A 600-Year Streamflow History in the Salinas Valley

A 600-Year Streamflow History in the Salinas Valley Reconstructed from Blue Oak Tree-Rings

R. Daniel Griffin

A 600-YEAR STREAMFLOW HISTORY IN THE SALINAS VALLEY RECONSTRUCTED FROM BLUE OAK TREE-RINGS

A thesis submitted in partial fulfillment of the requirements for the degree of

Master of Arts

By

Richard Daniel Griffin, B.S. University of Arkansas, 2002

May 2007 University of Arkansas

Abstract

The large scale agricultural industry of the Salinas Valley has been developed

around rich soils, coastal microclimates, and long-term groundwater overdraft. Seawater

intrusion in the lower valley threatens productive agricultural lands and municipal

freshwater supplies. Infiltration from the Salinas River and its tributaries provides the

principal recharge to the valley’s alluvial aquifers. Salinas Basin water managers have

goals for hydrologic balance. However, relatively short instrumental discharge records

provide limited perspective on hydroclimatic variability in the Salinas Valley,

particularly at longer timescales. The recently developed network of blue oak (Quercus

douglasii) tree-ring chronologies offers potential for investigating the pre-instrumental

history of Salinas River discharge. Over 1,000 blue oak specimens precisely dated with

the methods of dendrochronology have been used to develop 13 new tree-ring

chronologies for the Central Coast Ranges of California. These blue oak chronologies

reflect regional scale precipitation variability, and are also highly correlated with Salinas

River streamflow.

Bivariate regression was used to reconstruct Salinas River water year discharge at

Paso Robles from an average of the four longest blue oak chronologies in the region. The

595-year reconstruction explains 70% of the observed flow variance and provides new

paleoclimatic evidence for a long history of quasi-periodic moisture rhythms in the

California’s Central Coast Ranges. The reconstruction illustrates periods of surplus and

deficit discharge more persistent than those represented in the Paso Robles gauge record,

and indicates that decadal scale drought is not uncommon for the Salinas Valley.

Consecutive year surplus flow events were less frequent during the 20th century, while

consecutive year droughts were more common. According to this study, the most

persistent drought in 600 years (1917-1934) coincided with the first signs of seawater

intrusion in the lower Salinas Valley. A modern drought of comparable magnitude would

have severe implications for freshwater supply, groundwater recharge, seawater intrusion,

and agricultural production in “America’s Salad Bowl.”

This thesis is approved for recommendation to the Graduate Council Thesis Director: _____________________________________ David W. Stahle Thesis Committee: _____________________________________ David W. Stahle _____________________________________ Malcolm K. Cleaveland _____________________________________ Ralph K. Davis

Acknowledgements

I am pleased to thank my thesis committee members and all of the individuals who have

contributed this research. Foremost, I am deeply grateful to my mentor, David W. Stahle,

for his exhaustive guidance, daily inspiration, and enduring friendship. His relentless

pursuit of academic excellence sets a model for success. I am very thankful for Malcolm

K. Cleaveland. Over the years he has provided countless hours of instruction, assistance,

and priceless humor. I appreciate the insight, suggestions, and support offered by Ralph

K. Davis. I am indebted to my student colleagues, particularly Matthew D. Therrell,

Jesse R. Edmondson, Justin D. Pollan, Richard H. Styron, Dorian J. Burnette, and Mark

D. Spond, for their assistance, encouragement, and friendship, in both the field and the

laboratory. I thank Falko K. Fye for his assistance with composite mapping and climatic

interpretation. It has been my privilege to work in California, and I wish to recognize

the researchers whose collaborative expertise facilitated the Ancient Blue Oak Project:

Daniel R. Cayan, Michael D. Dettinger, David M. Meko, Kelly T. Redmond, and David

W. Stahle. We are grateful to the numerous private land owners and public land

managers who permitted access to the blue oak study sites. This research was funded by

the CALFED Ecosystem Restoration Program and the National Science Foundation.

I dedicate this work to my loving family.

vi

Table of Contents

Abstract ii Acknowledgements vi Introduction 1 Tree growth as a Proxy for Streamflow 1 California Blue Oak 4 Salinas Valley Water Resource Issues 6 Data and Methods 8 Tree-Ring Data 8 Streamflow Data 13 Calibration and Verification 14 Results and Discussion 15 Conclusions 21 Literature Cited 24 Figures 32 Tables 53 Appendices 60

vii

INTRODUCTION Tree-growth as a Proxy for Streamflow

Climatic fluctuations should be considered in water resources planning (e.g.

Cayan et al. 2003). However, instrumental records of streamflow do not always exhibit

the full range of climatic variability, particularly at longer timescales. Tree-ring

chronologies provide a high resolution and widely available natural proxy record for

reconstructing long-term hydroclimatic history. Chronologies developed from ancient

living trees and dead wood can extend thousands of years back in time (e.g., Stahle and

Cleaveland 1992; Grissino-Mayer 1996). Tree-ring chronologies have been widely used

to reconstruct past climatic conditions over North America (e.g., Cook et al. 1999; Cook

and Krusic 2004), and several studies have clearly demonstrated that the continental

United States is subject to multi-decadal climate variability beyond the range of that

exhibited in instrumental climate records (e.g., Woodhouse and Overpeck 1998; Stahle et

al. 2000; Cook et al. 2004).

Several previous studies have developed tree-ring based reconstructions of

streamflow for the exploration of pre-instrumental hydrologic conditions. Perhaps the

most well known tree-ring based discharge reconstruction was developed for the

Colorado River gauge at Lees Ferry (Stockton and Jacoby 1976). This reconstruction

extended from 1520-1961 and indicated that the Colorado River Compact, which

allocated 20.2 billion cubic meters (16.4 million acre-feet) of water per year based on a

relatively short dishcarge record, was not representative of long-term mean flow

conditions. The study concluded that the streamflow record used for the Compact was

actually based on the wettest period since the early 16th century, and that the Compact

may have over-allocated an average of some 2.6 billion cubic meters (2.1 million acre-

1

feet) per year to the stakeholders in the Colorado River basin. Woodhouse et al. (2006)

conducted a follow up investigation that utilized an expanded and updated network of

tree-ring chronologies in the development of a suite of tree-ring reconstructions of flow

in the Colorado River Basin, including the Lees Ferry gauge record. Woodhouse et al.

(2006) used a variety of reconstruction approaches, and concluded that the over-

estimation of annual mean Colorado River flow was closer to 1.6 billion cubic meters

(1.3 million acre feet). They confirmed that the Colorado River Compact was based on

one of the wettest periods in the last five centuries, and that some pre-20th century

droughts were more severe than any witnessed during the instrumental period.

Cook and Jacoby (1983) used canonical regression to reconstruct Potomac River

low flow (July-September) at Point of Rocks, Maryland, from 1730-1977. They found

several periods of about 50 years when the flow was above or below the median, and

determined that the majority of the gauged flow period was generally above the median.

This work demonstrated the potential for tree-ring reconstructions of streamflow in the

eastern United States.

Cleaveland and Stahle (1989) used tree-rings to reconstruct 1700-1980 annual

runoff of the White River at Clarendon, Arkansas. This analysis described the tendency

for consecutive years of surplus and deficit flow before and during the instrumental

period, including persistent deficit flow events more severe than any experienced during

the 20th century. The reconstruction also indicated a significantly greater variance in

flows during the 20th century, possibly reflecting changing climate and/or anthropogenic

disturbances to the watershed. Cleaveland (2000) developed a 963-year tree-ring

reconstruction of summer (June-August) flows for the same gauge record on the White

2

River at Clarendon, Arkansas. This work provided a longer perspective on discharge

variability and confirmed the findings of the previous study. White River extreme low

flow events tend to cluster in both the instrumental and reconstructed records. The White

River instrumental record did not capture the full range of duration or magnitude for

persistent flow extrema, and the 20th century may have experienced a greater number of

discharge extrema, both wet and dry.

In California, Meko et al. (2001) developed a tree-ring reconstruction of

Sacramento River discharge for 869-1977 from a network of tree-ring chronologies that

included multiple tree species. The goals of this study were to provide water managers

with long-term perspective on hydrologic drought to aid in “sensible water resources

planning” for the Sacramento Basin, a “vital source for surface water in California”

(Meko et al. 2001). Their results indicate that the instrumental flow history for the

Sacramento River contains representative drought extremes lasting 6-10 years, but that

the instrumental record does not accurately represent the magnitude of drought extrema

for longer or shorter averaging periods. The reconstruction indicates 1-5 year and 10+

year drought extrema more severe than represented by the instrumental record, further

advocating the use of proxy records in studying long term hydrologic regimes in the

western United States.

