Hacker News new | past | comments | ask | show | jobs | submit login
Show HN: High Resolution Terrain Maps of Cascade Volcanoes (ramblemaps.com)
48 points by mparr4 27 days ago | hide | past | web | favorite | 40 comments

By 2023, the USGS plans to have taken high-resolution LiDAR over the entire continental U.S., as well as creating 1-meter raster elevation products. Maps like these will be easier to create in the future.

Progress: https://www.usgs.gov/media/images/1-meter-dem-availability-c...

A fellow HNer and I started making high resolution terrain maps using LiDAR data.

We've got maps of the Cascades and my home state of Vermont.

We're software guys. We've got a custom website but we're using Shopify for our checkout. We'd love feedback on the maps, the website, or anything at all!

Wouldn't these more properly be labeled "relief" maps vs "elevation"?

Regardless of the terminology, they look stunning. Definitely appeal to the map geek in me.

Yes, that would absolutely be the proper name. For us I think it was just a matter of familiarity, does the lay person know what a "relief" map is?

It's more descriptive, but potentially at a cost of understanding.Though maybe we should be targeting the demographic who knows what a "relief" map is though... Unsure, just making this up as we go along!

Love Vermont. Makes me want to head up to Sunset Ridge for a hike ;)

By coincidence, I happened to be researching algorithms for generating 3D meshes from height maps for procedural terrain generation. While it looks like the classical Delaunay trianglization remains the best strategy. I wonder if the high fidelity of next gen sensor data will uncover new techniques. Most PCG techniques employ pretty noisy maps and never quite look "real".

Fast Polygonal Approximation of Terrains And Height Fields (1995)


Ha. I'll see you up on Sunset Ridge!

Very cool stuff. I'll give that paper a read.

These look great and I'm sure my display isn't even doing justice to them. I imagine you are producing these on-demand, so I wonder if it's possible to allow customers choices like lighting direction and ink color. The Mt. Shasta one for example is illuminated from the north which is interesting but not realistic.

Thanks, Jeff.

There's a weird trick that our brains play on us when the light comes from somewhere other than the north, the ridges and valleys look inverted.

There's been some research on this and the consensus is that sunlight coming from the northwest is best for making the terrain appear "natural".

If you rotate the map 180, you'll see what I mean. The deep gully coming down Shastina looks like a ridge, for example.

While we are printing these on-off, there is a fair bit of work involved in generating the source images: generating hillshade layers, photoshopping the results, etc, so I'm afraid we can't do custom lighting and colors right now.

Interesting. It seems like metastable illusions would be a problem for any lighting direction when the object is cone-shaped.

The maps are really cool. How are you guys managing to find customers given that the product is fairly niche?

Thanks, glad you like them.

We're figuring that out right now. We've got a FB ad campaign going and have gotten a few sales, but this line just went live on Wednesday. Our original maps (released at the end of March) were all of Vermont ski mountains and with ski season cut short by the pandemic, we think we missed the boat timing wise.

Our main priorities right now are expanding our geographic reach so that more people can find something that they like. We've got a few more series in the works that we should have up within the month that will greatly expand our offerings.

Then... FB ads, content marketing, target blogs and relevant traditional media? Maybe try and get prints in visitors centers, lodges, hotels, etc?

I would try REI and wta.org and follow the trail from there.

are you using headless?

if so, how is that going and was it worth the work?

We are not, we've got a full-on shopify site that we put on a subdomain, update the cart via the storefront API, and our checkout link takes you to the shopify checkout.

Headless, AFAIK, is a higher pricing tier. We're just sorta hacking the standard plan.

"Hello, My name is Misha. I am a map geek."

But what is a map geek?

A personal aesthetic where the visual, geometric and topological apects conspire to induce a overpowering feeling for the beauty of spatial possibility.

A spatial imagination, which can look at a map, but see the landscape. The symbolic abstractions vanish as a transparent veil over the underlying truth. A skill and a motivation also correlated with mathematics, physics, (computer) graphics, typography, chess, classical music, art and sculpture.

A specific intuitive understanding of topography, based on what it would look like on the ground, with nuances of intervisibility, and the operational task of planning of hiking routes. Perhaps also interwoven with knowledge of local geology, flora and fauna. Ultimately enabling navigating without maps, by the familiarity of the trail, the interlocking of landmark and timeline, and the experience of light and weather by the seasons. Striding out to the summit and rolling down to beat the sunset.

Can there be a map geek who is not also a hiker?

The mapgeekiness of the (probably male) introverted techie personality. Not just viewing and relating to the beauty, but a desire to know, capture and own, all of it. The collector of the specimens, the librarian of the catalog, the unconsulted oracle preoccupied with imaginings of unspoken theories.

