Laboratories 7 and 8

Inferring Climatic History and Historical Land Use From a Pollen Profile

Changes in climate are very important to our species. Many of our ancestors, all those then located in northern climates, adapted to the dramatically colder and more challenging environment of the glacial advances, when temperatures were dramatically colder and the amount of water bound in the widespread and very thick ice sheets that covered northern continents was phenomenally greater. The land here in Burlington, Vermont, was under a massive and continuous ice sheet that has left scars on local bedrock exposures as the ice moved across them. As the climate warmed and the glaciers receded, the resultant water flow dramatically altered the surface features of the state, and for a short period an arm of the sea filled the basin of Lake Champlain to some hundreds of feet above the present level. We are now faced with the possibility of even further melting of glaciers as a result of a greenhouse effect - the trapping of sun energy in the earth’s atmosphere by an increased carbon dioxide component of the atmosphere, the increase presumably due to the massive combustion of fossil fuels by our species.

Walking around the University of Vermont on a regular day, it is easy to forget how changeable climatic conditions are. However, changes in climate are possible to document by various indirect means; the common thread scientifically is to infer changes in climate from evidence of past climate trapped in various accumulated deposits that have built up over time. This is often done chemically, since different temperature and rainfall conditions lead to different chemical compositions of these sedimentary deposits. More interesting to biologists is the possibility of using fossils to document climatic change - and at least one kind of fossil is so common that Biology 2 students can use them to infer climatic change in the Vermont environment without fear of using up the fossils.

These fossils do not seem much like fossils at first - since we are accustomed to thinking of bones and petrified logs when we think of fossils. In this case, we are talking about the pollen grains of plants, which are blown about by the wind or carried by animals in the process of reproduction. They make great fossils, because the outer pollen coat–a dense, resistant layer that surrounds the living part of the pollen–is made of an inert protein called sporopollenin, one of the most resistant substances ever invented by living things. In addition to their durability, pollen grains have another important attribute - they are incredibly diverse in structure. In fact, there is so much diversity that often a species or at least a group of species in a genus can be recognized based solely on the shape and surface features of the pollen grains.

The diversity of pollen grains is in the structure of the pollen coats, which can be seen in most decent microscopes. There are three fundamental dimensions to pollen grain diversity: the structure of the wall, the sculpturing of the wall, and the furrows and pores that are impressed into the wall. In addition, in a few groups, the pollen grains are dispersed and remain in groups, which are very distinctive. Finally, the size of the grain is important: measuring size will be difficult in our work, we will have to work with relative size.

Pollen grains, in particular wind-dispersed pollen grains, are produced in incredibly large numbers. If you are familiar with the lakes of New England you have some idea of the numbers involved. Each year, in the middle of June, the white pine, the largest and commonest of the pines in our part of the world, releases its pollen as part of its annual reproductive process. On a windy day, you can see large, dense clouds of pollen being blown away from these trees. So much pollen is produced that the pollen, which is yellow, builds up as a continuous layer on the surface of the lake, coloring the lake itself yellow over large areas in quiet bays. As the pollen is blown across the surface of the lake, it accumulates at the land on the lake edge, creating a yellow line that persists through the rest of the summer.

It is the buildup of the whole array of wind-dispersed pollen grains from many species in bodies of water and wetlands, especially lakes and bogs, that makes it possible to infer climatic history from these fossils. Each new year’s production of pollen grains tends to accumulate over the last as sediments accumulate in the lake bottom. Gradually, deep sequences of year upon year of accumulation build up, so that an undisturbed lake bottom can contain an unbroken record of year-by-year pollen fallout through thousands of years. As the climate changes, the vegetation changes, since plant species are adapted to different and often narrowly defined climatic regimes. As the vegetation changes, the proportion of different pollen grains dispersed in the air changes - these changes are captured in the sediments accumulating in lakes and bogs, building a record of the changing vegetation in the region.

Plant ecologists have gradually evolved a way to sample pollen from the sediments that build up in lakes and bogs over the last hundred years, so that there are now a set of modern techniques in place for gathering pollen samples. The key goal is to collect an uncontaminated record of the pollen deposition in a region. The basic approach is to drill into the sediment with a peat sampler. These samplers are hollow; as they are pushed down into the lake bottom or bog the sampler fills with a column of sediment, which is an intact record of the accumulation at the site so long as the sampling is done very carefully. Deep columns of sediments can be collected that represent extensive periods of time - and the sediments can be dated using radioisotope decay measurements (McDowell et al. 1971 is an example).

