BirdReturns: How Big Data and Farmers are Protecting the Pacific Flyway

[Slide text: eBird From Concept to Conservation, The Cornell Lab of Ornithology- Brian L. Sullivan, The Nature Conservancy- Mark D. Reynolds; Images: Photo of a dunlin, list of bird species, and map of the world with purple dots marking many locations] 

[Steve] Good evening. Last time I stood up here I didn’t have sheets of paper, I had Jessie Barry standing right here telling me what to say, but now I have paper. The first thing I have to do is tell you who I am. I’m Steve Kelling, I’m the Director of Information Science here at the Lab of Ornithology, and today we’re going to be having a talk that’s that’s that’s really special for the projects at the Lab of Ornithology. Those are particularly, that I’m involved with, but I have to inform you all that we’re going to be live-streaming this tonight. 

So people everywhere have an opportunity to hear this presentation. The other, next thing I have to tell you is that the next seminar will be here on November 2nd, which is a Tuesday night, Monday night at 7:30, and it’ll be Gerrit Vyn, who’s going to be talking about the Lab’s new book called Living Bird. And I don’t know if you’ve seen that or had an opportunity to see that yet, but it’s it’s a very special book. So I encourage you all to come out for that. 

Also another place dear to my heart and in New York state, the Cayuga Bird Club this Saturday will be going up to Ferd’s Bog and another bog that’s cause Sabattus Bog in the Adirondacks, look for gray jays, boreal species. That’s a great place to see black-backed, three-toed woodpeckers. And they’ll stop for lunch and dinner, you should bring snacks, and it might snow.

[Laughter] 

All birders are welcome, and you can contact Carl Steckler. Carl, you want to stand up? And Carl will tell you how to get there. So we’re going to leave, we’re, they’re going to leave 7am from the Lab parking lot. 

Okay, so tonight I want to introduce two people, one’s an old friend and one’s a new friend, that’s going to talk about a project about conservation of bird populations in Central Valley of California. You’re first, Brian? Brian Sullivan, who is one of the project managers for eBird, lives in Monterey, California but works remotely here at the Lab. 

He’s been our link with Mark Reynolds, who is the Lead Science Officer for the Migratory Bird Initiative for The Nature Conservancy in California, and I know I screwed that up, I did pretty good. And they’re going to talk about a project called BirdReturns, which links citizen science with NASA Earth imagery, uses concepts and big data analysis techniques to solve a problem about how to keep fields flooded. A, rice fields flooded for shoreBirds. Brian? 

[Brian] All right. Thanks, Steve. Good evening everyone, everybody watching online, including my three-year-old at home, hi Ella.

[Laughter] 

I’m really excited to be here tonight. It was about ten years ago that I came into this room, and and gave a presentation about my vision for eBird. They were hiring a project leader position, and I was lucky enough to get that job along with Chris Wood. It’s been a pretty crazy, fun decade of work here at the Lab. 

I’m really lucky to be involved with everybody in this building. eBird is touched by everyone in this building, and the success of eBird is not only because of the team that works very closely on the project, but also because of the fact that it’s housed here at the Lab, and everybody here should take pride in what we’re about to talk about today. It’s it’s been a really fun project to work on, and continues to be a great challenge moving forward. 

So today I’m going to talk about the big data side of this story. How do we get the data that we need to deliver conservation action on the ground? And I’m going to talk a little bit about the story of eBird, how we engage the community, all the way down to the ways that we create data, scientific data, that can be used in conservation. Then I’m going to turn it over to Mark to show basically our highest profile exemplar of conservation action happening on the ground with eBird data. 

[Slide text: eBird, Our hypothesis: Data gathered by birders can help provide a better understanding of bird distribution and abundance across large spatial and temporal scales, ultimately leading to broad-scale science use and conservation action] 

So our hypothesis all along with eBird is that data collected by birders could be used for science and conservation. And it’s taken a long time to sort of figure out whether or not that’s true. When I started giving talks about eBird ten years ago I would say that with certainty, and boy I sure hoped it was true. And now ten years later it’s really beginning to become a reality. 

[Slide text: eBird, eBird Life Cycle: Engage the global birding community to collect “Big Data” 

Focus on data quality 

Provide open data access 

Analyze data 

Use results for on-the-ground conservation] 

And you know, the life cycle of eBird, if you think about sort of what we do with the project. We engage a broad global community of birders to collect big data. We have a never-ending focus on data quality, I’m going to talk a little bit about that. 

We have a central tenet of open data access, so everything that we collect is completely open for research conservation. We have a shop that’s tied to eBird that helps analyze the data and provide data sets that can be used for conservation. And ultimately the goal is bringing it all around, is when people use these data for conservation action, that’s when eBird’s a real success story. 

[Slide text: Community Engagement] 

So I’m going to talk a little bit about community engagement. This is the side that 10 years ago, Chris Wood and I, and Marshall Iliff, and others thought a lot about how do we get people to actually use this this product? 

[Slide text: eBird, Unique Locations; Image: World map with many areas in yellow and dots of yellow elsewhere] 

This project, these tools that we have. And this slide here is basically from outer space view of a, eBird’s world as it looks today. These are three million locations where people around the world have submitted checklists of birds. 

And as we zoom in you can see that there’s a lot of detail here.

[Image: Map of contiguous United States with dots of yellow and orange throughout, but an especially high concentration in the Northeast] 

If we zoom in on North America, and then here on the Northeast, 

[Image: Map of Northeast United States with yellow dots throughout] 

every one of these little yellow dots represents a location where people are collecting information about birds and submitting it to eBird. 

So we’ve been pretty successful in the last 10 years within the community engagement side. 

[Slide text: eBird, Statistics: More than… 250 million observations submitted ~8 million hours collecting data >5.5 million unique visitors 98% of known species All countries] 

We’ve collected 250 million observations, five and a half million people have used eBird, we’ve collected data on just about every species that’s existing in the world right now, from every country around the world. 

[Graph: Growth in Monthly eBird Observations and Checklists, showing almost exponential growth] 

And over the last ten years our growth curve has been fairly exponential. So we’ve we’ve had some success here, and one of the key pieces to the success was building tools for birders. 

[Slide text: eBird, Tools for Birders] 

You know, when we first got here eBird sort of was, you know, the technology of the internet was a little bit nascent. Everybody was on dial-up connections, the concept of eBird was that you should submit your data because it’s good for birds, and it’s going to be great for science and conservation. But there wasn’t much for birders in terms of getting something back for all that time and energy. At that time it took a long time to just enter your data online. And then when you did it sort of went into this black box, and you couldn’t get anything back from the project. 

So so Chris and I, and Marshall, and others came up with these ideas for tools that would engage the birding community and make eBird a little bit more sticky, as we say. So when people get invested in it, they begin to get things out of it that help them do what they like to do, which is go birding, keep track of what they see, things like that. 

