MOOC Innovative methods to assess the distribution of marine ecosystems

MODULE 5
Indirect approaches to characterise the distribution of
ecosystems.

5.1 Introduction to Module 5

In this Module, we will drive participants through some of the most common indirect
methods to characterize the distribution of ecosystems.
Three different approaches will be presented.
Firstly, we will study the basis of species distribution models, and we will review the most
common approaches applied to explain and predict the distribution of marine organisms.
Secondly, we present some innovative tools to classify the marine environment, based
on the biophysical characteristics of estuarine and coastal habitats.
And finally, we analyse the key elements to be considered in the design of field and
laboratory experiments, through the presentation of some case studies.

1
 MOOC Innovative methods to assess the distribution of marine ecosystems

5.2 Species distribution modelling: Ecological niche concept.

Even as saltmarshes are shrinking, our understanding of saltmarsh dynamics and

processes is growing, giving saltmarsh managers important insights about how to protect
and conserve this important and beautiful ecosystem.
These models generate hypotheses about what are the mechanisms controlling the
spatial patterns of species distributions based on the relationship between species
occurrence data and environmental predictors.
The theoretical framework for this approach is the ecological niche.
The niche concept has its roots in the Darwinian view of ecosystems. As a rough
preliminary definition, let me say that the niche describes how a species interacts within
an ecosystem. The niche of a species depends on both biotic and abiotic factors, which
affect the ability of the species to survive and endure.
For example, a macroalgae species needs certain conditions of sea surface temperature
and wave height. It also requires radiation, so this macroalgae species is limited to a
certain depth. In addition, this species requires rocky substrate to develop. Finally, the
substrate has to be available, without the competition of other macroalgae. So, there is
an area where all these conditions converge (the red dash area), and this is the
ecological niche of the macroalgae species.
One of the first ecologist who developed this idea of ecological niche was Grinell, who in
1917 referred to it as, the climatic and habitat requirements of a species, that is the
environmental factors, expressed geographically.
Several years later, Hutchinson provided a conceptualisation of the Grinnellian niche, as
a hypervolume of environmental states that allows a given species to establish,
reproduce and survive in a certain geographical region or biotic community.
According to this definition of ecological niche, the conditions where a species is able to
live is often broader than where the specie actually lives. Why? Because of the biotic
interactions. Let’s explain this concept with the BAM (biotic, abiotic and movements)
diagram.
Here you can observe the circle A (the blue one) that represents the region in the
geographic space where the suitable abiotic conditions occur; the circle B (the yellow
one) that represents the region where biotic conditions would allow existence of the
specie, and the circle M (the orange one) is the region that has been accessible to
dispersal or colonization by the specie
In this context, Hutchinson also introduced the concept of the fundamental niche as
comprising those environmental abiotic conditions that allow a species to maintain
sustainable population growth, that corresponds to the circle B, and the realised niche
as the part of the fundamental niche that can be occupied when competition or other
forms of biotic interactions are included, which is the intersection area between the three
circles.
Going back to species distribution models, they usually can be interpreted as describing
a species’ Grinnellian niche, which may not define the complete n‐dimensional
Hutchinson’s conceptualisation, but determine an approximate set of conditions in which

2
 MOOC Innovative methods to assess the distribution of marine ecosystems

species can occur. This is considered the best alternative for modelling distributions
between species and the environment.
Then, when we build models to explain and predict the distribution of organisms, we only
consider abiotic conditions, but we have to take into account that the species presence
also depend on dispersal limitation (such as barriers to migration) and biotic interactions
(such as competition or predation).

