Discharge measurements

Climatological and hydrologic field course

1 Background
Discharge (or surface runoff Qs) refers to the horizontal water flow occurring at the surface in
rivers and streams. It does not include the groundwater flow (or subsurface runoff Qg). Since
discharge is only defined in streams or rivers, water balance calculations are often performed
for the catchment area upstream of the discharge measurement site, toward which it is often
assumed to drain entirely. This means that the hydrological boundaries of the catchment are
assumed to correspond to the topographic watershed boundaries. In reality, these can be
crossed by the groundwater flow, making this assumption not perfectly valid. Moreover, the
catchment boundaries might also change with time or wetness. During the field course we only
measure the surface runoff.

Surface runoff can be caused by many processes. Under normal conditions, only subsurface
flow contributes to stream flow once it reaches the surface. However, topography, as well as
soil characteristics, can also determine the amount of discharge. For example, the hydraulic
conductivity – which depends strongly on the soil wetness - controls how easily the precipitated
water can percolate vertically into the ground. Depending on this property, during intense
precipitation, water might be unable to infiltrate into the soil, leading to infiltration-excess
runoff. Moreover, after prolonged rainfalls, soils might become saturated, and the remaining
precipitation can lead to saturation-excess runoff. The type of vegetation can influence the
physical properties of a soil as well, and experiments of land-cover changes at the catchment-
scale were found to influence the surface discharge.

Total (surface + subsurface) runoff Q is often expressed in volume of water per unit of time
(m3/s or l/s). Alternatively, it can be expressed in equivalent water depth at the catchment
scale, dividing Q by the catchment area. This is useful, or even essential, in water balance
studies where other fluxes (e.g. precipitation or evapotranspiration) already have the
dimension Length/Time (e.g. mm/d). Discharge can then be formulated as a function of other
surface water fluxes through the surface water balance equation:

𝑑𝑑𝑑𝑑
𝑄𝑄 = 𝑄𝑄𝑠𝑠 + 𝑄𝑄𝑔𝑔 = 𝑃𝑃 − 𝐸𝐸 − (1)
𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑
Here, P is precipitation, E is evapotranspiration, and 𝑑𝑑𝑑𝑑 stands for the temporal change in soil
moisture content. All terms are expressed in m3/s or in mm/d. Compared to the other terms in
the equation, Qs is the only one which is representative for the whole catchment area upstream
𝑑𝑑𝑑𝑑
of the discharge measurement site. Note that on climatic (multiannual) timescales, 𝑑𝑑𝑑𝑑 is often
considered to be zero, meaning that total runoff is equal to the difference between
precipitation and evapotranspiration.
2 Measurement methods
2.1 Bucket method
If one finds a spot where it is possible to capture all the water from the streamflow (for
example at a spillway), using a stopwatch and a bucket. Depending in the discharge, it might be
decided to either measure the time that it takes to fill the bucket, or measure the volume of
water collected in a predefined timespan.

2.2 Water height relationships

An often-used technique to quickly establish the discharge, is to take advantage of the fact that
water height is directly related to stream morphology and discharge. By introducing structures
into the river (such as weirs), the morphology is fixed, and a relationship can be determined
using any of the other methods presented here. Once this relationship is established, a simple
water heigh measurement suffices to determine the discharge.

This technique is often used for continuous (automated) discharge measurements. A pressure
transducers or ultrasonic depth sensor is installed next to point where the relationship was
established. These devices are completely automatic and log water height at high temporal
resolution.

The relationships were defined for both the Huwilerbach (Appendix I) and the Upper
Rietholzbach (equation 2).

𝑣𝑣 = 0.0198 ∗ (𝑑𝑑 − 17.7)1.3597 (2)

v [m/s] Stream velocity

d [cm] Water depth as indicated by the scale

2.3 Float method

This method can be used to get a very rough estimate of the runoff. With a floating object the
surface velocity of the river is measured. The mean river velocity can then be approximated by:

𝑣𝑣𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 ≈ 0.85 ∙ 𝑣𝑣𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 (3)

vmean [m/s] Stream velocity

vsurface [m/s] Flow velocity at the surface

Multiplied with the cross-sectional area, this gives you the discharge.
2.4 Manning’s equation
This method allows an approximate estimation of runoff without performing any velocity or
runoff measurements. The empirical “Manning’s equation” assumes that there is a relation
between the river geometry and velocity. Manning’s equation states:

1 2/3 1/2
𝑣𝑣 = 𝑅𝑅 𝑆𝑆 (4)
𝑛𝑛

With:

n Manning’s roughness coefficient

R The hydraulic radius 𝐴𝐴/𝑃𝑃𝑤𝑤
A Cross-sectional area of the flow
Pw Wetted perimeter (see Fig. 2)
S the slope of the water surface

This, again, needs to be multiplied by the cross-sectional area.

The advantage of this method is that it is very simple and fast to use. The disadvantage lies in
the empirical nature of the equation, so that it should only be used to get an idea of the
magnitude of runoff.

