Take a look at 40802 (c)(2)(B)(i)(II):
If an engineering and traffic survey was conducted more than seven years prior to the date of the alleged violation, and a registered engineer evaluates the section of the highway and determines that no significant changes in roadway or traffic conditions have occurred, including, but not limited to, changes in adjoining property or land use, roadway width, or traffic volume, 10 years.
According to the law, it's up to the engineer to decide whether or not the changes are significant. If so, then a new survey is warranted. In this survey, the engineer did note that a change was present, but thought that it wasn't significant. While the document claims that the policy requires a new speed survey, the law in 40802 doesn't.
Since the survey indicates other safety factors here, it's going to be hard to argue that the 5mph reduction isn't warranted. The interesting thing about these 85th percentile numbers is that they're all over the map. I'm not sure this helps the southbound drivers, but at least on the northbound side it looks like a higher speed limit might be warranted in a few sections. That said, VC 22361 authorizes separate speed limits on multiple lane highways with two or more separate roadways. If this highway doesn't qualify, then the speed limit must be the same on both sides. If that's true, then I'd get the raw data for these calculations and determine whether the aggregate 85th percentile speed for the sections in question is calculated correctly.