The Evolution of Climate Change Impact Studies on Hydrology and Water Resources in California

Abstract

Potential global climate change impacts on hydrology pose a threat to water resources systems throughout the world. The California
water system is especially vulnerable to global warming due to its dependence on mountain snow accumulation and the snowmelt
process. Since 1983, more than 60 studies have investigated climate change impacts on hydrology and water resources in California.
These studies can be categorized in three major fields: (1) Studies of historical trends of streamflow and snowpack in order
to determine if there is any evidence of climate change in the geophysical record; (2) Studies of potential future predicted
effects of climate change on streamflow and; (3) Studies that use those predicted changes in natural runoff to determine their
economic, ecologic, or institutional impacts. In this paper we review these studies with an emphasis on methodological procedures.
We provide for each category of studies a summary of significant conclusions and potential areas for future work.

Full text

The evolution of climate change impact studies
on hydrology and water resources in California
S. Vicuna &J. A. Dracup
Received: 28 July 2005 /Accepted: 5 September 2006 / Published online: 10 February 2007
#Springer Science + Business Media B.V. 2007
Abstract Potential global climate change impacts on hydrology pose a threat to water
resources systems throughout the world. The California water system is especially
vulnerable to global warming due to its dependence on mountain snow accumulation and
the snowmelt process. Since 1983, more than 60 studies have investigated climate change
impacts on hydrology and water resources in California. These studies can be categorized in
three major fields: (1) Studies of historical trends of streamflow and snowpack in order to
determine if there is any evidence of climate change in the geophysical record; (2) Studies
of potential future predicted effects of climate change on streamflow and; (3) Studies that
use those predicted changes in natural runoff to determine their economic, ecologic, or
institutional impacts. In this paper we review these studies with an emphasis on
methodological procedures. We provide for each category of studies a summary of
significant conclusions and potential areas for future work.
1 Introduction
The latest Intergovernmental Panel on Climate Change (IPCC) report reaffirms that the
climate is changing in ways that cannot be accounted for by natural variability and that
“global warming”is occurring (IPCC 2001). This global warming is likely to have
significant impacts on the hydrologic cycle, affecting water resources systems (Arnell 1999;
IPCC 2001). These impacts will vary for different regions of the earth. Regions that have a
Climatic Change (2007) 82:327–350
DOI 10.1007/s10584-006-9207-2
S. Vicuna (*)
Department of Civil & Environmental Engineering, University of California, 612 Davis Hall,
Mail Code 1710, Berkeley, CA 94720-1710, USA
e-mail: svicuna@berkeley.edu
J. A. Dracup
Department of Civil & Environmental Engineering, University of California, 625 Davis Hall,
Mail Code 1710, Berkeley, CA 94720-1710, USA
e-mail: dracup@ce.berkeley.edu

large fraction of runoff driven by snowmelt would be especially susceptible to changes in
temperature, because temperature determines the fraction of precipitation that falls as snow
and is the most important factor determining the timing of snowmelt runoff. The California
water system is highly dependent on snow storage and hence has great potential to suffer
from the effects of global warming. This potential risk has spawned over 60 studies in the
last 20 years in the field of climate change and its resulting impacts on water resources in
California. The following paper presents a review of these studies with the objective of
understanding how the field has evolved over time, what has been accomplished to date,
and what remains to be done. This review builds on previous assessments of the impacts of
climate change on Californian water resources (Gleick and Chalecki 1999; Wilkinson et al.
2002; Roos 2003; Kiparsky and Gleick 2003). Studies on the Colorado River were not
included in this review because, although changes in its streamflow affect directly
California, the Colorado River is a different hydrologic region than the rivers that are in the
State of California.
This paper is organized to present the three main categories representing the field of
climate change impacts on water resources. The first section covers studies that ask: what
evidence of climate change do we see in historical streamflows to date? The second
category covers studies that are focused on prediction of the potential future effect of
climate change on streamflow. Finally, the third set of studies uses those predicted changes
in natural runoff to predict their economic, ecologic, or institutional impacts. Presented in
Fig. 1is a timeline showing when this over 60 studies were published over the last two
decades. The timeline shows an early interest in the field that peaked in the early 1990s
followed by a slight decay in the mid-1990s and a rampantly increasing interest by the end
of the century that appears to be ongoing.
2 The historical record: is there evidence of the effects of climate change
on California hydrology?
It has been almost two decades since Roos (1987,1991) first brought attention to changes
that are occurring in California’s streamflow patterns. Roos looked specifically at the
Sacramento basin and determined that the seasonal fraction of runoff flowing through the
snowmelt/spring season (from April to July) was decreasing throughout the twentieth
century (see Fig. 2). This same behavior was confirmed by other studies using more
complex statistical measures and extended to other basins in the State (Fox et al. 1990;
Wahl 1991; Aguado et al. 1992; Pupacko 1993; Dettinger and Cayan 1995; Shelton and
Fridirici 1997; Shelton 1998; Freeman 2002) (see Fig. 3for a map showing which basins
have been studied).
Using seasonal fractional runoff as a measure of changes in streamflow patterns could be
misleading because it may either imply changes in the spring runoff or changes in the other
seasons or both (Wahl 1991). Avoiding this uncertainty, Cayan et al. (2001), Stewart et al.
(2004,2005) and Regonda et al. (2005) each present a different approach to measure
changes in streamflow patterns. Cayan et al. (2001) considered the ‘spring pulse,’defined
as the day when cumulative departure of daily streamflow from mean is the most negative.
Their study, covering several basins in the Sierra Nevada Mountains, correlated the spring
pulse with other measures of spring onset (e.g. flower blooming). Their results showed an
organized earlier spring onset. Stewart et al. (2004,2005) obtained a similar result using
flow-weighted timing, or ‘center of mass’of streamflow, as the metric to determine runoff
328 Climatic Change (2007) 82:327–350

timing (see Fig. 4). Regonda et al. (2005) used the “date (Julian day) on which 50% of the
water-year flow is equaled or exceeded.”All of these studies focused on different basins
throughout the Western U.S., finding a shift towards earlier snowmelt. However, results for
the Sierra Nevada basins were less statistically significant than in other regions, such as in
the Pacific Northwest.
0%
10%
20%
30%
40%
50%
60%
70%
1906
1912
1918
1924
1930
1936
1942
1948
1954
1960
1966
1972
1978
1984
1990
1996
2002
Percent of Annual Runoff
Fig. 2 Trend of April–July as percent of water year runoff for Sacramento four basin index (modified from
Roos 1991)
0
1
2
3
4
5
6
7
8
9
10
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
Year
Number of studies
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
Total
Fig. 1 Historical record of climate change and water resources/hydrology related studies in CA
Climatic Change (2007) 82:327–350 329

