The cache is not at the posted coordinates. This puzzle is best solved at a computer. You can probably manage it on a phone, but it may be more tedious.
Background
The parking structure at the Marymoor Village Station has room for a whopping 1403 vehicles. If it gets full and you have to wander all the way to the top, you may notice some interesting parking shade on the top floor. You can also see them from 520 if you chose to sit in traffic instead of utilizing this great new transit option. These solar canopies do double duty - they block sunlight in order to keep the cars below them cool (ok, this is Seattle - they form a pathway and protect you from the rain), but also utilize that sunlight to produce power.
The panels at this station are bifacial, meaning they can generate power from sunlight hitting either side of the panel. Wait a minute, the other side of the panel is facing away from the sun! Well, that's true, but it still receives a little reflected light. But isn't it more expensive? It turns out that most of the costs here are driven by the labor costs of installation. Even among materials, the panels play a small role. You need to buy all that framing, wiring, and some inverters to convert the DC current to AC. So even if the 'dark side' is only going to generate 10% extra power, sometimes it's worth it. Additionally, you can influence how much extra power you get by changing the substrate under the panels. In this case the panels are over a bright concrete that has pretty high reflectivity. The amount of light reflected by a surface is called its albedo.
The Puzzle
Your job is to find out just how much of a difference the back-side of the panel is going to make. Let's see it in action! We're going to use some nifty tools from the folks over at the National Renewable Energy Laboratory (NREL) to make it easy. Specifically, we're going to use an older, fairly simple model called PVWATTS since the results are less likely to change over time (and so I don't have to constantly update the puzzle).
https://pvwatts.nrel.gov
The simulation requires 2 components - the location and the system specs. It will find the solar resource data on its own - just give it the posted coordinates in a form it will understand: 47.6678, -122.1072. Put that in the 'Get Started' box. Confirm on the next page that it is using solar resource data for 'Lat, Lng: 47.65, -122.1' and hit the right arrow to go to the system specs.
For the system specs, I've come up with some estimates for an 8-panel canopy based on what I saw on site.
When I looked at the installation, I found Hanwha Q.PEAK Duo XL-G10.3 BFG 480W panels. You can read the spec sheet if you are interested, but it's not necessary for this puzzle. 8 panels * 480 Watts = 3.84 kW. The default system losses are a bit out of date; let's use a more optimistic 11.5. The panels are pretty close to flat here, so we'll use an 8 degree tilt. Finally, they aren't facing due south so let's use 160 degree azimuth (0 is North, 180 is due South).
| DC System Size (kW): |
3.84 |
| Module Type: |
Standard |
| Array Type: |
Fixed (open rack) |
| System Losses (%): |
11.5 |
| Tilt (deg): |
8 |
| Azimuth (deg): |
160 |
For now, skip the advanced parameters and go to the results. Since the station opened in May, let's use the May AC energy generated (rightmost column). That will give you the decimal minutes for the longitude.
Now let's go back and see how making the panels bifacial impacts the results. Be sure to use the orange arrow and not the browser navigation. Expand the advanced parameters and change Bifacial from 'no' to 'yes'. Finally, the albedo of the concrete will differ from satellite-derived albedo, so let's specify 0.68. A careful observer will note that we're making a signfiicant simplifying assumption here - that there are no cars parked on the top floor. Who knows what their albedo would be...
Go back to the results and look at the energy generated in May. How many more kilowatt hours were generated as a result of using bifacial panels? Use that number as the latitude decimal minutes.
N 47° 40.{bifacial benefit}′ W 122° 06.{without bifacial}′
Congratulations! You've learned a bit about PV panels and solar simulation and hopefully now have the coordinates for the cache!
If you've done all this and you're confident you did it correctly but you can't make the checker happy, shoot me a message. I'm counting on this tool's defaults and weather data source remaining stable - if NREL updates them I'll have to tweak the puzzle.