Download Species diversity I. Ecological communities Ecological communities

8-1
Community ecology
Lab 8: Species diversity
I. Ecological communities
Ecological communities are assemblages of populations of interacting species. People
conceptually recognize communities because they are perceptually obvious. Examples of
ecological communities include forests, prairies, wetlands, estuaries, lakes, deep ocean
hydrothermal vents, and coral reefs. The essential feature of communities is that they are
assemblages of species that predictably co-occur.
The community concept became a central focus for ecologists in the early 1900s because of the
work of Clements. Clements spurred the development of a probabilistic theory of community
development when he suggested that the predictability of community development (i.e.,
succession) was proof for the existence of a super-organism. This super-organism was proposed
to be composed of the assembled organisms, much like organisms are composed of assemblages
of tissues. Though the super-organism concept of Clements has widely lost favor among
ecologists, it is true that ecological systems are more than the sum of their parts.
The structure of ecological communities is measured with a number of different metrics. One
method of analyzing ecological communities is the construction of food webs, which address the
functional relationships among the species of a community. Species diversity and species
richness are also important measures of community structure. Species richness is a measure of
the number of species per unit area. Species diversity is a non-dimensional, numerical index
generated for a given community, which takes into account both richness and abundance of
individual species. Because of the problems associated with mathematical measures of species
richness that arise in certain situations, many ecologists elect to use measures of species diversity
to describe ecological communities.
II. Sampling effort curves and diversity indices
Ecologists face two main problems when quantifying differences in the abundances of species in
communities. First, the total number of species found correlates with the sample size because
you are more likely to find a rare species as you increase your sampling frequency. This means
that diversity cannot be compared between communities that were sampled at different
intensities. Second, the number of individuals representing a species may not be a good
indication of the functional importance of that species to the community. To some degree, the
functional roles that species play in a community vary in proportion to their overall abundance.
There are cases where this is not true, however. An excellent example of a species whose
functional role is not proportional to its overall abundance in a community is a keystone
predator. A keystone predator species may be represented by only a few individuals, but it plays
a critical role in structuring the community in which it lives. The Florida panther in the
Everglades is a good example of such a keystone predator. Thus, it is best to have some measure
of the functional roles of species in a community in addition to simple measures of the numbers
of individuals that represent each species.
8-2
One part of the discovery process in assessing communities is identifying when we have exerted
a sufficient sampling effort to determine species richness and diversity with some level of
confidence. Construction of a species-area curve (Figure 1) is one approach to determining
adequate sampling effort. Species-area curves plot the area examined with repeated samplings
(x-axis) versus the total number of species found in those samplings (y-axis). Alternatively, a
sampling effort curve (Figure 2) plots the cumulative number of individuals sampled (x-axis)
against the total number of species represented by those individuals (y-axis). Both curve types
address the same question (i.e., whether species richness is increasing or has leveled off in your
sample) but may be appropriate for different situations. For example, data from rapid ecological
assessments of species richness with far-reaching conservation consequences frequently do not
include accurate measures of area covered but do include numbers of individuals sampled. In
such a scenario, a sampling effort curve would be more informative than a species-area curve.
The result for both curve types is a line that increases steeply at first but eventually levels off at
an asymptote. The point at which the species-area and sampling effort curves level off is the
point where additional sampling is yielding no additional information about the number of
species. In other words, the leveling off point or asymptote represents the optimal sample size in
terms of area or individuals, depending on the type of curve. The total number of species in a
community strongly determines how large the sample should be to reach this optimum, though
the number of rare species also plays a critical role.
Area
(m^2)
0.25
0.5
0.75
1
1.25
1.5
1.75
2
2.25
2.5
2.75
3
3.25
3.5
3.75
Cumulative
# of Species
12
14
16
16
17
18
19
20
20
20
21
21
23
23
23
Figure 1: Herbaceous Plant Species-Area Curve for the FIU Environmental Preserve (shown
with data used to generate graph).
8-3
Plant
#
4
8
11
14
17
20
23
27
29
32
36
38
40
Cumulative
# of Species
3
5
6
9
12
15
16
17
17
18
18
19
19
Figure 2: Herbaceous Plant Sampling Effort Curve for the FIU Environmental Preserve (shown
with data used to generate graph).
