Live From Bird Cams Lab: Birds, Feeders, And Science
February 25, 2021[Title Slide: First line of text reads The Cornell Lab of Ornithology with a illustration
of a Yellow-bellied Sap-sucker next to it. Text below is the title that reads Birds,
Feeders, and Science Oh My: What Can We Discover Together? The text below that reads Bird Cams
Lab | Cornell Lab FeederWatch Cam. At the very bottom of the screen is the National
Science Foundation Logo with disclaimer text that reads This material is based upon work
supported by the National Science Foundation under Grant 1713225. Any opinions, findings,
and conclusions or recommendations expressed in this material are those of the author(s)
and do not necessarily reflect the views of the National Science Foundation.]
[Photo: Background is a screenshot of the Cornell FeederWatch cam with several different
bird species feeding on multiple feeders slightly covered in snow.]
[Text: The title of the webinar is over the photo and reads Birds, Feeders, and Science
Oh My! What Can We Discover Together? In the bottom right is the logo for the Bird Cams
project and the National Science Foundation logo.]
[Video: In the upper right hand corner is a video of the speaker, Rachael Mady.]
[Rachael] I want to officially welcome you to the webinar today “Birds, feeders, and
science oh my! What can we discover together?” My name is Rachael Mady and I am the Bird
Cams Lab Project Leader and I am going to be kind of leading us through the webinar
today and keeping us on track. I want to give a warm welcome to everyone that’s here and
also to everyone that’s watching the archived recording because I know many of you couldn’t
be here today, but we are still so thrilled to have you be a part of this community and
with that I would like to turn it over to Ben to introduce himself.
[Ben] Hi everybody my name is Ben Walters I work with the Cornell Lab’s Bird Cams program
as the Communication Specialist. um I’ve been working with the cam since 2016 now I think
and uh it’s been really fun I get to interact with everybody through you know communicating
via social media, making content cutting highlight clips, um you might have talked to me over
email if you email Bird Cams about anything um also I came from a research background
studying predator prey dynamics at the University of Western Ontario where I did my Masters
and I’m happy to be here today talking with everybody about the Cornell Lab FeederWatch
cam and this upcoming investigation
[Charles] I guess I’m the third person here.
[Ben] Yeah go ahead Charles.
[Charles] My name’s Charles Eldermire I’m the Bird Cams Project Leader at the Lab of
Ornithology and I’ve been actually working at the Lab since 2005. Made it over 15 years,
but since 2012 I’ve been focused completely on bird cams and uh as many of you know and
some of you probably don’t know which is one of the things you’re going to find out about
is for the last couple years bird cams has also been you know involved in this other
project that we’re calling Bird Cams Lab that Rachael’s going to give you some more information
- It’s been a really great way to interact with the community in a way that we never
had before so iIm really excited to be here. I’ll be keeping an eye on the Q&A in the chat
if you all start typing again at the same time I can’t read it all. I promise I’ll try
and if you’re asking questions just a reminder throw them in the Q&A because they queue up
for us and we won’t miss them. They don’t go away like the chat does when it scrolls
up really fast um but I will also be looking in the chat um and replying when necessary
so hand things back over to Rachael. I’m really happy to be here today
[Rachael] Thanks Charles. and yeah I am definitely the newer one of the Bird Cams team. So I’ve
only been here for three years or so and I was first a graduate student and now I’m thankful
to be part of the Bird Cams Lab team. So I we’re going to keep doing Bird Cams Lab stuff
as long as we can so with that, I’m going to share my screen with everyone again.
[Slide: A dark-gray background with the title “Roadmap” An icon of a map and a pin is shown
on the left with a numbered list on the right hand slide that reads 1. Overview of Bird
Cams Lab 2. Cornell Feeders Live 3. Wonder Board Highlights 4. Discussion. The speaker’s
video is in the top right corner.]
[Rachael] Because I want to orient you to the rest of this hour. If you ever see me
turn to the side, I’m keeping track with my notes because I want to make sure we cover
everything that we hope to today. And we’re first going to just give a brief overview
what Bird Cams Lab is many of you may be familiar may be familiar with bird cams but maybe not
as much as Bird Cams Lab. We probably have some newer people in today, and then I want
to dive into the specific investigation that we’re focusing in on which is a part of Bird
Cams Lab and that’s called Cornell Feeders Live, then I’m going to talk about the highlights
from discussions we’ve been having on the Wonder Board and if you don’t know what the
Wonder Board is you will soon. And then we’re going to open up the last half hour or so
for a discussion so that’s when we’re going to answer your questions in the Q&A and we’re
going to be addressing what you’re bringing up in the chat. So with that.
[Slide: Screenshot of the Bird Cams Lab website. At the top is the website title, “Bird Cams
Lab.” In the middle is a picture of a Red-tailed Hawk flying and facing the viewer against
a grey blurred-out background. Over top the picture is the text “Welcome to Bird Cams
Lab.” At the top there is also a navigation bar. The speaker’s video is in the top right
corner.]
[Rachael] Bird Cams lab so for Bird Cams lab is a part of bird cams and we initially created
Bird Cams Lab here at the Lab, Cornell lab of Ornithology, because we saw people watching
the Bird Cams, asking questions, collecting data, and we realized there needed to be a
space for this curiosity and this um ambition to collect data with the cams that take place
and we wanted to give people the opportunity to work with scientists, to co-create together
investigations on the cams, and so what we did was we created Bird Cams Lab, which you
can see here in a screenshot of the website and that allows people to jump into any part
of the scientific process that they want to.
[Slide: Now the screenshot is faded and there are give gray boxes going across the top each
with a word in them: Observe, Question, Collect, Analyze, Share. Speaker’s video is in the
top right-hand corner.]