3
 MOOC Innovative methods to assess the distribution of marine ecosystems

5.3 Types of Species Distribution Models (SDMs).

Nowadays, there are many algorithms with which to build species models. The simplest
way of classification refers to the type of distribution data needed as input, presence‐
absence or the intermediate type, presence-background algorithms. This last one implies
that background points are generated as pseudo-absences, usually in a random way.
Here we are going to explain in more detail five different types of algorithms.
Let’s start with the climatic envelopes approach. As we already known, species
distribution models are based in the concept of ecological niche, and this is one of the
simplest approaches to modelling niches, based on presence-only method.
Here dots represent occurrence records and the axis represent environmental
dimensions. The axes x could be the sea surface temperature and the axes y the salinity.
The model defines shapes in a multidimensional space that enclose environments
associated with known occurrences localities. An example is BIOCLIM model.
When both presence and absences are available, the regression‐based approaches are
the most commonly used in ecology and particularly in habitat suitability modelling.
Regression fit a curve through a set of points using some goodness‐of‐fit criterion. The
most well‐known examples are GLMs, GAMs and MARS. GLMs (represented by the
yellow line) relates the linear models with the response variable via a link function. GAMs
(green line) automatically fit response curves as closely as possible to the data given the
permitted level of smoothing. Finally, MARS (blue line) fit linear segments to the data.
The classification method is also applied when presences and absences are available.
Classification trees are based on classifying the observations into homogeneous groups,
divided on the basis of each of the explanatory variables. This way, the environmental
variables (for example SST and salinity) are divided at split points (p1, p2). Then, the
decision tree classifies suitability as positive (1) or negative (0) along each branch. Some
examples are RF, BRT or CART.
Another method are the neural networks.
An example of artificial neural network could be composed of neurons (blue circles)
connecting by weights (yellow lines). A machine learning algorithm is used to learn a
response based on the input environmental variables, that could be for example SST,
salinity, radiation, wave height and currents.
The last one is a presence-background method: the maximum entropy.
This states that the probability distribution that best represents the data is that with the
greatest entropy. An example is Maxent, one of the most used models in the last decade
that we are going to apply in the practical exercise.

4
 MOOC Innovative methods to assess the distribution of marine ecosystems

5.4 SDM applications.

We already know what species distribution models are, but what are their applications?
In ecology their applications are broad, in this lesson we are going to show some
examples.
The first one is cartography of species or habitats. Through SDMs we can develop
automatic probabilistic distribution maps, what is especially useful for large areas. Such
as a cartography of a macroalgae species distribution worldwide.
Another application is the understanding of the ecological requirements of the species.
As a result of the modelling, the species response curves are obtained, which allows to
better understand the ecology of species providing their tolerance limits for each
environmental predictor. From this response curve we infer that the macroalgae optimum
sea surface temperature is 15ºC.
With this information, it is possible to analyse the distribution of species, biogeography
and dispersive barriers, what will guide field sampling to the discovery of new
populations.
Another application is the detection of biodiversity hotspots, that allow to establish priority
areas for conservation measures.
SDMs are also useful to detect potential distribution of invasive species.
This figure is part of a thesis that aimed to analyze the spatial and temporal distribution
of estuarine vegetation at a local scale, integrating the study of interactions between
native and invasive species. Here it is map the present and future distribution of the
invasive shrub Baccharis halimifolia. We can also evaluate different management
scenarios. Here is analysed a restoration scenario where we consider to open a drainage
dyke. The SDM helps us to be awarded about the future vegetation distribution if we
open one or two gates.
SDM can be used to hindcast paleodistributions of species and ecological communities
for time periods and locations of prehistoric human occupation.
Lastly, one of the most common applications in the last years is to assess the potential
impacts of climate change on species.
In the scientific literature there are many examples of SDMs that project potential future
changes in the geographic ranges of species. Here you can observe the predicted
distribution of a seagrass species (Zostera marina) for mid and long term and two
scenarios of climate change (RCP4.5 and RCP8.5). Yellow colours indicate the seagrass
will maintain stable, orange colours that it will increase and red colours means it will be
loosed, according to the projections.
So, as you have seen SDMs are powerful tools for many ecological questions.

