I'd suggest the suddenness of it has a lot to do with output. Notice the flow fell fast along with the temp.
If you change the beginning date from 9-28 to 8-28, you'll see a similar steep decline around Sept. 18-19, which also coincides with a high water period.
My opinion would be this:
- Back in August/early September, the cool water reservoir was completely depleted and the lake rather uniformly warm from top to bottom. Release temps thus stayed relatively stable and high, regardless of where the water was being released from top or bottom.
- In fall, the water temperature ENTERING the lake begins to decrease with the cool nights. Cold water is heavier, so it goes straight to the bottom, while the older warmer water remains on top and takes a long time to cool. A cool fall rain allows more of this to happen more quickly, since more cool water enters.
- During or shortly after the storms, they are releasing lots of water to get the lake level back down. When they do this, water from one gate alone isn't enough. They release both top and bottom water. The bottom water is newly cooled, hence a decrease in temp during the event. After its over and they drop flows, they close the top and let the bottom go, so it cools even more.
- It's worth nothing that the PFBC is stocking trout on the 13th and likely coordinated for cooler water. So now that there's lots of cool, deep water after that big storm, they can utilize cool bottom release to get temps down like they did in the spring, without the concern of "saving it" since more is on the way.
At some point the cool nights cool the topwater too, and once the topwater reaches the temp of the bottom water, the thermocline will disappear and the lake will turn over. You will note turbidity rise when this happens. During winter, the bottom water is warmer than the topwater.