Several factors have been postulated as causing the trend in the decline of fractional
spring streamflow, including increases in winter precipitation and increases in spring
temperatures. The latter factor was found to explain earlier spring runoff by Cayan et al.
(2001), Stewart et al. (2004,2005) and Regonda et al. (2005). The increase in spring
temperature factor may also explain why the magnitude of the trends depends on the
altitude of the basin. Basins located well above the ‘freezing line’(above 2,000 m) show
less pronounced changes in runoff patterns than mid-altitude basins (1,000–2,000 m)
because they are less sensitive to changes in temperature, as shown by studies comparing
trends for different basins (Roos 1987,1991; Aguado et al. 1992; Pupacko 1993; Dettinger
and Cayan 1995; Mote et al. 2005; Regonda et al. 2005; Stewart et al. 2005). This may
explain why lower-in-altitude basins located in the northern Sierra Nevada Mountains have
experienced more pronounced trends than their higher-in-altitude counterpart basins in
southern Sierra Nevada. Whether these changes in runoff patterns are a signature of a
changing climate in California remains uncertain. Although some authors have suggested
that these trends are due to climate change (e.g Pupacko 1993; Shelton and Fridirici 1997;
Shelton 1998; Stewart et al. 2005), other have been more cautious, suggesting the
possibility that the length of the historical record only shows one realization of a low
frequency process such as the Pacific Decadal Oscillation (e.g. Aguado et al. 1992;
Dettinger and Cayan 1995; Cayan et al. 2001).
Another geophysical factor that has been studied for historical trends is the snowpack
level measured as the Snow Water Equivalent (SWE). Studies show a decreasing trend in
the latter half of the twentieth century in SWE for most snowmelt basins throughout the
Western U.S. that is consistent with temperature increases for the same period (Mote et al.
2005; Regonda et al. 2005).
Fig. 3 Spatial coverage
(black dots) of streamflow
trend studies in California back-
ground map from: http://faculty.
sierracollege.edu/ccox/images/
maps/CA_rivers _map.jpg
330 Climatic Change (2007) 82:327–350

It is the author’s opinion that an area of future research regarding the historic evidence
of climate change in the geophysical record is the interaction between vegetation, surface
runoff and the timing of streamflow in California. Some areas in the California Sierra
Nevada have experienced changes in land cover due mainly to deforestation. These
changes in land cover might have affected the albedo and the soil and vegetation
functions of waterheds and thus streamflow runoff patterns. We could not find any study
that considered trends in land cover vegetation to isolate this factor in streamflow trend
analysis.
Summarized in Table 1are the studies to date on historical trends of streamflow and
snowpack level data in California. We classified the studies in the table according to the
basins studied and measures used in the trend analysis.
3 Predicted future impacts on Californian natural streamflow due to climate change
3.1 Methodology
The methodologies used to assess climate change impacts on hydrology and water
resources systems have been addressed by Gleick (1986,1989) and Wood et al. (1997).
There are two major steps involved in this process: (1) Determining changes in temperature,
precipitation and other climatologic variables such as evapotranspiration and; (2) Using
these changes to determine the resulting changes in streamflow. See Fig. 5for a schematic
of the methodology.
Fig 4 Observed changes in the timing of the center of mass of flow (from Stewart et al. 2004). Note: Color
of the symbols corresponds to a given magnitude of the linear trend, which is given in terms of the
corresponding overall shift (days) for the 1948–2000 historical period. Larger circles indicate statistically
significant trends at the 90% level, smaller circles correspond to trend that do not meet the significance
thresholds at the 90% confidence level
Climatic Change (2007) 82:327–350 331

Table 1 Summary of studies on the historic trend of streamflow and snowpack data in California
Study Basin (length of record) Trend found (at significant level unless explicitly expressed) and possible cause examined
Roos 1987 Sacramento River Index (1906–1986);
Combined Kings and San Joaquin (1901–1986)
Reduction in fractional spring (snowmelt) runoff for both basins. More dramatic for
Sacramento Index which is lower than Kings-San Joaquin. Not significance analysis
performed. No cause explored.
Fox et al. 1990 Actual and unimpaired Delta outflow (1920–1986) Increase in July through November flows and decrease in April and May flows.
Increases in good agreement with increases in precipitation during the flood season.
Roos 1991 Sacramento River Index (1906–1990); Combined
Kings and San Joaquin (1901–1990)
Reduction in fractional snowmelt runoff for both basins. More dramatic for Sacramento
Index which is lower than Kings-San Joaquin. Not significance analysis performed.
No cause explored.
Wahl 1991 Several basins in Western U.S. including California’s
Sierra Nevada and Cascades (median record length
of 60 years finishing in 1989).
Reduction in fractional snowmelt runoff for 10 basins in CA. No cause explored
Aguado et al. 1992 Several basins in California, including Sierra Nevada
and Cascades (1948–1986).
Reduction in fractional snowmelt runoff. Possible causes (identified by regression
models): increase winter precipitation and spring temperature. Results more pronounced
for high altitude basins.
Pupacko 1993 American and Carson rivers (1939–1990) Increasing March and winter month’s streamflow. Possible cause: increase minimum
winter temperature. More pronounced trend for lower altitude basin (American)
Dettinger and
Cayan 1995
Smith, American, San Joaquin, Carson and Merced
rivers (1948–1991). 8-river index (1906–1990)
Reduction in fractional snowmelt runoff. Trend seems to have begun in the late 1940s
and is most pronounced for mid-altitude river basins (1000–2000m). Possible cause:
increase winter temperature.
Shelton and
Fridirici 1997
Unimpaired inflow into the Delta (1921–1992) Increasing March, September and October fractional runoff. Decrease in April and May
fractional runoff. No cause explored.
Shelton 1998 Unimpaired Sacramento river (1921–1994) Reduction in fractional snowmelt runoff. No cause explored.
332 Climatic Change (2007) 82:327–350

Cayan et al 2001 Several basins in western U.S. including California’s
Sierra Nevada (1948–1995).
There is a trend toward earlier snowmelt timing since the mid-1970s. Snowmelt timing is
measured as the “spring pulse,”which corresponds to the day when cumulative departure
from mean is most negative. This measure correlates well with other spring onset
measures (e.g. flower blooming). Possible causes are warm episodes associated with
larger-scale atmospheric conditions.
Freeman 2002 North Fork Feather and Yuba River (1952–1999) Increase in March runoff and decrease in May runoff. Decrease in snowpack in Lake
Spaulidng course. No cause explored.
Stewart et al. 2004 Several snowmelt dominated basins (279) in Western
U.S. including California’s Sierra Nevada (1948–2000).
A large majority of basins exhibit significant trends toward earlier snowmelt timing
during the last few decades. Results not very significant in Californian basins. Snowmelt
timing is measured as the “center of timing,”that corresponds to the centroid of the
hydrograph. The principal cause is an increasing spring temperature.
Mote et al. 2005 Several basins in western U.S. and Canada including
California’s Sierra Nevada and Cascades (1950–1997)
Decreasing springtime SWE in northern California. SWE decreases are related to increases
in temperature. Increasing SWE in California’s southern Sierra Nevada. SWE increases
occurred in high-altitude basins or in basins with high increases in precipitation.
Regonda et al. 2005 Several mountain basins in western U.S. including
California’s Sierra Nevada (1950–1999)
Both earlier (at mid-altitude basins) and later (at high-altitude basins, normally over
2,500 m) timing for peak flows (both not statistically significant in Californian basins).
Timing of snowmelt is measured as the date on which 50% of the water-year flow is
equaled or exceeded. Decreasing levels of SWE, especially for low basins. Causes:
temperature increases in low basins.
Stewart et al. 2005 Several snowmelt and non-snowmelt dominated basins
in western U.S., Canada and Alaska including
California’s Sierra Nevada, Cascades and southern
California (1948–2000).
A large majority of snowmelt basins have exhibited significant trends toward earlier
snowmelt timing (measured both as “spring pulse”and “center of timing, CT”) and
decreasing snowmelt fractional runoff. Trends are less pronounced for high-altitude
basins. The major cause of the timing trend has been winter and spring temperature
increases. Indicators of low frequency natural fluctuations, such as PDO, are not
sufficient to fully explain the observed timing changes. For cases of non-snowmelt-
dominated basins there is a trend toward later snowmelt timing (as CT).
Climatic Change (2007) 82:327–350 333