5
 MOOC Innovative methods to assess the distribution of marine ecosystems

5.5 Distribution modelling of macroalgae populations.

We know that the global climate is changing, as more frequent heat waves or sea level
rise, but how does this affect algae? This is the question that Camino Fernandez de la
Hoz asked herself in her doctoral thesis.
The changes in macroalgae distribution affect our daily lives more than we can imagine.
Macroalgae fields are important for many reasons, they are a refuge for many marine
animals, they are used by humans for food or medicines and they also protect our coast
by reducing the energy of waves. Therefore, to understand how climate change will affect
them is crucial.
Although a doctoral thesis is something highly complex, we are going to try to explain
you in a simple way how Camino achieved this goal. Then, these variables were
projected to 2100 in different climate change scenarios through the use of climatic
models. The distribution of different macroalgae species was also compiled.
In this point we introduce the SDM algorithms. The extrapolation to climate change
scenarios implies a high level of uncertainty, to reduce it a stepwise methodology was
developed to select the most transferable algorithms in time in the marine environment.
The SDM algorithms related the distribution of algae in the last 30 years with the data of
environmental variables in that period. Once this relationship was known, the same
algorithms were applied to the projected data to 2100 to know the distribution of
macroalgae species in a climate change scenario.
So, what we obtained from the SDM algorithms application?
The main results obtained showed the potential distribution of the species for the
historical period and for the projected period (2100). For example, for Saccorhiza
polyschides and Gelidium spinosum a remarkable decrease in habitat suitability was
predicted in the whole distribution area. Sargassum muticum southern areas reduced
their suitability, until disappear in the Iberian Peninsula; meanwhile probability increase
in northern areas. Pelvetia canaliculata probabilities of occurrence in the west of Britain
might decreased slightly, whereas the suitability of some areas around the Brittany and
the English Channel increased. The suitable habitat for Cystoseira baccata was
predicted to increase in the climate change scenario around UK and the Gulf of Biscay
This work helps to fill the gap in knowledge between macroalgae ecology and
environmental drivers on the one hand and between science and managers on the other.
With this information it is possible to anticipate changes, for example, to decide which
areas should be protected because, although they are not threatened today, they will be
in the future.

6
 MOOC Innovative methods to assess the distribution of marine ecosystems

5.6 Approaches to marine classification based on biophysical

characteristics.

In this video we are going to focus on marine classifications. Biophysical classification

consists on establish similar areas according to their physical, chemical and biological
characteristics, providing a comprehensive framework for organizing information about
coasts and oceans and their living systems.
So, why is important to classify the marine environment? It is important from both
conservation and planning perspectives. For example, marine classification is a useful
tool in the case we want to compare and predict the response of different ecosystems
and organisms to human activities across regions. It may also be useful for the
development of conservation strategies to preserve species in degraded or fragmented
areas, as well as in shifting habitats due to climate change.
The information used to classify the marine environment could be physical, chemical or
biological and there are multiple classification systems that applied different combination
of indicators and procedures to integrate such information.
Given the influence of physical and chemical factors on the distribution of species, it
could be advantageous to use these variables in classifications, especially in global
scales. One of the advantage of these factors is the possibility of a continuous data
acquisition against the lack of homogeneous reliable biological information all around a
large area. So, more easily measured physical or chemical variables could be used as
surrogates of biological patterns.
Physical and chemical classifications should be biologically validated. The criteria should
be an objective statistical demonstration, which proves that the derived classification
units are significantly similar or different, based on both environmental and biological
characteristics.
In this context, different methods have been applied to classify coastal waters at regional
and larger scales all over the world.
The first approach for a global division was carried out through the Large Marine
Ecosystem project (LME), defined in 1984. This concept followed years of discussion,
deliberation, and development by oceanographers, marine ecologists, geographers,
economists, fisheries scientists, and marine policy makers from around the world. They
grappled with how best to understand the variability of large ecosystems and how to
manage the oceans’ living resources for sustained productivity.
Finally, marine systems were worldwide divided, based on four linked ecological criteria:
bathymetry, hydrography, productivity and trophic relationships. At the end, 64 major
ecosystems were distinguished around the world.
If we focus on a region as the NE Atlantic coast, several classification systems have
been developed.
The first one is CORINE. It consists of an inventory of land cover in 44 classes. This
classification is produced by the majority of countries by visual interpretation of high-
resolution satellite imagery provided by Copernicus. The first version was referenced in
1990, with several updates, the last one in 2018.

