Message

Seagrass transitions

Feedback mechanisms

Seagrass bed dynamics are characterised by a set of positive feedback loops that reinforce seagrass dominance. However, external drivers can reverse some of these feedbacks, leading to a decline in seagrass and a possible regime shift (van der Heide et al. 2011). Such feedback loops are common in marine ecosystems that are dominated by an engineering species, as they create optimal environmental conditions for their own growth (van der Heide et al. 2011).

Seagrass dominated regime

  • Seagrass-turbidity feedback loop (local, well established): Seagrasses are ecosystem engineers and have the ability to promote their own growth by modifying the abiotic environment. They attenuate currents and trap suspended sediments and nutrients, which reduce turbidity and enhance light penetration (Duarte 2002; van der Heide et al. 2007 and 2011; Nyström et al. 2012). This reinforcing feedback loop (red loop in the diagram) plays an important role in maintaining seagrass dominance (de Boer 2007; van der Heide et al. 2011) as seagrass have, compared to other plants, unusually high requirements for good light conditions for photosynthesis and growth (Burkholder etal. 2007) . This feedback is particularly important in shallow areas that can easily become turbid due to strong wave action (van der Heide et al. 2011). Additionally, lowered current velocities and fine bottom sediment enables root attachment for new shoots, thus seagrass facilitate for their own expansion (Cardoso et al. 2004). With greater seagrass biomass the engineering function is augmented (van der Heide et al. 2011), which further reinforces seagrass dominance.
  • Seagrass-algae competition feedback loop (local, well established): Seagrasses compete with algae for space, light and nutrients (Duarte et al. 2006). The ability of seagrass beds to reduce nutrient levels in the water column prevents the proliferation of algae (Valentine and Duffy 2006; van der Heide et al. 2007). Thus, they prevent high turbidity levels from phytoplankton blooms, shading from benthic macroalgae and fouling by epiphytes (Burkholder et al. 2007). In this way they ensure sufficient light uptake for seagrass growth, and therefore, by maintaining oligotrophic conditions seagrasses reinforce their own dominance. This feedback loop is green in the causal loop diagram.
  • Seagrass-herbivore feedback loop (local, well-established): The structural complexity of seagrass beds provides algal herbivores with important habitats and refuges from predators (Duarte 2002; Valentine and Duffy 2006). Mesograzers such as small gastropods and crustaceans prevent seagrasses from being smothered by epiphytic algae by maintaining low algal cover through high grazing pressure (Valentine and Duffy 2006); and planktivores and filter feeders are important in keeping phytoplankton abundance low (Burkholder et al. 2007). It has been shown that seagrass beds with sufficient herbivory function are more resilient to nutrient input as algae are kept in low abundance (Valentine and Duffy 2006). As such, algal herbivores are important for keeping the beds healthy and to facilitate further seagrass expansion, which in turn will provide more habitats for herbivore communities. This loop is represented in blue in the diagram.

Algae dominated regime

  • Algae-seagrass feedback loop (local, well established): An increase in algal abundance, through nutrient loading or reduction in algal herbivores, can shift the feedback dynamics into favouring algae instead of seagrass dominance as algae are inherently better competitors under high nutrient conditions (green loop) (Valentine and Duffy 2006). In deeper coastal waters added nutrients facilitate phytoplankton blooms while in shallower waters blooms of macroalgae and epiphytes are favoured where greater wave action prevents phytoplankton blooms (Burkholder et al. 2007). The former increase turbidity and reduce light penetration (Valentine and Duffy 2006) and as such alter the seagrass-turbidity feedback loop (red loop) to undermine seagrass growth instead of enhancing it. Macroalgae rapidly develop thick canopies that shade seagrasses and epiphytes directly obstruct light from reaching the seagrass leaves (Burkholder et al. 2007). All three scenarios cause a decline in seagrass cover through light reduction and consequently will reduce seagrass engineering function, which further facilitates algal blooms as nutrients can be resuspended into the water column (Burkholder et al. 2007). Additionally, algal blooms can cause significant reduction of oxygen (hypoxia) in the water column and oxygen depletion (anoxia) in the sediments, causing an increase in hydrogen sulphide concentration which is directly toxic to seagrasses (Burkholder et al. 2007; Nyström et al. 2012). Hypoxia can also cause a decline in herbivore populations (Burkholder et al. 2007) and with a reduction in herbivory the algal dominance is further reinforced.
  • Grazer-Epiphytic algae feedback loop (local, well established): If the grazing pressure is reduced (as a result of e.g. overfishing) epiphytic algae is allowed to proliferate and will outcompete seagrass for light and nutrients (Valentine and Duffy 2006). With a reduction in seagrass cover grazer habitats and refuges are reduced making grazers more vulnerable to predation. Furthermore, recolonizing seagrasses have low shoot density and provide poor refuges for grazer populations, leaving them exposed to predation. With low grazing pressure epiphytic algae is favoured and will prevent seagrass re-growth (Valentine and Duffy 2006). This is a combination of the blue and the green feedback loops in the diagram.

