Bivouac.com   Home     Help   Index   SignUp   Login
Inclusion Rules for Mountains #3335
    Date first written: 2014.04.23   Review Date:2021.02.21

Table of Contents
1. Preface
2. Coverage Plans
3. Types of Areas
4. Qualifying Rules
5. Deletion Rules
6. Case Studies
7. Peaks Not to Put in
8. Techniques to Find Peaks
9. Inclusion Rules for "Ridge End" Peaks
10. Short Rules

1. Preface
This is a document you use when you are deliberately working on covering new areas, and charting all the relevant peaks. See discussion Inclusion Rules for Mountains for latest updates.

This is a document that gives rules for deciding how many peaks to put into the database in different types of areas. It could be called the "scope" of the database, or the "coverage rules".

The "method" of doing the prominences for an area is you look at the topo map, and mark in at least one "placeholder" peak every 5 km, as long as they are over some minimum prominence threshold. The prominence threshold depends on the type of area. Up in the Yukon, all we need is placeholders every 5 or 10 km, whereas an area of more interest we would chart peaks as close as 1 km, and as little as P100. Eg: Victoria, Vancouver, Banff

We need guidelines as to what peaks should be added to the database, and in some rare cases, what peaks should be deleted to reduce clutter. For any given peak, there are three possible statuses:

- peaks that should be in the database for sure.
  - peaks that are already in the database but aren't essential
  - peaks that should be deleted, to reduce clutter

The rules vary by area and circumstances. Therefore this document gives a set of "inclusion rules". These inclusion rules are then tested and reviewed by a chapter of case studies.

See also the chapter giving project status, province by province in the Prominence Editors Guide.
 

2. Coverage Plans
A "Coverage Plan" is a plan for all the peaks you should try and include if you are working on an area. Eg: All the peaks over P500, highest alpine peak in any 5 km square, any officially named peak, etc.

You need to define the following:
  Target Peaks (the ones you want for sure)
  Suspect peaks (ones that turn out to not meet the target criteria, but close
  Delete Peaks (Ones that clearly don't meet the criteria_

For any given peak, it is included or not included according to what you are trying to do in that area. This boils down to a set of "inclusion rules" for each type of area. What we want are guidelines so people don't pick some area of low interest and then start putting in every little bump.

I've been trying to define sensible inclusion rules for Bivouac for at least 10 years. Initially, the rule was a simple prominence limit: no peaks less than 300m prominence. That proved unworkable because in many frequently visited areas there are important peaks with lower prominence. For example Ben Lomond only has a prominence of 189m.

I can now see that the rules have to vary according to micro area. In an area surrounding a tricky traverse, it may be desirable to chart unnamed peaks as close as one per kilometer. For example this is what is desirable for the Powell Divide, in order to write good route descriptions.

So what comes out of this is a set of rules depending on the type of "micro area" we are talking about. For any given area, we can first decide what type of area it is, and then come up with a plan. For example, the Powell Divide area is an area where there are several important trip reports, which need peak names for detailed route descriptions. Once you decide what type of micro area you are in, any given peak can qualify in two ways: (1) As placeholders or (2) as high prominence peaks

  POPULATING A NEW AREA
 First define the target. eg: All P300 peaks and largest peak every 5 km alpine area.
 The general method is to pan around a given area and plot all candidate peaks. With current maps and computer displays, it seems you can't be sure you are getting all the P500 peaks unless you plot at least one peak every 5 km, such that when you look at the computer screen for any area, you know you've dealt with that area. I call those "candidate peaks". Candidate are ones that you put into an area that look like they might qualify as "target" peaks but may turn out to fall short.. But after the prominences are plotted it is still useful to leave these "candidate" peaks in the database, just so some other editor doesn't again have to check them.

While working on the Yukon, I discovered I needed 5 km placeholders. Placeholders are just the largest in an area. With 5 km, there will be at least one peak on the map when you have the 200m scale. The 200m scale bar is as far as you can zoom out and still see the detailed contour lines. The 500m maps are useless. You just don't spot them on the old raster map, because the numbers aren't on the contours. So perhaps the most efficient technique is just to set the map to 200 scale, and scroll around and put in the highest thing you see every time there is nothing on the screen.