7
 MOOC Innovative methods to assess the distribution of marine ecosystems

This is one of the most used classifications in Europe for several purposes: the
establishment of habitat protection legislation, the inventory of habitats in a
biogeographic region, country or site, the monitoring of biodiversity and the reporting or
description of species’ habitat requirements.
There are classifications developed in Conventions, as the OSPAR regions, established
in the Convention for the Protection of the Marine Environment of the North-East Atlantic.
At national scale there are numerous schemes available for the classification of coastal
waters. These classifications greatly vary depending on the region where they were
developed, on the physical and biological heterogeneity and on the availability of data.
The variables commonly used in the most relevant classifications are sea surface
temperature, wave exposure, currents, substrate composition and topography.
So, there is a wide range of biophysical classifications. We should select the one that fits
better with our spatial domains and objectives.

8
 MOOC Innovative methods to assess the distribution of marine ecosystems

5.7 Classification of NE Atlantic coasts and New Zealand estuaries.

For the development of this session, we have selected two examples that use different
approaches to classify different marine environments. We are going to describe in detail
both methodologies.
First I will describe the classification system of the Atlantic European coast proposed by
Ramos et al in 2012. This methodology was specifically developed to classify the rocky
intertidal coast from the viewpoint of macroalgae.
Let’s start by talking about variables. This classification system uses two variables, sea
temperature and light radiance, to divide the coast in broad geographic regions, called
biotypes by the authors. Besides, few additional variables are considered to recognize
some variability within biotypes. These additional variables are salinity, tidal range and
significant wave height.
Once the variables of interest have been established, what is the next step to follow? Of
course, characterizing the study area according to such variables and establishing
groups based on similarities. In this case, all European coast was characterized and
based on that information, it was classified in different typologies using a statistical
grouping procedure. As a result, 5 biotypes and 16 subtypological variants within them
were delimited. In the maps, you can see how they are spatially distributed along Europe.
The second classification system I want to talk about is the estuary classification system
that was created to classify New Zealand estuaries. This methodology was developed
by Hume et al. in 2007.
According to this methodology, a full classification for an individual estuary describes its
hydrodynamic processes, its catchment geology and land cover characteristics. In this
lesson, we will focus on describing the hydrodynamic classification. Here, you can see
the assignment procedure, in which five metrics that take into account water inputs and
morphology are used for assigning estuaries to eight different classes.
These classes cover many types of estuaries present in different parts of the world, from
lagoons to bays and fjords. The picture represents the distribution of all estuary classes
along New Zealand.

