Wildlife CSI (Compost Scene Investigation)

During last Thursday’s regularly scheduled field work, we had the pleasure of the company of several visitors.  Amy Coan, a Wildlife CSI – Summer 2013 participant on the Connecticut Counting Crows took part.  She teaches chemistry and environmental science at East Catholic High School in Manchester, CT.  Also joining us were Dr. Cameron Douglas, a recently arrived postdoctoral fellow in Trinity’s Environmental Science program, and his two research students Rose Rodriguez and Sarah Black.  Thanks to Cameron for sharing this image.

More CSI participants are visiting this week.  If you are able, we extend a welcome for you to join us for upcoming fieldwork on Wednesday, 17 July; Tuesday, 23 July; Monday, 29 July; or Friday, 16 August.  Field work will begin mid-morning and will run up to several hours.  Please contact me (scott.smedley@trincoll.edu), if interested.   For those unable to attend in-person, we are looking into the possibility of a live, virtual tour by hosting a webinar during one of the later field work sessions.  Stay tuned.

Wow!  We have some very productive contributors to Wildlife CSI.  In just over 10 days, you have collectively categorized over 25,000 images!  Thanks for your help with this research.  Congratulations on some strong performances.  Here are standings (click to enlarge) as of early this afternoon.  If your team has yet to contribute much, don’t lose heart.  Over five weeks remain.  You don’t have to be a record-breaking contributor to help.  Every image categorized moves us closer to our goal of a better understanding of scavenger ecology.

To address Chris Higginbotham’s question posted on the Facebook page, here is the image in question with blue jay (circled in blue) and mystery bird (circled in orange).  Again, this whole series benefits from use of consecutive images.

With our recent heat and humidity, frigid winter mornings are hard to recall.  This past winter, I was out early on many a cold morning following nighttime snowfall.   The goal was to collect freshly deposited cottontail rabbit fecal pellets (see image).  As the roughage of the woody, winter diet passes down the digestive tract of a rabbit, it scrapes off epithelial cells.  That epithelial DNA can then be used to distinguish between the two cottontails that occur in southern New England:  the imperiled New England cottontail (Sylvilagus transitionalis) and the ubiquitous, non-native Eastern cottontail (S. floridanus).  In the field these two species can’t be told apart conclusively based on their visual appearance.

Got the DNA results back today –  helpful for establishing the likely identity of the rabbits in our study, but not terribly exciting from a conservation perspective – all pellets from Eastern cottontails.  Thanks to the Connecticut Department of Energy and Environmental Protection for its collaboration in this effort.  Interested in further background on the New England cottontail?   Visit:  http://news.yahoo.com/seeking-save-peter-cottontail-extinction-145130105.html

Check out this recent image from the current round of our compost experiment.  Okay, they may be cute, but overabundant white-tailed deer are a problem in forests.  Experiments where exclosures prevent deer from accessing certain forest plots (see/hear a description of one such study at http://www.npr.org/2011/06/15/137192604/what-does-more-deer-mean-for-forests) have demonstrated that over-browsing by these herbivores can have a variety of negative impacts, including alteration of plant population structures, reduction of plant species diversity, and promotion of non-native, invasive plant species.  What options do we have to mitigate these adverse effects of deer in our human-influenced environment?

As mentioned in the CSI webinar, here is an aerial view of the study site, showing the three pile locations (yellow circles), and some other maps placing the site geographically.   Now, some questions for you.  Why these particular pile locations?  Why are the pile treatments (CON, MIX, VEG) randomly assigned among these locations between rounds of the experiment?

For those Wildlife CSI – Summer 2013 participants for whom it is geographically feasible, we invite you to visit the study site one of several occasions when we are doing field work there:  Thursday, 11 July; Wednesday, 17 July; Tuesday, 23 July; Monday, 29 July; Friday, 16 August.  Field work will begin mid-morning and will run several hours.  Please contact Scott Smedley (scott.smedley@trincoll.edu), if interested.

