Our planning for the Chile eclipse in Elqui Valley was extremely helpful, but it wasn’t perfect. Here I’ll discuss what I’ve learned, show you the best photos from this amazing astronomical event, and contemplate what I wish I had done differently.
The Final Images
I chose 8 images in the end, these are a mix between single raw exposures, and HDRs blended from the bracketed exposures. The results are better than I expected, but not exactly as I planned them.
The Camera Settings
I wrote on a previous blog post about my camera settings and preparation for this event. I pretty much stuck to this plan, except that after totality was over, I tried some other combinations on the fly. But before and during totality, I stuck to these settings:
ISO 640 - f/8 - 1/10 sec
1/10 sec being the middle of 7 brackets at ±1 exp, so the final exposures were:
1/80, 1/40, 1/20, 1/10, 1/5, 0.4, 0.8
This setting worked best, the ISO being low meant less noise.
ISO 2500 - f/16 - 1/10 sec
This was with the same exposure times of 1/80, 1/40, 1/20, 1/10, 1/5, 0.4, 0.8. Even though at f/16 the foreground hills were sharper, the extreme amount of noise at ISO 2500 made things blurrier in the end, I much prefer the f/8 versions. I only used this for the 400mm shots, not at 100mm, but I wish I had kept it all to the first setting.
If I could change something here, it would be to have more brackets. I think my middle point was spot on, I would not change that at all, but I wish I had one more stop of light towards each end for extra control. My 5D IV only allows for 7 brackets, but if I could’ve had 9…
Google Earth preview vs. Reality
Google Earth was very accurate, the only major difference was that the exact silhouette of the mountains was off by a bit due to the low resolution. I matched my mockups to my final shots here as a comparison. The position of the sun is off by a tiny bit, which could be critical if I was trying to line it up to something with more precision, but good enough for this shoot.
In my original plan, I wanted to have the sun right in the center of the V of the mountains, but I realized that would make us miss part of the totality, as the sun would move a full sun-width in 2.5 minutes and be partially occluded before totality ended. Because of that, I chose a slightly safer spot in which the sun would keep dropping to the bottom-left even after totality was over. In the end, I much prefer this un-centered composition, my gut was telling me to go centered, but it would’ve been a boring choice.
We scouted locations the day before, and only had one sunset and we happened to be at a different scouting spot for it. Next time it would be better to plan for 2 scouting days at least, something like this:
Day 1: Scout all possible locations, and decide which one is best.
Day 2: Do more scouting, but make sure to be at the pre-selected location at the time of the eclipse to make sure everything lines up exactly right.
Day 3: Eclipse time.
After our single scouting day, I wanted to know how different the sun position would be the day after, during the actual eclipse. Google Earth doesn’t show a precise sun shape to calculate from, so I used the app Sky Guide and took screenshots of different times and overlayed them in Procreate on my mobile phone, all while on the road with no cell data. I found this out:
The sun moves about one sun-width every 2.5 minutes, which was the duration of this eclipse. This image shows a 10 minute period, where I hand-drew some intervals.
The sun’s relative position in the sky shifted only a little bit between each day (here are July 2, 3 and 4).
This simple info helped me visualize things better. It’s good to have a round number and clear idea of how fast things will be moving. Knowing that the sun would be at about the same spot the next day eased my worries too, no surprises.
In hindsight, I should’ve been using The Photographer’s Ephemeris 3D for some of this planning. I only had the 2D version and didn’t consider looking into the 3D one until after someone mentioned it, after the eclipse had passed. I tried it today, and it’s simple and could’ve come in very handy. I still would’ve used Google Earth because it lets me import KMZ files and sync to my custom maps, but TPE 3D is much more accurate with sun position, scale, and shadows.
My Most Important Tool
Having a printout of our scouting locations and the camera settings was critical. During the scouting route, signal was sparse, and the printouts offered all the info that was needed for each location. And as the eclipse approached, and I was doing the last checks on the camera, I couldn’t remember the settings, I was too nervous, but I had this handy sheet in my pocket and it served me well.
Those Who Didn’t Plan
Our location was close to the Varillar cemetery, which is on top of a hill and with plenty of parking space. We knew from our planning that the cemetery would be under the shadow of the nearest hill during the eclipse, only 2 minutes before totality hit. I warned a lot of people who were going to park up there that they needed to hike up the hill behind them or they would miss the eclipse, and they did so, but more and more people kept showing up later, and I could only warn so many. At some point we had to hike to our spot on our own hill and leave the poor souls behind.
As totality approached, and the shadow of the hills draped over the cemetery, we could hear the distant and wretched howls of people realizing their terrible choice. A few lost souls escaped the graveyard, chasing after the edge of the shadow and catching a partially occluded glimpse of the eclipse every time they turned around during their sprint, but most of the unlucky and unprepared mortals clung high to the cemetery walls as if the extra few feet of height would make a difference.
May that image serve as a cautionary tale.