9
 MOOC Innovative methods to assess the distribution of marine ecosystems

5.8 Characterizing distributions: field and laboratory experiments.

In module 2 we introduced the concept of zonation and discussed some of the factors
that determine it. You may recall that we described how species can be distributed in
bands or groupings in the environment because of the strong gradients that exist, using
the intertidal zone as an example. You may also recall that we discussed how physical
factors, like desiccation and heat, determine the upper edge of an intertidal organism’s
range, while its lower limit is usually determined by interactions with other organisms
such as competition and predation.
Can you remember who demonstrated these ecological concepts and how?
In the 1960s Connell manipulated the presence of two species of barnacles on intertidal
rocky shores in Scotland and showed that the lower limit of the high intertidal species
Chthamalus stellatus was set by competition with the mid zone species Semibalanus
balanoides.
At about the same time Joseph Connell carried out his experiment in Scotland, another
biologist,Robert Paine, conducted an experiment in Washington by removing the
predatory seastar Pisaster ochraceus from an area of the intertidal shore. This
experiment showed that Pisaster was responsible for controlling the spread of mussels,
which would dominate the intertidal shore in the absence of Pisaster. These pioneering
investigations provided a framework for the rise of experimental field studies in recent
decades.
So what exactly is an experiment, and how can we design and carry out effective
experiments?
An experiment can be defined as the formal test of a hypothesis. A hypothesis is a
specific prediction, which describes, in concrete terms, what you expect will happen
under specific circumstances. A useful experimental hypothesis is formulated as a
precise and testable proposition: “if what I propose is true, then making this particular
change to the system would lead me to observe this specific response …”
Usually a hypothesis is based on some previous observation or knowledge by us or by
others, such as that found in the scientific literature. For example, Paine had observed
that at mid-intertidal rocky shores in the Washington coast, species diversity was low, as
mussels and adult barnacles formed a conspicuous band. At lower tidal levels, species
diversity increased abruptly, the rock being occupied by anemones, immature barnacles,
chitons, limpets, algae and sponges. Paine knew from the literature that the starfish
Pisaster ochraceus was likely to limit the distribution of mussels and barnacles. He
hypothesised that “if species richness at low shore levels results from the presence of
Pisaster, than removing Pisaster from those areas would lead to a decline in species
richness“.
Manipulative experiments use active interventions to produce some kind of measurable
response in the system. An experimenter deliberately changes or manipulates an
independent variable, in order to cause a change in a dependent response variable.
Manipulations are made to establish cause-effect relationships between independent
and dependent variables.

10
 MOOC Innovative methods to assess the distribution of marine ecosystems

In Paine’s experiment the independent variable was the presence of Pisaster (present
or absent), while the dependent response variable was the species richness in the
community.
Back at Paine’s study site, the effects of the manipulation were soon observed as
changes in the study system, which is what experimenters call results. At the start of the
experiment, 15 species occupied the low intertidal rock in the area. Three months later,
where Paine had removed the Pisaster, barnacles had colonized about 70% of the
available space. One year later, the barnacles were giving way to mussels. Two years
later, the mussels had solidly colonized the low shore area, creating mussel beds
similarly to what is found at the mid shore, and only 8 species remained in the area.
Removing Pisaster had led to a drastic change in species composition and distribution,
and eventually to a decline in species diversity. Paine concluded that the starfish
removed dominant competitors and controlled species richness.
Paine’s experiment had some limitations in scale and replication, and the full
interpretation of the results and of the underlying ecological mechanisms took actually a
few more years of work and thinking, which resulted in the definition of the term
“keystone” to describe species that exert top-down influence on lower trophic levels and
prevent species at lower trophic levels from monopolizing critical resources, such as
competition for space or key producer food sources.
Paine’s experiment was done in the field, using nature as a laboratory. In a field setting,
only the variables under study are controlled by the experimenter, while the others vary
naturally. Field experiments pose challenges to their execution and can be less
reproducible by other researchers, but have great predictive power, as the experimental
conditions are reflective of more natural conditions.
Experimental tests can be done also in the laboratory. In the laboratory all variables are
fully controlled by the experimenter, which simplifies the detection of cause-effect
relationships between those few variables that are allowed to change, but the broader
interpretation and generalization of the outcomes can be complicated by the artificial
setting of the study.
Manipulative experiments are not always feasible due to scale or ethical reasons. Would
you deliberately damage an oil tanker to study the effects of oil pollution on rocky shores?
Hopefully not!
What can be done in these cases are “natural” experiments that rigorously observe the
natural changes occurring in a system as a consequence of natural or anthropogenic
processes, such as the outbreak of an invasive species or a pollution accident.
Most impact studies tend to rely at least in part on well-designed natural experiments, as
it was the case for the severe oil spills that occurred at the Amoco Cadiz and Deepwater
Horizon, which impacted well studied coastlines, allowing scientists to fully quantify the
consequence.
Quantifying the response is the final crucial step! Several responses can be measured
in the systems. Response variables should be as much as possible quantitative, meaning
they can be expressed numerically by counting or measuring. Example of variables that
are frequently quantified to characterize changes in species distribution in marine
experiments include species percentage cover, density, biomass. Response variables

11
 MOOC Innovative methods to assess the distribution of marine ecosystems

can also include qualitative properties, for example the shape or colour of an individual,
or the presence or absence of a certain species.

