{"id":16247,"date":"2015-01-15T15:41:47","date_gmt":"2015-01-15T20:41:47","guid":{"rendered":"https:\/\/www.allaboutbirds.org\/news\/?p=16247"},"modified":"2019-10-10T14:53:30","modified_gmt":"2019-10-10T18:53:30","slug":"a-hundred-million-points-of-light","status":"publish","type":"post","link":"https:\/\/www.allaboutbirds.org\/news\/a-hundred-million-points-of-light\/","title":{"rendered":"EBird and A Hundred Million Points of Light"},"content":{"rendered":"<p><!--[gallery size=\"full\" ids=\"24889\"]\n\n--><\/p>\n<div class=\"wp-block-group sidebar-alignright sidebar-space order-bottom\"><div class=\"article-list list-style alignright\"><h2 class=\"article-list-header\">More From Living Bird<\/h2><ul><li class=\"article-item\"><div class=\"article-item-container\"><div class=\"article-item-media  content-living-bird-toc\"><figure class=\"article-item-media-ratio\"><img decoding=\"async\" src=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/05\/LB-TOC-winter15.jpg\" srcset=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/05\/LB-TOC-winter15-720x329.jpg 720w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/05\/LB-TOC-winter15-768x351.jpg 768w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/05\/LB-TOC-winter15-480x219.jpg 480w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/05\/LB-TOC-winter15.jpg 918w\" sizes=\"\" alt=\"Florida Scrub-Jay by Robert Stalnaker\" loading=\"lazy\" \/><\/figure><\/div><div class=\"article-item-body\"><span class=\"article-item-header\">Living Bird Winter 2015\u2014Table of Contents<\/span><\/div><\/div><\/li><li class=\"article-item\"><div class=\"article-item-container\"><div class=\"article-item-media  content-living-bird-toc\"><figure class=\"article-item-media-ratio\"><img decoding=\"async\" src=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2023\/07\/living-bird-latest.png\" srcset=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2023\/07\/living-bird-latest.png 720w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2023\/07\/living-bird-latest-240x180.png 240w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2023\/07\/living-bird-latest-480x360.png 480w\" sizes=\"\" alt=\"Living Bird-latest issue\" loading=\"lazy\" \/><\/figure><\/div><div class=\"article-item-body\"><span class=\"article-item-header\">Living Bird Magazine\u2014Latest Issue<\/span><\/div><\/div><\/li><li class=\"article-item\"><div class=\"article-item-container\"><div class=\"article-item-media  content-article\"><figure class=\"article-item-media-ratio\"><img decoding=\"async\" src=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2023\/07\/living-bird-acrhive.png\" srcset=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2023\/07\/living-bird-acrhive.png 720w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2023\/07\/living-bird-acrhive-240x180.png 240w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2023\/07\/living-bird-acrhive-480x360.png 480w\" sizes=\"\" alt=\"Living Bird archives\" loading=\"lazy\" \/><\/figure><\/div><div class=\"article-item-body\"><span class=\"article-item-header\">Living Bird Magazine Archives<\/span><\/div><\/div><\/li><\/ul><\/div><\/div>\n<div class=\"wp-block-group sidebar-alignright sidebar-space order-bottom\"><!--HubSpot Call-to-Action Code -->\r\n<span class=\"hs-cta-wrapper\" id=\"hs-cta-wrapper-096b8ce3-0e2d-46c5-bbf7-12de3323c8da\">\r\n    <span class=\"hs-cta-node hs-cta-096b8ce3-0e2d-46c5-bbf7-12de3323c8da\" id=\"hs-cta-096b8ce3-0e2d-46c5-bbf7-12de3323c8da\">\r\n        <!--[if lte IE 8]><div id=\"hs-cta-ie-element\"><\/div><![endif]-->\r\n        <a href=\"http:\/\/cta-redirect.hubspot.com\/cta\/redirect\/95627\/096b8ce3-0e2d-46c5-bbf7-12de3323c8da\" ><img decoding=\"async\" class=\"hs-cta-img\" id=\"hs-cta-img-096b8ce3-0e2d-46c5-bbf7-12de3323c8da\" style=\"border-width:0px;\" src=\"https:\/\/no-cache.hubspot.com\/cta\/default\/95627\/096b8ce3-0e2d-46c5-bbf7-12de3323c8da.png\"  alt=\"subscribe to Living Bird magazine\" loading=\"lazy\" \/><\/a>\r\n    <\/span>\r\n    <script charset=\"utf-8\" src=\"https:\/\/js.hscta.net\/cta\/current.js\"><\/script>\r\n    <script type=\"text\/javascript\">\r\n        hbspt.cta.load(95627, '096b8ce3-0e2d-46c5-bbf7-12de3323c8da', {});\r\n    <\/script>\r\n<\/span>\r\n<!-- end HubSpot Call-to-Action Code -->\r\n\r\n<\/div>\n<p>Oddly enough, when the new age of range maps began to dawn, it happened on an empty stretch of country road in eastern Montana. It was early September 2009. Four members of the Cornell Lab of Ornithology\u2019s eBird team\u2014Steve Kelling, Chris Wood, Marshall Iliff, and Brian Sullivan\u2014were out birding. A Barn Swallow dipped over a small stock pond. A pair of Eastern Kingbirds tended a couple of youngsters. In a grove of cottonwoods were a couple of Wilson\u2019s Warblers. All pretty standard stuff. Then an Orchard Oriole appeared that made the group almost giddy.