3.1.1 Determining changes in temperature and precipitation
There are two alternative approaches to determine the changes in temperature and
precipitation associated with climate change. The more simple and direct approach is to
develop hypothetical scenarios of changes in temperature and precipitation. The second
approach obtains these changes in temperature and precipitation as output from a ‘general
circulation model’(GCM).
Proposed hypothetical climate scenarios in these studies include changes in temperature
covering the plausible range for the twenty-first century (e.g. +2 to +5°C). Since projections
of precipitation are less consistent and include both increases and decreases, hypothetical
scenarios are selected within this range. The “hypothetical scenario”approach was used in
the earliest studies (Revelle and Waggoner 1983; Gleick 1987; Jeton et al. 1996) and a
recent study by Miller et al. (2003). However, it is important to mention that of all these
studies, only Revelle and Waggoner (1983) relied solely on the ‘hypothetical scenarios’
approach. The other studies also included output from GCMs. The advantage of the
hypothetical scenario approach is its simplicity in representing a wide range of alternative
scenarios. These scenarios can be used to determine the sensitivity of a particular basin to
changes in climate conditions and whether systems can adapt to the range of potential
impacts. However, it is difficult to assign probabilities associated with these scenarios so
that they can be used for making policy decisions in water resources management.
GCMs have evolved over the past 50 years since their original conception by Phillips
(1956). Currently GCMs are representations of the coupled atmosphere-land-ocean-ice
systems and their interactions. These models provide information on the response of the
atmosphere to different scenarios of greenhouse gas concentrations (IPCC 2001). Most
Physical Hydrologic
Models
Statistical Hydrologic
Models
Changes in temperature, precipitation and other
hydroclimatic factors
Hypothetic Scenarios. Arbitrary
changes in temp and
precipitation
Changes in runoff
C
CO
O
2
2
E
Em
mi
is
ss
si
io
on
n
S
Sc
ce
en
na
ar
ri
io
os
s
General Circulation Models
(GCMs)
Downscaling to
Regional Scale
H
Hi
is
st
to
or
ri
ic
ca
al
l
s
st
tr
re
ea
am
mf
fl
lo
ow
w,
,
p
pr
re
ec
ci
ip
pi
it
ta
at
ti
io
on
n
a
an
nd
d
t
te
em
mp
pe
er
ra
at
tu
ur
re
e
d
da
at
ta
a
Fig. 5 Methodology to evaluate hydrologic implications of climate change
334 Climatic Change (2007) 82:327–350

climate change studies of California used GCMs’output to represent climate change
conditions (Gleick 1987; Lettenmaier et al. 1988; Lettenmaier and Gan 1990; Tsuang and
Dracup 1991; Leung and Ghan 1999; McCabe and Wolock 1999; Miller et al. 1999;
Wolock and McCabe 1999; Hay et al. 2000; Miller and Kim 2000; Wilby and Dettinger
2000; Carpenter and Georgakakos 2001; Kim 2001; Kim et al. 2002; Knowles and Cayan
2002; Snyder et al. 2002; Huber-Lee et al. 2003; Miller et al. 2003; Van Rheenen et al.
2003; Dettinger 2004; Dettinger et al. 2004; Hayhoe et al. 2004; Knowles and Cayan 2004;
Leung et al. 2004; Stewart et al. 2004; Van Rheenen et al. 2004; Dettinger 2005;
Georgakakos et al. 2005; Kim 2005; Maurer and Duffy 2005). Several factors distinguish
these studies, the most important being the different GCMs used, different downscaling
methodology and the methods used to characterize uncertainty.
With respect to the choice of GCM output, it is interesting to note how the most
prominent GCMs used in earlier studies, (the models of the Goddard Institute for Space
Studies (GISS), and the Oregon State University (OSU)) were replaced by a different set of
GCMs in later studies. The current GCMs prominently used are the UK’s Hadley Center
(HadCM2 and HadCM3), the NCAR (CCM3 and PCM) and the Canadian Centre for
Climate Modeling and Analysis Canadian (CGCM1) models. According to experts in this
field, the reason behind this evolution in the GCMs used for climate change studies is
“computer resources”rather than a technical issue (D. Cayan, I. Fung and D Lettenmaier,
personal communication).
One major limitation to using GCM output data is that the spatial and temporal
resolution does not match the resolution needed for hydrologic models. For example, the
spatial resolution of GCMs (about 200 km) is too coarse to resolve complex orography and
sub-grid scale processes such as convective precipitation, which are of major relevance for
mountainous terrains like the California Sierra Nevada Mountains (Wilby and Dettinger
2000). Several methods have been developed to “downscale”or transfer GCM output to
surface variables at the river basin scale. The most common are: (a) delta/ratio
(perturbation) methods; (b) stochastic/statistical downscaling and (c) dynamic downscaling
or nested models (Wood et al. 1997). It’s interesting to note how the evolution of this field
of study in California has followed the evolution of the downscaling methodologies. Earlier
studies did not consider downscaling GCM output but instead used raw GCM output. The
delta/ratio method, that basically modifies the historic climate data time series (usually
temperature and precipitation) by applying a delta (ratio) obtained by comparing the GCM
output for both climate altered and controlled scenarios, then became the preferred method.
Recently, studies have tended to use more complex downscaling methods (either statistical
or dynamic). Improved downscaling methodologies have provided means of expanding the
temporal aspect of the analysis. Earlier studies relied solely on monthly perturbations of
historical time series, while later studies have explored the derivation of a ‘new’(not based
on historical data) time series of climate variables (either daily or monthly), including the
analysis of changes in the frequency of extreme events (e.g. flood or droughts) and in
interannual variability (e.g. Miller et al. 2003; Dettinger et al. 2004; Hayhoe et al. 2004;
Stewart et al. 2004; Van Rheenen et al. 2004; Kim 2005; Maurer and Duffy 2005).
GCMs consistently project an increase of temperature for California of between 2.5
and 9°C by 2100 (Dettinger 2004,2005). However, the GCMs exhibit greater variability in
their precipitation projections (see Fig. 6) (Dettinger 2004,2005). This inconsistency adds
a high degree of uncertainty to making water resources management decisions. Recent
studies have tried to tackle this problem using different approaches. Some have
considered using a various (and differing) GCMs’outputs, under multiple greenhouse gas
Climatic Change (2007) 82:327–350 335

emission scenarios (e.g. Miller et al. 2003;Hayhoeetal.2004; Leung et al. 2004). With
that approach they intend to bracket plausible changes, but they still do not explicitly
account for the uncertainty in projections.
Dettinger (2004,2005) focused explicitly on the uncertainty related to GCM projections
in California. He developed “projection distribution functions”(pdfs) based on a
resampling technique of 18 available projection scenarios (six models times three emission
scenarios each). Another approach is that by Maurer and Duffy (2005), whose study of 10
GCMs allowed them to assess statistically significant projections across all GCMs. The
explicit consideration of uncertainty, that is embedded in climate change projections, can be
taken into account by assigning probabilities to the different scenarios. The authors believe
that this is an important approach, as it will lead to better-informed decisions by the water
resource community in California.
Fig. 6 Ensemble of historical and future temperature and precipitation projections from six coupled ocean-
atmosphere general-circulation models (from Dettinger 2004,2005)
336 Climatic Change (2007) 82:327–350