[Rachael] So people can come into, like you tuning in today or watching the archive recording,
you can observe and ask questions with us, help us figure out what questions we want
to answer, collect the data, and then go through the rest of the process you can jump in at
the analyze and visualize stage of the data we collected, or help us share it out with
the rest of the community, and all of these parts are important but you can jump in and
out based on your interests. And one of the biggest tenants of Bird Cams Lab is that you
can come to this space with no prior knowledge, no prior experience of birds or science, but
we work together and we learn together, and we have created supports to help you and us
get through this process together and make discoveries on the cams. For now, so Bird
Cams Lab, we work through each of these steps in the scientific process for each investigation
and right now for Cornell Feeders Live we are in the observe and question phase.
[Slide: The last three gray boxes that each separately read, “Collect, Analyze, and Share,”
are faded and the words in the first two gray boxes (Observe, Question) are outlined in
red. The video of the speaker is in the top right-hand corner.]
[Rachael] So we’re starting here and next is data collection, but before we can collect
data we have to figure out what question we’re going to ask, and we’ll talk a little bit
more about that later in the hour. But for many of you who may be familiar with the Cornell
FeederWatch cam this is what you see.
[Slide: A screenshot of the Cornell FeederWatch cam with the title at the top “Cornell FeederWatch
cam” over it. There is a male Pileated Woodpecker in the middle (a black, white, and red bird)
and there is a male Northern Cardinal (small red bird with an orange beak) perched below
on a platform looking up at the Pileated Woodpecker. Behind the birds and the feeders they are
on is a frozen lake and some trees. The speaker’s video is still in the top right-hand corner.]
[Rachael] And for those of you who may not be familiar it’s a cam that’s located at the
Cornell Lab of Ornithology. And before I turn it over to Ben to kind of orient us to this
cam I want to put out a poll to see how many of you, how often you watch this cam. So a
poll should come up on your screen now and so how often do you watch this cam. Do you
watch it daily, have you never watched it? And Charles and Ben you are able to vote this
time if you would like to previously on webinars only people tuning in could vote but I have
allowed it so we all can vote. And you guys are quick I see in the chat that Mark it looks
like you watch a few times a week yes that is also that’s not an option but good to know.
I’ll let a couple more people put in their votes before I end the poll. Alright I’m gonna
go ahead and end the poll. It’s okay if you didn’t put in your vote. No worries, but I
want to keep us moving along. I’ll share out the results so you all can see what I’m seeing.
[Slide: Over the current slide is the poll results, which is a question followed by a
bar of color that corresponds to the number of people who selected each option, with the
bar being red for the one with the most people. The question is at the top, “How often do
you watch the Cornell FeederWatch cam?” The answer options and the percentage of responses
for each is shown below. 10% (18 people) for I’ve never watched it before, 16% (30 people)
for Once or twice, 16% (30 people) for Few times a year, 29% (53 people) for Few times
a month, 15% (28 people) for Nearly every day, 7% (12 people) for Daily, and 7% (13
people) for Multiple times a day.]
[Rachael] And it looks like we have everyone here, people who watch it multiple times a
day too you’ve never watched a cam before but you are here and we are so happy that
you are. And so welcome everyone! And to make sure we are all on that same page I’m going
to stop sharing those results and I’m going to turn it over to Ben so that he can kind
of orient us to what this cam is, where it is, and what we should know.
[Slide: The poll results go away and the speaker video is now of Ben. The speaker video is
still in the top right-hand corner.]
[Ben] Thanks Rachael. So um just like Rachael mentioned uh the Cornell FeederWatch cam is
aptly named because it’s at the it’s located right next to the Cornell Lab of Ornithology.
It’s actually located in a bird feeding garden that’s also sort of an observation garden
for people that are visiting the labs visitor center. So as you can see in in the picture
there is a pond in the background, it’s a very beautiful scene. Um we have about a 10
acre pond and surrounding that pond is what’s called Sapsucker Woods it’s a little uh bird
sanctuary area that attracts all sorts of forest residents and many of them come by
our bird feeders many of the birds. It’s a really beautiful camera, we just recently
installed a couple years ago at 4k cam so we get crystal clear images of the birds as
they arrive at the feeder and forage on the various offerings that we provide. One thing
we try to do is offer a pretty diverse set of both feeders and food types to attract
a lot of different species of birds. As you may know some species or some food types and
some feeders accommodate all sorts of birds whereas others are more specialized to attract
certain types of birds. So we try to offer a mix of those to get you know woodpeckers,
anything from you know goldfinches, we have cardinals as you can see in the picture, but
really all sorts of birds. We even get some fun migrants stopping by come spring, like
Baltimore orioles they come by and feed on the oranges that we offer. So it’s a really
great cam it’s also a really opportun a really great opportunity to be able to learn more
about the birds that we see through an investigation and I’m really excited to see what type of
things people are interested in um sort of learning about the birds that are visiting
this feeder because we have a lot of people watching a lot of people asking questions
all the time and now’s our chance to actually do a little bit of a deeper dive into the
birds that we watch every day. So it’s a really fun opportunity
[Rachael] Thanks for that sims so that’s the cam and many of you may already be familiar
with the investigation title, but we have called the investigation with this cam Cornell
Feeders Live. And you may be wondering why we called it that and that’s because we’re
going to be collecting data live, in real time, and we’ll be able to watch the cam and
collect data on the side and it’ll look something like this.
[Slide: A screenshot of the live data tagging tool to collect data for a previous investigation
with the Panama Fruit Feeder cam. At the top is the title “The Data Too.” In the middle
is a view of the cam with greenery in the background and a feeding table in the middle
accompanied by 2 hummingbird feeders. On the right hand side are 6 buttons with pictures
of bird species and one button with bananas. below the view of the cam in the middle are
two buttons that read “Field Guide” and “Tutorial.” Beside the two buttons is a log of observations
with times and the name of the button that was clicked. The speaker video is still in
the top right-hand corner.]
[Rachael] This is the data tool that we used for a previous investigation, with Panama
Live, and you can see that the view of the cam is right here, and we’re going to be able
to click buttons here on the right hand side to collect data from this cam, and so we decided
to call it Cornell Feeders Live. But before we can collect data, like I said in this investigation
we’re at the observe in question asking phase.
