An Introduction to Sampling Vegetation Analysis of Prairies by gregoria


									                                                       Biology 150 Prairie Restoration--Fall 2004

                  An Introduction to Sampling: Vegetation Analysis of Prairies

To begin our study of prairies, we will undertake the critical first step in any scientific endeavor,
careful description. The goal of descriptive studies is to seek patterns in nature, which are then best
tested experimentally (although well-planned descriptive studies can also be used to test hypotheses
when experimental approaches are not feasible). As you will see, descriptive studies require a great
deal of forethought and can be highly quantitative -- patterns are most precisely described
mathematically rather than simply verbally. In our first study, we will compare a reconstructed
prairie at CERA to a remnant prairie nearby. Our data might help us address whether the
reconstructed prairie at CERA is representative of "real" prairies, or whether different management
decisions made at CERA have different effects on the biological communities there. Hopefully,
they will raise a number of questions that we can later address experimentally.

One of the central challenges in ecology is to determine what factors control the abundances of
different kinds of organisms (community structure) and the movement of energy and materials
through ecosystems. These factors include the amount of incoming energy, the availability of
nutrients, the frequency of disturbance, and/or the nature of the community's feeding relationships
(called its trophic structure). One of the first concerns we should address if we want to measure
any aspect of a community is how we deal with spatial variation. How can one know that a
measurement of a physical or biological parameter is an accurate representation of a community
when it varies within a relatively small area?

What is a sample?

A sample is an independent subset of the population that you are studying. When you do a
biological study (experimental or descriptive), you are studying a population of organisms, be they
plants, animals, or fungi. If we were to analyze the prairie vegetation at CERA, the population
would be all of the plants growing in prairies at CERA.

Why sample?

We sample for one simple reason: in most biological studies, it is logistically impossible to
enumerate the entire population of organisms. Imagine having to individually count all of the plants
in even a small plot of prairie. Or trying to count the individuals in a population of mobile animals
such as butterflies. A biologist would only be able to do a single study in his or her lifetime! Not
surprisingly, sampling plays a very important role in biological studies.

The three types of sampling

There are three basic ways to sample a population: haphazard, random, and systematic sampling.
Haphazard sampling is the simplest type of sampling: you just go into the area you are studying
and arbitrarily decide where to take your samples. It is also the least desirable way to sample for
one simple reason: your samples may be biased. For example, you may unconsciously decide to
sample plants in areas where the vegetation is healthier, biasing your results. For logistical reasons,
you sometimes have no choice but to haphazardly sample, but it is never desirable because of the
potential for bias.

                                                        Biology 150 Prairie Restoration--Fall 2004

Random sampling is the best type of sampling because it is unbiased (unlike haphazard sampling).
When you randomly choose where to take your samples from, you can be assured that your samples
are an unbiased subset of the population. So if you remember only one thing about sampling, let it
be this: sample randomly whenever possible!

Systematic sampling falls between haphazard and randomly sampling, both in ease of sampling and
desirability. In systematic sampling, you take your samples at some regular interval (say every 2 m)
across the whole population you are sampling. Researchers studying sessile organisms such as
plants or intertidal marine invertebrates commonly use systematic sampling. For these organisms,
running a transect or series of transects across the population and taking samples at fixed intervals is
simply the easiest thing to do. While systematic sampling is exceedingly easy to do, it suffers from
the same problem as haphazard sampling: possible bias. If there is periodic variation in the
population you are studying, then systematic sampling can yield a biased set of samples. For
example, if soil moisture increases and then decreases every 2 m, then systematically sampling
every 2 m could result in a biased set of samples. Fortunately, there is little evidence of periodic
spatial variation in natural systems (although temporal variation in natural systems is often
periodic). Although systematic sampling is unlikely to be biased, random sampling is generally

How many samples?

Within reason, the more samples you take, the better, as your populations estimates will be more
precise. Having said that, you should gauge the number of samples you take by how variable
whatever you are sampling is. For example, if you wanted to look at the effect of fire on
populations of prairie plants and insects, you would take more insect than plant samples. Why?
Because insect populations vary more in time and space than plant populations do. To get good
estimates of insect populations, you would simply need more samples. Unfortunately, you often do
not have good information on the natural variation in whatever you are measuring. In these cases,
you will have to do the best you can.

