Macroscopic network properties and short-term dynamic simulations in coastal ecological systems at Fildes Bay (King George Island, Antarctica)

G Model ECOCOM 587 No. of Pages 13

Ecological Complexity xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Ecological Complexity journal homepage: www.elsevier.com/locate/ecocom

Original research article

Macroscopic network properties and short-term dynamic simulations in coastal ecological systems at Fildes Bay (King George Island, Antarctica) Marco Ortiza,* , Fernando Berriosa,b , Jorge Gonzáleza,b , Fabián Rodríguez-Zaragozac , Iván Gómezd a Instituto de Antofagasta (IA), Instituto de Ciencias Naturales AvH, Facultad de Ciencias del Mar & Recursos Biológicos, Universidad de Antofagasta, P.O. Box 170, Antofagasta, Chile b Programa de Doctorado en Ciencias Aplicadas, Mención Sistemas Marinos Costeros, Universidad de Antofagasta, Chile c Laboratorio de Ecosistemas Marinos y Acuicultura (LEMA), Departamento de Ecología, CUCBA, Universidad de Guadalajara, Carretera Guadalajara-Nogales Km. 15.5, Las Agujas Nextipac, Zapopan 45110, Jalisco, Mexico d Instituto de Ciencias Marinas y Limnólogicas, Universidad Austral de Chile, Casilla 567, Valdivia, Chile

A R T I C L E I N F O

Article history: Received 10 November 2015 Received in revised form 15 June 2016 Accepted 20 June 2016 Available online xxx Keywords: Ascendency Direct and indirect effects South Shetland Islands Subtidal benthic habitats

A B S T R A C T

A trophic mass-balanced of the benthic/pelagic system dominated by large brown macroalgae in Fildes Bay (Antarctica) was constructed by integrating biomass, production, food spectrum, and consumption related information. The resulting trophic model was used to determine the macroscopic (emergent) properties, overall health and propagation of dynamical higher order effects within this complex Antarctic ecological system in response to simulated impacts. The magnitude of the Relative Ascendency, Relative Overhead, and Redundancy values indicates that the coastal benthic/pelagic Fildes Bay system is likely to remain less developed and therefore more resistant to perturbations than other ecological systems dominated by brown macroalgae. In terms of model component contributions to the Ascendency, detritus accounted for
33% of the value, followed by the phyto-zooplankton complex (
26%), macroalgae (
19%), filter-feeders (
7%), small epifauna (5%), and top predators (2%). Short-term or transient Ecosim dynamical responses to increase the total mortality of each model component-given mixed and top-down vulnerabilities-revealed that changes in macroalgae levels had a limited impact on the other components of the system. The filter feeder, small epifauna and benthic fishe’s functional groups had the greatest effects on the remaining Fildes Bay system components. The magnitude of the System Recovery Time indicated that the Nacella concinna and small epifauna components would take the longest time to return to their initial state. Based on the outcomes obtained from the model, we suggest that this preliminary trophic model, including simulated impacts, provides promising possibilities for the determination of macroscopic baseline conditions and the most sensitive components of the Fildes Bay ecological system. ã 2016 Elsevier B.V. All rights reserved.

1. Introduction The Antarctic continent has been isolated geographically for 40–65 million years. The coastal ecosystems of the Antarctic Peninsula and the South Shetland Islands are home to a rich, dense array of flora and fauna that are physiologically adapted to cold waters and low-light conditions (Clarke et al., 2004; Dayton et al., 1974; Gili et al., 2001; Gómez et al., 2009). Many macroalgae,

* Corresponding author. E-mail addresses: [email protected], [email protected] (M. Ortiz).

invertebrate, demersal fish and marine mammal species can be found within these communities (Costa and Crocker, 1996; Gili et al., 2001; Iken et al., 1997; Klöser et al., 1993; Richardson, 1979). However, while these species and their habitats are protected by the “Antarctic-Environmental Protocol”, the Antarctic faces several stressors. The average air temperature on the Antarctic Peninsula has increased between 4  C and 5  C over the past 20 years (King, 1994; Stark, 1994). This change in the environment could have long-term consequences for sea ice and ice shelf-dynamics (i.e., glacial melting) (Vaughan and Doake, 1996; Smith and Stammerjohn, 2001) and could modify the amount of direct light macroalgae and other organisms that inhabit subtidal and intertidal

http://dx.doi.org/10.1016/j.ecocom.2016.06.003 1476-945X/ã 2016 Elsevier B.V. All rights reserved.

Please cite this article in press as: M. Ortiz, et al., Macroscopic network properties and short-term dynamic simulations in coastal ecological systems at Fildes Bay (King George Island, Antarctica), Ecol. Complex. (2016), http://dx.doi.org/10.1016/j.ecocom.2016.06.003

G Model ECOCOM 587 No. of Pages 13

2

M. Ortiz et al. / Ecological Complexity xxx (2016) xxx–xxx

systems are exposed to each day. Additionally, ultraviolet (UV) radiation levels have fluctuated from the norm since the development of the ozone hole in the Antarctica as a consequence of the anthropogenic release of atmospheric pollutants (i.e., fluorocarbons, halocarbons, chlorocarbons, dioxins and carbon dioxide) (Pessoa, 2012). Thus, Antarctic coastal ecosystems are under severe stress as a direct consequence of global warming, which could facilitate biological invasions, and the increase in UV radiation, which could reduce the growth of various macroalgae species (Richter et al., 2008). Historically, a diverse array of studies have examined environmental conditions, changes in the ecophysiology of key organisms under regional and global climate changes, population and community ecology, and trophic ecology on the Antarctic

Peninsula (Corbisier et al., 2004; Gili et al., 2001; Mincks et al., 2008; McClintock, 1994; Norkko et al., 2007; Pinkerton et al., 2010; Smith et al., 2007; Targett, 1981; Valdivia et al., 2014). Despite numerous attempts to establish trophic connections among the different benthic and demersal species that inhabit subtidal systems, few studies have addressed the macroscopic or systemic properties of coastal Antarctic ecosystems nor have they determined the most sensitive components of the ecological network. Therefore, complementary theoretical frameworks that integrate a finite and interconnected set of species and/or functional groups to represent and describe the structure and dynamics of such ecological systems are required (Pickitch et al., 2004). The application of network analyses based on multispecies or ecosystem models allows us to estimate the emergent or

Fig. 1. Study area and samples stations at Fildes Bay (King George Island, Antarctica) (GC = Glacier Collins, IA = Island Artigas; Island Shoa, HT = Half Three, EN = Estrecho Nelson, and GN = Glacier Nelson) (a); and trophic model diagram for the coastal ecological system of Fildes Bay. Vertical position approximates trophic level. The circle size is proportional to the compartment (populations and/or functional groups) biomass (g wet weight m2). For details see Table 1(b).

Please cite this article in press as: M. Ortiz, et al., Macroscopic network properties and short-term dynamic simulations in coastal ecological systems at Fildes Bay (King George Island, Antarctica), Ecol. Complex. (2016), http://dx.doi.org/10.1016/j.ecocom.2016.06.003