Five of the 36 predictor tree-ring chronologies used in the Sacramento River

reconstruction were blue oak (Quercus douglasii, Hook. and Arn.) chronologies

developed at the University of Arkansas Tree-Ring Laboratory (Meko et al. 2001;

ITRDB 2007). The three blue oak chronologies from the central Coast Ranges (Mount

Diablo, Pinnacles, and Pacheco Pass) were reported to be the most sensitive to water year

3

precipitation of all 36 chronologies considered. The regression model which included the

blue oak chronologies as predictors exhibited superior regression statistics (Meko et al.

2001), and a report to the California Department of Water Resources (Meko 2001)

advocated the expansion and extension of chronologies in the blue oak network for

improved understanding of long-term hydrologic drought in the Sacramento Basin.

Other studies of tree-ring reconstructed streamflow include Brito-Castillo et al.

(2003); Case and MacDonald (2003), Earle (1993), Gedalof et al. (2004), Graumlich et

al. (2003), Holmes et al. (1979), Jain et al. (2002), Jones et al. (1984), Lara et al. (2005),

Meko and Graybill (1995), Meko and Woodhouse (2005), Meko et al. (1995), Norton

(1987), Phipss (1983), Shen and Tabois (1995), Smith and Stockton (1981), Villanueva-

Diaz (2005), and Woodhouse (2001).

California Blue Oak

California is world famous for its trees, which include the oldest, tallest, and

largest individual trees yet documented (Lanner 1999; Van Pelt 2001). Unfortunately,

California’s endemic blue oak remains underappreciated (Stahle 2002; Allen-Diaz et al.

in press). This species is found at the lower forest border in the iconic foothills of the

interior Coast Ranges and in the Great Central Valley (Figure 1, Griffin and Critchfield

1972). Extending from the semi-arid grassland to mixed conifer forest, blue oak canopy

closure is highly variable, but generally changes from open savannas at low elevation to

dense woodlands at higher positions (Figure 2; Allen-Diaz et al. in press). Blue oak trees

seem to be remarkably well adapted to their drought-prone environments (Pavlik et al.

1991). They have small leaves that probably reduce evapotranspirative moisture loss and

4

can be shed under extremely dry summer conditions (Figure 3; Pavlik et al. 1991). The

annual growth rate of blue oak is very slow and is principally limited by winter-spring

moisture conditions (e.g., Stahle et al. 2001).

Blue oaks exhibit a stunted and knotty growth form (Figure 4) and are susceptible

to heartwood rot induced by at least four fungal agents (Inonotus dryophilus, Laetiporus

sulphureus, Hydnum erinaceum, Ganoderma applanatum; McDonald 1990). These trees

are not suited to timber production (McDonald 1990) and were never industrially logged

(Stahle 2002). Intensive field surveys indicate that old-growth blue oak are common

throughout the plant’s native distribution (Figure 1). Blue oak woodlands provide critical

wildlife habitat in California, often at the sprawling urban-wildland interface, with faunal

species richness higher than many terrestrial plant communities in the state (e.g., Allen-

Diaz et al. in press). Unfortunately, the ecological significance of old-growth blue oak is

not widely appreciated in California, and these native plant communities continue to be

threatened by a number of factors including overgrazing, agriculture expansion, firewood

cutting, urban, and exurban development (Figure 5, Allen-Diaz et al. in press).

Despite recent land use trends, drought stressed blue oaks remain abundant in

California, and this species has a demonstrated potential for tree-ring dating, long

chronology development, and climate reconstruction for improved insight into the

Golden State’s pre-instrumental streamflow and precipitation history (e.g. Meko et al.

2001; Stahle et al. 2002; Redmond et al. 2002). The University of Arkansas Tree-Ring

Laboratory has recently updated and expanded a network of 36 blue oak tree-ring

chronologies that are 300-700 years long and well distributed throughout the native range

of the species (Figure 6). These chronologies were all developed from old-growth blue

5

oak woodlands, and offer new opportunities for reconstructing California’s hydroclimatic

variability at a regional scale. The chronologies from the central Coast Ranges (Figure 6;

Table 1) are among the longest and most moisture-sensitive in the blue oak network, and

because they are so well correlated with Salinas River discharge records, an investigation

into pre-instrumental hydroclimatic variability in the Salinas Valley is easily justifiable.

Salinas Valley Water Resource Issues

The Salinas River Valley is the largest coastal basin in southern California (Figure

7; Planert and Williams 1995), and has a long history as a national agricultural center

(e.g., Anderson 2000). Beginning in the late 19th century, sugar beets and table

vegetables were produced in large quantities and distributed across the United States. By

the early 20th century, groundwater pumps were extracting an estimated 38 cubic meters

per minute from the Salinas’ alluvial aquifers (Anderson 2000). In 1924 the turbine

pump overcame the shallow depth limitations of the centrifugal pump and agricultural

expansion boomed. As absentee corporate growers realized the profit potential for

“America’s Salad Bowl,” the nation’s strongest vegetable industry emerged (Anderson

2000; Thompson and Reynolds 2002). Monterey County is the number one crop

producing county in the United States (USDA 2002), partially because growers in the

lower Salinas Valley utilize coastal microclimates, rich soils, and groundwater to produce

up to three crop rotations per year. In 2002, Monterey County’s $2.16 billion in crops

accounted for 11% of California’s crop product, and over two percent of the U.S. national

crop product (USDA 2002). Monterey County is the nation’s largest producer of lettuce,

spinach, broccoli, artichokes, assorted berries, and nuts (USDA 2002).

6

This region’s dynamic and industrialized agriculture is largely dependant on

groundwater irrigation from the valley’s interbedded alluvial aquifer system (Planert and

Williams 1995). Reduced hydraulic pressure and seawater intrusion were first

documented in the ‘180-foot’ aquifer in the mid-1930’s (Figure 8; Thompson and

Reynolds 2002). Seawater began intruding into the ‘400-foot’ aquifer (not shown) by the

late 1950s (Anderson 2000, MCWRA 2006). The Salinas Valley’s alluvial aquifers are

reported to be recharged primarily by precipitation and the subsequent infiltration of the

Salinas River and its tributaries (Planert and Williams 1995). Consequently, there is a

strong seasonal variation in hydrostatic pressure in wells that is aggravated by heavy

summer withdraws. Perennial dependability of hydrostatic well pressure (head) has been

a historical problem for Salinas Valley growers. In the mid-20th century three lakes were

constructed in the headwaters of the Salinas River system to modulate dry-season

recharge and slow seawater intrusion with controlled release of water from these

reservoirs (Thompson and Reynolds 2002). These lakes resulted in more dependable

hydrostatic pressure in the upper valley, where groundwater-based agriculture emerged.

As overdraft continued, a complex political divide developed between stakeholders in the

upper and lower basin (Thompson and Reynolds 2002).

Long term overdraft and saltwater intrusion and have already ruined thousands of

acres of agricultural lands and several municipal wells (Westra and Vinton 1992), and as

of 2005, saltwater was approaching the city of Salinas where approximately 80 percent of

the valley’s population lives (Figure 8; Thompson and Reynolds 2002; MCWRA 2006).

Now more than ever, seawater intrusion threatens freshwater supply to hundreds of

thousands of acres of prime farm land and the municipal supply for a population of over

7

200,000 people (MCWRA 2006; Thompson and Reynolds 2002; Westra and Vinton

1992).

Under heavy pressure from the California State Water Control Board, the

Monterey County Water Resources Association (MCWRA) has been developing and

implementing a large scale basin management plan with hopes of finding solutions to the

Salinas Valley’s long standing water resource issues of flood control, seawater intrusion,

and political divide (Thompson and Reynolds 2002; MCWRA 2006). Primary objectives

of this project include stoping seawater intrusion, providing adequate water supplies to

meet current and year 2030 needs, and to improving the hydrologic balance of the

groundwater aquifers in the Salinas Valley. Despite injections of some 37 million cubic

meters (M m3) per year of treated effluent since 1998, overdraft and saltwater intrusion

continue (Figure 8; MCWRA 2006). Well monitoring, geographic information systems,

and refined hydrologic modeling are the principal tools for the MCWRA (2006).

