Maybe under their collars.
Turns out that it wasn’t so hot just a few kilometres South of the city. Not when one eyeballs a plot of daily minimum and maximum temperatures over the past 15 years collected at the agricultural research station in Medina. WA’s Department of Agriculture operates a number of automated weather stations. The one in Medina has been operating for 15 years, albeit with some data collection problems in the early days.
A correspondent on Warwick Hughes’ blog may be able to explain the “warming”:
Philip Bradley Says:
January 29th, 2012 at 4:56 pm
The East Perth location of the official Perth site before 2004 is across the road (about 20 meters) from a large grassed park that is irrigated every night in summer. The 2004 move to an unirrigated location at Mt Lawley resulted in an approx 1.5C fall in nighttime temperatures.
Nighttime irrigation causes a near ground greenhouse warming effect.
The new Mt Lawley location is in middle of field of bare earth (sand), which will reflect more heat than grass, heating up the Stevenson screen nicely during the day.
So BoM is making comparisons between apples and oranges. They ought to know better.
The lame-stream press are of course falling over each other to predict imminent immolation of mankind.
UPDATE: I’ve managed to make 2011 the warmest year in, like evah!
NB: The dip in the curves around 2000 is a result of the automated station not working for 2 months in 1999. Maximum temperature was recorded at 0°C during outages and that has an effect when you take the running average or total which includes that period.
The green line shows “clearly” that 2011 was the hottest year. But the trick is to take an average of the temperature extremes over a year.
Statisticians and those who deal with the science of measurement can explain it better than I. Such jiggery-pokery to get the desired result (alarm) has no statistical or physical validity. For starters; John Brignell explains the Extreme Value Fallacy in his FAQ.
“Raw” data plotted earlier shows no such trend at a glance. The human brain is much quicker at recognizing patterns visually than going through abstract mathematics. If you don’t see a pattern or a trend at a glance, then there probably isn’t a real pattern or trend.
I’ve also plotted insolation (yellow), the amount of energy received from the sun as a running total over the same period as well as rainfall (blue), also as a running, accumulated total. The values for those two are shown on the right Y-axis.
When there’s sunshine, it gets warmer. Unless it’s been wet. (Correlation, not causation.)
No Nobel Prize in Physics for that observation. And no Pulitzer Prize for the pun.
On a hunch, I decided to investigate our very own hockey-stick phenomenon a little further… why the up-tick in the 2010/2011 average maxima?
So I plotted the running average of minimum temperatures (the darker green line below) and found an even sharper up-tick.
So it looks like the location has recently taken to not cooling off as much at night. As we don’t get 2°C of warming a year from a 2ppm increase in CO2. Besides; it’s an argicultural research station where there a lots of plants which love the demon gas.
When one is too close to the data, one loses perspective. Wood for the trees.
The view from above produced a useful perspective. A water treatment plant (Beenyup) within 500 metres to the East. One that has recently been expanded in capacity. In the years 2008 to 2010.