12
 MOOC Innovative methods to assess the distribution of marine ecosystems

5.9 Application based on field experimentation.

In this lesson, we will use an example of a field experiment carried out in a saltmarsh to
illustrate how the Scientific Method was used to characterise species distribution.
If you remember the previous examples with Joseph Connell and Robert Paine, an
experiment typically starts out with an interesting observation. This observation leads the
scientists to question the phenomenon. Then they formulate a testable hypothesis, and
design and carry out an experiment to test the hypothesis. The scientists then analyse
the results, which allows them to either accept or reject the hypothesis.
Using this new insight, they can develop a new theory. Sometimes, this leads to further
interesting questions and subsequent studies.
In our case, we observed an interesting and worrying phenomenon occurring in a few
Adriatic saltmarshes in recent years. We saw that the perennial Spartina grass species
that live in the lower marsh were gradually being replaced by the annual succulent
Salicornia veneta.
This is ecologically significant because Spartina is a perennial grass which has an
extensive root system that helps stabilise the soft soil, while Salicornia is an annual
species with a limited root system. This shift therefore, enhances soil erosion and leads
to loss of coastal protection functions of the marsh.
This observation led us to question, what could be driving this community shift. Based
on observations by other scientists reported in the scientific literature, we guessed the
shift could be due to climate change.
So we asked ourselves: Is the shift caused by climate change, specifically rising
temperatures and declining rainfall? And, will future stress from sea level rise make this
worse?
With our questions in mind, the next step was to formulate a testable hypothesis and
predict measurable effects. Our first hypothesis was: If climate change has an effect,
then higher temperatures (+2°C) and lower rainfall (-20%), will reduce the growth and
survival of Spartina spp.
We also hypothesised that, if sea level rise also has an effect, then increasing inundation
will change the responses of both Spartina and Salicornia. The direction of the effects of
this additional factor was difficult to predict because previous knowledge was lacking.
These hypotheses are testable because they can be proved or disproved with
experiments and data collection.
After we finalised the hypotheses, it was time to design and run our experiment. To do
this, we planted Spartina and Salicornia in pots in a saltmarsh, and subjected them to
different treatments where we manipulated the three independent variables,
temperature, rainfall and inundation. To simulate a 2 degree increase in temperature, we
constructed greenhouses over some of the pots. To simulate a reduction in rainfall, we
used funnels of different sizes to control the amount of rain reaching different pots.To
simulate different sea level rise scenarios, we situated the pots at different tidal levels on
the saltmarsh.

13
 MOOC Innovative methods to assess the distribution of marine ecosystems

We ran the experiment for 6 months, after which we quantified the effects of the
manipulations on the plants by measuring plant responses like plant dimensions and
biomass. If you recall, these parameters are also called dependent response variables.
With all the data collected, it was time to move to the next step: analysing the results.
We analysed the data using various statistical models. By comparing the growth of
Spartina and Salicornia in treatments with higher temperature and lower rainfall, we
could conclude that Spartina is indeed less tolerant than Salicornia to these stressful
conditions. On the other hand, Salicornia did extremely poorly in treatments with
increased inundation, indicating that Salicornia does not do well with future increased
inundation levels.
Overall, these results confirm that changing climate, in terms of higher temperatures and
lower rainfall is shifting the system to a dominance of annual succulents like Salicornia
in the region. So we were able to accept our 2 hypotheses.
We forecast that the species shift from Spartina to Salicornia, due to temperature rise
and lower rainfall, could increase the long-term vulnerability of the system to future
increasing inundation levels, because Salicornia is far less tolerant to rapid inundation
than Spartina.
These results led us to develop a new theory, which is, that heatwaves, droughts and
sea level rise caused by climate change, may combine to reduce the resilience of
saltmarshes. We wrote and published a paper about this study and the findings here
have inspired other experiments.

