Ok, my followup. There are some further citations I stumbled across while making this list which I need to track down, but these are some nice ones which I was able to obtain.

Much of what I found when I looked into this a while back were studies with poor methodologies and no real dig into the numbers. There's a paper by Sudeshna Mitra titled “Sun glare and road safety: An empirical investigation of intersection crashes" which looked at crashes in Arizona, but had the bizarre premise that peak hour traffic at intersections were the same for all legs in the AM and PM, i.e., you wouldn't have higher traffic east in the morning, and then west in the afternoon. Having spent several summers doing intersection counts that's just wrong. But it's the first paper I came across that looked into the issue.

A better series of papers come from Japan, with the downside that I could only find one that was in English. (Hagita, Kenji and Kenji Mori. “The effect of sun glare on traffic accidents in Chiba Prefecture, Japan.” Asian transport studies 3 (2014): 205-219.) These authors dig more into collisions and glare, focusing on sun position and collision types. They also point out weaknesses in earlier papers, including Mitra. This is a good one to read and is the only one with a focus on actual crash analysis. I wish I could read their other papers as they appear to go into it in more detail, but again, Japanese. :(

Another paper comes from Spain and focuses more on the when and why the sun can be a problem. They came up with the ‘cone’ of interference for sun glare. If you’re under 40 and the sun is within 19 degrees of your focal point the glare will interfere enough with your vision to be a problem. The angle increases with age, and they use 25 degrees for drivers over 60. (Jurado-Piña, Rafael et al. “Methodology to Analyze Sun Glare Related Safety Problems at Highway Tunnel Exits.” Journal of Transportation Engineering-ASCE 136 (2010): 545-553.) Of interest is their solar graph for the year, showing sun position by altitude, allowing you to place your cone in relation to the sun position for the year to see if you’re designing a future road problem. These authors have done other papers with focus on this graph for road design. Well worth a read.

I only have the cites in front of me at the moment, so numbers are from memory.