G Model ECOCOM 587 No. of Pages 13

M. Ortiz et al. / Ecological Complexity xxx (2016) xxx–xxx

macroscopic properties (sensu Odum, 1969; Ulanowicz, 1986, 1997), overall health (Costanza and Mageau, 1999) and to assess the propagation of direct and indirect effects, within complex ecological systems (Hawkins, 2004; Levins, 1998a). This systemlevel analysis should be considered as a complementary strategy to classic isolated-reductionist studies and other kind of network analyses (e.g. Calizza et al., 2015; Eklöf et al., 2013; Rossi et al., 2015). Recent studies have been conducted in Fildes Bay on King George Island (South Shetland Islands) to assess the temperature and UV stress tolerance of Antarctic macroalgae species (Huovinen and Gómez, 2013; Rautenberger et al., 2015) and to describe the spatial variation of intertidal and subtidal rocky-shore communities (Valdivia et al., 2014). These studies were conducted to quantify the effects of global climate change on the physiology of macroalgae, as well as the ecology of Antarctic marine species. However, a more holistic approach is required if we want to assess the macroscopic properties and ecosystem heath of the coastal ecological systems within Fildes Bay. Such an approach requires the construction of an ecological network based on the flow of energy and/or matter (Odum, 1969; Ulanowicz, 1986, 1997). The primary objective of this work was to build a trophic model that represents the predator-prey and/or consumer-resource relationships between the most abundant coastal species in the benthic/pelagic ecological system of Fildes Bay using Ecopath with Ecosim software package (Walters et al., 1997). The macroscopic system properties and dynamical simulations were used as sensitivity analyses in order to determine the most sensitive model compartments. The impacts on this ecological model system were entered as follows: (1) a steady increase in the total mortality of each model component until the fifth year in the simulation; and (2) an increase in the vulnerability of algae to consumption by herbivores from the mixed to top-down flow control mechanisms (from v = 0.3 to v = 1.0, respectively). The outcomes of this sensitivity analysis will be used to highlight what species and/or functional groups should be monitored for assessment the putative effects produced by global changes in this ecological system. Based on this model and the associated network analysis, the other system-level attributes evaluated included the following: (1) the consumption rates and species preyed on by the primary benthic predators in the system; (2) the macroscopic (emergent) properties of Fildes Bay’s coastal benthic/ pelagic system based on Ulanowicz’s theoretical framework; (3) the ecological redundancy within the system, which was defined as the number of species or functional groups with similar trophic functions (parallels flows) in the system (sensu Lawton, 1994); (4) the degree of resistance to disturbances and the resilience of the ecological system in response to different perturbation scenarios; and (5) the species and functional groups that would be most affected by simulated perturbations. 2. Materials and methods 2.1. Study area and bay characteristics Fildes Bay is located along King George Island, Antarctic Peninsula (Fig. 1a) and is 14 km long and 6–14 km wide. The surface of the bay and its coastal waters freeze regularly during the austral winter (from July to September). After late October, the sea-ice begins to crack and floating ice reaches the shore (Bick and Arlt, 2013; Valdivia et al., 2014). The subtidal communities are dominated by the brown macroalgae species Himmantothallus grandifolius and Desmarestia anceps; the red algae species Gigartina skottsbergii, Trematocarpus antarcticus and Plocamium cartilagineum; the grazer species (gastropod) Nacella concinna; the asteroid predator species Diplasterias brucei; and a variety of fish, sponge,

3

bryozoan and ascidian species (Targett, 1981; Huovinen and Gómez, 2013; Valdivia et al., 2014). 2.2. Ecopath with Ecosim (v.5.0) modelling approaches In this work, a trophic model was constructed using Ecopath with Ecosim software (www.Ecopath.org). Ecopath was first developed by Polovina (1984) and was further expanded by Christensen and Pauly (1992) and Walters et al. (1997). Ecopath permits a steady-state description of the matter/energy flows within an ecosystem at a particular time, while Ecosim enables dynamic simulations based on an Ecopath model to provide an estimation of ecosystem changes as a consequence of a set of perturbations. The multispecies models based on Ecopath and Ecosim have been widely used to describe and compare a variety of emergent properties (Ulanowicz, 1986, 1997) associated with ecosystems of varying spatial sizes, geographic locations, and complexities (e.g., Arias-González et al., 2011; Christensen and Pauly, 1992; Christensen and Walters, 2004; Guénette et al., 2008; Monaco and Ulanowicz, 1997; Ortiz, 2008a; Ortiz and Wolff, 2002; Pinkerton and Bradford-Grieve, 2014; Preikshot et al., 2013). For additional details related to the Ecopath and Ecosim frameworks, see Appendix A. 2.3. Selection of model components, sampling programmes and data sources Three intensive field studies were conducted during the austral summers of 2013, 2014 and 2015 to identify the biological components (species or functional groups) of the system model and to estimate the average biomass (B), average density, and food sources of the selected components (Valdivia et al., 2014). Sampling was performed to directly estimate the average biomass and density of the macrobenthic species (between 5 m and 30 m depth) at six stations within Fildes Bay (Fig. 1a). The production (P) and turnover rates (P/B) were estimated using the following allometric equation (Eq. (1)): "
#  Biomass 0:73 ð1Þ  Density Production ¼ Density where 0.73 is the average exponent regression of annual production on body-size for macrobenthic invertebrates (for more details see Warwick and Clarke, 1993). Food consumption rates were obtained from the literature (Cornejo-Donoso and Antezana, 2008; Ortiz, 2008a; Ortiz et al., 2015; Pinkerton et al., 2010). To determine the diets of N. concinna, Harpagifer antarcticus, Notothenia coriiceps, N. rosii and the asteroid species, the stomach and guts were revised and the gut contents were classified to the lowest possible taxonomic level; the frequency of occurrence of each food item was then calculated. Several studies examining the trophic ecology of several benthic and pelagic species were also used to determine the range of food consumed. Appendix B lists the sources for the data used and selected for the current work. A trophic model with 17 components was constructed for Fildes Bay. The components represent the most abundant individual species or functional groups composed of multiple species. Seven components represented the following individual species: the brown macroalgae H. grandifolius and D. anceps, the red algae G. skottsbergii, the herbivores N. concinna and Margarella sp., the echinoid Sterechinus neumayeri, and the asteroid D. brucei. The other components were functional groups that included several species. The Chlorophyta, Rhodophyta, and Phaeophyta groups included multiple species of green (e.g., Monostroma hariotii), red (e.g., T. antarcticus and P. cartilagineum), and brown (e.g., Ascoceira

Please cite this article in press as: M. Ortiz, et al., Macroscopic network properties and short-term dynamic simulations in coastal ecological systems at Fildes Bay (King George Island, Antarctica), Ecol. Complex. (2016), http://dx.doi.org/10.1016/j.ecocom.2016.06.003

G Model ECOCOM 587 No. of Pages 13

4

M. Ortiz et al. / Ecological Complexity xxx (2016) xxx–xxx

mirabilis and Halopteris obata) algae, respectively. The filter feeder (FF) group was comprised primarily of several clam, hydrozoan, bryozoan, and sponge species. The small epifauna (SE) component included several gastropod, nematode, and nemertean species. The group of other sea star species (SS) included principally O. validus and Odontaster sp. The benthic fishes (BF) group was comprised primarily of H. antarcticus, N. coriiceps and N. rosii. The three final groups were the phytoplankton (Phy), zooplankton (Zoo) and detritus (Det) groups (Table 1 and Fig. 1b). All the compartments are trophically linked by detritus primarily as microbial film and organic matter- because several studies have emphasised the importance of bacteria as food for various species of molluscs (e.g., Epstein, 1997; Grossmann and Reichardt, 1991; Plante and Mayer, 1994; Plante and Shriver, 1998), zooplankton (Epstein, 1997), and Echinodermata (Findlay and White, 1983). The interaction matrix for the benthic system is shown in Appendix C. The models were constructed to depict the trophic relationships between the most important species or functional groups in the benthic communities of Fildes Bay. Notably, the models excluded the energy flows from epiphytes and the microphytobenthos, in addition to those leading to seals and birds, because insufficient scientific information was available for these groups. Although these exclusions reduced the realism of the model’s configuration, the most dominant interdependencies and energy flows are reflected. Moreover, such a system-level error should not impede a comparative analysis of ecological systems placed under similar limitations (e.g., kelp forest of SE pacific coast). 2.4. Macrodescriptors and network properties Ecopath-based trophic models combine Polovina’s approach (1984) of estimating the biomass and food consumption of individual species or functional groups with that of Ulanowicz (1986, 1997) ecosystem and network analyses of flows between model components to calculate the emergent properties of ecosystems. The descriptors used within the model include Total System Throughput (TST), Ascendency (A), Overhead (Ov), Development Capacity (C), and A/C and Ov/C ratios. Total System Throughput describes the size or activity of a system in terms of flows and attempts to quantify the metabolism of a system. Ascendency Table 1 Parameter values entered (in bold) and estimated (standard) by Ecopath II software for the coastal ecological system of Fildes Bay (Antarctica). (Note: TL = trophic level, B = biomass [g wet weight m2], P/B = turnover rate [year1], Q/B = consumption rate [year1], EE = ecotrophic efficiency [dimensionless], and GE = gross efficiency [dimensionless]). Compartments Species/functional groups

TL

B

P/B

Q/B

EE

GE

(1) Benthic Fishes (BF) (2) Diplasterias brucei (Db) (3) Seastars (SS) (4) Small Epifauna (SE) (5) Sterechinus neumayeri (Sn) (6) Margarella sp. (Msp) (7) Nacella concinna (Nc) (8) Filter Feeders (FF) (9) Himmantothalus grandifolius (Hg) (10) Desmarestia anceps (Da) (11) Gigartina skottsbergii (Gs) (12) Phaeophyta (Phaeo) (13) Chlorophyta (Chloro) (14) Rhodophyta (Rhodo) (15) Zooplankton (Zoo) (16) Phytoplankton (Phy) (17) Detritus (Det)

