Research Project

How do habitat characteristics of a restored prairie vs. an abandoned cornfield affect adult monarch numbers, sex ratio, and use of habitat?

Male monarch

Zoe’ Rochester, Jack Sondreal, Eva Vasiljevic, Alex Pauhle, Stephanie Schmidt, and Karen Hanson
Amery Middle School, Amery WI
St. Croix Falls High School, St. Croix Falls WI

Abstract

We researched how habitat affects adult monarch numbers, sex ratio, and use of habitat. We studied three sites located in Polk County Wisconsin. The first site was D. D. Kennedy, a restored prairie between Amery and St. Croix Falls. The second site was an abandoned farm field located in Amery. The last site was another abandoned farm field located in a city-owned nature reserve in St. Croix Falls. Starting in early July 1999 we sampled each site approximately every two weeks until early September 1999. We studied the size and makeup of the monarch populations, milkweed abundance, and kept track of flowers blooming during each sampling day. We used mark-release-recapture techniques to study monarch population characteristics. Field transects were used to collect flowering species data. We also recorded the behavior and nectaring preference of the monarchs that we observed. Our data suggest that there were large numbers of monarchs at each site, although we had too few recaptures to make accurate estimates of population size. We found a large variety of flowering plants at all three sites.

Introduction

Polk County Wisconsin was originally a combination of oak savanna and prairie ecosystems (Wisconsin DNR, 1999). A prairie is a unique ecosystem. It is a combination of native grasses and wildflowers (Nichols, p.2). An oak savanna is similar to a prairie, but contains some mature oaks. "Savannah can be defined as having one tree or more per acre, but with less that 50% of the area covered by trees." (Murray, p.10). Scientists estimate that only one-half of one percent of the original Wisconsin prairie is still remaining. For the last 200 years the prairie/Savannah ecosystems have been replaced as settlers move into the area and build homes and plant crops. The original settlers purposely brought new plant species with them, such as red clover for teas and wines. They also brought non-native species like burdock and cocklebur unknowingly (Thomas, p.13). Many of these plants and grasses out-competed some of the original species.

In the last decade there has been growing interest in restoring land to its original plant communities. Residents of Wisconsin have been encouraged through grants and governmental support to restore their vacant land to native ecosystems. Most local farm fields are a mixture of a small number of native species and a large number of introduced non-native species.

Our study group was interested in the effects of prairie restoration on monarch butterflies. Originally we asked, how do habitat characteristics of a restored prairie vs. an abandoned cornfield affect adult monarch numbers, sex ratio, and use of habitat?

Our group consisted of two students and a teacher from Amery, Wisconsin and two students and a teacher from St. Croix Falls, Wisconsin. Our study sites were all located in Polk County, Wisconsin. The Amery site was a 54-acre old farm field located less than a quarter-mile from Amery Middle School. The St. Croix Falls site was a 6-acre old farm field which is part of an 89-acre tract of land donated to the City of St. Croix Falls. This site is located .5 miles from the St. Croix Falls Middle School. The last site was a 75-acre restored prairie located halfway between the schools.

Originally we produced the following table detailing expected similarities and differences between the two site types. We used this table to develop hypotheses on the characteristics of monarch populations at the three sites.

Prairie	Both	Old Fields
Less milkweed		More milkweed
More milkweed species		Mostly common milkweed
Native plant species		Non-native plant species
Larger		Smaller
	Nectar availability
More diverse nectar		Less diverse nectar
	Water Human disturbance

On studying the site we did find one error in our original data table. There was actually more milkweed in the restored prairie. On questioning local naturalists they explained to us that the common milkweed was one of the species that comes back in a restored prairie whether it is planted or not.

Hypotheses

We developed several hypotheses about the relationship between site characteristics and monarch population characteristics. These are listed below, along with some of the reasons for the alternative hypotheses.