3.1.2 Prediction of changes in natural runoff
Perturbed series of climatic data (mainly temperature and precipitation) are used to drive
hydrologic models to predict changes in streamflow runoff (see Fig. 5). There are two
alternative approaches: using either statistically or physically based hydrologic models.
Statistical hydrologic models determine potential future runoff values according to the
characteristics of the historical record. A statistical model used for climate change impact
assessment might determine (through regression or observational analysis) the relation
between streamflow runoff and climate variables such as temperature and precipitation. The
first study of the impacts of climate change on Californian water resources used a statistical
model that considered annual values of streamflow, temperature and precipitation (Revelle
and Waggoner 1983). Most recent studies use monthly time scales to determine changes in
the total annual volume as well as changes in the seasonal pattern (Cayan and Riddle 1993;
Duel 1994; Risbey and Entekhabi 1996; Peterson et al. 2000; Stewart et al. 2004). One
major limitation of these statistical approaches is that they do not incorporate the physical
mechanisms and processes that determine basin response to climate forcing. Also, the
statistical approach is limited to those levels of forcing that have historically occurred.
Physically-based hydrologic models have been the preferred tool to assess the impacts of
climate change in the California hydrology. Some researchers developed their own models
for their analysis (e.g. Gleick 1987; Tsuang and Dracup 1991; McCabe and Wolock 1999;
Wolock and McCabe 1999). However, most studies used previously developed physically
based models, including: the USGS Precipitation-Runoff Modeling System (PRMS); the
U.S. National Weather Service River Forecast System Sacramento Soil Moisture Accounting
and Anderson Snow Models (SAC-SMA); the Variable Infiltration Capacity (VIC) model;
and the Water Evaluation and Planning System (WEAP) model. The spatial parameteri-
zation of these models ranges from distributed to lumped, and the spatial resolution ranges
from regional (entire U.S. West Coast) down to subbasin.
Although some models include-state-of-the-art levels of complexity, there are important
factors that have not yet been taken into account. The most important of these missing
factors, in the author’s opinion, is the dynamic interaction between vegetated land cover
and climatic variables. Also, factors that affect watershed responses have been treated as
static parameters in the models, rather than changing with time or with future climate.
3.2 Predicted climate change impacts on Californian natural streamflows: most
significant results
Although the studies assessing the impacts of climate change on California hydrology
have differed in their methodological approach, their results tend to agree in certain
critical areas. Results consistently show that increasing temperatures associated with
climate change will impact Californian hydrology by changing the seasonal streamflow
pattern to an earlier (and shorter) spring snowmelt and an increase in winter runoff as a
fraction of total annual runoff (see Fig. 7). These impacts on hydrology vary by basin,
with the key parameter being the basin elevation relative to the ‘freezing line’during snow
accumulation and melt periods and the prediction of temperature increases. Most studies
have shown, coinciding with the ‘trend studies’presented in the first section, that basins
located at medium altitudes (northern Sierra Nevada Mountains) will be affected more by
climate change (see Lettenmaier et al. 1988; Lettenmaier and Gan 1990; Cayan and Riddle
1993; Duel 1994; Jeton et al. 1996; Wilby and Dettinger 2000; Kim 2001; Kim et al. 2002;
Climatic Change (2007) 82:327–350 337

Fig. 7 Mean-monthly streamflow rates in the Merced (a), Carson (b), and American (c) Rivers, in responses
to PCM-simulated climates during selected 29-year periods. Historical run is a PCM simulation with
historical radiative forcings imposed, the business-as-usual run is a simulation with future business-as-usual
increases in greenhouse gases, and the future-control run is a simulation with future greenhouse-gas
concentrations held constant at 1995 levels (modified from Dettinger et al. 2004)
338 Climatic Change (2007) 82:327–350

Knowles and Cayan 2002; Miller et al. 2003; Knowles and Cayan 2004; Leung et al. 2004;
Kim 2005). However, later studies are showing just the opposite, i.e. higher impacts in
higher elevation basins in southern Sierra Nevada Mountains (Dettinger et al. 2004; Hayhoe
et al. 2004; Van Rheenen et al. 2004). Two factors help explaining this inconsistency:
First, with regards to the “trend studies”it’s important to note that the measured historic
trend in temperature has shown increases of only 1–2°C for the last century. These are
low compared to the temperature increases predicted by the GCMs which are 4–8°C.
Secondly, most GCM’s projections used in previous studies showed either high increases
in winter precipitation (HadCM2) or just mild increases in temperature (around 2°C for
PCM) but none has shown before combinations of high increases in temperature with
decreases in precipitation. So a plausible explanation of this ‘supposed’inconsistency is
the following: historic temperature increases have not been high enough to perturb the
‘temperature insensibility’of high elevation basins like those found in the southern Sierra
Nevada Mountains. Also, results from previous studies have not been in the right
direction to cause this perturbation, because they either masked temperature impacts with
high increments in precipitation (e.g. HadCM2) or because they just didn’t cross the
temperature threshold to perturb the regime of high elevation basins (e.g. PCM). Later
studies are showing that the combination of sufficiently high temperature and decreases
in precipitation are enough to cause significant impacts in the hydrologic conditions in
the southern Sierra Nevada Mountains with impacts larger than their northern Sierra
Nevada counterpart basins.
Changes in total runoff volume depend on the precipitation prediction (scenario), which
depends on the GCM chosen, as previously discussed. Assessments using GCMs that
project wetter conditions (e.g. UK’s HadCM2) tend to produce higher overall streamflow
runoff as compared to “drier”GCMs (e.g. NCAR’s PCM). An example of such diverging
results is shown by Miller et al. (2003) (see Fig. 8). Other examples of such studies are:
Gleick (1987), Lettenmaier et al. (1988), Lettenmaier and Gan (1990), Jeton et al. (1996),
Dettinger (2004), Hayhoe et al. (2004) and Maurer and Duffy (2005). Improvements in the
characterization of uncertainty related to climate impact assessment would improve the
ability to use these diverging results in a manner useful for water resources management.
Table 2summarizes the studies to date on the impacts of climate change on hydrology in
California. We have sorted the studies presented in the table according to the different
methodological approaches they have used.
4 Future predicted impacts on Californian water resources systems
due to climate change
Most of the streamflow in California (especially that draining the Sierra Nevada) is
regulated by large reservoirs. Significant changes in the timing of streamflow that feeds
these reservoirs will alter their ability to serve their design functions under current operating
rules: flood control, water supply for agricultural, urban and industrial uses, hydropower
generation, environmental services, navigation and recreation. For example, an earlier and
shorter snowmelt spring runoff could make it more difficult to refill reservoir flood space
(determined by considering historical hydrologic conditions) during the late spring and
early summer, thus reducing the amount of water supply that can be delivered (Roos 2003).
The ultimate impact on California water resources and their associated functions will
depend on the ability of the man-made infrastructure to cope with these changes. The
Climatic Change (2007) 82:327–350 339

