Work project, 15 hp Theroretical Ecology Autumn Semester 2019
The Importance of Trait Mediated Indirect
Interactions in Marine Ecosystems.
3
The Importance of Trait Mediated Indirect
Interactions in Marine Ecosystems
Abstract
The importance of Trait Mediated Indirect Interactions (TMII) is increasingly being recognized. TMII are interactions between two species via a change in trait (behavioural, morphological, but not numerical) over a third one, which together cause ecological dynamics. Marine food webs have complex interactions, but TMII have not yet received great appreciation or application in marine conservation and management models. This article is a review about the different ways in which TMII can affect marine ecological dynamics. I summarize known examples of Behaviourally Mediated Indirect Interactions, Physiologically Mediated Indirect Interactions, and other types of indirect interactions such as initiated or mediated by parasites, in order to provide a better understanding about their functioning. I found that TMII are omnipresent in marine ecosystems and occur at all trophic levels, spanning from macro- to microorganisms. Furthermore, it includes many different taxa and guilds, and the mechanisms are highly diverse. Some of them enhance Density Mediated Effects, while others counteract them. Sometimes this results in the effects opposite of those expected, and often they extend further in the food web. Understanding of TMII is likely to be beneficial for marine conservation and management, due to the role of humans causing them or suffering its effects.
4
Table of Contents
Introduction……….5
Material and Methods……….5
Results……….6
Behaviourally Mediated Indirect Interactions………6
Top predator – mesopredator – prey interactions………6
Predator – prey – competitor interactions………11
Predator – prey – predator interactions………12
Predator – competitor – prey interactions………..13
Prey – predator – prey interactions……….13
Physiologically Mediated Indirect Interactions……….13
Toxicity and disease………..13
Induced morphological responses………14
Induced chemical responses……….15
Chemical signalling..……….16
Interference………..16
Change in selective pressures………..16
Other interactions……….17
Interactions initiated or mediated by parasites……….17
Microbial interactions……….19
Movement of nutrients and habitat coupling……….20
Discussion………20
Acknowledgement………21
5
Introduction
Understanding the dynamics and processes of an ecosystem has a vital importance to its conservation. It is widely recognised that the presence of a single species may have effects in the whole ecosystem through series of indirect interactions with other species. Despite that systems ecology usually assumes that these interactions work through population densities, the importance of non-consumptive effects is increasingly recognized.
Marine ecosystems represent 71% of the Earth’s surface and contain almost all known phyla. Marine food webs are often more complex than terrestrial ones, and tend to have more species of prey per predator, and more species of predator per prey. Moreover, many marine species such as invertebrates and fish have ontogenetic shifts in ecological niche and trophic level, so the potential interactions between species and groups should be numerous, complex, and diverse (Dill et al, 2003). They are also endangered ecosystems, due to climate change, pollution and overfishing. Trait Mediated Indirect Interactions (TMII) are known to amplify as well as counteract trophic interactions, so their importance to understanding, conservation and management of marine ecosystems should be large. However, they are not having a great appreciation or applicability in marine conservation and management models (Dill, 2017). There have been several definitions of TMII in the literature, sometimes differing more than slightly, which can lead to confusion (Abrahams, 2007). So, to provide a clear definition, in this report TMII are defined as “any effect caused by a change in a property (density, size, behaviour or mere presence) of a first species (called “initiator”), that via a change in trait (behavioural, physiological, or other) on a second species (the “transmitter”) affects a third species (called “receiver”) in the community”. Trait Mediated Indirect Interactions differ from “Density Mediated Indirect Interactions” (DMII), which focus on effects of density changes in the transmitter.
In some cases, the transmitter and the receiver belong to the same species. By using the above-mentioned definition strictly, they should not fall under TMII. However, they are anyway included them in the study to show “TMII-like” mechanisms that are potentially important in the sea. Moreover, in this report, each example of TMII is presented in isolation, but in natural ecosystems many interactions co-occur. Humans are considered the initiator species in very few cases, but in most of them, they are the underlying cause, at least indirectly, due to their activities and the influence on the marine ecosystems (global warming, direct fishing, transport of species, etc).
The purpose of this review is to describe the different ways in which Trait Mediated Indirect Interactions affect different marine species and ecosystems. Based on published literature I provide several examples of TMII, summarizing available information to improve the understanding of the ecological dynamics. The aim is to contribute to conservation ecology and highlight gaps in our current knowledge.
Materials and Methods
6
Results
Behaviourally mediated indirect interactions
One of the most studied type of non-consumptive indirect interactions in marine ecosystems and on land, are the behaviourally mediated indirect interactions. The most common behaviour is the adoption of a strategy to avoid predation from a prey transmitter because of the presence of an initiator predator, which can affect other species in many ways. These strategies may be a reduction in activity, change of habitat, or others.
Top predator – mesopredator – prey interactions.
Top predator – mesopredator – prey interactions are easy to grasp and as such they are intuitive examples of behaviourally mediated indirect interactions: they lead to trophic cascades where a top predator (initiator) that feeds on a mesopredator (transmitter), causes density changes in the mesopredator prey (receiver). Transmitter reduction by consumption may cause an increase in receiver density due to the reduced predation in a density mediated indirect interaction. However, prey (transmitters) usually don’t wait passively to be eaten by other species. To avoid it, many of them reduce their activity, having to find a trade-off between feeding and predation risk. That can positively affect their own prey (receiver) densities, in a Behaviourally Mediated Indirect Interaction (Fig. 1), causing an overestimation of the importance of Density Mediated Indirect Interactions and the consumptive effects. Interaction between consumptive and behaviourally mediated effects may even have a larger impact than both effects in isolation. For example, anti-predator behaviour may result in lost body condition due to lower foraging rates, that induces compromised individuals to take
greater risks that may result in falling victim to predators (Kizska et al, 2015)
There are several examples of such interactions in the literature, comprising many different taxa:
Sea urchins have shown an increase in cryptic (hiding in crevices) or escape behaviour in the presence of predators, both in field studies and in experiments simulating predator cues (Spyksma et al, 2017; Kintzing & Butler, 2014). In turn, the reduced activity by the sea urchins have been shown to cause increased biomass of algae. In both in laboratory as in field studies cues from a microcarnivorous fish, Oxyjulis califórnica, have been shown to reduce grazing from the limpet Lottia insessa on kelp (Haggerty et al, 2018). In coral reefs, high densities of top predator Nassau grouper Epinephelus striatus caused a reduction in the activity of two smaller groups of mesopredators, Cephalopholis fulva and C. cruentata, and the recruitment of many fish species in the reef such as parrot and damselfishes increased as a result (Stallings, 2008). Many marine mammals show a change in their diving behaviour when their predators (such as sharks) are present, optimizing energy gain and safety from predators by spending less time in more profitable but dangerous zones or decreasing their use of risky feeding tactics. This is in turn has consequences for their prey. For example, North-East Pacific harbour seals (Phoca vitulina) forage mainly in surface, despite that walleye pollock (Theragra
chalcogramma), which is usually found in deeper waters, are larger and more predictably
7 encountered than prey that can be found in surface, which would make it a more profitable prey for seals. But pollock overlaps spatially with sleeper sharks (Somniosus pacificus) so the risk of foraging for pollock is substantially higher for seals. (Wirsing et al, 2007) It means that the presence of seals has a positive effect over pollock. Of course, marine mammals can also act as predators, being the initiators of BMII by reducing the activity of their prey. For example, magpie morwong Cheilodactylus nigripes reduce their grazing rates on algae when New Zealand fur seals Arctocephalus forsteri are present (Kizska et al, 2016).