[Slide text: eBird, Personal record-keeping; Image: My eBird page showing Life List total species, checklists, ABA Area Total Ticks, and locations] 

So one of the key pieces that we put in right away was this My eBird page, which allows you to keep track of all your bird observations in one place. 

[Slide text: eBird, Streamlined data entry; Image: Data entry page with list of bird species and number of individuals seen next to some species] 

We spent a lot of time streamlining the data entry process. So we knew that it was a hurdle for people to enter data online, so we you know made this as easy as possible. 

[Slide text: eBird, Mobile data entry; Images: Cell phones with eBird app start page in three languages] 

And today we’re really focused on shifting this process to the mobile world. So we have an iOS app now for data entry that’s available in six or seven languages. 

We’re working feverishly on an Android version of this app, and this is really opening up eBird to new audiences around the world that, it’s really their only connection to the internet is through mobile devices now. 

[Slide text: eBird, Macaulay Library connection; Images: Photo of Costa’s hummingbird with description and spectrogram of Costa’s hummingbird call] 

We have a developing collaboration with the Macaulay Library that’s really focused on rich media. So it might even be tomorrow that we launch this functionality to the public for the first time, where you can upload photos and sounds directly to your eBird checklists and then they become embedded in the Macaulay Library archive that’s been in existence for a hundred years. So that’s not only creating an amazing scientific research collection, but it’s giving birders again tools that they find useful. 

There’s communities around the world whose only engagement is through the camera lens with with birds. They don’t think about using binoculars like we do traditionally to birdwatch, they’re using cameras. So tools like this will allow us to engage different sort of emerging communities of birders around the world. 

[Slide text: eBird, Discover birds; Image: “Explore a Region” eBird webpage, with search bar for entering a location to explore… Local Bird Activity, Trip Planning, and Regional Rankings] 

So that’s sort of the data input side, on the output side we have you know these millions and millions of bird observations. How do we organize them in a way that a birder would find interesting? I had a question just before the talk. How do I, how do I find recent observations of a species? I can’t do it. Well these tools that we developed on the Explore Data page are supposed to make that easier. So this is an example of one tool called the Region Explorer.

[Slide text: eBird, Region Explorer; Image: Tompkins County, NY region overview, showing 328 total species, 113,286 total checklists, and list of most recent sitings] 

You can type in any region here, state, county, and it gives you this dashboard overview of all the activity that’s happening in a region. So this is just a visualization for Tompkins County.

And you can see all the species that have been reported, who’s seeing them. It’s a great way to sort of just quickly visualize, you know, all these millions of records that are coming in. 

[Slide text: eBird, Bar Charts; Image: List of bird species with bar charts showing when during the year they are most likely to be in the area] 

We have tools that allow birders to understand status and distribution of birds, such as bar charts. These tools are really simple, they take the raw data and they plot it out over time, and you can see when birds are occurring in the region. 

So I showed up here on Saturday, you know third week of October, and you can see I’m a little bit too late for veerys, which are at the top of the list here. Most of the veerys are gone. I would have had to come in September to catch the veery migration. But tools like this are really useful for traveling birders to understand when to visit a region, the target species, things like that. 

[Slide text: eBird, Range Maps; Image: Map of Ithaca, NY and the surrounding area with blue and red markers where wood thrush sitings have been reported] 

We have a next generation version of what are traditionally called range maps, and these start at the global scale, so you can type in any species in the world and see their global range. And as you zoom in, you actually interact with real data points. So these blue and red markers on the map represent observations for wood thrush around here. The red ones are right around the Lab. 

[Image: Same map with sitings from Bluegrass Lane Natural Area over it] 

And as you interact with these points you can see how many were seen, who saw it. You can get to their checklist directly, it’s a great way to sort of interact, and again dig into these millions of records. 

It’s not lost on us that birders like rare birds, so we give them a series of tools to explore what rarities are being reported in a region. 

[Slide text: eBird, Sharing with other birders — rarities; Image: Prothonotary warbler rare bird siting for Myers Point in Tompkins County, NY, with photo and description] 

You can go look at this stuff on the eBird website, or you can even sign up for alerts that send emails hourly to your account if you’re that crazy as a birder. 

[Slide text: eBird, Birders + competitive spirit = more data; Image: Tompkins County Top 100 eBirders for 2015, showing the name, number of checklists, species, and most recent addition of each observer] 

One of the equations that we found to be true pretty early on in eBird was that birders plus competitive spirit equals more data. And it sounds kind of trivial, but every tool that we’ve built in eBird is centered around the idea of driving more data into the system. And this one, called the top 100, tends to drive birders a little bit crazy. But in drive driving them crazy and getting them into the field, they’re submitting lots and lots of complete checklists of birds that we can then use for scientific modeling and conservation. 

So as you can see if you look on the left column there in the middle left it says complete checklists, and at the top of that list is Jay McGowan. He submitted 1,100 complete checklists of birds just in Tompkins County this year. And this is happening in counties all over the US, and countries all over the world. So when birders get engaged in these kinds of competitive things it can really drive a lot of data into the system. 

[Image: Global Big Day, May 9, 2015 Tallies for the world, showing number of species, checklists, and participants for various regions] 

So this year we launched really what is the the world’s biggest bird, competitive bird competition. It sort of pits countries against countries instead of birders against birders. And in this case we launched this project called the Global Big Day, this year was on May 9th, and what we found was that this, the idea of sort of countries banding together to try to find as many species as they can in a single day really resonated in a lot of places around the world, in ways that we we haven’t been able to connect with these groups in the past. 

So a good example of Global Big Day success was in Brazil this year. Brazilians really enjoyed taking control over this competition, and and they pitted themselves against Ecuador and Peru in particular. And they were, you know, on social media this was really hot topic, who’s going to see the most birds, and it drove a lot of data into the system from places where we didn’t have data before. 

I think Chris Wood said somebody told him that they they thought the Global Big Day was as fun as watching the World Cup, so we thought that was pretty good to have a comment like that come from a Brazilian. 

[Slide text: eBird, Data Quality] 

But when you gather all these data you really have to worry about data quality. We spend a lot of our time as eBird project leaders, with Steve and others at the Lab, just really trying to constantly evolve our data quality process, because we know that observers have variable skills, there’s all sorts of variation in the way people go birding. We have to account for all of this in a modeling process. 

[Slide text: eBird Data Quality Process; Photos: a bird with a dog’s head photoshopped onto it, and a catbird] 

So we basically, and we we want the catbird but we don’t want the birddog. And to do that we’ve developed this tiered data quality system. 

[Audio: loon calling] 

That’s okay, that was a loon. 

[Slide text: eBird Data Quality Process -Automated data filters -Hundreds of experts review flagged records -Correspondence with experts improves observer abilities -Modeling observer differences improves results; Image: Perched fork-tailed flycatcher] 

We have automated data quality filters that are spread out around the world, and these you can think of as geographic regions in which people can submit data without it being flagged for review, and I’ll talk a little bit more about that. We have this quickly expanding network of regional editors that help vet these flagged records when they do get, you know, when they’re kicked into the system as unusual. And those experts correspond with birders in a region that are coming up, and that sort of expert-beginner communication shouldn’t be understated, because it’s really important that that happens. 

