Detecting Climate Change through Means and Extremes.

This study investigates how the climate is changing at individual weather stations around the world by measuring means and extremes. Extensive python code was written to read in the daily data from thousands of stations and compute average yearly temperature, precipitation, and extremes such as heat waves, cold spells, very wet and dry periods. A best-fit line and goodness-of-fit statistic were computed for each statistic at each station. All these linear trends were collected into histograms and geographic distribution plots. It was found that the earth is undergoing significant changes: 97% of all stations are increasing in temperature since 1950, averaging 4° F per century. All measures of extreme heat are increasing, and all measures of extreme cold are decreasing. Almost all stations are recording significant increases in the number of warm days (days above a reference 90th percentile) and heat waves (three consecutive warm days) and the warmest day of the year is getting hotter. There are fewer cold nights and cold spells. The number of frost nights is significantly decreasing, averaging 25 fewer frost nights per year over a century. Changes in precipitation vary by geography: some regions are drier while others are wetter. For example, precipitation changes are strongly varied in both the U.S. and Australia. This study confirms that the earth's climate is rapidly changing in both means and extremes. To help the public better understand climate change, this data is presented in an online map for interactive exploration at http://lillianpetersen.github.io. [ABSTRACT FROM AUTHOR]