Another possible strategy for prey to avoid predation is moving to a more secure habitat where food availability is not so high but the predation risk is lower. It would cause an increase in the receiver prey population density in the habitat where the predator is present, but also a decrease in the receiver prey population density in the habitat where the predator is absent (Fig. 2).
Sea snails Tegula funebralis flee from their intertidal pools in presence of waterborne cues from predatory seastar Leptasterias (Gravem and Morgan, 2016). Many marine mammal species effectuate migrations from long distances to avoid predation. Terrestrial predators such as wolves or bears are also able to induce habitat shifting in marine mammals, especially pinnipeds. In the Arctic Sea pinnipeds seek refuge from danger by entering water due to the risk from polar bears and humans, whereas Antarctic pinnipeds do so by hauling out on ice because the predation risk there comes from killer whales and leopard seals. (Estes et al, 2016). As predators, marine
mammals can cause habitat shifting in other species, such as emperor penguins Aptenodytes
forsteri and Adélie penguins Pygoscelis adeliae, which avoid foraging at night due to predation
risk by leopard seals (Kyzska et al, 2015). Marine reptiles are also important drivers of BMII. In Shark Bay, Australia, green turtles in good body condition choose safer but less profitable microhabitats when predation risk by sharks was high, but turtles in poorer conditions selected more profitable but also more risky microhabitats. In seasons when sharks are less abundant turtles in good condition also moved to more profitable zones. That has implications to the overall ecosystem dynamics, since their feeding and cropping can alter its community’s composition and nutrient cycles. As turtles are long living animals and they actually suffer low predation rates, the effect of BMII is much stronger than DMII (Heithaus et al, 2007).
Safer microhabitats are, however, not always those with lower predator density. If landscape features help to escape from predators, prey may choose that microhabitats despite that encounters may produce more often. In Shark Bay turtles, dugongs and dolphins choose zones close to deep waters in seasons with high densities of tiger sharks. That species can escape more easily in deep waters because of their superior manoeuvrability. Counterintuitively, this causes that the arrival of predators in an area increase the mesoconsumer abundance and the exploitation rates. On the contrary, cormorant’s escape abilities are not affected by underwater landscape (they just fly) so in “shark season”, they move to interior habitats where encounters with sharks are less common, despite that habitats are not as profitable. It shows that BMII initiated by one single species may have different effects according to the interaction between transmitter species and landscape features (Heithaus et al, 2009).
8 These interactions may not only affect individual species but also the whole habitat biodiversity. This is especially true when the receiver species is a strong competitor. If the receiver is affected negatively by TMII it can supress its dominance in an ecosystem and allow less competitive species to be released, increasing the biodiversity. In the other hand, when the receiver is positively affected, the ecosystem diversity is reduced. An experiment conducted by Wada et al (2013), showed a BMII initiated by the snail T. clavigera and transmited by the limpet S. sirius, which reduced grazing on Ulva algae, competitively superior. It caused a shift from a competitively inferior Lithoderma algae dominance to Ulva dominance (Fig. 3).
BMII may also affect a high number of species and ecosystem biodiversity when the affected species is an ecosystem engineer, improving spatial heterogeneity or facilitating the coexistence of many species. If that species is affected positively, the ecosystem diversity will increase, but when is affected negatively the diversity will decrease. An experiment conducted by Premo and Tyler (2013) showed that in the presence of predator clues, gastropod Ilyanassa
obsoleta and bivalve Mercenaria mercenaria reduced their activity. Both species had effects
on sediment processes, and their reduced activity caused reduced oxygenation and nitrogen removal in the case of M. mercenaria, and an increase in algal biomass and a decrease in ammonium release to the sediment in the case of I. obsolete. There is another example in Shark Bay. Dugongs may forage for seagrass in two different ways: they can crop, stripping leaves from the top of seagrass, or dig into the substrate to acquire the entire plant. The second tactic is more profitable for them in terms of energy intake, but it implies that it cannot scan for predators as often as with cropping. Moreover, digging produces sediment plums that may attract predators and hinder vigilance even more. So, in seasons when tiger sharks are more abundant, dugong dig very infrequently, but more often when they are more absent. Excavation facilitates patch succession, while cropping promotes persistence of perennial species, so by altering the way seagrasses are harvested by dugongs, tiger sharks are affecting the composition and structure of seagrass species (Wirsing et al, 2007).
On the other hand, BMII cannot only affect ecosystem biodiversity, but also be affected by it, since ecosystem interactions are not limited to linear chains but go through entire trophic levels. An experiment conducted by Byrnes et al (2005), showed that predator diversity strengthens the effect of trophic cascades through BMII. Predator invertebrate diversity, was positively correlated with kelp abundance and negatively correlated with herbivore invertebrate density. According to the results, it was caused by BMII, but no single predator reduced grazing of all herbivores. Predator diversity may also strengthen the trophic cascade providing redundancy to the loss of one component species or trough complementary effects on prey, which have to show different responses to different predators. In the same way, transmitter diversity weakens the strength of trophic cascades. Another experiment conducted by Duffy et al (2005) showed that BMII initiated by crabs over primary producers were much weaker with high diversity of herbivorous species. Of course, not all the interactions have the same strength. Another experiment conducted by Reiss et al (2014) using fish predators showed that different predators has varying effects on invertebrate prey and algae, but this differential effect weakens in presence of multiple predators.
Effects on algae (or the receiver in general) are, however, not limited to density. Reynolds & Sotka (2011) described how the presence of pinfish predator Langodon rhomboids olfactory cues reduced grazing by the amphipod Ampithoe longimana and had physiological effects on the alga Sargassum filipendula. This alga reduced its palatability in response to perceived
Figure 3: Trophic cascade consisting in snail Thais, limpet Siphonaria, and algae Ulva and
Lithoderma. Limpet’s preferential feeding on Ulva
9 grazing, but A. longimana reduced grazing in presence of L. rhombodis, which caused S.
filipendula tissue to be more palatable. More interestingly, chemical cues released from
damaged algae seem to attract pinfish, creating a cyclic tritrophic interaction that would increase algal fitness, and maybe a negative feedback that could have regulatory effects. If we also consider that this interaction could give S. filipendula a competitive advantage against another algae species, for example competing for light (Coleman et al, 2007), we can start to guess how complex TMII can be and how far they can extend. Other examples where presence of top predators altered physiology of prey were described by Palacios et al (2016). Mesopredator dottyback, Pseudochromis fuscus by continued attacking and foraging for damselfishes Pomacentrus amboinensis caused an increase in O2 uptake from the latter. But in presence of top predator coral trout Plectropomus leopardus, dottybacks reduced their activity, minimizing damselfishes stress and allowing them to maintain routine O2 uptake, in a BMII that benefit physiologically the receivers.