3.03 2.98 3.29 2.76 2.35 2.10 2.05 2.30 1.00

32.67 37.83 18.52 66.50 34.50 59.50 56.50 177.46 597.67

2.40 1.40 1.40 4.60 3.70 3.40 3.60 1.87 2.10

8.50 5.00 5.00 15.50 13.50 11.50 12.50 8.50 –

0.05 0.18 0.04 0.95 0.99 0.99 0.99 0.93 0.35

0.28 0.28 0.28 0.30 0.27 0.30 0.29 0.22 –

1.00 1.00 1.00 1.00 1.00 2.00 1.00 1.00

310.50 35.17 650.50 38.50 174.57 122.50 300.00 100.00

2.30 2.50 2.80 2.40 5.00 4.60 18.00 –

– – – – – 15.50 – –

0.38 0.42 0.39 0.36 0.32 0.99 0.54 0.07

– – – – – 0.30 – –

integrates both the size and organisation of a system. Organisation or complexity is measured by the Average Mutual Information (AMI), which refers to the number and diversity of weighted interactions among system components. Ascendency and Overhead are related to system stability (Christensen, 1995; Cropp and Gabric, 2002) and maturity (Cropp and Gabric, 2002; Fath et al., 2001; Perez-Espana and Arreguin-Sanchez, 2001; Ulanowicz and Abarca-Arenas, 1997). Overhead is the difference between Ascendency and the system’s Development Capacity, and it indicates the multiplicity of information pathways (as a measure of redundancy) that may be closely related to a system’s capacity to withstand perturbations (Christensen, 1995; Angelini and Petrere, 2000). Development Capacity quantifies the upper limit of the Ascendency, and the A/C and Ov/C ratios are also used as indicators of ecosystem development (Kaufman and Borrett, 2010) and a system’s ability to resist disturbances (Ulanowicz, 1986, 1997). The A/C and Ov/C ratios are inversely related; that is, as higher A/C and lower Ov/C indicates a more developed and mature ecosystem, but less resistant to perturbations. These macrodescriptors have all been widely used to describe and compare a variety of ecosystems of different sizes, geographic locations, and complexities (e.g., AriasGonzález et al., 2004; Bayle-Sempere et al., 2013; Díaz-Uribe et al., 2012; Heymans and Baird, 2000; Kaufman and Borrett, 2010; Li and Yang, 2011; Monaco and Ulanowicz, 1997; Panikkar and Khan, 2008; Patrício and Marques, 2006; Ortiz et al., 2015). 2.5. Balance of trophic models The model balancing procedure required two steps. The first step was to determine whether the model outputs were realistic, which was proven when the Ecotrophic Efficiency (EE) was <1.0 for all variables or components (Ricker, 1968). When an inconsistency was detected, the biomass values (average) were adjusted slightly within a standard deviation (1 SD) of the mean obtained from the field studies. In the case of Rhodophyta, Chlorophyta and Phaeophyta, the turnover rates (P/B) were estimated using the Ecopath software. When balancing the models, modification of the diet matrices was not required. The second step was to define the variation in the magnitude of the Gross Efficiency (GE), defined as the production/consumption ratio. This value was set to fluctuate between 0.1 and 0.3 for all components (for additional details, see Christensen et al., 2004). 2.6. Ecosim simulations and system recovery times The dynamic simulations were performed – as sensitivity analyses – to quantify the propagation of direct and indirect ecosystem effects and the System Recovery Time (SRT, an internal measure of stability) of the Fildes Bay ecological system. The impacts on the network were simulated as follows: (1) a steady increase in the total mortality (Z) of all components (see Eqs. (2) and (3)) which was set equivalent to 10%, 30% and 50% until the fifth year of the simulation. These three magnitudes were set only for prediction purposes as a measure of confidence; and (2) an increase in the vulnerability (v) of macroalgae to consumption by herbivores from v = 0.3 (mixed flow control) to v = 1.0 (top-down flow control). The vulnerability of the herbivore group was simultaneously simulated with the total mortality increments (10%, 30% and 50%) (see Fig. 2a). The simulations performed will permit to determine the most sensitive compartments, which could be considered in a putative monitoring program for assessment of impacts of the global change (i.e. increases in temperature and UV radiation) in Fildes Bay. Z ¼ MðnaturalmortalityÞ þ F ðf ishingmortalityÞ

ð2Þ

Please cite this article in press as: M. Ortiz, et al., Macroscopic network properties and short-term dynamic simulations in coastal ecological systems at Fildes Bay (King George Island, Antarctica), Ecol. Complex. (2016), http://dx.doi.org/10.1016/j.ecocom.2016.06.003

G Model ECOCOM 587 No. of Pages 13

M. Ortiz et al. / Ecological Complexity xxx (2016) xxx–xxx

5

Fig. 2. Simulated scenarios of steady increasing up to 10%, 30% and 50% of total mortality (Z) for each model compartment until fifth year. BM = baseline (Ecopath) mortality. For more details see Material and Methods (a); and simulated changes of Ascendency magnitudes (macroscopic properties) as response to the steady increase of total mortality (10%, 30% and 50% (b).

Production ¼ Biomass  Z

ð3Þ

The propagation of short-term or transient responses was determined by evaluating the changes in biomass for all of the components in the sixth year of the simulation, i.e., one year after the maximal increase in total mortality (Fig. 2a). All components, with the exception of macroalgae, were simulated using a mixed flow control mechanism (v = 0.3), which is considered to be more realistic than the bottom-up or top-down mechanisms (e.g., Hunter and Price, 1992; Krebs et al., 1995; Levins, 1998b; Masterson et al., 2008; Muhly et al., 2013). Furthermore, the use of a mixed flow control mechanism (both bottom-up and top-down) was recently shown to produce the highest certainties in predictions (Thompson et al., 2004; Ortiz, 2008b). Notably, the Ecosim simulations for a benthic system in northern Chile based on short-term dynamics indicated up to
60% certainty in the predictions using a mixed flow control mechanism (Ortiz, 2008b). 3. Results The coastal benthic/pelagic ecological system of Fildes Bay dominated by brown large macroalgae reached a Total System Throughput (TST) equals to 24,234.0 g ww m2 year1. The Development Capacity (C) accounted for 110,354.4 Flow bits, Ascendency (A) accounted for 32,953.9 Flow bits, and the A/C and Ov/C ratios were 29.8% and 70.1%, respectively (Table 2). Notably, the A/C value calculated for the Fildes Bay ecological system was one of the

Table 2 Network flow indices for the coastal ecological system of Fildes Bay (King George Island, Antarctica) after mass-balance model by Ecopath II. The units are in g wet weight and Flow bit is the product of flow (g wet weight m2 year1) and bits. Network flow indices Total System Throughput (TST) (g ww m2 year1) Ascendency (A) (g ww m2 year1 * bits) Overhead (Ov) (g ww m2 year1 * bits) Development Capacity (C) (g ww m2 year1 * bits) Average Mutual Information (AMI) Seastar and G. skottsbergii accounting for the lowest% of AMI Pathway redundancy (internal flows of Overhead) (%) A/C (%) Ai/Ci (%) Ov/C (%) Finn’s cycling index (FCI) (%) Finn’s mean path length (dimensionless) Food web connectance (dimensionless) Omnivory Index (OI) (dimensionless)

21432.00 29758.60 69355.02 99114.87 1.39 0.36 56.00 30.02 14.00 69.97 1.09 2.42 0.28 0.22

lowest compared to those obtained for other coastal areas along the Chilean coast and around the globe (Table 3). The difference between the A/C and the Ai/Ci ratios for the Fildes Bay model may indicate a dependency of this system on external connections (sensu Baird et al., 1991). Fig. 2b shows the changes estimated for Ascendency as a response to the steady increase in the total mortality within each group. In this case, Ascendency was negatively related to increasing mortality, and this macroscopic property may be also used to measure system’s resilience.

