U N I VE R S I T Y OF C O P E N H A G E N

F AC U L T Y OF HE AL T H AN D M E DI C AL S C I E NC E S

Macroecology and Community Ecology

SGBK20002E

Four hour written exam on 25 th January 2024

The exam assignment consists of 8 pages including this front page.

© This document is protected by copyright law.

Examination instructions:
The exam set consists of 4 problems and 8 pages.

Aids:
The exam is open book, taken as an ITX exam.

Practical matters:
• Read through the entire assignment before commencing your assignment.
• The assignment consists of multiple questions each with sub-questions. If the questions weigh
differently in the total grade, this is indicated in the beginning of each question.
• Do not copy/paste text or figures into your exam answers.
• You must answer all questions and sub-questions.

1
Problem 1: Range limits – 20 %
In 1923, the Norwegian botanist Rolf Nordhagen tallied the elevational distributions of all plant
species in Sikkilsdalen in the mountain area of Jotunheimen, Norway. In 2008, it was resurveyed by
researchers at the University of Bergen. Here is a summary of their results. The labels are
abbreviated scientific names. Significance testing was done using simple t-tests.

Position of elevational range Centre:

Figure 1.1: Species optimum elevation in 1923 vs 2008. The line indicates no change and deviations from the line
indicate a change in species optimum upwards (above the line) or downwards (below the line). Species with triangle
symbols show statistically significant changes in optimum elevation.

Position of elevational range limits:

2
3
Questions:
1. Describe the patterns in Figure 1. What abiotic phenomenon can explain the
temporal change apparent from the figure, and how?
Figure 1.1 shows changes in the optimal elevation of species between 1923 and 2008.
Points above the line indicate that the species' optimal elevation has shifted upwards, while
points below the line indicate a downward shift.
Patterns: In Figure 1.1, species optimum elevations have shifted upward over time. Species
above the diagonal line have experienced significant elevation increases.

A likely abiotic explanation is climate change, particularly rising temperatures, which can
drive species to higher elevations to avoid warmer conditions.
Abiotic phenomenon: Climate warming likely drives this change, as higher temperatures force
species to shift their optimal range upward to maintain suitable temperature and microclimate
conditions.

This is further supported by the fact that the study was performed in Norway, where the effects
of climate change are expected to be significant.

2. In the table are data for the trend in the upper limit of all 91 species, and lower limits
for 25 of these. What kind of ecological processes could drive the shift detailed in each
of the four table rows? Describe the mechanisms without referring to the numbers in
the table.
Competition: Species better adapted to the changing conditions may outcompete others.

Dispersal: The ability of species to spread to new suitable areas is crucial to how they can
adapt to climate change.

Physiological limitations: The tolerance of species to temperature and other factors can limit
their ability to spread to new areas.

Upper limit increases: Driven by species' adaptation to warming, leading to upward migration
into cooler, unoccupied zones.
Upper limit decreases: Competition or unsuitable abiotic conditions at higher altitudes might
force a contraction downward.
Lower limit increases: Warming can make lower elevations unsuitable, prompting upward
shifts.
Lower limit decreases: Reduction in competition or local adaptation to warmer low-altitude
conditions might permit expansion.

3. Based on the results in the table, what processes have been most important in
driving the patterns we observed in Fig 1.1?
Climate change is the most likely driver for the observed patterns in Fig 1.1.

Most important processes are upward shifts in upper limits driven by warming, as these
would affect the elevational range centers seen in the figure

4
4. The distance between the upper and the lower limits constitutes the elevational range
of each species. If you were to redo the experiment in the tropics, would you expect
the magnitude of elevational ranges to be the same or different? Detail why.
It is expected that elevational ranges in the tropics would be different due to smaller
temperature fluctuations and more competition for space. There is likely more limited
variation in temperature across elevations in the tropics compared to non-tropical areas,
which can affect how species are distributed in altitude.

Different: Elevational ranges in the tropics are often narrower due to steeper
environmental gradients. Species in tropical montane systems are more specialized and less
capable of shifting their ranges.