Milkweed Availability

Ha₁: Higher milkweed density will cause a higher female population (females are attracted to milkweed to lay eggs).
Ha₂: Higher milkweed density will cause a higher male population and smaller female population (females will avoid patrolling males)
Ha₃: Higher milkweed density will cause a higher overall monarch density, but it will not affect the sex ratio.
H₀: Milkweed density doesn’t affect monarch sex ratio or density.

Nectar Availability

Ha₁: Higher daily nectar availability causes higher monarch population.
H₀: Higher daily nectar availability will have no effect on monarch population size.

Nectar Diversity

Ha₁: More flower diversity will cause a higher monarch population (nectar is there for more time).
H₀: Flower diversity will have no effect on monarch population size.

Native vs. Exotic Nectar Plants

Ha₁: Monarchs prefer native plants (since they evolved in the presence of these species).
H₀: Monarchs nectar equally from exotic and native flowers.

Site Size

Ha₁: There will be a higher monarch density in a large site (monarchs are more likely to find a large site and stay there).
H₀: Site size does not affect monarch population density.

Methods

We collected monarch data from one of two types of sites each week. Every other week the entire six-member team collected data from the restored prairie, and on alternative weeks, as separate three-person teams from the two abandoned farm fields. Our methods consisted of the following:

Catch butterflies in the morning. Record the number of monarchs caught in the morning, and the behavior of the monarch when caught as nectaring, flying, or other. Record the sex of the captured butterflies, mark them with a number on their forewing, and release them.
Walk transects through the field. Count the number of blooming plants per square meter. Record the number of transects examined, and the species and the number of blooming plants per square meter. Record other plants that are observed to be in bloom on that date that did not happen to fall in a sample.
Catch butterflies in the afternoon. Record the number of monarchs caught in the afternoon and the behavior of the monarchs at the time of capture (as in step one). Record the sex of the afternoon-captured butterflies. Number and release the butterflies caught in the afternoon.
Calculate population size of the monarchs each week using the following formula (Where X = total population of monarchs):

# of butterflies caught in AM	=	# of marked monarchs caught in PM
X	=	Total # of monarchs caught in PM

Determine the population density of flowering plants using the following formula:

# of flowering plants = Density of Flowering Plants
# of transects

The following calculations were done one time:
- Determine the population density of milkweed using the formula stated in step # 5.
- Determine the area of each site in meters.

Results

Milkweed Availability

There was a higher density of milkweed at D. D. Kennedy, the restored prairie (Figure 1). All sites contained milkweed, and attracted monarch butterflies. Milkweed density varied from about 0.8-1.2 ramets per m², with the highest density at the restored prairie (D.D. Kennedy).

Monarch Populations

Numbers

We only recaptured monarchs on two. On both July 7 (at DDK) and September 16 (at Riegel), we caught a total of 12 butterflies and recaptured one. Using the MMR formula given above, we estimated that there were 72 monarchs at these sites on both dates. Monarch collection numbers and sex ratios changed from week to week (see Figures 2-4). This was due partially to skill of the collectors and weather on the sampling day, but is probably at lest partially reflective of the number of butterflies present at each site on our sampling dates.

Sex Ratios

We used chi-square tests to discover any difference in the sex ratio of monarchs between dates at each site and also between the sites (Tables 1, 2, and 3). At Riegel Park, the sex ratio did not change significantly throughout the summer, as our calculated chi-square value was lower than the critical value (Table 1). At D.D. Kennedy (Table 2), the calculated value slightly exceeded the critical value at a 90% confidence level. At Amery (Table 3), the calculated value was also slightly above the critical value at a 90% confidence level. The sex ratio appears to have remained constant at Riegel Park, and may have become less male biased over the course of the summer at DDK and Amery.

When we compared the overall sex ratios between the sites, it was a different matter. Both Riegel and Amery are male-dominated with Amery consisting of almost twice as many males as females (Table 4). D. D. Kennedy is female biased. Our chi-square value showed that these differences are statistically significant.