[Slide: A dark gray background with a screenshot in the middle of the Wonder Board webpage.
The title of the webpage is “Cornell Feeders Live Wonder Board.” Then there are two columns
of text. The column on the left has introductory text to what the board is and a discussion
feed with a place to enter a comment. The column on the right has a video feed of the
FeederWatch cam, a welcome message, some resources, and a twitter update field.]
[Rachael] And so right here you see a screenshot of what the Wonder Board looks like. And this
is on our Bird Cams Lab website right now, and people have come to this Wonder Board
forum and have posted down here in the “join discussion,” and have posted their thoughts,
questions, and we’ve started to narrow in on what we can actually investigate on this
cam. You can see the cam in the top right hand corner, so you can observe and ask questions
just about the same time which is how some of the best research questions are asked,
is when you watch things and you think about what you are curious and want to know more
about. And so the conversations on this wonder board for the past, we’ve done it for about
a week now. It’s been live and we’ve had more than 30 people post and we’ve had many more
vote where they can actually vote on the ones and up vote them so that the ones that are
more favored rise to the top. And we have run the gamut in terms of what people are
interested in learning more about, But I do want to share with everyone today the four
big buckets that appeared through all the question asking and wondering aloud.
[Slide: The screenshot of the Wonder Board is faded and is moved to the left-hand side.
Against a dark-gray background is white text that reads “Themes’ and then under that a
numbered list from 1 to 4 that reads “Arrival Patterns, Weather effects, Food preference,
Interactions. The speaker video is still in the top-right corner.]
[Rachael] The first being arrival patterns, where many people posted on the Wonder Board
and are interested. You may be interested tuning in today wondering about how birds
are arriving at the feeder and whether there’s some sort of schedule and this is actually
the number one topic in terms of what people were most interested in voting on and on the
Wonder Board. Next we have weather effects. So that also runs the gambit in terms of what
is weather because weather, depending on where you are in the world, could be anything from
a lot of snow to a lot of rain and anything in between, and there’s been a lot of great
discussion on what we mean when we say weather and thinking about how those different factors
can relate to what’s going on on the feeder. And then we have this idea of food preference
like Ben said there are a lot of different food types that we put on the feeder and we
do have the ability to manipulate them and to actually be pretty intentional with how
we put the food out. And we could maybe dig into how different species prefer different
types of food on the feeder as well. And then one that always seems to come up and almost
all the investigations we’ve done with feeders it’s, people have brought up or we’ve actually
investigated interactions and there’s been some interest as well in this investigation
in the question-asking phase that we’re in right now and understanding the frequency
of interactions and does that change throughout the day, um do interactions between birds
change in response to weather, in response to the food, and so that’s another area of
interest from people on the Wonder Board. And there are all possibilities for an actual
investigation here and something we’ve um started to talk about on the Wonder Board
and we can talk more about today is that these topics are all pretty feasible, but then we
start narrowing in on how we actually investigate it we start to have to think about what we
mean by the words we’re using, and how we’re actually going to collect the data when we
have the tool that we do, which is clicking buttons while we watch the cam. And I am curious
before we get into a discussion, um I am noticing I am coming through scratchy and I apologize
to all. This is the first time I’ve worn a scarf for a webinar and so I’m going to take
off subscribe and maybe the feedback will go away if it doesn’t then please let me know
um so hopefully I’m coming through clearer. And before we get into a discussion about
these themes and topics and about what’s coming to mind for you all, I want to launch a second
poll and I want to actually see how many of you have participated on this wonder board
in this investigation so far just to get a sense of who is here today. Alright looks
like we got about 93 of you have voted so I’ll end the poll, no worries if you didn’t
have a chance to vote.
[Slide: Over the current slide is the poll results, which is a question followed by a
bar of color that corresponds to the number of people who selected each option, with the
bar being red for the one with the most people. The title at the top is “Have you participated
in Cornell Feeders Live?” and the responses are 46 people (22%) for Yes, 148 people (71%)
for No, and 13 people (6%) for I’m not sure.]
[Rachael] It looks like we have everyone here again too we have many of you here today have
participated on the Wonder Board many of you have not, and many of you aren’t quite sure,
which is completely fine because the way the way the world is right now sometimes it can
be hard to keep track of things. So no worries, but this is Bird Cams Lab where we welcome
everyone to come participate at any point and so we’re so glad that you’re here today.
We invite you to ask questions in the Q&A and to share your thoughts in the chat. and
so I’ll stop sharing those results and I will open the floor. So Charles, if there’s anyone
asking questions in the Q&A or we can maybe touch on some of the questions that were submitted
ahead of time we can start digging into some of these themes or some other thoughts that
are coming from the community.
[Screen: No more slide is shown, instead it is the videos of the panel. Rachael is in
the top left corner, Charles is in the top right corner, and Ben is in the middle below.]
[Charles] Yeah that’s great! I think that um everyone’s been doing a good job of listening
and jotting a few things here and in the background I’ve just answered some questions about technical
things about the camera, since it’s not really a technical webinar, we didn’t really plan
to answer your. Feel free to keep asking them if you’re interested in those technical more
technical questions just email us at birdcams@cornell.edu and we have a whole email that kind of walks
through the general way that we do cameras that’ll be way more interesting to read than
to hear me um recite to you, that’s pretty dry stuff. There are actually a couple questions
that focus on Bird Cams Lab that have flowed into the Q&A um and one of the questions uh
well two of them actually are are focused on what does arrival even mean what’s an arrival
pattern um does arrival mean how they arrive? okay so we have one person saying, Stephen,
what’s an arrival pattern? Then Martha, Does arrive mean how they arrive or whether, rather
when they arrive, like time of day? Presence of other birds, etc? So um maybe you might
want to just take a moment to either touch on what those broad, maybe an example or two
from each of those categories, but starting with arrival.