14
 MOOC Innovative methods to assess the distribution of marine ecosystems

5.10 Application based on laboratory experimentation.

In this lesson, we will go through an example of how laboratory experiments are carried
out. As field experiments, those developed in the laboratory follow the same scientific
method. Just as a reminder, the scientific method starts with an interesting observation
which leads the scientist to question the phenomenon, formulate a hypothesis and finally
design the experiment to test this hypothesis. Then, the scientist must analyse the results
to verify whether the hypothesis is accepted or rejected and develop new theories.
In this example, we observed how the distribution of Gelidium corneum, a valuable red
seaweed, was changing in the Bay of Biscay. Significant biomass loss have been
reported in this area, to the extent that in some places this species is replaced.
As you can imagine, these shifts have several negative implications. For example,
biodiversity losses, reduction of coastal protection and of course, consequences in
economic terms.
The latter is of particular interest in northern Spain, as this species is used for the
production of microbiological agar-agar. Three different agents are involved: the
factories, the collectors at coast and the divers who collect it directly by plucking it from
the sea bottom.
These observations lead us to question which are the main drivers of these distributional
changes. Based on the scientific literature, it seems that a combination of factors may
be explaining the phenomenon, with climate change being a special influence.
So, different questions came to our minds…. Which is the main process (or processes)
involved in the distributional shifts and loss of biomass of this species? …….. Is this
process (or processes) affected by climate change, particularly rising temperatures and
increasing solar radiation?
A literature review helped us answer the first question. According to it, the vegetative
propagation process is considered the main mechanism of recruitment and it is also
involved in the maintenance of this species populations. As you can see in the image,
this process consisted of 5 different phases starting from the rhizoidal differentiation and
finishing with the development of erect fronds. After that, the next step was to formulate
a testable hypothesis.
In fact, two different hypotheses were made:
1) On the one hand, we hypothesized a higher re-attachment capacity under typical
temperate summer conditions (close to 20ºC) and low irradiance and the inhibition of the
process under simulated summer conditions (close to 25ºC).
2) On the other hand, we expected increasing irradiance to have a significant negative
impact on the survival capacity of re-attached fragments.
Once the hypotheses were formulated, the laboratory experiments to test them were
designed and carried out.
To do this, a system with different tanks was created. Each tank was provided with a
continuous flow of seawater, it also exemplified the different conditions of temperature
and light. Three different temperatures (15, 20 and 25ºC) and 3 irradiances were tested
(5-10, 55-60 and 95-100 µmol/m2/s), resulting in 9 specific environments. Inside the
15
 MOOC Innovative methods to assess the distribution of marine ecosystems

tanks, three concrete artificial substrata with 4 apical fragments of G. corneum were
placed.
We ran the experiment for 30 days, and afterwards different seaweed responses were
measured. Such as: the re-attachment capacity (estimated the ratio of branches attached
to the substrata from the total number of branches in each concrete disc) and the survival
capacity of rhizoids (estimated by assigning 0s to those apical fragments with less than
a 10% of damaged rhizoids (alive state) and 1s to those fragments with more than a 10%
showing damages).
When all the data was collected, we proceed to one of the final steps: analysing the
results.
Overall, the highest rates of re-attachment were identified at 20ºC and low irradiances.
Lower or higher temperatures, and higher levels of irradiance decreased the capacity of
this species to attach its filaments to the substrata. Once the fragments were re-attached,
higher rates of mortality were observed at increasing irradiance levels. However, few
differences were identified among different temperature treatments for this variable.
With these results, we confirmed that the higher rates of attachment are achieved at
temperate summer conditions and low irradiance levels. Nevertheless, the process was
not totally inhibited at higher temperatures, although the number of attachment structures
were lower. On the other hand, temperature does not seem to have any influence on the
survival of attached rhizoids, being exclusively affected by increasing levels of irradiance.
To conclude, it seems evident that the distributional shifts and biomass loss registered
for this species are caused by changes in climate affecting critical processes, such as
the vegetative propagation. Taking all of this into consideration, new questions came to
our minds. For example, does invasive or opportunistic species enhance the biomass
loss and distributional shifts?