analysis of the performance of the California water system under hypothetical hydrologic
scenarios such as climate change requires the aid of water resources systems models (also
called reservoir system analysis models).
The performance of the California water system under climate change scenarios was first
studied by Lettenmaier and Sheer (1991), and by Sandberg and Manza (1991). Both of
these groups examined the implications of climate change scenarios on the performance of
the State Water Project (SWP) and the Central Valley Project (CVP) using simulation
models. Most of the later studies on the impact of climate change on Californian water
resources have also relied on simulation models (Williams 1988; Dracup et al. 1993;Yao
and Georgakakos 2001; Knowles and Cayan 2002; Van Rheenen et al. 2003; Huber-Lee
et al. 2003; Brekke et al. 2004; Knowles and Cayan 2004; Quinn et al. 2004; Van Rheenen
et al. 2004). An exception to this approach is the optimization model (CALVIN) developed
by Lund et al. (2003). The models used in all of these studies also differ in terms of the
accuracy of their representation of the system being model (e.g., inclusion of groundwater
sources or accuracy of environmental regulation description). The results of these studies in
terms of water deliveries and reservoir storage also reflect the climatic projections used to
derive streamflow runoffs (see previous section). In this regard, predicted drier conditions
reduce the ability of the system to perform at historical levels, if no changes are made in
Fig. 8 Streamflow monthly climatologies based on the HadCM2 and the PCM (from Miller et al. 2003)
340 Climatic Change (2007) 82:327–350

Table 2 Summary of studies on the prediction of the impact of climate change impacts on hydrology in California
Study Determining changes in climatic variables Determining changes in climatic variables
Methodology GCM(s) chosen Downscaling method Methodology Model used (if applicable)
Revelle and Waggoner 1983 Hypothetic
scenarios
NA NA Regression NA
Gleick 1987 Hyp. Scen. +
GCMs
3 GCMs: GFDL, GISS,
NCAR CCM
No downscaling
method applied
Hydrologic Water budget model
Lettenmaier et al. 1988 GCMs 3 GCMs: GFDL, GISS,
OSU
No downscaling
method applied
Hydrologic U.S. National Weather Service
SAC-SMA and Anderson
Snow Model
Lettenmaier and Gan 1990 GCMs 3 GCMs: GFDL, GISS,
OSU
No downscaling
method applied
Hydrologic SAC-SMA
Tsuang and Dracup 1991 NA NA NA Hydrologic Energy based snowmelt
model.
Cayan and Riddle 1993 NA NA NA Regression NA
Duell 1994 NA NA NA Regression NA
Jeton et al. 1996 Hyp. Scen. +
GCMs
3 GCMs: GFDL, GISS,
OSU
No downscaling
method applied
Hydrologic Precipitation-Runoff Modeling
System (PRMS)
Risbey and Entekhabi 1996 NA NA NA Regression
(Observational)
NA
Leung and Ghan 1999 GCMs CCM3 Dynamic Hydrologic Not clear which
Miller et al. 1999 GCMs HadCM2 Dynamic Hydrologic TOPMODEL and Sacramento
Model
McCabe and Wolock 1999 GCMs CGCM1 and HadCM2 Interpolation using
VEMAP
Hydrologic Conceptual Snow model,
no changes in runoff
Wolock and McCabe 1999 GCMs CGCM1 and HadCM2 Interpolation using
VEMAP
Hydrologic Annual water balance model
Miller and Kim 2000 GCMs HadCM2 Dynamic Hydrologic TOPMODEL and Sacramento
Model
Hay et al. 2000 GCMs HadCM2 Delta/ratio and
Statistical
Hydrologic PRMS
Climatic Change (2007) 82:327–350 341

Table 2 (continued)
Study Determining changes in climatic variables Determining changes in climatic variables
Methodology GCM(s) chosen Downscaling method Methodology Model used (if applicable)
Peterson et al. 2000 NA NA NA Regression NA
Wilby and Dettinger 2000 GCMs HadCM2 Statistical Hydrologic PRMS
Carpenter and Georgakakos 2001 GCMs CGCM1 Statistical Hydrologic SAC-SMA
Kim 2001 GCMs HadCM2 Dynamic Hydrologic Not clear which
Kim et al. 2002 GCMs HadCM2 Dynamic Hydrologic Not clear which
Snyder et al. 2002 GCMs CCM3 Dynamic Hydrologic Not clear which
Knowles and Cayan 2002 GCMs PCM Statistical Hydrologic SAC-SMA (modified)
Van Rheenen et al. 2003 GCMs PCM Statistical Hydrologic Variable Infiltration Capacity
(VIC)
Huber-Lee et al. 2003 GCMs 2 GCMs: HadA2 and HadB2 Statistical Hydrologic WEAP
Miller et al. 2003 Hyp. Scen. + GCMs 2 GCMs: PCM and HadCM2 Statistical Hydrologic SAC-SMA
Dettinger 2004 GCMs 6 GCMs*3 Emission
Scenarios = 18 projections
No downscaling
method applied
Hydrologic Sensitivity Analysis from
Jeton et al (1996)
Dettinger et al. 2004 GCMs PCM Statistical Hydrologic PRMS
Van Rheenen et al. 2004 GCMs PCM Statistical Hydrologic VIC
Stewart et al. 2004 GCMs PCM Statistical Regression NA
Knowles and Cayan 2004 GCMs PCM Statistical Hydrologic SAC-SMA (modified)
Leung et al. 2004 GCMs PCM Dynamic Hydrologic Not clear which
Hayhoe et al. 2004 GCMs 2 GCMs: PCM and HadCM3 Statistical Hydrologic VIC
Maurer and Duffy 2005 GCMs 10 GCMs Statistical Hydrologic VIC
Kim 2005 GCMs HadCM2 Dynamic Hydrologic Not clear which
Dettinger 2005 GCMs 6 GCMs*3 Emission
Scenarios = 18 projections
No downscaling
method applied
Hydrologic Sensitivity Analysis from
Jeton et al (1996)
Georgakakos et al. 2005 GCMs CGCM1 Statistical Hydrologic SAC-SMA
342 Climatic Change (2007) 82:327–350

reservoir operating rules or other water management policies. On the other hand, predicted
wetter conditions would tend to overcome changes in the seasonality of streamflows,
producing an overall improvement of system performance. An example of these diverging
results in is presented in Fig. 9.
Frederick and Schwarz (1999) have not followed the approach discussed in the previous
paragraph but have estimated impacts to water resources without the aid of a water
resources system model. The approach followed by Frederick and Schwarz (1999) which
was part of the Water Sector report of the National Assessment Team for the U.S. Global
Change Research Program (Gleick 2000) considered changes in annual streamflow
upstream of reservoirs, and compared those with annual water demands to obtain water
scarcity indices with associated economic costs. A major drawback of this procedure is that
it does not account for changes in streamflow timing, which is a significant potential impact
on California.
Most authors have focused on water storage and deliveries from reservoirs to assess the
impacts of climate change on California water resource systems. However, it is worthwhile
noting some studies that have also included other means of measuring impacts. For
Fig. 9 Simulated monthly mean delivery levels (i.e. ratio of delivery to demand) for all CVP agricultural
users South of the delta for DRY, NORMAL and WET years, under control and two GCMs scenarios: a2025
and b2065 (from Brekke et al. 2004)
Climatic Change (2007) 82:327–350 343

