PubTalk 07/2019 — Taking the Pulse of Our Planet

Video Transcript
Download Video
Right-click and save to download

Detailed Description

Title: Taking the Pulse of our Planet: A 10-year status report from the USA National Phenology Network

  • Plants and animals in your backyard are sensitive indicators of climate variation and change
  • Understanding and predicting plant and animal seasonal activity, a science called phenology, helps humans adapt to a changing world
  • Learn about a national network of volunteer and professional observers tracking plant and animal phenology, and find out how we get involved


Date Taken:

Length: 01:26:01

Location Taken: Menlo Park, CA, US

Video Credits

Contact: Amelia Redhill -


… Geological Survey’s July 2019 public lecture. I’m Diane Garcia, and I’m with our Science Information Services. And I’m delighted to see such a full room. Thank you very much for attending. Before we get started, I always like to plug next month’s lecture. And we are going to have an August lecture on August 29th. And it is Pliocene World – Earth’s Climate 3 Million Years Ago and How it Relates to Our Future. So I hope you pick up a flier on the back table. And more importantly, I hope to see you here late August – August 29th. So, but what you’re here for is tonight’s lecture. And it’s Taking the Pulse of Our Planet – a 10-Year Status Report from the USA National Phenology Network. And it’s presented by Jake Weltzin. Jake assumed his position as executive director of the USA National Phenology Network in 2007. Jake’s interests in natural history developed as he grew up in Alaska and served as an exchange student in the Australian Outback. He obtained his bachelor of science from Colorado State University and his master’s from Texas A&M University and a Ph.D. from the University of Arizona. Following a postdoctoral fellowship at the University of Notre Dame, Jake went to the University of Tennessee where he served as assistant and then associate professor. Jake is interested in how the structure and function of plant communities and ecosystems might respond to global environmental change, including atmospheric chemistry, climate change, and biological invasions. His research spans temperate and tropical grasslands and savannahs, temperate woodlands, deciduous forest, and sub-boreal peatlands. His recent experience as a science administrator at the National Science Foundation underscored the need to foster large-scale science initiatives, such as the USA NPN, as its first executive director. Jake’s vision for the USA National Phenology Network is to develop a continental-scale instrument for integrative assessment of global change that simultaneously serves as an outreach and educational platform for citizens and educators. So the USGS monthly public lecture is pleased to bring you a program this evening about the National Phenology Network. I’m going to ask that you please hold your questions until the end. And now I’m going to ask for you to give Jake a warm round of applause to welcome him.