Barren sediment regime

  • Turbidity feedback (local, well established):The barren sediment regime is primarily maintained by the reversed seagrass-turbidity feedback loop (red loop) as the seagrass engineering function is reduced due to little or no seagrass biomass. Without dense seagrass cover currents are no longer attenuated, sediments and nutrients can readily be resuspended into the water column and coarse sediments constitute the bottom substrate (van der Heide et al. 2007 and 2011). This causes an increase in turbidity, so light levels may drop below seagrass tolerance (de Boer 2007), which inhibits seagrass recolonisation. Recolonisation is further constrained as new shoots are exposed to currents and can easily be uprooted, especially since the coarse sediments provide little hold for the roots (Cardoso et al. 2004).
  • Grazer-epiphytic algae feedback loop (local, well established): Even if abiotic conditions allow for seagrass recolonisation, success in bed establishment may be impaired by low algal grazing pressure.  New seagrass patches have low shoot density and provide poor refuges for grazer populations, which leaves the grazers exposed to predation. With low grazing pressure epiphytic algae is favoured and will prevent seagrass re-growth through out-competition for light (Valentine and Duffy 2006).  Without dense seagrass patches their engineering function is not re-established and this reinforces the sediment regime (van der Heide et al. 2007).

Drivers

Important shocks

  • Storms: (local to regional, well established): Since seagrasses have a high requirement for light and inhabit depths of 0-30 m, they are sensitive to strong storms, as these can cause increased turbidity and physical damage to the beds. This could mean a shift from seagrass to algal dominance, since algae are better competitors for light than seagrasses are (Duarte 2002; Orth et al. 2006).
  • Disease (local, well established): Seagrass cover can be decreased by sudden disease outbreaks. An example of this is the wasting disease, which is caused by the slime mould Labyrinthula zosterae and which can wipe out extensive seagrass beds, leaving a barren sediment regime. However, outbreaks of wasting disease is most commonly known to cause these large die-offs of seagrasses when they are already under some other form of stress, such as temperature increase, sea level rise or turbid water (Duarte 2002; Borum et al. 2004).
  • Human induced physical disturbance (local, well established): Physical disturbance is an important driver of seagrass loss, both as an external driver and as a human induced shock, such as the actual removal of beds in the case of e.g. port constructions or other forms of coastal developments (Borum et al. 2004).