5
Problem 2: Specific Leaf Area: 30 %
A group of researchers investigated the species composition of woody plant communities in a
topographically heterogeneous coastal landscape in a region with Mediterranean-type climate
(warm dry summers, mild wet winters). The landscape consisted of dry, rocky ridges alternating
with deeply cut stream ravines with moist, loamy soils at the bottoms. The sample consisted of 44
quadrats (“plots”) placed at different positions along the moisture gradient, harbouring a total of 54
tree and shrub species. A number of traits were measured for all species occurring in each quadrat.
Here we focus on a single trait: Specific Leaf Area (SLA), i.e. area of photosynthetic leaf surface
per unit leaf mass [cm2/g].

Based on the abundance of species in plots and the trait value for that species in that particular plot,
the scientists calculated two properties: 1) The mean SLA of all plant individuals in the plot
regardless of species (“Plot mean SLA”). Similarly, they calculated species’ mean trait values
across the plots, in which that particular species occurred, again weighted by the species relative
abundance in each plot (“Species SLA”). These values are plotted against each other in Fig. 2.1
(large black dots; for each plot, the mean trait value of each species in the plot is shown below in
grey).

Fig 2.1: Scatterplot of species trait values vs. plot mean trait values for log10-transformed values of Specific Leaf Area
in 44 woody plant communities. The small grey dots are trait values for each species in the individual plots, plotted
against the plot SLA (including all coexisting species in the plot). The larger black dots are species’ mean trait values.
A regression line is added for each species, showing how the SLA in each plot varies with the overall mean plot SLA.
The dashed line is X = Y.

In order to assess whether species co-existing in individual quadrats were more similar in terms of
SLA than expected under random assembly, a null model prediction was created by randomly
assigning species to communities (plots), while maintaining the species richness per plot, the
number of occurrences per species and the intraspecific distribution of both abundance and trait
values per species.

6
Fig 2.2:. Null prediction (dashed curve) and observed values (grey dots) of range (difference between largest and
smallest trait value in each community) in SLA in each of the 44 communities (plots) plotted against species richness of
the plot.

Questions
1. It is often assumed that trait variation within one species is smaller than variation
in the same trait across several species. Was that assumption justified by the
present data?
Figure 2.1 does not support the assumption that intraspecific variability in Specific
Leaf Area (SLA) is smaller than interspecific variability
The data shows that the variation in SLA within a species (intraspecific variation) can
be as large or larger than the variation in SLA between species (interspecific variation).
This is supported by Fig 2.1, where the regression lines show that a species' SLA
varies with the average SLA in a plot

Intraspecific vs. Interspecific Variation: It is important to distinguish between these two

types of variation. Intraspecific variation refers to the differences in a trait (SLA)
within a single species, while interspecific variation refers to the differences in the
same trait between different species.
Gray dots in Figure 2.1: Each gray dot represents the actual SLA value for a given
species in a specific plot. These dots show how much a species' SLA can vary
depending on the local environmental conditions in the plot.
Black dots: The black dots represent the average SLA value for a species across all plots.
These dots are useful for seeing the average difference between species, but they do
not show how much a single species' SLA can vary.
Regression Lines: The regression lines are crucial for understanding intraspecific
variation. They show how the SLA for a given species changes in relation to the
average SLA for all species in that particular plot.

7
Length of the regression lines: A long regression line indicates that a species' SLA is
very sensitive to changes in the environment and can vary considerably. A short
regression line indicates less sensitivity.
Spread of gray dots around the regression lines: This illustrates how much the SLA for
a given species varies across different plots. A large spread indicates high intraspecific
variation.
Why the data does not support the initial assumption:
Large Intraspecific Differences: Figure 2.1 shows that a single species' SLA can vary
significantly depending on the environment. The intraspecific variation, shown by the
spread of the gray dots around the regression lines, can be as large or even larger than
the differences between the average SLA values of different species. This means that a
species' SLA value is not fixed but is flexible and depends on the environment of the
plot in which the species is found.
Environmental Gradient: The landscape in the study has a moisture gradient, which
could explain the variation in SLA. SLA is an important functional trait that can be
influenced by environmental factors. Therefore, there will be variation in SLA for
species found at different locations along this gradient