And most recently we’ve begun modeling differences in observers’ abilities to detect birds. When we can understand that piece, and put it into the modeling process, it’s really going to improve our ability to model birds accurately. So evolving data quality and the data quality filters. 

[Slide text: eBird, Evolving data quality; Image: Map of contiguous United States with California outlined in red] 

When we started eBird we, our goal was to have one data quality filter per state in the US, and you can see the state of California is quite large soon we realized that wasn’t going to, that wasn’t going to work, so we broke it in half. 

[Slide text: eBird, Evolving data quality; Image: Map of California with counties, and red line through the middle dividing it into northern and southern halves] 

And then the next step was to break even counties within California down to smaller geographic regions.

[Slide text: eBird, Evolving data quality; Image: Satellite view of Riverside, California with one area highlighted in red] 

And right now this is the process we’re in with eBird. This is the stage of our data quality filters piece, where we’ve got a county like Riverside, California now has eight filter regions. And you can, you know, the idea of these filters is to flag records of birds that aren’t supposed to occur in these biogeographic regions. 

So if you’re in Riverside County and report a northern cardinal it’s going to get flagged. But there’s a lot more subtlety to it. It also flags birds that are early or late in their migration, it flags high counts, all of these things get looked at by our regional editors. 

[Slide text: eBird, Species accumulation curves; Graphs: Representative Individual SACs and Indices, Length of Checklist (Hours) versus Number of Species on Checklist] 

So the most exciting piece of this evolving data quality process recently has been the ability to understand differences in observers. And Steve Kelling and others just published a paper in PLOS ONE, PLOS ONE, on this notion that we can understand a person’s ability to detect birds by looking at this metric that’s called a species accumulation curve. 

And what you see on the screen here are three different individual’s species accumulation curves, and what a species accumulation curve is, it’s pretty simple, it’s the number of birds you see per unit time. So hours spent in the field, eventually you’re going to tail off on the number of different species you can detect. 

And very experienced birders tend to detect more species more quickly than beginning birders, and there’s also differences in the makeup of species that they can detect. But the key thing is that once we understand these differences we can start to build those into our modeling processes 

[Slide text: eBird, Observers improve over time; Graphs: Number of Checklists versus Estimated Number of Species with curves for various checklist numbers] 

and improve the performance of those models. 

And a key outcome of this paper was not only the modeling component, but also that over time we can for the first time actually measure how much better a person gets as they submit more data to a project like eBird. So these three, these two graphs right here show that as you submit more checklists, your species accumulation curve actually gets steeper, so you’re you’re learning, you’re getting to be a better birder. 

It’s a really significant outcome of this paper that our observers are actually improving over time through engagement with projects like eBird. 

[Slide text: eBird, Data Access] 

So we’ve got all this data, but it’s really important to not let it sit here on servers where no one can get to it. So we have to think about data access. And we developed early on the notion that we would you know always make data freely available for research conservation use 

[Slide text: eBird, Data Products — 40K+ Downloads eBird Observational Dataset (EOD) eBird Basic Dataset (EBD) eBird Reference Dataset (ERD) eBird Species Distribution Models (STEM)] 

but we had to figure out sort of which communities out there wanted eBird data, then we had to develop the appropriate data sets for them. So right now we have these four different data sets that package eBird data, all these millions of observations in different ways. And this list of datasets, I’m not going to get too into the weeds on these, but they’re basically very basic views of the data at the top, down to very marked up views of the data that have all sorts of covariates that are designed for more rigorous scientific analysis. 

But so far we’ve had 40,000 downloads of the eBird dataset. 

[Slide text: eBird, eBird open data access; Image: Global Biodiversity Information Facility webpage and eBird Data Access Available Datasets] 

And those data come through either direct access at eBird or partner organizations like the Global Biodiversity Information Facility, which is a, you know, all taxa data data clearing house, and they also serve eBird data up in a very simple form. So people can come to eBird and get the three more rigorous datasets, or they can go to GBIF and get the more simple view of the data. The point is there’s a lot of data downloads happening with these data right now. 

[Slide text: eBird, eBird DataUse — Project Type; Chart: Pie chart of project types, with Academic/student taking up 48%, NGO/Governmental 29%, General 20%, and Commercial 3%] 

So on eBird side of things we can understand a little bit about how people are, what this community of people downloading the data looks like, because we have a little bit of an access gateway on the data. We ask you who you are, if you have an organization, what your project is, things like that. So we can get metrics like this that tell us a little bit about how the data are being used. 

You can see that a huge proportion of the data use right now, almost fifty percent, is from the education, the academic and student community. Today we’re going to focus on the other side of this this pie chart, the twenty-nine percent NGOs and governmental agencies, as well as the general side. This is where people are accessing eBird data and using it more for conservation. 

[Slide text: eBird, Science] 

So another key component of this what we call the eBird enterprise is the science component. We need to be able to understand all the the benefits and drawbacks of the kinds of data that we collect, and we need to be able to model those data, you know, use them in different types of analyses. So these data right now are being used by people outside the Lab as well as inside the Lab. 

[Slide text: eBird, Status and Distribution; Images: Birds of the Sierra Nevada: Their Natural History, Status, and Distribution; American Birding North American Birds; Scientific paper applying large scale citizen science data for local management use] 

And really the eBird data, they provide this foundation of information on species occurrence, abundance, and distribution. These things are basically the building blocks of just about every conservation process that’s happening out there right now. You need to know where the animal lives, and in what numbers, and when in order to conserve it properly. 

So at a very basic level eBird data get used in publications like Birds of the Sierra Nevada, journals of bird record like North American Birds, lots of technical literature white papers, grey literature, things like that. But they’re also being used in more higher level scientific analysis. 

[Slide text: eBird, Evolution of Australian Honeyeaters, Miller et al. 2013, Ecology Letters; Images: Graph of mean annual temperature versus Log10 precipitation over time; Photos of two Australian honeyeaters] 

And this paper is a example of a data use of eBird where they explored the evolution of this radiation of Australian honeyeaters over time. 

[Slide text: eBird, Chickadee hybrid zone analysis, Taylor et al. 2014, Current Biology; Image: Maps of North America showing range of Poecile atricapillus and Poecile carolinensis, and Contact zone] 

In this case a grad student here in the Miller, or sorry in the Lab, used eBird data to understand the shift in northern, sorry Carolina chickadee and black-capped chickadee’s hybrid zone. So this was really amazing to me on a number of levels. One, who in here can identify a hybrid chickadee? Not me, they’re really hard. And also this shift that they detected, it was, it was actually 11 kilometers over 10 years, the hybrid zone shifting north. 

And they used a combination of genetic analysis to look at these captive birds, to look for, you know, this hybridization, but they also verified it using eBird. I just thought it was pretty amazing that they could detect such a small range shift, and a rather obscure, you know, hybrid taxon. 