Is this data available in raw format? One of my hobbies is foraging for plants + mushrooms, and I'm often struggling to find good enough topographic data. If this dataset is available, I'd appreciate some pointers!

Also: Are there any developer-friendly datasets for climate, terrain, and other IRL data? I've spent a lot of time curating stuff from NOAA and Data.gov to fit me needs. Huge time sink! If this doesn't exist then somebody should do this ;)

A kindred spirit! I also am a forager (from VT).

The "National Map" has a lot of the available high resolution elevation data (derived from LiDAR), but not all of it. Much of this data is collected by the states, so often there is more data on a state's geoportal. If you tell me what state you're interested in I can send you a link if I've got it.

For the national data: https://viewer.nationalmap.gov/basic/

Also have a look at the ASTER GDEM data for elevation and water body classification: https://asterweb.jpl.nasa.gov/gdem.asp. I have managed to successfully use this in one of my own projects, but it did take some processing to get it into a different data format for my purposes. Although this data has only about a 30-metre resolution, all the raw data does seem to be available for the vast majority of the globe (between 83 south and 83 north latitude).

ALOS seems pretty good to me (I'm a GIS noob though). They have free 30m and expensive higher res https://www.eorc.jaxa.jp/ALOS/en/aw3d30/

Are these 2D prints, or 3D relief prints with raised surfaces for the peaks? At first I thought they were 3D but the photos of the maps at an angle seems to suggest they're 2D.

They are 2D prints that are printed on materials and using inks that give them quite a bit of dimensionality. The acrylic option even catches the light differently as you move around it, creating a pretty cool effect.

You can get 3d mountain prints from TinyMtn - https://www.shapeways.com/shops/TinyMtn

If you could print on the maps the climbing routes from https://www.avenzamaps.com/maps/731848/mount-rainier-climbin... that would be amazing.

We've talked about making maps for climbers, it's certainly something we'd like to do. The more we here from folks who'd like that sort of thing the higher up the priority stack it goes!

Omg, that’s like a trophy on the wall - look where I went in 2020! That red line here. Better than a moose head on the wall.

Hey! Nice maps.

How did you go about your land cover masking / color assignment? I assume you're just using lidar here, but land cover retrieval can be tricky. Maybe by downscaling a coarser land cover datasource?

I ask because I work on this (segmentation, land cover) and just wonder what folks are up to!

We're just consumers of the land cover data for our Green Mountain Ski Map series. The details of how they generated the data are in this report: https://vcgi.nyc3.digitaloceanspaces.com/portal-pages/VT-Lan...

I've been banging my head against the wall trying to similarly classify other states and haven't had much luck yet. I was initially trying to do it just using orthoimagery, but I'm thinking pairing it with the LAS files will be helpful.

Got any resources you like for downsampling or other techniques?

> Mount Rainier, originally known as Tahoma, is the tallest volcano in the Cascade Range and the 2nd tallest mountain in the continental United States at 14,411 feet.

Yeah, it's not the 2nd tallest mountain in the continental US. There are a bunch of mountains in Colorado that are taller and the tallest one is in California

  Mount Whitney 14,494 feet
  Mount Elbert 14,433 feet
  Mount Massive 14,421 feet 
  Mount Harvard of Colorado 14,420 feet
  Mount Rainier 14,410 feet

You're absolutely correct. That's what I get for going off memory. Updating now.

just doing my job (ie correcting random people on the internet) :))

you are right, but some of us(climbers) don't count those due to Prominence issues. :) Otherwise tall ridgelines could have dozens of points littering up the lists.

I'm always amazed at how many different ways we can count mountains.

By prominence it is the second tallest in the continental US.

Very cool. I climbed a couple of those bad boys so I'd love to have a few of these.

Nice! Which ones?

Adams and Shasta for me. Am hoping to get back to the PNW for volcano season next year to do some ski mountaineering. I love these mountains.

Baker and Rainier. I'd love to do more. Shasta is high on the list for sure. I've done a fair amount of splitboard touring on the mountains up here but never on the big volcanos.

I've done Baker ski resort backcountry, but that's actually on Mt Shuksan (despite the name). Shuksan is not a volcano but widely considered one of the most beautiful mtns in the PNW.

Nice. I've been to the Baker resort, but didn't venture into the backcountry.

You gotta get on some volcanoes with the split next May!

I'd have to agree about Shuksan, a real stunner of a mountain!

nice maps and great idea.

Just shared to some hiking forums i use, hopefully you get some sales.

Thanks, glad you like them and I appreciate that very much!

Any of these for the moon or other planets?

Not yet! I looked into Olympus Mons but couldn't find data at high enough resolution that I'd be happy printing at. Definitely interested in extra-terrestrial maps!

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact