree of certainty for the predictions (Thompson et al., 2004; Ortiz, 2008b).

METHODS Study area The benthic system of La Rinconada Marine Reserve (231280 S 701300 W), located in the northern part of Antofagasta Bay (Mejillones Peninsula, Chile), was chosen for this study (Figure 1). Oceanographically, the reserve is influenced by three different currents/water masses: subAntarctic water (SAW), subtropical water (STW), and equatorial subsurface water (ESSW) (Escribano et al., 1995). The bottom is dominated by sand and gravel. Two ecological subsystems hosting different aggregations of species can be clearly recognized at 8–15 m depth (Figure 1). It is important to mention that near Mejillones Peninsula there is an important Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) DOI: 10.1002/aqc

TROPHIC MODEL OF MARINE RESERVE LA RINCONADA

Study area CHlLE

-70º5’

-70º4’

N

SS1

SS2

Rinconada Reserve

-23°5’

Antofagasta bay

100 80

-23°6’ %

60 40 20

SS1

SS2

-23°7’

2 km

Figure 1. Study area of La Rinconada Marine Reserve (SE Pacific), northern Chile. Both ecological subsystems (SS1 and SS2) are depicted. A species assemblage (cluster) is shown for both subsystems (SS1 and SS2).

upwelling centre that supplies nutrients to the coastal ecosystem (Escribano et al., 2004). The temperature of surface water ranges between 161C in winter and 201C in summer (Escribano et al., 2004).

Ecopath and Ecosim theoretical frameworks The balance of mass (energy or nutrients) for any species or functional group of the network can be represented as follows:     n X P Q EEi  Bj  DCji Yi BAi Ei ¼ 0 Bi  B i B i j¼1 ð1Þ Copyright r 2009 John Wiley & Sons, Ltd.

where Bi and Bj are the prey i and predator j biomasses, respectively; P/Bi is the productivity (production/biomass ratio), which is equivalent to total mortality (Z) (Allen, 1971); EEi is the ecotrophic efficiency, that is, the fraction of the total production of a group used in the system; Yi is the yield of fisheries per unit area and time (Y 5 fishing mortality  biomass); Q/Bj represents food consumption per unit biomass of j; DCji is the fraction of prey i in the average diet of predator j; BAi is the biomass accumulation rate for i; and Ei corresponds to the net migration of i (emigration minus immigration) (Christensen and Pauly, 1992; Walters et al., 1997; Christensen and Walters, 2004). Under this theoretical framework, the energy input and output of all living groups, Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) DOI: 10.1002/aqc

M. ORTIZ ET AL.

transference rate between compartments i and j. This parameter determines whether the flow control mechanism is top-down, bottom-up, or mixed, ranging from 1.0 (bottom-up) to c2.0 (top-down); a value of 2.0 represents a mixed control mechanism. Details concerning the Ecopath with Ecosim software package are given in Christensen and Pauly (1992), Walters et al. (1997), and Pauly et al. (2000). The Ecopath theoretical framework combines the approach of Polovina (1984) to estimate the biomass and food consumption of the ecosystem variables or functional groups with that of Ulanowicz (1986, 1997), a network analysis of flows among variables of the system, for calculating the macro-descriptors: total system throughput (T), ascendency (A), development capacity (C), and relative ascendency (A/C ratio). Throughput describes the vigour or size of a system and provides a measure of its metabolism. Ascendency integrates both size and organization of the systems, with the latter referring to the number and diversity of interactions between its components. The development capacity quantifies the upper limit for ascendency, whereas the A/C ratio describes the degree of maximum specialization that is actually achieved in the system (maturity index) (Baird and Ulanowicz, 1993; Costanza and Mageau, 1999). This ratio can also be used as the system’s ability to withstand disturbance (Ulanowicz, 1986, 1997). All these macro-descriptors have been widely used to

by definition, must be balanced. The energy balance is ensured within each variable or compartment group by using the following equation of Christensen et al. (2004): Q ¼ P þ R þ UAF

ð2Þ

where Q is consumption, P is production, R is respiration, and UAF is the unassimilated food of each variable or compartment in the system. The inclusion of biomass accumulation and migration factors in Equation (1) presents Ecopath models as an energy continuity approach rather than a strictly steady-state condition. This particular approach allows changes in the variables or compartments when the mathematical function is expressed in its dynamic form. In order to use Ecosim, an extension routine of Ecopath is included to define the consumption by compartment i; Qij is represented by the following equation: aij  vij  Bi  Bj ð3Þ Qij ¼ ð2vij þ aij  Bj Þ where aij represents the instantaneous mortality rate on prey i caused by a single unit of predator j biomass. Likewise, aij can be understood as the rate of effective search by predator j for prey i. Each aij is estimated directly from the corresponding Ecopath models as: aij ¼ Qi =ðBi  Bj Þ where (Qi 5 total consumption of i). The vij represents the

Table 1. Prey–predator and plant–grazer matrix (percentage of wet biomass in stomach of predators) for subsystem 1 (SS1) and subsystem 2 (SS2) used for the Ecopath with Ecosim software programme Prey-predator (A) Subsystem: SS1 (1) A. purpuratus (2) sA. purpuratus (3) T. dombeii (4) T. pannosa (5) A. ater (6) T. chocolata (7) L. magallanica (8) Cancer spp. (9) SEH (10) SEC (11) LE (12) Chlorophyta (13) Rhodophyta (14) Phaeophyta (15) Zooplankton (16) Phytoplankton (17) Detritus

(B) Subsystem: SS2 (1) A. purpuratus (2) sA. purpuratus (3) T. dombeii (4) T. pannosa (5) T. chocolata (6) L. magallanica (7) Cancer spp. (8) SEH (9) SEC (10) LE (11) Chlorophyta (12) Rhodophyta (13) Phaeophyta (14) Zooplankton (15) Phytoplankton (16) Detritus

1

2

3

4

5

6

7

8

10

10

5 5

4 2 10 10 1

8 7 5

30 20 3

9

10

11

10

5

5

3 5 10 20

15 15 10 10

10 10 5 5

2 20 14 6

10 1

20 10 2

15

15 2 40 5 85 15

85 15

70 30

85 15

85 15

10

3 40

60

10

29

33

1

2

3

4

5

6

7

8

9

10

14

5

5

8

5

5

15 10

58 20 10 1

15 30 15

25 20 10

15 10 10

5 5 5

14 12

2 10 10

10

15 10 2

95 5

15 2 40 5 85 15

85 15

Copyright r 2009 John Wiley & Sons, Ltd.