It is important to point out that BMII are not only present in animals with a common evolutionary history. Despite that native prey are sometimes naïve to cues from new predators, invasive animals may also induce anti-predator behaviour on native prey. That is more likely when prey react to more general than specific cues (chemical cues from damaged conspecifics, sight of big objects…). For example, invasive Pacific lionfish Pterois volitans has been shown to induce anti-predator responses in many herbivorous from the Caribbean, such as parrotfishes, reducing algal loss and even having positive effects to conservation of the entire ecosystem (Kindinger et al, 2017). In the opposite way, invasive species may react to native predators. Oyster drills Urosalpinx cinerea and Ocinebrina inornata respond hiding to the presence of native predatory crab Cancer productus by reacting to alarm cues from injured conspecifics. So, the presence of C. productus has a positive effect on native oysters, which recovery is made difficult because of the predation by the drills (Grason & Miner, 2012). Both cases show how BMII have influence in invasion biology.
It is hard to disentangle the effects of DMII and TMII. There are many factors that are needed to consider when conducting experiments or trying to predict the result of TMII:
As explained previously, many species avoid predation by fleeing to spaces where predators are not present. It means that by restricting prey dispersal in experiments, the importance of DMII can be overestimated, as shown the experiment conducted by Geraldie and Macreadie (2013) with toadfish predator Opsanus tau, mud crab mesopredator Panopeus herbstii and ribbed mussel prey Geukensia demissa. In that experiment, when mud crabs were restricted from dispersing, their consumption by toadfishes was 9 times higher, and they were also observed to move along mesocosm edges trying to disperse.
10 It is also important to point out that prey need to
be able to recognize cues from predators (visual, acoustical, mechanical, chemical or even electrical) and integrate them to show a response to their presence. Despite sounding obvious, it has many important implications when experiments are conducted, especially in laboratory experiments. Cue perception is environmentally affected. For example, sight is more difficult in turbid waters or in dark conditions, and chemical cues disperse more easily in high flow waters. That may cause an overestimation of BMII in many laboratory and field experiments, where prey would receive stronger cues and would show stronger responses (Fig 4). Secondly, prey are usually able to integrate different signals, which may have an additive or synergic effect. Many of the experiments only produce one type of cue (usually chemical) which may cause an underestimation of response. Also, integration of signals from multiple predators may cause different effects than only one. Responses may also be affected by other environmental signals, such as the perceived possible refugees, which may variate the intensity of the response (Weissburg et al, 2014).
BMII can have more implications than the apparent ones, as described by Alexander et al
(2013). The experiment replicated an intertidal system consisting in a fish high predator
Lipophrys pholis, an amphipod mesopredator Echinogammarus marinus and an isopod prey Jaera nordmanni. E. marinus was able to detect cues from L. pholis feeding in conspecifics
and reducing its activity, and that effect was transmitted to J. nordmanni altering the strengths of the functional response of E. marinus towards this prey resource in a BMII. But interestingly, the functional response in simple habitats was type 2, whereas in complex ones was type 3. That has a great importance in the stability of the system. A type 2 functional response is potentially destabilizing due to high risk of mortality to prey at low densities, which may drive prey to local extinction. But a type 3 functional response is stabilizing, due to the low mortality risk at low prey densities. Predator cues decreased the intensity of type 2 functional responses in simple and increased the strength of type 3 in complex ones. A possible explanation to that is that in presence of predators, E. marinus decreased their activity but in safer spaces the activity increased as a compensation, but anyway, it shows how BMII can lead to stabilizing a system.
11 predator densities, which in theory would increase escape behaviours but didn’t counteract the effects from protection from human predation. Despite this, in other ecosystems where human predation mainly focus on high order consumers, lower order consumers would magnify escape behaviours in protected areas (Rhoades et al, 2018)
As a final note, most of the described interactions have three trophic levels for reasons of simplicity, but the reach of TMII is not limited there. We can consider previous examples from Wada et al (2013) and Premo and Tyler (2013) as four chained interactions, since the receiver causes changes in biodiversity affecting fourth species. We can find also another interesting example in nearshore habitats near the Aleutian Islands. There was a decline in otter populations attributed to a shift in the killer whale diets, which seemly induced otters to move shoreward, to shallow waters where predation risk by killer whales was low. That induced anti-predator responses in sea urchin, which dispersed away from damaged urchins discarded by otters. That allowed kelp patches to form in areas vacated by sea urchins, in a four chained behaviourally initiated and twice transmitted trophic cascade (Peckarsky et al, 2008).
Predator -prey -competitor interactions
Predator presence may cause a prey to move to safer spaces, but that may cause competition with species that already inhabit that places, not only because of the resources, but also for the “free enemies space”. The competition may be strong enough to displace the first species from the safe habitat, reducing their numbers or activity because of the predator presence (Fig. 5).
Beerman et al (2018) described how the amphipod
Echinogammarus marinus reduced their activity
and concentrated in shelter microhabitats when the sea scorpion predator Taurulus bubalis was present. But when there was also an interspecific
competitor, Gammarus locusta, the probability of finding E. marinus in sheltered habitats was lower, and G. locusta was mainly found there. In the absence of predator, competitor presence did not affect E. marinus habitat choice. It is also possible for the receiver specie to be benefited due to the displace or reduced activity of its competitor (Fig. 6). Archaeogastropod Nerita
scabricosta avoid the areas where the neogastropod predator Purpura pansa is present, what
favours the settlement of Chthamalus barnacle larvae due to the reduced disturbance and benefits litorine gastropods too (Dill et al, 2003).
Behavioural response in presence of predators may however benefit species in the same trophic level of the transmitter in a different way, not only due to their absence or reduced activity. Common Indo‐ Pacific damselfish, Pomacentrus amboinensis aggressively defend nesting territories against potential egg predators such as moonwrasse
Thalassoma lunare. This fish is also a potential
predator on damselfish juveniles, so those juveniles who settle in male territories have higher survival rates. Moreover, due the change in selective predation, phenotypic selection favoured smaller individuals (McCormick and Meekan, 2007). Despite in this case there were interactions on two
populations of the same species, a similar interaction in which two different species participate is also plausible.
Figure 5: BMII between one predator species and two prey competitor species in which the receiver is negatively affected.
12
Predator -prey -predator interactions
Despite that we usually consider that two predators that feed on the same species are competing, sometimes appears a phenomenon called “predator facilitation” (Fig 7) where the presence of one type of predator facilitates the existence of another one. When that effect is reciprocal, it is called “apparent mutualism” (Dill et al, 2003).
Cetaceans and seabirds use to forage in close proximity, and rather than compete, seabirds benefit from this kind of associations. Prey aggregate near the surface in the presence of subsurface predators such as pinnipeds or cetaceans, which make them more accessible to seabirds, and also other mammals and teleosts, in a BMII. Several examples of this kind of interactions have been reported. Schools of pelagic fish such as herrings and sardines use to do that in presence of cetaceans and pinnipeds, becoming more vulnerable to seabirds. Deep dwelling preys, usually inaccessible to seabirds, have been documented to be driven to the surface from
long-finned pilot whales and balaenopterid whales such as fin whales and common minke whales. Grey whales also bring amphipods such as Ampelisca and Byblis to the surface, where are consumed by Fulmarus glacialis, Rissa tridactyla, Phalaropus fulicara and other seabirds. Rising populations of eastern-Pacific grey whales could benefit diving ducks such as Melanitta
perspicillata. Also in eastern-Pacific, Parkinson’s petrels seem to associate with rare
deep-diving delphinids in order to scavenge scraps. Common and roseate terns are usually found in proximity with dolphins and tuna, suggesting the same type of interaction. And seabirds are not the single group facilitated by cetaceans. Cape fur seals and sharks such as Carcharhinus
brachyurus, C. brevipinna and C. obscurus have been suggested to be facilitated by activity of
cetaceans such as common dolphins and Bryde’s whales during the winter migration of sardines in regions in the south-east of Africa (Kizska et al, 2015).