<\/p>\n<p>\u201cWe were all jumping up and down, like \u2018Oh my god I got Orchard Oriole in September, I can\u2019t believe it!\u2019\u201d recalled Marshall Iliff, who\u00a0co-leads eBird along with Wood and Sullivan. The bird flew off and the eBirders gave chase, but to no avail. Iliff snagged his pant leg on a barbed-wire fence, and nobody got photo documentation.<\/p>\n<p>As it turned out, that was okay. A little later in the day they saw another, and then another. In eastern Montana, maybe September Orchard Orioles weren\u2019t so rare after all.<\/p>\n<p>Why all the commotion about a bird that, as any field guide can tell you, breeds all over eastern and central North America? The answer is that field guides don\u2019t tell you everything you need to know\u2014at least not yet.<\/p>\n<p>In the Northeast, according to Wood, an Orchard Oriole sticking around until September is about as rare as a European vagrant such as a Curlew Sandpiper or a Ruff. But in the Great Plains, Orchard Orioles seem to migrate on a different schedule. Suddenly, the realization that they<em> hadn\u2019t<\/em> found a rarity that morning was more exciting than the thrill of thinking they had.<\/p>\n<div class=\"wp-block-image\"><figure class=\"alignright size-large\"><img decoding=\"async\" src=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/OOriole-Petersen.jpg\" srcset=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/OOriole-Petersen-720x497.jpg 720w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/OOriole-Petersen-480x331.jpg 480w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/OOriole-Petersen.jpg 742w\" sizes=\"\" alt=\"Orchard Oriole by LInda Petersen\" loading=\"lazy\" \/><figcaption>Ebird data are fueling occurrence maps that show far greater detail than has ever been possible before\u2014in some cases uncovering new migration patterns in birds such as the Orchard Oriole. <em>Photo by <a href=\"https:\/\/flic.kr\/p\/nBD2bQ\">Linda Petersen<\/a> via <a href=\"https:\/\/www.flickr.com\/groups\/birdshare\/\">Birdshare<\/a>.<\/em><\/figcaption>\n              <\/figure><\/div>\n<p>Possibly the only thing more exciting would be the prospect of a field guide that could illustrate these kinds of subtle patterns for anyone to discover. But how? You can cover up the range maps in most guides with a decent-sized postage stamp. The maps are drawn in about five colors, and they can\u2019t easily show when migration happens, let alone where the variations are.<\/p>\n<p>At least, that\u2019s what we expect from a field guide. But just in the last few years, research by the Cornell Lab\u2019s Information Science team has changed almost everything about what a range map can be. And they didn\u2019t even do it with birders in mind. It was an offshoot of a project to help conservation scientists understand at a fine scale where birds occur across an entire continent.<\/p>\n<p>The team was part of the first wave of ecologists to begin working in the realm of Big Data, using supercomputers to churn through mountains of eBird reports. Their work was funded by the National Science Foundation, augmented by data from NASA, and published in journals such as <em>Artificial Intelligence Magazine<\/em>. Their analyses ate up 2.5 million hours of processor time and returned more raw results than the Lab\u2019s computers could hold.<\/p>\n<div class=\"wp-block-image\"><figure class=\"alignright size-large\"><img decoding=\"async\" src=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/GuideBOok.jpg\" srcset=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/GuideBOok.jpg 489w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/GuideBOok-480x398.jpg 480w\" sizes=\"\" alt=\"Hand-drawn range maps quickly earned a spot in field guides after their debut in the Golden Guide\" loading=\"lazy\" \/><figcaption>Hand-drawn range maps quickly earned a spot in field guides after their debut in the Golden Guide.<\/figcaption>\n              <\/figure><\/div>\n<p>\u201cIt\u2019s really changed the way I think of range maps,\u201d Iliff says. \u201cThey used to be just a boundary around where a species lives. But we\u2019re getting close to the point where we\u2019re actually going to be able to ascribe a probability to every point around the world, on every day of the year, [for every species]. That\u2019s a totally different way of thinking about range maps for me.