70 30

85 15

55

10

7

3 40

24

33

95 5

Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) DOI: 10.1002/aqc

TROPHIC MODEL OF MARINE RESERVE LA RINCONADA

describe and compare a variety of ecosystems of different spatial sizes, geographic locations, and complexities (Monaco and Ulanowicz, 1997; Jarre-Teichmann and Christensen, 1998; Niquil et al., 1999; Heymans and Baird, 2000; Wolff et al., 2000; Ortiz and Wolff, 2002a, 2004; Arias-Gonza´lez et al., 2004; Patrı´ cio and Marques, 2006; Patrı´ cio et al., 2006).

Selection of model compartments, sampling programmes, and data sources Studies were carried out in the field between 2005 and 2007 at La Rinconada Marine Reserve in order to select the variables of each of the ecological systems and to estimate the biomass (B), catches (C), turnover rates (P/B), consumption rates (Q/B), and food items for the variables selected. Appendix A shows the source data for each of the compartments selected in the current work. Although most of the model compartments represent individual species, as is the case of the scallop A. purpuratus, and for the bivalves Tagelus dombeii, Transennella pannosa and Aulacomya ater, the carnivorous snail Thais chocolata, and the sea star Luidia magallanica, it was also necessary to adapt functional groups into compartments that included different species. The Rhodophyta, Chlorophyta, and Phaeophyta compartments included different species of red (e.g. Rhodymenia sp.), green (e.g. Ulva sp.), and brown (e.g. Glossophora kunthii) algae. The group Cancer spp. includes the crabs Cancer coronatus, C. polyodon, C. edwardsi and C. porteri; the large epifauna (LE) compartment includes the crabs Pagurus edwarsi and Eurypanopeus crenatus; the small epifauna herbivorous group (SEH) is made up of the molluscs Calyptraea trochiformis, Tegula sp., Fissurella sp., Chiton sp., Mitrella sp. and Colisella sp.; the small epifauna carnivorous group (SEC) includes the molluscs Aeneator fontanei, Nassarius gayi, Cancellaria sp., Oliva sp., Crassilabrum sp., Priene sp., Xanthochorus sp., Trigonostoma sp. and Nucula sp. In addition, the variable sArgopecten purpuratus (sAp) was incorporated into the models, representing, for purposes of the simulation only, the farming systems near the reserve. The diet matrixes for both subsystems (Table 1) show that all the compartments are trophically linked by detritus, primarily as microbial biofilm. Diverse studies have emphasized the importance of bacteria as food for various molluscs (Epstein, 1997; Plante and Shriver, 1998), zooplankton (Epstein, 1997), and echinodermata (Findlay and White, 1983). It is important to note that the models were constructed to grasp the trophic relationships for the most relevant species (commercial resources) inhabiting benthic communities, leaving out the flows from epiphytes, microphytobenthos, bacteria, and those leading to fishes, seals and birds due to insufficient scientific information. Although this reduced the realism of the model configuration, we think that the most relevant interdependencies and flows are reflected. Since the configuration of all models has followed the same principle, this systematic error should not impede a comparative analysis with other ecological systems under similar constraints.

Balancing and calibration of the models The first step in balancing the models was to determine if the model outputs were realistic, that is, to check if the ecotrophic efficiency (EE) was o1.0 for all variables or compartments. If an inconsistency was detected, the biomass values (annual Copyright r 2009 John Wiley & Sons, Ltd.

averages) were slightly changed within the confidence limits (standard deviation) obtained from field studies. Nevertheless, for the suspended scallop (sA. purpuratus), Chlorophyta, Rhodophyta, and Phaeophyta, the turnover rate (P/B) values were calculated by Ecopath. It was not necessary to modify the diet matrixes when balancing the models. As a second step, gross efficiency (GE) values were checked for consistency by comparing them with data from the literature.

System recovery time Stability is the ability of a system to return to a state of equilibrium after a disturbance (Holling, 1973). Ulanowicz’s theory (Ulanowicz, 1986, 1997) states that the ecosystem organization, in terms of relative ascendancy (A/C) and redundancy (internal flow of overhead), may be the most important attribute of system stability. Resilience has been conceptualized as the speed at which the entire system returns to its original state after it has been displaced from its original state (Pimm, 1982). Resistance describes the capacity of a system to withstand displacement (Begon et al., 1990). Therefore, stability includes both resilience and resistance. In the present contribution, we assume that system recovery time Table 2. Parameter values entered (in bold) and estimated (standard) for each ecological subsystem (SS1 and SS2) by Ecopath with Ecosim software programme Compartements Species/Funcional groups

TC

C

B

P/B

(A) Subsystem SS1 (1) A. purpuratus (2) sA. purpuratus (3) T. dombeii (4) T. pannosa (5) A. ater (6) T. chocolata (7) L. magallanica (8) Cancer spp. (9) SEH (10) SEC (11) LE (12) Chlorophyta (13) Rhodophyta (14) Phaeophyta (15) Zooplankton (16) Phytoplankton (17) Detritus

2.0 2.0 2.0 2.0 2.0 2.5 3.1 3.2 2.0 2.8 2.8 1.0 1.0 1.0 2.0 1.0 1.0

224.9 2.5 0.01 0.01 0.2 0.16

259.0 50.0 15.0 14.0 12.0 14.2 0.8 5.5 20.0 13.0 7.1 10.0 83.5 20.0 20.0 30.0 100.0

2.7 2.0 2.6 2.8 2.1 2.7 0.5 1.9 2.8 2.4 1.9 5.0 5.0 5.0 40.0 250.0

(B) Subsystem SS2 (1) A. purpuratus (2) sA. purpuratus (3) T. dombeii (4) T. pannosa (5) T. chocolata (6) L. magallanica (7) Cancer spp. (8) SEH (9) SEC (10) LE (11) Chlorophyta (12) Rhodophyta (13) Phaeophyta (14) Zooplankton (15) Phytoplankton (16) Detritus

2.0 2.0 2.0 2.0 2.5 3.1 3.1 2.0 2.8 2.8 1.0 1.0 1.0 2.0 1.0 1.0

46.1 2.5 0.2 0.2 0.5

53.6 25.0 314.2 78.7 48.9 0.01 5.0 20.0 15.0 12.5 10.0 255.0 20.0 20.0 30.0 100.0

2.7 2.0 2.0 2.8 2.7 0.5 1.9 2.8 2.4 1.9 5.0 5.0 5.0 40.0 250.0

Q/B

9.9 9.9 9.9 9.9 9.9 7.2 3.0 9.5 11.7 10.4 9.2

160.0

9.9 9.9 9.9 9.9 7.2 3.0 9.5 11.7 10.4 9.2

160.0

EE

0.36 0.03 0.86 0.88 0.89 0.72 0.06 0.10 0.82 0.73 0.72 0.70 0.01 0.94 0.01 0.80 0.19 0.57 0.05 0.18 0.42 0.26 0.06 0.10 0.99 0.99 0.84 0.70 0.004 0.94 0.04 0.87 0.34

Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) DOI: 10.1002/aqc

M. ORTIZ ET AL.

instantaneous direct and indirect effects and SRT as a response to an increase in fishing mortality (Fi). This was calculated for the first and second year of simulation for the scallop A. purpuratus; the bivalves T. dombeii, T. pannosa and A. ater; the carnivorous snail T. chocolata; the sea star L. magallanica; the group Cancer spp.; and the group Rhodophyta in the subsystems. The propagation of instantaneous effects was determined by evaluating the biomass of all the variables of both subsystems in the third year of simulation, that is, 1 year after the increase in fishing mortality. Finally, the FMSY values

(SRT), as obtained by our simulations, is a measure of the internal stability of the systems.

