>
If you take out two 3-year clusters in the last decade there is no trend at all.I would love to address each of your statements point by point, but it's been an interminably long week and I don't have the time. But this one I can knock out pretty fast.
I pulled the 1983 - 2017 series data from NIFC and created four different data sets from it, and charted each one. Each chart shows the values in red, the simple linear regression from 1983 to 2017 as a single red line, and a 5-year-running average in blue.
First, the raw data: https://ibb.co/iNQi8U ... there's a negligible change to the running average from 1983 to 1997, but from 1997 to 2009 there is a visually obvious increasing trend. The linear regression shows a 400% increase in acreage burned per year from 1983 to 2017.
Now, there are three different ways someone might "take out" 6 of the last 10 years. I think it's completely absurd to make any kind of argument by just deleting 60% of the data you don't like until the result looks the way you want it to look, but I'll play along. For each of the following charts, I just edited the same 6-highest years from 2007 to 2017.
One way to do this is to simply delete the years from the dataset. You pretend they didn't happen, and see what the overall data looks like then. And here it is: https://ibb.co/gRuzhp ... so here we see that there's still a clear upward trend, even with those years removed. The trend has been reduced to a mere 300% increase from 1983 to 2017, but there it is. You can see the obvious change in the data in the sharp decrease in the 5-year running average line.
A slightly more correct way to handle this is to decide that there's something wrong with the data for those years, and replace those years with approximate average values for the time period. This is occasionally done in climate science by people much smarter than me in cases where it's discovered that, say, a weather station sensor malfunctioned for a period of time. This chart looks like this: https://ibb.co/fvr1v9 ... it's fairly identical to the previous one.
Now, the other way we can do it, the most dishonest way we can do it, is to replace the values for each of those years with "0". As in, for that entire year, we subtract all wildfire activity, and pretend that the year went by with nary a spark. What does it look like then? https://ibb.co/hEWehp Ah! Now we finally get the flat trend we wanted. You can see the really obvious negative forcing on the 5-year running average when those years are left in the dataset but with "0" values.
There are answers to your other questions. The neat thing about news articles on this subject is that most of them can be traced back to some study, expert statement, or research paper. So if you're really interested in getting your questions answered, it shouldn't take more than a few hours' effort.
For example, one of my earlier links, https://theconversation.com/wildfires-in-west-have-gotten-bi..., included a nice graph of federal funds spent on fire suppression. At the bottom of that graph is a link to the NIFC data it was sourced from. So, there's a reasonable answer for "the amount of effort put towards fighting them".
And then you could look at that graph, and then look back at the increase-in-acreage graphs, and imagine what they would look like if federal spending on wildfire suppression were mostly flat, maybe just tracking with inflation.