- Well, thank you. I’m very, very glad to be here. This is actually my third lecture in this evening lecture series. I joined U.S. Geological Survey in 2007 in August. And, a few months later, they said, would you come be a speaker in our evening lecture series? And, oh, okay. Well, I don’t have much to talk about. I’m here to develop a new national continental-scale instrument called the Phenology Network that brings professional scientists and citizen scientists together across the nation to track the activity of plants of animals. When are birds migrating, and when are flowers blooming, and so on. So I think it was maybe just six months in. So you might say that I was introducing the world to the idea of phenology and maybe the term. Although I just call it smoke and mirrors. I went back and looked at my presentation. I was, like, oh my god. There’s only one slide on the Phenology Network in there. And it was just a screen shot of the home page. [laughter] It looked pretty bad. USA MPN. Big red circle there, and an arrow saying, this is really important. And then, two years later, they asked me to come back again. And I was – holy cow. So I took a peek at that presentation that I gave in 2010. And there was a bit more detail in there, but I was really interested in getting people to come and join the network and make these observations across the nation. So I probably spent a little bit too much time talking about the details. I look back on that. But at least we had a logo at that time. You know, I always say, start with a logo and go from there. So we’re actually on, like, our third version of the logo now. So we’ve really come along. Oh, and this is the third talk. So it’s been about 10 years. We actually just recently celebrated our 10th-year anniversary of the National Phenology Network. And I’ll describe phenology in a minute. I know that’s not an easy one. People come up and – you should change the name of your network. Well. [laughter] It’s a little bit late for that. So I just want to point out, I am getting a bit of feedback in the room, if maybe we can turn the game down back there, Mr. Sound Guy. And so here I am. Third one. Kind of giving an update in terms of, you know, what have we – what have we learned? But you may not know this, that actually this is my second time to give this talk today. So I gave it during lunch to staff and folks at U.S. Geological Survey. And then to you all – to the – to the broader public. And then I guess it’s streaming on YouTube right now, which is – which is fantastic. But what that means is, actually I haven’t been here to do this three times. I’ve been here to do this six times. This is my sixth presentation [laughter] on the Phenology Network in this Rambo auditorium. And I really – I think it’s really fantastic – a good opportunity. So, without much further ado – and am I coming through okay? Super. Okay, good. Okay, I’d like to just talk to you a little bit about how we go about taking the pulse of our planet. How do we pair professional, you know, paid people who are collecting phenological information and citizen scientists to understand what’s going on across the nation and to understand, what is the biological response to climatological drivers? Because the biological response translates fairly directly into what we do and what we like to do. We like to go fishing. We like to go exploring. We like to go walking outside. We don’t like ticks. We don’t like a lot of pollen. We don’t like a lot of mosquitoes. All of these things are biological components of the – of our world that affect us. So we’re trying to figure that out. What is the pulse of our planet? So what I’d like to do is sort of start relatively general with a little introduction to, what is this term “phenology” so that we can kind of get all up to speed together. Then I’ll talk a little bit about some of the climatological drivers. So I’m going to actually talk just a tiny bit about climate change and climate variation as one of the drivers that’s occurring at a – at a global scale, but translates right down into your back yard where phenological processes are occurring when you step back out there. Ah, the day lilies are out. Yes, they have accumulated enough heat now so that they can leaf out. They’re relatively late in Fort Collins this year. So what I’d like to do is start with what it isn’t. [laughter] And it ain’t phrenology. I did have hair when I started. And maybe phrenology being the study of bumps on the head and how it relates to moral character, I’ve probably been thinking about that for maybe a little bit too much. It’s not phonology – the study of sound. It’s phenology, derived from the term “phenos,” which is "to show" or "to appear" in Greek. So we keep keeping track – like, watching the stars. Even keeping track of, oh, when the olives are on the – on the trees as well. So not phonology. And actually, to be honest, I haven’t used this slide since – for years. And I like to think it’s because I don’t need to anymore. Everyone knows what phenology is. But I realized, when I looked at my 2010 presentation here, that this was one of the slides I used. I probably should always be using this slide, saying, this is not what it is. Rather, let’s go to a lexical definition. What does it say in the dictionary about what is phenology? Recurring plant and animal lifecycle stages and their timing and relationships with weather and climate. Seasonal patterns. Again, like, when are the birds laying eggs? When are they arriving – when are migratory birds arriving? When do we see fish in the stream? When actually should we try to go catch stripers? You know, depends on what you’re after. When should we be watching for that first tick? What about those hummingbirds? So really, I’ll be focusing a lot on plants here today, and individual organisms – like, leafing and flowering. But it extends to all manners of organisms – basically, everything around – everything around the Earth. And it also extends out to a global scale because we have plants – like on the right-hand side there – taking up carbon dioxide. And you can actually measure that in the pulse of the planet. And I’ll show you a bit more about that later on. So another way – a more simple way, perhaps, to describe phenology is just nature’s calendar. So I’ll talk a little bit more about Nature’s Notebook, which is how we’re recording nature’s calendar. We thought, should we call it nature’s notebook? Or nature’s almanac? Or nature’s calendar? Well, there’s a nature’s calendar in Europe, so we thought, oh, probably better not do that. And nature’s almanac sounds a bit more old-fashioned. Nature’s Notebook – I’ll talk more about that in a few minutes, which is actually the platform where you can go and be – participate and enter your data into the Phenology Network database. So phenology is just nature’s calendar. So how is the timing of flower visitation changing? What are the drivers? What about these migratory organisms? You know, when is the best time to go out and harvest animals from the landscape? What’s the best time to go see leaves falling in the autumn? So phenology is a complex term for something that’s relatively simple. It’s happening everywhere. It’s in your own backyard, and it’s really important. So thinking about – thinking about phenology and maybe kind of bringing it home a little bit, so we might like to watch the arrival of hummingbirds. There’s actually a project called Operation RubyThroat, where people are tracking the arrival of hummingbirds. We saw a monarch a little earlier. People are tracking monarchs through Monarch Watch and other projects like that. We have plant-animal interactions, like bees that are arriving – or, actually being – arriving on flatbed truck into the almond orchards. [laughter] So – from Florida and from the Dakotas and whatnot. So they’re here, and you rent a hive of bees to get your almonds pollinated. Otherwise, you’re going to lose your crop, so critically important. Insects that might be beneficial, or perhaps pests that are – like emerald ash borer. I live in Fort Collins, Colorado. Emerald ash borer has not arrived. All the ash trees in Fort Collins are wonderful along the streets. But when emerald ash borer arrives, the forest managers and city planners are going to need to be thinking very carefully about, how do we track the timing of emerald ash borers? Or when do we plant our gardens if we’re working with heirloom vegetables? So what I’d like to do is spend a minute or two just kind of trying to lay a little case as to how this might kind of work out there in the – in the real world. When we think – instead of just a single organism, thinking about organisms that are interacting with each other. And thinking about migratory birds. Say migratory birds arrive from migration, say, on this little timeline there on the left. They’ll lay eggs. The eggs will hatch. And then we need to feed those babies something. And so the parents will be collecting caterpillars and feeding the nestlings. And then – and then the caterpillar population will sort of decline. But that’s okay because the nestlings can go out and forage for themselves. But what happens is, this is a carefully and tightly evolved system of bird migration, where they actually time the migration to arrive at certain locations driven mostly by light. But whereas, the caterpillars and – the caterpillar system and the leafing of trees is driven by local conditions – so local weather, local climate. And so that can be relatively early. And so if we shift our climate to where it’s warmer, we end up perhaps pushing that abundance of caterpillars earlier in the year. But the question is, do the birds get the news? That is, as time – they should be delaying their arrival – or, moving their arrival earlier. Not necessarily, if they’re coming up from South America. So that’s just a case example, where we have actually been looking at that as scientists. Do we see this happening? There are some very good examples in Europe where they’ve been tracking this kind of thing for a long time. Here in the United States, we did a study recently about 40 species. Less than half were actually showing any sort of adaptability to changing environmental conditions. So I’ll talk a bit more about that as we go. But this is how it can get kind of complicated and shifts in one type of organism can affect another one you wouldn’t necessarily expect. And then we have this. We have biodiversity out there. We have all kinds of different organisms that are – that are important to us that we place value on for any number of reasons, and all of them have their own phenological cycles, and all of them are being driven by different drivers. And so the idea behind the Phenology Network was to try to untangle some of that. To try to figure out what’s going on. So what I’ll do here now is maybe change scales. Change the topic just a little bit to put a little context over why we’re trying to take the pulse of our planet. We take the pulse of the planet to figure out, is it healthy, you know, or not? Is that health changing over time? And one of the things that is changing is our planet, of course, is experiencing many changes where that blue marble out in space and – and we can see it, and we can track that, and we can know what’s going on. We may not understand all the processes, but we can understand what some of those changes are. And what I’d like to do is spend a few minutes just kind of putting this into the broader context. And I’ll always be scaling from individual plant all the way up to the globe, back down to the nation, back to the globe, et cetera, as I – as I talk. So here’s an example of one of the things that’s driving our changing planet. And it has direct and indirect impacts on plant and animal activity, or phenology. Changing atmospheres. Basically, the carbon dioxide concentration is increasing around the world. This is actually a data set from Mauna Loa. It starts in about 1980. Actually, the data go back to about 1957, to 2018. And what this is showing is the seasonal variation in concentration of carbon dioxide, but you can see it’s ticking up every year. So you’ve heard about global warming and greenhouse gases and things like that. And so that’s actually – this is actually a measurement of one of the greenhouse gases – this change in our planet that’s driving changes in our planet. This is carbon dioxide concentration. And, again, this is at Mauna Loa. But you can see this little tick going up and down. It’s actually driven by changes in vegetation activity, mostly in the northern hemisphere. So CO2 is very well-mixed globally. And so, when the concentrations go down in the spring and summer in the northern hemisphere, like in Canada and the eastern deciduous forests, it pulls CO2 out of the atmosphere. And in the wintertime, the plants go dormant. But they’re not dead. They’re still alive. They’re respiring, giving off carbon dioxide. And so you can see that seasonal cycle. That’s a phenological cycle in something as – you know, as global – a global fingerprint right there of CO2 out of Mauna Loa. And we can talk more about that, about how we understand CO2 concentrations, how that’s changing, what impacts that’s having. One of the impacts that we do know that it’s having is on global atmospheric temperatures. Svante Arrhenius in 1895 wrote that we have CO2 in the atmosphere, and it’s an important greenhouse gas. And if it increases, it will change the temperature because it holds temperature in. It’s like it holds infrared heat in on the Earth. So we’ve known for, you know, 120 or more years about the mechanism by which global warming might occur. And so then we can measure that. So we go out with some of the very best measurement devices in the world – all around the world, and we can look at, for example, changes in surface temperature for the contiguous – the lower United States. This is from the National Climate Assessment – U.S. Global Change Research Program’s National Climate Assessment. It was probably the top-notch assessment of climate in the U.S. that is conducted every four years at the behest of Congress. So this figure from 1895 to 2018 is showing the average temperature of the – of the U.S. relative to the long-term average, which is the black – the black line. It’s actually the 100-year average above the line in red is warmer. Below the line is cooler. And what you can see is, there has been a fair amount of variation – a lot of variation from one year to the next. Some years are warm. Some years are cold. Actually, this is a pretty warm year, isn’t it? Very, very warm June. It was 108 degrees in Paris. It was 59 degrees in Austin, Texas, this morning. Not 108. So there’s a lot of variation across space and through time, and we see that. But generally, in the last – since about 1985 or so, we’ve seen this just steady march. This is sort of the impact of these trace gases in the atmosphere that are warming up our world. So this is a U.S. temperature, but we see that around the world as well. And it’s probably happening in your backyard. So one of the ramifications of that is what we call the frost-free season. And, again, this is from the same source – this authoritative source on changing climatological conditions and impacts as well. So this is an impact of that warming temperature. This is from 1980 to 2018, it looks like, where above the line is the duration of the frost-free season – the time – actually, this is measured by satellite. It’s the time when the Earth – the pixels on the Earth are not frozen. So what is showing is, this is the time – the duration of that – of the frost-free season, meaning what you might also call the growing season, but when there is not ice. And you can see that that’s also going – a lot of variation as you go along, but there’s kind of been a steady trend for the last 25 years or so. And so we’re starting to kind of pick out that signal. So this is driving what’s going on in your backyard and how and when you might go out and do planting. Maybe a little bit less so here in coastal California. But there’s also a lot of variation, as we