Multispecies harvest assessment Ecopath’s mixed trophic impact (MTI) routine (Ulanowicz and Puccia, 1990) was used to make a preliminary evaluation of the propagation of direct and indirect effects in response to disturbances affecting species of commercial interest. Ecosim simulations were used to evaluate the propagation of

8

7

11

10

6

15

1

2

17

3

9

16

5

4

13

12

14

7

6 9

10

5

14

16

8

1

2

15

11

12

3

4

13

Figure 2. Trophic models for ecological subsystem 1(a), and subsystem 2(b). Vertical position approximates trophic levels. The box is proportional to the compartment biomass (g wet weight m2). Simple arrows represent the flow of matter among compartments and double arrows means flow to fisheries. The number in box corresponds to the species or functional groups (for details see Table 1). Copyright r 2009 John Wiley & Sons, Ltd.

Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) DOI: 10.1002/aqc

TROPHIC MODEL OF MARINE RESERVE LA RINCONADA

were determined only for commercially important bivalve species. All the dynamic simulations were carried out using a mixed flow control mechanism (vij) since the use of this control resulted in unique and stable FMSY values (Ortiz and Wolff, 2002b). Furthermore, Ortiz (2008b) recently demonstrated that the use of a mixed control gives the highest certainty in predictions. When taken separately, top-down and bottom-up controls could be considered to be a false dichotomy in ecology, distorting the capacity for understanding (sensu Levins, 1998b; Thompson et al., 2004).

where pi is the proportion of biomass of each species Bi with respect to the sum of the total biomass Bk. Therefore, in order to balance the overall effect and biomass, we established the topological keystone index (KS) for each species or functional group, integrating Equations (4) and (5) as follows: KSi ¼ log½ei ð1  pi Þ

Topological keystone species index The Libralato index (Libralato et al., 2006) is an extension of the mixed trophic impacts (MTI) (Ulanowicz and Puccia, 1990) since it uses the impact matrix. Since every impact can be quantitatively positive or negative, a new measure of overall effect is needed for each species or functional group ei using the following mathematical relationship: vffiffiffiffiffiffiffiffiffiffiffiffiffiffi uX u n 2 mij ei ¼ t

the total biomass of the food web was estimated using the following relationship: Bi pi ¼ Pn ð5Þ i Bk

ð4Þ

j6¼i

where mij corresponds to the elements of the MTI matrix and quantifies the direct and indirect impacts that each (impacting) species or group i has on any (impacted) group j of the food web. However, the effect of the change in biomass on the group itself (i.e. mii) is not included. The contribution of biomass from every species or functional group with respect to

ð6Þ

The other topological keystone index used in this work was developed by Jorda´n et al. (1999) and Jorda´n (2001). The Jorda´n index considers direct and indirect interaction in both directions (i.e. bottom-up and top-down). The keystone index of the xth species or functional group (Kx) is calculated as follows: n m X X 1 1 ð1 þ Kbc Þ þ ð1 þ Kte Þ ð7Þ Kx ¼ d f c c¼1 e¼1 e where n is the number of predators eating species x, dc is the number of prey of the cth predator, Kbc is the bottom-up keystone index of the cth predator, and symmetrically we have m as the number of prey eaten by species x, fe as the number of predators of its eth prey, and Kte as the top-down keystone index of the eth prey. Therefore, the keystone index (Kx) corresponds to the addition of bottom-up (Kbc) and topdown (Kte) components. For more details of this method, see

Table 3. Summary statistics after mass-balance process by Ecopath with Ecosim (A) and network flow indices (B) for each ecological subsystem (SS1 and SS2) Subsystems SS1

SS2

(A) Summary statistics Sum of all consumption (g m2 year1) Sum of all exports (g m2 year1) Sum of all respiratory flows (g m2 year1) Sum of all flows into detritus (g m2 year1) Total system through (g m2 year1) Sum of all production (g m2 year1) Mean trophic level of the catch Gross efficiency of fisheries (catch/net pp, %) Total net primary production (g m2 year1) Total primary production/Total respiration Net system production (g m2 year1) Total primary production (g m2 year1) Total biomass/total throughput Total biomass (exc. Detritus) (g m2 year1) Toal catches (g m2 year1)

7256.8 4104.3 3963.2 4800.0 20124.0 9908.0 2.0 0.028 8067.5 2.0 4104.3 14.1 0.029 574.1 227.8

8773.8 3417.1 5507.9 5062.6 22761.0 10436.0 2.0 0.006 8925.0 1.6 3417.1 9.8 0.04 907.9 49.5

(B) Netwok flow indices Ascendency (Total) Flowbits Overhead (Total) Flowbits Capacity (Total) Flowbits Pathway Redundancy (of Overhead) (%) A/C (%) Throughput cycled (exc. Detritus) (g m2 year1) Throughput cycled (inc. Detritus) (g m2 year1) Finn«s cycling index (FCI) (%) Average path length (APL) (dimensionless) Food web connectance (dimensionless) Omnivory index (OI) (dimensionless)

24375.1 55945.9 80321.0 50.8 30.3 7.0 1.4 2.4 2.5 0.2 0.1

29002.2 67845.7 96847.8 51.3 29.9 12.7 1.4 4.1 2.6 0.2 0.1

Copyright r 2009 John Wiley & Sons, Ltd.

Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) DOI: 10.1002/aqc

M. ORTIZ ET AL.

T. pannosa (78.7 g ww m2) also dominated in the SS2 (Table 2). Regarding predators, T. chocolata (48.9 g ww m2) and the large epifauna group stood out in SS2, followed by the group of Cancer spp. (5.5 g ww m2) in SS1 (Table 2) (Figure 2). With regard to the ecosystem structure of the ecological benthic systems, the SS2 had the highest magnitude of system throughput (T) (22761 g ww m2 year1) (Table 3). The two subsystems studied presented a similar magnitude of food web connectivity and omnivory index (OI), with comparable degrees of linearity in their topologic characteristics (Table 3). At the same time, both subsystems reached a mean trophic level of

RESULTS

1.50

1.00

1.00

0.50

0.50

T.chocolata

1.50

0.00

-0.50

-1.00

-1.00

-1.50

-1.50

1.50

1.50

1.00

1.00

0.50

0.50

0.00

Impacting population

Impacting population

-0.50 -1.00 -1.50

0.00 -0.50 -1.00 -1.50 1.50 1.00

0.50

0.50

Cancer spp.

1.00

0.00

0.00

-0.50

-0.50

-1.00

-1.00

-1.50

-1.50

1.50

1.50

1.00

1.00

0.50

0.50

Rhodophyta

T. pannosa

1.50

0.00

0.00

-0.50

-0.50

-1.00

-1.00

-1.50

-1.50 A. purpuratus sA. purpuratus T. dombeii T. pannosa A. ater T. chocolata L. magallanica Cancer spp. SEH SEC LE Chlorophyta Rhodophyta Phaeophyta Zooplankton Phytoplankton Detritus Fishery

A. ater

0.00

-0.50

L. magallanica

T. dombeii

A. purpuratus

The bivalve mollusc T. dombeii is the species with the highest biomass (314.2 g wet weight (ww) m2) in subsystem 2 (SS2). On the other hand, the scallop A. purpuratus (259.0 g ww m2) is the most abundant species in subsystem 1 (SS1). The functional group Rhodophyta (255.0 g ww m2) and the clam

A. purpuratus sA. purpuratus T. dombeii T. pannosa A. ater T. chocolata L. magallanica Cancer spp. SEH SEC LE Chlorophyta Rhodophyta Phaeophyta Zooplankton Phytoplankton Detritus Fishery

Jorda´n et al. (1999), Jorda´n (2000), and Vasas et al. (2007). It is important to note that in the current work we have calculated only the global topological keystone index (Kx).

Impacted population

Figure 3. Mixed trophic impacts (direct and indirect) as response to impacting A. purpuratus, T. dombeii, T. pannosa and A. ater (all commercial species). Ecological subsystem 1 (SS1) and subsystem 2 (SS2). Copyright r 2009 John Wiley & Sons, Ltd.

Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) DOI: 10.1002/aqc

1.50

1.00

1.00

0.50

0.50 T.chocolata

1.50

0.00

-0.50

-1.00

-1.00

-1.50

-1.50

1.50

1.50

1.00

1.00

0.50

0.50

0.00 -0.50

-1.50 1.50 1.00

0.00

-0.50

-1.50 1.50 1.00

Cancer spp.

0.50

0.00

-1.00

Impacting population

Impacting population

-1.00

0.50 0.00

-0.50

-0.50

-1.00

-1.00

-1.50

-1.50 A. purpuratus sA. purpuratus T. dombeii T. pannosa T. chocolata L. magallanica Cancer spp. SEH SEC LE Chlorophyta Rhodophyta Phaeophyta Zooplankton Phytoplankton Detritus Fishery

T. pannosa

0.00

-0.50

L. magallanica

T. dombeii

A. purpuratus

TROPHIC MODEL OF MARINE RESERVE LA RINCONADA

1.50 1.00

Rhodophyta

0.50 0.00 -0.50 -1.00

A. purpuratus sA. purpuratus T. dombeii T. pannosa T. chocolata L. magallanica Cancer spp. SEH SEC LE Chlorophyta Rhodophyta Phaeophyta Zooplankton Phytoplankton Detritus Fishery

-1.50

Impacted population

Figure 3. Continued.

fisheries equal to 2.0, which proves that harvesting is done almost exclusively over secondary producers, particularly the scallop A. purpuratus. According to the magnitudes of average path length (APL) and the omnivory index (OI), both subsystems presented a similar linearity of the trophic webs. The SS2 presented the highest magnitude of development capacity (C) and ascendency (A). However, with respect to the level of maturity and resistance to disturbances based on relative ascendency (A/C) and redundancy, both subsystems presented similar characteristics (Table 3). With regard to the contribution of each group to ascendency as a way of evaluating the contribution of each of the compartments to the overall structure and function of the system, in both benthic models, the phyto-zooplankton reached levels of magnitude higher than 40%, followed by detritus (28%), Copyright r 2009 John Wiley & Sons, Ltd.

the herbivore group (16%), macroalgae (10%) and carnivores (3%). The propagation of direct and indirect effects estimated using the mixed trophic impact (MTI) shows quantitative and qualitative differences between both subsystems (Figure 3). The SS2 presented a higher magnitude of changes with respect to SS1, highlighting the effects of the carnivorous snail T. chocolata and bivalve filter-feeders T. dombeii and T. pannosa. On the contrary, the dynamic simulations using Ecosim showed a very different response pattern to that obtained using MTI in which only the effects propagated by the scallop A. purpuratus stand out in the two subsystems (Figure 4). Figure 5 show the possible responses in biomass and the highest magnitudes of system recovery time (SRT) in both Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) DOI: 10.1002/aqc

M. ORTIZ ET AL. 200

100

100

0

0

-100

-100

-200

-200

200

200

Fishing on T. dombeii

100

100

0

0

-100

-100

-200 200

Fishing on T. pannosa

100

Final / initial biomass

Final / initial biomass

200

Fishing on A. purpuratus

Fishing on T. chocolata

Fishing on Cancer spp.

-200 200 Fishing on L. magallanica 100

0

0

-100

-100

-200

-200

200

200 Fishing on A. aterz

Fishing on Rhodophyta

0

0

-100

-100

-200

-200 A. purpuratus sA. purpuratus T. dombeii T. pannosa A. ater T. chocolata L. magallanica Cancer spp. SEH SEC LE Chlorophyta Rhodophyta Phaeophyta Zooplankton Phytoplankton Detritus

100

A. purpuratus sA. purpuratus T. dombeii T. pannosa A. ater T. chocolata L. magallanica Cancer spp. SEH SEC LE Chlorophyta Rhodophyta Phaeophyta Zooplankton Phytoplankton Detritus

100

Figure 4. Dynamical responses of biomass behaviour for ecological subsystem SS1 (a) and SS2 (b) subject to one year of increased (4  ) fishing mortality (between year 1 and 2 of simulation) for A. purpuratus, T. dombeii, T. pannosa and A. ater. (Note: the dynamical response of biomass was obtained at 3rd year of simulation using mixed controlling, u 5 2.0.)

subsystems analysed. The scallop A. purpuratus and the sea star L. magallanica presented the highest SRT magnitudes in SS1. However, the propagation of direct and indirect effects was clearly dissimilar. In SS2, on the contrary, the group of Cancer spp. and L. magallanica had the highest SRT values, propagating little effects in the remaining variables of the system (Figure 5). Table 4 summarizes the SRT magnitudes obtained in the simulations carried out in both subsystems. It can be observed that a disturbance that is present simultaneously in all the variables produces the highest SRT magnitudes, which are similar to those obtained for the sea star L. magallanica. Copyright r 2009 John Wiley & Sons, Ltd.

