Community Ecology

1
Community Ecology BCB331
Mark J Gibbons, Room Z108, BCB Department,
UWC Tel [phone]. Email mgibbons_at_uwc.ac.za
2
Individual species can influence communities in a
variety of ways
3
The role of inter-specific competition
IMPORTANT YES BUT HOW IMPORTANT
In a stable, homogeneous environment, where
species compete with each other on an ongoing
basis competitive interactions will reach
equilibrium IMPORTANT
But if other factors prevent equilibrium being
reached, competition may not be so important
4
Floder et al. (2002) Oecologia 133 395-401
Diversity higher in unstable environments
competitive exclusion prevented
5
Amongst terrestrial studies, most conducted in
temperate areas few dealt with phytophagous
insects
BEWARE people dont publish negative results,
people tend to study species that hint at
competition and journals dont publish all papers
submitted to them!
6
The case of phytophagous insects
Southern England
Lawton (1984) In Ecological Communities, Strong
et al. (Eds), Princeton, 67-100
Papua New Guinea
New Mexico
Competition rare vacant niches Implies
competition unimportant Maybe widespread amongst
herbivores
WHY?
7
Whilst competition is obviously important for
sessile species
8
Structuring power of competition
Even if competition intense, species concerned
may co-exist and competition need not determine
species composition of community
Unpredictable environments Patchy
resources Aggregated distributions
One approach to examining the role of competition
in structuring communities is to predict what
they should look like if inter-specific
competition played/plays a part in shaping them
and then to compare the results with real
communities.
9
One approach to examining the role of competition
in structuring communities is to predict what
they should look like if inter-specific
competition played/plays a part in shaping them
and then to compare the results with real
communities.
Predictions
Potential competitors that coexist should exhibit
niche differentiation
Niche differentiation by species may take form of
morphological differentiation
Within a community, potential competitors with
little differentiation would be unlikely to
coexist negative associations
Taking each of these in turn..
10
Potential competitors that coexist should exhibit
niche differentiation evidence from community
patterns
1 niche complimentarity
Anemonefish - Amphiprion
Nine species off PNG each typically associated
with just one species of anemone that is
aggressively defended
Anemones limiting resource results of
translocation experiments
Elliott Mariscal (2001) Marine Biology 138
23-36
Results show each anemonefish associated with
particular anemone AND characteristic preference
for a particular zone
11
Amphiprion percula
Amphiprion perideraion
12
This example suggests that WITHIN A GUILD niche
differentiation involves several dimensions and
species that occupy a similar position along one
dimension (anemone species) tend to differ along
another dimension (zone occupied)
13
2 niche differentiation in space
11 species of Macaranga in Borneo
14
Species also differentiated along a second niche
axis soil type (moisture and/or nutrients)
Evidence of complimentarity species with
similar light requirements differ in terms of
soils especially in the case of shade-tolerant
species
Davies et al (1998) J of Ecology 86 662-673
15
Dickie et al (2002) New Phytologist 156 527-535
16
3 Are patterns real or not? The NULL MODEL
approach
There is a temptation to interpret differences as
confirming the existence of competition BUT are
the differences big or regular enough to be
different from those expected by chance? Need to
construct a null model
A null model is like a null hypothesis it
provides a set of random data that can be used
to test observations against. A null model of a
community must retain certain characteristics of
the community under investigation but reassemble
components at random specifically excluding the
consequences of biological interactions.
17
Lizard communities in North America
Lawlor (1980) examined the dietary overlap
between lizards in ten communities and then asked
if these differed from those that would be
expected by chance alone.
HOW?
