You should use a breeding code whenever you see behavior that indicates the bird might be breeding, which generally depends on the time of year. Some breeding codes are strong evidence of breeding and can be used at any time of year, like code NE (nest with eggs). Other breeding codes are best used during a bird’s breeding season, like code P (pair in suitable habitat). You can use the Breeding Timeline to see when each species nests in Maryland and DC and when each species is overwintering or migrating. Avoid using weak codes like code H (in suitable habitat) or code S (singing in suitable habitat) during the species’ overwintering or migratory season.
Code F (flyover) should be used when a bird is flying overhead and not using the habitat below it. You should not use it when a bird is just flying. As you can imagine, this can be quite subjective, so if in doubt it is better not to use the code. This code is primarily used to help develop habitat association models, so one way to think about it is whether you would expect to see that species in the habitat you are in. If you would not expect this—say, a Double-crested Cormorant flying over while you are in a forest—it may qualify for code F.
Suitable habitat is the general ecosystem that a species would be expected to nest in. If you see a Rock Pigeon in the city, you can give it code H (in suitable habitat); it does not need to be under a bridge near an appropriate nesting site to receive this code. This code, despite being one of the most frequently used codes, is one that requires some basic knowledge of what each species uses for nesting habitat. Year-round residents may be in suitable nesting habitat all year, so code H should not be used during their non-breeding, overwintering season. Suitable nesting habitat for migratory species is found within their breeding range, not along their migratory route.
While there is considerable overlap, there are two main categories of bird vocalizations: songs and calls. Songs are generally considered to be loud, often long and complex vocal displays of territorial birds. They are often important for courtship and maintaining pair bonds. Calls are considered to be short and simple vocalizations used to communicate, among other things, distress, warnings, flight, or to maintain flock cohesion. For atlasing purposes, don’t get hung up too much on technical definitions. Basically, if it sounds like a song, you can use code S (singing). You can also use code S for woodpecker drumming or other territorial sounds.
Code S7 (a bird singing in the same place for 7+ days during its breeding season) should be used to describe a bird that you are reasonably sure is the same individual because it describes a bird that is likely to have a nesting territory. Some birds maintain winter territories, and some birds sing during migration, so it’s important to use this code at the right time of year. For example, a Song Sparrow singing in February may just be overwintering there, so using code S7 for that observation would not be a good choice.
Code M (7+ birds singing during their breeding season) is a good code to use for common birds. Similar to code S7, it should only be used during a species’ breeding season. If you hear seven or more birds singing in the same day and in the same block and it is during their nesting season, you can use code M.
Knowing the difference between two birds that are nesting together (a pair) and two birds that are just close to each other can be difficult. First, consider time of year. Most birds do not maintain pair bonds during the winter, so using code P (pair in suitable habitat) outside of a bird’s breeding season is not usually a good choice. Next, try to determine their sex and age; if they are juveniles or the same sex, you can safely rule out code P. If you cannot determine sex or age, you will have to rely more heavily on their behavior. A pair should be two birds that are on their own and not part of a larger flock and are generally moving together or perched close together. If you are watching a species that is normally solitary, two individuals close together is generally enough indication that they are a pair. If you are unsure about the birds in question, it’s better not to use code P.
Waterfowl are an exception to the general rule. They routinely display, form pairs, and even copulate well before they migrate north to their breeding grounds. Wait until after their migration is complete to use code P (and code C) for waterfowl.
An agitated bird (code A) is upset by a different species, generally because that species is perceived as a predator close to the nest. A territorial bird is upset that a competitor is inside its territory. Use code T (territorial) when a bird is upset by an individual of the same species and code A when a bird is upset by a different species. For example, a Red-winged Blackbird chasing a Northern Harrier out of a marsh would receive code A, while two American Robins tussling on the lawn or a Northern Cardinal fighting his reflection would get code T. Be aware that territorial or agitated behavior can occur outside of the breeding season when birds are defending wintering territories, so these codes are best used within each species’ breeding season.
Use code CN (carrying nesting material) when you see a bird transporting sticks, grass, string, or other nesting material. Use code NB (nest building) when you see a bird at the nest site building their nest. For wrens and woodpeckers, use code B (wren/woodpecker nest building) instead of code NB. Wrens and woodpeckers build dummy nests and roost holes respectively, so nest construction is not certain Confirmation.
You should use code ON (occupied nest) when you believe eggs are being incubated or chicks are being brooded but neither are visible. If a bird remains in a cavity or sitting on a nest for a considerable period of time (this is subjective, so use your best judgement and be conservative), you can use code ON. If a raptor is only perched on a nesting platform or if a bird does not remain in a cavity for long, you should use code N (visiting probable nest site). Keep in mind that some species will exchange incubation duties and many species use cavities to roost in at night. If a male enters a cavity and a female leaves, or vice versa, you can use code ON. If you see a bird enter a cavity late in the evening or leave a cavity early in the morning, especially outside of their breeding season, you should not apply a breeding code; this is likely a roost site.
