Tuesday, November 18, 2014

Fall 2014 fruit and seed work

I mentioned in a previous post that our analyses of seeds found in the fecal samples of birds requires context: some sort of compilation and rank abundance of all the ripe fruit available to birds to eat when they are present in the area. In that post, I described our fruit counting methods.

Fruit morphology
Another aspect of our work this fall was describing various morphological characteristics of the common species of fruit available at our site. Many factors go into fruit choice by birds. Among them are the size of the fruit, how many seeds it contains, and the size of the seeds.

Some of these data are available in the literature. As I searched for these metrics, I found that for some species different sources reported quite different numbers, some sources had data on one characteristic but not others, while measurements for some species were not to be found. The best, though not the easiest, solution was to take measurements ourselves. Weekly, we collected around 10 fruits from each of around 30 species, measured the fresh fruit, counted the seeds per fruit, and measured the seeds. Dana Wloch started this project last year, and it was expanded this year. I'll be finishing up with some of the late fruiting species next week, but so far we have measured about 3300 fruits and over 7000 seeds!

Some of our results are basic. For example, the average diameter of the five most common seeds in fecal samples (Common Buckthorn, Amur Honeysuckle, American Pokeweed, Riverbank Grape, and crabapple) is 8.2 mm. This agrees with other studies that have shown bird-dispersed fruits typically average about 8 mm.

Some results are much more intriguing. Glossy Buckthorn is a non-native species that can be quite invasive, especially in wetlands. A recent study from its native range in Sweden stated the average diameter of the fruits was 8.7 mm (458 fruits) with an average of 1.7 seeds per fruit. We measured 134 fruits here over two years, and the average both years was 7.7 mm. The average number of seeds for 168 fruits was 2.5.

Glossy Buckthorn fruit does not ripen simultaneously, so red,
unripe fruit are often on the same branch as black, ripe fruit.

Glossy Buckthorn has been shown to have evolved different morphological characteristics in different parts of its native range. This species has been present in North America for over 200 years. If our data is truly representative of the local population, it might suggest adaptation to a different suite of dispersers, exposure to more or better pollinators, and the higher seed set may be a factor in its success as an invader. All speculative at this point, but no doubt it will prompt me to continue measuring Glossy Buckthorn!

Seed samples
Meanwhile, during the fall season, we collected 453 fecal samples from 6 bird species. However, over 80% of the samples were from robins that we did not band. Several years ago, the University began cleaning off all sidewalks and roads on a daily basis in fall. In a number of areas, low spots in the pavement hold water and attract robins, which often then "leave a deposit." Collecting these samples in communal bathing and drinking areas, as well as along other paths where we observed robins foraging on the ground, is a convenient way to acquire a lot of data.

For catbirds, the top four species in our 41 samples this fall were:
  • American Pokeweed (native)
  • Riverbank Grape (native)
  • Amur Honeysuckle (non-native)
  • shrub dogwoods (native)
This closely follows the rankings for the 5 previous years combined. Pokeweed and grape have ranked #1 and #2, while the shrub dogwoods have been #3 and Amur Honeysuckle and Glossy Buckthorn are close together at #4 and #5.

Sample sizes for robins are much higher. This fall the 453 samples revealed the top four species as:
  • Amur Honeysuckle
  • Common Buckthorn (non-native)
  • crabapples (non-native)
  • Riverbank Grape
These four species are also the same top four species from over 900 samples from previous years, although Common Buckthorn has been the top ranked species for those 5 years.

Two other species in robin samples were notable. Previously, Multiflora Rose (non-native) was found in 4% of samples; this year it increased to 6%. This is interesting considering a number of Multiflora Rose has been reduced in the past year or so both by removal and from infection by rose rosette disease. We also have the native Illinois or Climbing Rose (Rosa setigera) here. There is overlap in the appearance of seeds of these two species, but there are fewer R. setigera, their hips are larger and "ripen" later than multiflora. I believe most of the seeds found in fall samples are probably multiflora.

A comparison of the large hips of the native Rosa setigera (left) and the
small hips of non-native Rosa multiflora (right).
Asiatic Bittersweet (non-native) was previously found in less than one percent of samples. This year, it was in 5% of samples, a rather large jump. This species has also been the target of removal the last couple of years. The capsules encasing the fruit usually open fairly late in the season, and did not seem much earlier this year (first date in samples Oct 8) than the average (Oct 16). Thus, neither increase in the number of plants nor early fruiting seems to explain the higher proportion of samples with bittersweet.

Asiatic Bittersweet is especially conspicuous when there are no leaves on trees.

One possible explanation for the increase in these two species is that robins are eating more of them because a great deal of their top-ranked species (buckthorn and honeysuckle) have been removed in the larger landscape.  However, annual fluctuations due to weather and crop size can be large, so it will take more years of sampling to see if this year was just a quirk, or if trends will start to appear.

I haven't done much digging through the numbers yet, but even this quick look is pretty interesting. 

Friday, November 14, 2014

Fall banding 2014 (last season?)

We wrapped up the 2014 fall banding season on 30 October. This was a tad earlier than usual, but the last week was marred by deer destroying two more nets, and the sighting of yet another "stray" cat.