Please cite this article in press as: M. Ortiz, et al., Macroscopic network properties and short-term dynamic simulations in coastal ecological systems at Fildes Bay (King George Island, Antarctica), Ecol. Complex. (2016), http://dx.doi.org/10.1016/j.ecocom.2016.06.003

G Model ECOCOM 587 No. of Pages 13

6

M. Ortiz et al. / Ecological Complexity xxx (2016) xxx–xxx

Table 3 Macroscopic properties (indices) for system development and organization derived from Ascendency network analysis for Fildes Bay (Antarctica) and other coastal ecosystems. Coastal Marine Ecosystems

Emergent or Macroscopic Ecosystem Properties

Overhead (flow bits)

Ov/C %

Ascendency theoretical framework (Ulanowicz, 1986, 1997) Throughput (g ww m2 year1)

Capacity (flow bits)

Ascendency (flow bits)

A/C %

A. Along the Chilean coast (SE Pacific) Benthic/pelagic coastal ecological system of Fildes Baya Kelp ecological system dominated by M. pyrifera, Antofagasta Peninsulab Kelp ecological system dominated by L. trabeculata, Antofagasta Peninsulab Kelp forest ecological system, Antofagasta Peninsulac Seagrass habitat ecological system of Tongoy Bayd Mud habita ecological system of Tongoy Bayd Benthic/pelagic ecological system of Tongoy Baye La Rinconada Marine Reserve coastal ecological system, Antofagasta Bayf Mejillones benthic/pelagic ecological system of Mejillones Bayg Antofagasta benthic/pelagic ecological system of Antofagasta Bayg

21452.5 72512.0 50105.0 85217.0 18746.6 17451.3 20834.9 20124.0 29429.8 37539.8

99415.9 207777.4 200609.4 332041.6 69270.4 59139.0 80689.8 80321.0 142897.9 170237.0

29749.3 93462.6 77613.5 117939.7 21557.8 19354.8 26312.6 24375.1 34395.1 48574.3

29.9 45.0 38.7 35.5 31.1 32.7 32.6 30.3 24.1 28.5

69665.4 112548.0 117678.9 211848.3 46991.0 39433.4 54377.2 55945.9 108353.1 121434.8

70.1 55.0 61.3 64.5 68.9 67.3 67.4 69.7 75.9 71.5

B. Around the world Coral reef ecosystem, Chinchorro Bank, Méxicoh Mangrove estuary of Caeté, Brazili Zostera meadows of Mondego Estuary, Portugalj Ems estuary in The Netherlandsk Benguela upwelling ecosystem, Namubial

148094.1 10558.6 10852.0 12980.0 8897.0

318400.0 44741.4 39126.0 6085.0 36041.0

178200.0 12261.6 16550.3 2327.0 17313.0

56.3 27.0 42.3 38.3 48.1

139800.0 31129.8 22575.7 3758.0 18728.0

43.7 63.0 57.7 61.7 51.9

a b c d e f g h i j k l

Current study. Ortiz (2008a). Ortiz (2010). Ortiz and Wolff (2002). Wolff (1994). Ortiz et al. (2010). Ortiz et al. (2015). Rodriguez-Zaragoza et al. (submitted). Wolff et al. (2000). Patrício and Marquez (2006). Baird and Ulanowicz (1993). Heymans and Baird (2000).

Ulanowicz (1997) proposed estimating Relative Ascendency by model component to evaluate the contribution of each group to the overall structure and function of the system. In this case, detritus accounted for
33%, followed by phyto-zooplankton at
26%, macroalgae at
19%, filter feeders at
7%, small epifauna at
5%, and top predators at
2%. Moreover, the seastar species groups, Chlorophyta and the red algae G. skottsbergii accounted for the complexity in the system; that is, they exhibited the lowest% of Average Mutual Information (AMI) within the Fildes model system (Table 2). The Ecosim short-term dynamic simulations showed that the macroalgae propagated more effects to other model components, using the three mortality scenarios and the two flow control mechanisms, than any other species or functional group. Nevertheless, the magnitude of these effects were the lower than those found for filter feeders, small epifauna and benthic fishes, which had the greatest quantitative effects on the other groups (Fig. 3). Used as a measure of resilience, System Recovery Time (SRT) values for the Fildes model system are summarised in Table 4. The macroalgae had the lowest SRT values (5.0–5.3 years) given the three mortality and two vulnerability scenarios. By contrast, the small epifauna (SRT = 14.0–16.25 years) and grazer N. concinna (SRT = 9.5–13.75 years) components would require the greatest amount of time to return to their initial conditions (Table 4). 4. Discussion Although we were well aware that the quantitative trophic model built and analysed in this study was a partial representation of the overall trophic makeup and interactions underlying the dynamics within Fildes Bay’s coastal benthic/pelagic ecological

system, such limitations, however, occur in any type of model and are independent of the model’s degree of complexity (Levins, 1966; Ortiz and Levins, 2011). In the present model, the following limitations were identified: (1) the model represented the austral summer condition only, as annual benthic/pelagic dynamics are unknown; (2) system complexity was reduced in relation to the composition of several functional groups, although the most abundant macroalgae, herbivore and carnivore species were represented; and (3) regardless of the inherent, well-known limitations and shortcomings of the Ecopath and Ecosim theoretical frameworks, the constructed model and its dynamic simulations represented underlying system processes based exclusively on short-term or transient dynamics. However, in spite of these constraints, the most important trophic relationships and energy/ matter flows were reflected in the model. Thus, the macroscopic properties of the system were adequately quantified and compared; and the most sensitive model compartments were also determined. Remarkably, the model indicated that >30% of total system biomass accumulated exclusively within the brown algae species D. anceps and H. grandifolius, which may indicate the relative stability of this system in the face of physical perturbations (e.g. coastal currents), particularly because these species occupy large patches of space, and the blades of H. grandifolius extend >20 m in length. In terms of the network and structure of the ecological system, the magnitude of the throughput (TST) estimated for this system (24,234.0 g ww m2 year1) (Table 3) was lower than that calculated for a kelp forest off the Antofagasta Peninsula (Ortiz, 2008a, 2010). However, compared to the benthic communities of Tongoy Bay (Ortiz and Wolff, 2002; Wolff, 1994) and different estuaries around the world (Baird and Ulanowicz, 1993; Patrício

Please cite this article in press as: M. Ortiz, et al., Macroscopic network properties and short-term dynamic simulations in coastal ecological systems at Fildes Bay (King George Island, Antarctica), Ecol. Complex. (2016), http://dx.doi.org/10.1016/j.ecocom.2016.06.003

G Model ECOCOM 587 No. of Pages 13

M. Ortiz et al. / Ecological Complexity xxx (2016) xxx–xxx

7

Increase of mortality Mixed flow control (v = 0.3) 10%

30%

50% Benthic Fishes

Benthic Fishes

Benthic Fishes 25 0 -25 25

Diplasterias brucei

Diplasterias brucei

Diplasterias brucei

0 -25 25

Seastars

Seastars

Seastars

0 -25 25

Small Epifauna

Small Epifauna

Small Epifauna

0

Final/Inital biomass

-25 25

Sterechinus neumayeri

Sterechinus neumayeri

Sterechinus neumayeri

Margarella sp.

Margarella sp.

Margarella sp.

Nacella concinna

Nacella concinna

Nacella concinna

0 -25 25 0 -25 25 0 -25 25

Filter feeders

Filter feeders

Filter feeders

0 -25 0.008 Himantothalus grandifolius

Himantothalus grandifolius

Himantothalus grandifolius

0.000 -0.008 0.008

Desmarestia anceps

Desmarestia anceps

Desmarestia anceps

0.000 -0.008 1

5

9

13

17

1

5

9

13

17 1

5

9

13

17

Functional groups 2

Fig. 3. Dynamical changes in biomass (g wet weight m ) of all model compartments subject to 4 years of increased mortality (between 1 and 5 year of simulation) under mixed (v = 0.3) and top-down (v = 1.0) flow control mechanisms (only for the grazers of macroalgae). The responses of biomass were obtained for the sixth year of the simulation. (Note: the numbers on the x-axis correspond to the species or functional groups of Table 1, and please note a difference scale on the y-axis.)

and Marques, 2006; Wolff et al., 2000), T was higher in the Fildes Bay system. Similarly, in terms of the system’s macroscopic properties, such as the A/C ratio, Ov/C ratio and Redundancy values, Fildes Bay would be a less developed system but is more resistant to disturbances than a kelp forest off the Mejillones Peninsula or the sea grass meadows of Tongoy Bay (Ortiz and Wolff, 2002; Ortiz, 2008a, 2010) (Table 3). This may be explained by the fact that Fildes Bay is negatively affected by the Antarctic’s austral winters, which are characterised by low temperatures and freezing, leading to a reduction in the herbivore biomass and thereby constraining the flow of energy/matter towards the upper trophic levels. Additionally, as shown in Table 3, the different estuaries and coral reefs

systems appeared to be more developed (A/C and Ov/C) but less resistant to perturbations compared to Fildes Bay and the benthic ecosystems studied along the Chilean coast. This latter comparison should be taken with a degree of caution because the trophic model constructed for Fildes Bay represents only a narrow temporal window, and unknown system characteristics may emerge during the rest of the year. The difference between the A/C and Ai/Ci ratios in the Fildes Bay model system may primarily be a consequence of the omission of the flows to birds and marine mammals from our analysis. The analysis of relative Ascendency by component revealed that those groups that principally contributed to the overall structure and function of the Fildes Bay system (i.e.,