[Slide text: eBird, White-throated Needletail; Image: Range map of white-throated needletail; Photo: Two white-throated needletails in flight] 

So inside the Lab we have a group upstairs where we have these really incredible scientists that are looking at eBird and starting to try to model the way birds move on the landscape. And this example is some work that Frank La Sorte did looking at the trajectories and speeds of migration of birds. And this example is using white-throated needletail, which is one of my favorite birds in the world. It’s this huge, sort of beautiful swift that nests in Asia, winters mainly in Australia. 

And you can see on this range map we don’t have a ton of data for this species. But you can see some points in the Himalayas, and some across Central Asia, and a lot from Australia.

[Slide text: eBird, White-throated Needletail; Image: Map of white-throated needletail sightings in different colors over time and one region circled] 

And what Frank has been able to do is look at the centroids for each day of a species’ range, and then plot that on a map. And he can see over time this trajectory of migration. What he shows for white-throated needletails is that they have this big looped migration that goes out over the Pacific and then comes up in spring on the west side of this trajectory. 

[Slide text: eBird, White-throated Needletail; Image: Same map with different colored horizontal bars along the loop] 

And they also, he also, can understand how fast these things are going. So the difference between these little plates, these little horizontal bars on the trajectory are one day intervals. So these, you know, the centroid of the range is moving very quickly during migration, then when they get to Australia they all get stacked up because they’re down there for a few months before they shoot back north. 

[Image: Map of North and South America with dots representing different bird species, moving along their migration routes] 

The neat thing about this type of analysis is that once we can understand how it works for one species we can look at a lot of species. And this map shows the migration trajectories or centroids of a species’ range for a suite of North American migrant birds that winter in Central and South America. And you can see the dates running across the bottom, so this is summer they’re all up in the northern parts of North America, and now it’s fall migration, they all start to move back down. 

And this is really significant for a number of reasons, because we’re seeing, you know, all of these species doing different things, but we can also see where these these trajectories concentrate, and highlight areas of conservation importance on a map for a suite of species. 

So this this is the first time we’ve been able to see this for a large number of birds at once. And it’s pretty, pretty fun to look at. And for the birders in the room you’re probably all going, “What’s that one doing this? What’s that?” That’s what we do every time we look at this. 

[Slide text: eBird, Distribution Models; Images: Map of North and Central America with expected eBird count for magnolia warblers shown in different colors and moving across the map over time, and magnolia warbler on a branch] 

So that centroid modeling approach is really, really neat, and really telling us a lot about these dynamics of migration. Another group here at the lab led by Daniel Fink has developed this next generation distribution modeling process called STEM. 

And without getting too technical about it, it takes your eBird data, and associates your observations of birds with a lot of different variables, landscape characteristics, all these things that may influence bird distribution, and then we create models that allow us to look at bird populations, the whole population, move across the landscape. And this is really for the for the first time, you know, the most significant thing to be able to watch entire populations of birds move through its entire life cycle, from breeding in the boreal forest like these magnolia warblers do, to wintering in these core areas in Central America. 

And this is a map that shows the abundance of magnolia warblers, so you can see in the summertime they’re mainly, you know, the high numbers are off in the eastern part of the boreal forest, they migrate through the eastern U.S., you know trans-gulf migrant, they come into the Yucatan and Belize and Guatemala, and they winter there in big numbers. So this, the predictive modeling process here is allowing us to do a lot in terms of understanding specifics of bird migration, but it’s also enabling a lot of conservation because under this really pretty visualization is tons and tons of data. 

There’s a million data points here on the landscape that we’re making predictions for magnolia warbler every week for every year, so we’re understanding a lot. And when we can do that, we can.

[Slide text: eBird, Distribution Models, Kelling warbler total; Image: Map of North and Central America with expected eBird count for warblers shown in different colors and moving across the map over time] 

Well this is a different visualization that I just wanted to show that just came hot off the press today. This one is actually like Frank’s centroids animation, this is a whole suite of migrant species that breed in the boreal forest in eastern North America and winter in Central America. And the thing that we are looking for here was trying to identify regions of conservation importance for this particular group of birds. As you can see, when they migrate south they’re piling into the Gulf Coast states, and then they winter again in that Belize-Guatemala region in really high numbers. 

So the idea of conservation is to understand that you know the most important place is on the landscape, and this kind of information allows us to get to it. I’m going to see if I can get this this to play.

[Slide text: eBird, Precision Science; Image: Zooming in from entire Earth view to New York showing predicted ovenbird presence during breeding season] 

So when we have this information, you know, we can look at it and look at these visualizations from from afar, but underneath is a lot of information, it’s very very fine scale. So as we zoom in on New York, right here on Ithaca, we’re looking at a visualization now of ovenbird during the breeding season. As you can see, as we zoom in we’re predicting where ovenbirds are occurring in very high resolution. And this ability is really transformational because it allows us to be very precise about where this bird lives and where it doesn’t. 

And this is what helped Mark and the TNC in California be so effective with their project BirdReturns. 

[Slide text: eBird Data Use — Conservation; Conservation Science → Conservation Planning (49) → Conservation Action- Site & Habitat Management (37), Species Management (2), Habitat Protection (30), Law and Policy (16) → Conservation Goal] 

So I’m going to start this conservation piece just by saying that there’s a lot of conservation action happening now with eBird data. In a recent study here at the Lab we found more than 150 tangible conservation actions happening as a result of eBird data, and those range from research and monitoring, conservation planning, to different types of actions like habitat management, species management, things like that. 

[Photo: Flooded fields from above] 

And Mark’s really going to focus his exemplar tonight on this habitat management component here, so with that I’m going to turn it over to Mark to talk a little bit more in detail about this project BirdReturns. And Mark, thanks a lot for being a great partner for us, for the Lab, and for eBird. This is, this is a one of our flagship success stories. So I’m really, really proud and glad that Mark’s here to be able to tell it to you all. Thanks. 

[Applause] 

[Mark] Thank you, Brian. And thank, thanks to everyone for coming out tonight. Brian’s conveyed some of the innovations and excitement around this revolution that you’re all participating in, which is citizen science, and birdwatching being able to contribute to big data that has applications in science and in conservation. And I can only say as a conservation scientist with The Nature Conservancy how exciting it has been to collaborate with the Lab, and with the whole eBird team, and Brian and Steve, on taking some of this rich data and applying it to some very acute conservation challenges we have not only in California but also in the world. 

So I’m going to talk a little bit about our work in California, 

[Slide text: The Nature Conservancy- Protecting Nature. Preserving Life; Image: Satellite view of North America with bird migration routes shown and California outlined] 

And the ways in which big data analytics can contribute to conservation on the ground in solving some of our biggest conservation challenges. One of them being migration. We all know that migration is, is a challenge because the birds don’t stay in one place, they move around. It’s a challenge in scale, and in scope. 