Of course, there are cases of predator facilitation that are not initiated by marine mammals. In south eastern estuaries of the USA, blue crabs Callinectes sapidus and spot Leiostomus
xanthurus (a fish) co-occur and share amphipods as a prey. Crabs cause amphipods to climb
up any available structures, facilitating spots to feed on them and increasing their survival. In rocky intertidal zone, sea urchins Strongylocentrotus purpuratus flee and climb upon one another's backs in order to avoid the asteroid Pycnopodia helianthoides. That facilitates sessile anemones Anthopleura xanthogrammica to fed over sea urchins, because sometimes they lose their grip and are swept by the waves into tide pools. Under other circumstances, there would be hard for anemones to fed on sea urchins (Dill et al, 2003).
Behaviour of a prey can, however, affect competition between two predator species in a different way than predator facilitation, as can show previously mentioned case of male damselfishes guarding territories. T. lunare was excluded from territories, but dottyback
Pseudochromis fuscus, another predator of small fishes but not eggs, was not (McCormick and
Meekan, 2007). That presumably benefit P. fuscus due to the lower competition.
13
Predator – competitor – prey interactions
Usually, competition between two predators is indirectly driven by their prey density. Despite that, it is possible a BMII where direct interactions between predators positively affect preys (Fig. 8), in the opposite way than previous examples where predator diversity strengthen the effects over prey.
O’Gorman et al (2008) conducted an experiment with six species of fish as predators, analysing their gut content to identify invertebrate preys showed that
increasing predator diversity positively affects preys and weakens the strength of trophic cascades, the opposite effect that Byrnes et al (2005) found (previously explained). One of the possibilities they suggested was the mutual interference between predators, due to intimidation, or avoidance due to the chance of intraguild predation (since trophic levels are not closed categories).
Prey – Predator – prey interactions
There are other BMII between species in the same trophic level that are driven by a common predator.
In Shark Bay, tiger sharks mediate a BMII between dugongs and bottlenose dolphins. Dugongs density is highest in shallow habitats, so tiger sharks prefer them. It causes dolphins and other shark prey to avoid that habitats because of the danger, despite they are more productive (Kizska et al, 2015).
Physiologically mediated indirect interactions
BMII are capable to affect marine ecological interactions in many different ways, but they are mainly driven by anti-predator behaviours. However, several TMII in marine ecosystems are not driven by behavioural changes but by physiological responses of the transmitter, and there is a high diversity of mechanisms in which Physiologically Mediated Indirect Interactions take form, such as toxicity, chemical signalling or induced morphological responses.
Toxicity and disease
Presence of invasive species such as algae or marine plants can affect negatively native ones due to toxic chemical compounds or modification of abiotic conditions, and indirectly affecting other species.
Sears & Ross (2010) monitored in northern New Zeeland the effects of the benthic dinoflagellate Ostreopsis siamensis (Fig. 9, a) on the predator – sea urchin- macroalgae trophic cascade. O. siamensis forms blooms and is highly toxic, and is spreading through the world due to ocean warming and human activities (Fig. 9: b), as a new source of anthropogenic stress. Historically, the main and most widespread anthropogenic stressor was overfishing of top-level predators. But nowadays, with multiple anthropogenic stressors acting on the ecosystems (such as nutrients, introduced species, and in this case, disease), they are likely to interact with overfishing and alter trophic cascades. When blooms were present, sea urchins
14 hold and then lose their spines (Fig. 9: C). That caused a decrease in sea urchin density, apparently by the facilitation of predation by fishes (Fig. 9: d). Due to that, macroalgal density increased, but only in reserves where fishing was prohibited. Outside the reserves, where fish density was very low, sea urchin and macroalgae density were not affected despite urchin health problems. That showed how TMII can interact with anthropogenic effects, not just in an additive way. Morover, high densities of O. siamensis can result in direct mortality of sea urchins, masking the effects of overfishing. A similar example can be found between the habitat modifying invasive algae Caulerpa taxifolia, who typically spreads forming dense monospecific beds, and the bivalve Anadara
trapezia, native from southeast Australia. Anadara’s predators use to forage in
unvegetated sediments, so Caulerpa initially provides protection from them. The abiotic conditions created by Caulerpa (low oxygen and high sulphides) also deterred their predators. But hypoxic conditions (probably created by associated reducing bacteria) also affect negatively to Anadara. Low oxygen force them to expose a bigger part of their body, and decreases the force of its abductor muscle, and its shell thickness and tissue health. That increased not only their vulnerability to predators, but also significantly their susceptibility to non-predatory death (Byers et al, 2010).
However, harmful algal blooms may also affect trophic interactions in the opposite way, by providing a chemical defence against predation. In northern California coast, five species of seabirds (black oystercatchers, godwits, sanderlings, whimbrels, and willets) have been demonstrated to avoid their two major invertebrate prey species, (sea mussels Mytilus
californianus and sand crabs Emerita analoga), when their concentrations of saxitoxin
exceeded certain ranges. Exactly the same result was found in south east Alaska with sea otters when their preferred prey, butter clams Saxidomus giganteus accumulated the same toxin. Saxitoxin, also known as Paralytic Shellfish Poisoning Toxin, is produced by dinoflagellates and is the most lethal and widespread toxin produced by harmful algal blooms. Both sea otters and seabirds, at intermediate prey toxicity, discarded prey tissues with higher concentrations of toxin and consumed the remaining body part. Female sea otters have also been observed giving the “healthy” pieces to pups. With high prey toxicity, they just shift to other preys, such as limpets or sea pens, that don’t accumulate toxins. That shift in behaviour is possible due to that species open or dismember their prey and contact it with their sensory organs before consuming them, allowing the predator to “test the food”. This is not the case for piscivorous mammals or seabirds, who usually consume their prey as a whole. They are the groups most commonly associated with mass mortalities due to toxins from harmful algal blooms (Kvitek and Bretz, 2004, 2005).
Induced morphological responses.
Some species respond not only behaviourally but also morphologically to the presence of predator cues. Freeman et al (2013) discovered that Australian native whelk Haustrum
vinosum reduced shell thickness in presence of crab Carcinus maenas cues (and also reduced
foraging). That is interesting because Carcinus is an invasive predator, and the study showed Figure 9: Blooms of Ostreopsis siamensis (a)
Ostreopsis cells, (b) the seaweed covered in Ostreopsis, (c) sea urchin with almost complete loss of
15 no difference between the responses of whelks that had a common history with that crab from 0, 20 and 100 years ago. Despite this, shell thickness does not seem to be a defensive advantage against predation, so it may be a secondary effect of the reduced foraging. So, TMII seem to be behaviourally mediated. Induced morphological defences have been, however, reported in other aquatic animals. Daphnia lumholtzi, a freshwater cladoceran, forms long head and tail spines in presence of fishes, in response to chemicals released by them. It’s an invasive species in North America, and that plasticity gives it a competitive advantage against other native species such as Daphnia pulicaria in presence of predators, whereas in absence of them D.
pulicaria is competitively superior (Engel and Tollrian, 2009). There is no reason to suppose
that a similar TMII wouldn’t be plausible in marine ecosystems.
Induced chemical responses.
