LinearRegression Model for Estimating Runoff Volumes Using Paleoflood and Modern StreamflowRecords o

1
Linear-Regression Model for Estimating Runoff
Volumes Using Paleoflood and Modern
Streamflow-Records on Snowmelt-Dominated Rivers
Jon Mason The Hopi Tribe Water Resources
Program Jodi Norris University of Wyoming now
with National Park Service
2
Modern drought and tree-ring drought
reconstructions have focused attention on dry
extremes
3
Amplitude of wet extremes is often underestimated
by tree rings
4

• At some point additional water ? additional tree
  growth

5

• At some point additional water ? additional tree
  growth
• Need to use different techniques for estimating
  extreme wet events

6

• At some point additional water ? additional tree
  growth
• Need to use different techniques for estimating
  extreme wet events
• Paleoflood investigations

7
Paleoflood investigations have traditionally
focused on estimating the peak instantaneous
discharge

• Uses of Paleoflood Data
• Flood frequency studies
• Inundation studies
• Flood-plain mapping

8
Methods have long been established to calculate
instantaneous discharge from attainable
paleoflood parameters

• Channel characteristics
• High water marks
• Roughness or retardation coefficient

9
At times it would be useful to know the total
volume of paleoflood waters in addition to the
peak discharge

• Design parameters for new reservoir construction
• Planning the operation of existing reservoirs

10
Linear-Regression Model for Estimating Runoff
Volume Colorado River at Lees Ferry case example

• Good continuous streamflow record from 1922-2005
• Straightforward flood hydrology snowmelt
  dominated peaks
• Good modern flood and reconstructed paleoflood
  record
• (OConnor et al, 1994 Topping et al, 2003)
• Important stream in southwest
• Historically important floods

11
Good continuous streamflow record from 1922 - 2005

• Annual peak-instantaneous discharge
• Daily mean discharge

12
Straightforward flood hydrology
13
Good modern flood and reconstructed paleoflood
record
OConnor, J.E., Ely, L.L., Wohl, E.E., Stevens,
L.E., Melis, T.S., Kale, V.S., and Baker, V.R.,
1994, A 4500-year record of large floods on the
Colorado River in the Grand Canyon, Arizona
Journal of Geology, v. 102, p. 19. Topping,
David J., Schmidt, John C., Vierra, L. E., Jr.,
2003, Computation and analysis of the
instantaneous-discharge record for the Colorado
River at Lees Ferry, Arizona May 8, 1921,
through September 30, 2000 U.S. Geological
Survey Professional Paper 1677, 118 p.
14
Important stream in southwest
15
Historically Important Floods at Lees Ferry
Estimated Peak Previously from Topping
- Estimated Year et al, 2003 (cfs) Peak (cfs)
1884 210,000 300,000 1921 170,000
220,000 1983 -- 116,000 1,200 -
1,600 BP 300,000 490,000 USBR, 1990

16
Process of estimating event volume from daily
data and peak flows

• Chose 60 day event length because Bureau of
  Reclamation used that length in their PMF
  estimates
• Used a moving window to identify the peak 60 day
  volume associated with the peak flow
• Compared the volume with the peak instantaneous
  discharge to see if there was a relationship
• Looked at both pre-dam and post-dam flows

17
Pre-dam peak flows (water year 1922-1963)
200000
150000
100000
50000
Average 60-day flow, cfs
25000
15000
10000
8000
17000
25000
50000
75000
100000
200000
300000
Annual peak flow, cfs
18
Pre-dam peak flows (water year 1922-1963)
Post-dam peak flows (water year 1963-2005)
Regression, r-squared .90
200000
150000
100000
50000
Average 60-day flow, cfs
25000
15000
10000
8000
17000
25000
50000
75000
100000
200000
300000
Annual peak flow, cfs
19
Pre-dam peak flows (water year 1922-1963)
Regression, r-squared .90
200000
150000
100000
50000
Average 60-day flow, cfs
25000
15000
10000
8000
17000
25000
50000
75000
100000
200000
300000
Annual peak flow, cfs
20
Pre-dam peak flows (water year 1922-1963)
Post-dam peak flows (water year 1963-2005)
Regression, r-squared .90
Estimated 1,000-year flood, 300,000 cfs peak
200000
Volume of 60 day flow 22.1 million acre feet
150000
100000
50000
Average 60-day flow, cfs
25000
15000
10000
8000
17000
25000
50000
75000
100000
200000
300000
Annual peak flow, cfs
21
USBR Probable Maximum Flood for Glen Canyon Dam

• 100-year snowfall
• Upper Limit Design Rainstorm over San Juan Basin
• Upper Limit Design Rainstorm over Boulder Basin

Source U.S. Bureau of Reclamation, 1990,
Colorado River Basin Probable Maximum Flood
Hoover and Glen Canyon Dams US Department of
Interior, Washington DC
22
Maximum 60-day inflow volume Source
(cfs) (ac-ft) USBR 547,000
16,485,000 Paleoflood 300,000
22,100,000 reconstruction
23
Glen Canyon Dam-Lake Powell Stats

• Crest elevation 3715.0 ft
• Total reservoir storage 27,000,000 acre-ft
• Spillway capacity 276,000 cfs

24
What about upstream diversions?
25

• What about upstream diversions?
• Total storage capacity of 4 largest upstream
  reservoirs 7,000,000 ac-ft
• Minimum January vacant storage requirement of 4
  reservoirs 1,300,000 ac-ft
• Depletion 1,000,000 ac-ft

26
Wet periods tend to come in clusters
Woodhouse et al., 2006
27
Wet periods tend to come in clusters

• Treating an extreme event as a single year event
  may be thinking on too small a temporal scale

Woodhouse et al., 2006
28
Conclusions

• Understanding the full spectrum of natural
  climate variability is important
• Amplitude of wet extremes is often
  underestimated by tree rings
• Paleoflood investigations on snow-melt
  dominated streams can provide useful volume
  estimates of extreme wet events
• The choice of driving mechanism (snowmelt vs.
  rainfall) is important in determining probable
  maximum flood estimates.
• Wet periods are often multi-year events,
  preparing for wet periods in the context of a
  single year may not be good enough

29
As we prepare for drought, are we
ignoring/forgetting wet events 1983 was 24 years
ago.
30
As we prepare for drought, are we
ignoring/forgetting wet events 1983 was 24 years
ago.
Colorado agriculture commissioner and member of
the state water conservation board "It's always
been my feeling that you store water as high as
you can because you can always release it later.
May 2005 U.S. Water News Online
31
QUESTIONS?