Table 1. Riegal Park
Observed				Expected
Date	M	F	% M	Date	M	F
7-13	9	3	75%	7-13	7.38	4.62
7-26	3	2	60%	7-26	3.08	1.92
8-21	7	3	70%	8-21	6.15	3.85
9-3	7	5	58%	9-3	7.38	4.62
9-16	6	7	46%	9-16	8	5

d.f. =4 χ 2 =1.6 χ 2_crit = 7.78 (90%)

Table 2. DD Kennedy
Observed				Expected
Date	M	F	% M	Date	M	F
7-7	9	4	69%	7-7	6.04	6.96
7-21	6	7	46%	7-21	6.02	6.98
7-29	3	11	21%	7-29	6.5	6.5
8-22	1	0	100%	8-22	.46	.46

d.f. =3 χ 2 =7.27 χ 2_crit =6.35 (90%)

Table 3. Amery
Observed				Expected
Date	M	F	% M	Date	M	F
7-14	24	14	63%	7-14	24.43	13.57
7-28	5	1	83%	7-28	3.86	2.14
8-5	5	2	71%	8-5	4.5	2.5
8-11	2	1	67%	8-11	1.93	1.07
9-7	9	7	56%	9-7	10.29	5.71
9-15	13	7	65%	9-15	12.86	7.14
9-21	14	22	39%	9-21	23.14	12.85

d.f. =6 χ 2 =11.64 χ 2_crit = 10.64 (90%)

Table 4. Site by Site Comparison of Sex Ratios
Observed				Expected
Site	M	F	% M	Site	M	F
Riegal	32	20	62%	Riegal	30.88	21.12
Amery	72	40	64%	Amery	66.52	45.48
DDKen	13	20	39%	DDKen	19.60	13.40

d.f. =2 χ 2 =9,8 χ 2_crit=6.965 (99%)

Butterfly Behavior

Although we wrote down nectaring preferences and monarch behavior on some sampling days, we failed to do so on others. However, on the days we did note preferences we were able to calculate the percentage of each behavior displayed. The results are as follows:

Table 5. Butterfly Behavior D.D. Kennedy
Date	# Collected	% Nectaring	Species of choice	% Flying	% Egg-laying
7/7	13	46%	Milkweed	46%	0%
7/29	14	35%	Not recorded	14%	21%
8/22	1	100%	Goldenrod	0%	0%

Table 6. Butterfly Behavior. Riegel Park
Date	# Collected	% Nectaring	Species of choice	% Flying	% Egg-laying
7/13	12	58%	Milkweed	25%	0%
7/26	5	40%	Milkweed	20%	40%
8/21	10	60%	Thistle	10%	0%
9/3	12	25%	Thistle	41%	0%
9/16	14	78%	White Aster	0%	0%

Monarch behavior varied throughout the summer. In July we observed more egg laying. In late August the butterflies spent more time nectaring. The weather on a given day greatly affected behavior and numbers collected. At D. D. Kennedy on August 22nd it was extremely cold. We collected only 1 butterfly (Figure 3). On July 29th it was extremely hot. We collected 13 butterflies in the morning. When we tried to recapture, the temperature had risen to 95 degrees. We only collected one butterfly.

On any given day at a specific site, there was often a flower preference. When the milkweed was blooming, many of the butterflies were collected on milkweed flowers. At Riegel on August 21st, most of the butterflies were nectaring on White Asters.

Nectar Diversity

We found a large variety of flowering plant species at all three sites (Table 7). The density of plants in bloom changed dramatically over the season, ranging from a low of 0.6 flowering plants per m² at DDK on July 7 to 6.3 flowering plants per m² at Amery on August 11.