2.5 Velocity-area method with current-meter

Principle:
Discharge is obtained by calculating the integral of the stream velocity 𝑣𝑣 over the cross-section
area of the flow A, where 𝑣𝑣 is measured perpendicular to the cross-section:

𝑄𝑄 = � 𝑣𝑣 𝑑𝑑𝑑𝑑 (5)

The velocity 𝑣𝑣 can be measured in discrete intervals along the cross-section by means of a
current-meter. A current-meter consists of a small propeller mounted on a pole that is
connected to a device which measures the frequency of the propeller rotation. This information
can be converted into a stream velocity using the provided look-up tables, which were
established during calibration experiments.

If we assume a purely laminar flow, the theory of fluid mechanics states that the stream
velocity is expected to vary vertically following a parabolic function because of the zero velocity
(no-slip condition) at the bottom of the stream bed. In case of a turbulent flow, we would get a
logarithmic function. For this reason, velocity should ideally be measured at several depths for
each interval along the river cross-section. Alternatively, a single measurement is best taken at
40% of the local depth (see Figure 1), which is generally the depth with the mean velocity.
Errors in the measurements can be introduced by random occurrences of turbulent eddies
during the establishment of the profile, by changes in water height or width of the stream.
During the experiment you should also take care to estimate the uncertainty of your
measurements.

Protocol:
1) Select an appropriate cross-section: ideally there should be no plants, no big stones and
the channel should be clearly defined.
2) Establish a point near the selected cross-section where the water level can be monitored
during the measurements.
3) Inspect and measure the cross-section. Make a clear graphical sketch of the profile.
4) Plan the measurement points and protocol based on the sketch of the profile (see Figure
1).
5) Perform the velocity measurements at the planned points. To do so, one should select a
propeller adapted to the stream velocity. Report the water level immediately before and
after each measurement, as well as any problems or noteworthy particularities
encountered during the measurements.
6) Repeat measurements several times.
7) Work out the measurements to obtain total discharge Q.

Figure 1: River cross-section and illustration of the compartments. (Source: www.montana.edu)

2.6 Velocity-area method using the dipping bar or „Tauchstab nach Jens“
The principle of this method is the same as for the current-meter. The dipping bar (Tauchstab)
can be used to measure the flow velocity based on the torque exerted on the bar by the
streamflow. A metallic horizontal stick can be interlocked in the dipping bar to counterbalance
the force of the flow. By varying the position of the horizontal stick, the torque it exerts on the
bar is changed. At the position where the bar is kept vertically, it balances the torque that is
created by the streamflow and the weight can be translated into a flow velocity using look-up
tables. The dipping bar can be used in water down to a depth of 60 cm. It should be noted that
contrary to the current-meter which measures velocity at a given depth, the force exerted on
the bar by the stream results from the integral of the flow velocity along the part of the bar that
is immersed in the water.

2.7 Tracer method (instantaneous injection)

Principle:
Tracers are often used in mountain streams with high turbulence and without clearly defined
channel. A good tracer needs to be non-reactive, besides it should be as harmless as possible
for the environment. During the field course, we will measure the discharge using sodium
chloride or salt (NaCl) as a tracer, by applying the instantaneous injection method. After having
defined an appropriate location to conduct the experiment, a known amount of tracer is
poured into the river in one go. This should be done upstream of where the measurements are
taken, at a distance sufficient enough from the first measurement location, so that the tracer
can get well mixed before reaching it.

The concentration of the tracer is then monitored by measuring the electric conductance in the
river. Ideally, the following conditions have to be fulfilled: (i) the background concentration of
the stream has to be known, (ii) it should not change during the experiment, (iii) the tracer is
totally intermixed with the water in the stream (no backwater effects), and (iv) there are no
water losses nor inputs between the injection and measurement locations. Using several
measurement sites and/or several measurement devices enables to check the results by
repeating the experiment.
The relationship between the electrical conductivity and the NaCl-concentration has to be
calibrated before starting the experiment. Then, one can plot the concentration as a function of
time. The resulting curve will typically look like that in Figure 2. Finally, the discharge can be
calculated using a mass balance equation:

𝑀𝑀
𝑄𝑄 = ∞ (6)
∫0 (𝑐𝑐(𝑡𝑡) − 𝑐𝑐0 ) 𝑑𝑑𝑑𝑑

where M is the injected mass of the tracer, 𝑐𝑐(𝑡𝑡) the concentration evolution over time and 𝑐𝑐0
the background concentration of the stream.

