Hydrol. Earth Syst. Sci., 16, 345356, 2012www.hydrol-earth-syst-sci.net/16/345/2012/doi:10.5194/hess-16-345-2012 Author(s) 2012. CC Attribution 3.0 License.
Hydrology andEarth System
Sciences
Technical Note: How image processing facilitates the rising bubbletechnique for discharge measurement
K. P. Hilgersom and W. M. J. Luxemburg
Water Resources Section, Faculty of Civil Engineering and Geosciences, Delft University of Technology, P.O. Box 5048,2600 GA Delft, The Netherlands
Correspondence to:K. P. Hilgersom ([email protected])
Received: 9 September 2011 Published in Hydrol. Earth Syst. Sci. Discuss.: 20 September 2011Revised: 21 January 2012 Accepted: 1 February 2012 Published: 6 February 2012
Abstract. In this article, we rehabilitate the integrating ris-ing bubble technique as an effective means of obtaining dis-charge measurements. SinceSargent(1981, 1982a), the tech-nique has not been applied widely, mainly as a result of prac-tical difficulties. We hypothesize that modern image process-ing techniques can greatly improve the rising bubble tech-nique. We applied the technique in both a laboratory setupand a field study, after determining the bubble rising velocityfor our nozzles in the specific case. During our measure-ments, we captured digital photographs of the bubble enve-lope at the water surface, each picture being a single mea-surement of the discharge. The photographs were correctedfor lens distortion and reprojected so that accurate distanceson water surface level could be obtained. This easy digi-tal procedure resulted in accurate discharge measurements,even when turbulence was involved and the averages of mul-tiple image analyses yielded good results. The study showsthat the rising bubble technique can be a preferable dischargegauging technique in some situations. Recent developmentsin image processing facilitate the method substantially.
1 Introduction
A large variety of discharge gauging techniques is appliedin the field of hydrological measurements. Each of themperforms best under specific stream properties, such as thegeometry of the cross-section and turbulence. The integrat-ing float method, one of these techniques, best resemblesthe frequently applied method using a float at the water sur-face. With the latter method, hydrologists measure streamflow by releasing a float at the water surface and calculatingthe velocity from its displacement over time. This method iseasy and quick but is considered relatively inaccurate, for an
uncertain correction factor is applied to estimate the depth-averaged velocity from the measured surface velocity.
To overcome this, the integrating float method has beendeveloped (Herschy, 1978, 1995), where a float is released atthe bottom of a river or canal. The float is assumed to risewith a constant velocity, so the depth-integrated horizontalvelocity can be determined from the floats displacement as itsurfaces. Air bubbles are a type of float that can be applied ina simple manner. This article discusses the merit of moderncomputational and photogrammetric techniques for applyingthe integrating float method and specifically the rising bubbletechnique to measure river discharge (Sargent, 1981; Shaw,1994).
Already in the 19th century, DAuria described a methodin which a sinker was released at water surface level to de-termine the depth-integrated velocity by the total horizontaldisplacement of the sinker as it reaches the river bed (Sar-gent, 1981). At the start of the twentieth century,Hajos(1904) described the reverse procedure: floats are releasedat the stream bed and their displacement can be found whenreaching the surface. Amongst others,Liu and Morris(1970)improved this method and investigated the influence of tur-bulence on the method.
Viol and Semenov(1964) were the first to apply this in-tegrating float method using air bubbles as floats, assumingthese air bubbles have a constant rising velocity. Their workformed the basis for research bySargent(1981, 1982a,b),who applied two photo cameras with autowinders to obtaina time series of photographs of surfacing air bubbles. Thehorizontal displacements of the air bubbles with respect totheir release points on the river bed could then be calcu-lated from the two pictures taken from different angles usinga standard photogrammetric technique. Repeating the pro-cedure provided an easy means to register how stream dis-charge evolves over time (Sargent, 1981, 1982a). Sargentswork meant a large advance with respect to applicability and
Published by Copernicus Publications on behalf of the European Geosciences Union.
346 K. P. Hilgersom and W. M. J. Luxemburg: Image processing facilitates the rising bubble technique
accuracy of the method. However, since Sargents publica-tions, the method has only been applied by a few authors(Toop et al., 1997; Yannopoulos et al., 2008), suggestingthat previous technologies made the method less easy andmore expensive to apply compared to other gauging meth-ods. A discussion of Sargents work confirms this (Sargentand Davis, 1981). We believe that modern techniques cansignificantly improve the rising bubble technique.