The existence of inducible chemical defences in plants and algae is well known, and give us an example about how physiological changes can affect other type of interactions such as competition:
Periwinkle Littorina littorea and isopod Idotea baltica both feed on the brown seaweed Fucus
vesiculosus, but the competitive advantage of these species for the resource is not symmetrical
due to a physiologically mediated indirect interaction driven by the algae. F. vesiculosus showed an inducible anti-herbivore response when grazed, but the two predator species are not equally affected. Idotea baltica showed a preference by previous non-grazed algae, but
Litorina littorea showed no preference between previously grazed and non-grazed algae. This
means that grazing by L. littorea has more negative effects on the Fucus than I. baltica. That asymmetrical TMII forces Idotea baltica to disperse, in order to find more palatable ungrazed algae (Yun et al, 2010). Another similar experiment conducted by Long et al (2007) with the same species but also with Litorina obtusata showed that the intensity of inducible responses may vary according to the inducer species. Inducible anti-herbivore response doesn’t have to be initiated by direct grazing. Codium platylobium has shown a weak trend to decrease its palatability due to the presence of grazer Turbo sarmaticus, despite not directly grazed by it, what suggest the possibility to react to waterborne cues originated by grazer, stimulating the production of secondary compounds (Díaz et al, 2006).
Surprisingly, some algae species have shown to stimulate grazing when previously grazed. Snails Paridotea rubra and Tricolia capensis preferred previously grazed pieces of Gracilaria
capensis and Hypnea spicifera, respectively, than intact ones. There has been suggested
different explanations, but all of them implies the existence of TMII. The algae may produce compounds that repel other grazers but attract those ones, what would mediate an interaction between another grazer as an initiator and those snails. There is also possible a complex interaction that finally would benefit the algae. For example, if the grazer cleans the algae of epiphytes or possible competitors, that would cause a heavier damage to it. Even if grazers are just attracted to algae by compounds that have not a defensive function, that would create a TMII with the grazer that feds on the algae in first place (Díaz et al, 2006). Benefit for the algae or plant is, however, more obvious when the attracted specie is a predator of the grazer. Coleman et al (2007) demonstrated in an experiment that waterborne cues released by
Ascophyllum nodosum (a marine plant) when grazed by Littorina obtusata, attracted Carcinus maenas (a crab predator). That TMII is interesting from an evolutionary perspective.
16
Chemical signalling
As said before, some species may respond to alarm cues from injured conspecifics. But when a species has the capacity to respond to alarm cues released from other species, it transforms in the receiver from a physiologically mediated indirect interaction initiated by a predator where the transmitter is the specie that released the alarm cue. Panulirus argus, a temporary reef-dweller spiny lobster, is able to recognize the alarm cues from injured individuals of P. guttatus a reef-obligate relative, and avoid shelters where they do (Briones-Fourzán et al, 2007). Similar mechanism has been suggested to occur in the brown alga Ascophyllum nodosum, which in response to waterborne cues increases the production of secondary metabolites, decreasing its palatability when neighbouring algae are grazed by Littorina obtusata (Díaz et al, 2006).
Interference
Another mechanism that could potentially generate a physiologically mediated indirect interaction is interference. In terrestrial ecosystems is well known that the presence of an invasive plant species can affect native ones interfering with pollination (because of their own pollen) and by doing so reducing their seed production, or difficult the dispersal and establishment of them, not limiting to competition in adult stages but exacerbating it. In marine ecosystems, where there are many sessile organisms and most of animals have external fecundation, there is no reason to assume that kind of non-lethal effects don’t exist. Heterospecific sperm can disrupt fertilization as well as heterospecific pollen, and sessile organisms will find difficulties to settle in presence of competitors in a similar way than plants. An experiment conducted using two sessile marine invertebrates, invasive Styela plicata and native Microcosmus squamiger, demonstrated that S. plicata had negative effects over settlement of M. squamiger. Seemingly, they were caused by active avoidance of the larvae or by some kind of allelopathy, but it had significant post-settlement effects. Larvae were forced to continue searching for a suitable habitats, and that not only increase the planktonic mortality rate. M. squamiger has non-feeding larvae, so larval swimming reduced its reserves, and post-metamorphic growth, what causes a decrease in their performance (Rius et al, 2009)
Change in selective pressures
Modification of phenotypic traits in a population due to changes in selective pressures can affect other species, and show how humans can also act as direct initiators of TMII. Overfishing of large bodied fish populations has caused a decline in their body sizes, even when their biomass has remained constant, and its effects can be wrongly attributed to a decrease in densities.
17 experiment used different densities of guppies from two different populations: one that suffered a low predatory pressure, and another one that suffered a high one. The last ones mature younger and at a small size than the first ones, and invested more resources into reproduction. The experiment showed that high predated guppies consumed more invertebrates while low predated ones consumed more algae, so algae were both directly and indirectly benefited from high predated guppies phenotypes. That even affected the nutrient flow. High predated guppies increased the nutrient recycling rate, since they excreted more nitrogen, what indirectly increased algal productivity (Travis and Lotterhos, 2013).
Other Interactions
Some TMII can’t be easily considered behaviourally mediated nor physiologically mediated. Three of them are described below in this review, but other mechanisms probably exist.
Interactions initiated or mediated by parasites
One important driver of Indirect Interactions are parasites (including pathogens). Marine ecosystems have some of the highest numbers of invasive species in the world and parasites have an important role in the success and impact of them, acting as initiators or transmitters, or affecting the invasion processes positively or negatively (Goedknegt et al, 2016). Some of the mechanisms in which parasites can generate TMII could be considered behaviourally, physiologically, or even density mediated, but due to the important roles of parasites in them, it’s easier to describe them together. It is possible to divide parasite interactions in the context of invasions in six ways (Fig 10):
Parasite release/reduction: The invasive species loses some or all their parasites in the invasion process, so it has a competitive advantage against native species, which carry their own parasites. As an example, in South Africa invasive mussel Mytilus galloprovincialis has lost its
parasites and is outcompeting the native mussel Perna perna, which suffer the infection of two trematodes that affects its growth and reproduction. The same happened in Arcachon Bay (France) with the invasive gastropod Cyclope neritea and in the Pacific coast of North America with the snail Batillaria attramentaria. They were both
much less infected by parasites than their native competitors: gastropod Nassarius reticulatus and snail Cerithidea californica, respectively (Goedknegt et al, 2016).
18 shad, an anadromous fish, transported the native marine protist Ichtyophonus (with low parasite-host specificity) into the Columbia River System. That allowed the parasite to infect the salmon Oncorhynchus tshawytscha, and potentially establish on freshwater systems. Parasite spillback can mediate indirect interactions between two species in different ways. For example, two species that don’t share resources but are infected by the same parasite in a form of apparent competition that may cause the exclusion of one of them. Or it may cause their co-existence if the strongest competitor is also more strongly affected by the parasite than the inferior one (Goedknegt et al, 2016).
Introduction of free living stages: Parasites are not always co-introduced with its original host, and they can infect directly native hosts, with cascading effects over other native species. For example, the parasitic isopod O. griffenis was introduced in the west coast of North America into ship ballast water and caused the collapse of the native mud-shrimp Upogebia
pugettensis. That had important consequences for the entire ecosystems, such as food
reduction for fish. Ballast water may be also a vector for human diseases. In 1991 a Cholera 01 strain was found in Alabama in oysters and fish, that had been seemly transported in ballast water from South America, that had suffered a spread of a similar strain half a year earlier (Goedknegt et al, 2016).