mentioned a little bit earlier, where the – this is from the Environmental Protection Agency. One of their indicators of their rates of temperature change – red being warmer, and this is by county – blues being cooler. And you can see there’s actually been a fair amount of warming, but it’s spatially variable at the national scale. In fact, that bluish area there in the southeast is called the warming hole. [chuckles] And we can talk more about some of the causes for that. It has to do a lot with atmospheric circulation. But, by and large, you can see that temperatures really are warming quite a bit across the nation, which is having impacts on you. So then how do we try to understand what the biological response – I’m not a climate scientist. I’m just putting those in there for context. I’m really a – I’m an ecologist. I study the interactions of organisms in their environment – more like a biologist. And we have this short of challenge of biodiversity, where there’s all these different organisms, all of which are responding different – in different ways to different drivers. If it’s – if it’s a desert grass, it might be responding to water precipitation changes. If it’s a migratory bird, it might be responding to different kinds of heating or warming. You know, or harvest, even. We can over-harvest some organisms, or they may be interacting. And so picking that all apart is the study of ecology. Phenology is a part of ecology. So it’s kind of a challenge. It’s also an opportunity. Now, we also have the challenge of geography. You know, we have a national – trying to get a national picture. And there’s a lot of variation out there. And this is just really – this is a map showing phenological regions across the United States. And it’s not important, other than to show you that, if the conditions here in Menlo Park, your seasonality – you don’t have a lot of variation from one season to the next. Actually, you have your rainy season. But it’s a lot different than up there in St. Paul. And a lot different than, say, in south Texas. And so all of these regions show that there’s a lot of different – there’s a lot variation when you measure using actually satellites in the seasonal signal of how vegetation acts. So that’s a backdrop in terms of how you would go about creating a national phenology network. Because if you’re going to try to understand, what is the pattern and driver of plants and animals that we hold value for, then you have to consider all this diversity and all that great geography, the different drivers, et cetera. So the challenge that I got when I came out of academia – I left National Science Foundation – was to start a new science and monitoring network in 2007 that would, again, bring folks together to try to collect, store, and then share phenological information back out with people with scientists and resource managers to use. The idea was to help advance science of phenology and our understanding of the world, inform decisions – like what natural resource decision- makers are making out there, say, at Don Edwards National Wildlife Refuge, or elsewhere, where they’re trying to making decisions – do we flood now? When should we be preparing for migratory organisms, et cetera? And then communicating and connecting. Trying to get the word out about some of the changes that we’re seeing. So then there’s just a little picture there of kind of how that works across scales. And that is, if you go from individual little cones – pollen cones on a juniper plant, you could – you could say, okay, that’s going to produce pollen now. But you might ask, okay, is the – is it just that one little pollen sac? Or is it all the tree? Or all the trees across a landscape, which is the second picture down? We can actually model that pollen production, and we can make predictions about when the pollen might be showing up in your nose. [laughter] And then actually go to the hospitals and say, okay, can you tell us when people are coming in with respiratory problems? This is a big problem in Albuquerque, New Mexico, surrounded by pollen. Or if you’re from central Texas on the Edwards Plateau. You know, my in-laws would have to sometimes go home because there was just way too much pollen, and they were out – they were out camping, and it was just impossible because there was too much pollen. They needed a phenological pollen forecast. I’m working on it. I’m working on it. [laughter] Or a tick forecast. I’d love to have a tick forecast. So how does this work, then? How do we build a national science and monitoring network focused on phenology across this challenging diversity and space? Well, this is what we call a workflow. Sometimes they call it a replicable workflow. In other words, you get to do it again. But that’s probably maybe a bit redundant. But we work primarily – starting with partners. What is the question that our partners have? And U.S. Geological Survey, of which – of which I am a part of, is a member of the Department of the Interior, and our science role is to, first and foremost, provide science and support of the other DOI bureaus, like Park Service, Fish and Wildlife Service, BLM, BOEM, et cetera. And so I’ve thrown up an image here just to – a representation of Fish and Wildlife Service, how that might be one of our partners. I’ll talk more about this as I go. And then we work together to find out, okay, what’s the question you have? Trying to figure out how to restore a wetland system. You need information about the species that are there and when they’re active, how many there are there, et cetera. So we define the data – their data needs. We then go build models. So there might be models that are already existing out there in the literature where people have investigated, when do the kestrels arrive here at this refuge, and when do the Siberian elms, which is an invasive species, produce their seeds? But sometimes we don’t have that. So we have to actually get people to go out and collect data to help build models. From that, we build products. Here, a map. You could call it a product. It could be a data set. It could be a map. I like to produce maps. They’re easy to communicate. I’ll show you several maps as we go along – communicate those back. And then there’s this part in the middle – validation. Which is really important because you’re taking data from across the nation and building a model and trying to describe the system. And we have to validate it. You don’t know whether it’s any good or not, necessarily. I mean, it’s really good to have a model. You understand the system better. But, by validating and testing it – testing what we expect against what we observe, that’s when we really start to learn. So I’ll talk a bit about that. So actually, this is the outline for my talk. It’s not linear. [laughs] The talk is sort of linear. But I’ll use this, and I’ll kind of try to work you through some of these so you can kind of better understand what we’re talking about. And then sort of, as I draw towards the last third of my talk, I’ll talk a little bit about how we advance science and inform decisions – kind of go back to those strategic goals that we have. Well, partnering is a big part of really our success. That’s why we’re sort of still around as this nascent nexus of sampling and science – the Phenology Network, where we have scientists, specialized networks, non-governmental organizations, federal agencies, resource managers, all working together – tribes – all working together to try to understand sort of what’s going on in my parcel of land or things that hold value to me. Citizen scientists are a big part of that. We actually have a lot of citizen scientists who are participating through this project, Nature’s Notebook, that I’ll be talking more about in a few minutes. That’s a critical part. Couldn’t do it at a national scale without the help of volunteers. So here are some of our volunteers, for example, who are working either as individuals or part of organized groups trying to answer their own question. Or they are part of national networks that are trying to understand what’s going on here. Like, the inventory and monitoring programs that Park Service and Fish and Wildlife Service have. They’re actually mandated to track the biological and cultural resources on their – on their federal lands. And another type of partner – I alluded to a little bit earlier – Fish and Wildlife Service. And this is at Valle de Oro National Wildlife Refuge. This is a very interesting urban wildlife refuge in the southwest that’s just basically just south of Albuquerque – downtown Albuquerque, New Mexico, right along the Rio Grande River, which comes down out of Colorado. And this is an old dairy, actually. Fish and Wildlife Service purchased the land, and they’re going to work to try to restore the riparian forest in this – in this system. But they’re also right there in a neighborhood. You can see houses back there. They’re right below the flyway where Albuquerque International Sunport, I think they call it, where that – and so the airport authority said, well, you can put a refuge there, but no net increase in birds, please. [laughter] So, you know, okay, well, are there birds, and when are they there? And actually, there’s really cool birds there. Sandhill cranes. American kestrels. And so, if you convert this system – this open system – to a riparian forest to try to deal with some of the flooding issues that are occurring right there in Albuquerque, you’re going to lose those species. So you need to understand, when are they there? What kind of habitat are they using, et cetera? So we can actually roll out Nature’s Notebook, and we are doing that at Valle de Oro National Wildlife Refuge to try to understand those patterns. And then help them make decisions about, how do they reforest and restore this old dairy system? Maybe we should just leave it like that. Okay, now we have our partners, and they have questions. When are the kestrels here? Are there kestrels here? People like to come and see kestrels. You can come – you can get on – there’s a small light rail that goes from downtown Albuquerque down to this refuge. They put a stop right there to bring people down and visit the refuge. It’s an urban wildlife refuge. A natural system designed to introduce people to the natural world. We use a project called Nature’s Notebook. This is an online user interface, basically, that allows you to go in, find – describe yourself a little bit, find a site where you’re going to track plant and animal activity, choose some plants and animals you’re going to track, look at the standard protocols, and contribute data right there. And then you can look at your data as well. You can track, where are my – where are my – where am I collecting data? What am I reporting here? You can go back and make changes. Oh, I realize the leaf wasn’t out, et cetera. So Nature’s Notebook. It’s Check it out. Or and go – and go check it out. And we’d love to have you – have you join us. So right now, we have hundreds of partners who are part of this. We’ve set it up so that, if you’re a local manager, or you have a natural area – we have many partners, actually, here in the Bay Area – then you can – you can join, and you can answer your own questions. So we’ll provide the standard protocols. You’re answering your questions. Maybe about flooding impacts or fire or species interactions or the timing of ticks or whatever. And you can share that information back with us so it contributes to the larger database while you’re answering your own questions. So we had about 450 partner organizations, like, the North American Pawpaw Growers Association. Anyone had a pawpaw? Fantastic forest fruit, native from the eastern deciduous forest. Tastes like pineapple, strawberry, and guava all mixed together. It’s phenomenal. But it has a very short window of ripeness. The raccoons love them. All the forest animals love them. And so the North American Pawpaw Growers Association approached us saying, we’d like to try to market. We need to know when we should be going out for these different varieties – Asimina triloba – and collecting the fruits and understanding when they’re ripe. Because it’s a narrow ripe window. So that’s another partner – a type of a – an interesting type of partner. We have thousands of sites across the nation where people are collecting data. Tens of thousands of observers. And we now have millions of records. And I had some little figures. I was going to show little traces of the number of records going up. It’s exciting. We get about 3 million records a year right now for about 16 million records total, or so, since – all collected through Nature’s Notebook since we rolled it out in about 2011. So we’re using this information. I’ll talk more about how we’re using this information a little bit as we go along. So that’s the platform. And, again, we’re inviting a variety of different kinds of partners to join us, either as individuals or groups. They might have – you might have a local phenology project with a leader. And so we have a little certification program. And the goal is to try to get the highest quality data from contributors – from volunteers and from scientists who are collecting data. We want standardized protocols. We want training about how to choose a site and how to identify organisms and how to collect phenological information. Is that leaf out yet? Oh, is that a leaf, or is that a – you know, what is that? Et cetera, so we have a variety of programs we use to support that to build data quality. And then here’s an example of where some of those data are being collected out across the nation. So each one of the little black dots represents where data have been collected, and the larger rings represent, okay, there’s foci. And you can see that we have some places we don’t have a lot of data. So we have to build models to interpolate between those data points. Some places where we have just an abundance of data. We could probably stop collecting data on red maple. People like red maple, and there’s a lot of data on red maple. [laughter] But that’s okay because we can also – if we continue doing that, we can continue to validate our models and improve them over time. Because the phenology of red maple is actually pretty flexible. It’s changing. So remember that little schematic – the workflow – the replicable workflow. The next step over was modeling. Trying to figure out, what is the relationship between those patterns that we’re observing and the driving variables, like changes in temperature, changes in precipitation, changes in day length, et cetera. So we work very hard, using a variety of different kinds of advanced statistical techniques – and some simpler statistical techniques – to understand what are those patterns. And so this is a case study that I’d like to just describe. And this is the only one I – the only time I use the word “model,” almost. And what I’ll do is I’ll try to tell a little story about that. This is Joe Caprio, and his spouse, who were asked to go to Montana to start a – basically a meteorological observation network up there – weather. In the ’50s. Because there weren’t a whole lot of thermometers up there, and they needed information about how the agricultural crops – what are the controls over apple production? You know, is it a good place to be producing apple crops and other kinds of crops? And there weren’t a lot of thermometers, so he realized everyone had a lilac by their back door. He said, I wonder if there’s a relationship between the temperature that we observe and when these lilacs are leafing and flowering? And so he spent some time developing a model. This was in the mid-’50s. It’s called a – like, a growing degree day model, where you accumulate heat through every single day from a weather station. You compare it to what you see on your lilac. And sure enough, he built a pretty good model. We’ve since made a few little adjustments to it. Basically, the same – structure is the same. And what he would do then is, he would extrapolate and draw – hand draw maps as to when the lilacs would be leafing and flowering across