Please cite this article in press as: M. Ortiz, et al., Macroscopic network properties and short-term dynamic simulations in coastal ecological systems at Fildes Bay (King George Island, Antarctica), Ecol. Complex. (2016), http://dx.doi.org/10.1016/j.ecocom.2016.06.003

G Model ECOCOM 587 No. of Pages 13

8

M. Ortiz et al. / Ecological Complexity xxx (2016) xxx–xxx

Increase of mortality Mixed flow control (v = 0.3) 10%

30%

50%

Gigartina skottsbergii

Gigartina skottsbergii

Gigartina skottsbergii

0.008 0.000 -0.008 0.008

Phaeophyta

Phaeophyta

Phaeophyta

Chlorophyta

Chlorophyta

Chlorophyta

Rhodophyta

Rhodophyta

Rhodophyta

Zooplankton

Zooplankton

Zooplankton

0.000

Final/Inital biomass

-0.008 0.008 0.000 -0.008 0.008 0.000 -0.008 25 0 -25 25

Phytoplankton

Phytoplankton

0 -25 1

5

9

13

17

1

5

9

13

17

1

5

9

13

17

Functional groups

Increase of mortality Increased vulnerability (v = 1.0) 10% 0.008

30%

50%

Himmontothalus grandifolius Himmontothalus grandifolius Himmontothalus grandifolius

0.000 -0.008 0.008

Desmarestia anceps

Desmarestia anceps

Desmarestia anceps

Gigartina skottsbergii

Gigartina skottsbergii

Gigartina skottsbergii

Phaeophyta

Phaeophyta

Phaeophyta

Chlorophyta

Chlorophyta

Chlorophyta

Rhodophyta

Rhodophyta

Rhodophyta

0.000 -0.008 0.008 Final/Inital biomass

0.000 -0.008 0.008 0.000 -0.008 0.008 0.000 -0.008 0.008 0.000 -0.008 1

5

9

13

17

1

5

9

13

17 1

5

9

13

17

Functional groups Fig. 3. (Continued)

Please cite this article in press as: M. Ortiz, et al., Macroscopic network properties and short-term dynamic simulations in coastal ecological systems at Fildes Bay (King George Island, Antarctica), Ecol. Complex. (2016), http://dx.doi.org/10.1016/j.ecocom.2016.06.003

G Model ECOCOM 587 No. of Pages 13

M. Ortiz et al. / Ecological Complexity xxx (2016) xxx–xxx

9

Table 4 Summary of the System Recovery Time (SRT) for the coastal ecological system of Fildes Bay (Antarctica), using mixed and top-down flow control mechanisms (v). System Recovery Time (years) Vulneability Mixed flow control (v = 0.5)

Top-down flow control (v = 1.0) Increase in total mortality (Production) 10%

30%

50%

10%

30%

50%

(1) Benthic Fishes (2) Diplasterias brucei (3) Seastar (4) Small Epifauna (5) Sterechinus neumayeri (6) Margarella sp. (7) Nacella concinna (8) Filter feeders (9) Himantothalus grandifolius (10) Desamarestia anceps (11) Gigartina skottsbergii (12) Phaeophyta (13) Chlorophyta (14) Rhodophyta (15) Zooplankton (16) Phytoplankton (17) Detritus

11.30 7.25 7.75 14.00 7.50 9.25 9.50 7.50 5.00 5.00 5.25 5.30 5.00 5.30 7.00 8.50 –

14.50 9.80 7.25 16.00 8.75 11.80 13.50 11.80 5.30 5.25 5.30 5.25 5.00 5.25 8.00 8.75 –

14.50 9.50 8.30 16.25 8.50 11.80 13.75 12.00 5.25 5.25 5.25 5.30 5.00 5.30 9.00 8.75 –

(1) Benthic Fishes (2) Diplasterias brucei (3) Seastar (4) Small Epifauna (5) Sterechinus neumayeri (6) Margarella sp. (7) Nacella concinna (8) Filter feeders (9) Himantothalus grandifolius (10) Desamarestia anceps (11) Gigartina skottsbergii (12) Phaeophyta (13) Chlorophyta (14) Rhodophyta (15) Zooplankton (16) Phytoplankton (17) Detritus

– – – – – – – – 5.00 5.00 5.25 5.30 – 5.30 – – –

– – – – – – – – 5.30 5.25 5.30 5.25 – 5.25 – – –

– – – – – – – – 5.25 5.25 5.25 5.30 – 5.30 – – –

Average: Global average:

7.52

8.83

8.97 8.44

Average: Global average:

5.15

5.25

5.25 5.22

detritus, the phyto-zooplankton complex and macroalgae) differed from those that contributed to the kelp forest (Mejillones Peninsula) (Ortiz, 2008a, 2010), but were similar to those within the benthic systems of Tongoy Bay and the La Rinconada Marine Reserve (Antofagasta Bay) (Ortiz and Wolff, 2002; Ortiz et al., 2010). This outcome indicates that although a significant amount of the Antarctic system’s biomass is concentrated in macroalgae, these macroalgae would contribute fewer nutrients to the coastal marine ecosystem than those within kelp forests (Duggins et al., 1989; Ortiz, 2008a, 2010). Interestingly, we were able to use the Ascendency value as a macroscopic-systemic resilience measure, and this value indicated that the modelled system would require less time to return to the initial condition (before disturbance) than the SRT values did (see Table 4 and Fig. 2b). Although it may be explained by the network’s topology and flow structure of the system, additional studies are needed to support robust conclusions. Based on Finn’s cycling index values, which correspond to the amount of flow in a system that is recycled with respect to Total System Throughput (Finn, 1976), the Fildes Bay benthic/pelagic coastal system presented a similar degree of system maturity when compared to the kelp forest of the Mejillones Peninsula and the benthic systems of Tongoy Bay and the La Rinconada Marine Reserve (Antofagasta Bay) (sensu Odum, 1969); however, this result did not hold for the other macroscopic comparisons. The Fildes Bay model system produced slightly higher omnivory index (OI) magnitudes and Finn’s mean path length (FPL) values than the kelp forest of the Antofagasta Peninsula and the benthic systems of Tongoy and Antofagasta Bay (Ortiz and Wolff, 2002; Ortiz et al., 2015) demonstrating that the Fildes Bay benthic/pelagic coastal ecological systems were characterised by a more articulated topological structure. One of the most unexpected results of the dynamic simulations was that although the macroalgae dominated the system’s biomass, this group had the least quantitative impact on the biomass of the remaining groups compared to other components such as filter feeders, small epifauna and benthic fishes. Despite this, the results also suggest that an eventual negative impact of

global changes on the system’s macroalgae would impact on other components of the system because these algae would propagate changes in biomass to nearly all of the other components. Additionally, the other seastar species, Chlorophyta and the red algae G. skottsbergii should also be monitored because these groups play an important role in promoting the complexity of flows (in term of AMI) within the Fildes Bay ecological system. The lowest System Recovery Time values calculated for macroalgae are consistent with that group’s weak quantitative effects on the system’s remaining components. However, the system would be less resilient when any disturbance affected the small epifauna and N. concinna. 5. Conclusions Macroscopic system properties such as A/C and Ov/C ratios and redundancy, which were estimated in the present study, would indicate that the Fildes Bay coastal ecological system would be less developed and organised but more resistant to perturbations compared to other benthic systems along the Chilean coast. This outcome may be a consequence of the Antarctic conditions acting on Fildes Bay, such as austral winters (freezing) and reduced energy/matter flows from macroalgae up to the top trophic levels. These results are relevant because they would demonstrate that the benthic/pelagic system of Fildes Bay has a different trophic network in terms of structure, size, and complexity. Although macroalgae propagated weak quantitative effects on the other components of the system, this group both positively and negatively impacted almost all other system components. This study suggests that the system’s macroalgae, small epifauna, other seastar species, G. skottsbergii, Chlorophyta and N. concinna would be the most sensitive model system components and they are likely candidates for additional research to quantify the effects of global change on Fildes Bay and the Antarctic Peninsula. The protection of the Antarctic environment, as a whole, should not only be focused on ensuring the health of biological populations and communities but should also consider the following factors: (1) changes in the macroscopic or emergent properties of coastal

Please cite this article in press as: M. Ortiz, et al., Macroscopic network properties and short-term dynamic simulations in coastal ecological systems at Fildes Bay (King George Island, Antarctica), Ecol. Complex. (2016), http://dx.doi.org/10.1016/j.ecocom.2016.06.003