The figure does not necessarily support the assumption that intraspecific variability is
smaller than interspecific variability. Instead, the length of regression lines and spread
of grey dots suggests substantial intraspecific variation across plots, which may even
rival or exceed interspecific differences. The large intraspecific variation suggests that
species have a high degree of plasticity in their SLA, allowing them to adapt to
different environmental conditions

2. Would you expect plot mean SLA value to vary with the overall environmental
gradient in the landscape? Why or why not?
Plot mean SLA will likely vary with the environmental gradient in the landscape, e.g.,
humidity. Since SLA is related to water availability and light, we can expect that dry areas
will have lower SLA than moist areas.

SLA is linked to water availability. In dry, rocky ridges, low SLA values (thicker,
drought-resistant leaves) are expected, while in moist ravines, higher SLA values (thin,
photosynthetically efficient leaves) dominate

3. Can the length of the regression lines in Fig 2.1 be taken as an indication of a niche
property of species? Why or why not?
The regression lines in Figure 2.1 should be seen as an indicator of SLA plasticity and
sensitivity to environmental change, rather than as a direct measure of niche breadth.

They indicate how much a species’ SLA responds to its environment rather than the range of
conditions that the species can tolerate.

A long regression line does not automatically mean that the species has a broad niche.
8
 While a long regression line can indicate environmental sensitivity, it does not give the full
picture of a species’ niche

in short, while the regression lines do provide valuable information about how a species' SLA
responds to environmental changes, they do not directly equate to niche breadth. The length of
the regression lines should be primarily understood as a measure of the species' SLA
plasticity and its sensitivity to environmental variation, rather than a direct measure of
niche breadth.

The length of regression lines can provide an indication of niche breadth, as it reflects
the species’ phenotypic plasticity or adaptability across environmental gradients. However,
the interpretation must account for the degree of environmental heterogeneity represented in
the dataset

4. Based on Fig 2.2., what can you infer about the dominant assembly process
in the investigated communities?
Fig. 2.2 shows that there is less variation in SLA within plots than expected
under random assembly. This suggests that environmental filtering is a dominant
process, where species with similar SLA values tend to occur together

Observed SLA ranges (grey dots) are narrower than the null prediction,
suggesting environmental filtering. This implies species with unsuitable SLA
values are excluded.

5. Propose an alternative hypothesis for a different process governing community

assembly (in the communities studied here or in different communities) and devise a
test that could discriminate between that and the pattern suggested in question 4.
An alternative hypothesis could be niche complementation, where species with different
SLA values tend to live closer together and utilize resources more efficiently.
A test could be to investigate if there is a correlation between the difference in SLA
between species in a plot and the diversity in the plot

Hypothesis: Competitive exclusion governs assembly, favoring divergence in traits.

Test: Compare observed SLA variance with a null model accounting for competitive
exclusion. Significant divergence indicates competition-driven assembly.

9
Problem 3: Richness of Neotropical palms - 30%
The tropics of the Americas exhibit the richest flora of all floristic regions. Palms (Arecaceae) is a
plant family with reasonably good data on species’ distribution across the Neotropics, which allows
for tests of what drives macroecological patterns.

Fig. 3.1: Palm species richness per 0.5° grid square (e). Panels a-d show separate richness patterns for species split into
four equally sized bins (quartiles) based on range size (i.e. panel a shows the 25% of species with the smallest ranges).
Correlation coefficients (r2) between species richness of quartiles and overall species richness are given.