example, Knowles and Cayan (2002,2004) studied the impacts on San Francisco Bay
salinity levels. Dracup et al. (1993) performed a series of studies on several water resources
functions such as hydropower generation (for CVP/SWP system), agricultural economic
costs and chinook salmon population.
The authors believe that there has been a dearth of studies related to climate change
impacts on water resources systems in California in the following areas: (1) The impact on
hydropower production in general and specifically in high altitude hydropower generation
facilities; (2) The impact associated with changes in reliability (and therefore economic
costs) among different water users (with different water rights and water sources) in the
California system and; (3) How the different reservoir objectives tradeoffs (e.g. hydropower
vs. water supply or flood control vs. water supply) could be altered under changes in
seasonal patterns predicted by climate change assessments.
A further important step in the analysis of predicted impacts to water resource systems
associated with climate change has been to devise changes in water resources management
practices to cope with predicted changes in streamflows. This approach was pursued by Van
Rheenen et al. (2004), who developed a series of mitigation strategies such as changing the
Table 3 Summary of studies on the impact of climate change on water resources (and associated functions)
in California
Study Model used Model spirit Impacts considered
Riebsame 1988 NA NA Policy Analysis. Flood and Drought reservoir
operations
Williams 1988 PROSIM* Simulation San Francisco Bay Salinity
Lettenmaier and Sheer
1991
WRMI Simulation Water supply deliveries
Sandberg and Manza
1991
PROSIM Simulation Water supply deliveries
Dracup et al. 1993 PROSIM Simulation Water supply deliveries + agricultural sector +
hydropower generation + Salmon population
Risbey 1998 NA NA Policy Analysis. Scenario comparison
Frederick and Schwarz
1999
NA NA Economic costs of annual changes in water
supplies. No water resources model used.
Haddad and Merrit 2001 NA NA Policy Analysis
Yao and Georgakakos
2001
Decision
Module
Simulation Reservoir performance under different
management and forecasting procedures
Van Rheenen et al. 2003 CVMod Simulation Reservoir water storage
Knowles and Cayan 2002 Historic releases Simulation Salinity levels in the Bay Delta
Huber-Lee et al. 2003 WEAP Simulation Economic Impacts related to scarcity costs +
Hydropower + Envrionmental Constraints
Lund et al. 2003 CALVIN Optimization Economic Impacts related to scarcity costs +
Hydropower + Envrionmental Constraints
Brekke et al. 2004 CalSim-II Hybrid Water supply deliveries and reservoir storage +
delta salinity
Quinn et al. 2004 CalSim-II Hybrid Water supply deliveries and reservoir storage +
delta salinity
Knowles and Cayan 2004 Historic releases Simulation Salinity levels in the Bay Delta
Van Rheenen et al. 2004 CVMod Simulation Reservoir releases and storage + environmental,
hydropower, flood control targets
344 Climatic Change (2007) 82:327–350

flood control rule curves of reservoir releases to lessen the impacts of climate change. They
concluded that even with the most comprehensive approaches,
achieving and maintaining status quo (control scenario climate) system performance in
the future would be nearly impossible, given the altered climate (change) scenario
hydrologies.
However, Yao and Georgakakos (2001) in research also prepared for the Water Sector
report of the National Assessment Team for the U.S. Global Change Research Program
(Gleick 2000) reached different conclusions. Yao and Georgakakos (2001) developed an
integrated forecast-decision system to assess the sensitivity of reservoir performance to
various forecast-management schemes under historical and future climate scenarios. Their
assessments are based on various combinations of inflow forecasting models, decision
rules, and climate scenarios. Their study demonstrated that
(1) reliable inflow forecasts and adaptive decision systems can substantially benefit
reservoir performance and (2) dynamic operational procedures can be effective climate
change coping strategies.
The authors believe that although these two studies reflect important advances, however
there are still some areas of work that should be addressed. These include studying potential
adaptation opportunities using, for example, a conjunctive use management approach or
optimization techniques (e.g. Stochastic Dynamic Programming) to determine whether
current reservoir release policies should be modified.
Finally a group of studies (Riebsame 1988; Risbey 1998; Haddad and Merrit 2001)
approached the impacts of climate change on California water resources not from a
quantitative perspective but from a qualitative policy-oriented approach. Riebsame (1988)
used an historic perspective to analyze the approaches taken by water managers in
California to adjust to climate variability and how those could be used in a climate change
scenario. Risbey (1998) performed a qualitative sensitivity analysis of the different
adaptation policies that can be undertaken in the Sacramento basin considering the
uncertainties embedded in climate and streamflow projections.
In Table 3we present a summary of studies on the topic of predicting climate change
impacts on water resources in California.
5 Conclusions
We have presented a review of studies in the field of climate change impacts on California
hydrology and water resources systems. These studies date from 1983 and can be
categorized in three major fields: (1) Studies of historical trends of streamflow and
snowpack in order to determine if there is any evidence of climate change in the
geophysical record; (2) Studies of potential future predicted effects of climate change on
streamflow and; (3) Studies that use those predicted changes in natural runoff to determine
their economic, ecologic, or institutional impacts.
In California we found more than 60 of these studies, with a majority focused on
predicting the hydrologic impacts of climate change. In Fig. 10 we show a distribution of
all these studies according to the classifications followed in this paper. The number and
breadth of studies permits a review of the different methodological issues.
Climatic Change (2007) 82:327–350 345

Major conclusions that can be derived from these studies are the following:
&In the last few decades there has been a trend towards an earlier timing of streamflow,
i.e. earlier spring pulse onset and earlier center of mass of the hydrograph, in the
California Sierra Nevada Mountains. These trends correlate well with an increasing
trend in temperature levels. Whether these trends are due to climate change or climate
variability is a subject of continuing debate.
&Several different approaches have been used to predict future streamflows related to
climate change. These approaches use either global climate models (GCMs) or
hypothetical scenarios to forecast changes in climatic variables. These changes of
climatic variables have been used to assess changes in natural runoff using different types
of hydrologic models (e.g. statistical or physically based). Results derived from these
studies consistently show a change in timing in streamflow runoff due to a consistent
increase in temperature. However, changes in the total volume of runoff are still not clear,
mainly due to uncertainties in future precipitation projections. The methodology used to
assess changes in hydrology has improved both in terms of downscaling outputs from
GCMs and in the characterization of uncertainty. Further improvements, such as the
refinement of the spatial and temporal resolution of the models, will enhance the ability
of policy makers to use this information in decisions-making.
&The forecasted hydrologic conditions associated with climate change will affect the
performance of the water infrastructure in California. Some research has addressed
these issues using water resource system models (either optimization or simulation) to
redistribute the hydrologic changes throughout the California water system. We believe
that there is a potential for significant research on the impacts of climate change on
California water resource systems.
Trend
23%
Hydrology
/ Trend
2%
Hydrology
51%
Hydrology
/ WR
9%
WR
15%
Fig. 10 1983–2005 studies on
climate change impacts on hy-
drology and water resources in
California, separated by type of
study. Nomenclature: Trend:
studies on the historic record;
Hydrology: on the potential future
impacts of climate change on
hydrology and; WR: on the
impacts to water resources sys-
tems. Some studies belong to two
categories (i.e. Hydrology/WR
and Hydrology/Trend)
346 Climatic Change (2007) 82:327–350