the western United States. And so this is from 1962, lilac leafing and flowering. So we’ve since taken that model and combined it with a few other species to make an index of spring. And I’ll be talking about that in some of the coming slides, and so I’m going to focus a lot on plants. I’m a plant ecologist. We have a lot of data on animals too. In fact, fully a third of the data are on animals. But I’ll be sort of focusing on that as a little thread as we go along. Kind of leveraging off of Caprio’s lilac phenology model and the more complicated Spring Index – not much more complicated Spring Index – [inaudible] spring model. So what we’ve done then is we’ve taken his model, that was basically done by hand, then it was also – it was done in Fortran and then MATLAB, and then now in Python is how we run it right now. And so this is basically Caprio’s model for the United States. And this is for the Spring Index leafing of lilacs plus a couple of honeysuckles that have been added in. This is kind of early indicators of the onset of spring. This is data from July 22nd. And this is a threshold model. So whenever there’s enough heat that accumulates, boom, out comes a leaf. And it just kind of slowly – and on come the flowers, slowly. And so you can actually – you can actually describe when that occurred across the nation. So each of the – the color ramp represents the date at which there was a leaf – a leaf that was supposed to have some out on a lilac or, actually, the Spring Index – kind of little index of early spring leafing plants across the nation. And, which you can see, it’s good to see that spring comes earlier in the south and later in the north, and that the mountains, where it’s cold, have even a later spring. And it’s happened all across the nation. So, if you were to continue this on up, by July, probably it’s all the way up to – would be all the way up to the – to the north slope of Canada and Alaska. And so – and this is just showing you that we now produce and deliver these data basically real time. I could have grabbed – and you could go online and get today’s data. Or you can actually get a six-day forecast, which we deliver. Or you can get the data back to 1900, which we also produce and deliver. But maybe what’s more interesting is the variation, or the anomaly. How does this compare to the long-term average? Like, okay, it’s just happening. Is this happening down there? Well, the question is, well, what was the year like? What was the spring like? Was it early? Was it late? I’m a scientist. I’m going to say it depends. [laughter] It depends on where you were. So what happened was, this year was kind of weird. It was a really – it was shaping up to be a really early spring across the southeast here. And so that’s what the red colors are showing. This is basically the same data, just compared to the 30-year average for the leaf at those – at each of those locations. So the – as soon as it turns on, we can compare it to the long-term average and color up a pixel. And so the red showed this, like, okay, it’s looking like we’re going to have another really early year like we did in 2017 when it was so warm across the nation. 2017, the whole map was just red. But then everything changed. And we had polar vortex, and we had shifts in large-scale atmospheric circulation, and also where the jet stream was sitting. And so we ended up with a very big delay in the onset of spring. And, in fact, as you know, it was cool – very cool and very wet. A lot of flooding that occurred. So this is kind of a retrospective, but we are generating this in real time so you can sort of track this. It’s a threshold model, so once the leafing happens, the flowering happens, well, then that’s done. But there are other models that are more dynamic that are produced on a daily basis. So you can see, okay, it’s warming up. Okay, it’s just sliding off. And those are growing [inaudible] models, which we also deliver this way. So these get picked up in a number of places. It’s kind of fun. I’ll show you a few examples as we go along. But national-scale inference of phenological activity, this is a little index of early leafing and flowering, but you can relate it to other taxa as well and/or develop models for other species. So just a – just a little story arc. Okay, so the next step really is to validate. And sometimes we don’t do that enough. I’d say we just don’t do that enough as scientists. We don’t often go out and validate what our models are. There’s different ways to do that. We’re really lucky because we have a lot of observational data, and it’s still rolling in – 3 million records a year that we can use to validate our model. And so we can compare, what did our model predict for, say, your backyard? If you’re in Buena Vista, Colorado, or in the – on the Santa Rita Experimental Range in Arizona, where you’ve got lilacs, or Reston, Virginia, what did the model say, and how is it doing? So we actually ask people to go out and be a part of our lilac observing campaign, either Lilac Watch or the Cloned Plants Project. Because we actually distribute cloned lilacs that are sterile around the nation so they all have the same genetic material. We do that for dogwoods as well. So you can be part of the Cloned Plants Project, and we bring that data in, and we can compare what we expected to what we see. And we have a number of different kinds of tools that we have available online, so you could actually go and say, okay, where’s the model doing well? Where is it doing poorly? Some of you are more schooled, actually, even than I. My staff is great. They used Tableau to create this little figure. So you can go in and click and set your time windows and whatnot and look to see whether lilac leafing or, in this case, lilac leafing across the entire time period data set that we have is early or late compared to what the model predicted. So I think it’s starting to understand better, where is the model performing well? Where is it performing poorly? How do we need to tweak that? Actually, what might be [inaudible]? Why is the model producing – doing well or poorly? Using a couple of different climatological drivers here. We’re using data from NOAA mostly. The National Center for Environmental Prediction – they also distribute data to The Weather Channel, which is where you get your weather forecasts. Or the National Weather Service. So we have historical data and contemporary data that we seam to integrate them together at different scales, depending on the application and timeframe. Built some maps. Did some validation. Now we’ve got to communicate those stories back out to our stakeholders and our partners. And so that’s a big part of what we do is getting the word out. That’s part of the reason why I’m here, for example, today. So we produce kind of standard documentation, like an information sheet that talks about actually another story I won’t talk about very much, but we’re starting to extend this and extend this out to other taxa. And, in this case, we’re actually working on developing what we call Pheno Forecasts – phenological forecasts for invasive and native insect pests like hemlock woolly adelgid, emerald ash borer, Asian long-horned beetle, et cetera. We don’t have ticks yet. We don’t have mosquitoes yet. Those are harder to model, but we’re working towards that. So here’s the information about phenological forecasts for invasive pest species that might be important if you’re, say, a forest manager in the east or in the southwest, where I understand there’s now apple maggots. So apple maggots are happening now in the southwest. That gives information for people. They can go out and start doing management. But, you know what’s most fun is when Al Roker does our communication for us. [laughter] So this is the Today show, and Al is giving his weather talk. And he pulled down the – this is in 2017 – pulled the data off of our – off of our servers, basically. We have an API, and so he – application program interface – and the data are available. And so he pulled it down. They made their own map. Like, did he just copy it and kind of color it by hand? It’s, like, no, no, no. It actually – we can see the pixels there. And so he used – he used the data, re-colored it. And so here he’s telling the story about early spring. And we’re, like, oh, he’s using the NPN logo – the old logo – that’s okay – because it’s a little older. [laughter] Always have a good logo. Logo, you got to – you’re not anything without a logo, right? Okay, so there we are. And he’s – look at how early it was – three weeks early. As much as three weeks early. Actually, parts of the United States, especially in the mid-Atlantic, were five weeks early. Five weeks earlier than the long-term average. It was a really super-anomalous spring. Just doesn’t happen that way. And we have other maps online. You can see, how often does that happen – kind of return intervals, we calculated. Fairly simple and straightforward to do that, but it’s kind of amazing how weird that was. And it was great because we had just rolled out our spring maps. So we tweeted a couple of times on Groundhog’s Day, and Al Roker was using it a few weeks later. Like, oh, yes. Have we arrived? It took a long – that’s why it takes 10 years before you can come back and give your third talk. Oh, and The New York Times picked us up. This – again, this is in spring of 2017. And they’re, like, holy cow. It’s just really early. It’s weird. Is it weird? Tell us again. Like, yeah, sure, we have a yardstick that we use. It’s called the Spring Index. It’s developed from some early leafing and flowering plants called lilacs and honeysuckles. And they said, okay, well, let’s see what we can do. So they pulled down our data and made this cool graphic. Like, wow, we should do this too. So we have animations, but we copied them. So this is showing actually just the days early. So anything that is – and they stopped the animation on March 6th, which is when it was – which was when it was published. It would have been great if they had kept going, but they only showed the days early, which I thought was kind of – you know, you should probably show where the places – where it’s also late. Have a different color, perhaps. But it was a simple animation. You know, get the word out on the – on The New York Times home page. Spring came early. Scientists say climate change is the culprit. We didn’t say that. We just said spring came early. But you could say climate change is the culprit. I don’t really know. So, anyway, that was very exciting. And then, even just like this year – this is a graphic from the Washington Post, and they did a different approach. A more static, staid approach, where they took 1983, 1984, 1985, and all the way up to 2018, and they just mapped out, this is what it looks like. And over time – so that’s later years down the lower – the lower corner. Actually through 2017. That lower right-hand corner is 2017, just so you can see how anomalous that year was. And you might think, okay, if there’s a trend, you should see that from upper left to lower right. Not really actually a trend in the phenological index that we’re – that we’re producing. It’s just a single moment in time when a leaf or a bloom comes on. So we need more complicated ways to describe, are there changes, and probe whether there really are changes. Spring is earlier and earlier. I’m, like, oh, not too sure. I’m a skeptic. So advancing science. So changing gears a little bit. Now, talking a little – just kind of stepping back and saying, okay, well, you’ve done all this stuff. Nice maps. Washington Post and New York Times pick you up. That’s great. But what have you done sort of for science, which you said you were going to try to do? And how are you informing decisions? Like, I don’t have a lot of time to go into a lot of detail, but I think it’s really important to say, okay, you know, this is actually working. So let’s take a look at that. So the Phenology Network actually has had phenomenal success, considering that there’s a bunch of volunteers spread out across the nation, who we hardly ever talk to, and we communicate to them through a variety of different social media and an online user interface called Nature’s Notebook, and they’re collecting and contributing millions of records every year. Kind of like eBird. Some of you probably are participants in eBird. Monarch Watch. iNaturalist. A whole bunch of different kinds of platforms out there nowadays. But scientists are grabbing a hold of data, picking it up, and using it. So a very nice paper in Biogeosciences a couple years ago. Probing the past 30-year phenology trend of U.S. deciduous forests. Actually, it should be “phenological trend,” because “phenology” is the study of the timing of lifecycle events, and “phenological” trend, so – but I’ll let them get away with it. Because it was a beautiful paper in Biogeosciences. Nature communication – Nature – a hotshot journal. Leaf onset in the northern hemisphere triggered by daytime temperature. Not nighttime temperature, not average temperature, but the daytime temperature – those highs during the day. Wow, we had – didn’t know. That’s kind of esoteric application. But you couldn’t answer that question until – and so now you know what to monitor – daytime temperature, which you have. They couldn’t do that without having a National Phenology Network – a national data set, which is what we provided. And then the one on the bottom – Phenology predicts the native and invasive range limits of common ragweed. Anybody here allergic to, you know, ragweed? It’s a real problem. And we need to know, where is it, and when is it active? And actually, they took that and extended it less to what is – what is it – when is it going up your nose, but more like what – how does the phenology – the timing of leafing and flowering relate to where – to where it is and where you find it? So that was a really interesting application. They just used a few data points. We didn’t have a lot of data points. They did some work in the lab and in a greenhouse. Put it all together and came up with this really cool paper in this premier journal, Global Change Biology. And then, of course, I had talked about this a little bit earlier. But this is actually where we’ve taken our phenological data that we – kind of similar to what the Washington Post did. And we said, well, you know, we haven’t really looked to see whether or not we see trends in this phenological data. People have looked before. We thought, let’s just gin this out for this public talk that I’m going to give next week and see what it looks like. So it’s – the results are provisional. It hasn’t been published anywhere. But this top map shows, for each one of the pixels, or a 4-kilometer-by- 4-kilometer area, what has been the trend in the timing of the – of the onset of spring – that Spring Index since 1981 to 2018. To sort of, like, last – to last year when they have a full data set. Green is showing later in time – up to later in time. Purple or blue is showing earlier. So if you – if you would expect the spring to come in earlier and earlier every year, then you would expect to see the map being – looking pretty purple. And it does. There’s not much green up there. Particularly a lot of purple in the west. But then we also had one more test. We said, okay, if you’re just looking and using a simple linear regression, which is a way to kind of model data points that might go like this, we’ll say is that line significant or not? Is that a significant slope? Is there a significant change? And we re-colored the map on the bottom. Should have used maybe a different color. It’s purple. But it’s showing a different thing where the trend is significant. So you can see there’s a lot of areas that are purple on the top of the map that don’t really show up as significant. Because there’s a small trend, but it’s not really – but it doesn’t stand up to the rigor of scientific and statistical analysis. But we do see a lot of warming trends occurring in the southwest U.S. And from – most recently from Tucson. Here you all are – this – the other side of the …

- Sierras.

- … the mountains – Sierras, thank you. And – where am I? Where am I? Who am I? And sixth talk. And so there’s some interesting trends. We have to dig into this. This was just produced last week, sort of on the fly. But here’s one we spent quite a bit more time working on. This was published in a paper – published in a journal called Ecosphere by Bill Monahan, who is now with Forest Service. He was with Park Service. And he was really interested in, how is spring changing across the whole national park system? Especially those parks that are valued for their natural resources. And so we looked at 267 national parks from 1900 to 2015. And we said, are there trends in the – sort of using the Spring Index as kind of this probe, or a yardstick, across the national park system. And so what you see here, all the blobs represent a national park or park system with a 30-kilometer buffer around it. And we asked the question, over time, are we getting earlier leafing or blooming of plants? Any blob that’s colored is showing in extreme earlier – shift towards earlier leafing and flowering, according to that standard yardstick. And so what’s that’s – that’s really important because, if you’re a park manager, and all the sudden, you have your flower festival two weeks late, and you have the firefly festival two weeks late, and you have invasive species coming two weeks earlier than the weed disposal crew gets there, you have to – you have problems. So you need to understand, how are things changing over time from one park to the next? So really, truly, half of the – three-quarters of the parks in the national park system are showing an earlier – shift towards earlier spring, and about half are showing this kind of very strong impact where it’s quite a bit earlier relative to the last 100 years or more. So let’s talk about, then, informing decisions – how you might use some of that – how you might use some of that kind of information. I have just, again, a few examples here. Here, what we’ve done is taken that same index and applied it to the entire continent. Because we’re really interested, for a paper we were working on, what is the trend over time across the continent within the migratory bird flyways? And so I’ve drawn black lines there to show sort of the four major migratory bird flyways. And we’re interested, okay, does it change in the timing of the seasonality as a bird migrates along, kind of going back to that very first figure. Okay, are the birds arriving earlier? How is the – how is the environment changing? And this is actually – was not published as part of the paper, but you can see some interesting patterns there. Warm colors, earlier – over the last 100 years. Cooler colors, spring is coming later over the last 100 years. And you can see that, up there in Canada – B.C. and Alberta – that there’s large portions of the – of that – of that part of the continent that have actually cooled according to this metric. And, again, these data have not been published. It’d be really interesting to go back and look at that. You do see a little bit of the warming hole there in the southeastern U.S. showing up. But, again, this is across, you know, 115 years, if we had the data. In Alaska, it was a little bit later. So some interesting patterns. And so what we did is, okay, well, let’s look at how migratory birds might be responding to a change in habitat. And this is a fairly complicated figure. I won’t spend a lot of time on it. But this was published in a paper in a journal called PLOS – PLOS One – that talks about the changes in the habitat when you have birds migrating from the south to the north, and is their habitat when they – their breeding habitat, for example, is that changing according to a standard metric? And so, if you look at, say, the warbler – lower right – or a crane – upper right, they move to migrate to new habitats, and the habitats – like, for the warbler, the bright colors, the warm colors there, in its breeding habitat, which is in the middle U.S. there, showing a strong shift towards earlier spring. And so the question is, how might that be impacting the birds? And we have no, actually, data on birds. It was just that seasonality of the habitat that they were landing in for their breeding range. So the story gets that, the fact that, okay, that is changing – that is – according to the standard metric, that is changing, and we see the – we see that going into the vegetation. But is it – how is it affecting vegetation? How is it affecting caterpillars? That might be the – how is it affecting birds? So there’s a lot of work more to be done, but an indication that we can start to see these very, very large-scale patterns, not only from your backyard but across the entire breeding ranges of species. I talked about this one just a little bit earlier, and just one little cool animation just to sort of show you what this might look like. This is one of our phenological forecasts. It’s actually for emerald ash borer. And this is – I put the whole year together. This is 2018. And so what it’s showing is the zones when the different lifecycle stages of the emerald ash borer might be met. And so we produce these on a daily basis and put it together, so what – you get an animation like this. And the different colors show the different lifecycle stages as to when it would be appropriate to go out and start doing – looking for emerald ash borer. If it’s – if you think it’s there, you need to know when to go out and look for it. Okay? Or when you might go out and actually treat the adults as they emerge. That’s a really easy time to identify that they’re there and to treat them if you need to treat, say, your street trees in Fort Collins. So this is an example of a phenological forecast and are working towards – I want to have a tick forecast. I’m kind of obsessed about tick forecasts because I think it would be just super useful to have it. You drive into a national park. Your phone lights up, and you get a warning – hey, there’s – it’s tick season. It’s important that you, you know, take precautions, especially in the northeast or other places where we have tick-borne diseases. It’s becoming a real problem. So the last tenet of our – of our strategic plan is communicating and connecting. Basically, I talked a lot about communication. It’s working really well. But, again, it’s very important to think of, okay, how do we then – so I won’t talk a whole lot more about it, but other than to just point out that one of the ways that we can take advantage of – well, actually, it’s critical that we have phenological information rolling in from folks who are helping us answer questions for our partners and to answer scientific questions. So we run a number of campaigns, like monarch – let’s see – Nectar Connectors on the upper left there. Mayfly Watch – I’ll show an example of Mayfly Watch. We talked a little about the lilac campaign. Our tree leaf campaign – Flowers for Bats and Shady Invaders. They’re testing whether or not invasive species are controlled by the amount of light and how that affects their reproduction. So we’ve run a number of different kinds of campaigns, and this is a way for you to get involved. So if you go to Nature’s Notebook, you kind of join a campaign – help collect data on some of these different campaign species. We have lots of species, but these are a few examples. So I talked a little bit about emerald ash borer. I mentioned hemlock woolly adelgid. And we have a campaign we’re running right now. It’s called Pest Patrol. So we’re producing these models of the phenological forecasts for those inspect pests, like I showed you for emerald ash borer. In order to validate that, we ask people, can you go out and look for emerald ash borer activity, record it through Nature’s Notebook, in this big database, using the same protocols. Everyone’s using the same protocol. We have the same definitions. It’s consistent. We have a place to store all those data at a national scale. This is – the lower right is showing – we had a little pilot project on hemlock woolly adelgid, which grows in ... ... a conifer – I can’t – hemlock. Hemlock. Eastern hemlock. And is decimating eastern hemlock forests. And the scientists at Cornell didn’t know, actually, what the phenological model was. And so they guessed. And so they gave us a couple guesses. And so we built that in to the – into a model framework and said, okay, we think, based on the best available information that we have, this is when things are going to be happening. And actually asked people to go out, and can you validate that? And it turns out actually, no, in some cases, they were finding hemlock woolly adelgids at much different accumulated temperatures than they would have predicated. So this is a case where the model was doing very well at all, but it was giving enormous amount of information back to the scientists, like, oh, okay, guess what? We just couldn’t get out there to be looking for hemlock woolly adelgids. And so we’ll work then to iteratively improve the model. This is another example. Mayfly Watch. You know, mayflies emerge from clean water at certain temperatures. So they’re driven by temperature. And, on the left-hand side – this is along the Mississippi River – this is a mayfly emergence. Okay, if you’ve ever lived in this particular region, you know. You’re, like, oh my god. It’s mayfly season. Temperatures must be warm enough in water, and we can actually measure that. And we can actually – we’re working with Fish and Wildlife Service along all of their string of refuges along the – along the Mississippi River locks and dams to try to understand, what are the drivers of mayflies. Because, when they come out, they can be a real problem. They can be a health and safety problem on bridges where you end up having to spend several thousand dollars to come out and bring the – bring the crews out – the city crews out to clean up the mayflies. Because you forgot, and you left the lights on on the bridge. And so when the mayflies emerge, and they left the lights on. Oh, no. They’ve got a big problem. And the mayflies collect there. So, you know what? If we had a mayfly model from Mayfly Watch and validated data pretty good, and we could deliver in the hands of – wait a second, my phone is ringing – oh, mayfly problem. Turn off the lights on the bridges. Okay, it’s dawn. Turn them back on. Okay, and avoid this particular problem. So we’re working actually – it’s kind of – it’s kind of a fun story, but we’re working actually with the Fish and Wildlife Service and with municipalities up and down the Mississippi River to try to understand, what are these drivers, and can we actually forecast mayflies so they can think about alternate ways of lighting bridges or dealing with the mayfly problem. Which is a boon to the environment, but not – depends on where it’s at. Okay, so what next? Good stories. Where do you go from here? Well, it’s working really well. We’ve had a fantastic 10 years. And we’ve really sort of – with a replicable workflow, we’ve really created a model, I think, for making sure that people can answer their own questions, help us collect data, share their data back with us, and that we can get high-quality data to be able to convert it into science and inform decisions. That’s the idea. And so we’re going to try to take that and extend it to other systems. So I talked a lot about ticks. Got to have a tick forecast. Agricultural production – when do you go and – when do you mow? How you deal with your old fields. Talked a little about that at the dairy near Albuquerque. When do you set hunting seasons? How do you actually know when you have organisms that are available out there for harvest? Some of us like to hunt. How do you manage birds around airports? We had a sad situation there, but if we knew when the birds were a problem, then we could actually start to do things like dissuading them from hanging out in flyways. Monarch butterflies. They need nectar and – on their migrations – and milkweed, especially. Are the milkweed – is the timing of the milkweed changing? And how is that affecting monarchs? West Nile virus and other diseases carried by mosquitoes. Pollen production, I mentioned. Understanding wildfire. Early spring, greater chances of wildfire. How can we incorporate phenological information from the biological environment into wildfire models? Invasive species and dust. So some places have a lot of dust problems, but it could be because of differences in invasive species and drawing and whatnot. So if we could extrapolate from the biological world out to the natural world, we can get a much better handle on our planet. And so then there’s – there’s new data that are coming out. This is a paper produced by the National Academy of Sciences. It was talking about how we can create next-generation Earth prediction systems. How can we get better, basically, weather forecasts and climatological forecasts? So we use the National Center for Environmental Prediction data, like Weather Service data, to do some more forecasting, but there are better models coming on board, like the Climate Forecast System from NOAA, version 3. The North American Multi-Model Ensemble. They can give us multi-month forecasts. And so we can start to make more informed decisions, not just days or even weeks ahead of time, but maybe even months. So this is a real opportunity to take the phenological data and the models that we’re developing and the systems we’ve developed and plug in better forecasts to understand the pulse of our planet. So with that, I’d like to say thank you very much. But it’s just me sitting up here saying thank you very much. And really, I’ve had a fantastic team. Actually, Theresa Crimmins has been with us since day one, in the lower right-hand side. And our staff has a very, very low turnover. We love working together. This is it. This is the – these are – this is the staff of the National Phenology Network at our National Coordinating Office. We work together. We really enjoy it. And there’s absolutely no way I could have given this presentation without having them kind of backing me up. It’s a great foundation and a great team. So, on behalf of all of us, thank you very much.