[Rachael] Definitely I think thank you for those wonderful questions that’s what i’m
talking about when we we start with somewhere and then we kind of narrow in on exactly what
we mean and I’ll just share some, so I said arrival patterns you pointed out arrival what
what does that mean some of the questions that people have been posting in the Wonder
Board fall under um some of these that I’ve summarized here on my other screen, so I’ll
share some of those. So many people wonder about which species arrive the earliest so
when they say arrive it’s land on the feeder so when you’re watching the feeder, who’s
the first one there? and then at the end of the day who’s the last one there? there’s
been quite a few people and I think a teacher also commented on the Wonder Board that her
classroom is interested in knowing the answer to that question. And then there’s also people
who have wondered about visitation patterns to the feeder throughout the day. And so for
a particular species they’re curious when do you see them land on the feeder and how
does that go through the day. And we can talk about what we mean by land on the feeder do
we mean on any of the feeders in view, do we meet on the platform, and that would be
something we want to talk about and make sure it’s clear and whatever data protocol we we
decide on. And then there’s also been people who just want to know when the feeder is the
busiest so they don’t really care exactly like what one species is doing, but they’re
just like when in the day are there a ton of species and completely understand that
because when you come to watch the feeder you want to see birds so it’d be really interesting
to know when to hone in on um in terms of visitation. So in previous investigations
we’ve narrowed down arrival as literally a bird landing on the feeder before um but that’s
something to talk about that’s definitely not something we have to do it could be we’re
coming in view of the feeder because we’ve done that before. So great great questions.
[Charles] And there’s also a bunch of people hopping in right now to say, “wait where is
the Wonder Board?” So I just put the link in the chat along with a link to sign up for
this investigation, and just a link to the general Bird Cams Lab website. But an interesting
thing, so just look for those in your chat folks um rather than me reading the http and
so on to you, um one question that I think is interesting that is worth spending a moment
or two on. Alicia says I’m a little nervous about doing a live study because I’m afraid
i might miss something. And she also asks uh is the study limited to six bird types?
[Ben] Yeah I can sort of touch on that one that’s a great question. First off we don’t
want anybody to be nervous about participating even though it isn’t a live environment, the
idea is that during the data collection process we’re hoping to have multiple people watching
the cam at any point in time. Right so if by chance you do there’s a lot of birds arriving
or you know for depending on what we’re collecting data on if a lot of stuff is happening, if
you missed something that’s okay because we’ve got other people hopefully that are going
to be there also collecting data and we can collate all of that data together and figure
out based on the observations of multiple people um observing the same time period,
get we get a better idea of what’s actually happening on the theater during that time
versus if it’s just a single person um you know going through and trying to identify
every bird that’s hopping um on the feeder if there’s you know a ton of action going
- So that’s the idea it’s crowd it’s a crowd-sourced type of data collection where we’re not relying
on any single one person to do all the heavy lifting that’s the great part about it as
as this you know Bird Cams Lab community is that we rely on each other to not only come
up with the questions, but also collect the data together. And um on to the second part
of the question are we limited to six types of birds? Not necessarily uh limited to six
types of birds but we are limited with the amount of of buttons that we have on the data
collection window. So Rachael showed um an image of the of the um an example of sort
of what live data collection looks like, and there was six buttons of different bird species
as well as a couple extra buttons that um also allowed us to tag other types of information
um now the information doesn’t have to be a species in particular it could be you know
a seed type, um it could be you know just a general characteristic of a species like
a large bird, medium-sized bird, small bird. It depends on what our question is that we’re
asking and then we can decide what buttons to use best use for to answer that question.
Um so you know if anybody else wants to add anything but that’s that’s typically what
we’re looking at and it’s an important thing to remember when we’re thinking about the
things that are actually plausible to investigate right because we are limited to these, um
you know six to eight number of buttons um in order to collect data with. So it’s harder
to do more sort of involved data collection where we’re we’re you know collecting a lot
of different types of information because we are collecting data live number one and
we have a limitation on the number of buttons.
[Charles] So surprisingly as you’re talking even more questions rolled in um and and lots
of uh some good things in the chat too. So just a reminder to anybody who might be coming
in new if you if you have questions try and stick them into the Q&A box rather than the
chat just because the the chat can sometimes move them out of view and it’s likely we’re
going to get more questions than we’ll be able to answer, so if we don’t get to your
question it’s not because we don’t like your question or we don’t like you. It’s just there’s
only so much time, but there’s um a few different we’re going to try and get through these kind
of thematically okay, and and um a few people are asking about um how long data collection
will last? and when it will be? And so maybe maybe we’ll jump into there and then talk
a little bit about how um I have a few questions sort of about how we support the investigation
as it goes so how long is this investigation actually happening for?
[Rachael] Great question. So we are hoping to do this investigation in the beginning
weeks of March that’s when we’re aiming to collect the data so after we decide on what
data we’ll collect we’ll need some time to actually make the buttons look like the data
we want to collect to make sure everything’s hooked up on the back end. And so we’re hoping
to do it for a couple weeks in March because of the timeline that we’re hoping for the
whole investigation because while we love doing this, and while we wish we could do
research projects for as long as we want, we have a timeline we’re trying to adhere
to because we’re funded by National Science Foundation grant and so the timeline we’ve
planned out for allows for data collection for about two weeks or so in March. And so
that will be the time where we watch it live and we click the buttons to collect the data
and when it starts is probably going to shift around a bit depending on the back end of
making sure all the buttons look like the way we want them to, and the supports like
a field guide or a tutorial are all laid out for everyone so that’s what we’re hoping for.
[Charles] Great and so we have sort of how long we’re shooting for now what’s the deal
with voting we had a question early on that we came we were talking about the questions
and the question was, “How do we vote on the wonder board and maybe if you wanted to spend
just a moment kind of clarifying the process we’re in right now and voting in general.