In really remote areas, where we don't expect any trip reports ever, all we want is to "catalog" high prominence peaks. There may be vast areas on the 1:50K where we don't catalog a single peak. The next level are areas where there may be one or more trip reports. In these areas we want enough peaks such that for any given part of the map, you can see a couple of peaks on the highest points. I call that basic "map coverage". (peaks every 5 km). Next level is an area where there are multiple trip reports. In these areas we want more coverage, perhaps with the highest points every 5 km charted. Next level is guidebook coverage where you want to chart every peak that is typically climbed as a separate objective. This results in peaks every 1 km. In an area where we are documenting a popular traverse, we may have peaks every 1 km, even if low prominence, and not objectives.

One other thing worth mentioning is that in the past we sometimes put in a bunch of insignificant peaks just to work out the prominence of higher peaks. But in low value areas, this is not desirable. We don't want to chart every bump on Vancouver Island, just to handle some officially named hills on the coast. The plan is not to chart every officially named hill, the plan is to chart significant peaks.

In summary, for any given area, one has to decide what coverage category that area calls for. Then you can point at any given peak and say: this one is in the database for "basic coverage", or "this one qualifies because of its prominence". Below are summaries of the coverage spacing and prominence thresholds for each type of area.

3. Types of Areas
TYPES OF AREAS

Remote Alpine Areas: (Obscure areas)
  - all peaks over P500 meters
  - placeholders: all peaks over P200 (General Minimum) and 5 km spacing
  - all officially named peaks (over P50 or 1 km spacing)
  - all peaks with standing name, over P100 and 1 km spacing
  - intermediate peaks

High Use area: (Or trip report area)
  - all peaks over P200
  - all placeholders over P100 and 5 km spacing
  - all officially named peaks
  - all standing names or useful names over P50 and 1 km spacing
  - all names useful to a trip report (along powell divide ridge).

4. Qualifying Rules
Below is a list of how a peak can "qualify": "Qualify" means the peak should be in our database, even with a different name. Typical question when looking at maps, or at other guidebooks is: "Should this peak be in the database? If the peak "qualifies" by any of the categories below, the answer is "Yes". For example, if a peak is a placeholder, it should be in the database.

  1. Major peak
     Any alpine peak above P500, regardless of spacing.

  2. Placeholder Peak
     Any alpine peak more than 5 km from any another.

  3. Greater Peak
     A "Greater" peak is any peak in the prominence walk that is both higher and greater prominence. For example, suppose in an area there was peak "A" of height 3000m. And you wanted to enter peak "B" at the end of the long south ridge of A. Suppose peak B is 2500m and you find the lowest saddle being 2300m. So you might think that you can just enter "B" and be done. But if north of your 2300m saddle, there was another peak elevation 2800m with a key saddle of elevation 2600m, it is a "greater" peak and must be entered. And then you'd have to change the line parent so that B -> C and C -> A.

  4. Intermediate Peak:
     An "intermediate peak" is one that would not qualify on it's own, but should be in the database to properly link some ridge end peak. Eg: If the official was P50, and there was a higher P80 in the prom walk to the nearest P200, it would need to be in the database

  5. Standing name
     A "standing name" peak is one named on maps, guidebooks or other major databases. Eg: In Baldwins Guide, or on BC Basemap or in John Roper's USA peak database). However it must still must be over P50 and 1 km spacing. (We don't want to clutter up the maps because somebody decides to name a bunch of spires or towers. The peak must have some information about it. Eg: In a case where somebody has just named everything on the map, we don't recognize that as a standing name, such peaks would have to qualify for other reasons as well.

  6. Special Interest
     A peak important to a trip report which otherwise wouldn't qualify. Eg: Peaks along a ridge traverse. Still must be over P50 and 1 km spacing.