- Thank you, Jake. I want to – we have time for questions, of course. And, of course, we’re going to ask you to please walk to the microphone to ask them so that our online viewers can hear the question.

- Thank you very much. I came with two other people who – we had a group of people who created our own monitoring site.

- Mm-hmm.

- In Palo Alto. And we monitored that site for a little over five years. And it’s a private site. And just wanted to come and offer, if anyone was interested in seeing how we set the site up, and we can take you over there and show you what it looks like.

- Is that through Nature’s Notebook? So you have a Nature’s Notebook site?

- Yes, we – yeah, we put all the data up on Nature’s Notebook for a little over five years.

- Oh. Great.

- And we had seven species of trees.

- Okay. Okay.

- And two samples of each.

- Excellent. Wow.

- Yeah.

- Thank you very much for your contribution. I feel a little bit bad. I should have done a little bit of research for, okay, who here might be participating in Nature’s Notebook, and what are the sites nearby? California has, I think, 3,000 registered sites, though. So got a lot of action here in California. So it’s really great to hear. I do know that, in the California Natural Reserve system, a number of the natural reserves are actually tracking blue oak and other taxa. So you’re not part of a natural reserve? One of the UC reserves? Okay. And that’s been fantastic. So that’s another kind of a network. It’s kind of a node of our network that really makes it a network as opposed to, you know, a small team of us from Tucson and Fort Collins. [chuckles] Thank you.

- So, first, thanks for coming out. I enjoyed your talk.

- Thank you.

- So, I’m a very heavy eBird user.

- Ah.

- And so some stuff with iNaturalist.

- Uh-huh.

- And I didn’t – you mentioned those, I heard, in your talk. But how much sharing of data goes on? Because, you know, I don’t …

- Yes.

- I don’t want to have to enter observations into six different systems.

- Yes.

- Right? And so there’s, like, the citizen scientist fatigue, you know …

- Yes.

- … in terms of recording data.

- [laughs] Yes.

- So just – can you talk to that a bit?