6 Caveat
A reduced version of this paper was previously presented at the Environmental & Water
Resources Institute Congress, American Society of Civil Engineers, Anchorage, AK, May
15–19. This reduced version of the paper was also published in the proceedings of the
conference (Dracup and Vicuna 2005).
Acknowledgments Funding was provided by the California Energy Commission through the California
Climate Change Center. We thank R. Leonardson for her thoughtful reviews. We acknowledged two
anonymous reviewers and P. Gleick for his extensive review and comments on this paper.
References
Aguado E, Cayan DR, Riddle LG, Roos M (1992) Climatic fluctuations and the timing of West Coast
streamflow. J Climate 5:1468–1483
Arnell NW (1999) Climate change and global water resources. Glob Environ Change 9(Suppl):S31–S49
Brekke LD, Miller NL, Bashford KE, Quinn NWT, Dracup JA (2004) Climate change impacts uncertainty for
water resources in the San Joaquin River Basin, California. J Am Water Resour Assoc 40(1):149–164
Carpenter TM, Georgakakos KP (2001) Assessment of Folsom lake response to historical and potential
future climate scenarios: 1. Forecasting. J Hydrol 249(1–4):148–175
Cayan DR, Riddle LG (1993) The influence of precipitation and temperature on seasonal streamflow in
California. Water Resour Res 29(4):1127–1140
Cayan DR, Kammerdiener S, Dettinger MD, Caprio J, Peterson DH (2001) Changes in the onset of spring in
the Western United States. Bull Am Meteorol Soc 82:399–415
Dettinger MD (2004) From climate change spaghetti to climate change distribution. Prepared for California
Energy Commission. Public Interest Research Program. Sacramento, CA. Available at: http://www.
energy.ca.gov/reports/2004-04-22_500-04-028.PDF
Dettinger MD (2005) From climate change spaghetti to climate change distributions for 21st century. San
Francisco Estuary and Watershed Science 3(1)
Dettinger MD, Cayan DR (1995) Large-scale atmospheric forcing of recent trends toward early snowmelt
runoff in California. J Climate 8(3):606–623
Dettinger MD, Cayan DR, Meyer M, Jeton AE (2004) Simulated hydrologic responses to climate variations
and change in the Merced, Carson, and American river Basins, Sierra Nevada, California, 1900–2099.
Clim Change 62(1–3):283–317
Dracup JA, Vicuna S (2005) An overview of hydrology and water resources studies on climate change: the
California experience. World Water Congress 2005. Impacts of Global Climate Change. World Water
and Environmental Resources Congress 2005. May 15–19, 2005, Anchorage, Alaska. doi: 10.1061/
40792(173)483
Dracup JA, Pelmulder SD, Howitt R, Horner G, Hanemann WM, Dumas CF, McCann R, Loomis J, Ise S
(1993) Integrated modeling of drought and global warming: impacts on selected California resources.
Report by the National Institute for Global Environmental Change. University of California, Davis, CA.,
p112
Duel LFW Jr (1994) The sensitivity of Northern Sierra Nevada streamflow to climate change. Water Resour
Bull 30(5):841–859
Fox JP, Mongan TR, Miller WJ (1990) Trends in freshwater inflow to San Francisco Bay from the
Sacramento-San Joaquin Delta. Water Resour Bull 26(1):101–116
Frederick KD, Schwarz GE (1999) Socioeconomic impacts of climate change on U.S. water supplies. J Am
Water Resour Assoc 35(6):1563–1583
Freeman GJ (2002) Looking for recent climatic trends and patterns in California’s Central Sierra. In
Proceedings 19th Annual Pacific Climate (PACLIM) Workshop, Pacific Grove, CA pp 35–48
Georgakakos KP, Graham NE, Carpenter TM, Georgakakos AP, Yao H (2005) Integrating climate-hydrology
forecasts and multi-objective reservoir management for Northern California. EOS, Trans, AGU 86(12)
Climatic Change (2007) 82:327–350 347

Gleick P (1986) Methods for evaluating the regional hydrologic impacts of global climatic changes. J Hydrol
88:97–116
Gleick PH (1987) Regional hydrologic consequences of increases in atmospheric CO
2
and other trace gases.
Clim Change 10:137–161
Gleick PH (1989) Climate change, hydrology, and water resources. Rev Geophys 27(3):329–344
Gleick PH (2000) US national assessment of the potential consequences of climate variability and change.
Sector: water resources. A Report of the National Assessment Synthesis Team. US Global Change
Research Program. Washington, D.C
Gleick PH, Chalecki EL (1999) The impacts of climatic change for water resources of the Colorado and
Sacramento-San Joaquin river basins. J Am Water Resour Assoc 35(6):1429–1441
Haddad BM, Merrit K (2001) Evaluating regional adaptation to climate change: the case of California water.
Advances in Economics of Environmental Resources 3:65–93
Hay LE, Wilby RL, Leavesley GH (2000) A comparison of delta change and downscaled GCM scenarios for
three mountainous basins in the United States. J Am Water Resour Assoc 36(2):387–397
Hayhoe K, Cayan DR, Field C, Frumhoff P, Maurer E, Miller N, Moser S, Schneider S, Cahill K, Cleland E,
Dale L, Drapek R, Hanemann RM, Kalkstein L, Lenihan J, Lunch C, Neilson R, Sheridan S, Verville J
(2004) Emissions pathways, climate change, and impacts on California. Proceedings of the National
Academy of Sciences (PNAS) 101(34):12422–12427
Huber-Lee A, Yates D, Purkey D, Yu W, Runkle B (2003) Water, climate, food, and environment in the
Sacramento Basin. Contribution to the project ADAPT-Adaptation strategies to changing environments,
Stockholm Environment Institute. Boston, MA
IPCC (2001) Climate change 2001: scientific basis. In: Metz B, et al (eds) Contribution of working group III
to the third assessment report of the intergovernmental panel on climate change. Published for the
Intergovernmental Panel on Climate Change (by) Cambridge University Press, Cambridge, UK; New
York, NY, USA
Jeton AE, Dettinger MD, Smith JL (1996) Potential effects of climate change on streamflow: Eastern and
Western slopes of the Sierra Nevada, California and Nevada. U.S. Geological Survey, Water Resources
Investigations Report 95–4260, pp 44
Kim J (2001) A nested modeling study of elevation-dependent climate change signals in California induced
by increased atmospheric CO
2
. Geophys Res Lett 28(15):2951–2954
Kim J (2005) A projection of the climate change induced by increased CO
2
on extreme hydrologic events in
the western U.S. Clim Change 68(1–2):153–168
Kim J, Kim T-K, Arritt RW, Miller NL (2002) Impacts of increased atmospheric CO
2
on the hydroclimate of
the Western United States. J Climate 15:1926–1942
Kiparsky M, Gleick PH (2003) Climate change and California water resources: a survey and summary of the
literature. Pacific Institute for Studies in Development, Environment, and Security, Oakland, California.
California Energy Commission. Report 500-04-073, Sacramento, CA
Knowles N, Cayan D (2002) Potential effects of global warming on the Sacramento/San Joaquin watershed
and the San Francisco estuary. Geophys Res Lett 29(18):1891
Knowles N, Cayan DR (2004) Elevational dependence of projected hydrologic changes in the San Francisco
estuary and watershed. Clim Change 62:319–336
Lettenmaier DP, Gan TY (1990) Hydrologic sensitivities of the Sacramento-San Joaquin river basin,
California, to global warming. Water Resour Res 26:69–86
Lettenmaier DP, Sheer DP (1991) Climate sensitivity of California water resources. J Water Resour Plan
Manage 117(1):108–125
Lettenmaier DP, Gan TY, Dawdy DR (1988) Interpretation of hydrologic effects of climate change in the
Sacramento-San Joaquin river basin, California. Prepared for U.S. Environmental Protection Agency.
Department of Civil Engineering, University of Washington, Seattle, WA
Leung LR, Ghan S (1999) Pacific northwest climate sensitivity simulated by a regional climate model driven
by a GCM. Part II: 2×CO
2
simulations. J Climate 12:2031–2053
Leung LR, Qian Y, Washington WM, Han J, Roads JO (2004) Mid-century ensemble regional climate
change scenarios for the Western United States. Clim Change 62:75-113
Lund JR, Howitt RE, Jenkins MW, Zhu T, Tanaka SK, Pulido M, Tauber M, Ritzema R, Ferriera I (2003)
Climate warming & California’s water future. Report 03-1. Center for Environmental and Water
Resource Engineering. University of California, Davis. California Energy Commission, Sacramento,
CA. Available at: http://cewre.engr.ucdavis.edu/Pub/Lund/03-1.pdf
Maurer EP, Duffy PB (2005) Uncertainty in projections of streamflow changes due to climate change in
California. Geophys Res Lett 32(3)
McCabe GJ, Wolock DM (1999) General-circulation-model simulations of future snowpack in the western
United States. J Am Water Resour Assoc 35(6):1473–1484
348 Climatic Change (2007) 82:327–350

