Biodiversity in the open ocean
Biodiversity Hotspots: history, definition and challenges
The concept of Biodiversity Hotspot was introduced to provide scientists and policy makers with a practical tool to face the problem of the current species extinction crisis (Koh et al. 2004; Briggs 2011). Since its introduction, the concept has received substantially coverage in the scientific literature, resulting in several slightly different definitions. The first definition was coined by Myers as “areas featuring an exceptional concentration of endemic species and experiencing an exceptional loss of habitat” (Myers 1990; Myers et al. 2000). In general Biodiversity Hotspots have been used to identify areas particularly rich in number of species, in rare, in threatened or in endemic species. Despite numerous modifications of the original definition, high species richness appears to be the basic requirement for an area to be classified as a hotspot. The drawback of this approach is that the taxonomic classification at species level is extremely difficult and time consuming, and often requires a multidisciplinary approach (morphological, genetic and molecular) and advanced technologies. Therefore rather than relying on taxonomy at species level to reflect species richness, more recent studies have investigated other measures potentially correlated to species richness, such as habitat heterogeneity, higher taxonomic levels, primary productivity, and functional diversity (Gaston 1996b; Kerr & Packer 1997; Mittelbach et al. 2001; Heino & Soininen 2007).
The assessment and application of some of these approaches are far more difficult in the marine environment than on land. The identification of pelagic hotspots in open ocean has been limited due to the difficulty of a) obtaining long-term distribution data on pelagic species; b) the dynamic character of pelagic habitats, and c) the difficulties in establish international relationships (Game et al. 2009). The ocean is a highly dynamic environment in constant change, therefore habitat based approaches are hard to apply. Additionally the remoteness nature of the ocean limits in-situ sampling to localised areas, limiting the scale of taxonomy based study.
Today, thanks to the introduction and increased availability of remote sensing information, some of these challenges have been overcome (Zacharias & Roff 2001; Louzao et al. 2011). New imagery and data sets are now enabling remote sensing, in conjunction with ecological models, to shed more light on some of the fundamental questions regarding biodiversity hotspot. For example recent studies noted that large assemblages of oceanic predators coincided with areas of frontal systems, that aggregate also plankton species (Olson & Hood 1994; Cotté et al. 2007; Kai et al. 2009; Cotté et al. 2013). In this study I proposed a new biodiversity proxy based on seascape patchiness of dominant phytoplankton types to estimate global marine biodiversity hotspot of primary producers. These hotspots seem to be related to high diversity assemblages of higher trophic level species.
What is biodiversity and why is important
The diversity of Nature has fascinated humankind since ancient times. Several Greek philosophers, among which Aristoteles, Thales, Pythagoras, Heraclitus, Parmenides, and Democritus, reflected on questions such as “Why are there so many kinds?”, “What is the relation of a kind to its individual representatives?”, “Are these kinds arranged in systematic ways?”, and “Why is there order in nature?”. Since the advent of Modern Science, attempting to describe and estimate biodiversity, along with understanding ecosystem characteristics that may determine/influence its patterns, has always been of primary concern for ecologists and conservationists. Darwin in primis observed and questioned the origins of geographical differences in the distribution of abundance and type of species (Darwin 1859). Concepts of rare and common species and high and low diversity assemblages emerged. Intuitively, assertion of inequality of species abundances and type, and changes with consequent loss of them have been present in the human thought.
However, defining and quantifying biological diversity, above all over space and time, in a way that allows an explanation for the observable natural patterns, is a challenging task with uncertain outputs. The reason for this difficulty is that biological diversity is a multifaceted concept that cannot be captured by a single parameter, and hence it has been defined and measured in different ways.
The first definition with juridical value appeared in 1992 in the Convention on Biological Diversity (http://www.cbd.int/), presented during the United Nations Conference on Environment and Development in Rio (« Earth Summit »). Here « Biological diversity » was defined as “the variability among living organisms from all sources including, inter alia, terrestrial, marine and other aquatic ecosystems and the ecological complexes of which they are part; this includes diversity within species, between species and of ecosystems”.
More widely, “biological diversity” refers to the variety of life at every level of complexity of its organization, from molecules to biomes (defined as contiguous areas with similar climatic conditions on the Earth, such as communities of plants, animals, and soil organisms) (Wilson 1992; Gaston 1996a; Mooney 2002; Hamilton 2005). In general it can be interpreted as the complexity of a system. Therefore evaluating biodiversity gives indications on the state of the ecosystem in a broad sense (Chesson 2000; Hamilton 2005).