Use code FL (fledged young) when you find a young bird that has recently fledged (meaning it left the nest) and is still dependent on its parents. Do not use it to describe a juvenile bird (meaning a bird hatched that summer) that is independent of its parents—even if it is still with its parents. For example, Canada Geese remain in family groups for most of the year, but you should not use code FL on them once the goslings can fly. Wild Turkey chicks can fly within a couple of weeks, but are still dependent on their mother for a few weeks after that and can receive code FL. Basically, if it looks like a baby bird, you can use code FL.
Confirmed codes—codes that involve nests, eggs, or chicks—are the gold standard of atlasing. If you find a nest, you know that bird nested there! But nests can also be time-consuming to find. Instead of spending a lot of time in your block finding as many Confirmed codes as possible, it is more efficient to look for Probable codes. By focusing on Probable codes, you will find Confirmed codes along the way and cover more ground.
Be especially aware of birds carrying nesting material (code CN), carrying food (code CF), and fledged young (code FL). The young of some birds, such as American Goldfinches or Great Horned Owls, have distinctive calls. Learning these can go a long way towards finding more Confirmed species in your block.
When you submit a complete checklist to eBird, you are reporting all the birds that you were able to identify. The opposite is true for an incomplete checklist; you are not reporting all the birds you identified. Each checklist also uses a ‘protocol’ to describe how you were atlasing. You can have Traveling, Stationary, or Incidental checklist protocols. Traveling or Stationary checklist protocols can be complete or incomplete, but Incidental checklist protocols are always incomplete. Since complete checklists are much more valuable for data analysis, you should submit complete checklists whenever possible—but don’t fudge it! If you submit a complete checklist even though it does not include all the birds you identified, you will be submitting misleading data.
Think of the difference between complete and incomplete checklists as the difference between your notebook and a conversation with your neighbor. In your notebook, you record everything you see; this is like a complete checklist. In a conversation with your neighbor, you only mention the highlights from your trip, such as an owl that you found; this is like an incomplete checklist. Both types are useful for different situations.
Incidental checklists should be used when you aren’t actively looking for birds; instead you had an ‘incidental’ observation of a bird, like an owl on the side of the road while you were driving. Incidental checklists are always considered incomplete because this protocol assumes that you did not have opportunity to detect other species that were around. It assumes that you are just reporting a notable sighting (like an owl on the side of the road). If you stopped and watched the owl for a few minutes, then you can submit a Stationary, complete checklist. Otherwise, using an Incidental checklist for this situation is the best choice.
Submitting a complete checklist means that you reported all of the birds you were able to identify. By submitting a complete checklist, you are also reporting zero for every species that you did not detect. This is tremendously valuable data; zeroes are just as important as the birds you did see. We can use this information to calculate how easy (or hard) it is to detect a bird, then use that detectability number to predict where a species occurs. Creating these predictions is only possible with complete checklists, so whenever you can, try to report all the birds you identify.
If I didn’t detect any birds on my eBird checklist, should I still submit it?
Yes, you should submit checklists even if you did not find any birds. This sometimes happens, especially with nocturnal checklists. Checklists with zero species are still useful for analysis, and they show other atlasers that someone has spent time atlasing in a block. For example, imagine a block that has no owls in it. You spend two hours at night listening for owls, but don’t find any. If you don’t submit that checklist, no one will know anyone was in that block looking for owls.
Your eBird checklists should contain observations that you personally made. You can encourage the original observer to join eBird and participate in the Atlas themselves, or you can go confirm their sighting. Of course, that is not always possible. If the observation is of a common bird (say, a Carolina Wren nesting in their backyard) in a well-atlased area, the Atlas will likely capture similar data for that species. If it is of a less-common species that is unlikely to be observed by someone else (say, a Mourning Warbler nesting in their backyard—wow! Lucky!), then you should contact your County Coordinator and notify them. If possible, provide the original observer’s name and contact information so that any subsequent questions can be directed to the original observer. If your County Coordinator deems it appropriate, they can enter the observation into a regionwide Atlas eBird account.
These second-hand observations are not as important in this Atlas as they have been in the past. In past atlases, it was important to find records of each species in as many blocks as possible because there was no good way to record zeroes—no one knew if a species wasn’t reported or if it wasn’t detected. The only solution was to look harder. Now, complete checklists, observer ID, and effort information all work together to provide information about zeroes. We can couple this information with habitat data to create detailed predictions, or models, about where each species nest in Maryland and DC. Second-hand observations don’t have the complete checklist, observer ID, or effort information that contribute towards creating these models, so they are much less valuable.