Main external direct drivers

  • Nutrient loading (local to regional, well established): A shift between seagrass to algal dominance occurs when light instead of nutrients becomes the limiting factor. Seagrasses are slow growing, very light-dependent organisms. They recycle much of their nutrients internally, whilst algae are more efficient in taking up excess nutrients from their surroundings. The direct effects of a higher nutrient load are an increase in fast growing macroalgae and a decrease in seagrasses due to the toxic effect that nitrates have on seagrasses. Indirectly, nutrient loading affects seagrasses by reducing light through the increase of phytoplankton in the water column (Duarte 1995). The decomposing of algae and seagrasses further fuels the phytoplankton blooms by releasing nutrients (Eklöf 2008). Hypoxia also causes both stress to seagrasses and a decline in herbivorous fish, leading to further algal dominance (Burkholder et al 2007; Eklöf 2008). Eutrophication is mainly driven by anthropogenic actions and land use, such as the extensive use of fertilizers in agriculture, aquaculture and sewage coming from the growing human population in coastal areas which are tightly linked to the export rate of nitrogen and contribute heavily to the nutrient input to oceans globally (Duarte 1995 and 2002).
  • Overfishing (local to regional, well established): Overfishing has been linked to market demand leading to an increase in commercial fishing pressure. It also often the result of poverty traps, increased human population, increased unemployment and coastal migration which in turn further increase the unsustainable use of fish stocks (Eklöf 2008). The effects of overfishing vary from grazers overgrazing seagrasses to loss of grazing pressure altogether. When predator pressure on grazers is released due to overfishing, the result is an increase in seagrass grazing, causing a decline and paving the way for algae dominance. However, most commonly overfishing results in a total loss of algal herbivory altogether due to cascading effects of removal of top predators that cause an increase in meso-predator populations which will reduce herbivore populations due to higher predation pressure. A reduction in herbivores releases algae from grazing pressure, which leads to algal overgrowth on seagrasses. This eventually suffocates them and since seagrasses engineer their own habitat, the loss will make the environment less suitable for a recolonisation, due to resuspension of sediments (Heck Jr and Valentine 2006; Duarte 2002). Seagrass beds are dependent on grazers to feed on algae before they get too thick to be eaten. If this grazing pressure is released, algae will grow thick and eventually prevent seagrass growth by reducing light conditions (Heck Jr and Valentine 2006).
  • Physical disturbance (local to regional, well established): Physical disturbance is a strong driver as well as a shock of loss of seagrass beds, as mentioned above. The slow processes of boating, anchoring, dredging and trawling affect seagrass beds negatively throughout a long period of time. They all cause water turbidity and resuspension of sediments as well as physical damage to the seagrasses. The greater part of these disturbances will result in an opportunity for algae to move in and shift the system from seagrass dominance to algal dominance (Borum et al. 2004). Levels of physical disturbance is related to human population growth and our increased activity in coastal and marine areas (Duarte 2002).
  • Siltation/Sediment loading (local to regional, well established): Soil erosion and in particular siltation (fine grain mud and clay particles suspended in the water column) increase turbidity, hence light conditions will no longer be optimal for seagrasses, which creates an opportunity for algal growth (Borum et al. 2004).
  • Aquaculture (local, well established): Aquaculture, today the fastest-growing food industry, is preferably placed close to the highly productive seagrass beds (Duarte 2002). They have significant impact on the shift from seagrasses to algal dominance by shading and deteriorating sediment condition. Since algae have less demand for light, a decrease in seagrass due to this deterioration enhances the risk for a shift of regimes. Fish cages also contribute to excess nutrient and organic matter which will further the decrease in seagrasses. Once the seagrasses start to decline, sediment stability in the area will decrease, thus creating a positive feedback loop speeding up seagrass loss and algal growth (Duarte 2002; Borum et al. 2004).

