Negative rainfall rate

[freddie]

My rainfall rate was briefly negative this morning (it wasn't raining at the time) - something I've not seen before.

negative-rain-rate.JPG

Any ideas?

[steve]

For Davis stations, Cumulus gets the rain rate from the console. I could check for it being negative and use zero instead; it's not the kind of thing you would normally expect a Davis station to do.

[freddie]

Mystery solved. My ISS battery went flat at about 0615 UTC this morning. As soon as the sun hit the solar panel on the ISS about 20-25 minutes later, power was restored. This evening when the charge from the sun had drained away, the ISS lost power again, and there was a repeat of the earlier symptoms after about 10 minutes without Cumulus getting a reading from the ISS.

Temperature and humidity flat-lined during the periods of time the ISS was without power. This is anticipated, as Cumulus will continue to store the last readings it received when it loses contact with the VP (unless you restart Cumulus, of course).

I guess the negative rain rate can be attributed to the above. I have since edited my data files to remove the spurious data.

[steve]

I was going to ask whether the other data was affected for the same period. The Davis DLL returns an 'error' value when a particular data item is unavailable, and Cumulus checks for this. It appears that the rain rate reading doesn't give the same error value. I've changed the code to check for negative values.