  7. Labelled Peak
     Any peak with a label should be in our database, even if it has another name. Eg: We need a peak for the thing labelled "Merchant Peak", even though the true Merchant peak is 1 km north. Only exception would be if the labelled peakis closer than 1 km, less than 50m prominence and has no other reason to be included.

When to match names?
 If we previously didn't have the peak, we may as well use the existing name, if suitable. Unless it is a "bad" name: Eg: We wouldn't rush out to propagate some 15 character non-pronouncable name. Or semantically confusing names such as "ridge". Here is a link to some of the name cases. See Name Cases

5. Deletion Rules
There can be two main reasons for deleting a peak:

1. Spacing: It is so close to another peak it messes up the Gmap. (Usually less than 1/3 km away. If this is a peak of interest, such as Mount Fee North Peak, convert the information to a Mountain Note on the main peak.

2. Unnecessary: Too many peaks in area of low interest
 This is the case where the peak may be spaced enough so as not to obscure the Gmap, but is well below what is required to ensure we've got all the P300 peaks. Eg: A peak just 1 km from a much larger peak that only has 40m prominence and in a remote part of the Yukon. I would just delete such a peak if there is no reason for it.

The previous chapter was about coverage rules and inclusion rules. Deletion rules are stricter, because once somebody has done the work of putting in a peak. In other words, in a remote area of low interest, you might never put in a peak only 2 km from another with only 130m prominence. But once it is there, you would probably leave it. However if it only had 40m prominence and was only 1 km away, you might just clean up the database by deleting it.

Glenn Woodsworth and I have deleted a few such obscure and useless peaks. My usual reason is they make the map unreadable. (Peaks closer than 1 km. ) In Glenn's case, they have been remote peaks up the coast that are not of any imaginable interest to anyone. Eg: very low prominence (less than 100m) and very close to some other peak (within a couple of km).

In general, in order to be deleted, a peak would need to exceed all the rules by a factor of two. Eg: If the rule was one peak every 2 km, and 100m, one might consider deleting a peak that was only 1 km from a higher peak, and less than 50m prominence. But in practice, deletions are few.

Cases:

  1. Sky Pilot subpeaks
     There are several low prominence rock towers such as Tombstone tower which were converted to Mountain notes on the Sky Pilot mountain page. This was to make the Gmap readable.

  2. Mount Fee North Peak.
     This is a case of getting rid of an unnamed north peak that is too close to the main peak. The distance from the main peak was less than 1/3 km, although the prominence is 63m. This was a marginal delete because the prominence is more than 50m. It messed up the default map display, but when you zoom in on the t4 display (20m) the peak is distinct.

6. Case Studies
This chapter examines sample clusters of peaks in various areas, to test the reasonableness of the rules outlined earlier. In each area, my goal is to derive an appropriate "spacing" and "Prominence limit". Eg: Peaks every 3 km, or anything over P300.

  1. Too many peaks on a ridge (Mirror Peak)
     Greg Jones put in long chains of peaks, but they are also in David P Jones page 371

      TRIM Jones Biv Latlong
     Height Height Name -------------------------------------------------------
      2723 2723 Riverside 50.463,-115.008 match
      2810 2805 EA72 50.477,-115.021 5 instead of 10
      2845 2845 Cultus 50.486,-115.032 match (Spot height)
      2810 2801 Mirror 50.492,-115.040 1 instead of 10
      2830 2841 Grizzly 50.497,-115.045 Extra 11?
      2810 2809 Spoon 50.504,-115.053 Fudged down 1
      2850 2858 Knife 50.508,-115.057 up 8?
      2930 2930 Platter 50.519,-115.076 match
      2978 2978 Aosta 50.529,-115.095 match
    The saddle heights are all off by 10 or 15m, do not match TRIM. There is a duplicate name FA05.

    But there is a trip report by Allan Schierman for Cultus NW1 (Peak 2841). In Bivouac the trip report track log shows he climbed what we now call EA94 which was 2841, but Glenn corrected it to 2840 but it should be interpolated to 2830.