Relatively short records of discharge (USGS 2007) provide limited perspective on

hydrologic regimes in the Salinas Valley. In this study, recently developed blue oak

chronologies were used to augment the instrumental record of Salinas River discharge for

the investigation of long-term hydrologic variability.

DATA and METHODS

Tree-Ring Data

Blue oak chronologies have been developed at 13 locations in the central Coast

Ranges of California (Figure 6; Table 1). Nine new chronologies were developed and

four previously existing blue oak records were extended and updated with methods

8

standard in dendrochronology. At each site, 40 living trees were randomly selected for

non-destructive increment core sampling in a 50 m by 200 m study plot using a modified

point quarter method (Cottom and Curtis 1956; Barbour et al. 1980; Clements-Peppers

2004). Additional old trees and dead wood were sought out and selectively sampled to

improve replication and extend the chronologies farther back in time (Figure 9). When

possible, 10 to 20 additional old trees were cored and 10 to 30 dead trees were sampled

with a chainsaw. Our field surveys indicate that in the absence of intense fire, dead blue

oak wood may withstand deterioration for over 100 years. At Figueroa Mountain, above

Mission Santa Ynez, wood recovered from an axe-cut stump was found to have an inner

ring date of 1292. Recovery of dead wood is critical in the development of the longest

possible blue oak chronologies.

Tree-ring collections were prepared using a sequence of increasingly finer-grit

sandpaper and microfinishing sanding films until the transverse surface was finely

polished. A perfectly polished surface is necessary to distinguish the microscopic

anatomy of annual rings in this slow growing oak. Given the complicated cellular

structure of blue oak at drought stressed sites, an annotated illustration and a careful

anatomical description of annual rings have been provided in Figure 10.

Specimens were examined with low magnification reflecting light microscopes,

and annual ring widths were evaluated using the skeleton plot method described by

Douglass (1941) and Stokes and Smiley (1996). Samples from multiple trees were

graphically compared and ring-width signatures were cross-matched back through time.

Crossdating issues associated with complex ring anatomy, false, or locally absent rings

were resolved and a graphical composite chronology was developed by averaging

9

skeleton plots from multiple trees. When cross matching of ring-width patterns was clear

and replicated among trees at a site, and then among sites in a region, calendar-year

dating was assigned to each annual growth ring. The outermost ring of living trees,

known to be from the present or most recent growing season, was used as the datum from

which growth was cross-synchronized between living and dead trees backwards through

time. As a final confirmation of accuracy, the new chronologies were compared with the

existing blue oak chronologies that were developed in the mid-1990’s at the University of

Arkansas Tree-Ring Laboratory (ITRDB 2007).

Dated ring-widths were measured under the microscope with a Velmex stage

micrometer (at 0.001mm precision). Program COFECHA (Holmes 1983; Grissino-

Mayer 2001) was used for crossdating quality control. This software correlates segments

of measured ring-width time series to identify possible dating and measurement

problems, and to assess cross dating signal strength. Potential errors flagged by

COFECHA were investigated and corrected if necessary. In some cases, if individual

trees were poorly correlated with all the others, they were excluded from the ring-width

dataset.

Using standard methods (Cook and Kairiukstis 1990), tree-ring chronologies were

computed from the raw ring-width data. Ring-width time series were detrended and

standardized to remove non-climatic age-related growth trend and differences in mean

growth rate among trees. E.R. Cook’s computer program ARSTAN (Cook 1985; LDEO

TRL 2006) was used to compute four types of mean ring-width chronologies from each

set of measured ring width data. The first type, the so called “raw” ring-width

chronology is simply mean annual ring width (in mm) without any detrending or

10

numerical treatment. In producing the other three types of chronologies, ARSTAN uses

empirical curve fitting to detrend each individual ring-width time series. First a negative

exponential or straight line was fit to the raw ring-width series using least-squares criteria

(Cook 1985). Ring width series were then detrended and standardized by dividing the

ring width value by the curve fit value (Fritts 2001; Cook 1985). The resulting indices

were detrended again with an inflexible smoothing spline designed to remove 67% of the

variance in a theoretical sine wave equal to the total length in each series (Cook and

Peters 1981). The “double detrended” ring-width indices have a mean of approximately

1.0 and were averaged with equal weight into the mean ring-width (so-called “standard”)

chronology.

For the third type, the so called “residual” chronology, ARSTAN uses

autoregressive modeling (Box and Jenkins 1976) to remove low order growth persistence

from the detrended and indexed ring-width data that is theorized to be of biological origin

(Cook 1985). This physiological memory can result in a lagged relationship where a

given year’s ring width is in some part a function of the previous year’s or years’

production and storage of energy reserves. This lagged growth response or

autocorrelation can extend over several years in some species (Cook 1985; 1987), but is

typically not present in most precipitation or streamflow data and it is desirable to remove

this biological effect prior to climatic reconstructions. Autocorrelation also violates some

of the statistical assumptions involved with the regression methods often used for climate

reconstruction from tree-rings (Fritts 2001, Cook 1985). ARSTAN models individual

ring width series to remove this autocorrelation, and produces so called “pre-whitened”

time series in which successive values have no autocorrelation. The result is the so

11

called “residual” chronology, which is the robust mean value function of the detrended,

indexed, and pre-whitened individual ring-width time series.

The fourth type is the so-called “ARSTAN” chronology. This time series is the

result of “reddening” the residual chronology by adding the modeled autoregressive

structure back into the residual chronology. The four types of tree-ring chronologies, the

raw ring-width, standard, residual, and ARSTAN chronologies were created for each of

the 13 blue oak collections from the central Coast Ranges. The four types of numerical

tree-ring chronologies and the component numerical ring-width series for each of the 13

sites will soon be available through the International Tree-Ring Data Bank (ITRDB

2007).

While the blue oak chronologies from the Central Coast Ranges generally are

more sensitive to precipitation variability than most other California blue oak records, a

few of these chronologies stand out as some of the most sensitive in the network (i.e.,

Mount Diablo, Pacheco Pass, Rock Springs Ranch, Pinnacles, and Los Lobos). These

chronologies are significantly correlated amongst each other (Table 2) and with Salinas

discharge records.

Residual tree-ring chronologies were selected for this study because any non-

climatic persistence structure was autoregressively modeled and removed (Cook 1985;

1987), and serially random residual chronologies are often better predictors for serially

random hydroclimate data. The four longest residual blue oak chronologies (Figure 11;

Table 1) were selected to maximize reconstruction length. None of these study sites were

located in the Salinas drainage basin, however the regional average chronology is

reflective of regional scale interannual moisture variability in the Central Coast Ranges.

12

Annual values were simply averaged for the common period, resulting in a regional

chronology with robust sample size through time (Figure 11; Appendix 1). The regional

average chronology is 595 years long (1409-2003), and exhibits regionally coherent

variability at interannual to multi-decadal scales. The ring-width indices of the regional

average chronology are normally distributed, and are not serially correlated. This

regional average chronology was used as the predictor for reconstructing Salinas River

discharge.

Streamflow Data

Five U.S. Geological Survey (USGS) gauges continuously measure flow on the

Salinas River, and monthly mean discharge time series are available online (Figure 12,

Table 3; USGS 2007). Water year mean flow records from the Salinas River and its

tributaries (not shown) are all highly correlated (Table 4), indicating basin-wide

coherence in inter-annual flow variability. USGS Research Hydrologist Michael

Dettinger (2004) recommended the Paso Robles gauge record for flow reconstruction

because it is upstream from more human interference than any other on the Salinas River.

The Paso Robles record is the second longest on the Salinas River (1940-present; Table

3), and the gauge is located in the upper valley where the Salinas is a gaining stream

(Planert and Williams 1995). A relatively small reservoir, Santa Margarita Lake, was

built upstream in 1941 and regulates low flows at Paso Robles (USGS 2007).