[Rachael] That is a great question so there’s kind of two different types of voting that
has happened in previous investigations and so the first one is upvoting and so when you’re
on the Wonder Board when you’re using the Disqus tool which is a tool we’ve embedded
into that webpage so that we can talk to each other there’s a little carrot underneath someone’s
comment on that board and you can click that that allows you to upvote something and then
there’s a way to on the board to filter based on best and then you can filter to see the
ones that have the most upvotes and so I can actually if where i can actually show everyone
what I mean by that because I think many of you today are not familiar with the Wonder
Board which is completely fine so let me so i’m just talking in the abstract share my
screen so you all can see exactly what i mean okay so i am going to share the Wonder Board
with you all is that coming through okay okay so this is the Wonder Board and if you go
down here and i’ll zoom in a little bit you can join the discussion by typing in here
once you have an account and then if i like something that someone says if i liked what
Alicia says i can click on this arrow and i can upvote it and then if i want to see
the board in terms of what other people have uploaded i can sort it by best and then that’s
how things kind of rise in the conversation and so this one had 14 so it rose to the top
and that’s the first type of voting the second type of voting has to do with actually selecting
the research question and so that’s when we use a different tool and we ask everyone to
vote and we actually select the research question or the topic depending on where we are and
narrowing that down and that’s when we start to make the decisions about data collection
so in previous investigations people have voted on the question itself on which species
we’re going to collect data on on the topic that they’re most interested in in terms of
like extra data to collect to do a sub question so those are the two different layers of voting
that you can participate in
[Charles] Perfect um people are very inquisitive today it’s great and thank you everyone out
there for asking questions um and there are a lot of them uh so I think one one sort of
a general category of question that we’re getting from a lot of different people involves
the idea that gosh if there’s a bunch of people taking data at the same time how do you tell
what’s going on so can you analyze it to avoid multiple sightings of the same bird is there
a minimum time or do I have to sign up for a shift um maybe you could one of you could
address kind of um sort of the philosophy behind this uh and then we’ll back into the
question after you’re done of why is it important to do this live so keep that in the back of
your head.
[Rachael] Ben, do you want to start or I can start.
[Ben] Sure I mean I can start um and then I might kick it over to you in terms of how
we analyze sort of all the raw data that’s coming in um so number one you don’t have
to sign up for shifts um there’s there’s no sort of minimal obligation that anybody has
in terms of if they want to participate it can be for uh you know they can share their
observations from a minute of the camera during our data collection period or they can watch
for hours at a time and and share what they see so there’s there’s no minimum or maximum
obligation that anyone has to uphold in order to be a participant. Um and like we said like
I said before you don’t have to worry about actually catching every single little thing
that you want that you see on camera. It’s great if you do um but don’t feel um you know
like you’ve you haven’t done a good job if you’ve missed that one cardinal that came
with the feeder um or you didn’t see what species it was that that stopped by the sunflower
seeds um because we the idea is that we have a bunch of people watching at one time, and
by having that crowd watching and all taking data we can sort of hone in on on on the story
that’s being told at the feeder during that time through everyone’s observations and um
I’ll kick it back over to you Rachael to share a little about how the how the data once we
pull the data how that works.
[Rachael] Yeah so that’s pulling the data is actually really exciting because it’s this
is something new and it’s not something that anyone has really dealt with before is like
what do you do with data collected in a live environment by multiple people and so this
has been something that we didn’t expect to be something that was an interesting challenge,
but became one and so far what we’ve done is we kind we collapse people’s observations
to, depending on the question and depending on the biology of the birds and depending
on what data we’re collecting, it will shift this little window, but the window of one
minute is something we’ve used for some previous investigations where if people say that something
happened within one minute of each other it’s the same observation and when we’re thinking
about birds being present on a feeder it seems like that mirrors what’s actually happening,
but because what happens is you have kind of these little peaks and observations and
sometimes there’s one person watching many times there’s like two people watching, and
so it’s interesting to see when they agree with each other or don’t agree with each other,
and there’s still room to explore in that area so we’ve come up with what we think is
the best practice for using it for Bird Cams Lab but there’s definitely room to grow and
definitely things to think about in terms of honing in on the best way to do it, but
it’s very interesting and in terms of why you would collect data live it, I mean there
are live cams all over the world and there are things that are happening on those cams
that people are picking up on and it’s an opportunity to really kind of tap into something
that’s already going on and engaging everyone that’s already watching and so that’s the
reason why we tried to surface Bird CamsLab and give the opportunity for this live data
collection
[Ben] Yeah sorry Charles I just wanted to add one thing about the the live environment
like the the draw of watching something live is something that you don’t get when you’re
watching something that’s you know pre-recorded right something if you’re watching a live
sporting event it’s not the same as if you’re watching a replay of the super bowl that happened
right um and that’s that’s what we see we see thousands and thousands and thousands
of people are watching live cameras and that that provides an opportunity to collect data
about birds about anything really, but in our case about birds. And so having uh tools
available for people to easily collect data while they’re watching if that’s something
they’re interested in is a you know could be a huge resource to learn more about all
of these species about all the things that you know people are passively watching at
the moment but they could be collecting data and so developing those tools is a really
important thing to get off the ground.
[Charles] Yeah and just to piggyback on that and to answer um a question a couple different
versions of this question have come up in the Q&A tool and Nancy wrote how can you get
accurate data if there are gaps in observations and this is something that I hear a lot from
people because, especially from people who aren’t necessarily scientists or haven’t had
to go gather data to try and solve a puzzle, especially in the biological world. And most
of what we know about the world there are giant gaps and what we do is we take lots
of samples of something and then use those samples to try and figure out what reality
is telling us right so it’s kind of like the the model I like to think of in my head is
a dot-to-dot picture so each one of our samples is a little dot on a page and the more dots
you get the more richness whatever it is that that is illustrating um it sort of like illustrates
right so the cool thing about live cams or recordings of live cams um is that we have
like every moment on that camera right but we don’t need every moment on that camera
to actually approximate reality all right so while censusing you know what’s happening
there is a is a neat goal it’s not actually the goal of data collection to get every single
instance but it’s it’s rather to um in part encourage people to participate because it’s
not just us learning from the data you’re collecting you’re learning from collecting
data right. It’s a two-way street that we’re hoping for through this project so that’s
my little uh that’s that’s my rant on sampling, which I really enjoy giving you can ask these
two they’ve probably heard me give it um a long time ago and recently but we’ve had a
couple of people actually ask about Zooniverse, and whether or not they need a Zooniverse
login or they just came from Zooniverse and does the tool work the same way? And I wonder
if you might just do a short brief for the people that don’t know what Zooniverse is
up to is this the same thing can they do they need this Zooniverse id to do this.
