Have you ever seen a big group of blackbirds in a large tree and estimated, say, 300-400 … only to have clearly 200 or more fly out of the tree, and it looks like nobody left? And then another couple hundred leave, and it still looks the same? Well, there are ways to get a better count.
Let’s face it, no counting technique for a large flock will give exactly the number of birds, but we can get close, and often we can get really close. For convenience let’s consider any technique that tabulates one bird at a time as a “count,” and those that tabulate by groups of birds at a time an “estimate.”
When submitting observations to eBird estimates are better than recording an “x,” and counts are better than estimates. Counts also validate and help improve future estimates. Here we will present some tips for doing counts of individual birds in large flocks. See below for links to articles on making quality estimates.
We’ll look at two easily available tools for counting birds one-by-one when there are lots and lots of birds: hand clicker-counters, which can be purchased for less than $10, and your camera.
Clicker-counters have many advantages over counting out loud or in your head:
- You won’t lose track of where you are in the count or skip over numbers.
- Clicker-counting is very much faster than oral or mental counting … helpful for many reasons.
- You pay attention to each bird, which is an excellent way to scan for unexpected rarities.
- With two clickers, one in each hand, it is super simple to count two species at a time.
- You easily can pause at any time with no fear of forgetting where you were.
- You can sometimes carry on a conversation while clicking.
- And, always be sure to reset a clicker to zero before each count.
Counting from photographs of flocks often is better than trying to count individual birds when they are in flight, moving about, densely packed, or about to disappear.
Photographs are most useful when the birds are large enough to clearly see individuals, and this may require multiple images. Overlapping areas can be marked with a pen, or the photos can be physically taped together or electronically stitched for the count.
There are several ways to count from photographs so you don’t miss birds or accidentally double count. Again, using a clicker makes the count much faster and less subject to error.
- On prints draw lines around groups of 20-30 birds for all the birds in the photo, and then put a check mark in each group once you have counted it. It is easiest to count if, when you draw the line, you enclose an area longer than wide rather than approximating a circle.
- On prints use a sharp felt tip pen to mark each bird as you make your count. This only works if the birds are fairly well spread apart and the marker doesn’t blur.
- Open the image in editing software and use a drawing tool to put a dot on each bird as you count.
One disadvantage of photographs is that they undercount when birds are hidden behind other birds. In those situations, if the flock is stable enough to count them in real life do so; the movement of individual birds can be used to see and count the hidden birds. One count of a flock of Snow Geese at Dead Creek WMA from multiple enlarged photographs yielded 4907 birds, whereas the count of live birds was 5657. Thus some 750 birds were hidden from view in the still photographs … birds that were behind other birds and birds that were walking in and out of shallow hollows.
When to count
In Vermont there are several common situations where there are large flocks that can be counted.
Birds in leafless trees (e.g. Grackles, Waxwings, Snow Buntings). These can be upwards of a few thousand and typically are quite underestimated. A February count from multiple, taped-together photographs of a solitary tree in Ferrisburgh yielded 3922 Snow Buntings. Sometimes back-lighted makes for easier counting.
Birds in leafed-out trees (e.g. Red-winged Blackbirds, Robins). Who knows how many are hiding there! Wait until they begin to exit the tree and photograph them as they stream out. If you are far enough away, you’ll be able to paste together the photos.
Birds on the ground, barn roofs, utility wires, ice (e.g. Geese, Gulls, Starlings, Swallows, Pipits). Take photos before they unexpectedly take flight. Clicker-count if convenient, or if photos will be hard to interpret, or if the birds are tightly packed and/or milling about. For birds on the ground, get as much height as you can in order to see behind the foreground birds. Even a few feet can make a big difference.
Birds on the water (e.g. Geese, Ducks, Gulls, Cormorants). Most folks seem to have a really hard time with estimating numbers on the water, with significant overestimates and underestimates being common. Photos can be useful or not depending on flock activity, wave action, heat distortion, distance, and your height above the water. For some species back-lighting helps, for others direct lighting is best (choose the best time of day, if possible). Clicker-counting is often much better than photographs since you can wait between clicks for birds that had dived, are hidden by waves, are masked by other birds, etc. Once accustomed to clicker-counting, whole-lake scans and extensive rafts can be counted with relative ease and confidence.
Birds in the air (e.g.Blackbirds, Bobolinks, Swallows, Starlings, waterfowl, shorebirds). Photographs are by far the better tool when the birds are not lost in background clutter. Get enough distance from the birds so that the entire flock or unit can be captured. “Burst” or related camera settings can be a real help. When species IDs are important frontal lighting is best. For other species back-lighting or other silhouetting can help the counting.
Using estimate-counts. In an estimate-count, birds are counted in units of 10, 50, 100, etc. or flocks are subdivided visually by successive halves until counting or estimating the number in one subdivision. Two fine articles explaining these techniques can be found on eBird:
Bird Counting 101 deals with basic concepts, why estimating numbers is important, and illustrates some examples of techniques for estimating single-species flocks. Bird Counting 201 gives approaches to estimating more difficult masses of birds, including large numbers, moving flocks, and mixed species flocks.
Finally, when there is just no time or means to do a precise count or a close estimate, decide upon a range such as “There are at least 700 Goldeneyes out there on the water, but I’m sure there are not over 1500.” Add that comment when you make your submission to eBird, then pick a number from within that range and use that number. You might feel comfortable with the minimum number, or you might choose the average. In either case it is usually much better from an eBird standpoint than submitting an “x”.
By Ian Worley, Vermont eBird Reviewer – Champlain Region