Is the RRBO banding program in jeopardy?
As noted in a previous post, this was a challenging season. Due to funding shortfalls, we cannot maintain the banding area as needed, and more importantly our hours have been cut. On banding days, all attention has to be devoted to tending the nets and the birds. We also had to spend time doing vegetation surveys when plants were fruiting throughout the season. As a result, time did not allow me to keep up with all my usual activities. Among things that went by the wayside were posting more regular updates to this blog and social media (I apologize!), and doing data analysis or writing the paper I am currently on, the results of our thrush diet study.

My obligation as a bander and researcher is to interpret and disseminate data we have already gathered; we need to make this a top priority. I have been tasked with raising at least $20,000 this year to maintain our current (reduced) schedule without further cuts. If we face more reductions in hours, we certainly will not be able to band next year. If we can top our goal, I hope we can resolve some of our time and resource issues and band at least part-time next fall.

How you can help

RRBO has over 2,000 people who follow us on Twitter, Facebook, or subscribe to our newsletter. Our $20,000 goal can be reached if everyone pitched in just $10! If everyone could spare $50 it would give us breathing room, and give RRBO the resources we need to accomplish our research goals.

You can visit the Support page on the RRBO web site for instructions on how to download a donation coupon, or donate online. If you'd like to donate via credit card directly through the University's secure web site just click this link. Thank you!

Will there be nets in the net lanes next year?

The fall 2014 banding summary, part 1: the birds
On to the birds. RRBO banded on 17 days between 18 August and 30 October. We utilized only about two-thirds of the number of nets we usually do, arranging them to avoid the most deer-prone areas to minimize destruction of nets and harm to birds. We still lost some nets but the damage occurred when they were closed and no injuries to birds resulted.

We captured 317 birds of 46 species. Since we tried to band on days when we hoped to maximize our captures of target species (this year, catbirds and robins), our capture rate* was 50.6 birds per 100 net-hours, the highest since 2006. However, cherry-picking days for good conditions, versus banding every day, means that we can't really compare these results with previous years.

We did succeed with our modest goal, as catbirds and robins were the top two species banded. In fact, despite the very limited schedule, our 38 new catbirds banded was a higher total than the past two years, which were dismal record lows. And we got over 40 seed samples from catbirds (more on that in the next post).

The highlight of the season was the Marsh Wren banded on October 9 described and shown in the previous post. A Blue-gray Gnatcatcher on September 15 and a Yellow-bellied Sapsucker on September 29 were other less-common species in the nets. While it's hard to judge without daily banding or surveys, all thrushes, but especially Hermit Thrushes, seemed scarce this fall, as were sparrows such as Lincoln's, Swamp, and even White-throats.

Although on my own time, I thought I would mention that I also banded at a Washtenaw County site previously used for the catbird project. Although using fewer nets than RRBO in 12 sessions for about half the number of hours, the capture rate was higher at 59.6 birds per 100 net-hours, a great result given that we did not choose our banding days based on conditions. We captured 167 birds of 27 species, with American Goldfinch and White-throated Sparrow being the top species. I am experimenting with this site for several reasons: to test the potential as a back-up for RRBO's campus site, to collect fecal samples to determine diet at a site less-invaded by non-native species to compare with RRBO's data, and to train an enthusiastic future scientist, Aspen Ellis.

Aspen is a UM-Ann Arbor student who has already accumulated a lot of varied ornithological experience: for many years she has prepared bird specimens at the University of Michigan's Museum of Zoology, and she has worked on Audubon's long-running Project Puffin as a research assistant the past two summers in the Gulf of Maine. Aspen's dedication to birds and science has been recognized by the Tim Schantz Memorial Foundation with a scholarship to Alaska to present her puffin experience, and the American Birding Association in their Young Birder of the Year contest. Read about her experiences and goals in her own words here at the eBird website.

Banding in Dearborn wasn't very practical for Aspen around her school schedule (and that of the family's cars!). But her clear commitment to bird conservation and science and her maturity spoke to me, and using the Washtenaw County site -- very near her home -- was the answer. It was a pleasure working with her. She is a fast learner and was mastering the basics as the season drew to an end; I hope we can continue the process next year.

Aspen and a female Eastern Towhee.

Part 2: the fruit data
The paper I am working on now is looking at seeds from fruits in the diets of over 300 Hermit and Swainson's Thrushes. I began with them because my previously published paper documented weight gains in these species. Meanwhile, we are still collecting data on other bird species. We have good samples sizes from catbirds (over 200), and a great data set for robins (well over 1000 from the fall seasons alone). Since catbirds and robins nest here, we would like to see if the fruit in their diets differs from the thrushes, which are only here during migration. Since we also have data for robins from summer and winter, we can also look at how their diets change throughout the year. We have also collected over 1000 specimens at the Washtenaw County site for comparison.

I will summarize a very demanding and interesting season of data on fruit availability and diet sampling in my next post.

A huge thank you to two people. First, Dana Wloch returned for yet another year of banding and seed processing. This year, measuring fruits and their seeds was added to her repertoire. In past years, she has been supported by grant money. This year, she came in several days a week as a volunteer. Truly, I would have gone insane without her.  At the Washtenaw site, my husband Darrin O'Brien forfeited his Sundays to help out banding and training Aspen. Darrin is a talented bander...and every new bander should get more than one perspective!

*In order to compare different locations or years that may operate the same number of hours but with more or fewer nets, capture rate is calculated by "net-hours." One net hour is one 12-meter net open one hour, or two 6-meter nets open one hour, etc. This rate is often expressed per 100 net-hours for more manageable numbers.