Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Rule of thumb for rating hikes using distance and elevation

1,479 views
Skip to first unread message

Edgar Rawl

unread,
Feb 28, 2001, 9:05:34 AM2/28/01
to
I recall seeing somewhere a rule of thumb that equates (roughly) every
1000' of climbing with so many miles of hiking. Does anyone know what
it is?

Eugene Miya

unread,
Feb 28, 2001, 2:40:33 PM2/28/01
to
In article <3A9C96C4...@earthlink.net>,

"Naismith's rule."

Time sub hike = 3 MPH * mileage + 2K ft/ hr * (# of 1K ft gained).

with variations for 2 MPH and 1K ft.

Then you can balance sides.

Footsie

unread,
Feb 28, 2001, 7:06:06 PM2/28/01
to
In "Rocky Mountain National Park Dayhiker's Guide" by Jerome Malitz,
(which is at my office, so I'm going by memory) He rates hikes for
difficulty with the method you mention. His formula is a thousand
feet elevation gained is equal to 2 miles hiked. So if you gain 2,000
feet in 4 miles on the way to a destination, the difficulty is a 10.
2 points for the gain, and 8 points for the 8 miles walked (4 in and 4
out).

.001 x elevation gain + 2x distance hiked to destinaton = difficulty
(assumes out and back route)

I have tried to elaborate on this by adding a points for trailhead
elevation. I give a point for every thousand feet over 8,000 (all the
trails in Rocky start at 8000 or higher) The rationale is that the
1000 ft of gain from 8000 to 9000 is MUCH easier than the 1000 feet
between 13,000 and 14,000.

I have a spreadsheet where I have applied this formula to trails
listen in a local guidebook, and can now sort by difficulty, as well
as distance, elevation gain, trailhead elevation, etc.

Good luck!

On Wed, 28 Feb 2001 14:05:34 GMT, Edgar Rawl <er...@earthlink.net>
wrote:

>I recall seeing somewhere a rule of thumb that equates (roughly) every
>1000' of climbing with so many miles of hiking. Does anyone know what
>it is?

change "example" to "home" to email

ken...@nojunk.rahul.net-

unread,
Feb 28, 2001, 7:34:55 PM2/28/01
to
In article <8l3r9tc0cguft4jpl...@4ax.com>,

Footsie <mgma...@example.com> wrote:
>In "Rocky Mountain National Park Dayhiker's Guide" by Jerome Malitz,
>(which is at my office, so I'm going by memory) He rates hikes for
>difficulty with the method you mention. His formula is a thousand
>feet elevation gained is equal to 2 miles hiked.

The formula I always used is to add 1 mile for every 1000 feet. If you
can hike 3 miles an hour on flat ground, you'll only do 2 miles an hour
if your trail climbs 500 feet/mile. By your formula, they'd be down to
1.5 mile an hour, which sounds kind of slow.

On the other hand, some people I know hike 2mph no matter how hilly.
They just get less tired on flatter terrain.
--
Ken Lee, http://www.rahul.net/kenton/

SPeacock

unread,
Feb 28, 2001, 9:25:08 PM2/28/01
to
What you really are looking for is the expenditure of energy...right?

Too much of the cerebrals going on here - time to get outdoors. This is
much like taking a GPS, magnetometer, and a cell phone along. Takes some
of the adventure out of it.

HOWEVER...joining in the good fun I pass along the following from some UK
forum about walking. I'd suggest the name of the usegroup, but don't think
they would understand our politics or the discussions about it...nor would
they tolerate it.

The following also gives you a bit of the taste of the level of discussions
on the other side of the pond. Enjoy.

************ quoted verbatum ************* (with a few minor edits)

There are probably many more modern studies but the only information
I have comes from an article in the March 1965 edition of 'The
Climber' by Gordon Waddell titled "The Energy Expenditure of Mountain
Walking".

He came up with the following formula which gives the energy
expenditure for a 10 stone young male over a 24 hour period allegedly
accurate to plus/minus 10%. For different weights (including gear)
adjust pro rata and it probably applies to both females and other
ages as well but as the underlying experiments were carried out on
males between 15 and 25 years of age the cautious scientist was not
prepared to stick his neck out and do more than speculate on its
wider application.

