Influence of Hydroclimate on Characteristics of Seasonal
Hydrograph Evolution in Snowmelt-Dominated River Systems
Margaret Matter
PhD Candidate
Department of Civil Engineering
Colorado State University
1. Introduction
The Upper Colorado River Basin (UCRB) is a major source
of water supply for seven states, but rapid development,
expanding types of demands, variable & limited water supply, &
limited accuracy & short lead time of water supply forecasts
require greater efficiency & effectiveness of water resource
planning & management. Earlier & more accurate forecasts
could help reduce uncertainty & risk in decision-making &
lengthen the period for planning more efficient & effective water
use strategies.
The goal of this research is to extend the lead time for
snowmelt runoff hydrograph estimation by 4-6 months (from
spring to the preceding fall), & at the same time increase the
accuracy of snowmelt runoff estimates in the UCRB.
2. Background
Snow comprises 50-80 percent of water supply in the CRB
[9], so Apr-Sep streamflow is predicted based on Dec-Feb (or
Mar) precipitation, when snowpack is most extensive [4]
Yet, total precipitation over a longer period of time, OctMar, is actually positively correlated with Apr-Sep flow [2,
5, 6].
This suggests that precipitation & temperature signals in
the fall are related to the same hydroclimatic conditions
that persist through winter and into spring.
AMO, PDO, & ENSO are major climate modes in the CRB.
ENSO is phase locked, or evolves, with the seasons [8],
Setting up in about Jul, & by Oct, conditions tend to
persist & can be predicted several months in advance [5, 3]
Influencing precipitation [e.g., 1, 7] & streamflow [e.g., 2, 6]
in Western U.S. rivers between fall & spring.
Yet predictability in the CRB is modulated by atmospheric
circulation, orographic effects, as well as climate modes, such
as PDO & AMO, that evolve over longer timescales [e.g., 5,
3], & are important drivers of climate signals with or without
contemporaneous ENSO extremes [3].
Thus, it is assumed that combined effects of temperature &
precipitation signals associated with major climate modes in
the CRB are nearly constant from fall to spring, so that
Beginning in fall, hydrographs may be expected to reflect
influences of precipitation & temperature associated with
hydroclimatic conditions that will drive snowpack
development through winter & early spring.
3. Objectives
The goal of this research is to:
Extend the lead time for snowmelt runoff hydrograph
estimation by 4-6 months (from spring to the preceding
fall), and at the same time
Increase the accuracy of snowmelt runoff estimates in
the UCRB.
5. Results (preliminary)
Gunnison River near Gunnison, CO.
Magnitude & Timing of F/EW 1/3 Flow Volume
We hypothesize that UCRB snowpack accumulation &
melt are driven by large scale climate modes, including
ENSO, PDO & AMO that establish & are detectable by
fall & persist into the following spring.
Historical records from largely unregulated reaches &
undeveloped time periods will be used to test this
Preliminary
hypothesis.
results
Statistical relations between fall-to-early presented
spring
hydrograph characteristics & hydroclimatic conditions
can be used to predict runoff from fall/winter
precipitation & snowpack accumulation.
Fig. 5-A Magnitude of F/EW 1/3-flow volume is
higher and occurs earlier in dry years than in wet
years, indicating that F/EW flow volume shifts
earlier and is lower in dry compared to wet years.
Magnitude & Timing of LW/ES 1/3 Flow Volume
4. Data & Methods
____Data for
Period of Record (Water Years)
Seasonal Analysis
Unimpaired
Validation
Gunnison River
nr Gunnison, CO
(09114500)
1911 – 1928
1944 - 1973
(18 yrs)
(30 yrs)
East River nr
Almont, CO
( 09112500)
1911 – 1912;
1917 - 1920;
1936 – 1949.
(20 yrs)
1950 – 1979
(30 yrs)
1921 – 1938
(20 yrs)
1941 – 1970
(30 yrs)
Colorado River at
Lees Ferry, AZ
(09380000)
Fig. 5-B Magnitude of LW/ES 1/3-flow volume is
lower & occurs earlier in dry years than in wet
years, indicating that snowmelt rate is higher in so
the flow volume shifts earlier, but the volume is
lower compared to wet years.
