Geoscience Reference
In-Depth Information
number of food webs, four of which are included in the current study: the
Broadstone Stream, Afon Hirnant, Tadnoll Brook and the Celtic Sea food
webs (
Woodward et al., 2010
). How prey size scales with predator size was
explored previously by
Woodward and Hildrew (2002), Woodward and
Warren (2007) and Costa (2009)
, and how the ratio between predator mass
and prey mass scales with predator mass was examined by
Barnes and
colleagues (2010) and Nakazawa et al. (2011)
.
Costa (2009)
also measured
how minimum prey size, maximum prey size and trophic niche breadth
scaled with predator size, and
Woodward and Hildrew (2002)
studied diet
width as a function of predator size.
In this study, we take the analysis of size structure in individual-based
food webs further, to highlight the benefits of a multifaceted perspective.
This is accomplished by characterising the patterns mentioned above and
other novel measures relevant to variation in the size of prey consumed by
a predator (predator as focal entity) and the size of predators that
consume a prey (prey as focal entity), in seven individual-based food
webs, three of which are newly described. The patterns observed in the
individual-based data were compared with those observed when the data
were aggregated to different extents to characterise the effect aggregation
has on our perception of size structure. We also explored the structure of
size-class-based food webs (
Woodward et al., 2010
) and compared them to
species-based ones with an equivalent number of nodes constructed for
the same systems. We hypothesised that an approach based on species
averaging might conceal important patterns in the size structure of eco-
logical communities. Further, as a size-class-based approach can reveal
patterns that are not observed using a species-based approach, the expec-
tation was that our combined approach would provide greater insights
into size structure in natural food webs than using either in isolation.
Finally, we discuss possible implications arising from the application of
these new perspectives for the parameterisation of dynamic models of
food webs.
To summarise, the aim of this investigation was to explore the effect on our
perception of the size structure of food webs of (i) resolution and (ii) group-
ing feeding interactions by either the sizes or the taxonomy of the individuals
involved. In order to address these aims, a systematic method to quantify the
size structure of a community was needed: we adopted the approach sug-
gested by
Yvon-Durocher and colleagues (2011)
and treated size structure as
a multidimensional description of a community. That is to say, the degree to
which a food web is size structured along any particular dimension (axis) can
be assessed through examining the allometric relationship between the body
mass of individuals, species or size classes and a chosen response variable
(e.g. that between TH and species body mass using data grouped by species
(Riede et al., 2011), or that between prey body mass and predator body mass
