An Introduction to Prominence
What | Why | Where | How | Example
"Je n'ai fait
celle-ci plus longue que parceque je n'ai pas eu le loisir de la faire
(I have only made this longer because I have not had time to make it shorter.)
Blaise Pascal, Lettres Provinciales
What is Prominence?
Minimum Reascent is perhaps the most used term, but Relative Height, Drop, Rise, and even Independent Height have their adherents and supporters. They can all mean the same thing, though they don't always, but how best to explain it all?
Some years after starting to play with mountain tables I was introduced to another term, Prominence, that has common usage not least amongst a group of peak-bagging enthusiasts on both sides of the Atlantic who converge on the Prominence mail list. I've opted into the use of their term, not least because one of their number  has produced the simplest and clearest definition yet of what it is we are all trying to describe:
So we're trying to measure how much of the height of a summit `belongs' to that summit, not in the geological sense of rock layers but in terms of how much a summit projects from it's surroundings .
[ 1: reference misplaced,
[ 2: "The Concise Oxford Dictionary" (10th edition, Oxford University Press, 1999) describes a prominence as "a thing that projects from something", and we're trying to put a value on that prominence, and to rate a mountain by that measure rather than its mere height.]
Many hillwalkers become hill-baggers -- perhaps this occurs when the idea of walking `all' the summits arises. This in turn gives rise to the need to define the area one is walking in, and to define just what makes summit `collectable'.
The tendency has been to pick a minimum height, and then, unless you intend to walk over every rock and tussock on every ridge (which might sound quite plausible until you happen upon certain moorland hills) you need to set some other criteria, usually a minimum all-round reascent.
The problem is that by concentrating on the mere height of a summit you give scant regard for how distinct it is from other summits. Munro's Tables, (1891 and revisions) probably the first serious tabling of any hills anywhere, lists all Scottish tops above 3000ft, inevitably ignore well separated peaks that fall short on altitude whilst including many a `molehill' that happens to be sat high on the shoulders of a higher peak. Even among the elite group of 280-odd summits deigned to be separate mountains, a minimum reascent or prominence of only 300ft (a little over 90m) will on average suffice for inclusion, whilst a 2999ft summit of any sort will be ignored. 
Over the decades newer tables of hills for various parts of Britain tended to use ever lower minimum heights whilst the minimum reascent required tended to rise. When Alan Dawson published his Relative Hills of Britain the summits of all Britain were tabled on the same basis, probably a first, and at the same time the two figures for height and minimum reascent had became the same thing. This was a major leap -- but the hills were still ranked by absolute height, i.e altitude.
Being at the time in the early stages of peak-bagging and not yet fully immersed when hearing of Dawson's tables, but being numerical and systematic of mindset, I was delighted with this development and was soon wondering when the next step might be taken -- tabling the hills without any regard for altitude, and ranking by reascent alone.  Dawson was still ranking summits such that, for example, a 1000m high summit with the minimum 150m reascent still outranked a 999m summit with 999m reascent. Maybe I'd get around to doing it myself.
I began in 1993 by drawing a sketch map of prominent Welsh peaks and their interconnecting ridges and separating cols -- the Significant Summits as I first knew them. After finding my original 1000ft qualifying minimum reascent revealed fewer tops than hoped for, stepping down to 500ft provided a sketch that was pleasingly recognizable as a skeletal frame of Wales, with the major river valleys clearly evident between the lines of the ridges. The Welsh coastline would wrap around about 5 to 15km from the outer peaks.
The sketch also showed what a nonsense political boundaries can be when considering hills and the like -- there were a couple of bits and pieces floating disjoint from the main pack, and this was going to be even more pronounced on its sister map of English summits with groups entirely cut off by the Severn from the English watershed proper.
Clearly, using rivers as boundaries was going to make far more sense. Extending the Wales sketch to the Severn (and to the Dee in the north) was easy and obvious but where to cut the Welsh watershed from the English was an open question.
One thought was to draw the line at the triplepoint where the watersheds met, but that could have been at a summit itself. Following the Berwyn ridge as it tapered out into England I found the lowest "col" in the watershed a little west of Market Drayton in Shropshire, just before it met the main north / south watershed of Britain.
It turns out this col (if you can call it that, being the lowest point in an area of generally low and rolling ground) is the lowest between Scafell Pike (highest summit in England) and Snowdon, suggesting itself to be the perfect place to define the furthest extent of "Greater Cambria" -- all of Wales and the bits of England on the Welsh side of the river. Of course a col has rivers either side of it, and the Weaver (which flows into the Mersey and not, as expected, the Dee) then defines the region to its north-east extent.
