Surfacestations.org volunteers Liz and Don Healy captured this official USHCN climate station of record (COOP ID #22659) in Douglas Arizona on their cross country trip this winter. As is typical with many USHCN stations using an MMTS instrument, cable laying issues put the sensor closer to the building. In this case the sensor is over gravel and only 11 feet from the south (sun facing) wall of the Douglas Fire Department.
Click for a larger image
Don Healy writes in his survey form:
“…the MMTS was 11 feet to the south of a one-plus story stucco building, 9 feet from a concrete sidewalk to the south, 21 feet from a very wide two-lane asphalt street (10th St.), 36 feet from the Dolores Avenue Sidewalk. Additionally, the MMTS was positioned about 11 feet from the east and west from two concrete walkways providing access to the building from 10th street. The MMTS was 9 feet to the east of an estimated 25-foot tall Mesquite tree whose crown reached the area immediately overhead the MMTS. [The person interviewed] indicated that the tree had been pruned recently. The landscaping under the MMTS was lighted-colored gravel.”
Given all the elements in that description, determining the representative temperature for the area would certainly be a complex job of bias deconstruction.
The official MMTS temperature sensor also has a nice view of the street and its surfaces:
The thermometer wasn’t always there, the previous location (8th street park) was better in terms of bias proximity, but still not the best.
Click for a larger image
The original Stevenson screen was in the fenced enclosure in the 8th street park. You can see the ball field in the background. According to NCDC’s Multi Metadata System (MMS) database, it was at this location up to 11-01-2001. That is when the equipment changed from max-min thermometers to the MMTS electronic one.
See the complete picture gallery here.
Unfortunately, even though the station has been moved to the front of the Douglas fire station on 10th street, there is no mention of it in the MMS location records:
In the plot of temperature for this station, using the raw USHCN data at GISS, it does not appear that the data after the move was consistent. However, while only 4 years of data seems to have been plotted (in red), the appearance of a step function seems to have emerged that has not been caught in USHCN adjustments at NOAA:
Click picture for larger image, see original image here
Perhaps our friend Atmoz can shed some light on the subject with his analysis skills which have been good at spotting such problems from a data perspective.