G Model ECOCOM 587 No. of Pages 13

10

M. Ortiz et al. / Ecological Complexity xxx (2016) xxx–xxx

benthic/pelagic systems (such as Ascendency, A/C, Ov/C, AMI, etc.), and (2) the propagation of any natural or anthropogenic interventions through the network and their direct and indirect influences on the other components of the system. The outcomes and conclusions presented in the current paper should be considered preliminary because the constructed model only represents austral summer conditions. An annual research programme is required to expand and refine our results. Acknowledgements This research was financed by the grant “Proyecto Anillo ART1101” (Comisión Nacional de Investigación Científica y Tecnológica, Chile, CONICYT-PIA) (Dir. Dr. Iván Gómez). We thank the Instituto Antártico Chileno (INACH) for logistic support during sampling period. We also thank the scientific diving team of Ignacio Garrido, María José Díaz, Jorge Holtheuer and Juan Bravo for collecting the biological samples. Marco Ortiz thanks Dra. Isabel Valdivia (Grant: RT-2214, INACH) for share the samples of benthic fishes. Appendix A. Ecopath and Ecosim theoretical framework. The basic equation of Ecopath can be represented as follows:

 n X P Q  EEi  Bj   DC ji  Y i  BAi  Ei ¼ 0 ð1Þ Bi  B i B j j¼1 2

where Bi and Bj are the biomasses (g wet weight [ww] m ) of preyi and predatorj, respectively; P/Bi (year1)is productivity (production/biomass ratio), which is equivalent to total mortality (Z) (Allen, 1971); EEi is ecotrophic efficiency (dimensionless) or 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 = fishing mortality x biomass) (g ww m2 year1); Q/Bj (year1) represents food consumption per unit biomass of j; DCji is the fraction of preyi in the average diet of predatorj; BAi is the biomass (g w w m2) accumulation rate for i; and Ei (year1) corresponds to the net migration (emigration minus immigration) of i (Christensen and Pauly, 1992). Under this theoretical framework, the energy input and output of all living groups, by definition, must be balanced. The energy balance is ensured within each variable or compartment group using the equation of Christensen et al. (2004): Q ¼ P þ R þ UAF

ð2Þ 2

1

where Q is consumption (g ww m year ), P is production (g ww m2 year1), R is respiration (g ww m2 year1), and UAF is

the unassimilated food (g ww m2 year1) of each variable or compartment in the system. Because of the biomass accumulation and migration factors in Eq. (1), Ecopath models offer an energy continuity rather than a strictly steady-state condition. This particular situation allows changes in the variables or compartments when the mathematical function is expressed in dynamic form. The Ecosim routine uses coupled linear differential equations representing production for each group with the following equation (Walters et al., 1997; Christensen and Walters, 2004): n n X X dBi ¼ gi  Q ji  Q ij þ Ii  ðMi þ F i þ ei Þ  Bi dt j¼1 j¼1

ð3Þ

where dBi/dt represents the growth rate during the time interval dt of group i in terms of its biomass (Bi), g is the net growth efficiency (production/consumption ratio) of pool i, Qji is the consumption (g ww m2 year1) of group j by biomass pool i, Qij is the consumption (g ww m2 year1) by biomass i by group j, I is the immigration (g ww m2 year1) of i, M is the simultaneous natural mortality (year1) of i, F is the instantaneous fishing mortality (year1) of pool i, and e is the emigration rate (year1) of compartment i. The link between predator and prey is key element in Ecosim simulations and is expressed in consumption or flow rates among linked biomass pools. The consumption by compartment (Qij) is represented by Eq. (4): aij  vij  Bi  Bj  Q ij ¼  2  vij þ aij  Bj

ð4Þ

where aij represents the instantaneous mortality rate (year1) 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 model by Eq. (5). The vij represents the vulnerability of a prey to be consumed by a predator or rate of transfer between compartments i and j. This parameter determines if the flow control mechanism is top-down (>2.0), bottom-up (1.0), or mixed (2.0). aij ¼ 

Qi  Bi  Bj

ð5Þ

Where Qi is total consumption of i. Details concerning the Ecopath with Ecosim software package (v. 5.0) are given in Christensen and Walters (2004). Appendix B.

Source of information by species and/or functional groups for the coastal ecological system of Fildes Bay (Antarctica). (B = biomass, P/ B = turnover rate, Q/B = consumption rate, EE = ecotrophic efficiency, and Diet). Compartments Species/functional groups

B1

P/B2

Q/B3

(1) Benthic Fishes (BF)

32.67

2.40

8.50

Diet4 Literature source 1,4

Field estimations for current work, 2,4Cornejo-Donoso and Antezana (2008), Pinkerton Palomares and Pauly (1998) 1,2 Field estimations for current work, 3,4Cornejo-Donoso and Antezana (2008), Pinkerton 1,2 Field estimations for current work, 3,4Cornejo-Donoso and Antezana (2008), Pinkerton 1,2 Field estimations for current work, 3,4Cornejo-Donoso and Antezana (2008), Pinkerton 4 Kaehler et al. (2000), Gili et al. (2001), Corbisier et al. (2004), Norkko et al. (2007), Mincks et al. (2008). 1,2 Field estimations for current work, 3,4Cornejo-Donoso and Antezana (2008), Pinkerton 4 Corbisier et al. (2004), Norkko et al. (2007), Mincks et al. (2008). 1,2 Field estimations for current work, 3,4Cornejo-Donoso and Antezana (2008), Pinkerton 4 Kaehler et al. (2000), Corbisier et al. (2004), Mincks et al. (2008).

et al. (2010),

3

(2) Diplasterias brucei (Db) (3) Seastars (SS) (4) Small Epifauna (SE)

37.83 18.52 66.50

1.40 1.40 4.60

5.00 5.00 15.50

(5) Sterechinus neumayeri (Sn)

34.50

3.70

13.50

(6) Margarella sp. (Msp)

59.50

3.40

11.50

et al. (2010) et al. (2010) et al. (2010),

et al. (2010),

et al. (2010),

Please cite this article in press as: M. Ortiz, et al., Macroscopic network properties and short-term dynamic simulations in coastal ecological systems at Fildes Bay (King George Island, Antarctica), Ecol. Complex. (2016), http://dx.doi.org/10.1016/j.ecocom.2016.06.003

G Model ECOCOM 587 No. of Pages 13

M. Ortiz et al. / Ecological Complexity xxx (2016) xxx–xxx

11

1,2

Field estimations for current work, 3,4Cornejo-Donoso and Antezana (2008), Pinkerton et al. (2010), 4 Kaehler et al. (2000), Corbisier et al. (2004), Mincks et al. (2008). 1,2 Field estimations for current work, 3,4Cornejo-Donoso and Antezana (2008), Pinkerton et al. (2010), 4 Kaehler et al. (2000) , Gili et al. (2001),Corbisier et al. (2004), Mincks et al. (2008). 1 Field estimations for current work, 2Estimated by Ecopath

(7) Nacella concinna (Nc)

56.50

3.60

12.50

(8) Filter Feeders (FF)

177.46

1.87

8.50

(9) Himmantothalus grandifolius (Hg) (10) Desmarestia anceps (Da) (11) Gigartina skottsbergii (Gs) (12) Phaeophyta (Phaeo) (13) Chlorophyta (Chloro) (14) Rhodophyta (Rhodo) (15) Zooplankton (Zoo) (16) Phytoplankton (Phy) (17) Detritus (Det)

597.67

2.10

–

310.50 35.17

2.30 2.50

– –

1

650.50 38.50 174.57 122.50 300.00 100.00

2.80 2.40 5.00 4.60 18.00 –

– – – 15.50 – –

1

1

Field estimations for current work, 2Estimated by Ecopath Field estimations for current work, 2Estimated by Ecopath

Field estimations for current work, 2Estimated by Ecopath Field estimations for current work, 2Estimated by Ecopath 1 Field estimations for current work, 2Estimated by Ecopath 1 Cornejo-Donoso and Antezana (2008), Pinkerton et al. (2010) 1 Cornejo-Donoso and Antezana (2008), Pinkerton et al. (2010) 1 Cornejo-Donoso and Antezana (2008), Pinkerton et al. (2010) 1

Note: B=biomass (g wet weight m2), P/B=turnover rates and Q/B=consumption rate, Diet.

Appendix C.