Monthly discharge statistics were obtained for the water years 1939-2003,

excluding 1966-1969 for which data were missing (Appendix 2; USGS 2007). Imperial

units of discharge (cubic feet per second) were converted to metric units (cubic meters

13

per second) and scaled to cumulative monthly and water year flow in million cubic

meters (M m3). Mean monthly Salinas River discharge at Paso Robles reflects the

region’s Mediterranean precipitation regime, with peak discharge in January, February,

and March (Figure 13a). The time series of water year discharge is serially random and

exhibits dramatic inter-annual variability, including notably wet and dry years (Figure

14). Sub-decadal drought variability is also represented by two persistent dry periods

(1944-1951, 1987-1992). The distribution of water year flows is positively skewed.

Double-mass analysis (Kohler 1949) was used to compare the Paso Robles data with the

long discharge record from Arroyo Seco. The results indicated no serious heterogeneity

in the Paso Robles flow data through time.

Calibration and Verification

Correlation analysis was used to model the relationship between the residual

chronologies and monthly, seasonal, and annual mean Salinas flow at Paso Robles.

These results indicate that the chronologies are generally best correlated with monthly

flows from January-June (Figure 13b), and that the chronologies are all significantly

correlated with water year cumulative discharge (Figure 15a).

Bivariate regression was used to calibrate the regional average blue oak

chronology indices with water year flow on the Salinas at Paso Robles (Table 5). A split

period analysis was used to assess the strength of the tree growth-streamflow relationship

through time. First, a calibration model was fit to the data from the early subperiod 1940-

1965:

ŷt = -151.38+232.44xt (1)

14

where ŷt is the estimated Salinas River discharge in year t, and xt is the regional average

tree-ring index value for year t. This model was then verified against the data of the late

subperiod (1970-2003; Tables 5 and 6). The halves were switched, and a calibration

model was fit to the late subperiod 1970-2003:

ŷt = -159.63+264.13xt (2)

The derived reconstruction was then verified against the observed data for the early time

period. Verification tests performed included the Pearson correlation, paired t-test, sign

test, cross products t-test, and the reduction of error and coefficient of efficiency statistics

(Tables 5 and 6; Fritts 2001; Steel and Torrie 1980; Draper and Smith 1981).

Verification results from the split period calibration exercise were satisfactory for both

subperiods (Table 6). Finally, a calibration model was fit to the entire period of overlap

between the instrumental and tree-ring records (Figure 15b):

ŷt = -155.84+250.10xt (3)

This calibration model appeared to be robust, with random regression residuals that

approximated a normal distribution (Table 5).

RESULTS and DISCUSSION

The blue oak based reconstruction explains 70% of the variance in observed water

year discharge on the Salinas River at Paso Robles (Figure 15b). The reconstruction

closely follows the instrumental record of drought, including the epic 1976-1977 event,

and the prolonged periods of deficit flow in the 1940’s and late 1980’s. Unfortunately,

the magnitude of peak flows observed at Paso Robles are not well reproduced,

particularly in 1941 (410 M m3) and 1983 (469 M m3). The highest reconstructed flow is

15

only 335 M m3 (1825; Appendix 2). During wet years, the timing and intensity of

precipitation and related surface runoff must affect the relationship between tree growth

and streamflow. Additionally, factors unrelated to moisture availability may limit tree

growth in wet years (e.g., soil nutrients). Because of these underestimations of observed

flow volume during the extremely wet years, the reconstruction probably also

underestimates the magnitude of other peak flows during the pre-instrumental period.

Nevertheless, the reconstruction is reasonably consistent with distribution and frequency

of observed high and low flows (Figures 15b and 16).

The full reconstruction (Figure 17) illustrates six centuries of dramatic interannual

variability in Salinas River discharge. For a historical perspective on reconstructed

drought, the lowest single year and n-year running mean flows (for n = 1, 3, 6, 10, 20,

and 30 years) are listed in Table 7. The same information for wet years is provided in

Table 8. These tabulations indicate that low flows are common on the Salinas River and

in the most extreme drought years, the Salinas runs dry at Paso Robles. Nineteen percent

(115 of 595 years) of annual flows were reconstructed as less than 10 M m3, and 16% (98

of 595 years) of values were reconstructed as zero flow. These zero year flow events are

well distributed throughout the reconstruction, but the highest flows are not distributed as

evenly. Fifteen of the 30 wettest years occur in the 19th and 20th Centuries (Table 8),

indicating an increase in the magnitude of surplus discharge years after 1800.

Exploratory analysis suggests that this change towards higher magnitude peak

flows was not an artifact of chronology development with respect to inflated sample size,

tree age, or detrending methods. This increase in high flow magnitude may be reflective

of low-frequency changes in the precipitation delivery mechanisms for California’s

16

Central Coast Ranges. Although these tree-ring data were collected in plant communities

with old-growth blue oak canopies, it is also reasonable to consider that this variance

trend may be associated with historical ecosystem disturbances with respect to

herbaceous ground cover changes. Allen et al. (in press) suggest that the replacement of

native perennials by exotic annual grasses causes reduced soil moisture late in the

growing season.

Extreme single year events are common in both the instrumental and

reconstructed time series, but persistent departures from normal may have more profound

socioeconomic implications. The ten-year smoothing spline emphasizes lower frequency

moisture variability, illustrating a long history of quasi-periodic climate rhythms in the

Salinas Valley (Figure 17). Several prolonged departures from around mean flow

conditions stand out, including periods of persistent drought and wetness more intense

than those exhibited in the instrumental record from 1940 to present (Figure 15b; Table

7).

The most persistent drought of the last 600 years, as indicated by the blue oak

trees, began in 1917 and lasted through 1934. Discharge was below average in 15 of 18

consecutive years, and mean flow for the period was only 53% of normal. Ironically, this

period coincides with the dawn of industrial scale agriculture, improvements in

groundwater extraction techniques, and the first signs of seawater intrusion in the Salinas

Valley. This remarkably dry time was preceded and followed by notably wet periods

(Figure 17). These events, particularly the drought, had widespread impacts in the

Salinas Valley, and were described by John Steinbeck. In the opening pages of East of

17

Eden, Steinbeck (1952) offers commentary on the cyclical nature and socioeconomic

effects of persistent climate extremes:

I have spoken of the rich years when the rainfall was plentiful. But there were dry years too, and they put a terror on the valley. The water came in a thirty-year cycle. There would be five or six wet and wonderful years when there might be nineteen to twenty-five inches of rain, and the land would shout with grass. Then would come six or seven pretty good years of twelve to sixteen inches of rain. And then the dry years would come, and sometimes there would be only seven or eight inches of rain. The land dried up and the grasses headed out miserably a few inches high and great bare scabby places appeared in the valley. The live oaks got a crusty look and the sage-brush was gray. The land cracked and the springs dried up and the cattle listlessly nibbled on dry twigs. Then the farmers and the ranchers would be filled with disgust for the Salinas Valley. The cows would grow thin and sometimes starve to death. People would have to haul water in barrels to their farms just for drinking. Some families would sell out for nearly nothing and move away. And it never failed that during the dry years the people forgot about the rich years, and during the wet years they lost all memory of the dry years. It was always that way.

Analysis of the discharge reconstruction indicates a significant (p < 0.01) spectral peak at

around 23 years which may account for up to five percent of the variance in the time

series (Figure 18; Jenkins and Watts 1968). The “30-year cycle” Steinbeck describes is

epitomized by the persistent events of the early 20th century, which appear to have been

unusual for the last 600 years.

Other decadal scale departures from average conditions stand out in the

reconstruction (Figure 17). The most persistent wet period in the last six centuries lasted

from 1738-1750. Mean discharge for the 13 year period was 156% of average. This wet

period was closely followed by the second most extreme drought event in the last six

centuries, a persistent drought that lasted from 1751 through 1765. Discharge in 14 of 15

consecutive years was below average, and mean flow for the period was only 52% of

normal.

Single year extremes of drought and wetness dominate the reconstruction, but it is

consecutive year drought events that most severely impact water storage systems (e.g.,

Lynch 1931; Jones 2006). Consecutive year events of surplus or deficit discharge,

18

respectively defined as greater than 120% and less than 80% of median flow are

illustrated in Figure 19. Return time intervals between events are skewed, indicating that

consecutive year events, both wet and dry, have a tendency to cluster. Note the clusters

of consecutive year droughts in the mid-1400’s, early-1500’s, mid-1600’s, and 1900’s.

Consecutive year droughts were more frequent during the 20th century than any other 100

year period in the reconstruction. Low-frequency changes in the magnitude and

frequency of consecutive year surplus flow events are also evident. Again, the 20th

century is unique with respect to the preceding 500 years: consecutive year events with

surplus flow exhibit a higher magnitude, but a reduced frequency.