Main external indirect drivers

  • Coastal development and Deforestation (local to regional, well established): Land use changes and practices upstream result in soil erosion that carries sediments downstream to the oceans, affecting water quality and turbidity. Coastal development leads to an additional input of nutrients and organic matter from sewage (Borum et al. 2004). Seagrasses are negatively affected both by the decrease in light penetration and by the increase in nutrients caused by this development, and can often result in a regime shift from seagrass to algae dominated regimes (Duarte 2002; Borum et al. 2004).
  • Climate change (global, well established – contested): Climate change may have severe effects on seagrass beds and some aspects of it can result in more favourable conditions for algae than for seagrasses. The aspects that have been shown to affect regime shifts in seagrasses are sea level rise, increased CO2 and temperature rise (Short and Neckles 1999, Duarte 2002). But the connection to actual regime shifts remains speculative (Short and Neckles 1999); climate change is correlated with other anthropogenic activities in marine areas (Borum et al. 2004) and most effects will vary spatially and depend on species in question. Temperature rise will affect photosynthesis in both algae and seagrasses and will depend on species thermal preference, but it has been shown that epiphytes growing on eelgrass will be favoured by higher sea temperatures (Short and Neckles 1999). Sea level rise will result in seagrasses losing their habitat and being forced to move in order to regain the light conditions needed and in addition the sea level rise will cause erosion that further will decrease light conditions for seagrasses (Short and Neckles 1999; Borum et al. 2004). An increase in CO2 levels might lead to an advantage for seagrass over algae since they are more CO2 limited, but this is contested since evidence is weak (Borum et al. 2004). Most significantly, climate change will increase the risk for more extreme weather with more frequent and bigger storms which will cause sediment resuspension decreasing light conditions together with physical disturbance (Short and Neckles 1999; Duarte 2002). It has been speculated that this in synergy with other anthropogenic and natural stressors can cause a decline in seagrasses (Short and Neckles 1999) and it is plausible to assume that this could cause a future regime shift. Further research is needed in order to establish what effects climate change will have on seagrass bed (Short and Neckles 1999).

Slow internal system change

  • Loss of connectivity (local to regional, contested): Connectivity between seagrass beds is important since the loss of extensive beds can cause a decline in other species and functional groups that migrate between patches of seagrasses i.e. grazers. Loss of these groups will cause a shift from a seagrass dominated regime to an algae dominated one, since loss of grazers will release algae from grazing pressure. The loss of seagrass patches can potentially negatively affect other nearby patches, due to the resulting increase in water turbidity, since they are sensitive to decreased light conditions as well as changes in water quality due to nutrients released from resuspension of sediments (Borum et al. 2004; van der Heide et al. 2007).

 

Summary of Drivers

# Driver (Name) Type (Direct, Indirect, Internal, Shock) Scale (local, regional, global) Uncertainty (speculative, proposed, well-established)
1 Storms Shock Local-regional Well established
2 Disease Shock Local Well established
3 Nutrient loading Direct Local-regional Well established
4 Overfishing Direct Local-global Well established
5 Physical disturbance Direct Regional Well established
6 Sediment loading Direct Local-regional Well established
7 Aquaculture Direct Local Well established
8 Coastal development and deforestation Indirect Local-regional Well established
9 Climate change Indirect Global Well established - contested
10 Loss of connectivity Internal system change Local-regional Contested

Key thresholds

Shift from seagrass to algal dominance

  • Light availability. Seagrasses require unusually high levels of light for their growth. However, the specific level of light required has not been determined, as it is case-specific (Burkholder 2007; van der Heide et al. 2011). Reduction in light is related to turbidity, to shading by macroalgae and fouling by epiphytic algae (Burkholder et al. 2007). The following thresholds are all related to the amount of available light.
  • Nutrient levels. Seagrasses require oligotrophic conditions. However, it is difficult to determine a critical threshold at which concentrations are detrimental for seagrasses, as it depends on other factors, such as current velocity and herbivory (Valentine and Duffy 2006; Burkholder et al. 2007). This threshold is related to light availability, as nutrient input can lead to an increase in turbidity due to eutrophication, to shading by macroalgae or to fouling by epiphytic algae (Orth et al. 2006; Burkholder et al. 2007).
  • Herbivory levels.  It is difficult to identify the threshold, but it can be reduced to the same threshold associated with light conditions, as the lack of herbivores can lead to shading by macroalgae, fouling by epiphytes and eutrophication (Burkholder, 2007).

Shift from seagrass to a barren sediment state

  • Seagrass density. Seagrasses need a certain density of shoots in meadows, below which they will not be able to modify the abiotic conditions of their ecosystem to benefit their own success (van der Heide et al. 2011).

Leverage points

Seagrass beds are often hysteretic systems; once they shift into an alternative state, bringing them back to the state dominated by seagrasses can require more than re-establishing the previous environmental conditions (Duarte 2002; van der Heide et al. 2007). Therefore, it is preferable to maintain or to build the resilience of these systems to prevent a regime shift, as trying to restore them once a shift has occurred can prove difficult if not impossible (Orth et al. 2006). Maintaining a high resilience in the system will allow it to overcome drivers that cannot be easily controlled or managed locally, such as the effects of climate change (Waycott 2008).