The Lake Champlain Valley is a real museum of the events of the glacial age. With luck, you will be able to use pollen core analysis to infer changes in the vegetation here in the Champlain valley during the recent past based on analysis of a core from a local Lake Champlain area pond sediment.

Recently, we went out to Lily Pad Pond, which is located on the Vermont National Guard’s Camp Johnson property in Colchester, Vermont. Here we used a pollen sampler to collect a column of the accumulated sediments in this pond. The pond has some open water, but much of the pond basin is filled with a muddy peat that has been accumulating since the lake level first dropped below this area some 11,000 years ago. The deepest sediments are reportedly some 10 m (meters); the core we extracted from the bog is about 7 m deep. Based on a possibly questionable comparison to a published pollen profile from another part of Vermont (McDowell et al. 1971, see Fig. 1) the bottom of this core is about 8000-9000 years old. We extracted the core in segments of 0.5 m; these are in the lab in pieces of plastic pipe, marked with the depth from which they came.

I. PURIFICATION OF THE POLLEN SAMPLE

You should work individually through this entire experiment. Each team member should keep in mind the two principal goals - keep the sample uncontaminated and keep the caustic compounds from spilling, and especially from getting on to people.

In the laboratory you will find that the pollen cores are laid out on the bench in their moisture-retaining saran wrap on pieces of pipe cut in half. The supportive half-pipe for each core is marked off in centimeters - your job as a team will be to prepare a sample from one of the points along the core. As you work together with the other people in the lab, you will construct a set of data for the whole core, but your work as an individual will be to develop the pollen frequency data for one place along the core - that is one depth in the original sediment.

A. SELECTION OF MATERIAL AND COARSE FILTRATION

1. Your teaching fellow will tell you which place along the core to work with. Make a careful note of the depth of the sediments that you are analyzing for later reference. At your assigned place on the core, break the core open to expose the pristine interior. Scrape off about 1 (one) cc of material (enough to make a good-sized pea), avoiding the outermost part of the core (the outer part may have been contaminated by the passage of the core up out of the corer). Push the core back together to conceal your sample site.

2. Place the peat in a plastic centrifuge tube, break down as much as possible with a stirring rod, and fill the tube one-third full with distilled water. Wait long enough for the peat to become thoroughly saturated, use a vortex to encourage mixing of water and peat.

3. Filter through two layers of cheesecloth squares into a second tube to remove the largest impurities (leaves, small twigs, etc.)

4. Bring all the tubes up to the same level with extra water, since some is lost to the cheesecloth.

B. BREAKING DOWN THE SEDIMENTS AND REMOVING HUMIC ACIDS - POTASSIUM HYDROXIDE TREATMENT.

1. Centrifuge at 3200 rpm for 4 minutes, pour off the supernatant (distilled water), and re-suspend the precipitated pellet at the bottom of the tube in 5 ml 10% Potassium hydroxide (KOH). Place in a hot water bath for 20 minutes.

In the potassium hydroxide solution, two things are happening that are critical to the process. First, the peat structure is disintegrated (more than you could accomplish with your stirring rod) so that you are dealing with a fine sediment; second, the organic soil colloids, which bind together the original organic substance at the large-molecule level, are broken up by attacking the humic acids in the sample. The heat hastens the process - it would happen over a day’s time at room temperature.

CAUTION: POTASSIUM HYDROXIDE IS HIGHLY CAUSTIC - IT WILL EAT YOUR CLOTHES AND SKIN IF IT COMES IN CONTACT.

2. Remove the tube from the hot water bath, fill with distilled water, and centrifuge (3200 rpm, 4 minutes).

C. STAINING AND MOUNTING ON SLIDES

1. Once again, pour off the liquid, this time into the waste receptacle for KOH. Add a drop of safranin stain to the centrifuge tube.

2. Clean a spatula from the supply in the lab using hot water, then wipe it absolutely dry with a paper towel to ensure that no contaminating pollen is present. Stir the pellet up gently with the spatula to mix the stain in. Let the stain soak into the pollen for five minutes. NO VORTEX.

3. Now you want to mount your sample on microscope slides for viewing. With a stirring rod, let about one drop of glycerol drop on to the center of a microscope slide. Then, with a disposable pipette remove a bit of the pellet from the bottom of the centrifuge tube and stir it slowly and thoroughly into the glycerol. Finally, cover the pollen suspension with a cover slip. Start by placing one edge of the coverslip on the slide near one edge of the glycerol suspension, then lower the coverslip so that it contacts the suspension and can be lowered slowly onto it from one side to the other. Use a probe to hold the coverslip off the slide as you lower it.

