A century ago, city planners didn't have the restraint that modern planners do, and as a result some older LA neighborhoods have some exceptionally steep streets. Random people on the internet usually claim that some combination of Fargo St, Baxter St (both in Silver Lake) and Eldred St (climbing Mt Washington from Highland Park) are the steepest, with Eldred barely taking the crown. I've never seen any justification of this, and being skeptical of things random people on the internet say, I decided to check.
The basic strategy is the same as the last time, but using real tools. I want to get a detailed-enough DEM, and to sample it along given street paths (from OSM). The SRTM data I used the last time isn't detailed enough: the higher-res SRTM data has cell about 30 meters wide. LA county has a publicly-available dataset with a higher resolution: cells about 3m wide. This should be sufficient.
After downloading the 2GB file, the DEM can be read into QGIS by opening the
dem_10ft/hdr.adf file as a raster layer. Note that it takes QGIS a bit of time
to load the whole thing. The Quick OSM plugin works reasonably well to import
OSM data into QGIS (can be installed through the Plugin manager in the Plugins
directory in QGIS). Once the DEM and the street geometry are loaded, we can
generate the elevation profile. There are several plugins that do this. With
some coaxing, I was able to get the "Profile tool" plugin to do what I needed
here.
Results
I checked Eldred, Fargo and Baxter. The term "steepest street" is poorly-defined, so based on a completely arbitrary measure or "steepest over a reasonable length" we have
- Eldred and Fargo tied for 1st
- Baxter 3rd
The sampled elevation-vs-horizontal travel data is here (everything in ft):
The elevation profiles for the 3 streets in their steepest sections looks like this (Baxter, Fargo starting at the 2 fwy, and going East; source):
We can see that as you travel West from Ave 50, Eldred St loses elevation, and then regains it quickly, getting more steep as it goes. We can also see that while the steepest section of Fargo is a bit steeper than Baxter, Baxter keeps going and has several more slightly-less-steep bumps.
If we plot the steepest sections of all 3 streets on top of one another, we get this (source):
Once again, Fargo is steeper than Baxter. The steepest section of Fargo is very similar to Eldred, but Fargo retains is steepness longer than Eldred does.
One could try to split hairs about whether Fargo or Eldred are steeper, but based on this data, they're even for all purposes.
Clearly this all is only valid if the source data is valid. It looks mostly reasonable. The DEM is detailed enough to show each street in a little valley, and this aligns well with Eldred. Fargo and Baxter lie in an area with a natural valley, but the street alignment ignores it cutting at an angle, which is consistent with the USGS topo. The absolute heights are fairly consistent with the USGS topo. I believe these results.
It's also possible that there's some other candidate steep street out there. Some random internet person claimed 28th Street in San Pedro had a steep section, but none was comparable to these 3. This is good enough for me and I'm stopping here.