One of the issues in measuring global temperature change is the placement of temperature monitoring stations. Some have suggested that stations once located in relatively unpopulated areas have become unduly influenced by the urban island heat effect. I live in Kansas City. When the radio gives the temperature, downtown is almost always a couple of degrees warmer than at the airport, fifteen miles from downtown in a much less densely populated area (note the cattle on some properties adjoining the airport.) As the population expands and engulfs the airport, you might expect to see the average temperature rise.
There are 100s of stations across the continental US, so one might question how much a handful of stations swallowed up urban heat islands would distort temperatures. (I'll have more about this in a later post.) However, only about 2% of the world's land mass is the US. How about other regions?
Brazil has a land area larger than the continental United States. Several months ago, Steve McIntyre of Climate Audit published a post on climate readings in Brazil. Only six stations have continuous records back to the 1930s. Only one is listed as rural. Here are charts of the temperature anomalies from the average for each station. Can you identify which one is the rural station?
Yup! Quixeramobim (and no, I have no idea how to pronounce that). This doesn't prove an urban island effect, but it does look suspicious.
Comments