4. Repeat the mounting several times to make two to four slides of your pollen extract. You can use the same pellet over and over as long as it lasts. Mark the slide carefully with the depth in cm from which the sample came and your initials, using a permanent black marker. If you make more than one slide, give each an individual identifying number or letter.

D. POLLEN IDENTIFICATION AND COUNTING

Perhaps the most challenging part of this lab is the identification of the pollen grains. The goal here is to identify them to the genus of plants from which they come. In theory, identification to genus is relatively easy, because genera of plants generally have distinctively different pollen grains. However, in practice, you will it difficult because pollen structure is a whole new world.

1. Working with your sample slides, learn to identify the eight common genera of pollens in the peat core. These common pollens document most dramatically the changes in the climate in Vermont since glaciation.

a. Remember the important components of variation in pollen structure: size, the structure of the wall, the sculpturing of the wall, and the furrows and pores that are impressed into the wall. For our purposes these variables reduce to a simpler set: size, shape, surface texture, and development of pores and grooves.

b. Using the diagram (see Fig. 2), locate and identify the following pollen grains. A simple, non-technical description of each follows the name.

i. Abies (fir): large-sized mickey mouse head with ears pointed up

ii. Betula (birch): small-sized triangular pollen with groove at each point of the triangle

iii. Fagus (beech): medium-sized sphere with three sets of pores and grooves

iv: Larix (larch): small-sized uninteresting sphere

v: Picea (spruce): medium-sized mickey mouse head with ears pointed up

vi: Pinus (pine): small-sized mickey mouse head with ears at side

vii: Quercus (hemlock): small-sized pollen with three deep grooves

viii: Tsuga (hemlock): large-sized sphere with bumpy surface

2. Now identify and count the pollen grains in these common genera on your slides. Use as many as it takes to develop the sample size your lab group chooses. The goal here is to count each pollen grain you see only once, and to count just enough pollen grains to have a defensible sample. This goal will take some thinking...

a. First of all, design a plan to count pollen grains under the microscope that avoids confusion. You need a plan that ensures counting each grain once and only once. Thus, you need to have an idea of where you have been and where you are going as you move the microscope slide to view new regions of the pollen preparation. Try out your own plan with a few preliminary counts - then discuss how to make the official counts with the rest of the lab group. When you have made a final decision, then proceed with the official count of your level in the core.

b. Second, develop a plan for deciding when you have counted enough pollen grains

.

You do not have to count all of the pollen grains on your slide, but you need to be sure that you have enough. Now the question arises, when do you know you have enough. Pollen samples typically have a small number of pollens in great abundance and a much larger number of species that are rare: hence the time you have spent looking at the common genera. For our purposes, and for most applications, it’s only the common ones that matter, because they are the only ones that will change dramatically enough from level to level in the sediment to be detectable. So the goal here is to count enough pollen grains to be sure that each of the common types is represented by some minimum useful number.

Generally speaking, quantitative statistical analysis works better if we have at least two of each common pollen grain, so you can yourself figure out when you have enough pollen grains. The way to do this is to begin your counting and keep track of the appearance of new kinds of pollen. Rare pollen grains (not in the common genera) will appear once or twice, then not again. The common grains will appear over and over. When all of the common grains are represented by at least five grains, you have enough. You can assume that grains you have not seen are in fact missing from the sample at this point.

c. Now design and draw up your final data sheet. You need a data sheet that allows you to tally the new pollen grains that you see, one by one, and that makes it easy to see when you have the minimum defensible number of pollen grains. You also need a column in which to write the final number, and a column at the right in which to compute the percentage of the sample that each genus represents. To make this computation, you will also need a space, at the bottom, in which to compute a sum of all pollen grains in your sample.

Remember, the final data sheet sets the standards for your work - it is the single instrument that defines your experimental work, based on the idea for the experiment and the procedures you have chosen to carry out the experiment. A good data sheet reflects careful thought about the original idea for the experiment and for the experimental test. A good data sheet well filled in is absolutely critical to this experiment, because you are contributing your data to the whole lab’s effort.

d. Go ahead and count your pollen.

E. CONSTRUCTING THE POLLEN DIAGRAM

The construction of a pollen diagram is in fact the analytic step in this experiment, since you are taking the raw data - the numbers that everyone has generated as they counted their pollen - and incorporating them into a visual summary. The goal of this summary is to allow you to easily envision the changes in percentages of the common genera of tree pollens in the sample over time.