- Yeah. That’s a – that’s a fantastic question. So I sort of alluded to the fact that we had this paper published where we looked at habitats. We knew nothing about the birds that might be actually using them. And that is a problem. The data integration, for a number of reasons, is more cumbersome. So think back to that conceptual figure that I showed, where you needed information on caterpillars, on birds, and on the habitat that the caterpillars – where you find the caterpillars there on the emerging leaves as they come out on the trees – like, oak trees. And so that’s three different data streams that might be collected by three different groups of either volunteers or professionals. There might be data in herbariums. There might be all kinds of data in eBird or iNat, et cetera. Or even more specific, there’s – I think there’s, like – there’s maybe hundreds of programs here in the – in the Bay Area that are – that are collecting data that way. And so – and fatigue is a – is a real problem. So data integration and fatigue. And so it would be really great to be able to start to pull these things together. So we had a postdoc in our lab, Jherime Kellerman, who actually said, look, I’ll take your – the phenological data, and the pattern of just leafing – just for deciduous trees in the eastern U.S. and compare them to the frequencies of observations from eBird. So he had to go to the eBird data set, pull it down. Go to our data set and pull it down. Mash it all together. And come up to see whether or not – it’s actually looking at some of those curves. Are there – are things changing through time? We didn’t quite get to the place where you’re going to publish a paper. We got some really cool patterns showing up there, but it was a lot of work, and we spent a lot of time as scientists fiddling with data and trying to – so what we’ve done is, within the Phenology Network, for about 1,200 species, we’ve worked to create standardized protocols so that we can have a new data resource and enable that kind of integration. But it is – it’s yet another project, but it’s a platform that allows observations of both plants and animals. And we talked to people at eBird early on. And, like, well, are you kind of creating a competition? It’s, like, no, no. Actually, we’re looking at, beyond just whether it’s there or not, we’re actually – what is it doing? And there’s opportunities to add how many there are. So I know eBird does that – how many there are. But then the question is, are they – are they migrating? Are they feeding? Are they reproducing? Are they dead? So we built in phenological information and created a richer ecological data set. But there was a lot of other networks out there doing phenology as well, especially in Europe. And so we’ve built special tools to help us compare our data to theirs to basically concatenate all different terms and fields together so that you can start to mesh those. And there’s now a portal called – plant phenology that allows you to integrate data from the National Ecological Observatory Network, the National Phenology Network, and the European phenological data using what we call ontologies to help us – help data sets integrate. And now there’s a user interface where you could actually have it all done automatically and publications about that – the pipeline as to how the data moves. So data integration is really – it’s kind of like a little bit, the new frontier. There are places where we – because we haven’t integrated, we don’t have the answers that we would love to have. Good science questions for graduate students and professors. Or anybody.

- Hi, Jake.

- Dr. Weiss.

- Yeah. Just – I was wondering, what’s your annual budget? And how much has been invested, specifically in your program, over the 10 years?

- Okay. Great. Thank you, Stu. I appreciate it. Well, this is us. I’m a USGS scientist. Everybody else on the team is [chuckles] from the University of Arizona. So I manage a cooperative agreement with the University of Arizona. The total budget is about $1.2 million a year. That’s my salary, the salaries that you see here, some overhead costs, and whatnot. So that’s today. It’s been stable for about the last five years or so. Resources come in a little bit from NASA. A couple of grants to the University of Arizona. Some money from Fish and Wildlife Service. So we don’t – can’t – they give us a little bit of money so we can accelerate their cool websites that we do for them. So the total is about – so far, the total investment is about $12 million. When I first started, we had a quarter of a million dollars, and I said, well, that sounds great. And they’re, like, well, that’s your salary too. Mm, okay. [laughter] That’s not so much. So, really, we’ve sort of built it through time. We’ve been successful. I’m not – you know, that was enough. It was the seed money. But that’s – so about $12 million total. So it sounds like a lot. A dollar per record. [laughs] 15 million records. But we’re catching up. We’re catching up. Thanks, Stu.

- Any last questions?

- Usually in a room – in any situation, it takes about seven seconds to truly plumb the questions out there. So we wait seven seconds in our office if somebody has a question. [chuckles] Please, go ahead.

- Just a question. Do you know about Calflora? Because they have been acquiring phenology information on California flora for a bunch of years. So I’m sure there are many, many millions of records.

- Yeah. That’s a very good question. I’ve heard of Calflora, and I’m not familiar – well enough familiar with the nature of the data or the standardization. We have had a lot of involvement with a variety of different California organizations. So my thought, to be honest, is there might be some significant differences there, but it would be very good. So there are a lot of other organizations that do collect phenological data. In fact, if you work with iNat, you – in some species, you actually have a chance to go over and add the phenological data in – link it over to the National Phenology Network, and the data are actually shared back over. On eBird, we haven’t bridged the gulf yet. They’re producing the best-quality bird data set. At a global scale, we’re producing the – we’re trying to understand what the resources are. So Calflora might be a really good one for us to work with. There’s actually a lot of different kinds of data sets out there.

- Yeah.

- Botanical gardens, for example. Many of them have great phenology data sets. Some of the California natural reserves. Actually, many of the Fish and Wildlife refuges – the national wildlife refuge systems have phenological data. Everybody used a different protocol, a different way to count coots, you know, or whatever. And those data are all on cards and on paper, and so sometimes we call those shoebox data sets. My guess is there’s probably several of you who are sitting on shoebox data sets. Meaning, like, you know, that you have been keeping your own data, like, on a barn door. When did the – when the combines arrive, or when did the martins arrive, you know, et cetera. So there’s all kinds of different data sets that are out there. We found that – we started out thinking, okay, we’re just going to integrate all these different data sets together. It’s going to be great. And, like, oh my gosh. You know, we realized, even that, as we changed our protocols through time, that that could create confusion. Because we had to tweak them early on, and we didn’t know how to – you know, what’s the best way to track a fish, you know, in a stream? So we had to change things over time. We realized when we worked with New York Botanical Garden, it took several years just to get the data re-homogenized back into the current protocol. So now we build better versioning, and we work to not change our protocols. But just – data integration is a – is a huge job. And there’s so many different data sets out there, like the Thoreau data set, and the Leopold data set, and the various and sundry data sets. Rather than try to gather those together, we say, here’s what we have, and if you have a specific question, we’ll help you find those kinds of data sets that are out there. So you actually have a data set registry tool. So you can say, I’m from this place. And we try to create a community of practice so they can figure out, okay, I’m really interested in coots. Okay, what phenology data sets are coot – of coots are out there and, you know, help that. And then that’s a grad student who’s going to integrate all those data and maybe share them back or publish them. So, good question. Thank you.

- Mine’s similar. You have a real challenge with data and having quality data.

- Mm-hmm.

- And so I’m kind of curious about war stories about – you know, you got a changing climate. You have invasive species. And then things that show up that you never expected. So I imagine you have a couple of stories with respect to surprises that you had to deal with.

- Yes. Thank you. Yes, indeed. One is a little different than what is actually happening out there. It was more like – it was a – it was a war story created by a scientist. [laughs] I met the first person – the person who wrote the first scientific paper off of our data set. I was so excited. I was, like, oh my gosh, how did you clean the data? And he looked at me, and he said, clean the data? [laughter] Oooh, no. And so [laughs] – and then, since then, we’ve had – you know, it’s great. It’s, like, oh, my gosh. Really fantastic patterns showing actually changes in phenology of red maple and other deciduous forest trees that we had a fair amount of data on. And projected that out using climate change scenarios as to what phenology might look like. And it was pretty stunning, especially, like, for red maple. It’s fairly plastic. It’s adaptable to the changing environment. And so he was projecting out to 2100 and showing this very, very large change in red maple. Other species are more recalcitrant. They don’t change as much. And in fact, there’s a paper in Britain where they actually did use great data like that and found that the timing of organisms, since they respond differently, shifted. And so you ended up kind of scrambling, creating kind of a novel phenological system. It was a really great – a really great paper where they had good data in the British Isles, where they’ve been doing this for a long time. And so – I lost – kind of lost my train of thought, there, but that was – that was one. And integrating – realizing that the great European data sets were constructed using a completely different phenological monitoring system and coding systems called the BBCH code. And here we had the MPN and the NEON data sets that were all integrated together. And to actually just – to realize, wow, we could look at taxa all the way around the northern hemisphere, including, you know, Europe and the United States, but there’s this very, very large difference in the coding. But since they were consistent, we’ve actually built that ontology I talked about a few minutes ago that helps us blend the two together. So those are a few examples. But people will always call us up and say, this is really weird. What’s going on? And we’re, like, well, we don’t really know [laughs] what’s going on. But it’s great because people start to become observers in the system, and they see – they see sort of interesting, weird things. Like big freezes that occur around Easter through Tennessee and Alabama that basically took out whole nut crop that year. Or big freezes that happened in – actually, in 2017, really early spring, sort of set the whole system up. And a – and a huge freeze came through and affected all the grape production. And so that affects how large grape juice drink makers might adjust their formula a little bit because they just, you know, lost a bunch of grape products. Or thinking about paying several hundred bucks per hive to get your almond orchards pollinated here. I had no idea. There’s a book called The Beekeeper’s Lament. Fantastic. Shipping beehives across the country on flatbed trucks to get them here to the almond orchards because they have such a high pollination requirement. And that’s a narrow window, right? You have about two weeks where those flowers are coming on. So you’ve got to get the hives there at the right time. And Blue Diamond Industries actually used to have – I’ve kind of telling lots of stories here now. You’ve kind of got me rolling. Blue Diamond Industries actually used to have – Blue Diamond – you know, the – makes the almonds, they used to actually have a phenological monitoring system. There’s remnants of it online. You can see that the growers themselves – Blue Diamond growers, actually, were tracking the phenology of almonds flowering and the puffy flowers as they come out and are available. So the industry itself was getting organized to have that kind of information. E&J – Ernest and Julio Gallo – used Landsat, which is a USGS product, to understand when to water particular parts of the fields out there for wine grape production here in California. So those are kind of some really cool surprises, where you have these applications kind of transmitting the scientific information and how you collect the information and process it into making decisions. Pretty neat. Thanks. Thanks for the question.