Cook et al. (1999) and Cook et al. (2004) have utilized a network of over 835

tree-ring chronologies to reconstruct a 2.5o latitude by 2.5o longitude grid of June-August

Palmer Drought Severity Index (PDSI; Palmer 1965) over North America, and these data

are available online. This dataset has been used to analyze continental scale moisture

conditions with respect to the most extreme years of Salinas Valley climate (Figure 20).

Composite maps were generated for the 10 individual wettest and 10 individual driest

years as indicated by the Salinas discharge reconstruction from 1645 to 2003, when all

286 gridpoints over North America were fully populated (Tables 7 and 8). During the 10

wettest years in the Salinas Valley, wet conditions were focused over California, but a

weak cell of drought was centered over the south central United States (Figure 20a).

During the 10 driest years in the Salinas Valley, intense drought is centered over southern

California, northern Baja California, and southern Nevada, while modest drought

extended generally across the continental United States (Figure 20b). This drought

pattern resembles one of the major modes of North American drought variability as

19

identified in instrumental (Karl and Koscielny 1982) and reconstructed PDSI (Cook et al.

1999).

The dynamics of ocean-atmospheric forcing on climate and streamflow variability

over the western U.S. is one of the hottest topics in climate science (e.g., Andrews et al.

2004; Cayan et al. 1998; Cayan et al. 1999; Dettinger et al. 1998; Dettinger et al. 2000;

Haston and Michaelsen 1997; Seager et al. 2007). Kaplan et al. (1998) have analyzed

instrumental records of sea surface temperature (SST) to develop a dataset of 1856-

present monthly temperature anomalies relative to averages for 1951-1980, presenting the

opportunity to investigate SST anomalies associated with extreme hydroclimatic

conditions. This data was used to map average sea surface temperatures during the 10

years of highest and lowest Salinas River discharge, as reconstructed since 1856. The

composite map for wet Salinas River flow extremes exhibits an El Niño-like pattern, with

warm waters in the eastern equatorial Pacific and off the west coast of the Americas

(Figure 21a). These results provide new evidence that the years of highest streamflow in

central California are associated with El Niño, as has been reported in numerous other

studies (e.g., Andrews et al. 2004; Dettinger and Diaz 2001; Cayan et al. 1999; Haston

and Michaelsen 1994). The composite map for dry years does not exhibit a strong pattern

of Pacific SST variability (Figure 21b). However, as described by Trewartha (1981) and

Knapp et al. (2004), cool SSTs north of Hawaii and warm water off the coast of the

Pacific Northwest would favor a strong North Pacific anticyclone (the Hawaiian High)

extending into western North America. This blocking high pressure ridge would reduce

the possibility for moisture advection into California (Trewartha 1981).

20

CONCLUSIONS

Tree-ring specimens were collected from over 900 living and dead blue oak trees

from old-growth oak woodlands in California’s Central Coast Ranges. These specimens

have been carefully prepared and permanently dated with the methods of

dendrochronology, and are available for future study through the University of Arkansas

Museum. The new blue oak chronologies are 300- to over 700-years long and

collectively reflect regional scale moisture history for the Salinas Valley. Dead wood

recovered from the forest floor proved critical in developing these long chronologies, and

it should be noted that many of the shorter blue oak records might be extended further

back in time with more exhaustive fieldwork based exploitation of such material. The

numerical tree-ring chronologies and the untreated component ring-width time series will

soon be publicly available through the International Tree-Ring Data Bank (ITRDB 2007).

These blue oak chronologies will be useful in reconstructing a spectrum of precipitation-

influenced environmental variables, and may improve perspective on the effects of

climate in California’s colorful socioeconomic history.

The Salinas Valley has an interesting historical record of drought and wetness

(e.g. Lynch 1931; Tavernetti 1947; Anderson 2000), but unfortunately records of

streamflow are relatively short. This reconstruction of Salinas River water year discharge

is nearly ten times longer than the instrumental record at Paso Robles, and constitutes the

first tree-ring based estimation of discharge in the Salinas Basin. The numerical time

series of observed and reconstructed discharge are available online through the National

Oceanic and Atmospheric Administration California Treeflow webpage (Griffin 2006).

The reconstruction was calibrated from the simple average of four residual (serially-

21

random) tree-ring chronologies proximate to the Salinas Basin. The regional average

chronology is well correlated with other gauging stations in the Salinas River system

(including the long record at Arroyo Seco) and with precipitation at Salinas, indicating

that the new blue oak data could be employed to develop a suite of hydroclimatic

reconstructions for modeling long-term history of aquifer recharge and water balance in

the lower Salinas Valley.

The reconstruction closely follows the instrumental gauge record at interannual to

sub-decadal timescales. Unfortunately, as is often the case, this tree-ring reconstruction

does not match the full magnitude of the most extreme high flows observed at Paso

Robles and may also underestimate the magnitude of pre-instrumental wet extrema. The

Salinas River often runs dry at Paso Robles, and the full range of drought variability

reconstructed from in the tree-ring data is truncated in 16% of years. Because these

chronologies are well correlated with regional and individual station rainfall records that

do not bottom out at zero, calibrating the regional average chronology with available

precipitation data would result in a more detailed record of the most extreme droughts in

the Salinas Valley history. Despite these details, the Salinas discharge reconstruction

provides a long paleoclimatic perspective, and points to flow extrema more persistent

than those represented in the instrumental record at Paso Robles.

This study suggests that in the Salinas Basin, the 20th century was unusual with

respect to climate variability during the last 600 years. Consecutive year deficit flow

events were more common during the 20th century, and the most persistent drought in the

entire reconstruction, centered around the 1920’s, coincides with the first records of

seawater intrusion in the lower Salinas Valley. The magnitude of peak flow appears to

22

have been higher after 1830, however, the frequency of consecutive year surplus flow

events was lower in the 20th century.

Climate fluctuations, particularly sustained drought, ought to be considered in the

planning of water resource management. It should be noted that the Salinas Valley’s

most persistent drought extremes, as indicated by old-growth blue oak trees, occurred

before the potential for groundwater irrigation and industrial scale agricultural

development had been fully realized. A modern drought of comparable magnitude and

persistence would have severe implications for freshwater supply, groundwater recharge,

seawater intrusion, and agricultural production in the Salinas Valley.

23

References Cited

Allen-Diaz, B., Standiford R, Jackson RD. in press. Oak woodlands and forests. in Terrestrial Vegetation of California. Editor M.G. Barbour. UC Press, Berkeley, CA.

Anderson, B. 2000. The Salinas Valley: A History of America’s Salad Bowl. Monterey

County Historical Society. 208 pp. Andrews, E.D., Antweiler, R.C., Neiman, P.J., and Ralph, F.M. 2004. Influence of ENSO

on flood frequency along the California Coast. Journal of Climate 17:337-348. Barbour, M., Burk, J., and Pitts, W. 1980. Terrestrial Plant Ecology. Benjamin

Cummings Publishing Co., Menlo Park, Ca. 604 pp. Box, G.E.P., and Jenkins, G.M. 1976. Time Series Analysis: Forecasting and Control.

Holden Day, 576 pp. Brito-Castillo, L., Diaz-Castro, S., Salinas-Zavala, C.A., and Douglas, A.V. 2003.

Reconstruction of long-term winter streamflow in the Gulf of California continental watershed. Journal of Hydrology 278: 39-50.

Case, R.A., and MacDonald, G.M. 2003. Tree ring reconstructions of streamflow for

three Canadian Prairie rivers. Journal of the American Water Resources Association 39: 703-716.

Cayan, D.R., Dettinger, M.D., Diaz, H.F., and Graham, N.E. 1998. Decadal variability of

precipitation over western North America. Journal of Climate 11: 3148-3166. Cayan, D.R., M.D. Dettinger, K.T. Redmond, G.J. McCabe, N. Knowles and D.H.

Peterson, D.H., 2003: The transboundary setting of California's water and hydropower systems: linkages between the Sierra Nevada, Columbia River, and Colorado River hydroclimates. In Diaz, H.F. and B. Woodhouse (eds.), Climate and Water - Transboundary challenges in the Americas, Kluwer Academic Publishers. pp. 237-262.