[Photo: Dunlin in shallow water] 

And so the example that I’m going to be talking repeatedly about tonight is this bird, this is a dunlin, it’s a medium-sized, rather nondescript shorebird, but it’s a good example of the kinds of species that really have a lot of trouble with the changes that we’re visiting on the planet, as we’re altering wetlands globally. 

So dunlin, of course, like many shorebirds in this hemisphere, breed in the Arctic, and then they come down the Pacific Flyway with millions of their of their fellows, and stop over in places and look for food, to put on resources to help them with the rest of this perilous journey that they’re taking that’s thousands of miles, and eventually they end up in California, and that’s a great thing. 

This this is a wondrous cycle that we’re all able to participate in of animal, participate in an experience of animal migration, except that there are problems all along the way, and it’s not just for dunlin.

[Photos: Various migratory birds that use the Pacific Flyway, including Canada geese, stilts, ibis, white-fronted geese, and more] 

The Pacific Flyway is, is traveled by about sixty five species of waterbirds, so from Aleutian Canada geese, white-fronted geese, small ducks, dabbling ducks, waders, other shorebirds all use this aerial superhighway to get from their breeding areas to wintering areas. The problem is when they get to California, California is innovative in so many things, and we’re also unfortunately a leader in global habitat destruction. 

[Slide text: 95% of historic habitat destroyed] 

So in the Central Valley we’ve lost about ninety-five percent of our wetland habitats. 

[Image: Map of California showing wetlands of the Central Valley pre-1900, with >4 M acres, 20-40M waterfowl, ??M shorebirds; and post-1960 with < 250K acres, 4-7 M waterfowl, ~1M shorebirds] 

And so for birds this is what the Central Valley looked like in pre-settlement times, in the 1800s. There was a rich tapestry of about four million acres of integrated wetland systems. And since that time we’ve lost about ninety-five percent of them. And with that we’ve seen some pretty precipitous declines in waterbirds, particularly waterfowl and shorebirds. But they haven’t winked out entirely. And part of it is because when they experience historically this kind of landscape, 

[Slide text: California Migratory Waterbird Habitats; Photo: Wetlands of Central Valley of California from above] 

now this is a bird’s-eye view of the Central Valley of California, 

[Slide text: California Migratory Waterbird Habitats; Photo: Mosaic of farms and wetlands of Central Valley of California from above] 

and it’s still a mosaic, and there’s a lot of opportunity for them to find habitat in this agricultural landscape that we’ve created in California, it just has to be managed correctly. 

And so part of what our vision for this landscape is is a way in which our small, but very important anchor sites of refuges work in concert with the agricultural landscape and the urban landscape to protect and preserve migration in perpetuity. 

So from the agricultural point of view, here’s what this habitat looks like, 

[Image: Satellite of Central Valley, California showing different types of agriculture] 

14 million acres of the Central Valley in the far northern part of the Central Valley here in blue, you’re seeing our rice growing region. So in California there are about 500,000 acres of rice lands, mostly in the Sacramento Valley. Moving south you see it changing to the teal colors of various field crops, and then in the southern part in the San Joaquin and the Tulare Valleys you’re seeing the the growing landscape is dominated by that red, that’s nut crops, and then also the purple, which is vines for table grapes and also for for vineyards. 

So this landscape has a lot of potential, but we’re also finding it to be a landscape that’s vastly less resilient than the existing, the pre-existing wetland landscape. So building resilience into this landscape, in terms of the kinds of crops that are grown, and how they’re managed is really what we need to do in order to secure migration basically for future generations. 

[Slide text: Dynamic Durability; Images: Maps of California in Year 1 and Year 2] 

And one way we’re thinking about this at The Nature Conservancy is we’ve done a pretty good job at identifying places that are important enough to be protected forever. And those are the places that we and our partners buy and restore, and eventually some of them become part of our protected areas, refuges and federal and state lands. And so that’s the anchor site, but what happens in the rest of this landscape? 

May, you may want that to change from year to year, you may want that to be responsive to climate change, to drought, and to various other factors. And so we think about trying to mimic the dynamics of bird migration, and how we do our conservation. So this is the the fundamental premise of dynamic conservation. So in one year we may need this kind of habitat 

[Image: Same maps of California, in Year 1 yellow circles are distributed throughout the Central Valley] 

just strewn all throughout the San Joaquin and the Sacramento Valleys. This could be an example like the years we’ve just been experiencing, with this dire drought, where we need habitat well distributed throughout the Central Valley. And that’s been the impetus for this work that we’ve been doing the last couple of years, to hopefully an El Niño year this year, very wet. 

[Image: Same maps of California, in Year 2 smaller yellow circles are distributed less evenly throughout the Central Valley, with one in the north, one in the center, and three in the south]

There may be abundant water available and wetland-like conditions throughout this landscape, and from a conservation point of view we may need to provide less additional habitat, and distribute it in certain ways that are responsive to what’s happening out there. 

[Slide text: Dynamic Conservation, Precision science + New conservation tools + Policy influence = Effective and efficient conservation at scale; Photos: Map of California, auctioneer, United States Capitol, and northern pintails on water] 

So the vision for dynamic conservation is that we can couple precision science, such as the kinds of models that Brian talked about, with new conservation tools, which are really much more fluid, and less about protecting something in perpetuity, and more about working with the working parts of the landscape, the farmers and the agriculturalists, and getting them to do beneficial things for wildlife. But maybe just at the right place at the right time. 

If we can do that, we can build momentum behind changes in policy that would be necessary to really scale this to the whole Flyway scale. So that’s the vision of dynamic conservation, now where does this precision science come from? 

[Photos: Long-billed curlew in flight with radio telemetry backpack, map showing location dots, attaching the backpack to the bird, and satellite in space with view of Earth] 

Well, um, fortunately for birds, as you all know, they’re very well studied. We know more about birds than we do about almost any other taxa. And um, we typically at The Nature Conservancy work with research partnerships, and try and figure out where and when the animals and plants that we’re trying to protect need habitat. And traditionally this has been working with long-term demographic studies, or radio telemetry. These are great datasets but they usually give us really precise information about only a handful of species. 

And again on the Pacific Flyway we’re interested in protecting all 65 species of our waterbirds, keeping those common birds common, keeping those ecosystem services part of our agroecosystems. And there really wasn’t a source of information to develop the precision science around this until this. 

[Slide text: Northern Pintail, Images: Map of North and Central America with relative abundance of northern pintails shown in different colors and moving across the map over time, northern pintail] 

And so one more, really wonderful visualization of eBird data. This is for the northern, northern pintail. The northern pintail’s a duck that’s very important to us in California. They’re a whole Arctic duck, but in North America they’re wintering populations, about sixty percent of the population winters in the Central Valley of California. And this is a duck that’s declined precipitously since the 1950s and still hasn’t reached its recovery targets. So it’s a species that we’re very interested in. 

