Welcome to EnviroDIY, a community for do-it-yourself environmental science and monitoring. EnviroDIY is part of WikiWatershed, an initiative of Stroud Water Research Center designed to help people advance knowledge and stewardship of fresh water.
New to EnviroDIY? Start here

Artifactual troughs and flickers in recorded stream depth

Home › Forums › Monitor My Watershed › Artifactual troughs and flickers in recorded stream depth

Tagged: ,

Viewing 11 reply threads
  • Author
    Posts
    • 2024-09-15 at 4:37 PM #18653
      D_Manning
      Participant

        Beginning two weeks ago (September 2) my EnviroDIY station (VCRP1) equipped with a METER CTD sensor began displaying troughs and flickers in recorded stream depth that don’t correspond to true changes in depth as actually measured. Weirdly, the troughs and flickers occur at precisely the same times in almost precisely the same patterns on a diurnal basis, as if they’re linked to some clock function. I see no relation to variations in cellular signal strength, and the battery charge is fine. I’ve attached a photo of a few days of the data. Any ideas??

        Attachments:

        1. Screenshot-2024-09-15-at-4.17.19 PM.png

      • 2024-09-16 at 9:32 AM #18655
        Shannon Hicks
        Moderator

          Have you been to the station to clean the sensor lately?  Also, I’d recommend replacing the logger’s battery with a fully-charged one.  Sometimes when those battery packs are starting to fail (from corrosion on the terminals under the plastic wrapper) they will generate noise which can sometimes affect the sensor readings.

        • 2024-09-20 at 8:17 AM #18662
          D_Manning
          Participant

            Yeah – the sensor was cleaned recently. I had also switched the logger on and off. I’ve now replaced the battery. No luck – the trough and flickers, though slightly modified, still occur. I’ll try to substitute in another sensor, possibly another logger, in order to isolate the problem.

          • 2024-09-20 at 9:55 AM #18663
            Shannon Hicks
            Moderator

              You can also try disconnecting the solar panel for a few days since you’ve got a fully-charged battery on it.  Then that’ll tell you if there’s any correlation between the sensor noise and charging activity.  On a few rare occasions, a failing solar panel ended up generating noise on CTD sensor data.

            • 2024-09-20 at 9:55 AM #18664
              Shannon Hicks
              Moderator

                You can also try disconnecting the solar panel for a few days since you’ve got a fully-charged battery on it.  Then that’ll tell you if there’s any correlation between the sensor noise and charging activity.  On a few rare occasions, a failing solar panel ended up generating noise on CTD sensor data.

              • 2024-10-08 at 7:05 AM #18676
                D_Manning
                Participant

                  No luck. I disconnected the solar panel, but am still seeing artifactual troughs. It’s reminiscent of the much more extreme variations in reported stream depth for the Pine Creek station, which calmed with a new sensor. The latter might have been related to a frayed wire from the sensor, I’m not sure.

                • 2024-10-10 at 10:54 PM #18689
                  neilh20
                  Participant

                    Just wondering if you could post a link to where the data is being collected, and look at it as a system.

                  • 2024-10-11 at 7:30 AM #18690
                    D_Manning
                    Participant

                      Sure. https://monitormywatershed.org/sites/VCRP1/

                      Best way to view it is with time series visualization. Exclude interruption of pattern by a rain event and having the sensor out recently over a recorded reading.

                    • 2024-10-14 at 8:04 PM #18693
                      neilh20
                      Participant

                        As you say from 2nd sept – something changed, after that it looks to me like its is triggered by the temperature changes. Its just a visual analysis, but also using the conductivity. With a rain event the conductivity also typically changes, and of course temperature can change.

                        For daily diurnal cycle after sept 2 the temperature starts varying by about 3C a day. before that it was about 2C a day.

                        The Metros_hyrdos21 Gen2 specification says “NOTE: Depth measurement accuracy assumes no abrupt temperature variations”  .

                        All depth sensors/piezo resistive sensors have a temperature co-efficient, and typically have some form of temperature compensation to correct for changing temperature.   For some reason something changed Sept 2 – you might have to forward the info to Metros and see if they can make any suggestions

                        Depth and temperature

                        Attachments:

                        1. Screenshot-2024-10-14-170239.png

                      • 2024-10-14 at 8:35 PM #18695
                        neilh20
                        Participant

                          .

                        • 2024-10-15 at 9:26 AM #18697
                          D_Manning
                          Participant

                            Thanks. Temperature coefficients, compensations – that’s interesting. I have the sensor sitting in a bucket of stream water right now – should see more rapid changes in temperature. I’ve a note into the METER Co.

                          • 2024-11-18 at 8:51 PM #18750
                            neilh20
                            Participant

                              @d_manning just wondering if you have had any insights 🙂

                          • Author
                            Posts
                          Viewing 11 reply threads
                          • You must be logged in to reply to this topic.