Author Topic: When does meteorologic institutes measure days precipitation?  (Read 1080 times)

0 Members and 1 Guest are viewing this topic.

Offline IngemarS

  • Member
  • *
  • Posts: 48
    • Österåsen, Fjällsjö Weather Station
Hi,

In Sweden I have found out that SMHI (Swedish Meteorologic and Hydrologic Institutue) measures daily precipitation not between midnight and next midnight. They measure at 6pm to the next 5:59am.

Is this the way other countries´does it? 
Davis Vantage Pro2 Plus, VWS, WXSIM

Offline Mattk

  • Forecaster
  • *****
  • Posts: 2136
Re: When does meteorologic institutes measure days precipitation?
« Reply #1 on: August 11, 2018, 06:00:22 AM »
It varies, in some countries it is from 9am to 9am with an interim recording for the 6 hours to 3pm. Midnight to midnight as per what many consumer type electronics provide is more for simplicity with the distinction between today/yesterday etc. When manual observations were the go no observer was keen or going to be bothered to get up at midnight just to read the rainfall, it was dark, cold and out of normal working hours. 

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: When does meteorologic institutes measure days precipitation?
« Reply #2 on: August 11, 2018, 08:16:35 AM »
In the US the CoCoRaHS observers report at 7AM. Since it it done manually I suppose that is why it is done at that time. They actually let you report whenever you want but they only want your data if it was done between 6 AM and 9 AM. And they want you to pick a time and stick with it always. But they prefer 7 AM.

But that is just CoCoRaHS. Everyone else in the US goes by the calendar day and so at midnight it resets. Please correct me if I'm wrong. I'm just not aware of it being done differently in the US.

I think anyone with automated measuring equipment should stick to midnight as a reset. Since it is automated it is reporting and updating at every minute of the day a running total. It wouldn't make sense to reset in the middle of the day. Unless you want to have numbers to compare to manual readings like from CoCoRaHS.

When you upload to cloud service they just take your data. You don't get to pick when to reset. Most (or probably all) reset at midnight. So if you change your console reset time or your weather software, that would be a number that you only see, because the data being uploaded is just data and your custom reset has no effect other than for you at that device.

Also a common thing to do is to analyze and make note of how much rainfall over a longer period of time. For example, how much rain in the past week or a few days. Like if there had been a hurricane that brought rain for 2 or 3 days then news media often reports total rain over those days. They are just adding it up for the media report.

« Last Edit: July 24, 2020, 08:37:53 AM by galfert »
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline IngemarS

  • Member
  • *
  • Posts: 48
    • Österåsen, Fjällsjö Weather Station
Re: When does meteorologic institutes measure days precipitation?
« Reply #3 on: August 11, 2018, 10:34:05 AM »
Thanks Mattk and galfert!

I really had to look up "Cocohrahs", nice with volunteers!

Yes, in "historic" times it made sense with 6AM, and reports by telephone were made at around 7 AM. But SMHI still says 6 AM makes sense because of the rain period of any day is more likely to occur evenings and nights when temperature falls. Another thing that points to this is that last days precipitation data is not available until around 10AM...

Yes and Sweden has old style manual weather stations, along with a bit better and then a new standard.

I still see all stations reporting precipitation with that skew. From what I can see in my data that skew is more of a true thing during summer half year. But that is only data from my station.

Also, I think that in any country using DST the don´t use the DST midnight, but rather keep to the normal? I was recommended by an American professor in meteorology to keep the Standard time on my ws. This because days when DST et c occurs is not on the same dates, which makes comparisons between dates more problematic. I even do comparisons "days after winter solstice", "days after spring equinox", ... since where I live, N 63.8 E 15.9, we have a swing of -31C to +31C and every day in theory should count for a change of 0.339C per day, and I study the inertia of temperatures. The graph is an example of such an offset from from Winter Solstice . There is a tiny hand written red line that shows we first loose max temps and then the average of max'es, average, and so on.         
« Last Edit: August 11, 2018, 10:48:01 AM by IngemarS »
Davis Vantage Pro2 Plus, VWS, WXSIM