Nowadays, the need for attention on this aspect of biological studies is even more urgent, due to the increasing loss, at an alarming rate, of diversity at all levels of the ecosystem (Myers et al. 2000; Myers & Worm 2003; Butchart et al. 2010; Mora et al. 2011; Cardinale et al. 2012; Hooper et al. 2012). Indeed, evidence is increasing that biodiversity influences productivity, ecosystem processes and services and its stability and resilience, underlining the importance of characterization and preservation of diversity-rich environments (McGrady-Steed et al. 1997; Peterson et al. 1999; Tilman 2000; Cardinale et al. 2006; Worm et al. 2006; Cadotte et al. 2011; Flynn et al. 2011).
What do we know about global patterns of biodiversity in the open ocean
Robert May noted that if aliens visited our planet, one of their first questions would be, “How many distinct life forms (species) does your planet have? He also pointed out that we would be “embarrassed” by the uncertainty in our answer underlying our limited progress with this research topic thus far”(Hamilton 2005; Mora et al. 2011).
In summary, we know very little about biodiversity in the open ocean, this is primarily due to the extent and remoteness of the oceanic environment coupled with the difficulty in acquiring detailed information about the whole marine community, particularly over a spatio-temporal scale that can be useful to clarify the contribution of environmental processes in determining biodiversity patterns. Recent predictions indicate that at least 8.7 million species of eukaryotes are found on Earth, including 2.2 million marine species, of which only about 9% have been taxonomically classified (Mora et al. 2011). However, these are luckily to be only underestimations, as the more recent phylogenetic studies have revealed an overall underestimated biodiversity in most groups of oceanic species. In particular, higher diversity has been detected in planktonic species, which questions the traditional morphological approach applied to define species in this group.
Limited information is also available on the distribution of biodiversity and what influences its patterns in the open ocean (Tittensor et al. 2010). Previous studies about global patterns of ocean biodiversity found general relationships with latitude, temperature, energetic gradients (areas of sudden variation in flow direction or strength) and with environmental variability (so-called macro-ecological patterns) (Gaston 2000). The general rule of life on earth is a decreasing diversity with increasing latitude (Longhurst 2010). In several taxonomic groups a so called “latitudinal species diversity gradient” has been observed (Rombouts et al. 2009; Tittensor et al. 2010; Berke et al. 2014). This pattern was also reported for plankton (Yasuhara et al. 2012), although several exceptions exist. Indeed, the latitudinal gradient may vary once other variables such as for example longitudinal shifts, depth and topography come into play.
Temperature in particular appeared to be one of the most important variables in describing biodiversity patterns (Rutherford et al. 1999). The temperature hypothesis postulates that: “higher temperatures increased metabolic rates may promote higher rates of speciation leading to greater diversity, or that range limits are set by thermal tolerance, with more species tolerant of warm conditions” (Rohde 1992; Currie et al. 2004; Allen et al. 2007). However, the positive correlation between diversity and temperature has not been observed within all species and regions (Yasuhara et al. 2012). Indeed, at local level also the effect of temperature on biodiversity may lose value for more important physical and biological forces. In these areas, for example, high biodiversity values may be better explained by other more important ecological drivers such as turbulent features, nutrient availability and stratification. Upwelling regions are among the most classic examples; here cold waters rich in nutrients are brought to the surface layers by local physical dynamics, creating indirectly areas of high biodiversity (Barton et al. 2010).
Finally, a no-less important mechanism which may explain local variation in species diversity is dispersal, commonly defined as the active or passive movement of individuals from its birth site to a new area (Clayton et al. 2013; Levy et al. 2014). However, dispersal is a very complex mechanism which may result either in a declining diversity, through direct competition for habitat and resources, or in increasing diversity for those communities whose composition is driven by species interactions and competition (Cadotte 2006).
In summary there is no single mechanism that can adequately explain any given biodiversity pattern. Biodiversity patterns may be affected by different forces at different spatio-temporal scales, larger scale processes may affect smaller ones, and finally variability is at the base of biodiversity.
Which kind of biodiversity to measure
In the previous section I described what biodiversity is, and highlight some knowledge gaps. The first step needed in order to proceed further, is to pin the concept down. We cannot even think of superficially carve the mystery of biodiversity and even less understand accurately how much and fast are we loosing biodiversity, if we cannot measure it. Several indices, metrics, algorithms and models have been proposed as tools to study biodiversity. However, any effort to measure biodiversity has to rapidly face a common problem: what kind of biodiversity do we want to measure?
Today, the increasing number of definitions available in the scientific literature is probably the only increasing component of biodiversity. The concept of “species richness” as basic arithmetic count of the number of species found in a specific area, still remains the more simple and pure facet of biodiversity. For many years taxonomy has relied only on the use of morphological features to identify and distinguish different species. However, as molecular and microscopic techniques became widely available, it became evident that morphological features alone cannot always guarantee a correct classification at species level. In some cases for example, such as for cryptic species (species morphologically identic but reproductively isolated) morphological features alone cannot guarantee a correct taxonomic distinction. On the other hand, phylogenetic techniques are not always applicable.