16
 MOOC Innovative methods to assess the distribution of marine ecosystems

5.11 Application based on mesocosm experimentation.

Welcome to this new lesson. In the next few minutes I will be talking about Ecohydraulics
modelling of living plants for wave and current attenuation.
I will first introduce the concept of mesocosm. A mesocosm (meso- or 'medium' and -
cosm 'world') is any experimental system that examines the natural environment under
controlled conditions, providing a link between field surveys and highly controlled
laboratory experiments.
Compared to laboratory approaches, mesocosm studies have the advantage that they
maintain a natural community under conditions closer to their natural ones, taking into
account relevant aspects from ‘the real world’.
This lesson presents a real mesocosm experience carried out under the FP 7 project
THESEUS and published by Lara et al., 2016, Losada et al., 2016 and Maza et al., 2015,
to better understand the interactions between fluid flows and marine vegetation. The
distinctive aspect of this eco-hydraulic modeling was the use of real saltmarsh
vegetation, to determine the efficiency of different plant features in the dissipation of
waves and currents energy.
The interactions between flows and vegetation are fundamental for the understanding of
aquatic environments. The measurement of fluid flow is always challenging, but when
biological elements, such as plants, are introduced into the equation, the challenge is
even greater, because while aquatic flora modifies the flow, the flows in turn also change
the shape and behavior of the vegetation.
The seaward edges of European temperate saltmarshes are typically dominated by 2
pioneer grass species: Puccinellia maritima and Spartina anglica which are effectively
the architects of saltmarsh development. Puccinellia maritima is found in the lower-mid
marsh and can be considered a flexible plant, with a high degree of bending, while
Spartina anglica is a stiff plant that occurs in pioneer-lower marsh.
To evaluate the effect of Shoot stiffness, Vegetation density and Standing biomass, on
energy dissipation, different vegetation densities were tested. The meadow of Puccinellia
maritima was thinned in two sequential steps, obtaining 66% and 33% of the original
density. While Spartina anglica was thinned only once.
The 100% densities were representative of field conditions and lower densities were
used to study the influence of this parameter in energy attenuation.
The experiments were conducted in the Cantabrian Coastal and Ocean Basin at the
Environmental Hydraulic Institute “IH Cantabria” in Santander (Spain). 28 free surface
gauges were disposed to measure water surface and 3 ADVs, placed before, in the
middle and at the end of the meadow, were used to record the flow velocity.
This video shows the disposal of boxes to create the meadow of Spartina in the central
pit of the basin.
The influence of the relative water depth (hr), which is defined as the ratio between water
depth and vegetation height, was examined using four different water depths from 0.40
m to 1m, which matches water depths during flooding events.

17
 MOOC Innovative methods to assess the distribution of marine ecosystems

Simultaneously, the influence of wave parameters were examined by testing 6 different

regular and irregular wave conditions, combined with positive and negative current
conditions. The wave conditions were selected attending to field conditions under
average storm events.
1) The main finding of the experiment was a new analytical formulation for the
assessment of wave damping under the combined effect of waves, and both following
and opposing currents.
2) The study showed that the wave damping coefficient decreases rapidly as the plant
becomes more submerged and occupies less volume of the water column, and that
higher values of density and biomass lead to higher attenuation rates, especially in the
case of stiff vegetation (i.e. Spartina anglina).

18
MOOC Innovative methods to assess the distribution of marine ecosystems

19