Streamflow has been measured continuously at stream gaging stations at the outlets of nine watersheds (W1-9). The height of water passing over a 90 or 120 degree sharp-crested V-notch weir is measured with a float and pulley placed in a stilling well. Stage heights are converted to volumetric discharge using standard, published equations (Hertzler, 1938, King,1954). At weirs 1-4 and 9, when stage heights are above the V notch, high flow calculations are made based on the geometry of the upper rectangular section of the weir outlet. High flows at stations 5-8 are measured by rectangular San Dimas Flumes, oriented upstream and in series with the V-notch weirs. Flume rating curve tables, specific to widths of 2, 3 and 4 feet, were provided by the San Dimas Experimental Forest in Glendora, California where the namesake flume was developed. At weirs 1-8, low flow (stage heights < [0.15 – 0.5 ft]) values are produced with lookup tables developed from weir specific empirical calibrations. During winter, plywood hoods are placed over the V-notches and propane gas heaters keep the V-notches, flumes, and stilling wells ice-free. Over the complete record of streamflow, several technological changes have been made.
1955-2012: Stage heights at the stream gages were recorded continuously with Stevens A-35 mechanical recorders at the weirs, and Belfort FW-1 mechanical recorders at the flumes. The paper strip charts were changed every 2 weeks. Gaps in the record (e.g., due to problems such as debris and ice in the V-notch, stopped clock, poor ink trace, and erroneous resets of stage height) were filled by tracing the ink line from a chart of a similar nearby weir. The analog charts were digitized at Hubbard Brook using a Summagraphics digitizer. For each day, 2 to 130 points on the chart were digitized, depending on the rate of change of flow. During high flow periods digitized points were taken with no more than a 0.150 ft change on the strip chart. Daily streamflow values were calculated from instantaneous values using trapezoidal integration (Bailey et al. 2003).
2013-present: On January 1, 2013, we transitioned to digital streamflow sensors that record stage every 5 minutes (288 points per day). The first digital sensors to measure weir stage, Campbell Scientific CS410 Shaft Encoders (resolution 0.01 ft), were installed in 2010 and operated in parallel with the charts for several years. In 2016, Amass PSE optical shaft encoders (resolution 0.003 ft) were installed and the current version of the published data transitions from CS410 to Amass encoder data on 2017-10-04. Streamflow measurements at the flumes transitioned from mechanical chart recorders to digital shaft encoders about 2013, digital shaft encoders to sonic and radar distance sensors about 2015, and from Max Botix to Judd sonic sensors in 2018 and 2019. Transitioning from shaft encoders to sonic sensors eliminated the need for stilling wells for the flumes, which were difficult to keep ice free during winter. The digital shaft encoders at the weirs and flumes utilize the same float and weight around a pulley technology as the mechanical chart recorders. For 2013-2015, gaps in stage height were filled using a regression relationship with a nearby watershed or by linear interpolation for shorter gaps. Once stage height was gap-filled, specific discharge (mm) was calculated. For 2016-present, specific discharge was calculated first, and gaps were filled with the median of an ensemble of predictions from available watersheds (See et al. 2020). Daily streamflow values are calculated from instantaneous values using rectangular integration.
References
Hertzler, R. A. 1938. Determination of a formula for the 120° V-notch weir. Civil Engin. 8: 756-757.
King, H.W. 1954. Handbook of hydraulics. Ed. 4, 556 pp., illus. New York: McGraw-Hill Book Co., Inc.
See C.R., Green M.B., Yanai R.D., Bailey A.S., Campbell J.L., Hayward J. 2020. Quantifying uncertainty in annual runoff due to missing data. PeerJ 8:e9531 https://doi.org/10.7717/peerj.9531
Bailey, A.S., J.W. Hornbeck, J.L. Campbell, and C. Eagar. 2003. Hydrometeorological Database for Hubbard Brook Experimental Forest: 1955-2000. GTR NE-305
