Rumney Traffic Map

Hey Rumney Climbers,
I've been working on collecting data for this thing since early March, and it is finally done. The map should be pretty self explanatory, but basically it predicts traffic at all the crags in Rumney (sans buffalo pit and gem hunter, which I have the data for but they're too far away to fit). Tell me what you think, it initially was a small project for school but I blew it way out of proportion and just went all the way with it.
If you want the ArcGIS data or Excel I can give you that as well.

Viewable and downloadable Here (google docs)

Available for download here too - document.li/y4Dl (non-google for you paranoid folks)

Edit; I realized I can post it here too:

Map

Looks good. Its what I expected to see for the results. Not to many surprises.

yeah, can you share the details, i'm interested.

Nice Eli!

thanks for sharing... how'd you collect data? ArcGIS data wld be cool too. Do you have any GIS maps of NH with LiDAR /bedrock geo data?

I'd also be curious to hear methodology. Did you just sit here with a clicker for an afternoon?

If you still have the raw data, I'd also be curious to see a different cut of this. It looks like your coloration is (I'm assuming?) on a sheer headcount basis, but it would be neat to see it as a percentage of available routes. As an example, Below New Wave has a middling yellow on your map, but there are only 6 routes there: it's rare that I walk by and don't see people on every single one of them.

I'd also be curious to hear methodology. Did you just sit here with a clicker for an afternoon?
it would be neat to see it as a percentage of available routes.

Very cool map. I'm also interested in your methodology.

Regarding the suggestion of percentages--that is built in through the proportional size of the circles. It would be possible to relate this data as a percentage, but that would probably clutter the map and make it somewhat harder to read.

Pretty cool - expected for the most part (Waimea, MC, Bonsai, OC)

Scott Phil wrote:Regarding the suggestion of percentages--that is built in through the proportional size of the circles.

I think we're talking past one another. Here's what I'm saying:

My interpretation of this (Eli, stop me if I'm wrong here) is that colors are assigned on an absolute basis. So when Below New Wave is the same color as (say) the Hinterlands, it means that the two see roughly the same number of climbers. But the Hinterlands has a great many routes, so the actual impact of this many people is lessened.

Essentially what I'm getting at is if you're looking to ask "will I have a good time?", the metric you really care about is perceived crowding--climbers divided by routes--rather than a total headcount.

Lavery wrote: I think we're talking past one another. Here's what I'm saying: My interpretation of this (Eli, stop me if I'm wrong here) is that colors are assigned on an absolute basis. So when Below New Wave is the same color as (say) the Hinterlands, it means that the two see roughly the same number of climbers. But the Hinterlands has a great many routes, so the actual impact of this many people is lessened. Essentially what I'm getting at is if you're looking to ask "will I have a good time?", the metric you really care about is perceived crowding--climbers divided by routes--rather than a total headcount.

Yeah, a climbers vs available routes heat chart might be more meaningful.

Of course now that I say that, if a wall has 7 5.13+ routes and 2 5.9s, availability might be relative since 5.13 is unavailable to the majority of climbers.

Lavery wrote: I think we're talking past one another. Here's what I'm saying: My interpretation of this (Eli, stop me if I'm wrong here) is that colors are assigned on an absolute basis. So when Below New Wave is the same color as (say) the Hinterlands, it means that the two see roughly the same number of climbers. But the Hinterlands has a great many routes, so the actual impact of this many people is lessened. Essentially what I'm getting at is if you're looking to ask "will I have a good time?", the metric you really care about is perceived crowding--climbers divided by routes--rather than a total headcount.

Good points--and I totally agree with the important question of "will I have a good time?"

The challenge is how to clearly convey this information graphically. For example, it might help if the traffic data were portrayed as actual numbers rather than a range of low to high. In theory, you could also depict the traffic colors in relationship to the number of climbs though depending on how it was done that would potentially make the map harder to read. Even then, it would depend on the popularity of individual routes for each area. For example, an area with low traffic may still have a line at the base of a popular route.