Once you have adequately sampled the species in a community, you can also calculate an index
to quantify the species diversity in that community. The two most common indices of species
diversity are Simpson’s index (D) and the Shannon-Wiener index (H). Higher values of D and
H represent greater diversity. Both indices are calculated from the proportions (pi) of individuals
in the total sample (Ntotal) that are represented by a given species (i), such that...
pi = ni / Ntotal
…for each species. Simpson's index (D) is calculated as…
D = 1 / Σ pi2
The Shannon-Wiener index (H) is calculated as…
H = - Σ [pi * log (pi)]
Statistical comparisons of Shannon-Wiener index (H) values in two communities can be made
using an adaptation of the t-test where…
t = (H1 – H2) / sH1-H2
8-4
In order to calculate sH1-H2, the variance (s2H) of the Shannon-Wiener index for each community
needs to be calculated. This is achieved by using the total number of individuals of all species
(Ntotal) in the community of interest along with the number of individuals of each species (ni).
Then, the variance equation is…
s2H =
Σ ni log2 ni – (Σ ni log ni)2/Ntotal
(Ntotal)2
Once s2H has been calculated for each of the two communities, then sH1-H2 can be calculated as…
sH1-H2 = √(s2H1 + s2H2)
Lastly, comparisons of H values require calculation of specialized degrees of freedom. The
formula for the specialized degrees of freedom is…
(s2H1 + s2H2)2
df = -----------------------(s2H1)2 + (s2H2)2
N1
N2
All of these calculations can be made in Excel. Try matching the example data below before
calculating values for your own data:
SITE 1
Species
A
B
C
D
E
F
Abundance
n
47
35
7
5
3
2
99
N = SUM of n
=B4/99
p
0.474747475
0.353535354
0.070707071
0.050505051
0.03030303
0.02020202
Simpson's index (D) = 1/(!(p^2)) =
Shannon-Weiner (H) = -1*(!(log(p)*p)) =
S2H1 = [!n * [log(n)]2 - (![n * log(n)]2)/N] / N2
=C4^2
p^2
0.225385165
0.124987246
0.00499949
0.00255076
0.000918274
0.000408122
0.359249056
SUM of p^2
2.783584209
=LOG(C4)
=E4*C4
=LOG(B4)
log p
(log p)*p
log n
-0.323537337 -0.153598534 1.672097858
-0.45156715 -0.159644952 1.544068044
-1.150537155 -0.081351112 0.84509804
-1.29666519 -0.065488141 0.698970004
-1.51851394 -0.046015574 0.477121255
-1.694605199 -0.034234448 0.301029996
-0.540332761
SUM of log(p)*p
=(LOG(B4))^2 =B4*H4
=B4*G4
(log n)2
n * (log n)2 n * log n
2.795911247 131.4078286
78.58859932
2.384146126 83.4451144
54.04238155
0.714190697 4.999334881
5.91568628
0.488559067 2.442795335
3.494850022
0.227644692 0.682934075
1.431363764
0.090619058 0.181238117
0.602059991
223.1592454
144.0749409
209.6726122
2
!n * [log(n)] ![n * log(n)]
![n * log(n)] 2 / N
0.540332761
0.001376047
SITE 2
Species
A
B
C
D
E
F
Simpson's index (D) = 1/(!(p^2)) =
Shannon-Weiner (H) = -1*(!(log(p)*p)) =
S2H2 = [!n * [log(n)]2 - (![n * log(n)]2)/N] / N2
Abundance =n/N
n
p
48 0.457142857
23 0.219047619
11 0.104761905
13 0.123809524
8 0.076190476
2 0.019047619
105
SUM of N
=C4^2
p^2
0.208979592
0.047981859
0.010975057
0.015328798
0.005804989
0.000362812
0.289433107
SUM of p^2
3.455029771
=LOG(C4)
=E4*C4
=LOG(B4)
=(LOG(B4))^2
log p
(log p)*p
log n
(log n)2
-0.339948062 -0.155404828 1.681241237 2.826572098
-0.659461463 -0.144453463 1.361727836 1.854302699
-0.979796614
-0.10264536 1.041392685 1.084498725
-0.907245947 -0.112325689 1.113943352 1.240869792
-1.118099312 -0.085188519 0.903089987 0.815571525
-1.720159303 -0.032764939 0.301029996 0.090619058
-0.632782798
SUM of log(p)*p
=B4*H4
=B4*G4
n * (log n)2 n * log n
135.6754607
80.69957939
42.64896209
31.31974023
11.92948597
11.45531954
16.1313073
14.48126358
6.524572197
7.224719896
0.181238117
0.602059991
213.0910264
145.7826826
202.4056243
!n * [log(n)]2 ![n * log(n)]
![n * log(n)]2 / N
0.632782798
0.000969197
SH1-H2= "(S2H1+S2H2) =
t = (H1 - H2)/SH1-H2=
1 (s2H1 + s2H2)2 =
2 (s2H1)2/N =
3 (s2H2)2/N =
df = 1 / (2+3)
p value
0.048427721
-1.909031354
5.50017E-06
1.91263E-08
8.94613E-09
195.9277649
0.057593241
0.05<p<0.10
fail to reject H0
The diversities are not significantly different.