[Rachael] Welcome everyone from Zooniverse I’m so glad that you are with us today um
great questions and especially because we just finished collecting data for an investigation
on Zooniverse very relevant. This tool and this investigation is different than Zooniverse
so what we did for the last investigation, some of you may be familiar was Battling Birds
and we collected data on archive video clips and this data collection we needed to have
a place where we could document some detailed information about each clip and so this one
we are going to be live and this is going to be on our bird cams website, and so this
one will be separate from Zooniverse and we’re not going to use Zooniverse as the tool, but
we have for past investigations and oh we could always go back to Zooniverse if in the
future it makes sense to collect data there. But for this one we’re going to be right where
we have been before with Bird Cams and collect data live over on our platform. So you don’t
need to create a Zooniverse login the only login you need to create to participate on
the Wonder Board is a Disqus so you can see all the comments on there, but you can’t actually
post or interact unless you create a login it’s free. And then for data collection will
ask you to create a what’s called a Cass login and that has to do with the Cornell Lab of
Ornithology and we’ll have there are resources that walk you through how to do that and that’s
also free as well.
[Charles] So one thing that several people have asked about is actually wait a second
I don’t actually know the birds that well at your feeder do you guys have any resources
how can i learn where can I learn how am i going to be able to know that I’m doing a
good job of data collection i don’t know ben if you want to talk at all about what we have
available for them sure yeah so one thing that the Cornell Lab has is a ton of online
resources um not only will we probably provide some sort of tutorial to help you get associated
with the things we’re looking forward to collecting in this project um so so we’ll have something
there specific to you know the question we eventually decide on that will help lead you
through the type of delay data we’re looking towards collecting um but if there’s a bird
for example that you don’t know on our on our website our our cam site all about birds.org
cams if you uh navigate to any of one of our feeder cam pages at the bottom of the page
it’ll be a species info tab and that species info tab will contain a picture of every single
species that we’ve actually seen visit that feeder up to this point that we’re aware of
um the cam goes live 24 7 so there might have been a a one-off rare species that’s visited
that might have slipped through the cracks but all that we’re aware of and so you can
click on those birds to direct you to resources on our online field guides or through ebird
that can give you all sorts of information about how to id these birds give you a little
bit more information on their natural history their range you could go as deep as you want
to
[Charles] So a couple of questions that actually focus on what we hope to learn from doing
this what what do we um what what’s interesting about it in terms of what we what a community
might find and um is this something we’re going to publish in a peer-reviewed journal
so sort of like importance of the research interestingness of the research maybe you
could um touch on those themes.
[Rachael] Yeah great questions and something that we should always be thinking about when
we take on research is what what’s the point and it’s usually the question we get it’s
also what’s the point so there’s kind of two sides of a coin for Bird Cams Lab one side
is the actual research we’re doing with the cams and um we do hope to actually answer
a question that’s interesting and that is relevant and that we can share out with the
scientific community and that’s part of the way we encourage our community to walk through
and choose which question we do so for instance weather is really really interesting but there
has been quite a bit of research um about it in terms of the effects of temperature
and in terms of the effects of wind so maybe if we do weather we don’t focus on those variables
and we focus on a different one or maybe we focus on arrival patterns for species that
we don’t know a lot about and so that’s kind of figuring out what’s interesting in terms
of previous research in terms of this we are using this on one cam and so in terms of publishable
sometimes what we do can influence the next study after us and so for instance we did
a study with hawk happenings where we learned something really cool about the behavior of
the birds on the cam but we did realize that it’s one cam it’s one family over one season
and so we can we it’s kind of a starting point the pilot study as some would call for um
scientific research if anyone out there has ever done any research they may be familiar
with the idea of a pilot study where you try things out and you see what works and then
you go to the next step where you kind of expand it and so that’s what this investigation
could do so that’s in terms of the biological side which is pretty exciting the other side
has to do with Bird Cams Lab as a whole and so we started this project because we wanted
to try out this idea of co-creation where we all work together professional and non-professional
scientists work together to create something and go through the whole scientific process
together and at the end of the day how does that affect what we learn in terms of about
science and knowledge and so there’s this overarching goal of Bird Cams Lab as well
because we’re funded by the national science foundation to understand informal learning
in everyone that’s a part of this so there’s kind of two sides of the research coin and
we do hope to share that with the community both in potentially published scientifically
peer-reviewed papers as well as other mediums as well so social media posts online and other
ways that we can disseminate the information.
[Charles] Those are all good points and it brings I think a couple people back to the
question so I had a nice little conversation in the Q&A with Nancy about like wait if you’re
not sampling all the time what about the question you know like if nobody’s taking data at sunset
nobody’s watching the camera and your questions about sunset and it raises a valid point how
do we get to that question and somebody else asked about voting again for the question
I thought it might be worth bringing up again maybe how do we get to the point where uh
the community has a question that will that that their collection their data collection
efforts will then be focused on.
