Latitudinal Variation in Food Webs leading to Top Predators in the Pacific
reports (PDF): FY
2009, FY 2008,
OF PROPOSED RESEARCH
(1) Compare the trophic pathways supporting large pelagic fishes from
tropical and temperate waters of the Pacific Ocean using existing data
on stomach contents of tunas and billfish from these regions
(2) Provide qualitative models of the resulting trophic pathways and
identify species and interactions that are key to the system's structure
A trophodynamics study (PFRP Project
#659559) has demonstrated significant heterogeneity in trophic pathways
across the tropical Pacific, associated with either the prevailing mesoscale
oceanography or the seabed topography. These results have the potential
to not only identify shifts in ecosystem structure (climate change),
but also may help to define stock structure of widely distributed tropical
tunas. A similar, but more localized study off eastern Australia (Young
et al. 2004) is also beginning to reveal differences not only within
the region but also in comparison with the study in the tropical Pacific
(Olson et al. 2005). Given the proximity of the two studies geographically
and the potential migration of tunas between the regions, project researchers
will conduct a study that compares the trophodynamics of the pelagic
ecosystems of these regions using largely existing data sets. The analysis
would combine statistical comparisons with qualitative models to determine
similarities or otherwise of the two regions. Such comparisons may offer
insight into the ecosystem impacts of potential climate change expressed
as ocean warming.
Researchers will compare stomach contents of top predators from tropical
and temperate waters of the western, central, and eastern Pacific Ocean
to examine latitudinal differences in the trophic pathways of these
regions. The results will be used to develop qualitative models (Dambacher
et al. 2002) of the trophic flows within each region.
Compilation of data sets
Researchers will examine three recent data sets of predator prey matrices:
one from the western tropical Pacific Ocean (10°N to 20°S), one from
temperate waters from the same region (20° to 35° S), and one from the
eastern Pacific Ocean (20° N to 20° S). Preliminary assessment of these
data sets showed existing trophic data for the majority of pelagic fish
species found in the three areas, a total of more than 100 taxa). In
particular, researchers found ~10 predator species [target species,
including yellowfin and bigeye tunas; bycatch species, including dolphin
fish, wahoo, skipjack and albacore] for which detailed trophic data
was available from all three regions. Other species for which trophic
data was available from at least two of the three regions included swordfish
and species of Carangidae (jacks).
There are also a series of global environmental data sets accessible
from CSIRO Hobart which can be accessed for the relevant temporal and
spatial scales. To establish in detail which data to use in the analyses,
a PI meeting in Hobart is proposed for September 2006. Because the eventual
qualitative models will require an understanding of lower trophic level
complexity for the different regions, researchers will also consider
available data and literature sources that will be required to fill
in these lower trophic levels.
Analysis will initially consist of a series of comparisons using classification
and regression tree (CART) analysis, to identify significant groupings
of food web structure and the main environmental correlates that can
be used to distinguish them. Classification and regression trees are
ideally suited for the analysis of complex ecological data (De'ath and
Food web data sets from the different regions will be individually analyzed
in terms of their predator-prey links. This will be done in multiple
ways, one of which will be to consider only if the species is or is
not consumed by a predator, and to ignore its relative importance to
the predator's diet. Here analysis of network structure is purely qualitative,
and link values from prey to predator equal +1 or 0. Another analysis
will consider the proportion, by wet weight, which a prey species represents
in the predator's diet, thus weighting more prevalent prey and minimizing
the effect of rarely consumed species. Alternately, a blending of the
two considers the qualitative network with rare prey items excluded.
Analysis of network structure by each of these different means will
be used to identify commonalities and differences between the food webs
of the different regions.
A novel aspect of this work is the inclusion of additional information
into the aggregation algorithm, and thus researchers will explore the
influence factors associated with life history, allometry, and habitat
(e.g. SST) on system structure. For instance, short-lived species can
be expected to respond relatively fast to environmental perturbations,
and thus their prey and predators can be expected to experience indirect
effects of a perturbation more quickly than prey and predators associated
with long-lived species. Comparing foodwebs thus aggregated from each
of the different regions can identify structural differences related
to responsiveness of the system to environmental change. Similarly,
incorporating information on ocean habitat in the analysis of network
structure can potentially reveal important patterns of association that
are spatially structured.
1 funding for this 2-year project estimated to be available mid-2006.
J.M., H.W. Li and P.A. Rossignol. 2002. Relevance of community structure
in assessing indeterminacy of ecological predictions. Ecology
G., and K.E. Fabricius. 2000. Classification and regression trees: a
powerful yet simple technique for ecological data analysis. Ecology
81 (11): 3178-3192.
R., J. Young, V. Allain, and F. Galván-Magaña. 2005. OFCCP workshop
on the application of stable isotopes in pelagic ecosystems, La Paz,
B.C.S., Mexico, 31 May-1 June 2004. GLOBEC Newsletter 11 (1): 42-44.
J. W., Hobday, A. J. , Dambacher, J. D. (2004) Determining ecological
effects of longline fishing in the Eastern Tuna and Billfish Fishery.
FRDC Research Project, 2004/063.