Table 7. Flowering Plants (Top 4 Bloomers)
Date	Bloom density/m²	#1	#2	#3	#4	Total # of species
AMERY
7/14/99						15
7/28/99	.9	Canada Thistle 60%	Butter & Eggs 27%	Fleabane .1%		22
8/11/99	6.3	Goldenrod 90%	Butter & Eggs .1%	Fleabane .02%		14
RIEGEL
7/13/99	1.98	Fleabane 40%	Yarrow 19%	White 12%	Bergamont 10%	15
8/21/99	3.2	Goldenrod 49%	White 40%	Fleabane 8%	P. Everlast 1%	20
9/3/99	.96	Goldenrod 58%	Aster 12%	P. Everlast 10%	Thistle 8%	13
9/16/99	2.56	Goldenrod 89%	Cinquefoil 6%	Aster 1%	Part. Pea 1%	11
D.D. Kennedy
7/7/99	.6	H. Allysum 38%	Milkweed 26%	Fleabane 23%	Yarrow 7%	20
7/21/99	1.29	Fleabane 55%	B. Susan 15%	Yarrow 8%	Cinquefoil 6%	24
7/29/99	1.37	Fleabane 52%	B.Susan 21%	P. Conefl. 8%	Bergemont 6%	18
8/22/99	2	Goldenrod 70%	Grey Cone 11%	Oxeye D. 11%	P. Conefl. 6%	12

Exotic vs Native Species

As shown in Table 8, there were more native than non-native plants flowering at all three sites. We used the chi-square test to determine if butterflies nectared on plants according to their abundance, or if they preferred native plants disproportionately. 100% of the butterflies nectared on native species at Riegel and D.D. Kennedy, and 95% of the species sampled at Riegel and D. D. Kennedy were native. We have no data for nectaring preference at Amery. The expected values are almost exactly the same as the observed values.

The major difference between the prairie and old farm field was the species of grasses present. The restored prairie contained mostly warm-weather grasses such as Big Blue Stem, Little Blue Stem, Switch Grass, and Indian Grass. The old farm field contained mostly cold-weather grasses such as Timothy and Quack Grass.

Table 8. Exotic vs. Native Plants Flowering (total numbers)
	Riegel	Amery	D.D.Kennedy
Exotic	1	12	36*
Native	186	207	312

*Ignoring 161 Rabbits Foot Clover. This is not a nectaring source for Monarchs. It also is an extremely small plant, found on the very edge of the prairie.

Table 9. Numbers of Monarchs observed nectaring on native and exotic plants
Site	Native	Exotic
Riegel	24	0
D.D. Kennedy	7	0

Table 10. Exotic vs. Native Butterfly Preference Expected Data (assuming no preference)
Site	Native	Exotic
Riegel	24	0
D.D. Kennedy	6.7	.3

Discussion

Originally we were interested in this topic because we felt that prairie restoration might have some adverse affect on monarch numbers, if, for example, there were less "weedy" milkweed in a restored prairie. We chose a site near each of our schools to facilitate student research. Both of the school sites were old farm fields full of "weeds". The restored prairie site was at a site in between our two schools. The seed mix had been chosen and planted 5 years before. When we formed our original hypotheses, we actually thought that the restored prairie would not contain a lot of milkweed. This was not true. The prairie had a milkweed density of 1.2 plants/m². The old farm fields had an average milkweed density of .8 plants/m². The size of the three sites also varied. The Riegel Park sampling site was 5 acres, Amery was 54 acres, and D.D. Kennedy was 75 acres.

In our original hypotheses on milkweed availability, we addressed how milkweed density would affect monarch populations. Because we only recaptured two butterflies all summer, it was impossible for us to calculate the size of the monarch populations. There were many factors that contributed to our inability to recapture monarchs. Some of these would be the researchers’ lack of experience collecting butterflies, poor weather conditions, time constraints, and possibly a very large monarch population. We also could not draw any conclusions on how site size affected monarch population for the same reason.