Chris Higginbotham posted yesterday on the CSI Facebook page about his doberman causing him to inadvertently hit the “save data” button before categorizing an image.  Human-dog interactions go back a long ways.   As a paper that appeared this spring argues, this human-canine association might originate with wolves scavenging from ancient compost piles.

http://www.bbc.co.uk/news/science-environment-21142870

Nearing the end of the first full week of CSI-Summer 2013, here is the complete listing of team standings (as of 7 a.m. Eastern this morning).  Click on table to enlarge. We have some impressive performances at this point.  We’re still early in the contest, so if your team has yet to make a showing, there is still ample opportunity to be a contender for the prizes.  If you’re having trouble with the qualifying quiz or other questions, please let us help.

Although next week’s network discussion will likely still focus some on procedural issues, we will also begin to consider related ecological concepts and think about applying citizen science in teaching.

Thanks for your contributions to this project.  Based on the comments received, sounds like you’re having some fun.

 

Based on a question from Amy Coan, this post deals with a topic that we can all relate to.

Question

What do we do if we make a mistake?  I made a silly mistake on photo #697.  I think that my finger slipped when I was choosing a number from the drop down menu.  There were very obviously 3 crows in the photo, but we I was hitting the SAVE DATA button, I noticed that for some reason I put down one crow.  Don’t worry- it won’t affect your research- I don’t think that anyone else will make a mistake with that photo!

Anyway, I tried hitting the Back button, but it didn’t look like there was anything that I could do about correcting my mistake.

Please let me know if there is anything that I can do in the future in case this happens again.

Response

I can relate to the situation you’re describing, as I have done that myself.   Once the categorization of a particular image has be entered into the database by hitting “save data,” it cannot be corrected by the submitter.  As you recognize, the 80% agreement threshold (among the five independent categorizations) that we employ should cover for the occasional data entry error of this nature.

As I categorize images, I use the following approach to minimize my data entry errors:

1.)  When the focal image is served, I take a look to get a feel for what is going on.

2.)  I then go to Consecutive Images to clarify what I saw in Step 1 and possibly expose hidden animals (throughout image, be particularly aware of background) by their movement between consecutive images.

3.)  Returning to my focal image (#1 at bottom of frame) in the Consecutive Images), I establish what animals are visible (either completely or partially) in that image.

4.)  Leaving Consecutive Images, I return to the initial page with its pull-down categorization menus.  I enter the relevant species and their counts.

5.)  I then look at the image one last time and account for what I just entered in the menus.   This is my ploy to reduce data entry errors of the type that you describe.  If my menu selections and the image content are in agreement, I “save data.”  If not, I correct my mistake and then submit.

Are you scratching your head about a certain animal in an image that you encountered?  Want some input, so you’ll be ready the next time you see it?  Here’s what you can do to send that image to us.

1.) Each image file has a unique url.  To get to it, just click on the image to enlarge it in the Research Mode (the window with the pull down menus, not the consecutive image window).  For example, that will display following url: internet2.trincoll.edu/CompRes/DisplayImage.aspx?u=http://REINDEER.cc.trincoll.edu/CompostResearch/Actual/R9/MIX_R9S_101012_08-54-51_V_04_10.JPG

2.)  Then delete the first part up to, but not including, the inner “http”, thus leaving http://reindeer.cc.trincoll.edu/CompostResearch/Actual/R9/MIX_R9S_101012_08-54-51_V_04_10.JPG

3.) That url can now be sent in an email or posted on the Facebook page or blog.  By pasting the url into a browser’s address bar and hitting enter, it will display that image.

The actual image in this example was the basis of a question by Chris Higginbotham.

Question

I’ve seen this bird several times, and I just want to make sure I’m identifying it correctly. I’ve been calling it a “red-shouldered hawk”, because it doesn’t really seem to fit the description of any other types, but it seems much lighter in color on the breast and belly than the pictures in the field guide, or for that matter other images I’ve been able to find of this species.

 Response

From the image it is apparent that the bird is indeed a red-shouldered – immature without the darker rusty orange breast of an adult, although a hint of orange is appearing in its plumage. The streaking on the breast is more evenly distributed in the immature red-shouldered than in the immature red-tailed where it is in more of a concentrated belly band.  So this image would be categorized as “red-shouldered hawk, 1”; “crow, 1”.  See the crow tail exiting on right?