- Yeah. So, with data coming from so many different sources, some of them might be higher quality than others. Some of them might have a higher level of accuracy than others. And, when you put them together, there’s the potential to create – instead of something useful, just a big mess.

- Yes.

- So how do you work on that problem?

- That’s a great question. We have a whole page on our website devoted to data quality – descriptions of the processes that we use. And we developed a strategy to deal with data quality. You have two parts. We basically break it apart. We have quality assurance and quality control. Quality assurance is working to create systems to get – to prevent people from making mistakes or whatever before the data come in. Quality control is where, okay, now we have this piece of data. Do we know whether it’s any good or not? There actually might be a third part. I just sort of realized that we create tools after the fact that allow people to kind of post hoc go in and try to work with the data and slice it and determine which piece of data is not so good. Maybe that would be part of quality control. But we do that sort of not – we ourselves don’t do that actively. Rather, we provide information to others to be able to do that. So quality assurance might be things like pre-filled dates. So, for someone – for example, if someone said, oh, I saw this flower, and then I saw this leaf, as they’re entering data. What we know for that particular species, that actually the leaf comes on before the flower, then they’ll get a little flag. They’ll try to enter a date in the future. That won’t allow them to do that. Other kind of simple things that we can do to prevent mistakes from occurring. Those that might be mistakes. Or we could – we actually do a lot of training before people even go to the field, say, with our certified leader program, to try to improve data quality. So, before anybody even goes to the field to make an incorrect interpretation, perhaps, or enter the data incorrectly. So those are all quality assurance. Quality control then allows us to do some out-of-range testing, for example – pretty easy to do. Except for things like blue spruce, which is everywhere. Red maple is also everywhere. Except there is a red maple in Phoenix. I know it’s not Acer rubrum. It’s somebody’s red maple. It has red leaves on it. So it’s out of – so we flag it. Because we don’t know whether it’s actually really – someone could have planted a red maple there – truly Acer rubrum L – Linnaeus. Because, in Nunavut, which is what the old Northwest Territories, people started saying, what are these weird red-breasted birds doing out here all the sudden? They’re at the very northern end of Canada. It turns out it was robins. [laughter] Robins had actually started – they were – they were vagrants – or, what’s the word – on occasion, seen on occasion. But, because of the conditions changing, you actually had a shift in robins, where they’re becoming part of the – part of the natural history of the area, if you will. Relatively recently. So we don’t necessarily know whether something – if it’s out of place or whether or not it’s actually a good piece of data. So we flag it with some quality control flagging. There’s one more piece that I’ll tell you. So, if you have a piece of data, and it’s – the question is, did you see a flower on your lilac? We ask people to say yes or no, or I couldn’t tell, or I didn’t look, basically. And so, if you have – because that absence data. We ask them, tell us – did you look or not? They might say, no, I didn’t look today. Okay, good. So that means it – absent – you don’t – you can’t make – you don’t have to make inference as to whether it was or not. So if you have a piece of data that goes like this – someone’s watching it through time. Yes. That’s a lot different than somebody who’s going out weekly. No. No. No. No. No. Yes. Yes. Yes. Yes. No. No. No. That’s a – oooh, that’s a lovely set of data right there on, say, leafing or flowering of red maple. And so we actually – that’s how we collect the data. We distribute it that way. So that takes a lot of processing. So when I make a pretty map, we’ve had to deal with all of that. But it really allows someone to be able to kind of go out and parse all that out. It might go yes, yes, yes, yes, yes, no, yes, yes, yes, yes. Okay, what’s going on there? So the scientist who’s doing the data analysis can make inference based on a string of data. So those are a few examples of how we work to manage the data sort of up front once we have it, or how we process it after the fact. Thank you.

- One quick question. Looking 10 years in the future, say, the way the Apollo program did their 10-year within-the-decade moon shot idea, if you look 10 years in the future from today, what particular challenges or successes or things like landing people on the moon – that kind of out-there goal do you think is possible and that you’d like to see?

- Wow. Thank you. When I first started this project, I had to tell myself it was a 30-year project. Okay, this is a 30-year project. And when I said to a group of people at a workshop, when we were getting there, to define – the kind of early days – I said, you know, I’m the first executive director of the National Phenology Network. And several people came up, and they’re, like, are you leaving? [laughter] No, no, no. I’m just the first one because this is a 30-year project, and I don’t want to be doing this 30 years from now. I’ll probably be – I’ll probably reach my phenological maximum. [laughter] I guess. I hope so. I hope I’m the first. And so, we’re 10 years in. So we’re a third of the way in, you could say. And, if something happened, we have tools now that will basically – we flip a switch, and the data basically gets summarized and then put into a trusted data repository. The data are actually there. We just expose it with a metadata record. And so it’ll be there for posterity with all of this description as to how the data were collected. So if that were to happen. But we have 20 more years. Or maybe 30. Maybe it’s always 30 years. And I think it’d be really fantastic to have longer-lead forecasts. I’d like to have a six-month forecast of whether it’s going to be a bad fire year in California. Okay? So we deliver that information. Actually, we develop in collaboration with the Forest Service and the state of California. And they know what to do about where to put tankers and how many new – you know, how many new airplanes, and where to stage. And whether they get the hotshot crews ready. Okay, we’re going to have to bring the hotshot crews over from Nevada this year, et cetera. I would like to have a multi-month forecast of tick seasonality with my phone vibrating in my pocket telling me, you better watch out. Stop at the ranger station. Because the rangers, they know that too. Get some – get some tick repellant, et cetera. Because Lyme is far worse now and spread in the last 10 years, you know, out to some places that we didn’t expect it. Or other kinds of tick-borne diseases. Exotic diseases are really important. Right now, there’s all kinds of pollen data sets out there across the nation. Pollen. And human health data sets. Never the twain shall meet, as it turns out. But if we could get organized and create, like, a national pollen network – another NPN – actually, we are because NPN is the New Parents Network. [laughter] So they got there first. And we’re, like, okay, we’ll be USA NPN. We’re just one of many national phenology networks anyway. And so, if we could create a national pollen network to create standards to share pollen data from the many locations where it’s been collected in different – on rooftops of hospitals and different experimental stations, et cetera, and start to create forecasts of pollen. Is this going to be a bad pollen year? When is pollen going to start? Is there going to be a pollen bomb? We deliver the data to industry. The industry starts the great warehouse on wheels and gets your Claritin into the [laughter] – into the stores on time. And that’s going to take just that cultural – that’s going to take cultural change. Not just scientific wherewithal or data management wherewithal. It’s going to take cultural change so people will start to share their pollen data. So right now, I talked about that leaf – that tree – deciduous tree campaign – the leaf campaign that we have. Our hope is actually, if we know when the leaf occurs, and when the flowers are out, that we would be able to do pollen – people probably don’t really want to monitor pollen. That’s hard to do. It’s hard to see, et cetera. Or, like, the junipers. But if we have some phenological index, then we can peg that to when that pollen is going to be produced, and we can plug that in. So we can actually produce good pollen forecasts. Because pollen is going to be a huge problem as well. So those are a few examples of where, in 10 years, I actually think we might be there. I think some really great partnering with industry would be fantastic, where we have – we are delivering biological information at a national scale at a daily time step with huge amounts of resolution and well-defined workflows and data quality, et cetera. And then we have, say, a big meteorological organization that comes in and says, okay, we’d like to integrate our data with yours. And we have now, you know, 35-day forecasts. Okay, that’s good. And the third – the third-party client might be a huge conglomerate that’s trying to minimize how much – how much, say, herbicide needs to get applied, or how they manage – how they manage diseases. Maybe it’s a city. Or a whole region. So I do have to say that, if I did mention any trade names here – any brand names, that does not – that does not infer that I – how do I – what do I write in my [laughter] – does not infer sponsorship by USGS. So, in other words, there’s some really great opportunities out there to work collaboratively with industry and with members of the public and with scientists to bring together – so understand this kind of – how things are changing at local to national to global scales.

- Well, I want to thank you all for coming to this great talk by Jake. And I want to encourage you to return on August 29th to hear about Earth’s climate 3 million years ago. And let’s all give Jake one last welcome.


[inaudible background conversations]

- Hi, hi, hi.