[ 3: There will be a link
here to an article on the arbitrary nature of Munro's Tables and
attempts over the last 30 years to rationalize them, when that is
[ 4: Tables with a minimum prominence and no minimum height criteria were created for areas outwith these Isles certainly as early as the 1980s. I am to date unaware whether or not any went so far as to rank these tables purely by prominence -- please drop me a mail if you have any information regarding such.]
Where do these tables cover?
Since realizing the obvious, that political boundaries can be mighty unhelpful in terms of defining groups of hills, I intended to extend the tables to cover all this group of isles that sit adrift just to the west of mainland europe but still on the continental shelf, separated only from the continent by a surfeit of ocean since the retreat of the last ice age.
Thus the Significant Summits tables on these pages include all the isles from the Scillys to the Shetlands, from Kent to Kerry to Kilda, and not forgetting the Isle of Man lurking quietly in the Irish Sea.
The Channel Islands are excluded because they would most certainly be on the far side of the river Seine should the tides recede come the next ice age. Rockall is technically a solitary mid-Atlantic lump, somewhere beyond the continental slope, and the Faroes too are further up the same mid-Atlantic ridge rather than adjoined to Europe proper, though I'd like to see their hills tabled too.
How to measure Prominence?
For a detailed example, see below. For a quick example, take the most commonly used mapping in all these Isles, the 1:50 000 series -- Landranger in GB, and Discover / Discovery in Northern Ireland / the Irish Republic, each produced by their respective Ordnance Surveys.
Pick a summit on any such map at hand. Now all these maps in their current form have contour lines depicting altitude at 10 metre intervals (though the OS(GB) has the regrettable habit of not drawing all contours at this scale when they would be especially tightly packed, presumably on some grounds of readability, yet they often obscure the same area with an overdose of crag symbols instead -- go figure).
Now count how many contours surround your chosen summit and no higher ground . If you count, say, 8 contours, then your top will have a prominence of roughly 80m -- thats 5m above plus 5m below every contour line counted .
The Significant Summits and most other tables that take any account of reascent all rely on slightly more rigorous checking and counting than the method above, but the principle is the same. Tables of summits elsewhere in the world tend to worry less about the odd metre and pretty much keep to counting contours and make one adjustment for the exact summit height.
[ 5: The first list of any
sort of peaks in England and Wales gave a rough account in this way
(albeit in its second edition) -- J. Rooke Corbett, listing the
peaks at least 2500ft high gave not only each summit's height but also
the number of 50ft contour lines that encircled each summit and none
higher. This is quite possibly the first account ever given of
prominence, given as secondary to a summit's height though it
[ 6 For a note of how accurately we can truthfully measure prominence in the UK, see Brimful of Error, but only if you're prepared to be fiddle around with statistics.]
Impractical real-life example of what Prominence measures.
Note, in case I haven't pointed it out elsewhere, that the prominence of the highest point of an island is precisely the height of its summit above sea-level. That much is easy. To make the rest of it easy, we're going to make an island out of Helvellyn in England's Lake District.
Before we do, imagine yourself for a minute on your favourite hilltop and imagine the world flooding and the tide rising higher and higher. Every so often the waters advancing up a pair of valleys will meet at the col lying between the two. At this moment, the land on either side of the valleys is cut into two separate islands. When the waters have barely risen to the point where they just cut off your summit from all higher ground -- in other words, when your summit is now the highest on its island -- then the remaining height difference between the surface of the lapping waters and your hilltop refuge is that hilltop's prominence.
Now, to our specific example:
Just to ram the point of prominence home, continuing from above:
Note that for Lower Man, the 3035ft (925m) peak on Helvellyn's north shoulder, the tide would have to rise to as much as 895m before it was cut off from Helvellyn, giving Lower Man a prominence of barely 30 (thirty) metres, whereas Fairfield to the south, only 2864ft (873m) high, would be cut off from Helvellyn and all other higher ground when the waters had reached the 574m col beneath Helvellyn's Dollywagon Pike -- so Fairfield has a prominence of about 300m, ten times greater than that of Lower Man, yet it is Lower Man gets an honourable mention for being one of the 3000ft'ers outside of Scotland.
That is what Prominence is about -- recognizing each hilltop for its presence and relegating any other small-but-high bumps to being the mere features (interesting or otherwise) that they truly are.
© Alun Fisher 2001-2003