i found this neat site with photos of the locations of temperature sensors used by GISTEMP. it certainly is enlightening. i wonder how much of our raw data over the last 30 years is affected by all the bad placement of the temperature sensors by GISTEMP?
from a comment on one of the pictures:
here are the 4000 stations that are currently used in the GISTEMP data set used for climate modeling. 1221 of them are in the USA.
These weather stations only sample the micro-climate surrounding the sensor. Assume an area around the sensor of, on average about 25 square meters…5m X 5m….with wind effects, maybe that is too small? OK how about 100 square meters but that is broad since the temperature sensor bulb really only measures the volume of air immediately around the sensor, which is quite small, but wind does carry and mix heat from nearby sources, so eventually it evens out over an area.
So then, with 100 square meters of area around a sensor being the representative area sample in the micro climate, multiplied by 4000 sensors worldwide, we can agree that roughly 400,000 meters squared or 400 square kilometers squared of earths surface is actually represented in the meteorological station data used in computer modeling of temperature; compare this to the actual area of earth’s surface 510,065,600 square kilometers.
This demonstrates the weakness of using this data set to measure or predict a global mean temperature. The area being covered by weather station data is only .00008% of the earths total surface.
Global decisions are being rendered on that .00008%
i find that it is significant that over 25% of all the temperature sensors are located in the u.s. wouldn’t that alone skew the sample?