Today’s exercise

You will be divided into 3-person research teams, each of which will undertake different
measurements of the two prairies. Please read the descriptions of each of these, however, as you
may find other teams' results useful in interpreting your own (when you write your paper). Before
you begin, make sure you read the example of the “Planning Investigation Form” which is attached.

Sampling scheme for this exercise -- Taking a truly random sample in an area requires sampling
every position with equal probability. That is often impractical, especially for very large areas. For
this study, we will define a large rectangular study area, choose random positions along the long
edge of it as endpoints of transects that cross the short dimension of the rectangle. You’ll sample
random positions along these transects (see diagram at end of handout). Since each research team
will want to sample the area in a representative fashion, members of teams 3-8 will be spread out
across the area and actually working along with members of other research teams. After the
description of the teams below, you’ll see how this works.

                                                        Biology 150 Prairie Restoration--Fall 2004

Team 1 -- Soil conditions

Physical conditions have important effects on the performance of individual organisms and thus on
the dynamics of their populations. Temperature, of course, affects a multitude of physiological and
biochemical processes. Organisms that live partly or entirely in soil, such as vascular plants,
bacteria, fungi, soil invertebrates, and burrowing vertebrates, are fundamental parts of the
movement of energy and nutrients through the prairie ecosystems. Soil conditions are both a
reflection and a determinant of such important processes as photosynthesis (for plants), herbivory
and decomposition. Two of the most important parameters that affect, these processes are
availability of water and temperature. Note: for this first comparative study, we are sampling
prairies on different day, so comparisons of air temperature (in particular) between prairies
may not be valid. Keep this in mind when you analyze your data. For the second project,
which involves sampling an experiment all on the same day, this won’t be a problem.

       1. Make sure you have a meter tape, a meter stick, a random number table, a soil
       penetrometer, a digital thermometer (for air temperature), a soil thermometer, a soil sampler,
       small paper bags and a marking pen.

       2. Review the operation of your instruments.

       3. At each sampling point (see above procedure), take the air temperature at the soil surface
       (without pushing away all the vegetation). Then record soil temperatures at two depths by
       inserting the soil thermometer in soil at 1 cm and 4 cm. Measure soil hardness using the

       4. You will measure soil moisture and soil organic content back at the lab. Take a soil
       sample by brushing away lose litter from the surface and then inserting the sampler down
       into the soil so the mark on the side is at the surface. Pull the sampler out WITHOUT
       TWISTING IT if possible. Remove the upper 5 centimeters of the core from the sample and
       place it in a labeled ziplock plastic bag. Seal it to prevent water loss. We’ll keep these until
       you’ve obtained samples from both prairies, at which time you’ll determine soil moisture
       and organic matter content back on campus (see last page of handout).

Group 2 -- Above-ground plant biomass and necromass per area

These biological features of the environment are ecosystem-level properties, as they measure the
amount of material (and energy) tied up in different components of the living environment,
measured on a per-area basis. Biomass is currently living material, which in the prairie consists
mainly of green plant shoots. Necromass is formerly-living material that is not yet decomposed. At
this time of year, necromass consists of "litter," pieces of dead plants from last year that have fallen
onto the soil surface.

       1. Assemble a meter tape, a random number table, two 0.5 X 0.5 meter quadrat frames, a
       pair of clippers for each group member (do not lose these!), paper bags, and a marking pen.

       2. Choose 2 random sampling points along each transect. At each sampling point, place the
       quadrat frame on the soil surface and clip off all plant material at the base. Put dead

                                                       Biology 150 Prairie Restoration--Fall 2004

       material (including plant litter and animal carcasses) into one labeled bag (lab day, group
       names, prairie, and sample number) and put live material into another. Clip the material into
       small pieces for ease of handling. The samples will be oven-dried back at the CERA lab for
       48 hours.

       3. After drying, measure the mass of each full bag -- then empty the bag and get the mass of
       the bag itself.

Groups 3-9 -- Responses of individual species

Individual species respond in different ways to local soil and management conditions. Abundance
is measured as stem density, the # of stems per unit area (usually expressed as per sq. meter).
Height of a plant is a measure of plant vigor and number of flowers an estimate of reproductive
potential. Interesting species to monitor would be prairie-dependent species (i.e., those that are rare
in disturbed habitats) and invasive species (i.e., alien species that invade disturbed prairies and
displace native species). Your instructors will help you identify a species to focus upon so that we
have a representative group of grasses and forbs.