Cayan, D.R., Redmond, K.T., and Riddle, L.G. 1999. ENSO and hydrologic extremes in

the western United States. Journal of Climate 12:2881-2893. Clements-Peppers, K. 2004. Old-Growth Forests in the Western Cross Timbers of Texas.

Ph.D. dissertation, University of Arkansas. Fayetteville, Fayetteville, AR. 171 pp. Cook, E.R. 1985. A Time Series Analysis Approach to Tree Ring Standardization. Ph.D.

dissertation, University of Arizona, Tucson, Tucson, AZ. 171 pp.

24

Cook, E.R. 1987. The decomposition of tree-ring series for environmental studies. Tree- Ring Bulletin 47:37-59.

Cook, E.R. and Jacoby, G.C. 1983. Potomac River streamflow since 1730 as

reconstructed by tree rings. Journal of Climate and Applied Meteorology 22:1659-1672.

Cook, E.R. and Kairiukstis, L.A. 1990. Methods of Dendrochronology. Dordrecht,

Kluwer Academic Publishers. 394 pp. Cook, E.R. and P.J. Krusic. 2004. The North American Drought Atlas. Lamont-Doherty Earth Observatory and the National Science Foundation. Data available online:

http://ingrid.ldeo.columbia.edu/SOURCES/.LDEO/.TRL/.NADA2004/.pdsi- atlas.html. Last accessed April 10, 2007.

Cook, E.R., Meko, D.M., Stahle, D.W., and Cleaveland, M.K. 1999. Drought

reconstructions for the continental United States. Journal of Climate 12:1145-1162.

Cook, E.R. and Peters, K. 1981. The smoothing spline: a new approach to standardizing

forest interior tree-ring width series for dendroclimatic studies. Tree-Ring Bulletin 41: 45-53.

Cook, E.R., Woodhouse, C.A., Eakin, C.M, Meko, D.M, and Stahle, D.W. 2004. Long-

term aridity changes in the western United States. Science 306:1015-1018. data available online: http://www.ncdc.noaa.gov/paleo/pdsi.html. Last accessed April 11, 2007.

Cottom, G., and Curtis, J.T. 1956. The use of distance measures in phytosociological

sampling. Ecology 37:451-460. Cleaveland, M.K., and Stahle, D.W. 1989. Tree-ring analysis of surplus and deficit runoff

in the White River, Arkansas. Water Resources Research 25:1391-1401. Cleaveland, M.K. 1999. A 963-year reconstruction of summer (JJA) streamflow in the

White River, Arkansas, USA, from tree-rings. The Holocene 10:33-41. Dettinger, M.D., Cayan, D.R., Diaz, H.F., and Meko, D.M. 1998. North-south

precipitation patterns in western North American on interannual-to-decadal time scales. Journal of Climate 11:3095-3111.

Dettinger, M.D., and Diaz, H.F. 2000. Global characteristics of streamflow seasonality

and variability. Journal of Hydrometeorology 1:289-310.

25

Dettinger, M.D., McCabe, G.J., and Morego, J.A. 2000. Multiscale hydrologic variability associated with El Niño/Southern Oscillation. In El Niño-Southern Oscillation, Diaz, H.F., and Markgraf, V. Editors. Cambridge University Press. pp 113-148.

Dettinger, M.D, 2004. U.S. Geological Survey, Scripps Institution of Oceanography, UC

San Diego, 9500 Gilman Drive La Jolla, CA, 92093. personal communication. Douglass, A.E. 1941. Crossdating in dendrochronology. Journal of Forestry 39:825-831. Draper, N.R. and Smith, H. 1981. Applied Regression Analysis (second edition). John

Wiley, New York. 709 pp. Earle, C.J. 1993. Asynchronous droughts in California streamflow as reconstructed from

tree rings. Quaternary Research 39:290-299. Fritts, H.C. 2001. Tree-Rings and Climate. Blackburn Press, Caldwell, NJ. 576 pp. Gedalof, Z., Peterson, D.L., and Mantua, N.J. 2004. Columbia River flow and drought

since 1750. Journal of the American Water Resources Association 40:1579-1592. Graumlich, L.J., Pisaric, M.F.J., Waggoner, L.A., Littell, J.S., and King, J.C. 2003. Upper

Yellowstone River flow and teleconnections with Pacific basin climate variability during the past three centuries. Climatic Change 59: 245-262.

Griffin, J.R., and Critchfield, W.B. 1972. The Distribution of Forest Trees in California.

U.S. Forest Service Research Paper PSW-82. 114 pp. Griffin, R.D.. 2006. Salinas River Annual Flow Reconstruction. International Tree-Ring

Data Bank. IGBP PAGES/World Data Center for Paleoclimatology Data Contribution Series #2006-103. NOAA/NGDC Paleoclimatology Program, Boulder CO, USA. Available online: http://www.ncdc.noaa.gov/paleo/streamflow/ca/salinas.htmlLast accessed March 30, 2007

Griffin, R.D., and Stahle, D.W. 2005. Ancient Blue Oak Predictive Model Design and

Testing. Unpublished report by the University of Arkansas Tree-Ring Laboratory. 20 pp. Available online: http://uark.edu/blueoak/data/TASK%203,%20BO_MODEL_report.pdf. Last accessed March 5, 2007.

Grissino-Mayer, H.D. 1996. A 2129-year reconstruction of precipitation for northwestern

New Mexico, USA. In Tree Rings, Environment, and Humanity. Dean, J.S., Meko, D.M., and Swetnam, T.W., Editors. pp 191-204.

Grissino-Mayer, H.D. 2001. Evaluating crossdating accuracy: a manual and tutorial for

the computer program COFECHA. Tree-Ring Research 57:205-221.

26

Haston L, and Michaelsen J. 1994 Long-Term Central Coastal California Precipitation Variability and Relationships to El Niño-Southern Oscillation. Journal of Climate 7: 1373–1387

Haston, L. and Michaelsen, J. 1997. Spatial and temporal variability of southern

California precipitation over the last 400 years and relationships to atmospheric circulation patterns. Journal of Climate 10:1836-1852.

Holmes, R.L. 1983. Computer-assisted quality control in tree-ring dating and

measurement. Tree-Ring Bulletin 43:69-78. Holmes, R.L., Stockton, C.W., and LaMarche, Jr., V.C. 1979. Extension of riverflow

records in Argentina from long tree-ring chronologies. Water Resources Bulletin 15:1081-1085.

International Tree-Ring Data Bank (ITRDB) 2007. Contributors of the International

Tree-Ring Data Bank, IGBP PAGES/World Data Center for Paleoclimatology, NOAA/NCDC Paleoclimatology Program, Boulder, Colorado, USA. Website: http://www.ncdc.noaa.gov/paleo/treering.html. Last accessed March 16, 2007.

Jain, S., Woodhouse, C.A., and Hoerling, M.P. 2002. Multidecadal streamflow regimes in

the interior western United States: Implications for the vulnerability of water resources. Geophysical Research Letters 29:321-324.

Jenkins, G.M., and Watts, D.G. 1968. Spectral Analysis and its Applications. Holden-

Day. 524 pp. Jones, J. 2006. California Department of Water Resources, Box 942836 Sacramento, CA

94236. personal communication. Jones, P.D., Briffa, K.R., and Pilcher, J.R. 1984. Riverflow reconstruction from tree rings

in southern Britain. Journal of Climatology 4:461-472. Karl, T.R., and Koscielny, A.J. 1982. Drought in the United States. Journal of

Climatology 2:313-329. Kaplan, A., Cane, M.A., Kushnir, Y., Clement, A.C., Blumenthal, M.B., and

Rajagopalan, B. 1998. Analysis of global sea surface temperature 1856-1991. Journal of Climatology 2: 18 567-18 589. Data available online: http://ingrid.ldeo.columbia.edu/SOURCES/.KAPLAN/.RSA_MOHSST5.cuf/.dataset_documentation.html. Last accessed April 10, 2007.

Knapp, P.A., Soule, P.T., and Grissino-Mayer, H.D. 2004. Occurrence of sustained

droughts in the interior Pacific Northwest (A.D. 1733-1980) inferred from tree-ring data. Journal of Climate 17:140-150.

27

Kohler , M.A. 1949. On the use of double-mass analysis for testing the consistency of

meteorological records and for making required adjustments. Bulletin of the American Meteorological Society 30:188-189.

