Department of Physics, Chemistry and Biology (IFM)

A range is a geographical area where a species can be found, and the distribution and abundance of species in that area broadly depends on the biotic and abiotic factors present in that area, such as environmental conditions (Van der Putten 2012). Several studies show that climate change is a major factor influencing these environmental conditions. For instance, rise in temperature has resulted in species range shifts (Blois et al. 2013; Parmesan et al. 1999; Parmesan and Yohe 2003). Range shifts have been observed in both terrestrial and aquatic environments (Chen et al. 2011; Hunt and Stabeno 2002; Parmesan and Yohe 2003). Range shifts can occur at various rates, magnitudes and directions depending on the species involved (Schloss, Nunez, and Lawler 2012), but generally as the climatic conditions change geographical location species are shifting their distributions to higher latitudes and/or altitudes (Root et al. 2003; Walther et al. 2002).

These changes vary across species and trophic levels, potentially resulting in altered biotic interactions, which leads to changes in the food-web structures and resulting in changes at ecosystem level (Menéndez et al. 2008). Temperature changes also influences the rate of biochemical and physiological processes in marine ecosystems (Pörtner and Farrell 2008). For example, studies in marine systems have shown that some species responded to warming (mostly prolonged periods) by reducing their body size (Jochum et al. 2012; Smith, Betancourand Brown 1995; Ohlberger et al. 2012; Naya and Cook 2017), which may result in changed populations, interactions and communities in the ecosystem (Doney et al. 2012). 

Descriptions of feeding relationships among species dates as back as late 1800s, and the quantitative as well as comparative research on potential generalities in the network structure of food webs started in the late 1970s whereas the use of food web started since 1980s (Dunne 2005).  However, early studies of food webs used food web data with artifacts of missing species and interactions (Lawton and Warren 1988; Winemiller 1990), and often poorly resolved webs with few species (20 or less) (Martinez and Havens 1993). However, since 1990s there has been a massive development in the quality, diversity and resolution of food-web data and analysis methods (Dunne 2005).

A food web is a graphical representation of an ecological network,  mapping all species and their feeding interactions in an ecosystem (Dunne 2005). It shows feeding interactions using a directed edge: an edge from node A to B indicates that energy flows from species A (prey) to species B (consumer). To represent interactions in food webs matrices are often used, as that opens up for an important set of mathematical tools as well as mathematical analyses. The matrix is made by having the columns and rows represent the species, and a 1 is assigned to row i and column j, if species j feeds on species i, and if not then a 0 is assigned to the entity (Dunne 2005). The matrix is a useful tool for performing quantitative and qualitative analysis of food webs.

With the increasing complexity of food web data, there is a need to simplify the network structures to understand and identify the interaction patterns and major pathways in the community (Sander et al. 2015). One can use community detection approach,  which provides mean of partitioning a network in communities, for example detection of modules i.e. groups of nodes (species) that are densely connected to each other but sparsely connected to other dense groups in the network (Yang, Liu, and Liu 2010). This approach plays a crucial role in providing useful information on the functionality of the complex networks since communities are also referred as  modules, which consist of groups of nodes that can for example share similar function or common properties in the network (Fortunato 2010). Krause and co-auther (2003)showed that when community detection was used in food webs, food webs tend to be in compartments than they are expected to be by chance. Such compartmentalization may play a key role in ecological communities’ resilience to perturbation (Stouffer and Bascompte 2011). Further, several other studies have shown that compartments in food webs reflect spatial differences among species, for example different habitat preferences (Allesina et al. 2005; Krause et al. 2003; Newman 2006). 

Another concept type of groups in ecological networks (food webs) is that of trophic role, where species are considered as trophic species when the species share the same predators and prey (Luczkovich et al. 2003). Here, species are considered equivalent and grouped together if they are preyed upon by the same groups of species which in turn are also preyed upon by the same group of species  (Allesina and Pascual 2009).

The group model (Allesina and Pascual 2009) is a framework that takes into accounts both characteristics of network structure at the same time to detect groups. It is a probabilistic model and also can be used to investigate similarity present among species within an ecological network (Allesina and Pascual 2009).