8-5
III. Objectives
You will use the quadrat sampling technique to quantify plant species diversity in two campus
communities selected by your TA. From the data taken by your entire lab, you will complete the
following objectives:
1) Generate a species-area curve for each community to determine whether or not your lab
samples approach the optimal size areas.
2) Calculate both D and H for each community using the combined class data.
3) Compare the Shannon-Wiener index (H) values for both communities using the
specialized t-test.
IV. Instructions
Before setting out to measure species diversity in the selected campus communities, complete the
following pre-field instructions:
1) Generate several hypotheses as a class that you can test with the data generated in
today’s lab.
2) Set up field data sheets for today’s protocol. Remember the importance of noting
total numbers of individuals for each species in each quadrat and, thus, the
importance of keeping a record of all plant species you find.
3) Divide into groups and work as teams in the field. Work should be divided up so that
all team members get to experience each aspect of the exercise.
4) Be sure that you have all the field sampling equipment you need.
5) All field teams should participate in sampling as many quadrats as possible. After
sampling for this exercise, your TA will pool the data from all groups to generate a
larger data set for each community that you sampled.
8-6
Field Instructions:
Since each group will be contributing information to your entire section’s data set, it is
absolutely critical that everyone in the lab gives each species the same label or name. Therefore,
before any sampling can begin, you need to develop a plant identification key for your study
area. The best way to do this is to select a voucher specimen for each species you find and
identify it — either by genus and species, if possible, or by some other identifier (e.g., A, B, C,
etc. or Amy, Bob, Chuck, etc.). Then, be sure that every group has access to your collection of
voucher specimens so that individuals are counted as the correct species. This is VERY
important. Take as much time as necessary to generate this collection of voucher specimens, and
be sure everyone in your lab agrees on the identification labels you will use for each species.
You will be using randomly placed 0.25 m2 quadrats (50 cm X 50 cm) to sample each
community. Spend as much time as possible in each community to generate a robust data set.
Each time you randomly locate your quadrat, first identify all species present in the quadrat using
your voucher specimens. Then, count and record the number of individuals of each species
present. If your group identifies a plant that is NOT in your lab voucher collection, it may be a
rare species. It is VERY important that you add this plant to the voucher collection, and be sure
that everyone in the lab knows that you have done this. Consult with your TA before picking
any rare plants. Your TA will condense all group data sets into a single lab section data set for
analysis.
Further Reading
Crist, T. O., J. A. Veech, J. C. Gering, K. S. Summerville. 2003. Partitioning species
diversity across landscapes and regions: A hierarchical analysis of alpha, beta, and
gamma diversity. American Naturalist 162(6):734-743.
Pineda, E. and G. Halffter. 2004. Species diversity and habitat fragmentation: frogs in a
tropical montane landscape in Mexico. Biological Conservation 117(5):499-508.
Swamy, P. S., S. M. Sundarapandian, P. Chandrasekar, S. Chandrasekaran. 2000.
Plant species diversity and tree population structure of a humid tropical forest in Tamil
Nadu, India. Biodiversity and Conservation 9(12):1643-1669.