Evapotranspiration issues (Davis)

[steve]

I've changed the way I read/process the ET value in 1.9.1 to get around the problem where the Davis ET reading represents a different item depending on whether it is 'live' or from the logger. I am now using the annual ET total (in the same way that I do for rainfall) and calculating the daily figure relative to that.

But I have noticed that the annual ET figure supplied by the Davis DLL jumps around. When it changes, it changes on the hour; e.g. at 0100 this morning, my annual total jumped up from 78.99 to 1743.60, with a corresponding jump in the Cumulus calculation for the daily total (I use 0900 as my start of day). The annual total then started going down with each hourly update, until at 1200 it reverted to 79.04, resulting in the daily calculation going negative.

Has anyone else noticed this? You may not have spotted it at the time, but you may see it if you look in your monthly log files. To find the figures, look towards the end of each entry. The last five fields are:

# Evapotranspiration
# Annual Evapotranspiration
# Apparent temperature
# Current theoretical max solar radiation
# Hours of sunshine so far today

[steve]

Actually, I may be able to answer my own question. I'd looked to see if this was a known issue with the version of the DLL I'm using, but didn't think to check the console firmware versions. One of the changes in firmware 1.90 is "Fixed clearing of yearly ET history data". I'm still using 1.82; I think an upgrade is in order...

[Gordon-Loomberah]

I've just installed build 962 and see: Evapotranspiration Today 1001.40 mm

Hard to imagine when its been drizzle and showers all day so far! Maybe its the annual total, with wrong tag...

I think I saw this problem mentioned somewhere recently, but haven't been able to find it again to see if there is a fix.

[steve]

What firmware version are you running?

[Gordon-Loomberah]

1.9 (US) English on the VP2

[steve]

It could be down to something I changed between the two versions you've been running. See how it goes after the first day. But I haven't actually checked my data since I upgraded my firmware.

Edit: I've looked at the annual ET figure in my Nov log, and it's been behaving (i.e. only increasing) since I upgraded to firmware 1.90, and the daily figure has been correspondingly correct.

[Gordon-Loomberah]

OK, ET is back to a sensible number for today 0.66mm @midday sounds right, as its been overcast with some rain earlier this morning.