As for the eventual presence of keystone species in SS1 and SS2, both indexes showed quite different results: according to Jorda´n’s index, L. magallanica and Cancer spp. are the keystone species in SS1, whereas Libralato’s index shows the keystone species to be phytoplankton (Table 5). In SS2, the respective keystone species were detritus (Jorda´n index) and phytoplankton (Libralato index) (Table 5).

DISCUSSION The results obtained for the distribution of biomasses in La Rinconada Marine Reserve clearly established that both Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) DOI: 10.1002/aqc

TROPHIC MODEL OF MARINE RESERVE LA RINCONADA

200

200 Fishing on A. purpuratus

100

0

0

-100

-100

-200

-200

200

200

Fishing on T. dombeii

Fishing on L. magallanica

100

100

0

0

-100

-100

Final / initial biomass

-200 200 Fishing on T. pannosa 100

-200 200 Fishing on Rhodophyta 100

0

0

-100

-100

-200

-200 A. purpuratus sA. purpuratus T. dombeii T. pannosa T. chocolata L. magallanica Cancer spp. SEH SEC LE Chlorophyta Rhodophyta Phaeophyta Zooplankton Phytoplankton Detritus Fishery

Final / initial biomass

Fishing on Cancer spp.

100

200 Fishing on T. chocolata 100

0

-100

A. purpuratus sA. purpuratus T. dombeii T. pannosa T. chocolata L. magallanica Cancer spp. SEH SEC LE Chlorophyta Rhodophyta Phaeophyta Zooplankton Phytoplankton Detritus Fishery

-200

Figure 4. Continued.

subsystems studied are dominated by commercially important species: the scallop A. purpuratus in SS1 and the clam T. dombeii in SS2. This natural distribution pattern allows the development of complementary non-spatially-superimposed fishing management, reducing disturbances associated with human intervention in the global benthic system. It is important to mention that in SS2, along with the high abundance levels of T. dombeii; the clam T. pannosa and the snail T. chocolata are also distinguished (both the latter species are economically valuable), thus the implementation of a multispecies management programme in the marine reserve is required. Copyright r 2009 John Wiley & Sons, Ltd.

In terms of the estimates of system throughput (T), SS2 had a slightly higher magnitude than that obtained in SS1, despite the fact that both systems are dominated by different benthic species. The T magnitudes estimated for both subsystems are very close to those calculated for the benthic ecological systems in Tongoy Bay (Ortiz and Wolff, 2002a). It is also important to indicate that T estimations were also higher than those described for other ecosystem models such as estuaries and mangroves (Wolff et al., 2000; Vega-Cendejas and ArreguinSanchez, 2001; Patricio and Marques, 2006; Patricio et al., 2006) but lower than those obtained in coral reefs (AriasGonza´lez et al., 2004). Trophic webs presented topological Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) DOI: 10.1002/aqc

M. ORTIZ ET AL.

Figure 5. Dynamical changes of biomass at each ecological subsystem subject to 1 year of increased (4  ) fishing mortality under mixed controlling (u 5 2.0). The responses of the species with the largest system recovery time (SRT) are shown.

Table 4. Summary of the system recovery time (SRT) for each ecological subsystem (SS1 and SS2) using mixed flow control mechanism (u 5 2.0) Mixed flow control (v 5 2.0) Ecological subsystem

System recovery time (year)

La Rinconada SS1 Fishing Argopecten purpuratus Tagelus dombeii Transanella pannosa Aulacomya ater Thais chocolata Cancer spp. L. magallanica Rhodophyta All

15 5 5 6 6 11 16 7 16

La Rinconada SS2 Fishing Argopecten purpuratus Tagelus dombeii Transanella pannosa Thais chocolata Cancer spp. L. magallanica Rhodophyta All

10 5 5 6 11 16 8 16

characteristics very similar to those described for the Tongoy Bay benthic systems (Ortiz and Wolff, 2002a) and similar to the trophic webs of kelp communities off Mejillones Peninsula of northern Chile (Ortiz, 2008a). The magnitudes of capacity (C) that were calculated are within the extreme values described for the Tongoy Bay Copyright r 2009 John Wiley & Sons, Ltd.

subsystems (Ortiz and Wolff, 2002a). However, these were much lower than the values obtained for the kelp model systems of Mejillones Peninsula (Ortiz, 2008a). The relative ascendancy (A/C), considered to be an index of maturity as well as of the system’s ability to withstand disturbances (Ulanowicz, 1986, 1997), shows that both subsystems have similar characteristics and are, in turn, very like those obtained in the benthic systems of Tongoy Bay (Ortiz and Wolff, 2002a) and other ecological systems (Baird and Ulanowicz, 1993; Heymans and Baird, 2000; Wolff et al., 2000; Vega-Cendejas and Arreguin-Sa´nchez, 2001). However, both subsystems would be less resistant to the kelp benthic models of Mejillones Peninsula (Ortiz, 2008a). It is important to indicate that the conclusions based on A/C ratios should be taken cautiously due to the negative relationship between ascendency and maturity described by Christensen (1995). Ulanowicz (1997) proposed estimating the relative ascendency of each group as a way of evaluating the contribution of each of the compartments to the overall structure and function of the system. In this sense, the results obtained are consistent with the oceanographic properties of the marine reserve, which is characterized by the permanent influence of nutrient-rich waters coming from the upwelling centre located off Mejillones Peninsula (Escribano et al., 2004). Similar results have been published for the benthic systems of Tongoy Bay (central Chile) (Ortiz and Wolff, 2002a), also located very near an upwelling centre off Lengua Punta de Vaca Peninsula (Daneri et al., 2000; Montecinos and Quiroz, 2000). Unfortunately, the direct and indirect effects estimated using MTI and Ecosim showed very different response patterns, with only four benthic species, all economically Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) DOI: 10.1002/aqc