[Ben] sure um I can take a first step at that and ratio you can add anything if you want
so so that’s a really good question so we’re talking about this investigation we’re talking
about the Wonder Board but how do we get from what we’re talking about on the Wonder Board
to actually finding out what we agree upon in terms of what we’re what we’re honing in
on for this next live investigation on the Cornell feeders so first um you know we definitely
invite all of you to continue the discussion on the Wonder Board because that’s where it
all starts we’ve already outlined some of the um the main you know interesting topics
that have come up in the community so far and we’re excited to continue to talk about
those and and how we might um number one if they’re feasible given the tools we have and
number two how we might refine a question into something that we can actually start
collecting data on with the live tools so once um you know the the conversation continues
on the wonder board and we start to get these like themes that are popping up that we see
are sort of the main interest what we’ll do is we’ll start honing those honing in on those
themes a little bit more into something that’s more testable and then we’ll allow the community
to vote on those various themes so we’ll send out an email to the community telling you
it’s time to vote on the next question for the Cornell Live investigation and you’ll
have your opportunity to you know put in your two cents on which question you’re most interested
in um investigating next on the feeders so once that happens once we have our question
in hand we’ll build out the tool for the live investigation we’ll get all of our buttons
ready and then we’ll be in communication with you to eventually start the data collection
process
[Rachael] yeah that was a great way to kind of walk through that process Ben and I’ll
point out that depending on the investigation it’s kind of because it’s a co-created space
it’s a bit of give and take and so in the last one we were working with a researcher
Dr. Miller we kind of had the research question we were filling in all the holes in terms
of like what data did we need and for this one we’re all really working together we don’t
have that question set out so that’s why we have kind of these buckets of themes that
have arisen from the discussions and now it needs to be continued back and forth to figure
out the actual details and so depending on how the rest of today goes and how the wonder
board goes with you all will kind of determine what direction we go whether we need to take
a vote on the topic and then from there think about the details or if we’re really we do
seem to be going in towards a topic and we can already vote on the details in terms of
like what wording we need or what specific type of data we need so like you mentioned
Nancy for the arrival like are we gonna focus on the people the birds that come in the morning
and evening and if we do we have to make sure we’re watching then or are we instead going
to look at a pattern throughout the day and maybe it doesn’t matter as much that we get
those exact times and so it’s a great point the question is really going to determine
how we collect the data because we can’t they’re very much connected.
[Charles] And can you remind people uh one more time in the timeline how much longer
you you expect the wonder board will be going and when a vote might be coming
[Rachael] Yes so hoping to have the Wonder Board going for about a week more and as we
prep to get the recording out to people and we talk on the wonder we’re getting a vote
out next week and so depending on how discussion goes whether we need to create a design board
which we’ve done in other investigations which continues to help us narrow or if we can stay
on the Wonder Board that’ll be something we figure out over the next couple days but then
next week hoping to do a vote because we want to start collecting data and we know you do
too so that is the timeline we’re hoping for which some may argue is like I just got here
like can I contribute you really can you really really can like regardless of where you get
to jump in you have value in this.
[Charles] Awesome thank you and it’s I think people are excited about the opportunity um
and a few people have actually asked they’ve come back to well why why are you guys doing
this cam? You guys have a lot of feeder cams um and this wonder board is just focused on
this feeder cam why did you guys choose this one?
[Ben] um well number one um i think the Cornell Lab FeederWatch cam is is really interesting
because it is our most viewed camera so a lot of people are sort of attracted to the
setup and the camera and they they love watching the scenery there so there’s already a large
audience that we can pull from and help us collect data throughout the data collection
period on the live cam also we’re just interested in um you know based on the on the different
seasons that we go through with our projects um you know some cams might be more pertinent
for a certain timeline than others um right now the Cornell Lab FeederWatch cam is bustling
with all sorts of birds there’s a lot of activity um you know we’re going to be looking for
some spring migrants coming in soon to add to the diversity of the feeders and we’re
also interested in comparing sort of the um the live annotation process on various different
cameras so we’ve already done a live investigation on our panama fruit feeder cam a live investigation
during the nesting period on our Cornell hawks cam and we just want to see what that experience
is like on a different cam that also has a really high viewership
[Charles] Great a couple people have pointed out this seems like it’d be a great uh opportunity
for um artificial intelligence or uh machine learning um to also collect data i don’t know
if either of you um have anything you want to share about that.
[Rachael] Sure i’ll just touch on that there is definitely talk of that and there are people
at the lab that are thinking about those things right now we’re focusing on working with people
in our community to collect the data since they’re already doing it on their own and
they already want to investigate the things that they are interested in and so i definitely
think there’s a place for AI and there is a lot of talk of implementing that potentially
with Bird cams Lab potentially with cams in the world and so that’s kind of i think a
next step um but Bird cams Lab right now is focusing on the community aspect of investigations
together so AI is definitely super super cool but for now we’ll focus on working together
to collect the data
[Charles] I think another another way to frame it is AI artificial intelligence might be
really good at maybe categorizing birds um but it may not be very good at developing
interesting questions or analyzing results and um you know one of the goals of this whole
project is to potentially provide new avenues for anybody anywhere with a computer really
to be able to engage in the scientific process and learn and so so while like machine learning
probably has a role to play in pure data collection at this point um thankfully um our brains
are are better uh at generating questions and doing the rest of the scientific process
right now very true very true so we’re we’re getting through most of the bird cam’s lab
questions and what a flurry from everybody out there uh we do have some other questions
just about the feeder set up we have about seven minutes left and i wonder if we could
take the next couple minutes and just rifle through a few of them before hitting up the
end and we got a few questions and I’ll give you two two big ones squirrels why don’t we
see more squirrels on the feeders and related what are you feeding the birds there because
we get all these birds we don’t get starlings that’s a third part why don’t you guys have
starlings you got starlings and squirrels why don’t we have them and what are you feeding
them.