And we’ve long been looking for a way to really understand where and when it needs habitat, not just in California, but across the Flyway. And when I first saw these visualizations from the eBird data from the Lab, I thought this is it. This is how we’re going to be able to pinpoint those places where we can work with landowners and make a difference. 

[Images: Map of California with the Central Valley outlined showing probability of northern pintail presence over time, and northern pintail in flight] 

So a few years ago we started a partnership with the Lab here and started taking some of this data and downscaling it to California, to our study area in the Central Valley. And already just even at this level it starts to reveal patterns that we’ve never seen before that gives us opportunity to think about the timing and the placement of additional conservation. So this is a pintail loop running throughout the summer, we’ll take it again through fall and just I look at a couple of the places that are important here. 

Sorry about that. So it has pintail coming into California, they’re one of the fastest flying ducks and one of the earliest arrivers in California. And we can already see that the rice lands and the refuge system in the Sacramento Valley popping in that early part of the fall, and then filtering into the rest of the state. And so this information is conservation actionable for us. 

[Image: Map of part of the Central Valley of California showing STEM probability of northern pintail presence] 

Even though we had that sense we still wanted to verify it, and so to reiterate some of the points Brian made about data quality, we took this initial data set of modeling outputs from eBird, and put that against a really rich dataset of telemetry locations for northern pintail, shown here in purple. And fortunately we see amazing concordance. This is data that’s very painstaking to to get, long-term studies of radio-marked pintail. Researchers driving up and down the Central Valley, and fixed-wing aircraft too, and you can see it really maps on well to the model predictions, which gives us the confidence that this information is conservation actionable. 

[Images: Map of the Central Valley of California showing agricultural lands, protected areas, and refuge complexes; a large flock of birds on and above agricultural land; and an agricultural field] 

So we can take those data and again look at where and when birds most need habitat, something we’ve never really been able to do before with this level of sophistication. So here I’m showing a map of the Central Valley, the agricultural lands are in brown, and that small but really important network of protected areas is in green here. 

[Images: Same map of Central Valley with northern pintail fall probability added, two northern pintails in flight] 

And we can put the pintail distribution for a specific time slice on top of that, and see how much of that protected areas landscape is actually benefiting the pintail, and what needs to happen in order for them to have habitat in the agricultural lands. Now we can take that for a variety of species now, and look at what their needs are week by week. 

[Images: Same map of Central Valley with long-billed curlew fall probability added, a long-billed curlew in flight] 

And so this is another bird that we’re, we very much care about in California and elsewhere. This is a long-billed curlew, North America’s largest shorebird, a species that’s declining, and it depends on an agricultural matrix as well as pasture land for its habitat. And we can see that it’s, at this particular time period, does not very well map on to the protected areas landscape. So this is a species for which private lands conservation is really essential. 

Okay, so we can take all of these data and start to ask some of these really important conservation questions regarding when and where do we work, 

[Slide text: Point Blue Conservation science for a healthy planet, The Nature Conservancy Protecting nature. Preserving life; Images: Satellite in space, diagram of a satellite, and map of the northern Central Valley showing water] 

but we also wanted to put it together with data about the habitat. Where and when is the habitat available? What condition is it in? And so to do that we started thinking about water availability and surface water. Where is it on the landscape? When is it available? And when is it correlated with the predictions for the birds? 

And we started a project with Point Blue Conservation Science, one of our conservation partners in California, and they had developed a methodology for looking at Landsat satellite imagery for the presence of surface water, an algorithmic approach to all of this Landsat data. This is the world’s oldest Earth-observing satellite system from the 1970s that takes imagery of the same location on the planet every 14 days. And so it’s a huge data set, and we’ve now mined this back to the year 2000. 

[Image: Map of the Central Valley of California showing surface water probability over time] 

And are able to make these kinds of maps of where and when surface water is available in this very dry, heavily altered landscape of the Central Valley. So now this gives us the ability to look at places that we have high predictions for birds and low predictions for water, and go in there and try and make a difference, and make better habitat for the birds. 

[Images: Map of Sacramento Valley showing probability of least sandpiper and high probability of surface water, least sandpiper in shallow water] 

So this is really the the scientific underpinning of BirdReturns, this habitat restoration and enhancement project I’ll be referring to for the rest of the talk. And this is an example of the least sandpiper, where we have the predicted occurrences here in orange and the water availability underneath in blue. And it gives us the ability to look at high to medium predicted occurrences for that bird and places where it could have better habitat. 

And that gives us the power of precisely locating when and where we need to go talk with those farmers, and try it and align their practices with the needs of the birds. 

[Slide text: Aligning with Farming; Image: Rice Field Activity, Potential Shorebird Habitat, and Shorebirds throughout the year] 

So we piloted this in California rice growing region, because rice is a flood managed crop, and there’s already a well known scientific basis for rice serving as a proxy to wetlands, because its flood managed. The issues with rice are it’s only beneficial at certain points in the year. So this is the annual cycle of rice in the, in the top from the field preparation and seeding of the fields, all the way through when the rice is growing and when it’s harvested.

And at certain times of year there’s water on the landscape, and it’s not covered up with a crop, and those are the times a year when shorebirds can use that habitat. There are also times of year when that water is pulled down and those fields are dry, but there’s still water available, and those fields could be kept wet a little bit longer. 

So looking at the shorebird potential needs gives us the ability to precisely target activities those farmers could do to specific weeks of the year, particularly in the early fall and in the early spring. 

[Slide text: Habitat Auctions; Images: Chart of habitat auctions with bidding and contracting in August, fall practices in September and October, and spring practices in February and March; shorebird, duck, and geese photos] 

And so armed with this information, we went into grange halls and county libraries all over the Central Valley and met with farmers, and talked with them about whether they would be interested in working with The Nature Conservancy, and having us help them create habitat during specific times of the year. And we were really surprised that farmers were pretty excited about this, especially because we gave them the option of submitting a competitive bid for the cost that it would, that they would incur in providing that habitat. 

So this is our auction period. We have auctions that we run in September through October in two week increments, and in February and March. And I’ll talk a little bit more about how this works. 

[Slide text: Reverse Auction; Images: Auctioneer, four figures selling to one figure buying with the goal being lowest price] 

We do this in a format that’s called a reverse auction. So most of you know what a forward auction is, it’s when you have a precious item that you sell to the highest bidder. So the little graphic in here of the guy, that auctioneer, that’s from the sale of Edvard Munch’s the, The Scream, which at the time was the most expensive painting ever sold at auction. And so The Scream, there’s only one of it, you all want it, you all bid against each other, the price goes up. 

In a reverse auction we are the buyer, and we want something from each of you. And because of that you’re all competing to give me what I want at the least price, so it’s just as its name is the reverse auction, and the goal is the lowest price. But we still wanted the highest quality, so we weren’t trying to not pay the farmers what that habitat was worth, but we’re trying to get the most habitat for our very scarce conservation resources. 