Prey-predator (and plant-grazer) interaction matrix for the coastal ecological system of Fildes Bay (Antarctic).The information is given in proportion. Prey\Predator (1) BF (2) Db (3) SS (4) SE (5) Sn (6) Msp (7) Nc (8) FF (9) Hg (10) Da (11) Gs (12) Phaeo (13) Chloro (14) Rhodo (15) Zoo (16) Phy (17) Detritus

1

2

3

0.15 0.15

0.30 0.10 0.15 0.15

0.10 0.01 0.20 0.05 0.30 0.30

0.05 0.05 0.01 0.05 0.01 0.03

0.05 0.05 0.01 0.05 0.01 0.03

0.40

0.10

0.10

0.04

4

5

0.01 0.10 0.10 0.10 0.30 0.06 0.05 0.01 0.10 0.01 0.10

0.20

0.06

0.15 0.05 0.01 0.20 0.01 0.15 0.20 0.03

6

7

8

0.20 0.10 0.015 0.30 0.01 0.06 0.10 0.20 0.015

0.20 0.15 0.01 0.40 0.01 0.07 0.05 0.10 0.01

0.30 0.60 0.10

References Allen, K.R., 1971. Relation between production and biomass. J. Fish. Res. Board Can. 28, 1573–1581. Angelini, R., Petrere Jr., M., 2000. A model for the plankton system of the Broa reservoir, Sao Carlos. Braz. Ecol. Model. 126, 131–137. Arias-González, J., et al., 2011. Predicted impact of the invasive lionfish Pterois volitans on the food web of a Caribbean coral reef. Environ. Res. 111 (7), 917–925. Arias-González, J., et al., 2004. Trophic models for investigation of fishing effect on coral ref ecosystems. Ecol. Model. 172, 197–212. Baird, D., Ulanowicz, R., 1993. Comparative study on the trophic structure, cycling and ecosystem properties of four tidal estuaries. Mar. Ecol. Progr. Ser. 99, 221– 237. Baird, D., McGlade, J., Ulanowicz, R., 1991. The comparative ecology of six marine ecosystems. Philos. Trans. R. Soc. Lond. 333, 15–29. Bayle-Sempere, J., et al., 2013. Trophic structure and energy fluxes around a Mediterranean fish farm. Ecol. Model. 248, 135–147. Bick, A., Arlt, G., 2013. Description of intertidal macro-and meiobenthic assemblages in maxwell bay, king george island, south shetland islands, southern ocean. Polar Biol. 36, 673–689. Calizza, E., et al., 2015. Effect of multiple disturbances on food web vulnerability to biodiversity loss in detritus-based systems. Ecosphere 6 (7), 1–20. Christensen, V., Pauly, D., 1992. Ecopath II: a software for balancing steady-state ecosystem models and calculating network characteristics. Ecol. Model. 61, 169–185. Christensen, V., Walters, C., 2004. Ecopath with Ecosim: methods, capabilities and limitations. Ecol. Model. 172, 109–139.

9

10

11

12

13

14

15

0.90 0.10

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, pp. 154 available online at http://www.ecopath.org and http:// www.fisheries.ubc.ca. Christensen, V., 1995. Ecosystem maturity-towards quantification. Ecol. Model. 77, 3–32. Clarke, A., et al., 2004. Evolution and diversity of the benthic fauna of the Southern Ocean continental shelf. Antarct. Sci. 16, 559–568. Corbisier, T., et al., 2004. Trophic relationships in the nearshore zone of Martel Inlet (King George Island, Antarctica): d13C stable-isotope analysis. Polar Biol. 27, 75– 82. Cornejo-Donoso, J., Antezana, T., 2008. Preliminary trophic model of the Antarctic Peninsula ecosystem (sub-area CCAMLR 48.1). Ecol. Model. 218, 1–17. Costa, D., Crocker, D., 1996. Marine mammals of the southern ocean. In: Ross, R., Hofmann, E., Quentin, L. (Eds.), Foundations for Ecological Research West of the Antarctic Peninsula. Antarctic Research Series, vol. 70. American Geophysical Union, Washington DC, pp. 287–301. Costanza, R., Mageau, M., 1999. What is a healthy ecosystem? Aquat. Ecol. 33, 105– 115. Cropp, R., Gabric, A., 2002. Ecosystem adaptation: do ecosystems maximize resilience? Ecology 83, 2019–2026. Díaz-Uribe, J.G., et al., 2012. An integrated ecosystem trophic model for the North and Central Gulf of California: an alternative view for endemic species conservation. Ecol. Model. 230, 73–91. Dayton, P., et al., 1974. Biological accommodation in the benthic community at McMurdo Sound, Antarctic. Ecol. Monogr. 44, 105–128. Duggins, D.O., Simenstad, C.A., Estes, J.A., 1989. Magnification of secondary production by kelp detritus in coastal marine ecosystems. Science 245, 101–232.

Please cite this article in press as: M. Ortiz, et al., Macroscopic network properties and short-term dynamic simulations in coastal ecological systems at Fildes Bay (King George Island, Antarctica), Ecol. Complex. (2016), http://dx.doi.org/10.1016/j.ecocom.2016.06.003

G Model ECOCOM 587 No. of Pages 13

12

M. Ortiz et al. / Ecological Complexity xxx (2016) xxx–xxx

Eklöf, A., et al., 2013. Secondary extinctions in food webs: a Bayesian network approach. Methods Ecol. Evol. 4, 760–770. Epstein, S., 1997. Microbial food web in marine sediments: I. Trophic interactions and grazing rates in two tidal flat communities. Microb. Ecol. 34, 188–198. Fath, B., Patten, B., Choi, J., 2001. Complementarity of ecological goal functions. J. Theor. Biol. 208, 493–506. Findlay, R., White, D., 1983. The effects of feeding by the sand dollar Mellita quinquiesperforata (Leske) on the benthic microbial community. J. Exp. Mar. Biol. Ecol. 72, 25–41. Finn, J.T., 1976. Measures of ecosystem structure and function derived from analysis of flows. J. Theor. Biol. 56 (2), 363–380. Gómez, I., et al., 2009. Light and temperature demands of marine benthic microalgae and seaweeds in the polar regions. Bot. Mar. 52, 593–608. Gili, J.-M., et al., 2001. Are Antarctic suspensión-feeding communities different from those elsewhere in the world? Polar Biol. 24, 473–485. Grossmann, S., Reichardt, W., 1991. Impact of Arenicola marina on bacteria in intertidal sediments. Mar. Ecol. Prog. Ser. 77, 85–93. Guénette, S., Christensen, V., Pauly, D., 2008. Trophic modelling of the Peruvian upwelling ecosystem: towards reconciliation of multiples datasets. Prog. Oceanogr. 79, 326–335. Hawkins, S., 2004. Scaling up. the role of species and habitat patches in functioning of coastal ecosystems. Aquat. Conserv. Mar. Freshw. Ecosyst. 14, 217–219. Heymans, J., Baird, D., 2000. A carbon flow model and network analysis of the northern Benguela upwelling system, Namibia. Ecol. Model. 126, 9–32. Hunter, M.D., Price, P.W., 1992. Playing chutes and ladders: bottom-up and topdown forces in natural communities. Ecology 73, 724–732. Huovinen, P., Gómez, I., 2013. Photosysthetic characteristics and UV stress tolerance of Antarctic seaweeds along the depth gradient. Polar Biol. 36, 1319–1332. Iken, K., et al., 1997. Grazing by the Antarctic fish Notothenia coriiceps: evidence for selective feeding on macroalgae. Antarct. Sci. 9 (4), 386–391. Kaehler, S., Pakhomov, E., McQuaid, C., 2000. Trophic structure of the marine food web at Prince Edward Islands (Southern Ocean) determined by d13C and d15N analysis. Mar. Ecol. Prog. Ser. 208, 13–20. Kaufman, A., Borrett, S., 2010. Ecosystem network analysis indicators are generally robust to parameter uncertainty in a phosphorus model of Lake Sidney Lanier, USA. Ecol. Model. 221, 1230–1238. King, J., 1994. Recentr climate variability in the vicinity of the Antarctic Peninsula. Int. J. Climatol. 14, 357–369. Klöser, H., et al., 1993. Seasonal variation of algal growth conditions in sheltered Antarctic bays: the example of Potter Cove (King George Island, South Shetlands). J. Mar. Syst. 4, 289–301. Krebs, C.J., et al., 1995. Impact of food and predation on the snowshore hare cycle. Science 269, 1112–1115. Lawton, J., 1994. What do species do in ecosystems? Oikos 71, 367–374. Levins, R., 1966. The strategy of model building in population biology. Am. Sci. 54 (4), 421–431. Levins, R., 1998a. Qualitative mathematics for understanding, prediction, and intervention in complex ecosystems. In: Raport, D., Costanza, R., Epstein, P., Gaudet, C., Levins, R. (Eds.), Ecosystem Health. Blackwell Science, MA, pp. 178– 204. Levins, R., 1998b. The internal and external explanatory theories. Sci. Cult. 7 (4), 557–582. Li, Y., Yang, Z., 2011. Quantifying the sustainability of water use systems: calculating the balance between network efficiency and resilience. Ecol. Model. 222, 1771– 1780. Masterson, P., et al., 2008. Interaction of top-down and bottom-up factors in intertidal rockpools: effects on early successional macroalgal community composition, abundance and productivity. J. Exp. Mar. Biol. Ecol. 363 (1–2), 12– 20. McClintock, J., 1994. Trophic biology of Antarctic shallow-water echinoderms. Mar. Ecol. Prog. Ser. 111, 191–202. Mincks, S., et al., 2008. Trophic structure on the West Antarctic Peninsula shelf: detritivory and benthic inertia revealed by d13C and d15N analysis. Deep-Sea Res. II 55, 2502–2514. Monaco, M., Ulanowicz, R., 1997. Comparative ecosystem trophic structure of three U.S. mid-Atlantic estuarios. Mar. Ecol. Prog. Ser. 161, 239–254. Muhly, T., Hebblewhite, M., Paton, D., Pitt, J., Boyce, M., Musiani, M., 2013. Humans strengthen bottom-up effects and weaken trophic cascades in a terrestrial food web. PLoS One doi:http://dx.doi.org/10.1371/journal.pone.0064311. Norkko, A., et al., 2007. Trophic structure of coastal Antarctic food webs associated with changes in sea ice and food supply. Ecology 88 (11), 2810–2820. Odum, E., 1969. The strategy of ecosystem development. Science 104, 262–270. Ortiz, M., Levins, R., 2011. Re-stocking practices and illegal fishing in northern Chile (SE Pacific coast): a study case. Oikos 120, 1402–1412. Ortiz, M., Wolff, M., 2002. Trophic models of four benthic communities in Tongot Bay (Chile): comparative análisis and preliminary assessment of management strategies. J. Exp. Mar. Biol. Ecol. 268, 205–235. Ortiz, M., et al., 2010. Trophic mass balanced models and dynamic simulations of benthic communities from La Rinconada Marine Reserve off northern Chile: network properties and multispecies harvest scenario assessment. Aquat. Conserv. Mar. Freshw. Ecosyst. 20 (1), 58–73. Ortiz, M., et al., 2015. Mass balanced trophic models and short-term dynamical simulations for benthic ecological systems of Mejillones and Antofagasta bays (SE Pacific): comparative network structure and assessment of human impacts. Ecol. Model. 309–310, 153–162.