TROPHIC MODEL OF MARINE RESERVE LA RINCONADA

Table 5. Network keystone indices (Jorda´n, 2001 and Libralato et al., 2006) for each ecological subsystem (SS1 and SS2) (Subsystem/species (A) SS1 (1) A. purpuratus (2) sA. purpuratus (3) T. dombeii (4) T. pannosa (5) A. ater (6) T. chocolata (7) L. magallanica (8) Cancer spp. (9) SHE (10) SEC (11) LE (12) Chlorophyta (13) Rhodophyta (14) Phaeophyta (15) Zooplankton (16) Phytoplankton (17) Detritus (B) SS2 (1) A. purpuratus (2) sA. purpuratus (3) T. dombeii (4) T. pannosa (5) T. chocolata (6) L. magallanica (7) Cancer spp. (8) SHE (9) SEC (10) LE (11) Chlorophyta (12) Rhodophyta (13) Phaeophyta (14) Zooplankton (15) Phytoplankton (16) Detritus

Jorda´n’s index Kx

Libralato’s index KSi

1.082 0.226 1.082 1.08 0.77 2.21 6.89 6.89 4.08 4.24 4.48 0.37 0.37 0.37 0.40 5.02 5.87

() 0.36 () 2.14 () 0.73 () 0.73 () 0.84 () 0.29 () 0.30 () 0.25 () 0.65 (1) 0.07 () 0.38 () 0.85 () 1.77 () 0.43 () 1.16 (1) 0.15 () 0.27

1.217 0.258 1.217 1.217 2.330 3.914 3.914 4.217 4.057 3.914 0.392 0.392 0.392 0.392 4.433 5.392

() 0.19 () 1.45 () 0.68 () 0.60 (1) 0.09 () 0.30 () 0.28 () 0.75 (1) 0.06 () 0.37 () 0.85 () 1.84 () 0.42 () 1.14 (1) 0.13 () 0.25

important, standing out. Therefore, it is important to note that, of all the species, the scallop A. purpuratus is the one presenting the highest commercial value and it is illegally exploited nowadays (Ortiz, pers. obs.). Regarding the system recovery time (SRT) magnitudes obtained, the carnivorous sea star L. magallanica had the longest SRT in both subsystems, coinciding with the keystone species behaviour presented in SS1 (sensu Jorda´n’s index; Jorda´n, 2001) (Table 5). It is worth pointing out that no convergence was observed between both keystone indexes and only the character of keystone species (or functional groups) for the group of phytoplankton coincided in both subsystems using Libralato’s index. The latter coincides with the highest ascendency concentration in the phyto-zooplankton complex typical of upwelling-dominated ecosystems. Thus the design and implementation of any exploitation management strategy in La Rinconada Marine Reserve should take into consideration at least the complexity detected in this work using different theoretical bodies of analysis, especially considering the fact that the economically and commercially valuable species (except the sea star L. magallanica) are those that propagate the greatest direct and indirect effects on the subsystems. The trophic models and the simulations carried out in the present work are the first attempt to describe the main flows of Copyright r 2009 John Wiley & Sons, Ltd.

matter and energy occurring in the principal benthic systems of La Rinconada Marine Reserve. Despite this, it is important to stress that new efforts should be made to extend the knowledge obtained so far through the evaluation of the relative importance of bacteria, DOM, and POM in the benthic trophic webs. In this sense, it would also be relevant to know the degree of influence on the benthic webs that the unpredictable aggregations of fish occurring in the reserve could cause. Finally, it is important to note that, despite the known limitations and shortcomings of the Ecopath and Ecosim theoretical frameworks (Christensen and Walters, 2004), it is suggested that both models, constructed as simulations in the present contribution, represent the underlying phenomena in the ecological systems studied only when their short-term dynamics are considered.

ACKNOWLEDGEMENTS This contribution was financed by CORFO (Chile), Grant No. 04CR7IPM-01.

REFERENCES Allen KR. 1971. Relation between production and biomass. Journal of the Fisheries Research Board, Canada 28: 1573–1581. Arias-Gonza´lez J, Nun˜ez-Lara E, Gonza´lez-Salas C, Galzin R. 2004. Trophic models for investigation of fishing effect on coral reef ecosystems. Ecological Modelling 172: 197–212. Avendan˜o M, Cantillanez M. 1996. Efectos de la pesca clandestine sobre Argopecten purpuratus (Lamarck, 1819) en el banco de la Rinconada, II Regio´n. Revista Ciencia y Tecnologı´a del Mar 19: 57–65. Avendan˜o M, Cantillanez M. 2005. Growth and demographic structure of Argopecten purpuratus (Lamarck, 1819) in the La Rinconada Marine Reserve, Antofagasta, Chile. Ciencias Marinas 31:491–503. Avendan˜o M, Le Penenc M. 1998. Presencia de ce´lula auxiliar en la gametoge´nesis de Argopecten purpuratus. Estudios Oceanolo´gicos 17: 13–16. Avendan˜o M, Le Pennec M, Cantillanez M. 2001. Anormalidades de larvas de Argopecten purpuratus (Lamarck, 1819) (Mollusca:Pectinidae) una causal de los problemas en la produccio´n artificial de semillas. Estudios Oceanolo´gicos 20: 32–42. Avendan˜o M, Cantillanez M, Thouzeau G, Pen˜a J. 2007. Artificial collection and early growth of spat of the scallop Argopecten purpuratus (Lamarck, 1819) in the La Rinconada Marine Reserve, Antofagasta, Chile. Scientia Marina 71: 197–205. Baird D, Ulanowicz R. 1993. Comparative study on the trophic structure, cycling and ecosystem properties of four tidal estuaries. Marine Ecology Progress Series 99: 221–237. Begon M, Harper JL, Townsend CR. 1990. Ecology: Individuals, Populations and Communities, 2nd edn. Blackwell Scientific Publications: Oxford. Cantillanez M, Avendan˜o M, Thouzeau G, Le Pennec M. 2005. Reproductive cycle of Argopecten purpuratus (Bivalvia: Pectinidae) in La Rinconada Marine Reserve (Antofagasta, Chile): response to environmental effects of El Nin˜o and La Nin˜a. Aquaculture 246: 181–195. Christensen V. 1995. Ecosystem maturity-towards quantification. Ecological Modelling 77: 3–32. Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) DOI: 10.1002/aqc