\u201d<\/p>\n<p>In a happy offshoot of all that advanced analysis, the eBird team has created something of direct, utilitarian benefit to birders at large. In the coming years, this new breed of range map is likely to change your experience of going bird watching (especially if you travel). It may even help you with bird identification without ever requiring you to look at a map.<\/p>\n<div class=\"wp-block-image\">\n              <figure class=\"size-large alignnone\"><img decoding=\"async\" src=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/eBirdMAP-1280x645.jpg\" srcset=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/eBirdMAP-720x363.jpg 720w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/eBirdMAP-768x387.jpg 768w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/eBirdMAP.jpg 1280w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/eBirdMAP-480x242.jpg 480w\" sizes=\"(max-width: 1280px) 100vw, 1280px\" alt=\"eBird Orchard Oriole occurrence maps\" loading=\"lazy\" \/><figcaption>The new maps represent a shift from using an expert\u2019s judgment to using data augmented by models. An Orchard Oriole map starts with where the species has and hasn\u2019t been seen (left). Models use environmental data sets to predict where else Orchard Orioles occur\u2014for each week of the year. Comparing midsummer (top right) to early fall (bottom right) clearly shows different timing between Northeast and Great Plains orioles.<\/figcaption>\n              <\/figure>\n            <\/div>\n<h3>From Ink to Pixels<\/h3>\n<p>The very first published range maps were made of words. Even in the early 20th century, when Arthur Bent began his classic series of species accounts, his range maps were sentences. His Orchard Oriole goes like this: \u201cManitoba and southern Ontario to northern South America.\u201d<\/p>\n<div class=\"wp-block-group legacy-sidebar sidebar-alignright has-lightgray-background-color has-background\">\n<h4>A Mesmerizing Migration Map<\/h4>\n<div class=\"wp-block-image\">\n              <figure class=\"size-large alignnone\"><img decoding=\"async\" src=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2016\/01\/la-sorte-map-118-spp-64-725.gif\" srcset=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2016\/01\/la-sorte-map-118-spp-64-725-720x720.gif 720w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2016\/01\/la-sorte-map-118-spp-64-725-480x480.gif 480w\" sizes=\"\" alt=\"animated map of bird species migrating in western hemisphere\" loading=\"lazy\" \/><figcaption>Each dot represents a single bird species; the location represents the average of the population for each day of the year (see paper for a more precise explanation of the &#8220;average location&#8221;). Here&#8217;s <a href=\"https:\/\/www.allaboutbirds.org\/news\/mesmerizing-migration-map-which-species-is-which\/\">a\u00a0key to\u00a0which species is which<\/a>.<\/figcaption>\n              <\/figure>\n            <\/div>\n<\/div>\n<p>Actual maps began to appear in North American guides only in 1949, with the first Golden Guide. Bird watchers quickly saw their usefulness for narrowing down identifications, and they became de rigueur in modern field guides.<\/p>\n<p>Range maps are still made of ink and paper today. They are typically the work of a single expert, who draws them by hand using a combination of personal experience, data from journals and breeding bird atlases, and word of mouth.<\/p>\n<p>This approach has its advantages\u2014the human brain is supremely good at collating information, making decisions, and extracting patterns. But a hand-drawn map also has weaknesses. Pull out a magnifying glass to look at your hometown, and you\u2019re still just looking at a blush of color on a piece of paper. It still says only that the bird either should or shouldn\u2019t occur in your general area. And at heart, it\u2019s still a judgment call. If you want to know why, you\u2019ll need to call up the expert and ask.<\/p>\n<p>Tomorrow\u2019s maps will be made of pixels and data. Already, <a href=\"http:\/\/ebird.org\/content\/ebird\/occurrence\/\">55 prototypes are viewable at eBird<\/a>; more are coming. Each point of color on the map is there to represent a specific probability of occurrence, and each represents just the tip of a pyramid of data lying underneath the map. If a conservation scientist wants to know why any part of the map looks the way it does, he or she can look into the data and figure it out.<\/p>\n<p>Because the new range maps exist as data, they even have the potential to do their work invisibly. The Cornell Lab\u2019s free Merlin Bird ID app uses a birder\u2019s location and the date to automatically tailor its ID suggestions to the most likely species\u2014doing just what bird-walk leaders have been suggesting for decades. (The app still allows you to pull up maps for each species, too.)