Management plans should act at a scale that includes not only the seagrass beds themselves, but the processes and factors that affect them, such as water quality and land use in surrounding watersheds (Orth et al. 2006). Because many drivers affect seagrass meadows in synergy, the management of these ecosystems require an integrated approach (Borum et al. 2004).The following are important leverage points to prevent a loss of resilience of the system and to maintain the reinforcing feedback loops that maintain the system in the more desired state. Although the literature reviewed does not discuss the actors involved in these leverage points, it can be assumed that the agents involved in most of the management policies are policy makers, for the creation of legislation and local governments for the enforcement of these regulations. In each of the leverage points there would be specific actors, such as farmers in the case of nutrient loading, fishers in the case overfishing and tourism-related agents and coastal population in the case of physical disturbance.

Limitation of the input of nutrients and other pollutants (local to regional, well established)

Reducing the input of nutrients would favour seagrass dominance, as they thrive in oligotrophic conditions, while algae are stronger competitors in conditions of high nutrient concentrations (Duarte 2002). There are few documented examples of degraded seagrass meadows recovering following a reduction in the input of nutrients (Burkholder et al. 2007), therefore this should be a preventive measure, rather than a corrective one.

Because it is difficult to identify point-sources of nutrient loading, the possibility of assigning legal responsibility for the loss of seagrasses associated with this driver is limited (Duarte 2002). Further, as nutrients travel with the currents, regulating policies should affect not only the areas to be protected, but should act at a larger scale, ranging from regional to international (Kenworthy et al. 2006). Actions that should be implemented to reduce the input of nutrients from surrounding watersheds are the following: treatment of urban and industrial sewage (Borum et al. 2004); reduction of fertilizers in agriculture (Duarte 2002); and protection of wetlands to intercept and reduce agricultural runoff (Duarte 2002; Borum et al. 2004). Where sewage treatment is not possible, it should be disposed in areas with an efficient water exchange for the dilution of the nutrients (Borum et al. 2004). Many management actions and policies for regulating nutrient loading can also be useful for reducing other types of pollution, such as organic loading, which can also lead to seagrass decline. For instance, sewage treatment plants also remove organic waste through mechanical treatments (Borum et al. 2004).

Management of fisheries (local to regional, well established)

Controlling fishing activity is important to prevent overfishing, which can lead to a reduction in herbivores, through trophic cascades from the removal of top predators (Scheffer 2005). As herbivores control the population of algae (Duarte 1995 and 2002; Elköf 2008; Waycott 2008), maintaining the grazers in the trophic web can improve the capacity of seagrass beds to handle high levels of nutrients, enhancing their resilience to nutrient loading (Duarte 2002). Consequently, it is possible that controlling fishing activity is a more effective measure for avoiding a regime shift than reducing nutrient input (Valentine and Duffy 2006).

Due to the mobility of fish, seagrass beds can be affected by the lack of regulating policies in other geographic areas (Kenworthy et al. 2006). Hence these policies should not be limited to the local scale, but should be applied at a regional to international level, as with policies regarding nutrient loading. One way of controlling overfishing is the creation of marine protected areas. However, this can lead to socio-economic problems related to the coastal societies that depend on fishing activities. Management should take into account the social drivers e.g. poverty, food demand etc. that lead to overfishing (Eklöf 2008). It has also been proposed to reintroduce top predators as a way of restoring the food web, in order to break the reinforcing feedbacks that maintain the dominance of algae (Munkes 2005).

Limitation of physical disturbance (local, well-established)

Limiting and regulating human activities related to physical disturbances of the coastal area is important to reduce damages produced by direct removal or fragmentation of seagrass beds, and by the reduction of water transparency caused by siltation. For instance, activities such as sand reclamation, dredging, trawling and the construction of coastal infrastructures such as bridges and piers should not be carried out on seagrass meadows. In addition, these activities should be carried out with appropriate equipment to minimize siltation. Dredging and sand reclamation should be limited to short periods, as seagrasses cannot survive high turbidity levels for longer periods (Borum et al. 2004). These activities can be regulated through legislation, and for activities such as boating and anchoring they can also be limited through awareness of the value of these ecosystems (Borum et al. 2004).