[Slide text: BirdReturns “Experiment” 1. Will the birds respond? 2. Can we attract enough birds to run the “experiment”? 3. Will the auction deliver cost-efficient habitat? Image: Northern pintails on water] 

This is a big experiment, so even even with this bolstering of our strategy with this this rich data and analysis behind it, we still had a lot of uncertainty. Would this work? If we provided habitat these times of year, would birds respond? Would the habitat be any good? Could we get enough farmers to participate that we could investigate this competitive auction tool? And would the auction actually be better than if we just paid everybody the same price? And so those were the three hypotheses that we had when we first did this in the spring of 2014. 

[Slide text: BirdReturns Pilot Spring 2014 -40 Rice farms -2” -4” H2O shorebird depth – ~10,000 acres -Feb-March 2014 ; Image: Satellite of bird locations and water probability for the region of Sacramento Valley in Jan-March] 

We were really excited to have a group of willing farmers that wanted to work with us, they, we had 53 bids submitted. For an initial effort it was a pretty great success, it covered about 40 individual rice farms. We ended up selecting about eighty percent of the bids that were submitted, and we were asking farmers again to provide habitat just for shorebirds, two to four inches of water depth. There’s, there’s a lot of habitat programs out there, and our wildlife refuges are mostly focused on habitat for waterfowl, so we wanted to focus on shorebirds because they mostly are not getting their needs addressed by other conservation programs.

We had about 10,000 acres in our initial offering of this program, and again it ran from February through March. This is an inset here of the Sacramento Valley, and also of the the water availability shown in blue here, and our three main areas in which we had farms participating. It was really good geographic coverage, and included small farms, big farms, industrial farms, family farms, so for a relatively small sample we were really pleased with the diversity of participation. 

So what were our results? 

[Slide text: BirdReturns Spring 2014 Results – ~11,000 acres of high-quality spring habitat -53 bids received, contracted with 80% -Auction was robust and cost-efficient -Cost 25% less than uniform pricing -Cost < 1% of protection strategy; Image: Shorebirds in water] 

Again about 11,000 acres were in the program, we contracted with about eighty percent of the bids. The auction proved to be really pretty cost effective. It cost about twenty-five percent less than if we had paid everybody the same price. And our, in our structuring of the auction we, we gave people some feedback on their initial bids, and gave them an opportunity to come down. That also drove the price down about an initial, an additional fifteen percent. 

And relative to the cost of protection, we do a lot of protection at The Nature Conservancy, and we’ve actually purchased conservation easements on rice farms for birds. This cost about one percent, less than one percent of the per acre cost of buying this property either in fee or putting a conservation easement on it, so pretty cost effective. 

[Slide text: Monitoring Design; Image: View of farm from above with certain areas outlined in red] 

So what about the birds? Well, we had to go fairly deep in the science here to really verify that this was creating good bird habitat. So we designed a sampling program to have really trained professional avian scientists go visit points on the landscape and count the birds that were in these fields. We also worked with our farming community to do some counts on fields that weren’t in our program, because we were interested in the comparison. So we have some treatment fields and then a small number of control fields to compare with. 

So this is this is some of our points. And we do 200 meter point counts where we count all the birds that we were able to see in that, in that radius on those fields. 

[Photo: Man standing near a flooded agricultural field entering data into a tablet next to a camera on a tripod] 

This is my colleague Greg Olette, he’s a senior scientist at The Nature Conservancy and he’s been leading our monitoring program. Here’s Greg entering some data from one of our BirdReturns fields on an iPad, and it goes directly up into the cloud, and we can view it on our servers later that day. So this has been a project that spurred a lot of innovation for how we just do basic field work, and it’s been really terrific that way because we were able to get results fairly rapidly. 

[Graphs: Bird Responses Spring 2014, Species Richness, Shorebird Density, Wader Density, and Waterfowl Density comparing treatment and control fields] 

So this is the bird responses from spring of 2014. In the upper left species richness, about, over twice as many species in our treatment fields as in the control fields, not too surprising, most of the treatment fields have more water on them in the control fields. Shorebird densities were about 30 times higher on average in the treatment fields than on the control fields, and these densities particularly later in the year are higher than anything that’s been reported in the literature for California rice lands. 

So by putting habitat at the right place at the right time we were really benefiting more birds at higher densities. Below are just some systems, summary stats on on waders. These are non-target birds, but yet were using our fields as well. And waterfowl as well.

[Slide text: Dunlin Response Spring 2014; Graph: Time period versus average number of Dunlin per hectare for treatment and control fields; Photo: Dunlin standing in shallow water] 

The hero of spring of 2014 was the dunlin. We had amazing responses of dunlins. In fact one of our field techs observed a flock of 20,000 dunlin on one of our fields, which is estimated to be about twenty percent of the entire Central Valley dunlin population for that time period on one field. 

So emboldened with this information and this, the success of having farmers participate in the program 

[Slide text: Expanding the Program; Graph: Number of bids, Acres bid, and Dollar amount in Year 1 and Year 2 for accepted and denied/ineligible bids] 

we’ve extended the program, and we have made it basically a year-round program, and we’re now starting our third year of doing this kind of work. Expanding the program has meant a couple of things for us. One, more participation means that the auction can function better. We can accept more bids, but we can also reject more bids. We have more acres bid, we can be more selective, we can pick higher quality habitat. Throughout we’ve also been trying to make the most of scarce conservation dollars and of precious water in the drought, so we’ve been trying to use this as a model program of adaptation and efficiency even as we’ve been trying to extend it. 

[Slide text: BirdReturns Impact 2014-15, Season and Acres- Spring 2014- 11,000, Fall 2014- 6,100, Winter 2014-2015- 5,000, Spring 2015- 6,000, Fall 2015- 7,200; ~150 farmers ~35K ac wetlands -Over 50 species of waterbirds -6K point counts, 737,990 total birds observed, 256,643 shorebirds] 

So our, to date we’ve created about 35,000 acres of temporary wetland habitat. This is habitat that would not have been there, absent this effort. We now have a growing community of about 150 farmers, not all of them have been in every auction cycle. A lot of new people coming in, word is getting out about this as being something that that people can do in a way that doesn’t undercut their bottom line and that they want to participate in. Many of our longtime participants are now really interested in the birds, and are really looking forward to having them come back the next time that they’re in the auction. 

We have over 50 species that are that are benefiting from this, and we have now a mountain of bird observational data from our treatment fields and our control fields. 

[Slide text: Dynamic Conservation & Migratory Birds -New tools- habitat when & where birds need it most -Able to achieve dynamic habitat at scale -Optimizes conservation investments -Scientific foundations for adaptive management; Photos: shorebirds and waders] 

So what’s next for this? Well, I think that this is an example of the power of citizen science, big data, and conservation coming together around some of the big conservation issues we have globally. This is just one of many examples where I think we’re developing new approaches and new tools to really match the dynamics of the natural world with better and more efficient conservation. In this case we’re trying to put habitat when and where the birds need it most, and trying to achieve greater results at scale. We’re also trying to optimize conservation investments, and throughout this lay strong scientific foundations for adapt, for adaptive management. 