<\/p>\n<p>Eventually, range maps will allow you to zoom in to see local features or zoom out to see an entire hemisphere. You\u2019ll be able to filter them, or drill down to see individual checklists, or look at photos uploaded by birders. They\u2019ll show the movements of birds\u2014migration and postbreeding dispersal\u2014not with static color but with movement. Range maps will finally come to life, like the portraits on the walls in a Harry Potter novel.<br \/>\n<div class=\"wp-block-image\"><figure class=\"alignright size-large\"><img decoding=\"async\" src=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/BlackpollMap.jpg\" srcset=\"https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/BlackpollMap-720x642.jpg 720w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/BlackpollMap-768x685.jpg 768w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/BlackpollMap-480x428.jpg 480w, https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/BlackpollMap.jpg 899w\" sizes=\"\" alt=\"Blackpoll Warbler occurrence and photo by tfells via Birdshare\" loading=\"lazy\" \/><figcaption>The next step is to model an entire hemisphere, even where data are scarce. Early results for Blackpoll Warbler show promise, picking up the birds\u2019 spring route through Florida and their fall leap from the Mid-Atlantic to Venezuela. Rare sightings in the West can trick the models, as can sparse birding coverage of Amazonia in winter\u2014but future models will address this by adding environmental data. <em>Map from Cornell Lab Information Science Program, photo by <a href=\"https:\/\/flic.kr\/p\/HoTFNR\">tfells<\/a> via \n<a href=\"https:\/\/www.flickr.com\/groups\/birdshare\/\">Birdshare<\/a>.<\/em><\/figcaption>\n              <\/figure><\/div>\n<h3>Ninety-Seven Percent of the World&#8217;s Species<\/h3>\n<p>\u201cThe only way you can conserve biodiversity is knowing where species occur,\u201d is how Steve Kelling explains why he\u2019s interested in making range maps. Kelling directs the Lab\u2019s Information Science program and has shepherded eBird from its beginnings in 1997 to the global database that now provides the raw material for the new range maps.<\/p>\n<p>When eBird launched, in 2002, it grew slowly at first, and then exponentially. It helped that the program had been designed to appeal to birders\u2019 inherent love of keeping lists. Birders can enter reports from anywhere and at any time of the year. Participating is free, instantly accessible on the web (and on smartphone apps), and the program is filled with tools that let birders explore and share what they and others have contributed.<\/p>\n<p>It took 10 years for eBird to get its first 100 million observations; it took 2 years to get the second 100 million. By 2014, eBirders had spent a collective 1,700 years watching birds, and had reported more than 10,000 of the world\u2019s roughly 10,300 species. It\u2019s now one of the world\u2019s largest collections of ecological data\u2014meaning bird watchers are among the most sophisticated sensors of planetary health that the world has ever produced.\u00a0Paul Lehman has a few qualms about all those data, and he\u2019s not alone. For the past two decades, Lehman has drawn the range maps for most of the North American field guides. (He likes to joke that he owes the Eurasian Collared-Dove and its prolonged range expansion for keeping him in business.)<\/p>\n<p>\u201cBirders get tripped up as often by not appreciating status and distribution as by the field marks that they learn in a field guide,\u201d Lehman says. That\u2019s an argument for studying your field guide, but it\u2019s also a concern about using birders\u2019 sightings to generate range maps, he says.<\/p>\n<p>\u201cIf you were to go to Plum Island [Massachusetts] this weekend and ask everyone to send you their eBird checklist,\u201d Lehman says, \u201chow many people would even be close to each other on [counts of ] Herring Gull, Semipalmated Sandpiper, or all the Yellow-rumped Warblers in the myrtle thickets? Somebody with good ears, how many creepers and kinglets will they record versus somebody with poor ears?\u201d<\/p>\n<p>Questions about data quality are among the most frequent that the eBird team hears. The prospect of individual data points being inaccurate just gives people the heebie-jeebies. Fortunately, the maps are much more than a raw compilation of eBird sightings.