    Option 1: Only essential peaks + P300
     Delete all low prom peaks except EA94 (Grizzly), Cultus, and Riverside. Then fix all the heights of saddles. Tell Glenn and Greg. As a general principle on such a long chain of peaks, only put in the peaks mentioned in the trip report, and put in a prominence walk. So FA17,
     Only peaks > P300 unless named. Reason: Because it is too hard to maintain such a list of peaks.

    Option 2: Only P200 peaks + peak in report This would delete 3 peaks: mirror and the rogue FA05 and EA72.

    If in future somebody climbs another, then put it in.

     

  2. Powell Divide Area (British Columbia traverse area)
     This is an area where there are detailed trip reports and route reports, and tricky route finding. Therefore, we chart peaks all along the main ridge about every kilometer, even though their prominence is often less than 100m. The peaks are not objectives in themselves, but are important milestones on the traverse. The peaks were assigned two letter codes starting with "PD" (Powell Divide) from north to south:

     Peak Prominence ------------------------------------
     PD2 100m
     Crossroads (PD3) 180m
     Edlinton 159m
     PD4 38m (for spacing, more than 1 km from anything)
     PD5 256m
     PD6 88m
     Danelaw 334m
     Princess 63m
     PD7 118m
     PD8 88m
     PD9 65m
     PD10 48m two km from previous
     PD11 85m
     PD12 38m key corner in route

     All these peaks are valid. The general plan is to name a peak every 1 km along the traverse, as they are all milestones. In some cases as low prominence as 38m (PD12). It is worth mentioning that the existing "guidebook" map of this area made by Western Forest Products also charted the peaks to the same density. They followed the old method of using the height as an identifier, but of course, the flaw in this naming plan is that their map no longer matches the heights on the TRIM maps, so the "PD" numbers are much more convenient.

  3. Washington State
     Target: (peaks above treeline)
      - Every peak 300m prominence
      - Every 200m and 5 km separation
      Marginal:
      - 200 and 2 km
      Delete
      - anything less than 200m prom and 2 km separation. Eg: 107m and 1.5 km

    Working Peaks: Sometimes peaks get put in that look like they meet requirements until you work out the prominence. It's worth keeping these, just so someone isn't tempted to add it in the future.

    Deletions:
      Must be at least 200 prominence and less than 2 km separation I deleted anything with less than 2 km separation less than 150m prominence.

  4. Trafalgar Peak Area
     (a British Columbia guidebook area) See True Blue Mountain
     I investigated a few peaks in the database for British Columbia that were low prominence. In this area True Blue peak was only 43m in prominence, so I was wondering if such a peak should even be in the database. But I soon discovered that this area is close to Kaslo and has trip reports that specifically refer to True Blue and other peaks. At first glance, I thought that perhaps I had gone too far and added too many peaks in this area several years ago, but when I reviewed the situation, I saw that they are all useful for a route description of the trip report Sandra had written. Therefore it is valid to have charted so many unnamed peaks. The derived density was to chart peaks every 2 or 3 km, if over 100m prominence.

     - The "true summit" rule means Josephine Peak had to be entered, in order to provide the parent.
     - Put in spacer peaks, (to bring density up to density of official names, which means peaks every 3 km.

  5. Halifax Hills in Nova Scotia
     There are a bunch of very minor hills charted in Halifax, which were put in to do the prominence walk from an officially named hill. Have a look at:

    Southdale Hill 40m
     Shearwater Hill 20m
     Millar Mountain (official)

    But although "miller Mountain" is an official name, I don't think it is necessary, particularly if it causes us to chart numerous other peaks just to do the prominence walk. These are not mountaineering destinations, they are just areas in Halifax like Little Mountain in Vancouver. They would not be in any guidebook, and there would never be a trip report. They are not alpine.

  6. Colorado
     I examined several areas in Colorado. The general rule for Colorado is to have the officially named peaks, with spacing at least 3 km. And then within that spacing, it is reasonable to put in other higher prominence peaks, if over P300.