Parasite co-introduction: Parasite may be introduced together with their original host and only affect it, and any native species. It may give a competitive advantage to native species. Thought rare, there are examples of that. Two species of snapper (Lutjanus) were introduced to Hawaiian Islands from French Polynesia in 1950s and they carried six monogenean species that have not infected any native species (at least to be known). Unfortunately, the effects of co-introduction without spillover are not well known yet, but they may have important ecological implications. For example, they may take the form of strong regulative effects over invasive populations, and by doing so, affect competitive interactions between native and invasive species (Goedknegt et al, 2016).
Parasite spillover: parasites can be co-introduced with their original host but also infect naïve native species, what will cause negative effects over them, as emergent diseases. There are many known cases, both micro and macroparasites and from 14 different taxa. Usually they are direct parasites, so they only need one potential species to be present. Host identity is diverse too, with fish, crustaceans, and molluscs. Most of that introductions are related with aquiculture or fishing practices, such as restocking of wild populations or the moving of life organisms for aquiculture, and parasite spillover can affect hosts both in the wild as in the aquiculture settings. The number of native hosts may also be high. As an example, the global eel trade caused the nematode Anguillicola crassus to infect seven different eel species in four different continents. Even if it may be considered an indirect effect by itself, parasite spillover can have indirect consequences over more species. For example, in the Chesapeake Bay mud crabs Eurypanopeus depressus feed significantly less over mussels when infected by the barnacle Loxothylacus panopaei, that have been introduced in the 1960 from the Gulf of Mexico. That released crab predation intensity on mussels (Goedknegt et al, 2016).
Parasite transmission interference: Invader species may be neither a host nor a parasite but reduce the parasite infections, for example, if the new species predates on parasite free-living infective stages. The invasive crab Hemigrapsus takanoi actively prey on the trematode H.
elongata cercariae (infective stage), creating a DMII on their hosts. But it may create a physical
or chemical barrier that interferes with the transmission in a TMII. For example, trematode
Himasthla elongate free-living stages become entangled in the invasive seaweed Sargassum muticum, avoiding them to infect the native mussel Mytilus edulis. Another possibility is when
19 These mechanisms are not mutually exclusive, and they can act synergistically or counteract themselves (Goedknegt et al, 2016).
Not all TMII initiated or mediated by parasites are related to invasions. In New Zealand coasts, trematodes Maritrema novaezealandensis and Philophthalmus sp. are known to induce behavioural changes on intertidal gastropod Zeacumantus subcarinatus. M.
novaezealandensis infected snails take significantly shorter time to exit the water when they
perceive cues from predator crabs and injured conspecifics than uninfected snails, whereas
Philophthalmus infected snails take longer. That seems to be more a pathologic side effect of
the infection than an adaptive manipulation by the parasite (even when they are known cases of parasites manipulating host behaviour to their benefit), but it drives TMII. Snails infected by M. novaezealandensis seem to be more vulnerable to predation by seabirds, while
Philophthalmus infected ones are more vulnerable to crab predation. Those interactions are
more complex since both parasites have seabirds as definitive host, but M. novaezelandensis has crustaceans as second intermediate hosts (Kamilla and Poulin, 2012).
Physiological changes induced by parasites are also known. Typical effects related with parasite infections such as a decrease in fecundity, body mass and condition can mediate TMII to other species due to the reduction of prey, competitor or predator density or their ability to escape, compete, or predate. But some parasites are known to induce other traits in their hosts not directly related with low health conditions. For example, in hypersaline wetlands in Spain and Portugal, cestodes have been reported to colour their brine shrimps hosts (Artemia spp.) bright red, as well as induce positive phototaxis increasing surface time. That make shrimps more visible to birds, the cestodes definitive hosts (Goedknegt et al, 2016). Presumably, similar interactions would exist in many marine ecosystems.
More intense are responses driven by ecosystem engineer hosts. Hydrobia ulvae is an herbivore mud-snail that lives in soft-bottom intertidal systems and a potential host for different species of trematodes. It is considered an ecosystem engineer due to its grazing and bioturbation activity. Mouritsen and Haun (2008) showed that increase of parasite infection rates in populations had similar effects on the diatom community structure as a decrease in the total density of snails. Those effects extended to other faunal species, but while snail density was shown to affect two species, there were seven species that were indirectly affected by the infection rates of a snail.
Microbial interactions
TMII are not limited to pluricellular organisms. Microorganisms also show dynamics than can be considered DMII and TMII, with protozoans as initiators, bacterium as transmitters and viruses as receivers. Protozoans reduce bacterial density, in a DMII that negatively affect viruses because of the lower numbers of potential hosts, and direct consumption of viruses in lysogenic cycle (Fig 11 a & b). But that reduction may have also positive effects for viruses. Predation by protozoans increases nutrient availability so enhance growth of surviving bacteria.
That benefits viruses, which have shorter latent periods and release a higher number of virions per bacterial lysis. Also, predation decreases diversity, which allows viruses to attack a major proportion of bacteria, and reduce the capacity of them to adapt against that, in a TMII that weakens the effect of DMII on viruses (Fig 11 c). This may be a reason that allows the coexistence of protozoans and viruses in that systems, despite they share the same resource.
20 Interactions between those tree groups are, however, more complex, since we are not considering individual species but entire groups, due to logistical limitations. There are bacteria and protozoans that fed on free-“living” viral particles, viruses that infect protozoans, and even viruses that parasites another viruses, needing to co-infect an amoeba to be able to reproduce. That interactions create a complex cycle, where it is hard to determine which are initiators, receivers or transmitters, but may ultimately lead to the coexistence of those groups. (Miki & Jacquet, 2010).
Movement of nutrients and habitat coupling,
The last way in which interactions between species can be mediated is trough movement of nutrients within or between depth strata, habitats or ecosystems. That’s especially true in the case of cetaceans and other marine megafauna, due to their big size and their capacity to move through long distances (Estes et al, 2016). Faeces and carcasses, as well as regurgitation, are important sources of nutrients to many species in different ecosystems (sometimes limiting ones, such as iron). Whale, seal and manatee faeces are known to stimulate production in some ecosystems, translocating nutrients from different habitats and coupling them, what ultimately enhance fish populations. Whale carcasses that fall to the sea floor or move nearshore are massive pulses of organic matter critical to sustain certain populations. That processes are closely linked with behavioural traits, since their spatiotemporal distribution is dependent of migratory and foraging decisions (Kizska et al, 2015), so any influence that could alter that behaviours, such as fishing or the presence of predators, will ultimately affect that interactions.
Discussion
This study shows that Trait Mediated Indirect Interactions are at least as important as DMII in marine ecosystems. There is a large diversity of mechanisms, which are present worldwide and occur at all trophic levels, taxa and guilds, and even in the microbial world. Since their effects may improve or counteract Density Mediated Interactions, and often extend further in the food web, they are a key to understanding ecological dynamics. Growing evidence shows that without considering TMII, our knowledge about marine ecosystems cannot be considered complete. There are however many difficulties to determine with accuracy how DMII and TMII interact.