<\/p>\n<div class=\"wp-block-group legacy-sidebar sidebar-alignright has-lightgray-background-color has-background\">\n<h4>Solving the Area 51 Effect<\/h4>\n<p>STEM models are good, but they\u2019re not perfect. Perhaps surprisingly, the most common type of mistake is not missing where birds are, but incorrectly predicting them in places they aren\u2019t. Wesley Hochachka, an ecologist on the analysis team, calls this the \u201cArea 51 effect\u201d because it tends to happen only in sparsely populated regions such as central Nevada, site of the secret government base.<\/p>\n<p>The problem is a lack of data, especially negative data, the term for all the species on your checklist that you leave blank. Models need both types: positive data to know where a species occurs, and negative data to know where it doesn\u2019t.<\/p>\n<p>One Area 51 example is the Oak Titmouse. They\u2019re common on dry hillsides in California, where tons of local eBirders report them. When the models look at Nevada, they see similar hillside habitats but few eBird checklists. So they guess that the bird should be present. There\u2019s little evidence in eBird either way, so there\u2019s nothing to correct the model\u2019s guess. There is a simple fix (theoretically, anyway): get more people to go eBirding in these underbirded places.<\/p>\n<\/div>\n<p>Without models, eBird data would produce maps made up of thousands of little dots clustered around where people live and where they go birding. If you see a group of American Goldfinches in the trees behind your house, but you don\u2019t look in the trees in your neighbor\u2019s yard, does that mean they aren\u2019t there? To go from data points to a smooth range map, you need a model.<\/p>\n<p>\u201cNo matter how much data you collect,\u201d Kelling says, \u201cwe\u2019re never going to count all the birds, and we\u2019re always going to have biases to population centers.\u201d In Kelling\u2019s view, the biggest breakthrough in creating the new range maps was not accumulating eBird\u2019s mountain of data, it was figuring out how to model it.<\/p>\n<p>For this, a statistician on the eBird team named Daniel Fink invented something called a STEM model, or \u201cspatiotemporal exploratory model.\u201d He takes eBird\u2019s millions of observations and associates them with other big datasets on climatic variables, ecological conditions, and human factors.<\/p>\n<p>\u201cSome of the raw data [that are available] are pretty mindblowing,\u201d Fink says. \u201cSatellites are going around the earth every couple\u00a0of days, and you can sort of Lego it all together.\u201d His models correlate these environmental variables with where sightings have occurred\u2014and, crucially, where sightings haven\u2019t occurred as well (see \u201cSolving the Area 51 Effect,\u201d at left). Then the models look for other parts of the landscape that match those variables. If the match is close enough, the model includes it as part of the range map.<\/p>\n<p>Of course there\u2019s a catch. Birds often change where they live in different seasons or parts of their range. Tree Swallows winter in wooded wetlands in the South, but not in frozen Michigan and Wisconsin. Indigo Buntings look for insects along forest edges in early summer, but by late summer they\u2019re in fields eating seeds. How is a computer model to know?<\/p>\n<p>Fink solved the problem with what he calls a \u201cslice-and-dice\u201d approach. He chops the United States up into about 100 smaller regions. For each one, he runs up to 200 models looking for correlations between bird detections and environmental variables. The models find correlations that are valid for that area and time of year, but the analysis prevents them from extrapolating further.<\/p>\n<p>Fink averages those 200 model results to get a single probability for a certain area at a certain week of the year. That number becomes a pixel of color on the map. To get the next pixel, he runs all those models again. Then he does it all again for the next week of the year. And all of that again for the next species of bird. That\u2019s what fills up 2.5 million hours on a supercomputer.<\/p>\n<h3>The Orchard Oriole Harry Potter Range Map<\/h3>\n<div class=\"wp-block-group legacy-sidebar sidebar-alignright has-lightgray-background-color has-background\">\n<h4>How to Use eBird to Find Birds<\/h4>\n<p>The best way to get started as an eBirder is to use it to help you find birds. Curious about a bird\u2019s range? Type in its name and explore an interactive map of all its eBird sightings. Planning a road trip and looking for local specialties? Use Hotspot Explorer to home in on places with the most birds. Get a list of which species to expect with eBird Targets and find out their frequency through the year with eBird\u2019s bar charts. Once you\u2019ve gotten acquainted, you may want to start entering your own sightings.