      Peak Height Prominence -----------------------------
     Bx35447 3979m, 181m
     Bx35448 3981m 257m

     These have spot heights, and is 3 km away from its parent Eureka Mountain. So it fits in with the average density of official names in the area. Would be useful to tie in Gibson Peak, which is not yet in the database. 38.073129,-105.712466

  7. Colorado (Adams area)
     There are two Adams SE1 peaks put in by Mike Cleven:

      ID Name Prominence ------------------------------------
     #25451 Adams SE1 110m
     #25452 Adams SE1 91m


     These were put in by Mike Cleven in 2005, but now I think they could be removed. The density of peaks in this area is too high and the map is cluttered. Spacing should be 5 km unless over P300. I wouldn't even bother with Little Baldy, (an official name) since it is less than 100m prominence.

  8. Kitlope Meander
     (British Columbia Remote Trip Area)
     The Kitlope area is a remote coastal icecap that has had several trips over the years. It is a huge area, with very few official names. So my question was: what density of peaks should we chart in this area? For example, should the database chart any of the peaks along Dave Williams 100 km trip report called "A Kitlope Meander"? Previous to the report, the area only had P300 documented, and there is not a single milestone peak on their route.

    My first conclusion was that it would be good to name peaks every 10 km along the route. So I charted 3 peaks along their route, and two of them turned out to be P500 peaks which should have been in the database already:


      Prominence ---------------------------
      #35509 Kt2 763m
      Blue Fork 540m
      Kt3 479m
    But then I started looking at the alpine area around this traverse, and came to the conclusion that the coverage should be one peak every 10 km square, as long as they are over 250m prominence. (One peak every 100 square km). This resulted in putting in 20 peaks. I just put them in from the 1:50K map, and then Glenn Woodsworth figured out the proper prominences using TRIM maps. In this process we discovered at least 4 or 5 P500 peaks, so it was certainly worthwhile to do the coverage. 10 km square is a good coverage target, because you can get the whole surrounding area visible on a 1:50K map, and be reasonably certain you've got the highest point in that area.

  9. Mount Marion Area (Goat Range)
     Dan Richardson asked when it is appropriate to delete peaks. He gave three examples of potential deletes in the alpine area around Mount Marion.

      ID Name Prom Nearest Peak -------------------------------------
      13733 Mari-13 161 3 km
      15696 Mari-3 120 2 km
      15739 Mari-7 113 1 km

    The area is west of Duncan Lake. There is one trip report up Meadow Mtn by Sandra, and Dan Richardson has investigated some of the roads. So I'd say this area is a "Potential Local Exploration Area". If I lived close, I'd go and investigate that area. Therefore if I had put in the peaks in this area, I may well have put Mari-13 and Mari-3 because both are more than 2 km from any other. I might not have put in Mari-7. Although it meets the prominence criteria, there is a higher peak within 1 km. Would I delete it? Probably not. If it was less than 100m prominence, I'd be more tempted to delete it.

  10. South Mamquam area
     This is an example of a "Local Exploration Area", where we want peaks every 1 km distance, as long as they have 80m prominence.

  11. Sky Pilot Area
     I removed Ledgling, Nai, and Tombstone tower to reduce map clutter. They are less than 50m prominence, and less than 1/2 km spacing. What I did was put them as "Mountain Notes" on the Sky Pilot record. I categorize them as "rock tower features" on a given peak, not as peaks themselves. Note that they don't come close to qualifying as peaks, as they are only 300m apart.

     Feature Prominence Spacing -------------------------------------
     The Nai 25m 500m from Ledge
     Ledgling 25m 200m from Ledge
     Tombstone Tower 30m 200m from Ledge

  12. Valhallas -Devil's Range
     In this are there are lots of little bumps. Eg; Satan Peak is less than 0.4 km from Diablo, but it's prominence is 88 meters. This is a high interest area, and in Canada. For this area, the rule is 50m prominence, and .5 km. Even with that, they are a mess on any map.

  13. Boston Peak - Horseshoe basin trail
     See Boston Peak
     This is a high interest area, so the minimum cutoff would be peaks at least 1 km apart and 50m prominence. So True Horseshoe would not qualify. Ray Borbon has suggested that the peak at 48.500688,-121.041579 which he calls Shark Fin should be included. But it has very little prominence so I think any info on it should be included withBoston Peak. Similarly, I'm thinking that True Horseshoe should not be in the database, but should be a comment on Horseshoe Peak.