(a) (b)

(e)

(c) (d)

Questions
1. Based on Figure 3.1, describe the spatial distribution of palm richness, and
how it is related to the distribution of small- and large-ranged palm species,
including any richness hotspots.
There is a hotspot for palm diversity in the northwestern part of South America.
Species with small ranges tend to be more concentrated in diversity hotspots. Fig.
3.1 shows that the areas where palms are most species-rich are also those that
contain a large number of species with small ranges

Richness peaks in equatorial regions, particularly the Amazon basin. Small-ranged

species dominate hotspots, while large-ranged species contribute more evenly
across the area

2. Formulate two hypotheses for how water-energy relationships and topographical

complexity, respectively, could each be mechanistic drivers of the distribution of
palm richness.
Water-energy: Areas with higher rainfall and evapotranspiration support a higher

10
 diversity of palms.

Water-energy: High precipitation and evapotranspiration promote productivity and

niche diversity, increasing richness.

Topographic complexity: Topographic variation can create different microhabitats

that promote higher species diversity

Topographical complexity: Diverse microclimates and niches in complex landscapes

enable high species coexistence.

3. What do the correlation results in Table 3.1 (next page) indicate about which of
the two hypotheses has more support as an explaining factor for patterns of palm
diversity? Provide your interpretation of why the results may or may not be
different for small- vs. large-ranged species.
Table 3.1 shows that precipitation and evapotranspiration have a stronger correlation
with overall palm richness, especially for species with large ranges. Topographic
complexity seems to be more relevant for species with small ranges, but the overall
picture indicates that water-energy plays a greater role in palm diversity

Water-energy hypothesis is supported, as annual precipitation and evapotranspiration

show strong correlations with overall richness.

Topography's influence is weaker, particularly for large-ranged species, possibly due

to their broader ecological tolerance.

4. The results in Table 3.1 only provide individual correlations. What could be a
statistical way to test the relative importance of each predictor?
A multivariate statistical approach, such as multiple regression, would be necessary to
test the relative importance of each factor

Conduct a multiple regression or variance partitioning analysis to assess the

contribution of each predictor while controlling for others.

11
Table 3.1. Pearson correlation (r) between species richness (all, small-ranged (1st range size quartile) and large-ranged
(4th range size quartile)) and four predictors at cell grain size 0.5°.
All species 1st quartile 4th quartile

Annual precipitation (mm) 0.75 0.19 0.74

Actual evapotranspiration (mm yr-1) 0.74 0.11 0.71

Latitude (°) -0.76 -0.08 -0.8

Altitudinal range (m) 0.01 0.25 -0.11

12
Problem 4: Large mammal extinctions - 20%
Svenning et al. studied the effect of large mammal extinctions. Mammal extinctions over the past
50,000 years are thought to have implications on current vegetation structure and ecosystem
processes in temperate Europe.

Fig. 4.1: Current and estimated present-natural diversity patterns for (A and B) megaherbivores (≥1,000 kg), (C and D)
large herbivores (45–999 kg), and (E and F) large carnivores (>21.5 kg). The term “present-natural” refers to the state
that a phenomenon would be in today in the complete absence of human influence through time. For this mapping,
omnivores were classified as carnivores when meat constitutes a major part of their diet and as herbivores otherwise.

Questions:
1. Briefly discuss how reintroduction of regionally extinct herbivores could affect current
vegetation structure with a focus on plant functional types and adaptations.
Reintroduction of regionally extinct herbivores can change the structure of the vegetation by
altering plant functional types and adaptations. For example, grazing can limit tree growth and
benefit grasses. It can also lead to changes in nutrient cycling and plant species composition
Impacts:

13
• Grazing promotes grass dominance, reducing woody plant cover.
• Browsing controls tree seedlings, maintaining open habitats.
• Functional traits like spines or tough leaves might be selected for.

2. Briefly discuss how predator reintroductions may affect herbivore impact, in current and
rewilded areas in relation to grazing/browsing impact and body size.
Reintroduction of predators can change the impact of herbivores, partly by creating a "fear of
predation," which can change the grazing and browsing behavior of herbivores.
Predators can also lead to changes in the number of herbivores, and thus indirectly change their
impact on the vegetation
Effects:
• Predators reduce overgrazing by regulating herbivore populations.
• Larger herbivores may shift grazing pressure due to predation risk (landscape of fear).
• This creates spatial heterogeneity in vegetation.

14