Lara, A., Urrutia, R., Villalba, R., Luckman, B.H., Soto, D., Aravena, J.C., McPheef, J.,

Wolodarsky, A., Pezoa, and L., León, J. 2005. The potential use of tree-rings to reconstruct streamflow and estuarine salinity in the Valdivian Rainforest eco-region, Chile . Dendrochronologia 22(3): 155-161.

Lanner, R.M. 1999. Conifers of California. Cachuma Press, Los Olivos, CA. 274 pp. Lamont Doherty Earth Observatory Tree-Ring Laboratory (LDEO TRL) 2006. website:

http://www.ldeo.columbia.edu/res/fac/trl/public/publicSoftware.html. Last accessed December 17, 2006.

Lynch, H.B., 1931. Rainfall and stream runoff in southern California since 1769. Los

Angeles, California: The Metropolitan Water District of Southern California, 31 pp.

McDonald, 1990. Quercus douglasii Hook. & Arn. – Blue Oak. In Silvics of North

America, Vol. 2, Hardwoods. R.M. Burns and B.H Honkala. (Tech. Coords.) USDA Forest Service, Agric. Handbook. pp. 490-495.

Meko, D.M, Stockton, C.W., and Boggess, W.R. 1995. The tree-ring record of severe

sustained drought. Water Resources Bulletin 31:789-801. Meko, D.M., and Graybill, D.A. 1995. Tree-ring reconstruction of Upper Gila River

discharge. Water Resources Bulletin 31:605-616. Meko, D.M. 2001. Reconstructed Sacramento River System Runoff from Tree Rings.

Report for the California Department of Water Resources under agreement no. B81923-SAP#4600000193. Laboratory of Tree-Ring Research, University of Arizona, Tucson. 41 pp.

Meko, D.M. 2006. Laboratory of Tree-Ring Research, University of Arizona, Tucson

personal communication. Meko, D.M., and Woodhouse, C.A. 2005. Tree-ring footprint of joint hydrologic drought

in Sacramento and Upper Colorado river basins, western USA. Journal of Hydrology 308:196-213.

28

Meko, D.M., Therrell, M.D., Baisan, C.H., and Hughes, M.K., 2001, Sacramento River flow reconstructed to A.D. 869 from tree rings. Journal of the American Water Resources Association 37:1029-1040.

Monterey County Water Resources Association (MCWRA). 2006. website:

http://www.mcwra.co.monterey.ca.us/ Last Accessed December 17, 2006. Norton, D.A. 1987. Reconstruction of past riverflow and precipitation in Canterbury,

New Zealand, from analysis of tree rings. Journal of Hydrology 26:161-174. Palmer, W.C., 1965. Meteorological Drought. Weather Bureau Research Paper 45, U.S.

Department of Commerce, Washington D.C. 58 pp. Pavlik, B.M., Muick, M.C., Johnson, S.G., and Popper, M. 1991. The oaks of California.

Cachuma Press, Los Olivos, CA. 184 pp.

Phipps, R.L. 1983. Streamflow of the Occoquan River in Virginia as reconstructed from tree-ring series. Water Resources Bulletin 19:735-743.

Planert, M., and Williams, J.S. 1995. Groundwater Atlas of the United States: California

Nevada. Professional Report Ha 730-B. U.S. Geological Survey, Reston, VA. Redmond, K.T., Stahle, D.W., Therrell, M.D., Cayan, D.R., and Dettinger, M.D., 2002,

400 years of California Central Valley precipitation reconstructed from blue oaks, Proceedings, January 2002, 13th Symposium on Global Change and Climate Variations. American Meteorological Society, Orlando, FL, pp. 20-23.

Schweingruber, F.H., 1990. Microscopic Wood Anatomy. Swiss Federal Institute for

Forest, Snow, and Landscape Research, Birmensdorf, Switzerland. 226 pp. Schweingruber, F.H., Borner, A., and Schulze, E.-D. 2006. Atlas of Woody Plant Stems.

Springer-Verlag. Berlin. 229 pp. Seager, R., Ting, M.F., Held, I.M, Kushnir, Y., Lu, J., Vecchi, G., Huang, H, Harnik, N.,

Leetmaa A., Lau N.-C., Li C., Velez, J., and Naik,. N. 2007. Model projections of an imminent transition to a more arid climate in southwestern North America. Science DOI: 10.1126/science.1139601

Shen, H.S., and Tabios, III, G.Q. 1995. Drought analysis with reservoirs using tree-ring

reconstructed flows. Journal of Hydraulic Engineering 12:413-421. Smith, L.P., and Stockton, C.W. 1981. Reconstructed streamflow for the Salt and Verde

Rivers from tree-ring data. Water Resources Bulletin 17:939-947. Stahle, D.W. 1996. Tree rings and ancient forest history. In: Eastern Old-Growth Forests,

edited by M.B. Davis, Island Press, Washington D.C., pp. 321-343

29

Stahle, D.W. 2002. The unsung ancients. Natural History February 2002. pp. 49-52. Stahle, D.W., and Cleaveland, M.K. 1992. Reconstruction and analysis of spring rainfall

over the Southeastern U.S. for the past 1000 years. Bulletin of the American Meteorological Society 73: 1947–1961.

Stahle, D.W., Cleaveland, M.K., Griffin, R.D., Spond, M.D., Fye, F.K., Culpepper, R.B.

and Patton, D. 2006. Decadal Drought Effects on Endangered Woodpecker Habitat, Eos 87:121.

Stahle, D.W., Cook, E.R., Cleaveland, M.K., Therrell, M.D., Meko, D.M., Grissino-

Mayer, H.D., Watson, E., and Luckman, B.H. 2000. Tree-ring data document 16th century megadrought over North America. Eos 81:121,125.

Stahle, D.W., Therrell, M.D. , Cleaveland, M.K., Cayan, D.R., Dettinger, M.D., and

Knowles, N. 2001. Ancient blue oaks reveal human impact on San Francisco Bay Salinity. Eos 82:144-145.

Steel, R.G.D. and Torrie, J.H. 1980. Principles and Procedures of Statistics (Second

Edition). New York: McGraw-Hill. 633 pp. Steinbeck, J. 1952. East of Eden. Viking Press, New York. 602 pp. Stephenson, J.R., and Calcarone, G.M. 1999. Southern California mountains and

foothills assessment: habitat and species conservation issues. General Technical Report GTR-PSW-172. Albany CA: Pacific southwest Research Station, Forest Service, U.S. Department of Agriculture. 402 pp.

Stockton, C.W., and Jacoby, G.C.. 1976. Long-Term Surface-Water Supply and

Streamflow Trends in the Upper Colorado River Basin. Lake Powell Res. Proj. Bull. 18. Los Angeles: University of California institute of Geophysics and Planetary Physics. 70 pp.

Stokes, M.A. and Smiley, T.L. 1996. An Introduction to Tree-Ring Dating. University of

Arizona Press. Tucson. 73 pp. Stuart, J.D., and Sawyer, J.O. 2001. Trees and Shrubs of California. University of

California Press, Berkeley. 467 pp. Tavernetti, A.A. 1947. The Rainfall of Monterey County. University of California

Agricultural Extension, Salinas. Trewartha, G.T. 1981. The Earth’s Problem Climates. University of Wisconsin Press,

Madison. 371 pp.

30

Thompson, J.G., and Reynolds, R.J. 2002. Cultural evolution and water management in

the Salinas River Valley. Journal of the American Water Resource Association 38: 1661-1667.

United States Department of Agriculture (USDA). 2002. Census of Agriculture website:

http://www.nass.usda.gov/Census_of_Agriculture/ Last Accessed December 17, 2006.

United States Geological Survey (USGS) 2007. website: http://water.usgs.gov/ Last

Accessed April 10, 2007. Van Pelt, R. 2001. Forest Giants of the Pacific Coast. University of Washington Press,

Seattle. 200 p. Villanueva-Diaz, J., Luckman, B.H., Stahle, D.W., Therrell, M.D., Cleaveland, M.K.,

Cerano-Paredes, J., Gutierrez-Garcia, G., Estrada-Avalos, J., and Jasso-Ibarra, R. 2005. Hydroclimatic variability of the upper Nazas basin: Water management implications for the irrigated area of the Comarca Lagunera, Mexico. Dendrochronologia 22:215-223.

