Right -- I'm guessing that the 18% figure for Humboldt coincides pretty exactly with the number of <ahem> "undocumented agricultural workers."
On the "discouraged workers" thing, the problem is that there's no number that is relevant. If you quit working for a job, you're either homeless and eating at soup kitchens or dead; or (more likely in the U.S.) you've found a different set of accommodations -- your spouse still works, and you stay at home with the kids; you retire early; you move in with your sister and help around the house; you're just out of college and you do the "boomerang" twenty-something bum living with the 'rents with no rent thing. I know people, more or less, in all those categories, and I wouldn't really call them "unemployed" per se.
Like so many things, the term "unemployed" does not describe a binary categorization (damn, that old trope rears its head again), so the raw number isn't very meaningful in and of itself. The month-to-month changes matter, however, once you impose some artificial criteria to make the categories binary.
I don't trust the news or the government, but I trust their economic news and figures more than any other areas. If the Wall Street Journal starts giving bad economic and business news, investors will get very angry. You or I might not know, but somebody will, somebody that has a lot more money than you or I and somebody who cares more.