[Ben] So I can um take a stab at answering that because I am the person that manages
the feeder on a day to day basis so I’m there on site every day and i get to see what goes
on around the feeder and help sort of um keep the you know work towards keeping the birds
on the feeder and the rest of the squirrels and stuff off let me start by saying we do
get squirrels and we do get starlings it’s almost impossible to keep um squirrels off
of the feeder unless you have a feeder that is like a good at least 10 to 12 feet from
any jumping platform um and while we do do trimmings and stuff of the vegetation um around
the feeder and we try to place it in a really good spot that allows um enough space away
from jumping platforms scrolls are very industrious creatures they will get on the feeder if there
is you know any space available for them too um we also use baffles underneath our pole
system that supports the feeder that helps keep the the squirrels from climbing up the
feeders from underneath um and if anybody’s interested at looking at a picture of the
setup sort of from a different angle you can email us at birdcams@cornell.edu and i can
send you a photo so number one baffles providing enough distance away from any jumping platforms
um they recommend 10 to 12 feet but i would go even further if you have the opportunity
to i realize that not everybody has enough space to provide um that much room for their
bird feeder um i know that you know i in my backyard i’ve had trouble with squirrels um
even just dropping down from the treetops mostly than anything else um but uh starlings
on the other hand um there are certain things you can do to keep starlings from visiting
your feeders uh it’s not really an option for us because we have a big platform and
that’s going to invite pretty much any type of species to come visit including starlings
now some of the things you can do to sort of dissuade starlings from coming into your
feeder is by using feeder types that that um aren’t necessarily good for attracting
starlings so if you have a hanging feeder for example with a small perch and a small
feeding port the straws aren’t going to be able to land on there and actually feed out
of that small port that’s you know going to restrict larger birds from visiting sterling’s
also really like suet but they have a hard time accessing suit if it’s an upside down
suit feeder so if you have a feeder that hangs down with the cage at the at the the opening
of the bottom um is well the suits only open at the bottom that will allow birds that are
used to like hitching up tree trunks like nut hatches and woodpeckers to access it but
the starlings aren’t nimble enough for that so really it’s it’s a lot about how you know
thinking back and and designing your your bird feeder setup for the birds that you’re
most interested in attracting it’s not always going to work but you can experiment with
different things offering different food types a great resource that the lab has from our
project feeder watch team is the common feeder birds resource maybe we can get a link in
the chat about that basically it provides a great set of information about which food
and feeder types are best to attract certain species um so if you’re looking to attract
certain species and not others you can sort of use that resource to help inform you um
i forget what the second question was about um.
[Charles] It’s about what you’re feeding and we’re we’re about 458.
[Ben] Okay so uh quickly what we all of our food and feeders are supplied by our cam sponsor
Wild Birds Unlimited in our feeder tray we use their No Mess No Millet blend which is
a mix of hold sunflower hulled peanut hulled tree nuts so it’s a mix of that doesn’t have
any shells so that’s why it’s called no mess because the birds can eat everything we also
offer suet we have some shelled peanuts in the feeder right now we also have some different
seed characters that are made up of different things like safflower and other seeds but
we switch out arrangements throughout the year and we just try to mix things up to provide
a lot of opportunity for a good diversity of birds so if you want to check out that
resource that’ll help you out from feeder watch.
[Charles] So we’re barreling on to the very end of this thing thanks everyone for sticking
around still about a couple hundred people online here um i thought we’d take it out
there’s there’s a couple questions here that uh that i think are worth asking or answering
answering before we head off and one of them was how do i actually find the most recent
stuff on the Wonder Board so that I’m not duplicating which if you if you might give
a quick um i think there’s a button there right you can you can sort by most recent
is that right?
[Rachael] Yeah so on the Wonder Board if you look at the top right above where it says
join discussion on the right hand side you can sort by best you can sort by newest or
you can sort by oldest and so you can see kind of new old or your favorites and so that’s
how i would recommend doing that.
[Charles] And then yeah i saw that you answered this question typing Rachael, but i thought
it’d be a great question to take us out on as well which is do we have a role to play
after the data collection stage is complete I thought that’s a great way to to sort of
give a little bit more context to what Bird Cams Lab is.
[Rachael] Definitely yeah so we’re coming right up on the hour and that is that question
is really at the crux of what Bird Cams Lab is and why it’s so cool, in my opinion, is
that you don’t just get to collect data with us if you want to you get to to be a part
of the entire scientific process so you can help us ask the research question or you can
help us share it at the end or you can jump in anywhere in between you can collect data
you can visualize it with us we want you to be a part of it the whole way because we’ve
found especially through the past couple years that the community whether it’s professional
or non-professional scientist we all have really interesting ideas and really interesting
interpretations so we invite everyone to continue on with us along the way and we communicate
primarily through email, social media, so you can always stay tuned on the latest Bird
cams lab and with that i’ll put up our closing screen here so that you can make sure to sign
up if you are interested.
[Slide: Dark-gray background with a screenshot of three blue-jays on the platform feeder
on the Cornell FeederWatch cam. The video of the speaker is in the top right corner.
Text read “Thank you for tuning in” as a title. Below it reads “Sign up: https://bit.ly/BCLCornellFeedersLive
Follow on Twitter: @BirdCamsLab Contact us: birdcams@cornell.edu.]
[Rachael] And if you’re not already part of our email list sign up here at that first
link and then make sure you follow us on Twitter if you’re there as well but you don’t have
to have social media at all to be a part of this. We do stay up to date with everyone
via email pretty consistently. And then if you have questions after this or anytime in
the future you know where to find us birdcams@cornell.edu
[Ben] Thanks for joining everybody have a great rest of your day.
[Charles] Yeah very curious group it’s great great talking with you.
[Rachael] Bye everyone thank you so so much for being a part of this.
End of Transcript
Watch Bird Cams staff share highlights from the Cornell Feeders Live investigation, and answer the community’s questions.
On February 23, 244 people tuned in to learn more about the newest Bird Cams Lab investigation, Cornell Feeders Live, and the cam it’s based on, the Cornell FeederWatch cam. Bird Cams staff highlighted discussions on the Wonder Board and answered questions submitted beforehand and during the live webinar, leading to a lively discussion that could have easily continued on past the allotted hour.
The next step in the Cornell Feeders Live investigation is to refine the questions that the community is most interested in investigating. After that, we’ll take a vote before moving on to data collection.
Thank you to everyone who either attended the live event or watched the archived recording. Please share any remaining questions or ideas you have in the forum below. Especially if you weren’t able to attend live, we encourage you to share.