Miller NL, Kim J (2000) Climate change sensitivity analysis for two California waterhseds: addendum to
downscaled climate and streamflow study of the Southwestern United States. J Am Water Resour Assoc
36(3):657–661
Miller NL, Kim J, Hartman RK, Farrara J (1999) Downscaled climate and streamflow study of the
Southwestern United States. J Am Water Resour Assoc 35(6):1525–1537
Miller NL, Bashford KE, Strem E (2003) Potential impacts of climate change on California hydrology. J Am
Water Resour Assoc 39(4):771–784
Mote PW, Hamlet AF, Clark MP, Lettenmaier DP (2005) Declining mountain snowpack in Western North
America. Bull Am Meteorol Soc 86(1):39–49
Peterson DH, Smith RE, Dettinger MD, Cayan DR, Riddle L (2000) An organized signal in snowmelt runoff
over the western United States. J Am Water Resour Assoc 36(2):421–432
Phillips N (1956) The general circulation of the atmosphere: a numerical experiment. Q J R Meteorol Soc
82:123–164
Pupacko A (1993) Variations in northern Sierra Nevada streamflow: implications of climate change. Water
Resour Bull 29(2):283–290
Quinn NWT, Brekke LD, Miller NL, Heinzer T, Hidalgo H, Dracup JA (2004) Model integration for
assessing future hydroclimate impacts on water resources, agricultural production and environmental
quality in the San Joaquin Basin, California. Environ Model Softw 19:305–316
Regonda SK, Rajagopalan B, Clark M, Pitlick J (2005) Seasonal cycle shifts in hydroclimatology over the
Western United States. J Climate 18(2):372–384
Revelle RR, Waggoner PE (1983) Effects of a carbon dioxide induced climatic change on water supplies in
the Western United States. In: Changing climate. National Academy of Sciences Press, Washington, D.C
Riebsame WE (1988) Adjusting water resources management to climate change. Clim Change 13(1):69–97
Risbey JS (1998) Sensitivities of water supply planning decisions to streamflow and climate scenario
uncertainty. Water Policy 1:321–340
Risbey JS, Entekhabi D (1996) Observed Sacramento Basin streamflow response to precipitation and
temperature changes and its relevance to climate change impact studies. J Hydrol 184:209–223
Roos M (1987) Possible changers in California snowmelt patterns. In: Proceedings Fourth Annual Pacific
Climate (PACLIM) Workshop, Pacific Grove, CA, pp 22–31
Roos M (1991) A trend of decreasing snowmelt runoff in northern California. In: Proceedings 59th Western
Snow Conference, Juneau, AK, pp 29–36
Roos M (2003) The effects of global climate change on California water resources. A report for the Energy
California Commission. Public Interest Energy Research Program. Research Development and
Demonstration Plan. Sacramento, CA
Sandberg J, Manza P (1991) Evaluation of central valley project water supply and delivery system.
Global Change Response Program, U.S. Department of Interior, Bureau of Reclamation, Sacramento,
California
Shelton ML (1998) Seasonal hydroclimate change in the Sacramento river Basin, California. Phys Geogr 19
(3):239–255
Shelton ML, Fridirici RM (1997) Decadal changes of inflow to the Sacramento-San Joaquin Delta,
California. Phys Geogr 18(3):215–231
Snyder MA, Bell JB, Sloan LC, Duffy PB, Govindasamy B (2002) Climate responses to a doubling of
atmospheric carbon dioxide for a climatically vulnerable region. Geophys Res Lett 29(11):1514–1517
Stewart IT, Cayan DR, Dettinger MD (2004) Changes in snowmelt runoff timing in western North America
under a ‘business as usual’climate change scenario. Clim Change 62(1–3):217–232
Stewart IT, Cayan DR, Dettinger MD (2005) Changes toward earlier streamflow timing across Western North
America. J Climate 18(8):1136–1155
Tsuang BJ, Dracup JA (1991) Effect of global warming on Sierra Nevada mountain snow storage. In:
Proceedings 59th Western Snow Conference, Juneau, AK, pp 17–28
Van Rheenen NT, Palmer RN, Hahn M (2003) Evaluating potential climate change impacts on water resource
system operations: case studies of Portland, Oregon and Central Valley, California. Water Resources
Update 124:35–50
Van Rheenen NT, Wood AW, Palmer RN, Lettenmaier DP (2004) Potential implications of PCM climate
change scenarios for Sacramento-San Joaquin River Basin hydrology and water resources. Clim Change
62(1–3):257–281
Wahl KL (1991) Is April to June runoff really decreasing in the western United States? In: Proceedings 59th
Western Snow Conference, Juneau, AK, pp 67–78
Wilby RL, Dettinger MD (2000) Streamflow changes in the Sierra Nevada, California, simulated using
statistically downscaled general circulation model output. In: McLaren S, Kniven D (eds) Linking
climate change to land surface change. Kluwer Academic Pub., pp 99–121
Climatic Change (2007) 82:327–350 349

Wilkinson R, Clarke K, Goodchild M, Reichman J, Dozier J (2002) The potential consequences of climate
variability and change for California: the California regional assessment. U.S. Global Change Research
Program, Washington, DC
Williams PB (1988) The impacts of climate change on the salinity of San Francisco Bay. Philip Williams and
Associates for Environmental Protection Agency
Wolock DM, McCabe GJ (1999) Estimates of runoff using water-balance and atmospheric general circulation
models. J Am Water Resour Assoc 35(6):1341–1350
Wood AW, Lettenmaier DP, Palmer RN (1997) Assessing climate change implications for water resources
planning. Clim Change 37:203–228
Yao H, Georgakakos A (2001) Assessment of Folsom Lake response to historical and potential future climate
scenarios 2. Reservoir management. J Hydrol 249(1–4):176–196
350 Climatic Change (2007) 82:327–350