1. Review and Rating Discharge Measurements David S. Mueller Office of Surface Water March 2010

2. Overview Considerations in Rating a Measurement Considerations in Rating a Measurement Six Review Steps Six Review Steps Gather information for rating as you review Gather information for rating as you review Three Examples Three Examples

3. Rating Process Quantitative assessment of measurement uncertainty Quantitative assessment of measurement uncertainty Currently being worked on Currently being worked on Current practice is subjective Current practice is subjective Where we can quantify we will Where we can quantify we will Where we can’t we will make qualitative estimates Where we can’t we will make qualitative estimates

4. Rating Considerations Variability of transect discharges Variability of transect discharges Pattern in transect discharges Pattern in transect discharges Accuracy of top and bottom extrapolation Accuracy of top and bottom extrapolation Accuracy of edge extrapolation Accuracy of edge extrapolation Quantity and distribution of missing data Quantity and distribution of missing data Quality of boat navigation reference Quality of boat navigation reference Bias or noise in bottom track Bias or noise in bottom track Noise in GPS data Noise in GPS data

5. Spread in Transect Discharges A coefficient of variation (COV) is computed by the software (Std./|Avg.|) A coefficient of variation (COV) is computed by the software (Std./|Avg.|) The COV represents the sampled random error in the discharge measurement. The COV represents the sampled random error in the discharge measurement. Dividing the COV by the sqrt of the number of passes and multiplying by the correct value will provide the 95% uncertainty associated with random errors as measured Dividing the COV by the sqrt of the number of passes and multiplying by the correct value will provide the 95% uncertainty associated with random errors as measured For 4 passes: 1.6*COV For 4 passes: 1.6*COV For 8 passes: 0.8*COV For 8 passes: 0.8*COV This does not include any bias errors This does not include any bias errors This is probably and optimistic estimate This is probably and optimistic estimate

6. Example of Random Error 95% Uncertainy (Random Error) = 0.05*0.8 = 4%

7. Patterns in Transects Random errors should not display a pattern Random errors should not display a pattern A pattern in flow may indicate the need to adjust measurement procedures A pattern in flow may indicate the need to adjust measurement procedures Rapidly rising or falling – average fewer transects Rapidly rising or falling – average fewer transects Periodic – use more transects to capture variability Periodic – use more transects to capture variability

8. Top Extrapolation Average appropriate number of ensembles Average appropriate number of ensembles Evaluate discharge profile Evaluate discharge profile Check field notes for wind effects Check field notes for wind effects Try different extrapolation methods and compare change in total discharge Try different extrapolation methods and compare change in total discharge Little change – accurate extrapolation likely Little change – accurate extrapolation likely Significant change – how good is the data on which the selection is made Significant change – how good is the data on which the selection is made Surface effects captured? Surface effects captured? Field notes consistent with pattern? Field notes consistent with pattern? What percentage of flow is in the top portion What percentage of flow is in the top portion Subjective judgment Subjective judgment

9. Bottom Extrapolation Smooth streambed Smooth streambed Accurate extrapolation likely Accurate extrapolation likely Rough streambed Rough streambed Extrapolation uncertain Extrapolation uncertain

10. Edge Estimates What percentage of flow is in the edges? What percentage of flow is in the edges? May be small enough to disregard accuracy May be small enough to disregard accuracy Recommend no more than 5% in each edge Recommend no more than 5% in each edge Subjective judgment Subjective judgment Large edge estimates Large edge estimates Irregular edge shapes Irregular edge shapes Edge Q = 25% of total

11. Invalid Data Percent of invalid data Percent of invalid data Distribution of invalid data Distribution of invalid data How well is that portion of the flow estimated How well is that portion of the flow estimated

12. Summary of Rating Approach Consider the data Consider the data Consider the hydraulics Consider the hydraulics Consider site conditions Consider site conditions Use statistics, if possible Use statistics, if possible Use your judgment Use your judgment Pay attention to potential errors (as a percentage) as you review the data Pay attention to potential errors (as a percentage) as you review the data

13. Steps of Review 1. QA/QC 2. Composite tabular 3. Stick Ship Track 4. Velocity contour 5. Extrapolation 6. Discharge summary

14. QA/QC ADCP Test Error Free? ADCP Test Error Free? Compass Calibration Compass Calibration Calibration and evaluation completed Calibration and evaluation completed < 1 degree error < 1 degree error Important for Loop and GPS Important for Loop and GPS Moving-Bed Test Moving-Bed Test Loop Loop Requires compass calibration Requires compass calibration Use LC to evaluate Use LC to evaluate Stationary Stationary Duration and Deployment Duration and Deployment SMBA required to evaluate StreamPro SMBA required to evaluate StreamPro

