21 August 2015

Wishful trendline on Tamino wildfire graph?

Stats guru Tamino is like the bizarro superman, the anti-hero, of the statistics world, using his statistical powers to mislead his readers rather than inform. In a 2 1/2 year old post, he attempts to debunk a Washington Post article.

He graphs US wildland fire acreage and adds a curiously extreme red trendline. It would be nice to know if that red line was algorithm generated or custom-manipulated.

Look how close to the top of the data range his trendline gets just before the graph ends at the right. At the year 2012 the data and the trendline are almost equal, but the trendline should stick more to the middle of the data range, like Tamino does for the earlier years of the graph.

Could this be some sort of wishful thinking on Tamino's part, really hoping wildfires will increase, to prove CO2 causes Armageddon? As if a degree or two of air warming can increase wood fires with an ignition temperature of 300C.

(They'll claim warming causes more fire via more drought like the current California one. But NOAA says the oceans, not AGW, caused this drought.)

Next, compare the wildfire acreage with rising CO2, or temperature, and you'll see there's not much relation.

Temp (red line), CO2 (green line)
If CO2 causes more fires why did rising CO2 cause falling wildfires up until the year 1990?

Thinking Tamino's trendline is a bit skewed, I graphed the data and drew freehand a trendline (red and purple line) I believe is fairer and more toward the middle of the data range at both ends of the graph, not just one end. Using only data to 2012 to match Tamino's graph:

When you overlay mine with his you can see his line is much steeper toward the right and really takes off:

Tamino's post was written in 2013 so it is expected that the last two years of data wouldn't be on his graph.

2013 and 2014 data did not pan out as Tamino had so wished; doom-a-geddon once again postponed. Here's with those two years added:

These two more recent years had a downturn in wildfire acreage. With these two years added the trendline (black and white checker) should be even less extreme than Tamino's trendline:

Now if you take just the last few years and put a linear trendline on it, the acreage is actually falling, totally defying Tamino's trendline. Greenie activists foiled again by Gaia!

Tamino objected to the Washington Post article author George Will using wildfire count rather than acreage to show a declining trend. It's true acreage may indeed be a better measure of total wildfire, but even that trend doesn't show any particular relation to either temperature or CO2.

And in any case it's only the wildland area of one country. The global data is limited, but what we do have shows no uptrend:

The drought in America's west is more likely the cause of increased wildfires, and the drought is caused by ocean cycles, not CO2 or global average temperature which hasn't warmed for 18 1/2 years. Worldwide, as well as no fire increase, there's also no drought increase:

The above graph from the Bob Tisdale blog. See also the drought section of the extreme weather page here. That page summarises the utter lack of climate crisis in just about any climate metric you care to mention. Now the lack of global fire increase can be added to the long list of failed AGW predictions.

Also worth a look is the following page regarding precipitation trends around the world. That page is hilariously named Precipitation Patterns due to Global Heating yet when you look at the graphs some go up some go down and some areas stay the same, with no particular trend overall.

One day, a few years from now, the AGW scare will be dead and Tamino and his fellow 'world changers' will have to move on to some other scam to foist on the people at large.

No comments:

Post a Comment