7. Peaks Not to Put in
Although in theory all you need are a list of rules as to which peaks "Qualify", another way to express things is to list what peaks NOT to put in.

  1. Low Prominence peaks near larger peak
     Unnamed peak less than 50m prominence and less than 2 or 3 km from a larger peak

  2. Peaks too close
     In general peaks should be spaced out 5 km apart. However, there are often ridges where there are two peaks only 2 km apart. However such peaks should be reasonably prominent (200m). No point in putting in a bunch of peaks only P50, and less than 5 km apart. (Unless the ridge is a specific traverse, and each little peak is discussed, like in Powell Divide.

  3. Non alpine peaks
     Don't put in little non-alpine peaks like Little Mountain in Vancouver. Reason is because it would just clutter the map and necessitate a bunch more bumps just to do the prominence. A peak like Mont Royal in Montreal (prominence 233m) can be entered, but not some 50m prominence hills.

  4. Peaks less than 300m

8. Techniques to Find Peaks
The technique you use depends on the province, because the maps are different in different provinces. So below are some notes for BC and for the Yukon.

British Columbia
First stage is to find the peaks, then when they are all in, second stage is to link them together. It is often easier to spot peaks on the 1:50K map. Typically, for a remote alpine area in BC, I look at the 1:50 map, and pick what appears to be the highest point every 5 or 10 km or so. I initially let the system generate a name, such as FF17. I often put in the height in feet. This seems easier to find the high points rather than doing it from the Trim. I just look for peaks with at least one or two of the dark 500' major contour lines close by.

Periodically I zoom out to the 250K map, just to make sure I've got coverage of the highest thing in every 10 km square. (Only Alpine squares)


  Title Status: Temporary
  Full Title: Kt20
  Abbrev name: Kt20
  Peak Height: 6550 feet
  Province BC
  Country Canada

Next stage is to refine all the newly inserted peaks on the trim map.

Third stage is I assign names to any P500 that may have been found, and run the naming generator, such that the minor peaks are numbered according to which P500 they belong to.

Yukon Technique
At this time, there are vast areas of the Yukon where we have not charted a single peak. What we would like is to plot at least the highest peak in every 10 km square of alpine area. But it is not easy to pick out the highest peak from the maps. So the technique is to go into each 10 km square, and plot what looks like the highest. That peak becomes the representative for that square. At this point, I'm not doing prominences, but only trying to get the peaks.

Ideally, there would be a single map such that we could display each 10 km area on the screen, and then scan around with our eyes and pick out the highest one. But there is no such simple map. Sometimes you find them with Topo Canada, and sometimes with the 1:50K. I tried to work out a system for finding the peaks.

Often, it is best to start with the 1:50K map, at a 1 km scale, and just investigate likely looking areas. The difficulty with the Topo Canada map is it only shows spot heights when you zoom in to 200m. Furthermore the contour positions seem to vary when you zoom in.

Here are some notes that show myself putting in a typical sequence of peaks:

I start by putting in a peak I called Tulugaq which was 1458m. I observed that 1458m is high for that area, so anything over 1200 is probably worth charting. So I charted 4 more peaks nearby, 1239m 1358m, 1297, 1190. They are about 10 km apart.

Then I started just going into every 10 km square, and looking for the highest. So I found YT1 (which had a spot height), and then YT2 south of there with no spot height.

It's hard to spot the high points until you chart them. Eg: on the 1:50, I could see there was nothing in that area, but when I zoomed in, I saw 1374m. The 1:50 seems to be better than Topo Canada if there is a spot height, because the spot height will be visible even at 1 km scale, whereas the spots don't show up till 200m on the Topo Canada.

So then I found a few more, just by cruising the 1:50K map. Just set scale to 1km, and if there is no peak charted, look for the highest point. Eg: I can see that Bear is 1297, and above 1000m contour.

