Movement Ecology

An animal’s home range represents the area it uses for routine activity over various temporal scales. Exploring an animal’s home range offers insight into species’ spatial ecology and provides critical information for conservation planning.

1) Horizontal dimension of movements

Detections of young-of-the-year blacktip reef sharks (Carcharhinus melanopterus) actively tracked with acoustic telemetry around Moorea Island, French Polynesia revealing very limited home range (i.e., minimum convex polygon MCP = 0.02 km2) and KUD 50% = 0.02 km2 and KUD 95% =0.14 km2 (Bouyoucos et al. 2020).

Changes in space use (kernel utilization distributions) during incoming and outgoing tides by 33 acoustically tagged grey reef sharks. Circles represent the location of acoustic receivers (Papastamatiou et al. 2021)

While sharks can spend a lot of time in restricted areas, they are able to migrate over large scales along the coasts or open waters.

Large scale annual migrations of Port-Jackson sharks (Heterodontus portusjacksoni) in New South Wales, Australia (Bass et al. 2017)

2) Vertical dimension of movements

Fish lives in a 3 dimensional space at sea and therefore the vertical dimension of their use of space is very important.

Vertical profiles of a sawshark in Tasmania (from Burke et al. 2020)

3) Energy landscape

Animal locomotory costs will partially depend on the habitats they move through or reside in, and will contribute to their overall fitness and energy surplus. The spatial representation of animal's cost of transport (COT, energy spent to move a unit distance) is termed the energy landscape, and may explain animal behaviour, distribution patterns, movement paths and foraging ranges. Furthermore, while energy landscapes can explain habitat selection at the local level, these effects may scale up and provide an understanding of global patterns of distribution or even animal migrations. Within marine environments, these landscapes can be dynamic as water currents will influence animal power requirements and can change rapidly over diel and tidal cycles. In channels and along slopes with strong currents, updraft zones may reduce energy expenditure of negatively buoyant fishes that are also obligate swimmers.

During incoming tides, sharks form tight groups and display shuttling behaviour (moving to the front of the group and letting the current move them to the back) to maintain themselves in these potential updraft zones. During outgoing tides, group dispersion increases, swimming depths decrease and shuttling behaviours cease. These changes are likely due to shifts in the nature and location of the updraft zones, as well as turbulence during outgoing tides. Using a biomechanical model, we estimate that routine metabolic rates for sharks may be reduced by 10%–15% when in updraft zones.

Grey reef sharks save energy using predicted updraft zones in channels and ‘surfing the slope’. Analogous to birds using wind-driven updraft zones, negatively buoyant marine animals may use current-induced updraft zones to reduce energy expenditure. Updrafts should be incorporated into dynamic energy landscapes and may partially explain the distribution, behaviour and potentially abundance of marine predators.

Predicted location of updrafts and shark habitat use during incoming tides. Colour contours are the predicted ratio of vertical and tidal velocities. The size of the circles represents the proportion of time sharks spend at that depth at South-North positions of the channel (Papastamatiou et al. 2021).

4) Movement networks


  • Using network analyses to characterise movements

Network analyses are being used with increasing regularity to describe the spatial ecology and movement patterns of aquatic animals, including sharks, obtained through acoustic telemetry (Mourier et al., 2018). Here, nodes represent locations which are connected by the movements of individuals. Network analyses can then be used to investigate the structure and dynamics of animal movements, and identify spatial hotspots and central areas in their spatial ecology.

Network analyses offer a set of quantitative metrics or test statistics that allow researchers to characterise and analyse its structure. These are used to measure structural properties at the node, group or network level. centrality metrics are also useful for movement networks to better identify the central locations most pertinent to conservation and management, or identify centres of activity (e.g. area where most movements converge).

Centres of activity of bull sharks. Maps of average node-based metrics for each receiver: (a) closeness; (b) strength; and (c) continuous residency time (CRT). (d) Monthly TOPSIS scores for each receiver which integrates the three node-based metrics mentioned above (a high TOPSIS score means that the receiver has a high value for all three metrics) (Mourier et al. 2021).

  • Aggregating individual networks that have similar structural properties

It is important to consider individual variability in movement networks to avoid the loss of important information on movement properties within the studied population. Previous research demonstrated the importance of considering individual variability of movement and space-use for understanding the drivers of socio-spatial, ecological and evolutionary dynamics of animal populations. We explore this using a reducibility procedure to simplify and condense individuals that correlate strongly in their movement topologies, while reducing the amount of redundant information in the network.

Aggregating individual movement networks can lead to the loss of information, potentially misleading our interpretation of movement patterns (Mourier et al. 2019). Reducibility procedure defining clusters of topologically similar individual movement networks for (a) grey reef sharks and (b) blacktip reef sharks. Movement networks within clusters were then combined as a layer within the multilayer network before being aggregated across layers. Nodes are coloured by degree centrality (yellow to red = low to high).

5) The multilayer nature of movement networks


Animal movement patterns are increasingly analysed as spatial networks. Currently, structures of complex movements are typically represented as a single-layer (or monoplex) network. However, aggregating individual movements, to generate population-level inferences, considerably reduces information on how individual or species variability influences spatial connectivity and thus identifying the mechanisms driving network structure remains difficult.

I recently developed an approach which proposes incorporating the recent conceptual advances in multilayer network analyses with the existing movement network approach to improve our understanding of the complex interaction between spatial and/or social drivers of animal movement patterns.

Briefly, we apply multilayer network analyses linking individual movement networks (i.e. layers) to the probabilities of social contact between individuals (i.e. interlayer edges) in order to explore the functional significance of different locations to an animal’s ecology. This approach provides a novel and holistic framework incorporating individual variability in behaviour and inter-individual interactions. This approach can be used in applied ecology and conservation to better assess the ecological significance of variable space use by mobile animals within a population. The uptake of multilayer networks will significantly broaden our understanding of long-term ecological and evolutionary processes, particularly in the context of information or disease transfer between individuals.

Left: The multi-layer nature of movement networks constituted of multiple individual movement networks from a same or different species. The movement network of a species is often represented as an aggregated spatial network of many individuals from the same species, forming a monolayer network (grey). However, it can be represented as a multilayer network with each layer representing the movement network of an individual; here a multilayer network constituted of two individual movement networks (yellow and green layers) linked by interlayer edges which can represent frequency of contact between individuals at a particular node. Right: Bivariate map of nodes centrality ranks. Each black dot corresponds to a node located on a two-dimension map according to its social and spatial centrality score. Colours are representative of the functional centrality with nodes only important in terms of space use in pink, nodes in green only important in terms of sociality and nodes in yellow representative of both combined spatial and social centrality. Below is represented the corresponding aggregated movement network of both species where node colour is representative to their functional centrality score (i.e. their colour in the bivariate map) and edges represent the movements between nodes (for clearer visibility edge size is not proportional to edge weight).(Mourier et al. 2019)