Behaviourally Mediated Indirect Interactions can take several directions (Top predator– mesopredator–prey; predator-prey-predator…), and some of them not included in this review (prey-predator-competitor) could exist even if not described yet. Due to the high number of species any single species can interact with, it is difficult to predict what the outcome will be. Behaviour itself may be challenging to predict unless known before, as when prey move to spaces with higher predator density because it is easier to escape there. When BMII intensify DMII or just take the same direction, it is hard to measure the effect of each one separately, especially if intensity of BMII is affected by density, as in the experiment of Wada et al (2015). Looking at the results of Gravem and Morgan (2016) in which snails only stop feeding during short periods of time, it is possible to suggest that BMII could have just a more immediate regulatory effect, but when the causes persist over time the interaction is maintained due to density changes. However, in a study by Babcock et al (2010) it was shown that indirect effects take longer time to manifest than direct ones, suggesting that behaviour can delay effects on lower trophic levels. In natural ecosystems, hiding animals may find moments in which predators are not present when they can abandon their refugees and forage, a chance that they not have in laboratories.
21 however, hard to predict if BMII will weaken or counteract DMII without a wide knowledge of the targeted species interactions and their environmental conditions, as show the examples of behavioural facilitation and suggest the difference in anti-predator behaviour in different marine reserves found by Rohades et al (2018). The issue becomes even more complicated with Physiologically Mediated Indirect Interactions, due to the diversity of mechanisms in which they can act (poisoning, chemical signalling, morphological changes…). For comparison, BMII mainly consist of anti-predation behaviour. The examples about harmful algal blooms and fish size reduction caused by overfishing show how under apparently similar circumstances, physiologically mediated indirect interactions may have completely opposite effects. Fortunately, other interactions such as those related to parasites are much more constrained, and therefore they are less likely to appear unexpectedly.
Further studies will be needed to incorporate the effects of TMII in a strong model to marine conservation and management. However, this doesn’t mean that knowledge about different ways in which TMII can affect an ecosystem, and the examples described in this review, are completely useless until that day. Just considering the existence of different ways in which species could interact that would cause different results, and having some clues that allow marine conservationists to suspect them, can have significant benefits for theory and practice of marine ecosystems conservation and management.
Acknowledgment
I thank Agneta Anderson for her help and guidance. She always had a simple solution for all the problems I encountered during this project.
References
Abrahams, P. A. 2007. Defining and measuring the impacts of dynamic traits on interspecific interactions. Ecology. Vol. 88 (10), p. 2555-2562. doi: 10.1890/06-1381.1
Alexander, M. E.; Dick, J. T. A.; O'Connor, N. E. 2013. Trait-mediated indirect interactions in a marine intertidal system as quantified by functional responses. Oikos. Vol. 122 (11), p. 1521-1531. doi: 10.1111/j.1600-0706.2013.00472.x
Babcock, R. C.; Shears, N. T.; Alcala, A. C.; Barrett, N. S.; Edgar, G. J.; Lafferty, K. D.; McClanahan, T. R.; Russ, G. R. 2010. Decadal trends in marine reserves reveal differential rates of change in direct and indirect effects. Proceedings of the National Academy of
Sciences of the United States of America. Vol. 107 (43), p. 18256-18261. doi:
10.1073/pnas.0908012107
Beermann, J.; Boos, K.; Gutow, L.; Boersma, M.; Peralta, A. C. 2018. Combined effects of predator cues and competition define habitat choice and food consumption of amphipod mesograzers. Oecologia. Vol. 186 (3), p. 645-654. doi: 10.1007/s00442-017-4056-4
22 Byrnes, J.; Stachowicz, J. J.; Hultgren, K. M.; Hughes, A. R.; Olyarnik, S. V.; Thornber, C. S. 2005. Predator diversity strengthens trophic cascades in kelp forests by modifying herbivore behaviour. Ecology letters. Vol 9 (1), p. 61-71. doi: 10.1111/j.1461-0248.2005.00842.x
Briones-Fourzán, P.; Ramirez-Zaldivar, E.; Lozano-Alvarez, E. 2008. Influence of Conspecific and Heterospecific Aggregation Cues and Alarm Odors on Shelter Choice by Syntopic Spiny Lobsters. The Biological bulletin. Vol. 215 (2), p. 182-190. doi: 10.2307/25470699
Coleman, R.; Ramchunder, S. J.; Davies, K. M. Moody, A. J.; Foggo, A. 2007. Herbivore‐ induced infochemicals influence foraging behaviour in two intertidal predators. Oecologia. Vol. 151, p. 454-463. doi: 10.1007/s00442-006-0601-2
Díaz, E.; Gueldenzoph, C; Molis, M; McQuaid, C; Wahl, M. 2006. Variability in grazer-mediated defensive responses of green and red macroalgae on the south coast of South Africa. Marine biology. Vol. 149 (6), p. 1301-1311. doi: 10.1007/s00227-006-0320-6 Dill, L. M. 2017. Behavioural ecology and marine conservation: a bridge over troubled water?
ICES journal of marine science. Vol. 74 (6), p. 1514-1521. doi: 10.1093/icesjms/fsx034
Dill, L. M.; Heithaus, M. R.; Walters, C. J. 2003. Behaviorally mediated indirect interactions in marine communities and their conservation implications. Ecology. Vol. 84 (5), p. 1151-1157. doi: 10.1890/0012-9658(2003)084[1151:BMIIIM]2.0.CO;2
Engel, K.; Tollrian, R. 2009. Inducible defences as key adaptations for the successful invasion of Daphnia lumholtzi in North America? Proceedings. Vol. 276 (1663), p. 1865-1873. doi: 10.1098/rspb.2008.1861
Estes, J. A.; Heithaus, M; McCauley, D. J.; Rasher, D. B.; Worm, B. 2016. Megafaunal Impacts on Structure and Function of Ocean Ecosystems. Annual review of environment and
resources. Vol. 41, p. 83-116. doi: 10.1146/annurev-environ-110615-085622
Freeman, A. S.; Wright, J. T.; Hewitt, C. L.; Campbell, M. L.; Szeto, K. 2013. A gastropod's induced behavioral and morphological responses to invasive Carcinus maenas in Australia indicate a lack of novelty advantage. Biological invasions. Vol. 15 (8), p. 1795-1805. doi: 10.1007/s10530-013-0409-z
Geraldi, N. R.; Macreadie, P. I. 2013. Restricting Prey Dispersal Can Overestimate the Importance of Predation in Trophic Cascades. PloS one. Vol.8(2) doi: 10.1371/journal.pone.0055100
Goedknegt, M. A.; Feis, M. E.; Wegner, K. M.; Luttikhuizen, P. C.; Buschbaum, C.; Camphuysen, K.; van der Meer, J.; Thieltges, D. W. 2016. Parasites and marine invasions: Ecological and evolutionary perspectives. Journal of sea research. Vol. 113, p. 11-27. doi: 10.1016/j.seares.2015.12.003