M. ORTIZ ET AL.

Christensen V, Pauly D. 1992. Ecopath II: a software for balancing steady-state ecosystem models and calculating network characteristics. Ecological Modelling 61: 169–185. Christensen V, Pauly D. 2004. Placing fisheries in their ecosystem context, an introduction. Ecological Modelling 172: 103–107. Christensen V, Walters C. 2004. Ecopath with Ecosim: methods, capabilities and limitations. Ecological Modelling 172: 109–139. Christensen V, Walters C, Pauly D. 2004. Ecopath with Ecosim: a User’s Guide. Fisheries Centre Research Reports, vol. 12. University of British Columbia, Vancouver (available online at http://www.ecopath.org and http:// www.fisheries.ubc.ca). Costanza R, Mageau M. 1999. What is a healthy ecosystem? Aquatic Ecology 33: 105–115. Daneri G, Dellarossa V, Quin˜ones R, Jacob B, Montero P, Ulloa O. 2000. Primary production and community respiration in the Humboldt Current System off Chile and associated oceanic areas. Marine Ecology Progress Series 197: 41–49. Epstein S. 1997. Microbial food web in marine sediments: I. Trophic interactions and grazing rates in two tidal flat communities. Microbial Ecology 34: 188–198. Escribano R, Rodriguez L, Irribarren C. 1995. Temporal variability of sea temperature in Bay of Antofagasta, northern Chile (1991–1995). Estudios Oceanolo´gicos 14: 39–47. Escribano R, Rosales S, Blanco JL. 2004. Understanding upwelling circulation off Antofagasta (northern Chile): a three-dimensional numerical-modeling approach. Continental Shellfish Research 24: 37–53. Findlay R, White D. 1983. The effects of feeding by the sand dollar Mellita quinquiesperforata (Leske) on the benthic microbial community. Journal of Experimental Marine Biology and Ecology 72: 25–41. Francis R, Hixon M, Clarke M, Murawski S, Ralston S. 2007. Ten commandments for ecosystem-based fisheries scientists. Fisheries 32: 217–233. Gaedke U. 1995. A comparison of whole-community and ecosystem approaches (biomass size distributions, food web analysis, network analysis, simulation models) to study the structure, function and regulation of pelagic food webs. Journal of Plankton Research 6: 1273–1305. Hall SJ. 1999a. Managing fisheries within ecosystems: can the role of reference points be expanded? Aquatic Conservation: Marine and Freshwater Ecosystems 9: 579–583. Hall SJ. 1999b. The Effects of Fishing on Marine Ecosystems and Communities. Blackwell Science: Oxford. Hawkins S. 2004. Scaling up. The role of species and habitat patches in functioning of coastal ecosystems. Aquatic Conservation: Marine and Freshwater Ecosystems 14: 217–219. Heymans J, Baird D. 2000. A carbon flow model and network analysis of the northern Benguela upwelling system, Namibia. Ecological Modelling 126: 9–32. Holling CS. 1973. Resilience and stability of ecological systems. Annual Review of Ecological Systems 4: 1–23. Jarre-Teichmann A, Christensen V. 1998. Comparative modelling of trophic flows in four large upwelling ecosystems: global versus local effects. In Global Versus Local Changes in Upwelling Systems, Durand MH, Cury P, Mendelssohn R, Roy C, Bakun A, Pauly D (eds). ORSTOM Editions: Paris; 425–443. Jorda´n F. 2001. Trophic fields. Community Ecology 2: 181–185. Jorda´n F, Taka´cs-Sa´nta A, Molna´r I. 1999. A reliability theoretical quest for keytones. Oikos 86: 453–462. Jørgensen SE. 1992. Integration of Ecosystem Theories: A Pattern. Kluwer: Dordrecht. Copyright r 2009 John Wiley & Sons, Ltd.

Jørgensen SE. 2000. Application of exergy and specific exergy as ecological indicators of coastal areas. Aquatic Ecosystems Health Management 3: 419–430. Lawton J. 1994. What do species do in ecosystems? Oikos 71: 367–374. Levins R. 1974. The qualitative analysis of partially specified systems. Annals of the New York Academy of Sciences 231: 123–138. Levins R. 1998a. Qualitative mathematics for understanding, prediction, and intervention in complex ecosystems. In Ecosystem Health, Raport D, Costanza R, Epstein P, Gaudet C, Levins R (eds). Blackwell Science: MA; 178–204. Levins R. 1998b. The internal and external explanatory theories. Science as Culture 7: 557–582. Libralato S, Christensen V, Pauly D. 2006. A method for identifying keystone species in food web models. Ecological Modelling 195: 153–171. Marin V, Rodriguez L, Vallejo L, Fuenteseca J, Oyarce E. 1993. Efectos de la surgencia costera sobre la preductividad primaria primaveral de Bahı´ a Mejillones del Sur (Antofagasta, Chile). Revista Chilena de Historia Natural 66: 479–491. Mills L, Soule` M, Doak F. 1993. The key-stone-species concept in ecology and conservation. Bioscience 43: 219–224. Monaco M, Ulanowicz R. 1997. Comparative ecosystem trophic structure of three U.S. mid-Atlantic estuaries. Marine Ecology Progress Series 161: 239–254. Montecinos V, Quiroz D. 2000. Specific primary production and phytoplankton cell size structure in an upwelling area off the coast of Chile (301). Aquatic Science 62: 364–380. Niquil N, Arias-Gonza´lez J, Delesalle B, Ulanowicz R. 1999. Characterization of the planktonic food web of Tarapoto Atoll lagoon, using network ana´lisis. Oecologia 118: 232–241. Ortiz M. 2008a. Mass balanced and dynamic simulations of trophic models of kelp ecosystems near the Mejillones Peninsula of northern Chile (SE Pacific): comparative network structure and assessment of harvest strategies. Ecological Modelling 216: 31–46. Ortiz M. 2008b. The effect of a crab predator (Cancer porteri) on secondary producers versus ecological model predictions in Tongoy Bay (SE Pacific coast): implications to management and fisheries. Aquatic Conservation: Marine and Freshwater Ecosystems 18: 923–929. Ortiz M, Wolff M. 2002a. Trophic models of four benthic communities in Tongoy Bay (Chile): comparative ana´lisis and preliminary assessment of management strategies. Journal of Experimental Marine Biology and Ecology 268: 205–235. Ortiz M, Wolff M. 2002b. Dynamical simulation of massbalance trophic models for benthic communities of north-central Chile: assessment of resilience time under alternative management scenarios. Ecological Modelling 148: 277–291. Ortiz M, Wolff M. 2004. Qualitative modeling for the Caete´ Mangrove Estuary (North Brazil): a preliminary approach to an integrated eco-social analysis. Estuarine, Coastal and Shelf Science 61: 243–250. Paine RT. 1966. Food web complexity and species diversity. The American Naturalist 100: 65–75. Paine RT. 1969. A note on trophic complexity and community stability. The American Naturalist 103: 91–93. Patrı´ cio J, Marques JC. 2006. Mass balanced models of the food web in three areas along a gradient of eutrophication symptoms in the south arm of the Mondego estuary (Portugal). Ecological Modelling 197: 21–34. Patrı´ cio J, Ulanowicz R, Pardal M, Marques J. 2006. Ascendency as ecological indicator for environmental quality assessment at the ecosystem level: a case study. Hydrobiologia 555: 19–30. Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) DOI: 10.1002/aqc

