Low flow characteristics

Question 1

General characteristics

Answer 1

 Intensity → minimum flow

 Duration → number of days, weeks or months

 Magnitude → deficit relative to a threshold e.g. for planning or regulation of reservoirs

 Minimum water level → for navigation or ecological aspects

 Frequency → risk regarding failure of water supply

 Occurrence time → season dependent damage e.g. agriculture

 Spatial extent → economic and social impact, possibilities for compensation

Question 2

Specific characteristics for low flows NQ:

Answer 2

 Low flow characteristics are estimated from stream flow data, either from flow duration curves (FDC) or from flow hydrographs (FH)

 Usually daily time steps are used, but also monthly or other time intervals will work

 When FH are used the elements are taken from a base time interval, which is usually 1 year, but also seasons can be considered

 For low flow statistics annual series are preferred for partial duration series because of the high persistence of low flows

Question 3

(A) Characteristics from flow duration curve (FDC)

Answer 3

 FDC are constructed by ranking the data, and for each value the frequency of exceedance is computed as rank / sample size (if data are ranked from largest to smallest value)

 Flow exceedance quantiles (percentiles) Qα can then be obtained from the FDC  discharge which is exceeded α - percent of the period of record e.g. Q90, Q80

Question 4

(B) Characteristics from flow hydrographs (FH)

Answer 4

a) n-day average annual minimum flow: NMQ(n-day) in m3/s
b) Duration of non-exceedance of a threshold QS
c) Deficit volume

Question 5

a) n-day average annual minimum flow: NMQ(n-day) in m3/s

Answer 5

Is derived from a series of annual minima of the n-day average flow, e.g. 1-day, 7-days, 30-days

• Selection of n depends on the specific purpose of the analysis (e.g. n = 1, 7, 14, 30 days are typical values)
• Remark 1: for n=1d observations may be severely inconsistent because of short-term impacts like regulation, withdrawals, etc.
• Remark 2: Statistical analysis requires consistent, homogeneous and independent data (see course “Statistical methods”)

Question 6

b) Duration of non-exceedance of a threshold QS

Answer 6

• Dmax in days → longest duration for which flow Q falls below threshold QS within base period
• ΣD in days → sum of all durations for all drought events for which flow Q falls below the threshold QS within base period

Question 7

c) Deficit volume

Answer 7

• Vmax in m3 → largest deficit volume for the drought events; accumulated difference between threshold QS and Q(t) over time t for base period
• ΣV in m3 → sum of all deficit volumes for all drought events within the base period

Question 8

Low flow characteristics from flow hydrograph

Answer 8

max D
Sum D
max V
Sum V

Question 9

Frequency analysis

Answer 9

 Low flow frequency analysis is methodologically comparable to flood frequency analysis

 For design low flows also the return period T is used. However,mcompared to floods (cp. Eq. 6.1) the events of interest have a high exceedance probability. So it follows:
TDraught = 1/Pne = 1/(1-Pe)

Question 10

Frequency analysis: Typical PD’s

Answer 10

 Logarithmic Normal-distribution (LN2)
 (Log.) Extreme value distribution type III (Weibull) (LE3)
 (Log.) Pearson III distribution (LP3)
 Generalised extreme value (GEV)