If you are atlasing near a block boundary, you may observe birds in the other block. You should include these birds on your current checklist (this is required for you to submit a complete checklist), but you should not include a breeding code for those individuals. Instead, finish your checklist and make a new, incomplete checklist located where you saw those birds and use the appropriate breeding code. For example, imagine you are atlasing along a lakeshore. The block boundary runs along the middle of the lake. You only want to include breeding codes for the block you are in, not the neighboring block. As you are atlasing, you see a pair of Ospreys perched on a nesting platform across the lake. You enter both Ospreys into your checklist, but you do not include a breeding code for them. After you finish your walk along the lakeshore, you create an incomplete checklist located at the site of the nesting platform. You record two Ospreys on that checklist and include the breeding code N (visiting probable nest site).
If you are worried that a bird you found might be disturbed if people learn where it is, you have a couple of options to minimize this risk. First, you can wait eight days to submit your checklist so it will not be included on any eBird rarity or needs email alerts. You can also wait until the end of the nesting season to submit your checklist. A third option is to hide your checklist, but it’s important to note that if you hide your checklist it is also removed from the Atlas database; your observation will not count towards the Atlas. If you hide a checklist, please be sure to unhide it at the end of the nesting season.
If the observation you are concerned about is of a Sensitive species, you can submit your checklist as normal. Sensitive species in Maryland and DC include Black Rail (year round), Northern Goshawk (April–August), Barn Owl (year round), Long-eared Owl (April–September), Short-eared Owl (May–September), Northern Saw-whet Owl (April–September), and Swainson’s Warbler (April–August). These species will not show up on any email alerts, their location on the map is buffered to 150 square miles, and they are concealed on the checklist they were reported on.
If you are unsure whether a bird you found might be susceptible to disturbance, contact your County Coordinator for guidance.
If you had permission to atlas on private land but are concerned that posting a checklist on eBird might encourage other birders to unintentionally trespass, you should include the word ‘private’ in the location title (e.g., Smith Farm—private). This is an unmistakable indication to anyone who views your checklist that the location requires landowner permission before accessing. A second option is to post the location to the nearest road; this can be a good solution if the landowner is uncomfortable with a checklist being posted to their property. If you feel that neither of these options are sufficient, you can also wait to submit your checklist until after the nesting season is finished. Try to avoid submitting checklists at the block level; these checklists are far less valuable for any subsequent analyses.
Your eBird checklists only contribute to the Atlas if you submit them through the Atlas portal. The Atlas portal is essentially a label on your checklist that says it was collected following Atlas methods. If you are aware of breeding behavior, intend to use breeding codes, and are following Atlas methods, then you should submit your checklist to the Atlas portal. Since these requirements probably apply to all of your birding while the Atlas is happening, it basically means you can submit all of your checklists to the Atlas portal.
You can find instructions for how to change portals in Appendix C of the Atlas Handbook, available at ebird.org/atlasmddc/about/handbooks.
If you hear a single owl hooting, you should use code S (singing). If you hear a duet (listen for this in Great Horned and Barred Owls), you should use code C (courtship display). If you are atlasing at night and using playback (review the Atlas Handbook for guidelines on how to do this safely and effectively), use code T (territorial) on any owls that respond aggressively.
Anyone can atlas anywhere in Maryland and DC, at any time. By becoming the primary atlaser for a block, you agree to make sure the block completion targets are met by the end of the Atlas. Thank you to everyone who has agreed to do this! This helps County Coordinators know which blocks are less likely to receive sufficient effort by the end of the Atlas. Unlike in past atlases, as a primary atlaser you are not responsible for all of the block’s observations and you do not need to verify other people’s observations. Feel free to atlas anywhere and to sign up for as many blocks as you can manage. You can see which blocks are available using the Atlaser Block Tool and you can contact your County Coordinator to sign up.
Unlike species or abundance observations, there are no filters that flag unusual breeding codes for eBird reviewers. Instead, reviewers monitor incoming checklists and recent breeding codes submitted in each block. This manual review system is bound to miss some improperly coded species, but that’s okay. Each year, we will filter the dataset using computer scripts and reinterpret unlikely breeding statuses recorded in eBird. This includes things like early or late codes (e.g., code P (pair) for Hooded Mergansers in February), records from well outside where a species is expected to breed (e.g., code S (singing) for a Yellow-rumped Warbler in Harford County), and codes that don’t apply to a given species (e.g., code S for a Black Vulture). The bulk of the quality control will occur during each non-breeding season and prior to our analyses, rather than immediately after an observation has been submitted.