<\/p>\n<p>Get started at <a href=\"http:\/\/bit.ly\/eBirdExplore\">http:\/\/bit.ly\/eBirdExplore<\/a>.<br \/>\n<\/br><\/p>\n<h4>How to Enter the Perfect eBird Checklist<\/h4>\n<p>With just a few easy steps you can ensure that your checklist contains all the data scientists need to conduct their analyses:<\/p>\n<p><strong>Report All the Species You See <\/strong><\/p>\n<p>When you indicate your list includes all the species you saw, then analysts can also infer all the species you didn\u2019t see\u2014 and this \u201cnegative data\u201d is extremely helpful for the models (See \u201cSolving the Area 51 Effect\u201d on page 36).<\/p>\n<p><strong>Record Your Effort <\/strong><\/p>\n<p>Your species list is just part of what scientists need. They also need to know how long you spent, how far you went, and how many birders were with you.<\/p>\n<p><strong>Estimate Numbers <\/strong><\/p>\n<p>Making even a rough count of how many birds you saw is more helpful than just checking off a species. The eBird site has tips on how to estimate numbers.<\/p>\n<p><strong>Break Up Your Checklists <\/strong><\/p>\n<p>Entering a checklist for each stop on a birding tour makes the data more site-relevant. That\u2019s more valuable than entering one giant list for your outing.<\/p>\n<p><strong>Enter Your Data in the Field <\/strong><\/p>\n<p>A smartphone app such as BirdLog makes it all easier. It lets you skip a paper checklist and enter birds while they\u2019re fresh in your mind, alerts you to take notes if you\u2019ve seen something unusual, and keeps track of time spent<\/p>\n<\/div>\n<p>A couple of years after that morning in Montana, the same four birders gathered around a computer monitor along with Fink to look at the new maps. The picture\u2014a map of the continental United States\u2014started quiet and black, showing January, when all the Orchard Orioles were in Central America. Then the map began to tick through the year. Spring arrived with a flash of orange up the Mississippi and across the East. A brief moment of stillness indicated birds settling down for the breeding season. By late summer the color had drained out of the Northeast, leaving an orange glow of orioles across the Great Plains into September.<\/p>\n<p>\u201cThat was the first point where you could see the division in timing,\u201d Wood said, \u201cthat suggested there were these two different populations of Orchard Orioles that nobody that we know of had ever really understood before.\u201d<\/p>\n<p>The map had communicated in just a few seconds what few birders had ever put together. In the history of discoveries it was a small one, but one of hundreds now being made visible.<\/p>\n<p>Take the Eastern Phoebe. \u201cYou could go to [regional books on] birds of Florida, Georgia, Texas, Colorado, and get a feel for the migration of Eastern Phoebes,\u201d Iliff says. \u201cBut suddenly with a STEM map you\u2019re piecing things together in a way that no one ever has before.\u201d<\/p>\n<p>Same with Yellow Warbler. \u201cSomeone like Paul Lehman might be able to tell you that Yellow Warblers in the West arrive earlier than ones in the East, but when you see that on a map it\u2019s like a month-and-a-half difference,\u201d Iliff says. \u201cIn my mind there\u2019s no birder in the country who can understand some of the big-scale differences with migration phenology like you can when you see it on a map.\u201d<\/p>\n<h3>Pretty, Powerful<\/h3>\n<p>As pretty and as helpful for birders as they are, the really meaningful advance of STEM maps is how they signal the arrival of Big Data\u2013type analyses to the fields of conservation and basic ornithology.<\/p>\n<p>\u201cEvery time you look at one of these heat maps, you come\u00a0away with a question,\u201d Kelling told me, and those lead to new hypotheses and new avenues of research. \u201cSo our goal is just to crank these [models] out annually\u2014hundreds of them. And then researchers\u2026can look at that data and start generating and testing hypotheses.\u201d<\/p>\n<p>Scientists working on the <em>State of the Birds Report<\/em> in 2011 and 2013 did just that. For 350 species, they determined how much of each species\u2019 range is owned by a variety of public and private landowners. For the first time, land managers had a specific, personalized picture of which species were most important for them to pay attention to.