The accuracy for my Lacrosse 2010 temperature sensor is +/- 2 degrees F, while the readouts are in tenths of a degree.
Brian has a temperature correction capability in WD (presumably linear with temperature), but before using this you need to be able to determine what the offset really is.
Any suggestions on calibration? I’ve got a mercury thermometer in 1 degree graduations that I can place beside the Lacrosse sensor, but that’s not really accurate enough.
I guess one could pay a fee and take it to a calibration laboratory, but that’s a little excessive.
Yeah, I guess that’s the only practical way to do it. It makes one wonder though if they’re familiar with something we were taught in engineering school about ‘significant digits’ … it shouldn’t read out in tenths unless it’s really that accurate.
That could be why the NWS reports to the nearest degree, rarther than the nearest tenth of a degree. Also, in my experience with significant figures you estimate past the closest graduation. Since it is graduated to the nearest degree, you “estimate” to the tenth, since you can notice if it is between the graduations.