E = 100 (10 + R + 2C + 4H)

Where E = energy expenditure in Kcals/24 hours
R = distance on roads in miles/day
C = distance cross country in miles/day
H = height climbed in 1000s of feet/day

("The equation, however, may not hold for very slow strolls or
cross-country running.")

NB a Kilocalorie is what food fadists mean when they refer to a
'calorie'. If my dictionary is to be believed 'Calorie' (note the
uppercase C) is an acceptable alternative to Kilocalorie but either
way they are obsolete units consigned by the ISO to the same dustbin
as the Imperial units used in the rest of the formula. Those not
fully up to speed with the modern world need to know that the only
acceptable unit for energy in these increasingly intolerant times is
the Joule. :-(

As this formula relates to a 24 hour period perhaps the easiest way
to calculate the excess energy expended is to deduct from the answer
your best estimate of your normal energy expenditure. Waddell had the
following examples which may help:

Occupation Energy Expenditure
(Kcals/day)

Basal (lying in bed) 1500
Sedentary occupation 2500-3000
Heavy manual labour 4000-4500
Most stenuous manual labour 7000

*****************

So what does this mean in real terms? Well basically the assumption is
that
you'll use 1000 calories per day no matter what you do, so 1000 calories
are
added to the result. As for the rest;

1 road mile = 100 calories
1 cross country mile = 200 calories
100 feet of ascent = 40 calories

To take the Snowdon Horseshoe as an example, when I did it I measured
approximately 6.9 miles and 3600 feet of ascent.

So 6.9 x 200 = 1380 and 36x40 = 1440.
1380 + 1440 = 2820 calories for the walk.
Add 1000 calories gives you 3820 calories.

Not the sort of calculation you can easily do in your head as you stroll
along.

So what's this in metric? Well the conversion seems a little more complex
at first, but when you bear in mind that the formula is only rough anyway
(allegedly accurate to plus/minus 10%) it can actually work out simpler.
Here are the horrible numbers;

1 road kilometre = 62 calories
1 cross country kilometre = 124 calories
100 metres = 131 calories

Rather than messing about with those numbers I prefer a simpler point
system. Note that the calories used over 1 cross country kilometre (124)
are virtually the same as 100 metres of ascent (131). For this reason I
like to think of 1km of distance as being equal to 100m of ascent. This is

much more manageable. (I don't bother with road kilometres since I rarely
ever walk on roads.)

Taking the Snowdon Horseshoe example again, the metric figures are roughly
11 kilometres and 1100 metres of ascent.

So 11 km = 11 points and 1100 metres = 11 points.
11 points + 11 points = 22 points.
So I rate the Snowdon Horseshoe as a 22 point walk.

I don't bother converting to calories, I just compare walks using this
points system. As a simple rule of thumb;

A 10 point walk is easy.
A 20 point walk is moderate.
A 30 point walk is difficult.

If you want to convert points to calories it's quite easy, if you split the

difference between 124 calories (for 1km distance) and 131 calories (for
100m ascent) you get 127.5. If you multiply 22 points by 127.5 you get
2805. Add 1000 and you get 3805, only 15 calories different to the 3820
calculated originally (bear in mind the rounding errors due to metric
conversion of the distance and ascent). This works quite well.

If you want to work it out in your head as you stroll along, simply divide
the points by 4, add the result to the original number, add two zeros then
add 1000.

So 22 points divided by 4 = 5.5
5.5 added to 22 = 27.5
Add two zeros to 27.5 = 2750
Add 1000 to 2750 = 3750 calories

The result is 70 calories less than the original result, but pretty close
considering you can do it in your head, and well within the +/-10% accuracy

of the original formula.

******************
I am only interested in the last two parts to Roger's equation -
200KCal per mile of walking and 400KCal per 1000' of ascent.