<\/p>\n<p>Maps of ranges are one thing, but how do you know where a species\u2019 strongholds are? By modeling abundance, rather than simple occurrence. That was the goal of a postdoctoral associate named Alison Johnston, in 2013. Her models allowed The Nature Conservancy to set up temporary wetlands in California for some 200,000 migrating shorebirds (see \u201cMoneyball for Shorebirds,\u201d Autumn 2014).<\/p>\n<p>Someday, she says, a birder might be able to ask a computer, \u201cWhich three places can I go today to see the most numbers or species of birds?\u201d Likewise, a conservationist would be able to ask, \u201cIf I have enough money to create three land reserves, where could I put them to protect the most birds?\u201d<\/p>\n<p>The range maps started off with U.S. birds because that\u2019s where eBird has the most data. But global participation is growing, and Kelling has his sights on creating year-round maps for all 800 North American species across the entire Western Hemisphere. (They\u2019ve already made them for a few species, including Barn Swallow, Black-throated Blue Warbler, and Blackpoll Warbler.) To do this, Fink had to make another advance in STEM models: giving them the ability to adjust their resolution depending on how much data are available.<\/p>\n<div class=\"wp-block-group legacy-sidebar sidebar-alignright has-lightgray-background-color has-background\">\n<h4>Listen<\/h4>\n<p>Arthor Hugh Powell is a science editor at <em>Living Bird<\/em>. <a href=\"http:\/\/tinyurl.com\/mcct7qq\">Listen to\u00a0an interview with him about this article<\/a>.<\/p>\n<\/div>\n<p>These improvements will circle back to birders in the same way the initial maps did. Imagine a birding trip to Peru, where 1,800 species vie for your attention. \u201cAs good as that bird guide is,\u201d says Iliff, \u201cit\u2019s got almost 2,000 species in it. Nobody can absorb that in studying for a trip.\u201d<\/p>\n<p>In five years, Wood says, maybe there\u2019ll be a Merlin-like app that uses STEM-generated ranges to help you understand which species are in the running. It\u2019s the thing that bird watchers and their field guides have been pursuing for decades: tools that don\u2019t necessarily do the identification for you, but narrow down the possibilities and give you a place to start. The more data that pour into eBird (currently at the rate of several million observations per month), the more accurate the models will get. There\u2019s just one catch, and it\u2019s also the data. In the time since the first STEM models were run, the amount of data in eBird has doubled. The prospect of newly calculated models is enough to make a statistician salivate. But first they\u2019re going to need a few million more hours on a supercomputer.<br \/>\n<!--http:\/\/digital.livingbird.org\/livingbird\/winter_2015?#pg32--><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Oddly enough, when the new age of range maps began to dawn, it happened on an empty stretch of country road in eastern Montana. It was early September 2009. Four<a class=\"excerpt-read-more\" href=\"https:\/\/www.allaboutbirds.org\/news\/a-hundred-million-points-of-light\/\" title=\"ReadEBird and A Hundred Million Points of Light\">&#8230; Read more &raquo;<\/a><\/p>\n","protected":false},"author":8,"featured_media":16250,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_tec_requires_first_save":true,"_birdpress_living_bird_toc":0,"_birdpress_living_bird_toc_title":"","_birdpress_featured_image":false,"_birdpress_hero_toggle":false,"_birdpress_hero_type":"image","_birdpress_hero_image_type":"image","_birdpress_hero_style":"default","_birdpress_hero_ratio":"","_birdpress_hero_h1":"","_birdpress_hero_media_id":0,"_birdpress_hero_media_array_id":[],"_birdpress_hero_media_array":[],"_birdpress_hero_media":0,"_birdpress_hero_video_id":0,"_birdpress_hero_video":0,"_birdpress_hero_youtube":"","_birdpress_hero_content":true,"_birdpress_hero_byline":"","_birdpress_hero_byline_bottom":"","_birdpress_hero_button_link":"","_birdpress_hero_button_text":"","_birdpress_hero_button_color":"","_birdpress_hero_date":false,"original_guid":"","_birdpress_hide_search":false,"_birdpress_page_width":"","_birdpress_global_cta":false,"_birdpress_widget_sidebar":"","_birdpress_next_article":0,"_birdpress_next_article_title":"","_birdpress_prev_article":0,"_birdpress_prev_article_title":"","_birdpress_sub_navigation_id":0,"_birdpress_sub_navigation":"","_birdpress_sub_navigation_title":false,"_birdpress_anchor_navigation_id":0,"_birdpress_anchor_navigation":"","_birdpress_postType":"both","_birdpress_categoryID":0,"_birdpress_tagID":0,"_birdpress_parentPostID":0,"_birdpress_parentPostTitle":"","_birdpress_menuID":0,"_birdpress_menuName":"","_birdpress_listHeader":"","_birdpress_listLayout":"card-display","_birdpress_listColumns":"","_birdpress_maxItems":12,"_birdpress_listPaginate":true,"_birdpress_displaySort":true,"_birdpress_sortOrder":"DESC","_birdpress_sortBy":"date","_birdpress_listID":"","_birdpress_listClass":"","_birdpress_displayImages":true,"_birdpress_displayCaptions":false,"_birdpress_displayExcerpts":false,"_birdpress_attTop":"","_birdpress_attBottom":"","_birdpress_showLogos":false,"_birdpress_post_logo":0,"_EventAllDay":false,"_EventTimezone":"","_EventStartDate":"","_EventEndDate":"","_EventStartDateUTC":"","_EventEndDateUTC":"","_EventShowMap":false,"_EventShowMapLink":false,"_EventURL":"","_EventCost":"","_EventCostDescription":"","_EventCurrencySymbol":"","_EventCurrencyCode":"","_EventCurrencyPosition":"","_EventDateTimeSeparator":"","_EventTimeRangeSeparator":"","_EventOrganizerID":[],"_EventVenueID":[],"_OrganizerEmail":"","_OrganizerPhone":"","_OrganizerWebsite":"","_VenueAddress":"","_VenueCity":"","_VenueCountry":"","_VenueProvince":"","_VenueState":"","_VenueZip":"","_VenuePhone":"","_VenueURL":"","_VenueStateProvince":"","_VenueLat":"","_VenueLng":"","_VenueShowMap":false,"_VenueShowMapLink":false,"_tribe_blocks_recurrence_rules":"","_tribe_blocks_recurrence_description":"","_tribe_blocks_recurrence_exclusions":"","wds_primary_category":0,"wds_primary_topic":0,"wds_primary_content-format":0,"wds_primary_cornell-lab-project":0,"wds_primary_host-project":0,"wds_primary_read-more-tag":0,"footnotes":""},"categories":[1],"tags":[],"topic":[1049,998,1043,1119],"content-format":[1055],"cornell-lab-project":[1069],"host-project":[1137],"read-more-tag":[],"class_list":["post-16247","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-uncategorized","topic-citizen-science-get-involved","topic-news-and-features","topic-science-conservation-news-and-features","topic-try-this","content-format-article","cornell-lab-project-living-bird-magazine","host-project-living-bird"],"metadata":{"associated-posts":[""],"wpa_off":[""],"_edit_lock":["1530817752:4"],"_edit_last":["4"],"_thumbnail_id":["16250"],"wdsi_message_id":[""],"wdsi_do_not_show":[""],"custom-byline":["<h5>By Hugh Powell<\/h5>\r\n<h6>From the <a href=\"https:\/\/www.allaboutbirds.org\/news\/living-bird-winter-2015-table-of-contents\/\">Winter 2015<\/a>\u00a0issue of <a href=\"https:\/\/join.birds.cornell.edu\/ea-action\/action?ea.client.id=1806&amp;ea.campaign.id=24577\">Living Bird<\/a> magazine.<\/h6>\r\n<small>In eBird\u2019s first 10 years of existence, bird watchers contributed 100 million observations. It took only 2 more years to reach the next 100 million.<\/small>"],"banner-video":[""],"banner-image":["https:\/\/www.allaboutbirds.org\/news\/wp-content\/uploads\/2015\/01\/100Million.jpg"],"fallback-videobanner-image":[""],"original_guid":[""],"banner-text-style":["light"],"featured-image-display":["no"],"_birdpress_next_article":[""],"_birdpress_prev_article":[""],"_webdados_fb_open_graph_specific_image":[""],"_wds_meta-robots-adv":[",,,"],"banner-style":["default"]},"_links":{"self":[{"href":"https:\/\/www.allaboutbirds.org\/news\/wp-json\/wp\/v2\/posts\/16247","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.allaboutbirds.org\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.allaboutbirds.org\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.allaboutbirds.org\/news\/wp-json\/wp\/v2\/users\/8"}],"replies":[{"embeddable":true,"href":"https:\/\/www.allaboutbirds.org\/news\/wp-json\/wp\/v2\/comments?post=16247"}],"version-history":[{"count":0,"href":"https:\/\/www.allaboutbirds.org\/news\/wp-json\/wp\/v2\/posts\/16247\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.allaboutbirds.org\/news\/wp-json\/wp\/v2\/media\/16250"}],"wp:attachment":[{"href":"https:\/\/www.allaboutbirds.org\/news\/wp-json\/wp\/v2\/media?parent=16247"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.allaboutbirds.org\/news\/wp-json\/wp\/v2\/categories?post=16247"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.allaboutbirds.org\/news\/wp-json\/wp\/v2\/tags?post=16247"},{"taxonomy":"topic","embeddable":true,"href":"https:\/\/www.allaboutbirds.org\/news\/wp-json\/wp\/v2\/topic?post=16247"},{"taxonomy":"content-format","embeddable":true,"href":"https:\/\/www.allaboutbirds.org\/news\/wp-json\/wp\/v2\/content-format?post=16247"},{"taxonomy":"cornell-lab-project","embeddable":true,"href":"https:\/\/www.allaboutbirds.org\/news\/wp-json\/wp\/v2\/cornell-lab-project?post=16247"},{"taxonomy":"host-project","embeddable":true,"href":"https:\/\/www.allaboutbirds.org\/news\/wp-json\/wp\/v2\/host-project?post=16247"},{"taxonomy":"read-more-tag","embeddable":true,"href":"https:\/\/www.allaboutbirds.org\/news\/wp-json\/wp\/v2\/read-more-tag?post=16247"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}