TROPHIC MODEL OF MARINE RESERVE LA RINCONADA

Ulanowicz R. 1986. Growth and Development: Ecosystems Phenomenology. Springer: New York. Ulanowicz R. 1997. Ecology, the Ascendant Perspective. Complexity in Ecological Systems Series. Columbia University Press: New York. Ulanowicz R, Puccia Ch. 1990. Mixed trophic impacts in ecosystems. Ceonoces 5: 7–16. Vasas V, Lancelot Ch, Rousseau V, Jorda´n F. 2007. Eutrophication and overfishing in temperate nearshore pelagic food web: a network perspective. Marine Ecology Progress Series 336: 1–14. Vega-Cendejas M, Arreguı´ n-Sa´nchez F. 2001. Energy fluxes in a mangrove ecosystem from a coastal lagoon in Yucatan Peninsula, Me´xico. Ecological Modelling 137: 119–133. Walters C, Christensen V, Pauly D. 1997. Structuring dynamic models of exploited ecosystems from trophic mass-balance assessments. Reviews in Fish Biology and Fisheries 7: 139–172. Wolff M. 1994. A trophic model for Tongoy Bay — a system exposed to suspended scallop culture (Northern Chile). Journal of Experimental Marine Biology and Ecology 182: 149–168. Wolff M, Koch V, Isaac V. 2000. A trophic flow model of the Caete´ mangrove estuary (North Brazil) with considerations for the sustainable use of its resources. Estuarine Coastal Shelf Sciences 50: 789–803.

Pauly D, Christensen V, Walters C. 2000. Ecopath, Ecosim, and Ecospace as tools for evaluating ecosystem impact of fisheries. ICES Journal of Marine Sciences 57: 697–706. Pikitch E, Santora C, Babcock E, Bakum A, Bonfil R, Conover D, Dayton P, Doukakis P, Fluharty D, Heneman B et al. 2004. Ecosystem-based fishery management. Science 305: 346–347. Pimm S. 1982. Food Webs. Chapman and Hall: New York. Pinneger JK, Polunin NV. 2004. Predicting indirect effects of fishing in Mediterranean rocky littoral communities using a dynamic simulation model. Ecological Modelling 172: 249–267. Plante C, Shriver A. 1998. Patterns of differential digestion of bacteria in deposit feeders: a test of resource partitioning. Marine Ecology Progress Series 163: 253–258. Polovina J. 1984. Model of a coral reef ecosystem I. ECOPATH model and its application to French Frigate Shoals. Coral Reefs 3: 1–11. Rodriguez L, Marı´ n V, Farı´ as M, Oyarce E. 1991. Identification of an upwelling zone by remote sensing and in situ measurements. Mejillones del Sur Bay (AntofagastaChile). Scientia Marina 55: 467–473. Thompson RC, Norton TA, Hawkins SJ. 2004. Physical stress and biological control regulate the producer-consumer balance in intertidal biofilms. Ecology 85: 1372–1382. Scotti M, Podan J, Jorda´n F. 2007. Weighting, scale dependence and indirect effects in ecological networks: a comparative study. Ecological Complexity 4: 148–159.

APPENDIX A: MODELS DATA SOURCES Compartments Species/Functional groups

Parameter C

(1) A. purpuratus SS1 subsystem SS2 subsystems (2) sA. purpuratus SS1 subsystem SS2 subsystems (3) T. dombeii SS1 subsystem SS2 subsystems (4) T. pannosa SS1 subsystem SS2 subsystems (5) A. ater SS1 subsystem SS2 subsystems (6) T. chocolata SS1 subsystem SS2 subsystems (7) L. magallanica SS1 subsystem SS2 subsystems (8) Cancer spp. SS1 subsystem SS2 subsystems (9) SEH SS1 subsystem SS2 subsystems (10) SEC SS1 subsystem SS2 subsystems (11) LE SS1 subsystem SS2 subsystems (12) Chlorophyta SS1 subsystem SS2 subsystems

1

2

B

3

P/B

4

Q/B

Diet5

Literature source

224.9 46.1

259.0 53.6

2.7 2.7

9.9 9.9

12,3

2.5 2.5

50.0 25.0

2.0 2.0

9.9 9.9

12

0.01 0.2

15.0 314.2

2.6 2.0

9.9 9.9

1,3,5

0.01 0.2

14.0 78.7

2.8 2.8

9.9 9.9

1,3,5

0.2

12.0

2.1

9.9

1,3,5

0.16 0.5

14.2 48.9

2.7 2.7

7.2 7.2

1,3,4,5

0.8 0.0

0.5 0.5

3.0 3.0

1,3,5

5.5 5.0

1.9 1.9

9.5 9.5

1,3,4,5

20.0 20.0

2.8 2.8

11.7 11.7

1,3,5

13.0 15.0

2.4 2.4

10.4 10.4

1,3,5

7.1 12.5

1.9 1.9

9.2 9.2

1,3,5

10.0 10.0

5.0 5.0

Copyright r 2009 John Wiley & Sons, Ltd.

Field estimations for current work Ortiz and Wolff (2002a)

4,5

Field estimations for cuurent work Estimated by Ecopath, 4,5Ortiz and Wolff (2002a)

3

Field estimations for current work Ortiz and Wolff (2002a)

4

Field estimations for current work Ortiz and Wolff (2002a)

4

Field estimations for current work Ortiz and Wolff (2002a)

4

Field estimations for current work

Field estimations for current work Ortiz and Wolff (2002a)

4

Field estimations for current work

Field estimations for current work Ortiz and Wolff (2002a)

4

Field estimations for current work Ortiz and Wolff (2002a)

4

Field estimations for current work Ortiz and Wolff (2002a)

4 1

Field estimations for current work Estimated by Ecopath

3

Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) DOI: 10.1002/aqc

M. ORTIZ ET AL.

(13) Rhodophyta SS1 subsystem SS2 subsystems (14) Phaeophyta SS1 subsystem SS2 subsystems (15) Zooplankton SS1 subsystem SS2 subsystems (16) Phytoplankton SS1 subsystem SS2 subsystems

Copyright r 2009 John Wiley & Sons, Ltd.

83.5 255.0

5.0 5.0

1

20.0 20.0

5.0 5.0

1

20.0 20.0

40.0 40.0

30.0 30.0

250.0

Field estimations for current work Estimated by Ecopath

3

Field estimations for current work Estimated by Ecopath

3

160.0 160.0

1,3,4,5

Ortiz and Wolff (2002a)

1,3,4,5

Ortiz and Wolff (2002a)

250.0

Aquatic Conserv: Mar. Freshw. Ecosyst. (2009) DOI: 10.1002/aqc