To address the potential of the rising bubble technique asa modern and usable method, we introduce the applicationof image processing techniques. With this novel approach,we obtain quantitative data from photographs that were takenduring tests in a laboratory flume, and case studies in thechannelized river Reitdiep (The Netherlands) and the riverSure (Grand Duchy of Luxembourg). We demonstrate thatthe use of a single photo or video camera is sufficient toperform accurate measurements and that automation of theprocess provides the opportunity to perform continuous mea-surements with this gauging technique. Measurement resultsfrom the case study show that the method performs accu-rately in the specific situations.
Rather than a manual on measuring with the rising bub-ble technique, the aim of this publication is to show how theaddition of digital image processing techniques supports themethod.
2 Study site
We tested the rising bubble technique both in a laboratoryflume and in the field. The field study consisted of measure-ments in two distinct situations: a lock with a rectangularcross-section and and a natural stream with a more variablecross-section.
2.1 Lock measurements
The first field study was performed in the ProvincialeSluis (532014 N, 61750 E), a lock near the village ofZoutkamp (province of Groningen, The Netherlands), on9 October 2009. This 9-meter-wide lock is constructed onthe point where the channelized river Reitdiep enters theLauwersmeer (see Fig.1). At the time that this lake was stillpart of the Wadden Sea, the lock allowed ships to transfer be-tween the tidally influenced sea level and the controlled waterlevel of the Reitdiep. Since the Lauwersmeer was dammedin 1969, the lock is no longer required. Nevertheless, itsrectangular cross-section makes the lock suitable to accu-rately perform automatic discharge measurements. For thisreason, the lock is equipped with an acoustic discharge mea-surement device (ADM). Both the availability of data fromthe ADM and the locks geometry make this site suitable forfirst testing the rising bubble technique. Besides, the bridgeabove the lock makes it possible for us to take photographsfrom above the water surface.
Fig. 1. Map of the Lauwersmeer; the dark grey lines represent thesurrounding embankment.
The Lauwersmeer discharges into the Wadden Sea dur-ing low tide at sea by means of the R. J. Cleveringsluizen.Opening these sluices strongly enhances the discharge of theReitdiep near Zoutkamp. To apply the rising bubble tech-nique at varying stream discharges, the R. J. Cleveringsluizenwere opened during part of our field measurements.
2.2 Natural stream measurements
We studied the techniques performance for a natural streamin the Sure River (Grand Duchy of Luxembourg) on10 May 2011. The study site is located between the villagesof Boulaide and Bigonville, upstream from the lake Lac dela Haute-Sure (location: 495158 N, 54822 E). Locally,the river is 12.5 m wide and has a natural bed that is shallowon the right half and deeper on the left (facing downstream).
3 Materials and methods
3.1 Method
The principle of the integrating float method is that the timeT needed for a float to surface withvz equals the timeneeded for the float to displace over distanceL with a depth-averaged horizontal velocityvx (see Fig.2):
T =D
vz=
L
vx, (1)
Hydrol. Earth Syst. Sci., 16, 345356, 2012 www.hydrol-earth-syst-sci.net/16/345/2012/
K. P. Hilgersom and W. M. J. Luxemburg: Image processing facilitates the rising bubble technique 347
L
D
x
z
typical velocity profile
float path
float release
Longitudinal cross-section
Fig. 2. Float path occurring with a typical velocity profile overdepth.
whereD denotes the vertical distance between streambedand water surface. From Eq. (1), we find for the specificdischargeq (i.e. the depth-integrated horizontal velocity):
q =
D0
vxdz = vx D = vz L, (2)
as is described bySargent(1982a). Equation (2) shows thatthe specific discharge can be calculated from the constantvz and the displacement of the float while surfacing (L).The latter parameter varies over the width of the stream (y-direction), and several floats are needed to determineL(y).Integrating the measured displacements leads to the total dis-chargeQ:
Q =
w0
qdy = vz w
0L(y)dy = vz A, (3)
A being the surface of the bubble envelope (see Fig.3).
3.2 Application
For the rising bubble technique, air bubbles are the specificfloats. Air bubbles are easier to release from the bottom
