Hi Capt - well, I took the total anyway, and entered it into good old Excel as a function of time so I could look at a plot, and see this:
Now, there might be something there, and there might not be. I will say that it doesn't look entirely random to me, but you need to be careful.
Before proposing that there is cyclic behavior though, you would like a causal hypothesis for the cyclic behavior. In other words, what would make the fatalities fluctuate in a cyclic manner?
Think of this as a system, with the output being fatalities, and the input being some sort of environmental driver for the system. If you believe it's appetite, then look for data on prey populations over this period. If you believe it's bear numbers themselves, then look for numbers on bear population over this period. If you believe it's due to crowding/habitat loss, then look at numbers for development activity over this period. What you would like to see is that for something that you can believe mechanistically is the system driver, that you would find a statistically significant correlation. If it's news, that would be worth a research publication, by the way.
I have just looked at the total, here, I will look at the larger systems, since it may be that mixing the data from what may operate independently under independent driving forces, essentially confounds the data. Will get back with that.