Because we could not estimate butterfly population sizes, we could not draw any conclusions on the effect of nectar availability and nectar diversity on monarch populations. We did find that on given dates the monarchs showed a preference for certain types of flowers. As shown in Tables 5 & 6, the majority of monarchs preferred common milkweed in the month of July. In August, they preferred thistle and goldenrod. In the month of September, the majority of butterflies collected were on white asters. Our data on flower choice are somewhat incomplete. This is due to a lack of communication, organization, and record keeping on our group’s part. We lost some of our data and forgot to collect some of it on certain dates. We did find that there was a large diversity of flowers blooming at each of the three sites. It is difficult to draw any other conclusions from this data, because we did not sample often enough.

D.D. Kennedy, a restored prairie, had a female-biased population, while the abandoned cornfields were male-biased. The reason for this could be the higher density of milkweed at D.D. Kennedy, which is graphically presented in Figure 1. As there is no great variance between flower species at the three sites, nectar preferences probably do not affect this.

As far as native vs. exotic nectaring preference, we found that 100% of the Monarchs chose to nectar on native flowers. Because 95% of the flowers sampled at the three sites were native species, we could not show any significance using the chi-square test. Monarch nectaring preference was also our most incomplete data. We got so excited when we caught butterflies that we often forgot to record what type of plant the monarch chose to nectar on. We also lost some of this data in the field.

One of our original hypotheses addressed sex ratio as related to milkweed density. Our data suggest that there may be a correlation between a high milkweed density and a high female population. This may be a valuable direction for future research.

In conclusion, we really were not able to shed light on most of our original hypotheses. However, we learned incredible amounts of information about monarch behavior and flowering species of western Wisconsin. Our research really produced more questions than answers. Some of these are:

How does temperature affect nectaring behavior?
Do all prairies contain a large density of milkweed?
How long does an individual monarch stay on one field?
Why were there more female monarchs in the restored prairie?

Acknowledgements

We thank the following:

Karen Oberhauser and her staff for their knowledge and support for this process.
The Amery and St. Croix Falls School Boards provided financial and moral support.
The City of St. Croix Falls for allowing us to use Riegel Park.
Parents for all the extra trips to school.
Karry Redlich for picking us up at the airport late at night.
Rachael Hanson and Robert Rochester for their data collecting assistance.

Literature Cited

Murray, Molly Fifield. 1993. Prairie Restoration for Wisconsin Schools. University of Wisconsin-Madison Arboretum.

Nichols, Stan. 1978. Prairie Primer. University of Wisconsin Extension.

Thomas, Donna, 1993. "Growing Native". Wisconsin Natural Resources.

Wisconsin DNR. October 1999. http://www.dnr.state.wi.us/org/at/et/geo/map_gal/landcov/orgveg.htm

Biographies

	Alex Pauhle is a freshman at St. Croix Falls High School. Some of his interests include national politics, the environment, and current events. Some of his hobbies include playing the piano and drums, doing homework, and web site design. Alex also enjoys debate.
	Steph Schmidt is a junior from St. Croix Falls High School. She has two parents and three brothers who also attend school in St. Croix Falls. Steph enjoys many things; playing the piano & organ, singing, collecting wildflowers, walking, jogging, science, reading, writing, and drawing. She plans to be a piano major in college.
	Karen Hanson is a teacher at St. Croix Falls Middle and High Schools. The subjects that she teaches are physical science, earth science, and environmental science. Her interests include walking, softball, gardening, Monarch butterflies, and learning new things.
	Jack Sondreal is an eighth grader at Amery Middle School. He plays many sports, basketball, tennis, and football. Jack enjoyed catching butterflies, especially when there were many of them. He enjoys playing with his BIG dog, Cosmo.
	Eva Vasiljevic was born on November 12th, 1985 in Sarajevo, Bosnia. She plays volleyball, soccer, swims, and plays tennis. Eva liked catching the butterflies. She has an older brother who lives in Canada. Eva took care of over one hundred developing eggs and larvae in a tent in her backyard during the summer of 1999.
	Zoe' Rochester has taught 7th grade science for ten years. She has been a science explorer since her childhood days in the woods and beaches of northern Minnesota. She enjoys the intimacy of the outdoors in all seasons.