Grason, E. W.; Miner, B. G. 2012. Behavioral plasticity in an invaded system: non-native whelks recognize risk from native crabs. Oecologia. Vol. 169 (1), p. 105-115. doi: 10.1007/s00442-011-2188-5
Gravem, S. A.; Morgan, S. G. 2016. Prey state alters trait‐mediated indirect interactions in rocky tide pools. Functional Ecology. Vol 30, p. 1574-1582. doi: 10.1111/1365-2435.12628 Haggerty, M. B.; Anderson T. W.; Long, J. D. 2018. Fish predators reduce kelp frond loss via a
23 Heithaus, M. R.; Frid, A.; Wirsing, A. J.; Dill, L. M.; Fourqurean, J. W.; Burkholder, D.; Thomson, J.; Bejder, L. 2007. State-dependent risk-taking by green sea turtles mediates top-down effects of tiger shark intimidation in a marine ecosystem. The journal of animal
ecology. Vol. 76 (5), p. 837-844. doi: 10.1111/j.1365-2656.2007.01260.x
Heithaus M. R.; Wirsing, A. J.; Burkholder, D.; Thomson, J.; Dill, L. M. 2009. Towards a predictive framework for predator risk effects: the interaction of landscape features and prey escape tactics. The journal of animal ecology. Vol. 78 (3), p. 556-562. doi: 10.1111/j.1365-2656.2008.01512.x
Kamiya, T.; Poulin, R. 2012. Parasite-induced behavioural changes to the trade-off between foraging and predator evasion in a marine snail. Journal of experimental marine biology
and ecology. Vol. 438, p. 61-67. doi: 10.1016/j.jembe.2012.09.002
Kindinger, T.L.; Albins, M.A. 2017. Consumptive and non-consumptive effects of an invasive marine predator on native coral-reef herbivores. Biological Invasions. Vol. 19 (1), p. 131-146. doi: 10.1007/s10530-016-1268-1
Kintzing, M. D.; Butler, M. J. 2014. Effects of predation upon the long-spined sea urchin Diadema antillarum by the spotted spiny lobster Panulirus guttatus. Marine ecology
progress series. Vol.495, p.185-191. doi: 10.3354/meps10568
Kiszka, J. J.; Heithaus, M. R.; Wirsing, A. J. 2015. Behavioural drivers of the ecological roles and importance of marine mammals. Marine ecology progress series. Vol. 523, p, 267–281. doi: 10.3354/meps11180
Kvitek, R.; Bretz, C. 2004 Shorebird foraging behavior, diet, and abundance vary with harmful algal bloom toxin concentrations in invertebrate prey. Marine ecology progress series. Vol.293, p.303-309
Kvitek, R.; Bretz, C. 2005 Harmful algal bloom toxins protect bivalve populations from sea otter predation. Marine ecology progress series. Vol. 271, p. 233-243. doi: 10.3354/meps271233
Long, J. D.; Hamilton, R. S.; Mitchell, J. L. 2007. Asymmetric competition via induced resistance: Specialist herbivores indirectly suppress generalist preference and populations
Ecology. Vol. 88 (5), p. 1232-1240. doi: 10.1890/06-1585
McCormick, M. I.; Meekan, M. G. 2007. Social facilitation of selective mortality. Ecology. Vol. 88 (6), p. 1562-1570 doi: 10.1890/06-0830
Miki, T.; Jacquet, S. 2010. Indirect interactions in the microbial world: specificities and similarities to plant-insect systems. Population ecology. Vol. 52 (4), p. 475-483. doi: 10.1007/s10144-010-0235-4
Mouritsen, K. M.; Haun, S. C. B. 2008. Community regulation by herbivore parasitism and density: Trait-mediated indirect interactions in the intertidal. Journal of experimental
marine biology and ecology. Vol. 367 (2), p. 236-246. doi: 10.1016/j.jembe.2008.10.009
O’Gorman, E. J.; Enright, R. A.; Emmerson, M. C. 2008. Predator diversity enhances secondary production and decreases the likelihood of trophic cascades. Oecologia. Vol. 158 (3), p. 557-567. doi: 10.1007/s00442-008-1165-0
24 Peckarsky, BL; Abrams, PA; Bolnick, DI; Dill, LM; Grabowski, JH; Luttbeg, B; Orrock, JL; Peacor, SD; Preisser, EL; Schmitz, OJ; Trussell, GC. 2008. Revisiting the classics: Considering nonconsumptive effects in textbook examples of predator-prey interactions.
Ecology, Vol. 89 (9), p. 2416-2425. doi: 10.1890/07-1131.1
Premo, K. M.; Tyler, A. C. 2013. Threat of predation alters the ability of benthic invertebrates to modify sediment biogeochemistry and benthic microalgal abundance.Marine ecology
progress series. Vol. 494, p. 29-39. doi: 10.3354/meps10561
Reiss, K.; Herriot, M. B.; Eriksson, B. K. 2014. Multiple fish predators: effects of identity, density, and nutrients on lower trophic levels. Marine ecology progress series. Vol. 497, p. 1-12. doi: 10.3354/meps10622
Reynolds, P. L.; Sotka, E. E. 2011. Non-consumptive predator effects indirectly influence marine plant biomass and palatability. Journal of ecology. Vol. 99 (5), p. 1272-1281. doi: 10.1111/j.1365-2745.2011.01837.x
Rhoades, O. K.; Lonhart, S. I.; Stachowicz, J. J. 2018. Fished species uniformly reduced escape behaviors in response to protection. Biological conservation. Vol. 226, p. 238-246. doi: 10.1016/j.biocon.2018.06.030
Rius, M.; Turon, X.; Marshall, D. J. 2009. Non-lethal effects of an invasive species in the marine environment: the importance of early life-history stages. Oecologia. Vol. 159 (4), p. 873-882. doi: 10.1007/s00442-008-1256-y
Shackell, N. L.; Frank, K. T.; Fisher, J. A. D.; Petrie, B; Leggett, W. C. 2010. Decline in top predator body size and changing climate alter trophic structure in an oceanic ecosystem.
Proceedings. Vol. 277 (1686), p. 1353-1360. doi: 10.1098/rspb.2009.1020
Shears, N. T.; Ross, P. M. 2010. Toxic cascades: multiple anthropogenic stressors have complex and unanticipated interactive effects on temperate reefs. Ecology letters. Vol. 13 (9), p. 1149-1159. doi: 10.1111/j.1461-0248.2010.01512.x
Spyksma, A. J. P.; Taylor, R. B.; Shears, N. T. 2017. Predation cues rather than resource availability promote cryptic behaviour in a habitat-forming sea urchin. Oecologia. Vol. 183 (3), p. 821-829. doi: 10.1007/s00442-017-3809-4
Stallings, C. D. 2008. Indirect effects of an exploited predator on recruitment of coral-reef fishes Ecology. Vol. 89 (8), p. 2090-2095. doi: 10.1890/07-1671.1
Travis, J.; Lotterhos, K. E. 2013. Using experiments and models to untangle direct and indirect effects: is there a hope for understanding fishery systems? Bulletin of Marine Science. Vol. 89 (1), p. 317-335. doi: 10.5343/bms.2012.1024
Wada, Y.; Iwasaki, K.; Yusa, Y. 2013. Changes in algal community structure via density- and trait-mediated indirect interactions in a marine ecosystem. Ecology. Vol. 94 (11), p. 2567-2574. doi: 10.1890/12-0725.1
Wada, Y.; Iwasaki, K.; Yusa, Y. 2015. Prey density affects strengths of density- and trait-mediated indirect interactions of predators on an algal community. Journal of
experimental marine biology and ecology. Vol. 468, p. 67-73. doi: 10.1016/j.jembe.2015.03.015
Weissburg, M.; Smee, Delbert, L.; Ferner, M. C. 2014. The Sensory Ecology of Nonconsumptive Predator Effects. The American naturalist. Vol. 184 (2), p. 141-157. doi: 10.1086/676644 Wirsing, A. J.; Heithaus, M. R.; Frid, A.; Dill, L. M. 2008. Seascapes of fear: evaluating
sublethal predator effects experienced and generated by marine mammals. Marine
