NCDC Frequently Asked Questions
If the measuring equipment changes at a site, does that change the records? Does it change the normals?
[ Return to FAQs ]
Changes at the site never change what was originally observed, but these observations are subject to edits based on established criteria.
For normals, if the new equipment does not record weather elements in exactly the same way as the old, and causes a change in how weather is recorded relative to the previous instrumentation, then it does change the normals. If the equipment MOVES, this will also often cause apparent changes in climate. Location moves (anywhere from a few hundred feet to a few miles) can, in fact, cause greater changes than do instrumental changes.
The following white paper (United States Climate Normals, 1971–2000: Inhomogeneity Adjustment Methodology) [PDF] is available regarding procedures for adjusting station data to account for inhomogeneities due to changes in station locations, instrumentation, time of observation, surrounding environment, observing practice, sensor drift, etc. The purpose of such adjustments is to produce a time series and normals statistics that are representative of the observing practices as of the end of the normals period (December 2000), since these are the conditions under which future observations will likely be compared.
Since climate fluctuates constantly, there are real changes in "normals" due to climate variations, and there are fake changes in "normals" due to artificial things like sensor changes, equipment moves, methodological issues, and so on. The ONLY way to distinguish between fake and real variations is to have records from another piece of equipment at one site or the other that just keeps measuring the same way.