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. Ecol. Model. 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. Aquat. Conserv. Mar. Freshw. Ecosyst. 18, 923–929. Ortiz, M., 2010. Dynamical and spatial models of kelp forest of Macrocystis integrifolia and Lessonia trabeculata (SE Pacific) for assessment harvest scenarios: short-term responses. Aquat. Conserv. Mar. Freshw. Ecosyst. 20, 494– 506. Palomares, M., Pauly, D., 1998. Predicting food consumption of fish population as function of mortality, food type, morphometrics, temperature and salinity. Mar. Freshw. Res. 49, 447–453. Panikkar, P., Khan, M., 2008. Comparative mass-balanced trophic models to assess the impact of environmental management measures in a tropical reservoir ecosystem. Ecol. Model. 212, 280–291. Patrício, J., Marques, J.C., 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). Ecol. Model. 197, 21–34. Perez-Espana, H., Arreguin-Sanchez, F., 2001. An inverse relationship between stability and maturity in models of aquatic ecosystems. Ecol. Model. 145, 189– 196. Pessoa, M., 2012. Algae and aquatic macrophytes responses to cope to ultraviolet radiation—a review. Emirates J. Food Agric. 24 (6), 527–545. Pickitch, E., et al., 2004. Ecosystem-based fishery management. Science 305, 346– 347. Pinkerton, M., Bradford-Grieve, J., 2014. Characterizing foodweb structure to identify potential ecosystem effects of fishing in the Ross Sea, Antarctica. ICES J. Mar. Sci. 71 (7), 1542–1553. Pinkerton, M., Bradford-Grieve, J., Hanchet, S., 2010. A balanced model of the food web of the Ross Sea, Antarctic. CCAMLR Sci. 17, 1–31. Plante, C., Mayer, L., 1994. Distribution and efficiency of bacteriolysis in the gut of Arenicola marina and three additional deposit feeders. Mar. Ecol. Prog. Ser. 109, 183–194. Plante, C., Shriver, A., 1998. Patterns of differential digestion of bacteria in deposit feeders: a test of resource partitioning. Mar. Ecol. Prog. Ser. 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. Preikshot, D., Beamish, R., Neville, C., 2013. A dynamic model describing ecosystemlevel changes in the Strait of Giorgia from 1960 to 2010. Prog. Oceanogr. 115, 28– 40. Rautenberger, R., Huovinen, P., Gómez, I., 2015. Effects of increased seawater temperature on UV tolerance of Antarctic marine macroalgae. Mar. Biol. 162, 1087–1097. Richardson, M., 1979. The distribution of Antarctic marine macroalgae related to depth and substrate. Br. Antarct. Surv. Bull. 49, 1–13. Richter, A., Wuttke, S., Zacher, K., 2008. Two years of in situ UV measurements at Dallmann Laboratory/Jubany Station. In: Wiencke, C., Ferreyra, G., Abele, D., Marenssi, S. (Eds.), The Antarctic Ecosystem of Potter Cove, King George Island. Synopsis of research performed 1999–2006 at Dalmann Laboratory and Jubanu Station, vol. 571. Report of Polar Marine Research, pp. 12–19. Ricker, W.E., 1968. Food from the sea. In: Committee on Resources and Man (Ed.), Resource and Man, US National Academy of Sciences. W.H. Freeman, San Francisco, CA, pp. 87–108 (Chapter 5). Rossi, L., et al., 2015. Predator and detritivore niche width helps to explain biocomplexity of experimental detritus-based food webs in four aquatic and terrestrial ecosystems. Ecol. Complex. 23, 14–24. Smith, R., Stammerjohn, S., 2001. Variations of surface air temperature and sea ice extent in the Western Antarctic Peninsula (WAP) region. Ann. Glaciol. 33, 493– 500. Smith, W., Ainley, D., Cattaneo-Vietti, R., 2007. Trophic interactions within the Ross Sea continental shelf ecosystem. Philos. Trans. R. Soc. B 362, 95–111. Stark, P., 1994. Climate warning in the central Antarctic Peninsula area. Weather 49, 215–220. Targett, T., 1981. Trophic ecology and structure of coastal Antarctic fish communities. Mar. Ecol. Prog. Ser. 4, 243–263. Thompson, R.C., Norton, T.A., Hawkins, S.J., 2004. Physical stress and biological control regulate the producer-consumer balance in intertidal biofilms. Ecology 85, 1372–1382. Ulanowicz, R., Abarca-Arenas, L., 1997. An informational synthesis of ecosystem structure and function. Ecol. Model. 95, 1–10. 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. Valdivia, N., et al., 2014. Up, down, and all around: scale-dependent spatial variation in rocky-shore communities of Fildes Peninsula, King George Island, Antarctica. PLoS One 9 (6), e100714. doi:http://dx.doi.org/10.1371/journal. pone.0100714. Vaughan, D., Doake, C., 1996. Recent atmospheric warming and retreat of ice shelves on the Antarctic Peninsula. Nature 379, 328–331.

Please cite this article in press as: M. Ortiz, et al., Macroscopic network properties and short-term dynamic simulations in coastal ecological systems at Fildes Bay (King George Island, Antarctica), Ecol. Complex. (2016), http://dx.doi.org/10.1016/j.ecocom.2016.06.003

G Model ECOCOM 587 No. of Pages 13

M. Ortiz et al. / Ecological Complexity xxx (2016) xxx–xxx Walters, C., Christensen, V., Pauly, D., 1997. Structuring dynamic models of exploited ecosystems from trophic mass-balance assessments. Rev. Fish Biol. Fish. 7, 139– 172. Warwick, R., Clarke, K., 1993. Comparing the severity of disturbance: a metaanalysis of marine macrobenthic community data. Mar. Ecol. Prog. Ser. 92, 221– 231.

13

Wolff, M., Koch, V., Isaac, V., 2000. A trophic flow model of the Caeté mangrove estuary (North Brazil) with considerations for the sustainable use of its resources. Estuar. Coast. Shelf Sci. 50, 789–803. Wolff, M., 1994. A trophic model for Tongoy Bay—a system exposed to suspended scallop culture (Northern Chile). J. Exp. Mar. Biol. Ecol. 182, 149–168.

Please cite this article in press as: M. Ortiz, et al., Macroscopic network properties and short-term dynamic simulations in coastal ecological systems at Fildes Bay (King George Island, Antarctica), Ecol. Complex. (2016), http://dx.doi.org/10.1016/j.ecocom.2016.06.003