Wednesday, 5 February 2020

Out for the count


I love counting birds. And plants. And moths. Not forgetting butterflies of course. Always have done, ever since my early birding days, a discipline that was eagerly embraced, primarily as a means to gather data, but, increasingly, as means to an end. It is soothing, therapeutic and calming. Numbers themselves are mantras to be uttered - either out loud or in the head - and is very much a way of being in the now. Ommm...

Pseuds Corner over, and now onto the meat of this post... how do you count, or more precisely, how do you accurately count? 15 Rooks clearly on view in a field holds no barrier to a precise count being made, but what of a flock of several hundred finches that are feeding in the stubble, or thousands of waders scattered across an estuary -all moving, many rotating in and out of view?

There is no doubt that experience does play a part in compiling meaningful, accurate counts in such circumstances. Practice banishes the traps that can be fallen into, such as ‘double-counting'. A knowledge of an area will alert you to ‘hidden’ channels, dips in fields and favoured bolt-holes that will reward diligent observation and reveal their out-of-sight birds. Also, the understanding of the need to shift your viewpoint, in order to facilitate an accurate outcome, is crucial. And a count can go on all day, it doesn’t, and possibly shouldn't, be attempted immediately.

If a flock of pipits are flushed as you walk across a field you will subconsciously estimate how many there are. At this point the estimation will be just that, maybe ‘20-25’. If they obligingly fly around in an arc before landing, you might be given the chance to count them individually.’20-25’ then becomes a precise ‘24’. Over time, your flock estimates will be honed to a point where ‘250’ or ‘175’ will be close enough to be considered valid counts. Most people round-up (or down) in units of 5 or 10. After all, an estimate of 234 seems a bit odd, doesn’t it? There’s invalid precision involved there! My estimates are happily always on the low side. Subsequent accurate counts, when allowable, have always shown that I underestimate. I think most birders do. This was born out time and again when I was chasing the Hawfinch flocks a couple of winters ago, when sensible and accurate assessments were paramount with such historical numbers.

I posted about counting gulls at Beddington SF a while back, and how we reached our final numbers:

“On arrival it was obvious that the landfill site was operational, as thousands of gulls were present. After settling down, having had an hour or more scanning and also having walked across to the enclosed lagoons, the time came to estimate those numbers present. My gut reaction was to plump for a total larid mass of, lets say, seven or eight thousand individuals, based on nothing more than educated guesswork. However, when I tried to add a little bit of common sense to the process, it only helped to prove that my initial estimate was woefully short. I was quite happy to claim a fairly positive count of 375 Lesser Black-backed Gulls. So I used this as a starting point - a number not estimated but actually counted out, one gull at a time. I reckoned that there were at least double the number of Common Gulls present, so I arrived at a total of 750 for them. Pete, Johnny and Frank now joined in. For every Common Gull we had seen there were at least 15, maybe even 20 Black-headed present. Taking the lower figure of 15 and multiplying this by the number of Common Gulls we reached a total of 11,250 Black-headed. Herring Gulls were plentiful, but not as numerous as Black-heads, so we settled on a total of 4,500. Other gull species present were: Great Black-backed (15), Yellow-legged (2), Mediterranean (1) plus a rather fine first-winter Iceland. So, the gull grand total was in fact 16,894, rather more than my initial hunch”.

The incident above was replicated somewhat at Pulborough Brooks a couple of days ago. My first scan across the South Brooks from Hail’s View showed hundreds of ducks on view, and these birds were easily counted, sitting out in clear view on the water, no mere estimates were needed. But it was a different matter when looking out over the North Brooks, as thousands of duck were on show, near and far, scattered over islands, open water, dykes and frequently in flight. I had time on my hands, so spent a while just getting to grips with any flocks that were leaving the area - these were identified and counted. I then divided up the panorama before me into manageable sections, using dyke lines and islands as marker points, and counting within them. Working from left to right I had to keep half an eye on any birds, having been counted, that then moved into my uncounted area (so that they wouldn’t be double-counted), but also aware of any uncounted birds that were flying off and away -  these were then tackled before returning to the core count. It was the ornithological equivalent of keeping plates spinning in the air.

Only after an hour of this could I hope to have some idea of the numbers present. Being stood at one spot for so long meant that some hidden rafts of duck were not missed as they eventually swam into view, and using the gull example above, was able to assess my counts against a ‘known’ quantity - in this case, Pintail. This species was more easily countable, being not nearly as mobile as the Teal and Wigeon flocks, and because they are largish, elegant birds, they stood out. The count of 580 Pintail therefore was used as a marker. I could assess that for every Pintail there were at least four Wigeon. Simple maths took that up to 2,320, which tallied up remarkably well with my Wigeon estimate. And with that figure in mind, and my observation that there were 25% more Teal then Wigeon, my Teal estimate of 3,000 wasn’t that wide of the mark. This may not be scientific, but it certainly gives an honest and useable count. I was also happy in the knowledge that there would be many duck that I would have missed, hidden behind island, deep in channels and way in the distance. - now and again I would get tantalising glimpses of them, but little more. So my final counts were, without doubt, bare minimums of what were actually present.

Large flocks in flight can be awkward to count. If they are in the air for sometime, then you will get several attempts, but if all else fails count 50, or 100, then assess what percentage of the flock you have counted. Then multiply. When stress-testing this method it has always held true.

There may be no way of ever hitting a true, 100% accurate count of a large flock. 10,000 Starlings at a roost might genuinely be 9,979 or 10,032. It could be quite a bit lower or higher. But to record a number within meaningful parameters is way better than just recording the gathering as being 'many' or 'high'. What on earth do those phrases convey?

4 comments:

Gavin Haig said...

Many? High? A couple of my favourite words Steve, along with 'lots', 'few' and 'stacks' :-)

Some of those techniques are new to me though, and if I ever stop being an idler I'll try them.

KReid said...

many years ago - in the seawatching hide at Dungeness, counting brent geese:
BB (retired maths teacher) asks young assistant warden, "How do you count that flock of geese?' - I reply "start with individuals then groups of 10, or 50 etc". Unconvinced, the retired maths teacher turns to the warden and asks, "how do you count them, Mr Warden?". The reply, which silenced the hide for some time, was simple, "Accurately!".

Steve Gale said...

Gav - you can try these techniques out on your 'lots' of Caspian Gulls!

Is that you Keith? I recognise Messers Banson and McMinn in your anecdote...

KReid said...

It is indeed, and your recognition of the characters is spot on, I also remember the total number of brent geese was something like something like 10 298 (the simple days before statistics and confidence limits!)