But then I was cruising Topo Canada at 500m scale, and I see a 1300m. When I zoom in, I suddenly see a 1440 m contour, and just to the west a 1634 spot height!! And then a 1520 m peak. So at this point, I am just cruising, and picking off high peaks.

Then I cruise east, and I see a 1399 spot height on the 1:50K. So I chart that as YT12. At this point, my method of finding the high points is hit and miss. If there was a higher point not marked by spot height, I wouldn't find it on the first pass. The different maps and scales all reveal certain peaks at different scales. The only high points that appear at all scales are the ones I plot. So in some cases there might be 2 or 3 peaks in a 10 km area, as I find the highest.

9. Inclusion Rules for "Ridge End" Peaks
2015: I've written elsewhere about the inclusion rule which says if you've got a given peak, you also need any peak of higher prominence going up the prominence walk. But this chapter raises some other issues which relate more to inclusion. It documents the issues I've run into while doing an area of mountaineering interest such as Nirvana or the Telkwa range around Mount Forster.

I've noticed while doing the Telkwa range that it is desirable to include "ridge end" peaks, even though they are low prominence. Especially if they have a big drop into the valley on one side such that they are a landmark. Eg: the last peak on a long ridge arm.

The advantage of having such peaks is so we can refer to them in trip reports. It makes it easy to refer to the ridge. Another term for these peaks is "outlier" peaks. The rule is that if such a peak is a long way from any other peak then it gets extra "points".

Now, the question is: given that you put in an "outlier", how many of the other bumps on the ridge do you need? I think the old rule was that you should include any peak that is higher prominence than the outlier. Classic case is Denys Peak in the Telkwa range. I wanted some way to refer to the ridge between Denys and Sunset Creeks. The furthest peak out I called Denys, so we'd have a name for the ridge. It could also be the Forster SW ridge, since there is also a long NW ridge.

In any event, I "marked out" the three ridges in the area by putting in the peaks, and in each case, assigned a creek name to the westernmost peak. as Denys, Sunset and Atlantis (Glacis).

It is desirable to "mark out" significant ridges by putting in the peaks on them. A peak is significant if:

- it has high absolute prominence
 - it is a high point of an area (eg: highest thing in 5 km)
 - it is the end of a ridge we want to discuss
 - it has a big drop on at least one side, so as to make it significant looking from the valley below
 - it is a key point on a ridge traverse

SW Ridge
  Denys 118m
  Forster SW3 160m
  Forster W2 240m

NW Ridges
  Sunsets 190m
  Forster NW2 181m

Atlantis 120m
  Forster NW4 272m
  Forster NW2 181m

NE Ridges
  Melissen 60m
  Triangle 120m

Dockrill NE3 169m
  Dockrill NE1 175m
  Dockrill 243m

East Ridge
  Emerson 427m

10. Short Rules
Here is an attempt to summarize: Just the same rules again:

How can a peak qualify?

  1. Any over P500, regardless of spacing
  2. Any highest peak within 5 km
  3. Any over P200 and 2 km from any higher peak.
  4. Any spaced 1 km on a traverse route
  5. Any Officially Named peak
  6. Any Intermediate peak
  7. Any "standing name" peak (in guidebook or map)

  Just because a peak qualifies doesn't necessarily mean we've found all such peaks. Below is the "coverage plan"

  All peaks over P500 in Canada
  At least one peak every 5 km (coverage) in alpine areas of BC
  All peaks over P200 in any area with trip report
  All Officially named peaks (unless closer than 1 km)

Existing Names:
 A class 1 assigned name is less than 8 chars and pronouncable.
 In general, if we don't already have a name, we adopt guidebook or map names
 
  - use guidebook name if less than 8 characters
  - use geographic name (creek, etc)
  - assign provisional name - less than 8 chars, pronouncable, not duplicated within 200 km.

In general, the closer the spacing, the more prominent is the threshold. Eg: a peak that was more than 5 km away but only P100 might be entered, but not a whole chain of P100 peaks only 1 km apart on an obscure, non travelled ridge.