We’ve had the benefit and misfortune of running this program during the greatest drought in California history, and we were fully expecting that we might go open for business and have no farmers show up because of the risk they would be taking. That was not the case and we’ve again proved that this can be one way that we adapt to drought conditions and the new normal of life and a, change in California. 

[Video: Thousands of dunlins flying against a blue sky and over a flooded agricultural field]

So I want to close on this image here. If I can get this to run. So this murmuration of thousands of dunlins is flying over a field that wouldn’t have been there absent the contribution that you all make as birdwatchers through eBird to citizen science, to big data, and to the ability that conservation groups can have to try and make this landscape a little less harsh for these animals. Kara de Bellas, who was one of our field techs took this last spring, and sent this to all of us on the team. And for those of us that don’t get to get out in the field every day this was really a great thing to see. 

So thank you all for coming out tonight, and Brian and I would be happy to take some questions if we have some time. 

[Applause] 

[Mark] I’m going to go ahead and repeat your question since we’re streaming. So the question was do the farmers get any other benefit besides the monetary benefit, do the birds provide any benefit? And we believe they do. There are a number of studies that have looked at the effects of concentrations of birds on rice fields in helping the farmers to decompose the rice stubble, which is a, which is really why they flood their fields in the middle of the winter is to break down that stubble so they can get ready for planting in the next season. A lot of these birds are also eating a lot of invertebrates, and some of those undoubtedly are agricultural pests. That’s less well studied, but the the effect on decomposition has been studied experimentally, so yes.

So the question is if um any of the farmers who submit a bid to our program who aren’t successful, or may be rejected, if they pursue additional avenues for conservation, and many of them do. We’re not the only game in town. There are other programs that they can go to if they’re interested in receiving some compensation for habitat benefit. Um it’s not something that we are able to track, we mostly track what’s happening within our own program. 

How do we, so many hands, how do we do this? Okay yeah remember you had your hand up, we’ll get to everybody. Well the funding for this program has come from our generous donors, so this has been a privately funded program. And we are very interested in the vision for the sustainability of programs like this. And so what we’re exploring now is how to use these private dollars to actually influence how public money is allocated for this kind of bird conservation. So we can drive larger impact. It’s it’s something that that we’re always interested in doing with private money is trying to extend the impact, and for this program it’s something that state and federal agencies are very interested in, so we’re very hopeful that there’s a long-term future for this. 

Okay. So the question is regarding how the drought in California has affected migration for shorebirds and what was the second part? Yeah and whether it’s affected the BirdReturns program and the farmers. So the drought we’re experiencing California is unprecedented, and it has effects on communities, and on farms, and agriculture, and on everyone in California, and and also our wildlife. So this has been no joke, and it’s been a real challenge to to work in this environment and to try and, you know, further all of our goals for a prosperous California. 

The ways that we know that it’s been affecting the farming community, we’ve had about twenty five percent of the rice lands fallowed, I think this year it’s going to be about thirty-five percent, and a lot of those fallow lands don’t provide this kind of habitat once they’re taken out of rice farming. So that’s sort of a double whammy for the economy of the valley as well as for the opportunity to do good things for wildlife. In terms of how the drought is expressing itself, the water availability analysis I pointed to that we’re doing with Point Blue, we’re finding that over the period of droughts from 2000 to the last year, we’re able to see the signature of those droughts in that water availability signature. And it’s most profound oddly enough at the times of year that are most important to shorebird migration. The fall and the spring are the times of year where we have the, less water available because of the drought than we have in non-drought years. 

In terms of how it’s affecting the birds, that’s a big challenge because we were only able to really understand what’s happening with the birds when they’re on our fields, and we may be attracting more birds because there’s less habitat available, that’s certainly one hypothesis we’re investigating. We’ve also been working with Point Blue on radio telemetry of birds and looking at the movements of dunlin and of dowitcher relative to habitat availability and drought. That analysis is is in process. These are big questions and every year is going to be different in California. This year we’ll we’ll see whether it’s another dry one or whether there’s more habitat available. 

The question is have we experimented with any other habitats besides rice fields? We are just starting to take on some work that’s outside of the the rice growing region. We have some work that we’re doing exploring providing beneficial habitat for cranes in corn. The Nature Conservancy owns a very large corn farm in the, on the Sacramento-San Joaquin Delta called Staten Island, not this Staten Island, and so we’re very interested in the long-term future of sandhill cranes in these agroecosystems and we have some fields that are outside of our Staten Island Preserve that are being managed for cranes that are that are corn fields. So we’re aggressively looking for new crop types to try this out in. 

Sure yeah. And so the question is how do we manage the contradiction and controversy between water and birds in in California? Did I get there right? Well, I think that this is this is a new reality for conservation, and the drought in California is very serious, but it’s a, it’s analogous to conservation issues all over the planet where we have trade-offs, and we have people making a living on the landscape and those needs need to be factored into how we do our conservation. I think we’re all aware that in this age of humans dominating the earth, everything’s going to be about trade-offs. It’s no longer as simple as it once was to identify something important for nature and you know draw a circle around it and protect it. These are dynamic systems, and trade-offs is really what it’s all about. So in this case we’re trying to balance the needs of Californians, other uses of water with the needs for birds. By being more precise about when and where we use less water, and we’re able to be more efficient, and hopefully benefit wildlife along the way. 

Yeah. So um two questions. One is regarding water quality, are there water quality issues with BirdReturns and bird habitats in the California Central Valley? And the second relates to whether the bird, the birds are providing any other ecosystem benefits in terms of nitrogen and fertilizer through their presence on the fields. So water quality issues are present throughout the Central Valley. In the Sacramento Valley, I think most people are aware of now a couple decades ago, there was a set of water quality issues at the Kesterson Wildlife Refuge in California, a lot of birds were killed, that was selenium concentration, and that was in the San Joaquin Valley, which has agricultural drain water issues which aren’t present in the Sacramento Valley. So we don’t have those kinds of issues. 

The Sacramento Valley, in addition to supporting rice farming, also is the drinking water basin for the city of Sacramento. And so they have a lot of water quality regulations in terms of the kinds of pesticides and herbicides that are used in the rice ecosystems, so at least at that level this is also an area inhabited and used for drinking water, so we feel somewhat confident about water quality issues that way. Chris Elphick, who’s a leading shorebird scientist, and in many ways the godfather of agroecosystems shorebird ecology, has studied shorebirds in California for a couple of decades, and he’s assured us that there are no known examples of massive die-off of shorebirds from from any sort of herbicide or pesticide issues. It’s something we’re concerned about and aware of, and that we want to study more deeply in the future, but but for right now I think we’re just trying to meet the basic habitat needs. 

[Steve] Well, thanks. It’s been a great talk. Mark and Brian. We’ll be around a little bit afterwards if you have other questions but I think it’s a good time to thank… 

[Applause]

End of transcript