       1. Meet your two transect team partners (who each will be looking at a different plant
       species) at the head of your transect. Each group will need access to a meter tape, counters, a
       meter stick, and ruler or calipers.

       2. Choose 4 random positions along the transect using the random number table.

       3. At each position, use the meter tape to define a 2-meter diameter circle around the
       sampling point. Count the number stems of your species within that area and record it. If
       your species is too abundant to count, do a circle of 1 meter radius (but do this for ALL

       4. Measure the height of the plant closest to the sampling point (even if it’s outside the 2
       meter boundary) and, if possible, count the number of flowers or inflorescences (groups of
       flowers) on that plant.

       5. Move on to the next sampling point. If you finish you’re the points on your transect,
       move to the next one down and repeat.

scientific notebook can be found on page 17 of Investigations.

                                                       Biology 150 Prairie Restoration--Fall 2004

                  Organizing the Vegetation Survey (or Prairie Square Dance)

I have defined the area we will sample in the prairie as a rectangle that is 125 meters on the long
side and 30 meters on the short side. I have picked 12 random numbers between 1 and 125 to
define positions along the long side; these will be the end points of twelve 30-meter transects across
the area (which are conveniently numbered 1-12):

Transect        Position (meters)
            1               9
            2              13
            3              20
            4              26
            5              31
            6              45
            7              63
            8              73
            9              77
           10              91
           11             107
           12             112

Note: Research teams 1 and 2 (Soil and Biomass/Necromass) will not follow the directions
below. They will instead work with their research team members to sample 2 random point
along each odd-numbered transect.

Research Teams 3-9

Meet with your Research team to confirm your assigned plant species (make sure you agree what
they look like and what you are measuring).

Research Teams 3, 4 and 5

Split up, sending one member to Transect 1, one to Transect 5 and one to Transect 9. You now
have a 3-person Transect Team, each member of which is focusing on a different plant species.
Take data at each of four random points along the transect. If you have time, Go to transect 3, 7 or
11 and do some more sampling along that transect.

Research Teams 6, 7, 8, and 9

For Research Teams 6 and 7: split up, sending one member to Transect 2, one to Transect 6 and one
to Transect 10. For Research Team 8, split up, going to Transect 2 and 6. For Research Team 9,
split up, going to Transects 6 and 10. You now have a 3-4 member Transect Team, each member of
which is focusing on a different plant species. If you have time, go to transect 4, 8 or 12 and do
some more sampling along that transect.

                                             Biology 150 Prairie Restoration--Fall 2004

                                    Research Teams

Team 1   Will A., Mike, Sarah F.    Soil conditions

Team 2   Sarah B., Carolyn, Christina Bio/Necromass

Team 3   Will M-S, Anna, Ben        Andropogon gerardii (Big Bluestem)
                                    Sorghastrum nutans (Indian Grass)

Team 4   Emily G., Kate K., Cory    Solidago canadensis var. scabra (tall goldenrod)
                                    Solidago speciosa (Showy goldenrod)
                                    Solidago rigida (Stiff goldenrod)

Team 5   Evan, Katie R., Jenny      Dalea purperea (Purple prairie clover)
                                    Lespedeza capitata (Round-headed bush clover)
                                    Trifolium pratense (Red clover)

Team 6   Emily R., Jordan, Julia    Amorpha canescens (Lead plant)
                                    Liatris pychnostachya (Prairie Blazing Star)

Team 7   Emily S., Sam, Emma        Schizachyrium scoparium (Little Bluestem)
                                    Sporobolus heterolepis (Prairie Dropseed)
                                    Bouteloua curtipendula (Side-oats Grama)

Team 8   Meagan and Sarah S.        Ratibida pinnata (Grey-headed Coneflower)
                                    Echinacea pallida (Pale Purple Coneflower)
                                    Silphium integrifolium (Rosin weed)

Team 9   Lizzy and Becca            Daucus carota (Queen Anne’s Lace)
                                    Ambrosia artemisiifolia (Common Ragweed)


To top