The plan is to enter all of the data on the chalkboard, then for you to construct your own complete pollen profile using everyone’s data. Consequently it is critically important for your counts and identifications to be as high-quality as possible.

1. Assembling the complete data set: You, your fellow labmates, and your teaching fellow will design the final data sheet for the chalkboard. The horizontal axis should be time from oldest (left side of the board) to youngest (right side of the board). The vertical axis will be the eight genera. (This is the exact opposite of the usual custom - see the example pollen diagram - but your data will fit into the chalkboard better the other way.) You will be responsible for your layer in the core, which means that you will fill your numbers in for your eight genera in one vertical column.

2. Constructing your own pollen diagram: Once the data are all entered on the chalkboard, make your own copy of the whole data set in your notebook. Convert your count data to percentage data by figuring out the percentage of each of the pollens, where 100% is the sum of the counts for all eight tree species. If time permits, construct the pollen diagram from these data while you are still in the lab. Follow this plan...

a. The sense of the pollen diagram is to show the changes in percentages of each of the eight genera of trees over time with a line connecting the percentages of a single genus through time. Thus the whole diagram is a set of varying lines in a graph, with total numbers and relative numbers varying up through the core and thus up through time.

b. For each genus, plot percent of total pollen against time, with the time axis vertical from oldest (deepest) to newest (nearest the surface). Combine all eight diagrams in a single profile that allows you to compare the percentages of the eight genera of trees. Use the sample pollen profile as a model for your own work.

F. INFERENCES AND LIMITATIONS

Now comes the intellectually challenging part. Decide by inspecting your pollen diagram whether you can infer changes in vegetation surrounding Lily Pad Pond based on the pollen percentages in the core. Look first for obvious patterns like the vertical location of the maximum percentage of pollen of each genus. Are some maxima early and others late? Then look for direct and indirect correspondence between different genera - are there genera that are often high or low in percentage at the same time? Are there genera that are mirror images - one is common when the other is rare and vice versa? In this way you can build a set of inferences about which genera occur together and when they occur together.

Now seek an explanation for the pattern you are seeing relative to the distribution of these genera of trees in present day Vermont. This job you must do outside of the laboratory. You will have to somehow document the distribution of these genera in the state. This might best be done by thinking about any experience you have with the high mountains of the state, such as on a ski trip. What genera are found most commonly highest on the mountains, what genera are found here in the valleys? You may also want to consult one of the general texts on Vermont vegetation, such as Johnson’s Nature of Vermont. You may also want to talk to people who know the vegetation of Vermont well. Use whatever sources you can to develop an understanding of the present-day distribution of these trees in Vermont.

Finally, take time to think about the weaknesses of this experiment. What factors could influence the variation in the frequency of the pollen besides the changes in vegetation over time? There are several, so think carefully. Are any of these factors evident in the pollen profile that you have constructed? Work towards developing an intelligent set of ideas about the limitations on the inferences you have made about the vegetation changes. These ideas on vegetation change and the limitations to inferences about vegetation change will be important to develop, since they will form the final section, the discussion, of your lab report on this laboratory.

G. USEFUL REFERENCES

Birks, H. J. B. 1980. Quaternary Palaeoecology. Baltimore: University Park Press.

Faegri, Kn. 1975. Textbook of Pollen Analysis. 3d ed. New York: Hafner Press.

Johnson, C. 1980. The Nature of Vermont. Hanover, NH: University Press of New England.

Johnson, C. 1985. Bogs of the Northeast. Hanover, NH: University Press of New England.

McDowell, L.L., R.M.Dole Jr., M. Howard Jr., and R.A. Farrington. 1971. Palynology and radiocarbon chronology of Bugbee Wildflower Sanctuary and Natural Area, Caledonia County, Vermont. Pollen et Spores 13(1):73-91.

Moore, P.D. 1991. Pollen Analysis. 2d ed. Oxford: Blackwell Scientific Publications.

Moore, P.D. 1978. An illustrated guide to pollen analysis. London: Hodder and Stoughton.

West, R.G. 1971. Studying the past by pollen analysis. London: Oxford University Press.

H. SYNOPSIS of the McDowell et al. 1971 results:

Zone	Time period	Pollen trends	Climate
A	10,500-9300 ybp	spruce and fir common	cooler and moister than present
B	9,300-7,500 ybp	pine high, spruce and fir rarer	warmer and drier than zone A
C-1	7,500-4,000 ybp	pine lower; hemlock, birch, oak, and beech commoner	hipsothermal: warmer than present
C-2	4,000-1,600 ybp	maximums for hemlock, pine, and oak