The curved lines going from lower left to upper right are the RH.
The vertical scale is moisture content in grains.
Horizontal scale is temperature in F.
Now here's an easy example. Let's say your air is 80 degrees and 60%RH. If you plot the 80 degrees by going to the right on the bottom of the chart to 80 and follow that line up, find where it intersects with the 60%RH curved line. Mark that point. Now if you go straight left or right you can see what the RH is at various temperatures.
At 75 degrees you will have 70%RH. At 100 degrees it would be 32%RH.
When you measure slab RH, you are measuring the relative humidity in the air in the bottom of the hole. While this sounds like a slam dunk for calculating the RH at any given temp if you know a starting temp, unfortunately it doesn't quite work that way.
I have tested slabs for months as they go through temperature swings and my numbers did not match the chart. Some of this may have to do with hysteresis, some may have to do with moisture suppression, or maybe it's something with moisture releasing from the concrete and reabsorption, I haven't studied the issue yet. Howard Kanare was rumored to be working on it so I'm sure not going to...
What I am getting at is you can't extrapolate that a slab at 80 degrees and 60%RH will read 90%RH at 68 degrees. While it works well in the atmosphere, it hasn't been working in the test hole.... This is also why it is important to take your readings under service conditions! If you say a slab is fine to put flooring on when it is at 80 degrees and 60%RH, then the building is cooled and running at 68 degree slab temp, you can bet your RH will NOT be at 60% in that slab. If you have a failure and someone tests after you at the lower temperature, they will get a higher reading.
I test at service conditions and use extrapolation very carefully. I'm sure Jason can jump in here and explain the relation between temp and humidity in the space read by the probes, and tell us if we have a method to extrapolate to different temperatures accurately yet.