In the interest of full disclosure this is all academic for me as I've not yet climbed at Rumney, though I used to make maps for a living.

looks like it's a map of "number of routes per crag." It might be better named that. Or perhaps "route density map", but that's not quite the right term, because a large crag will have more routes. Traffic map might be a little misleading, because that implies the number of people on a given day or hour. This map shows number of route developers over the decades.

I guess you make the assumption that # routes predicts # of people at any given time

it's a beautiful map. pretty cool. love gis

Last night I wrote a long post explaining everything, hit submit - and mountainproject timed out. Here it goes again, I'll definitely hit copy before I hit submit this time around.

I guess I'll start by explaining how the map was made.

Methodology:
In early March I installed the ArcGIS Collector App on my phone, and began the process of making this map. I started by identifying criteria that make different crags have high or low traffic (which turns out to be a very subjective thing). The criteria I identified were: Rock Quality, Number of Routes, Past Popularity, and Hike time. I then determined that the best way to rank crags would be to create a point system for each category, and average the total points each crag scored in each category. Each category was ranked 1-5, 5 being the highest score and 1 being the lowest, we'll get into specifics in a minute.

Rock Quality: Totally subjective, but pretty easy to see that places like Waimea would score a 5 (pretty much perfect rock) and places like Left of the Venus Wall would score a 1 (Pretty much the shittiest rock/lichen and moss covered/chossy/sharp as hell/things that make it not enjoyable). It is a subjective measure; a great example of this is Below the New Wave - I think that the rock quality there is not superb, but not bad at all. I gave it a 4. (I'll explain why I chose Below the New Wave later)

Past Popularity: Again, relatively subjective but with a pretty decent consensus. Past popularity is scored with a 5 being the most popular, and 1 being the crags that no humans climb at. The Parking Lot Wall and The Meadows are big standouts here, they're always packed (they both scored 5s). Conversely the Infinity Wall probably has only seen me as a visitor this year, so it scored a 1. To continue with Below the New Wave, it scored a 4 in this category, because I do very often see climbers there, but it is by no means a destination crag.

Hike Time: Everyone knows that sport climbers hate hiking. If you argue with me I'll present you with the Prudential. It is THE destination crag in the West Crags, but you almost never see people there. For Hike Time it scored a 1 because it takes quite a long time to get to whether you go through the Black Jacks or the Hinterlands. The opposite crags are things like The Meadows, Parking Lot Wall, or 5.8 Crag (Which take no time or effort to get to). Here I gave Below the New Wave a 4 again because it does require a small amount of effort, but isn't bad at all.

Number of Routes: Here is by far the most confusing category but also the most objective, because of the way I split it up. I did this by trying to find natural breaks in the data, which led to the following breakdown(from memory I can't find my actual note about this one):

1 = 1 -7 Routes
2 = 8 - 14 Routes
3 = 15 - 20 Routes
4 = 21 - 28 Routes
5 = 29 - 54 Routes

As you can see from the break down, this is what adds a some of confusion to the map - Below the New Wave only has 6 routes, so even though it scored a 4 in all the other categories the 1 in this category takes down the average significantly, putting it on par with some definitely less popular crags. Because of this I would say that my map may be better at estimating the number of parties per crag as opposed to the number of likely available routes.

I went out and geo-tagged all 42 Crags with the Collector App over the span of a couple months, and then spent a couple sessions with the data making the map. Things that I left off the map entirely are crags like The Monolith, Hail Vader, and Hohe Hinterland (theres a few more I think that would be silly to put on - see Utopia Ledges). I did wind up going out to The Buffalo Pit, Gem Hunter, and a couple other totally ridiculous ones (I found a lot of relatively untouched rock/boulders in the process). Obviously this is imperfect, but I think it is really interesting and I'd love to do it for other crags, perhaps to get involved in planning erosion control and trail building.