To overcome this problem, biologists around the world have found different strategies to explain observed biodiversity patterns. For example in microbial studies, where the classification at species level is highly questionable, higher-taxon or trait-based classifications are commonly used as proxy of overall diversity (Green et al. 2008). The main levels of biodiversity traditionally considered are: morphological, phylogenetic, molecular and functional diversity. These levels however are not totally separated but rather one level may depend on the diversity of the other (Lankau & Strauss 2007).
Morphological diversity (assessment of diversity based on differences in physical features) is the most straightforward, it is often used to identify taxonomic classes, such as species, or to identify some functional traits used to classify functional groups. When differences or limitation exist in the methodologies applied to identify certain physical features, morphological diversity becomes a misleading method to estimate biodiversity of certain communities such as plankton. In these cases diversity must be assessed at genetic level.
Phylogenetic diversity derives from modern phylogenies that use DNA sequence data and explicit evolutionary relationships among taxa. The diversity is based on the evolutionary distance between the taxa, which is the time that each taxa has evolved independently. This distance represents a proxy for the magnitude of phenotypic differences between any two taxa (Cavender‐Bares et al. 2009). This kind of diversity has received great attention for conservation and community ecology studies. It also gives insights on the processes determining the structure and assemblage of the community (e.g. communities constrained by competition interactions are more likely to be composed by distant related taxa while communities constrained by tolerance to environmental conditions are more likely to be composed by closely related taxa). However, as it is essential to investigate the greatest possible variety of biological features, the preferred approach would be to support evidence in phylogenetic diversity with morphological distinctness (Vane-Wright et al. 1991). In some cases, such as for microbic communities, the combination of genetic and morphological analyses still does not guarantee a correct taxonomic classification. That is why for microbial communities molecular or functional diversity are most used.
Molecular diversity (definition of taxa on the basis of DNA or RNA characteristics or biochemical compounds) is the finest and most critical level of diversity because it is the origin of all diversity (Lankau & Strauss 2007). Molecular methods are an advantage when organisms are rare, cryptic or difficult to identify (Appeltans et al. 2012). New high throughput techniques allow to use metagenomics and DNA barcoding to study biodiversity of all environments (Hingamp et al. 2013). These techniques detect DNA sequence variations in particular regions of mitochondrial, chloroplast and nuclear DNA depending on the resolution required, as markers. Nuclear genes are more conserved compared to the others and have a slow evolutionary rate, therefore they are used to reveal more ancient evolutionary processes (Simon et al. 1994). The most rapidly evolving sequences are non-coding regions of DNA, used for instance for population studies. For protist and bacterial diversity, ribosomal genes are used (Zehr 2011).
Functional diversity (Cadotte et al. 2011) is based on the degree to which coexisting species vary in terms of their functional traits. It emphasizes the phenotypic difference among taxa while discounting phylogenetic relatedness, even if trait diversity is often closely associated with phylogenetic diversity. It affects species performances and ecosystem functioning (Chapin III et al. 2000; Garnier et al. 2007). Its influence is primarily expected through complementarity or enhancement of ecosystem processes caused by increased efficiency and specialisation of resource use by organisms with a high degree of trait dissimilarity. Experimental, theoretical and observational studies show that the maintenance of ecosystem processes depend primarily on the functional diversity rather than the overall diversity (Hooper et al. 2005).
1.2.5 How do we measure biodiversity: Indexes, Rank Abundance Distributions (RADs), Q-matrixes
As pointed out in the previous paragraph, identifying meaningful measures to evaluate biodiversity is essential for detecting changes that could pinpoint acting ecological processes. In this regard diversity analysis in ecology has always been a highly debated field with many different ideas on how to numerically characterize biological diversity (Magurran & McGill 2011). Multiple methods and biodiversity index have been used and introduced in ecological studies (Magurran & McGill 2011). Generally all these methods have been tested and advantages and limitations clearly described on either empirical or theoretical grounds, however recommendations of experts differ in describing which method to use. The choice of the method to be used depends on the aspect being investigated. Biodiversity can be described in terms of numbers of entities (e.g. how many genotypes, species, or ecosystems), the evenness of their distribution, the differences in some of their characteristics, often related to functional traits, and their interactions (Magurran 2004). The best approach, to have a complete understanding of an ecosystem, is to test different measures of diversity.
The main metrics used in ecology to study biodiversity can be classified depending on the amount of information we are able to detect and on the type of investigation that can suit. Following an ascending order of the amount of information on the community captured by the method we can use indices, rank abundance distributions (RADs) and ecological distances (Q-matrices).
