Slashdot Mirror
This Was America's Warmest Winter On Record (slate.com)
hondo77 writes: On Tuesday, the National Oceanic and Atmospheric Administration released its official assessment of December, January, and February's temperatures across the United States, and the results are striking: Not a single state in the U.S. had a cooler than average winter. (NOAA treats Alaska and Hawaii separately, due to shorter weather data records there -- though both states were significantly warmer than normal this winter. Weather records for the contiguous United States go back to 1895.) NOAA blames the recent warm weather on a record-strength El Nino "and other climate patterns," most notably, global warming. As a whole, this winter in the lower 48 was about 4.6 degrees Fahrenheit warmer than the 20th century average: a sharp contrast to the previous back-to-back frigid polar vortex winters, especially in the Northeast.
And where is it measured? In the ovens or in the made up data from East Anglia University? Thermometers from 100 years ago?
Such old instruments would do just fine. But in case you don't have those:
Take a long, glass tube with constant cross-section and a reservoir at the bottom. Fill with a substance that's chemically stable and liquid at both freezing & boiling point of water. Even though out of fashion these days, mercury is a good choice. Pump space above the liquid vacuum, and seal hermetically.
Go to a point @ sea level, take a bucket of pure water, put your thermometer in it, and cool such that some ice floats in it, some water is also in there, and temperature is stable and evenly distributed. Mark the liquid level in your thermometer with "0".
Now bring the water in the bucket to a boil, again wait until temperature is evenly distributed, and mark the liquid level in your thermometer with "100". Afterwards, divide the space between markings "0" and "100" in 100 equal parts, and (if possible) sub-divide each part in .1, .2, ..., .9 markings. Using the 0-100 part as reference, extend a bit below "0" until you get to the part of your glass tube where cross-section isn't constant anymore.
Now find a place to do measurements: some standard height from the ground (1.5m?), shielded from the sun but allowing for -some- airflow, NOT in a place where nearby human structures or activity will f**k up the readings, thermometer mounted such that it'll reach equilibrium with surrounding air temperature, and can be read without influencing the reading.
Then take a notepad, and once (or more) each day, go up to the location. Note date, time, place of reading, and your best estimate at what the thermometer shows. If reading doesn't make sense, investigate why. From time to time, check or re-calibrate thermometer if necessary.
As you see, it takes effort and attention to detail to get good readings. I'm sure modern weather-people will have higher-accuracy instruments, automated setups, and a wealth of number-crunching equipment to make sense of the data. And US-based folks might want to add a Fahrenheit scale for the locals. But none of the above is rocket science, and even a century ago (or 2? or 3? or 5?) people knew how to do this, took notes, and sometimes preserved those records. So unless you can show their methods were flawed somehow, their readings are as valid as what you'd get today @ the same place. Even if that old data has to be taken with a grain of salt, it's still data points that could be meaningful. Or even accurate. Regardless how old a thermometer was used.