15. Composite Tabular Number of Ensembles Number of Ensembles > 150 ??? > 150 ??? Lost Ensembles Lost Ensembles Communications error Communications error Bad Ensembles Bad Ensembles Distribution more important than percentage Distribution more important than percentage Distribution evaluated in velocity contour plot Distribution evaluated in velocity contour plot Reasonable temperature Reasonable temperature

16. Stick Ship Track Check with appropriate reference (BT, GGA, VTG) Check with appropriate reference (BT, GGA, VTG) Track and sticks in correct direction Track and sticks in correct direction Magnetic variation Magnetic variation Heading time series Heading time series Sticks representative of flow Sticks representative of flow WT thresholds WT thresholds Irregular ship track? Irregular ship track? BT thresholds BT thresholds BT 4-beam solutions BT 4-beam solutions

17. Velocity Contour Patterns in velocity distribution Patterns in velocity distribution Nonuniform Nonuniform Unnatural Unnatural WT Thresholds WT Thresholds Spikes in bottom profile Spikes in bottom profile Turn on Screen Depth Turn on Screen Depth Distribution of invalid and unmeasured areas Distribution of invalid and unmeasured areas Qualitative assessment of uncertainty Qualitative assessment of uncertainty

18. Velocity Extrapolation Average ensembles (10-20) Average ensembles (10-20) Evaluate several transects Evaluate several transects Use Power/Power as default Use Power/Power as default Find reasons to deviate from Power/Power Find reasons to deviate from Power/Power Don’t waste time where there is little difference Don’t waste time where there is little difference In doubt try different options an evaluate change in discharge In doubt try different options an evaluate change in discharge Helps evaluate potential uncertainty from extrapolation Helps evaluate potential uncertainty from extrapolation Person collecting data gets benefit of doubt Person collecting data gets benefit of doubt

19. Discharge Summary Transects within 5% Transects within 5% Directional bias Directional bias GPS used GPS used Magvar correction Magvar correction Consistency Consistency Discharges (top, meas., bottom, left, right) Discharges (top, meas., bottom, left, right) Width, Area Width, Area Boat speed, Flow Speed, Flow Dir. Boat speed, Flow Speed, Flow Dir.

20. Example 1 – Compass Cal. Compass Evaluation is Good! Compass Evaluation is Good!

21. Example 1 – Moving-Bed Test Loop looks good Loop looks good LC finds no problems with loop LC finds no problems with loop Lost BT Lost BT Compass Error Compass Error No moving-bed correction required No moving-bed correction required

22. Example 1 – Composite Tabular > 300 ensembles > 300 ensembles No lost ensembles No lost ensembles 0-1 bad ensembles 0-1 bad ensembles Temperature reasonable Temperature reasonable

23. Example 1 – Ship Track / Contour Ship track consistent Ship track consistent No spikes in streambed in contour plot No spikes in streambed in contour plot No hydraulically unusual patterns No hydraulically unusual patterns Unmeasured areas reasonable in size Unmeasured areas reasonable in size

24. Example 1 - Extrapolation Top extrapolation a bit uncertain Top extrapolation a bit uncertain Try Power/Power Try Power/Power Changed Q = 0.6% Changed Q = 0.6% Const / No Slip = OK Const / No Slip = OK

25. Example 1 – Discharge Summary All within 5% All within 5% 1 negative Q in left edge 1 negative Q in left edge Everything else very consistent Everything else very consistent

26. Example 1 - Rating Base uncertainty Base uncertainty 1.6*0.01= 1.6% 1.6*0.01= 1.6% Invalid Data Invalid Data None None Extrapolation Extrapolation 0.5-1% uncertainty in extrapolation 0.5-1% uncertainty in extrapolation Top is only about 10% of total Top is only about 10% of total Edges Edges About 1% of total Q About 1% of total Q Final assessment Final assessment Rate this measurement “GOOD” Rate this measurement “GOOD”

27. Example 2 – QA/QC No test errors No test errors Compass Cal Compass Cal High pitch and roll High pitch and roll Moving-bed tests Moving-bed tests Bottom track problems Bottom track problems

28. Example 2 – Bottom Track Issue Time series plots show spikes Time series plots show spikes 3-beam solution 3-beam solution High vertical velocity High vertical velocity

29. Example 2 – Composite Tabular Too few ensembles Too few ensembles Short duration Short duration Bad ensembles Bad ensembles

30. Example 2 – Ship Track / Contour Irregular ship track Irregular ship track Thresholds didn’t help Thresholds didn’t help No spikes in streambed No spikes in streambed Invalid data Invalid data Not many Not many High percentage because of so few ensembles High percentage because of so few ensembles Estimates probably reasonable Estimates probably reasonable Large unmeasured areas Large unmeasured areas % measured < 20% % measured < 20%