Areas where I'm probably very wrong:
Crags like Monsters from the Id are incredibly popular, but only for the 1% of us that crush 5.13 (not me). Because of things like this, my past popularity numbers are likely inflated in places (see monster getting a 3 in past.pop) and deflated in other places (see the Northwest Passage getting a 1). As with anything that is subjective it has the power to become incredibly inaccurate, especially if the project is left to the hands of one person (in this case myself). As far as actual traffic; I never did a head count, I've just been climbing at Rumney for years and basically went off of those observations.

Notes about the map:
- The symbols are sized porportionally to the number of routes per crag (1-54) in an attempt to make the symbol about the size of the actual crag when it is overlaid on the imagery basemap. That was the only reason I did it that way (that and I wanted the bigger crags to stand out).
- The numbers on the symbols are a reference only to the Crag Key, so that you can tell what you're looking at.

I made the map because I don't like climbing in crowded areas very often, and I wanted to have a good time. It was a lot of work, but I learned a lot in the process, did a cool project that I care about, and hopefully shared some valuable information with the community. The data is a mess, but I still have all of it and am willing to share.

Lavery wrote: I think we're talking past one another. Here's what I'm saying: My interpretation of this (Eli, stop me if I'm wrong here) is that colors are assigned on an absolute basis. So when Below New Wave is the same color as (say) the Hinterlands, it means that the two see roughly the same number of climbers. But the Hinterlands has a great many routes, so the actual impact of this many people is lessened. Essentially what I'm getting at is if you're looking to ask "will I have a good time?", the metric you really care about is perceived crowding--climbers divided by routes--rather than a total headcount.

You're correct.

Here is my data, it may be a mess to repair the sources for your copy of arcgis, but I wish you luck either way.

Only 14mb

Great idea! Last year the USFS was asking people to complete a survey about where they went and climbed, how many in party, how many routes and how long, etc, so a form of the usage data is out there.

Eric LaRoche wrote:Of course now that I say that, if a wall has 7 5.13+ routes and 2 5.9s, availability might be relative since 5.13 is unavailable to the majority of climbers.

This is the most important thing - route availability for routes that you actually care to do. It's better to give raw data and allow individual extrapolation, IMO

I assumed you used the MP data on the number of people in, say, the past year who ticked routes. Wouldn't that be the best practical way to figure out how much traffic an area gets? It seems that factoring in things like rock quality doesn't tell much about how many people are actually climbing there.

Jon Frisby wrote: This is the most important thing - route availability for routes that you actually care to do. It's better to give raw data and allow individual extrapolation, IMO

It is a little difficult to not be an individual when you are the only one working on something, and the raw data is totally subjective anyway. It should be noted that Rumney attracts a lot of hard sport crushers, so it may not be that invalid.

Bill Shubert wrote:I assumed you used the MP data on the number of people in, say, the past year who ticked routes. Wouldn't that be the best practical way to figure out how much traffic an area gets? It seems that factoring in things like rock quality doesn't tell much about how many people are actually climbing there.

I felt like rock quality was a valid critera, because it is certainly a deterrent for some people. Also it isn't necessarily just rock quality that I was looking at, but more rock/route quality (number of stars and if I've pulled blocks/rocks off climbs at different crags)

I do appreciate the feed back, and would love to improve my map at some point.

The end goal is hopefully that different crag-maintenance organizations can more easily plan for trails and trail upkeep. Basically the idea is that climbing is growing, and things are going to have to be maintained - I believe that major crags are going to have to undergo some planning process, and that resources like this could prove to be really helpful.

JohnnyG wrote:looks like it's a map of "number of routes per crag." It might be better named that. Or perhaps "route density map", but that's not quite the right term, because a large crag will have more routes. Traffic map might be a little misleading, because that implies the number of people on a given day or hour. This map shows number of route developers over the decades. I guess you make the assumption that # routes predicts # of people at any given time it's a beautiful map. pretty cool. love gis

Getting back to this - The traffic prediction comes from other data that I put into it, along with the number of routes. The symbol size changes mostly just to reflect the size of the area we're dealing with. A number of routes per crag map would have been a hell of a lot more simple to make. The actual data appears in the color spectrum - but the number of routes has an influence on that as well, which is why you see the correlation so clearly.