Patterns in Flow Variability ~ ICDF/DCDF Ratios
Methods
Magnitude & Timing of 1/3 & 1/2 Flow Volumes for
Fall/Early Winter (F/EW) & Late Winter/Early Spring
(LW/ES) Seasons.
Hydrograph Slope for F/EW & LW/ES Seasons
(Indicates rate of change over seasons).
Ratios of Increasing to Decreasing Daily Changes in
Flow (ICDF/DCDF); daily ratios over the water year
(Indicates trends in flow variability).
Fig. 5-C Average ICDF/DCDF for Nov increases as
hydroclimate becomes wetter (ie, larger ABY),
indicating that flow variability in Nov is higher in
wet years than in dry years.
4. Discussion
Results show that timing of fall/early winter flow volumes
and trends in early fall flow fluctuations (i.e., Nov
ICDF/DCDF) both exhibit fairly strong relationships with
ABY, or magnitude of the following spring snowmelt runoff
(Rho = 0.63 and 0.71, respectively). In dry conditions, timing
of seasonal flow volumes is earlier in F/EW and LW/ES, but
flow volume magnitude is larger in F/EW (Fig. 5-A) and
smaller in LW/ES (Fig. 5-B) compared to wet conditions.
Warmer temperatures accompanying dry hydroclimatic
conditions may result in precipitation falling as rain rather
than snow in the fall, and during the winter, less snowpack
development. Trends in flow variability (Fig. 5-C) show
greater variability (e.g., higher magnitude of increases in
flow) during early fall in wetter years than in drier years,
which may reflect shifts in temporal patterns of precipitation
distribution between wet and dry hydroclimatic conditions.
5. Conclusions
Thus, preliminary results for the UCRB show that
streamflow records for largely unimpaired basin conditions
exhibit characteristic features in hydrograph evolution
between fall and early spring that are in response to
precipitation and temperature signals associated with
hydroclimatic conditions that establish by fall and persist into
spring, influencing snowpack development and early
snowmelt. These results have application in skilled water
supply forecasting to improve accuracy and increase lead
time up to 4-6 months prior to current April forecasts.
6. Funding Source
Upper Colorado Region ~ U.S. Bureau of Reclamation
7. References
1. Cayan. D.R. and Webb, R.H 1992. El Nino/Southern Oscillation and Streamflow in the
Western U.S. In H.F. Diaz and V. Markgraf (eds). Historical and Paleoclimatic Aspects of the
Southern Oscillation. Cambridge University Press, 29-68.
2.Cayan, D.R., Redmond, K.T., and Riddle, L.G. 1999. ENSO and Hydrologic Extremes in the
Western United States. American Meteorological Society 12: 2881-2893.
3. Gershunov, A. and Cayan, D.R. 2003. Heavy Daily Precipitation Frequency over the Contiguous
United States: Sources of Climatic Variability and Seasonal Predictability. J. Climate 16: 2752-2765.
4. Gutler, D.S. and Rosen, R.D. 1992. Interannual Variability of Wintertime Snow Cover across the
Northern Hemisphere. J. Climate 5: 1441-1447.
5. Gutzler, D.S., Kann, D.M., and Thornbrugh, C. 2002. Modulation of ENSO-Based LongLead Outlooks of Southwestern U.S. Winter Precipitation by the Pacific Decadal Oscillation.
Weather and Forecasting 17: 1164-1172.
6. Hidalgo, H.G. and Dracup, J.A. 2003. ENSO and PDO Effects on Hydroclimatic Variation of
the Upper Colorado River Basin. J. Hydrometeorology 4: 5-23.
7. McCabe, G.J. and Dettinger, M.D. 1999. Decadal variations in the strength of ENSO
Teleconnections with precipitation in the western United States. Internat. J. Climatology 19:
1399-1410.
8. Neelin, J.D., Jin, F., and Syu, H. 2000. Variations in ENSO Phase Locking. J. Climate 13: 2570-2590.
9. Woodhouse, C.A. 2003. A 431-Yr Reconstruction of Western Colorado Snowpack from Tree Rings.
J. Climate 16: 1551-1561.