31. Example 2 - Extrapolation Averaged 10 Averaged 10 Highly variable Highly variable Difference between Power/Power and Constant / No Slip is 3.6% Difference between Power/Power and Constant / No Slip is 3.6% Uncertain about extrapolation accuracy Uncertain about extrapolation accuracy

32. Example 2 – Discharge Summary Not within 5% Not within 5% 8 passes 8 passes Consistent edges Consistent edges Inconsistent width and area Inconsistent width and area Bottom track problems Bottom track problems High Q COV High Q COV

33. Example 2 - Rating Base uncertainty Base uncertainty 0.14 * 0.8 = 10% 0.14 * 0.8 = 10% Invalid data Invalid data Some but not significant Some but not significant Too few ensembles (or too few transects) Too few ensembles (or too few transects) Bottom track problems Bottom track problems Large unmeasured areas Large unmeasured areas Extrapolation Extrapolation Power/Power to Constant/No Slip = 3.6% Power/Power to Constant/No Slip = 3.6% Edges Edges Reasonable Reasonable Final Assessment Final Assessment Rate this measurement “POOR” Rate this measurement “POOR”

34. Example 3 – QA/QC No errors in ADCP test No errors in ADCP test Good compass calibration Good compass calibration Two loop tests Two loop tests

35. Example 3 – Composite Tabular High percentage of bad ensembles High percentage of bad ensembles

36. Example 3 – Composite Tabular Very few bad ensembles with reference set to GGA!! Very few bad ensembles with reference set to GGA!!

37. Example 3 – Ship Track / Contour Everything looks good with reference set to GGA Everything looks good with reference set to GGA

38. Example 3 – Ship Track / Contour With bottom track we need to qualitatively assess the effect of the invalid data on uncertainty With bottom track we need to qualitatively assess the effect of the invalid data on uncertainty

39. Example 3 - Extrapolation Given the variability among the profile plots constant / no slip seems reasonable. Given the variability among the profile plots constant / no slip seems reasonable. Changing to power / power made no difference in the final Q Changing to power / power made no difference in the final Q Might justify power / power Might justify power / power

40. Example 3 – Discharge Summary Small directional bias Small directional bias Typical of magvar or compass error with GPS Typical of magvar or compass error with GPS Adjusting magvar 2.5 degrees removed directional bias and changed Q by 1.3 cfs. Adjusting magvar 2.5 degrees removed directional bias and changed Q by 1.3 cfs. Edge discharge are inconsistent but small relative to total Q Edge discharge are inconsistent but small relative to total Q

41. Example 3 – Discharge Summary Bottom track referenced discharge are lower indicating a moving bed was present. Bottom track referenced discharge are lower indicating a moving bed was present. Edge discharges are inconsistent but small Edge discharges are inconsistent but small Would require correction with LC Would require correction with LC

42. Example 3 - Rating GPS (GGA or VTG) should be used as the reference. GPS (GGA or VTG) should be used as the reference. Base uncertainty Base uncertainty 0.03 * 1.6 = 4.8% 0.03 * 1.6 = 4.8% Value inflated due to directional bias Value inflated due to directional bias No significant invalid data No significant invalid data Extrapolation and edges would make little difference Extrapolation and edges would make little difference Final assessment Final assessment Could rate this measurement “Good” wouldn’t argue much if rated “Fair” Could rate this measurement “Good” wouldn’t argue much if rated “Fair”

43. Example 3 - Rating If GPS were not available and BT had to be used the rating would change. If GPS were not available and BT had to be used the rating would change. Loops Loops High percentage of invalid bottom track High percentage of invalid bottom track 5-8% moving-bed bias 5-8% moving-bed bias Base uncertainty Base uncertainty 0.02 * 1.4 = 3.2% 0.02 * 1.4 = 3.2% Invalid data Invalid data Represent a significant percentage of the flow Represent a significant percentage of the flow Likely estimated well from neighboring data Likely estimated well from neighboring data Edges Edges Inconsistent Inconsistent Small percentage Small percentage Extrapolation Extrapolation Errors very small Errors very small Final assessment Final assessment Loop correction probably within 4% + 3.2% random error Loop correction probably within 4% + 3.2% random error I would probably rate this measurement “Fair” I would probably rate this measurement “Fair”

44. Summary Consider potential uncertainty or errors as you review the data Consider potential uncertainty or errors as you review the data Keep review to a minimum unless errors are identified or suspected Keep review to a minimum unless errors are identified or suspected Consider the percent of effect of an error or uncertainty will have on the discharge Consider the percent of effect of an error or uncertainty will have on the discharge Extrapolation uncertainty can be large in shallow streams Extrapolation uncertainty can be large in shallow streams Use the COV in the discharge summary as the base minimum uncertainty Use the COV in the discharge summary as the base minimum uncertainty Contact OSW if you have unusual data or you have questions Contact OSW if you have unusual data or you have questions

45. Questions???