Methods of stream flow measurement

Bucket method

The whole stream is diverted into a container of known volume and the filling time is measured. $Q = V/t$. It is accurate and used for very small flows (below about 10 L/s).

Float (velocity-area) method

Mark a straight length of channel (say 10–20 m), measure the time $t$ for a float to travel it, and the average cross-sectional area $A$. Surface velocity $v_s = L/t$; mean velocity $v = k\,v_s$ with $k \approx 0.8$–0.85. $Q = A\,v$. It is simple but approximate.

Weir method

A sharp-edged rectangular or V-notch weir is built across the stream and the head $H$ over the crest is measured. For a rectangular weir with end contractions $Q = 1.84\,(L - 0.1nH)H^{3/2}$ and for a 90° V-notch $Q = 1.4H^{5/2}$ (m³/s, metres). Accurate for small and medium streams.

Salt dilution method

A known mass of salt is dissolved and poured into the stream at one point, and the conductivity is recorded downstream where it is fully mixed. The area under the conductivity-time curve gives $Q$. It suits turbulent mountain streams where other methods are difficult.

Current meter

A propeller meter measures velocity at several points (usually at 0.6 of depth) in verticals across the section. $Q = \sum v_i A_i$ (velocity-area method). It gives the best accuracy but needs equipment.

Flow duration curve (FDC)

It is a plot of flow against the percentage of time (or days) the flow is equalled or exceeded, constructed from daily or monthly flow records sorted in descending order.

 Q
 |\
 | \___
 |     \____
 |          \______
 +------------------ % time
 0       50      100

For micro hydro the design flow is normally taken at the flow available 90–95 % of the time (dry season) so that power is dependable, and the area under the curve gives the energy available.