Now, I reckon that it takes me about twice as long to climb a typical
(discuss) hill as it does to descend. So, according to Naismith, it
should take 2 hours to climb up 2000' whilst walking 3 miles but only
one hour to get down again. (This doesn't seem steep enough for a
typical hill, BTW). Anyway, the energy used going up the hill will
be 600KCal for walking and 800KCal for climbing, equals 1400KCal. To
get down again, we have just the 600KCal for walking. The power
going uphill is therefore 700KCal per hour and going downhill it is
600KCal per hour. In other words, about the same. (Not as daft as
it first appears).

It would be interesting to do ones own analysis on this. Having
worked out the energy expended and the duration of the walk one can
calculate ones own power. Seemingly I walk at 600-700KCal per hour
(Naismith roughly works for me) so a typical 6-8 hour walk in
Munroland will cause me to expend 3600 to 5600KCal. (Plus the
1000KCal standing charge). This doesn't suprise me.

I have heard it said that LDP walkers expend 7000 to 8000KCal per
day. Their days are longer - maybe even 12 hours - and the distance
perhaps 25 miles but only perhaps 1-2000' of ascent. This works out
at 6400-6800KCal and a work rate of 550KCal/hr for a 12-hour walk or
660KCal/hr for a 10 hour walk over the same features.


Roger Rowlett

unread,
Feb 28, 2001, 10:07:14 PM2/28/01
to
Eugene Miya" <eug...@cse.ucsc.edu> wrote:
| Time sub hike = 3 MPH * mileage + 2K ft/ hr * (# of 1K ft gained).

This formula more accurately would be:
Hours = (mileage / 3 MPH) + (feet / 2000 FPH)


Gary S.

unread,
Feb 28, 2001, 10:28:23 PM2/28/01
to
On Wed, 28 Feb 2001 14:05:34 GMT, Edgar Rawl <er...@earthlink.net>
wrote:

>I recall seeing somewhere a rule of thumb that equates (roughly) every


>1000' of climbing with so many miles of hiking. Does anyone know what
>it is?
>

The AMC guidebooks equate 1000' elevation gain to 1 mile horizontal as
a rough rule of thumb.

Moderate downhill will be somewhat easier, but steep downhill can end
up as slow as steep uphill.

And if it applies anywhere, it would be here: YMMV.

Happy trails,
Gary
------------------------------------------------
Chill airs and wintry winds! My ear
Has grown familiar with your song;
I hear it in the opening year, --
I listen and it cheers me long.
--Henry Wadsworth Longfellow
------------------------------------------------
Gary D. Schwartz, Needham, MA, USA
Please reply to: garyDOTschwartzATpoboxDOTcom

Ben

unread,
Mar 1, 2001, 8:10:23 PM3/1/01
to
I just did a 42km hike starting and finishing at 1200m elevation (Katoomba
to Jenolan caves for anyone in Sydney).
Strangely the hike is MUCH harder one way than the other. Without a map
though I can't work out if the rating system would show this up or if it is
subjective...

Ben C.

Roger Rowlett <roger....@americasroof.com> wrote in message
news:3a9db...@nntp2.nac.net...

Pete Hickey

unread,
Mar 1, 2001, 9:11:33 PM3/1/01
to
Roger Rowlett <roger....@americasroof.com> wrote in message
news:3a9db...@nntp2.nac.net...

> This formula more accurately would be:


> Hours = (mileage / 3 MPH) + (feet / 2000 FPH)

In article <3qCn6.55058$o85.3...@news-server.bigpond.net.au>,


Ben <fir...@bigpond.net.au> wrote:
>I just did a 42km hike starting and finishing at 1200m elevation (Katoomba
>to Jenolan caves for anyone in Sydney).
>Strangely the hike is MUCH harder one way than the other. Without a map
>though I can't work out if the rating system would show this up or if it is
>subjective...

Not strange to me. The formula is an approximation. It
is linear. The linarity does not reflect reality. It
approximates it within a range.


--
Pete Hickey | Pe...@mudhead.uottawa.CA
Communication Services | XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
University of Ottawa | XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
Ottawa,Ont. Canada K1N 6N5| Contents of sig censored by management

Patrick Dragon

unread,
Oct 20, 2020, 8:21:38 AM10/20/20
to
Those units don't work
0 new messages