Climate change mitigation (global, speculative)

Mitigating climate change would limit the increase of the sea level and storms, reducing coastal erosion, to which seagrass beds are very sensitive (Marba and Duarte 1995 in Duarte 2002). It would also reduce thermal and saline stress on seagrasses, thus reducing the loss of resilience to other stressors and shocks (Borum et al. 2004). Given the global nature of climate change, measures to mitigate it should be tackled at an international scale (Borum et al. 2004). Furthermore, the effects of this driver are not easy to control locally, therefore it is important to maintain a high resilience in the system, so as to be able to overcome the effects of climate change (Waycott 2008).

Public awareness (local to regional, well-established)

The ecosystem services provided by seagrass meadows are under-appreciated, hence their protection is not considered a priority. An increase in awareness of the ecological functions they carry out and of the services they provide to society would create a new paradigm in society, ensuring a more effective implementation of conservation policies (Duarte 2002; Orth et al. 2006; Eklöf 2008). An example of this paradigm change can be seen in tourism, which can act as a driver of degradation but which is now becoming a promoter for the protection of coastal ecosystems (Duarte 2002; Eklöf 2008).

Transplantation (local, well-established)

Once a regime shift has taken place, it is possible to resort to methods for transplanting seagrasses to encourage and speed up their recolonisation. For transplantations to be successful in recolonizing a non-vegetated area, the areas treated must be big enough so as to modify the environmental conditions and re-initiate a positive feedback that will maintain the population (van der Heide 2007). However, the success rate of these techniques is not high (Duarte 2002; Orth et al. 2006) and can cause damage to the donor populations (Duarte 2002). Furthermore, these methods are economically costly, which is an important drawback, particularly as it is expected that much of the seagrass decline will occur in developing countries (Duarte 2002).

 

In order to be able to develop effective management plans it is necessary to increase knowledge of seagrass processes and their response to different drivers.This would help develop a forecast of the future state of these ecosystems and of the future threats they will endure, and to develop indicators of decline (Duarte 2002; Orth et al. 2006).

Summary of Ecosystem Service impacts on different User Groups

References (if available)
Provisioning Services
Freshwater 0
Food Crops 0
Feed, Fuel and Fibre Crops 0
Livestock 0
Fisheries -
Yes
Yes
Yes
Yes
Terrados et al. (2004)rnDuarte (2002)rn
Wild Food & Products -
Yes
Timber 0
Woodfuel 0
Hydropower 0
Regulating Services
Air Quality Regulation 0
Climate Regulation -
Yes
Yes
Yes
Yes
Water Purification -
Yes
Yes
Soil Erosion Regulation -
Yes
Yes
Pest & Disease Regulation 0
Pollination 0
Protection against Natural Hazards -
Yes
Yes
Yes
Yes
Cultural Services
Recreation +/-
Yes
Yes
Aesthetic Values -
Yes
Yes
Cognitive & Educational ?
Spiritual & Inspirational ?

Uncertainties and unresolved issues

In order to reduce uncertainties it would be necessary to improve our knowledge on seagrass processes and on their response to natural and anthropogenic drivers. Developing a worldwide monitoring network would allow to detect declines in these ecosystems in response to stressors and to forecast the synergistic effects of present and future drivers (Duarte 2002; Orth et al. 2006). A large uncertainty regarding the regime shifts between seagrass and algae or barren sediment regimes are climate change and its potential function as a driver. It is established that there are aspects that affect seagrass beds, such as increased temperature, CO2-levels and UV-radiation as well as changes in salinity and sea level rise. But how much is dependent on species and on geographic location. In addition, the effects are changing dependent on actions taken to mediate them. More research is needed to build understanding on how climate will affect seagrasses and the potential shift between regimes (Short and Neckles 1999).