Indices differ primarily in the importance they give to the number of categories and their abundance (Magurran 2004). The categorization can include not only species, but it can reflect guild composition, trophic structure, functional diversity, phylogenetic diversity, molecular diversity and morphological diversity (Magurran 2004). As meaningful aspects of complexity, involving interactions among individuals and between individuals and their environment, depend on population abundance, the frequencies of the different classes are considered when processes determining the structure and functioning of an ecosystem are investigated. Diversity indices establish equivalence relations among communities, depending on the aspect of compositional complexity measured. Among the diversity indices that consider both species richness and abundance the most used one are the Shannon Entropy Index (Shannon 1948), where species are weighted by the logarithm of their abundances, and the Simpson Index (Simpson 1949), used primarily to quantify the biodiversity of an habitat by taking into account the number of species present, as well as the abundance of each species (Table 1).
Table of contents :
Chapter 1 Introduction
1.1 Summary of the context
1.2 Biodiversity in the open ocean
1.2.1 Biodiversity Hotspots: history, definition and challenges
1.2.2 What is biodiversity and why is important
1.2.3 What do we know about global patterns of biodiversity in the open ocean
1.2.4 Which kind of biodiversity to measure
1.2.5 How do we measure biodiversity: Indexes, Rank Abundance Distributions (RADs), Q-matrixes
1.2.6 Introducing the concept of functional types to study plankton community ecology34
1.3 Plankton as keystone component in the functioning of the marine ecosystem
1.3.1 Overview of plankton ecology and biogeography
1.3.2 Phytoplankton biology and distribution
1.3.3 Oceanographic structures shaping plankton distribution
1.3.4 The interaction between phytoplankton and the turbulent dynamics of the ocean
1.3.5 Remote sensing a promising tool to investigate phytoplankton distribution
Chapter 2 Materials and Methods
2.1 Modelling approaches to measure global biodiversity
2.1.1 The Darwin model: coupled physical and ecological model of the global ocean
2.1.2 Computation and relationship of the ‘local’ and ‘seascape’ diversity in the model
2.2 Remote sensing approaches to measure global biodiversity
2.2.1 Remote sensing information to describe the marine environment
184.108.40.206 The Sea Surface Height
220.127.116.11 Principles to investigate transport processes: Eulerian and Lagrangian
18.104.22.168 Lagrangian coherent structures to study transport processes
22.214.171.124 Sea Surface Temperature
126.96.36.199 Ocean Colour
2.2.2 Remote sensing information about biodiversity
2.2.3 Computation of a spatial-based diversity index and its relationship to local diversity
2.2.4 Data analysis
2.3 In situ global biodiversity information
2.3.1 Atlantic Meridional Transect
188.8.131.52 The project, the sampling design and collection
184.108.40.206 Morphological diversity from inverted microscopy analysis
220.127.116.11 The model approach and data integration from global databases
18.104.22.168 Relationship of remote sensed diversity of primary producers with diversity of consumers
2.3.3 Tara ocean expedition and global high throughput information
22.214.171.124 Morphological diversity from high throughput imaging
126.96.36.199 Molecular diversity from barcoding
Chapter 3 Definition and robustness of a new biodiversity proxy. A study based on the ECCO2-Darwin circulation model.
3.2 What models and observations tell us about biodiversity and its drivers in the ocean?
3.4 How a local alpha diversity and an area-based diversity relate and why
3.4.1 Quantitative relationship between local and seascape diversity of virtual species
3.4.2 Environmental factors and accuracy of the estimation of the proxy
3.5 Discussion and conclusion
Chapter 4 Ecological relevance of remote sensing. Biodiversity hotspots estimated from space
4.3 Plankton community dominance and diversity
4.3.1 Reanalysis of chl spectra: towards an information on biogeography and diversity of planktonic communities
4.3.2 Remote sensed τ diversity and global plankton biodiversity hotspots.
4.3.3 Remote sensed temporal patterns, stability and ecological successions
4.4 Discussion and conclusion
Chapter 5 Can we derive information on higher levels of the trophic chain?
5.3 Bottom up effect of plankton seascape diversity on higher levels of the trophic chain
5.3.1 Remote sensing phytoplankton biodiversity and top predators aggregation
5.3.2 Covariance and congruence of primary production and consumers’ diversity
5.4 Discussion and conclusion
Chapter 6 Perspectives-linking remote sensing to in-situ high-throughput information of plankton community.
6.2 Tara cases study caveats and pitfalls
6.2.1 Results: Morphological diversity structured by hydrographic context
6.2.2 Results: Morphological and genetic diversities: crossvalidation
6.2.3 Results: the shape of the community by abundance distributions
6.3 Discussion and conclusion
Chapter 7 General Conclusions and perspectives
7.1 General conclusions and perspectives
7.2 Implications for management and conservation
Chapter 8 References