2) Calculated dietary overlap between every pair
of species in each community
3) Calculated mean dietary overlap between
species in each community
Lawlor (1980) American Naturalist 116 394-408
18
Four null models used that retained different
aspects of the food environment
Model 1
Minimal amount of initial structure retained
Only original number of species and number of
dietary items retained. Otherwise, all
electivities, including zeros, assigned a random
number between 0-1. Repeated 100 times
Individual overlap between species in a community
then calculated, as too mean overlap per community
Niche breadth and overlap increased wrt observed
19
Model 2
Original number of species and number of dietary
items retained ONLY electivities gt 0 assigned a
random number between 0-1. Repeated 100 times
Individual overlap between species in a community
then calculated, as too mean overlap per community
Niche breadth and overlap increased and different
wrt observed
20
Original number of species, number of dietary
items and electivities retained just randomly
reassigned to different diet items. Repeated 100
times
Model 3
Individual overlap between species in a community
then calculated, as too mean overlap per community
21
Model 4
Original number of species, number of dietary
items and electivities retained non-zero values
just randomly reassigned to other non-zero diet
items. Repeated 100 times
Original Data
Null Model Data
Individual overlap between species in a community
then calculated, as too mean overlap per community
This model retains the greatest amount of the
original structure in the system
22
Taken overall then there is a significant
difference between the observed patterns and
those simulated by the null models implying
that biological interactions (in this case
interspecific competition) have played a part in
structuring the communities observed
23
Niche differentiation may take form of
morphological differentiation evidence from
community patterns
Where niche differentiation results in
morphological differentiation, niche spacing
should be reflected in morphological differences
between species in a guild
In animal guilds, adjacent species tend to show
regular differences in body size or in size of
feeding structures Ratios of 2.0 for mass and
1.3 for length
Cuckoo doves 1.9 (mass) Bumblebees 1.32
(proboscis length) Weasels 1.23 (canine
length) Fossil brachiopods 1.48-1.57 (body
outline length)
24
How do you test if patterns real?
Construct a null model..
Hermoyian et al (2002) Geology 30 15-18
25
Inter-specific competition may often act through
a process of selective extinction too similar
species fail to persist together
Over the period 1860-1980, 18 pairs of introduced
con-generic passerine birds were present on the
same Hawaiian Island at the same time
Serinus species
Moulton Pimm (1986) In Community Ecology,
Diamond Case (Eds), Harper Row, 80-97
Of these, six pairs persisted together, three
pairs both went extinct, and in the other nine,
one of the two species went extinct When one of
the two species went extinct, the species pair
were morphologically more similar to each other
than when neither species went extinct 9
difference in bill length as oppose to 22
difference.
26
Within a community, potential competitors with
little differentiation would be unlikely to
coexist negative associations evidence from
community patterns
Diamond (1975) In Ecology Evolution of
Communities, Cody Diamond (Eds), Belknap,
342-444
Checkerboard distribution of two cuckoo-doves in
the Bismarck archipelago. NO island has both
species!
27
How do you test if patterns real?
Construct a null model..
Compare the pattern of species co-occurrences at
a suite of locations with that which would be
expected by chance
For example
Index of association between all pairs of native
and (separately) alien plant species found on 23
islands in Lake Manapouri (New Zealand)
constructed
dik (Oik Eik) / SDik
dik Association Index Oik Observed number of
island shared by species I and k Eik Expected
number of islands shared by species I and k SDik
Standard deviation of expected number
Wilson (1988) J of Ecology 76 1030-1042
28
The distribution of these values was then
compared to the distribution of values obtained
if the species were randomly distributed amongst
the islands (keeping number of species on each
island, and the total number of islands occupied
by a species, at the observed level)
Native Species Significantly more negative
associations than expected by chance
competitive exclusion Significantly more positive
associations than expected by chance probably
based on common microhabitats
29
The distribution of these values was then
compared to the distribution of values obtained
if the species were randomly distributed amongst
the islands (keeping number of species on each
island, and the total number of islands occupied
by a species, at the observed level)
Alien Species No significant differences between
observed and expected distribution of
associations interactions have not yet reached
equilibrium OR generally weedy and generalist
nature
30
Role of interspecific competition an appraisal
Possible and plausible explanation for some
organisation in some communities BUT not all
Why? Current competition not widely demonstrated
Ghost of Competition Past too easy to
uncritically invoke to explain patterns
Communities chosen for study may not be typical
subjective. Studies in which competition not
demonstrated may not have been published!
Patterns may have alternative explanations
Patterns may have arisen by chance!
Role of competition will vary from community to
community important in species rich, stable
vertebrate communities unimportant in
phytophagous insects
Other interactions may also play a role
31
THE END
Image acknowledgements http//www.google.com