How not to measure temperature, part 49. Alaska’s COOP Stations

Earlier I wrote up an essay on the NOAA climate station at Cordova, AK.

Click thumbnail at left for a larger image. 

  
This station was directly next to the village diesel power plant. That station also happens to be part of the NASA GISS surface temperature record used for climate research. The problem is the proximity to nearby human caused heat sources, which may not be accurately adjusted for in the record. Of course the real issue is that if the stations were properly setup and maintained by NOAA, paying attention to their own 100 foot rule, such potential bias would not be an issue. Today I’d like to show you a few other NOAA climate stations in Alaska.

Click thumbnails below for larger images.

Thanks to John Papineau for these photographs

English Bay - note the MMTS temperature sensor within about 1 foot of the building.No cold winter nights for this sensor!

Moose Pass - note the concrete structure which is a fish hatchery NCDC record says: HATCHERY, OUTSIDE & 3 MI NW OF PO AT MOOSE PASS, AK

Susitna Landing - note proximity to building this was installed on May 21st, 2003 NCDC record says: FLAT GRAVEL AREA NEAR CONFLUENCE OF KASHWITNA AND SUSITNA RIVERS. How would a researcher know about the building proximity from this?

Seward 19N - note proximity to building NCDC Record says: OBSERVERS HOME, OUTSIDE & 19.5 MI N OF PO AT SEWARD, AK Again, how would a researcher know about the building proximity?

Seward #2 - note proximity to street and shading issues. You can see the station location in Google Earth.

Tutka Bay - note proximity to building and weathering of old Stevenson Screen shelter.

As I’ve been saying, the MMTS temperature sensor and it’s cable is systematically forcing measurements closer to human influences. They problem clearly is not unique to the continental United States as these photos from Alaska demonstrate.

In all of Alaska’s open wilderness, are these truly representative of the climate? It seems that every station is close to the small packets of towns and villages that dot Alaska, and necessarily so, since a human observer is required to read and record the thermometer.

Surely though, a better job at station siting could have been done.