Table created using field notes and visual interpretation of time series on wqdatalive and using plot_ly in 2_interactive_qaqc_each_parm.R (see interactive plot of Turbidity.side.scatter below).
X = parameter invalid (sensor dry, sedimentation, malfunction) CAL = calibration problem; fixable by back-calibrating to RV values VAR = Values erratic or variable; potential issue with quality RNG = values out of range of the sensor
Start | End | Parameter | Screen.label | Comment.short | Comment.long |
---|---|---|---|---|---|
10/14/2022 19:30 | 11/8/2022 12:30 | ALL | X | Sonde dry | Sonde.dry |
11/9/2022 14:20 | 11/15/2022 10:00 | ALL | X | Sonde buried | Sonde buried; cleaned by SG on 2022.11.15 |
10/15/2022 12:00 | 11/14/2022 18:10 | TRYP | CAL | TRYP cal off | TRYP calibration off by 10x; use RV and standard calibration |
10/15/2022 12:00 | 11/14/2022 18:10 | Turb.Manta | CAL | TURB.Manta cal off | TURB Manta calibration error; cal log shows negative values for 100 NTU. Use RV and standard calibration |
11/8/2022 19:10 | 11/9/2022 0:30 | Turb.SS | X | TURB.SS.TU error | TURB OBS SS low during peak of storm. Possible occlusion of the sensor |
11/8/2022 19:10 | 11/9/2022 0:30 | Turb.BS | X | TURB.BS.TU error | TURB OBS BS low during peak of storm. Possible occlusion of the sensor |
11/9/2022 0:40 | 11/9/2022 5:30 | Turb.SS | VAR | TURB.SS.TU.variable | TURB OBS SS high and variable. May need additional smooothing |
11/9/2022 0:40 | 11/9/2022 5:30 | Turb.BS | VAR | TURB.BS.TU.variable | TURB OBS BS high and variable. May need additional smooothing |
11/8/2022 17:10 | 11/9/2022 3:00 | Turb.Manta | RNG | TURB.Manta out of range | TURB.Manta out of range |
11/9/2022 11:19 | 11/9/2022 11:21 | Turb.Manta | RNG | TURB.Manta out of range | TURB.Manta out of range |
Example of interactive plot used to determine QAQC break dates and times, here for Turbidity OBS side-scatter.
Plot of all raw data (black line), and final raw data (blue). Gray boxes indicate periods where all parameters were not valid. This can be caused by the sonde being dry, by significant biofouling, or by sedimentation on the sonde.
For November 14, 2022 download, the sonde was dry until Nov 8, and had significant sedimentation on Nov 9:
2022-11-14 notes: Most parameters had valid values over the period ending 2022-11-14. TRYP and CDOM were low due to miscalibration; the raw “return values” were therefore back-calculated using the calibration logs, and a second value of TRYP and CDOM calculated using the long-term stable calibration values for estimating ppb from RV. Plots below are the raw ppb values prior to recalibration.
Next: recalculate TRYP and CDOM using default calibrations. Calculate % ww using the TRYP-WW regression from the WW lab addition experiments.
Turbidity values followed the same pattern for all 3 sensors but had
some unlikely value drops at the storm peak. The turbidity sensors then
read reasonable values on the falling limb, until burial.
VAR = variable, possible quality problems RNG = variable out of range CAL = requires recalibration X = variable invalid
TRYP requires recalculation with different calibration parameters–should be higher than that for the raw sewage present at the beginning of the storm.
###6. Recalibrate TRYP, CDOM, TURB as needed For the 2022-11-14 download, TRYP, CDOM and TURB had issues with calibration, so were re-calculated using standardized calibration parameters, which are based on a compilation of all calibrations performed with the Manta sondes.
###7. Correct TRYP for turbidity. Based on laboratory experiments adding bentonite to TRYP 100ppb standard.
###7. Percent wastewater calculation Two estimates of %ww, based on
1) TRYP and 2) CDOM from laboratory experiments.