Figure 2: Typical evolution of the tracer

Integration concentration during a discharge
Tracer concentration

measurement using the instantaneous

Background
injection method
concentration

Time
Protocol:
1) Establish the relation between the electric conductivity and the NaCl-concentration. First,
fill water of the river into a bucket and measure the conductivity. Then add precisely
known amounts of salt to the bucket, stir well and measure the conductivity again. By
adding more and more salt to the bucket and measuring the conductivity you can
establish the calibration curve.
2) Select a location along the river with enough turbulence so that the tracer gets well mixed.
Estimate the distance 𝐿𝐿 which is necessary to intermix the tracer using e.g.:
Hull (1958): 𝐿𝐿 = 50 ∙ �𝑄𝑄
Fischer (1966): 𝐿𝐿 = 100 ∙ 𝑊𝑊 2 /𝑑𝑑
where 𝑄𝑄 is the discharge, 𝑊𝑊 the channel width and 𝑑𝑑 the water table depth.
3) Place the sensors for the electric conductivity at different locations downstream of the
injection point.
4) Inject the previously dissolved tracer instantaneously into the stream.
5) Measure the concentration evolution during the whole tracer transit.
6) Calculate Q.
3 Data analysis
Depending on the state of the streams, the discharge will be determined for the Rietholzbach
and Huwilerbach rivers using several of the methods described above. The results will be
reported, compared and discussed during the last two days of the field course. For this analysis,
you will obtain access to the automatic reference measurements from the Swiss Federal Office
for the Environment for both Rietholzbach and Huwilerbach.

During the data analysis you will investigate different aspects of runoff measurements, e.g.
year-to-year and seasonal variability of runoff, the influence of significant meteorological
events on runoff or a thorough uncertainty estimation of your measurements.
Appendix I – P-Q relationship Huwilerbach

W in Q in W in Q in W in Q in W in Q in W in Q in W in Q in
cm l/s cm l/s cm l/s cm l/s cm l/s cm l/s
0.0 0.00 10.0 1.63 20.0 7.44 30.0 25.71 40.0 61.93 50.0 119.82
0.5 0.00 10.5 1.84 20.5 7.99 30.5 27.06 40.5 64.28 50.5 123.34
1.0 0.01 11.0 2.07 21.0 8.58 31.0 28.45 41.0 66.68 51.0 126.93
1.5 0.01 11.5 2.31 21.5 9.20 31.5 29.90 41.5 69.15 51.5 130.58
2.0 0.03 12.0 2.57 22.0 9.86 32.0 31.39 42.0 71.66 52.0 134.29
2.5 0.05 12.5 2.69 22.5 10.56 32.5 32.92 42.5 74.24 52.5 138.06
3.0 0.08 13.0 2.83 23.0 11.29 33.0 34.51 43.0 76.87 53.0 141.90
3.5 0.12 13.5 2.98 23.5 12.06 33.5 36.14 43.5 79.55 53.5 145.80
4.0 0.17 14.0 3.17 24.0 12.86 34.0 37.82 44.0 82.30 54.0 149.77
4.5 0.22 14.5 3.37 24.5 13.71 34.5 39.55 44.5 85.10 54.5 153.81
5.0 0.29 15.0 3.60 25.0 14.59 35.0 41.32 45.0 87.96 55.0 157.90
5.5 0.37 15.5 3.85 25.5 15.52 35.5 43.15 45.5 90.87 55.5 162.07
6.0 0.46 16.0 4.14 26.0 16.48 36.0 45.03 46.0 93.85 56.0 166.30
6.5 0.56 16.5 4.44 26.5 17.49 36.5 46.96 46.5 96.88 56.5 170.60
7.0 0.67 17.0 4.78 27.0 18.53 37.0 48.94 47.0 99.98 57.0 174.96
7.5 0.79 17.5 5.15 27.5 19.62 37.5 50.97 47.5 103.13 57.5 179.39
8.0 0.93 18.0 5.54 28.0 20.75 38.0 53.06 48.0 106.35 58.0 183.89
8.5 1.09 18.5 5.97 28.5 21.92 38.5 55.19 48.5 109.63 58.5 188.45
9.0 1.25 19.0 6.42 29.0 23.14 39.0 57.38 49.0 112.96 59.0 193.09
9.5 1.44 19.5 6.91 29.5 24.40 39.5 59.63 49.5 116.36 59.5 197.79
Appendix II – Manning’s n for river channels

Manning's n for Channels (Chow, 1959).

Type of Channel and Description Minimum Normal Maximum

Natural streams - minor streams (top width at floodstage < 100 ft)

1. Main Channels

a. clean, straight, full stage, no rifts or deep pools 0.025 0.030 0.033

b. same as above, but more stones and weeds 0.030 0.035 0.040

c. clean, winding, some pools and shoals 0.033 0.040 0.045

d. same as above, but some weeds and stones 0.035 0.045 0.050
e. same as above, lower stages, more ineffective
0.040 0.048 0.055
slopes and sections
f. same as "d" with more stones 0.045 0.050 0.060

g. sluggish reaches, weedy, deep pools 0.050 0.070 0.080

h. very weedy reaches, deep pools, or floodways
0.075 0.100 0.150
with heavy stand of timber and underbrush
2. Mountain streams, no vegetation in channel, banks usually steep, trees and brush along
banks submerged at high stages

a. bottom: gravels, cobbles, and few boulders 0.030 0.040 0.050

b. bottom: cobbles with large boulders 0.040 0.050 0.070

Source: www.fsl.orst.edu/geowater/FX3/help/8_Hydraulic_Reference/Mannings_n_Tables.htm