Watson, E., and Luckman, B.H. 2005. An exploration of the controls of pre-instrumental

streamflow using multiple tree-ring proxies. Dendrochronologia 22: 225-234. Westra R., and Vinton, R. 1992. Final Social and Economic Present Condition Report

for the Salinas River Water Resource Management Plan Study. Prepared by the U.S. Bureau of Reclamation, Salinas California.

Woodhouse, C.A. 2001. A tree-ring reconstruction of streamflow for the Colorado Front

Range. Journal of the American Water Resources Association 37: 561-570. Woodhouse, C.A., Gray, S.T., and Meko, D.M. 2006. Updated streamflow

reconstructions for the upper Colorado River Basin. Water Resources Research 42: W05415.

Woodhouse, C.A., and Overpeck, J.T. 1998. 2000 years of drought variability in the

central United States. Bulletin of the American Meteorological Society 79:2693-2714.

31

Figure 1. The natural range of California’s endemic blue oak (Quercus douglasii) is indicated in green, as mapped by Griffin and Critchfield (1972). Blue oak communities are found in 39 of California’s 58 counties. The species occupies the foothills ecotone between low elevation grasslands and higher-elevation mixed conifer forests in the interior foothills of the Coast Ranges, and on the western foothills of the southern Cascades and Sierra Nevada. Disjunct populations occur on Santa Cruz and Santa Catalina islands (not mapped). Intensive field surveys indicate that old-growth blue oak communities are widely distributed throughout this native range.

32

Ffp

a

igure 2. Canopy closure in blue oak dominated communities is highly variable, but generally changes rom open savannas at lower elevations(a, Pacheco State Park) to denser woodlands at more elevated ositions (b, Rock Springs Ranch; e.g., Allen-Diaz et al. in press).

33

a b

Figure 3. Blue oak exhibits a host of moisture conserving mechanisms, including small drought deciduous leaves (Pavlik et al 1991). Blue oak leaves are alternate and simple, and are typically 4 to 10 cm long with five to seven lobes and shallow sinuses (e.g., Stuart and Sawyer 2001). The common name blue oak comes from the blue-green hue of the upper leaf surface (a). The lower leaf surface (b) is a paler green color.

34

Figure 4. Daniel Griffin measures the diameter of an ancient blue oak tree at Figueroa Mountain on the Los Padres National Forest. This tree exhibits readily identifiable physical characteristics of an old growth oak, as described by Stahle (1996), including a missing main leader, heavy limbs, dead limbs, knots, hollow voids, and slight lateral twist of the stem. Blue oak trees typically have this non-commercial growth form and these communities were never logged for industrial scale timber production (Stahle 2002).

35

a

b

Figure 5. Historically, blue oak woodlands were utilized principally for grazing. In recent years, profit incentives have threatened the conversion of old-growth blue oak woodlands for viticulture (a, upper Salinas Valley), and exurban sprawl (Allen-Diaz et al. in press). Blue oak trees have also been harvested locally for firewood (b, Lassen National Forest). This stump has 553 annual growth rings and represents one of the longest lived blue oaks yet documented. The inner and outermost rings have eroded away, so this tree was likely over 600 years old when it was cut.

36

Figure 6. The new network of blue oak chronologies developed from ancient oak woodlands in California (triangles). Most of these 300- to 700- year long chronologies are highly correlated with winter-spring moisture conditions (Meko 2006). Thirteen chronologies in the Central Coast Ranges (Red Triangles, Table 1) are all significantly correlated with discharge records at Paso Robles and other gauging stations on the Salinas River (Figure 12).

37

Figure 7. The Salinas Valley is the largest coastal basin in southern California (Planert and Williams 1995). The watershed drains northwest into Monterey Bay, and is bounded by mountain ranges to the east, south, and west. The Salinas River headwaters are in the La Panza Range at the southern end of the valley. The Salinas’ principal tributaries are the Nacimiento River, San Antonio River, and Arroyo Seco, all of which drain from the Santa Lucia Range, west of the valley. Groundwater-dependent crop agriculture is highly developed in the central and northern Salinas Valley. The four longest blue oak chronologies recently developed in the Central Coast Ranges (red triangles; Table 1) were used to reconstruct discharge at Paso Robles in the upper basin, the gauge which is above more human interference than any other on the Salinas River.

38

Figure 8. Historic seawater intrusion has been mapped for the 180-foot aquifer of the Salinas Valley by the Monterey County Water Resources Association. Reduced hydraulic pressure and seawater intrusion were first documented in the 180-foot aquifer in the mid-1930’s (Thompson and Reynolds 2002), although it is not mapped here until 1944. In 2005, seawater was less than two kilometers from the city of Salinas.

39

a b

Figure 9. Old living trees and dead wood were sought out and selectively sampled to improve replication and extend the chronologies further back in time. Daniel Griffin (a) using a Swedish increment borer to non-destructively extract a pencil thin core specimen from an ancient living blue oak tree, and David Stahle (b) uses a chainsaw to collect a specimen from a standing dead tree.

40

Figure 10. This photomicrograph illustrates the anatomical features of annual rings on the transverse surface of a precisely dated blue oak specimen (1576–1587) from Rock Springs Ranch in the central Diablo Range. This 12 µm thick section was prepared with a Reichert sliding microtome, and was stained with Safranin, so that lignified structures appear red.

Blue oak xylem (wood) is ring porous, heavily lignified, and remarkably dense. Annual rings in oak trees are most easily recognized by the identification of earlywood (EW), latewood (LW), and the boundary of terminal parenchyma cells which demarcate the end of the annual growth ring. Earlywood is characterized by large diameter vessels that congregate in a tangential row at the onset of ring formation, immediately following the distinct boundary of terminal parenchyma in the preceding ring. In particularly dry years (i.e. 1579, 1585) and sometimes late in life, very little, if any, late is added each year. Latewood is principally composed of heavily lignified fiber cells (bright red) and smaller diameter vessels. Axial paratracheal parenchyma cells (blue-gray color) form a matrix around earlywood and latewood vessel structures, often appearing with a “flame-like” distribution in the latewood (Schweingruber 1990). Dark colored terminal parenchyma cells are found in a tangential row at the end of the latewood, marking the end of the annual growth ring.

At drought stressed sites, such as Rock Springs Ranch, blue oak ring-widths are highly variable and sensitive to growing season precipitation. Note the wide rings of 1577 and 1583 and the extremely narrow rings from 1579, and 1585. Crossdating in dendrochronology proves that the growth ring for 1580 is completely missing in this particular sample. The absent ring indicates that this tree’s vascular cambium did not activate locally during the 1580 growing season. To our knowledge, blue oak is one of the only Quercus species shown to produce locally absent or “missing” rings, which may attest to its extreme drought tolerance.

41

Figure 11. The four longest blue oak chronologies from the Central Coast Ranges are presented as graphical time series with annual index values in red, and a smoothing spline version emphasizing 10-year variability in black. These chronologies are the result of numerical detrending and indexing of ring-width measurements (Cook 1985). Resulting ring-width indices have a mean value of about 1.0. Index values above the mean represent wet years and values below the mean represent dry years.

The regional average blue oak chronology computed as a simple average of the four records, presented at bottom, was used as the predictor for reconstructing Salinas River Discharge. The regional average chronology is also highly correlated with discharge at Paso Robles and other gauging stations on the Salinas River. Note the years of poorest tree growth (1444, 1571, 1580, 1585, 1613, 1782, 1795, 1829, 1864, 1877, and 1898) and the years of best tree growth (1441, 1524, 1656, 1825, 1868, 1878, 1941, 1958, and 1983). The regional chronology also reflects regional moisture variability at sub-decadal (e.g., late 1980s), and decadal (e.g., 1920s) timescales.

42

Figure 12. This map illustrates the watershed of the Salinas (light grey, bounded by blue dashes), and the distribution of the five continuous flow gauges maintained by the USGS (2007) on the Salinas River (black circles). These discharge records vary in length, but are all highly inter-correlated (0.93-0.99), indicating basin wide homogeneity in interannual flow variability. Tables 2 and 3 provide metadata about the Salinas River flow records and the correlations between the time series. Discharge data for the Salinas River’s tributaries are also available, including a particularly long record for Arroyo Seco (1902-present).

43

a b

Figure 13. The hydrograph of mean monthly Salinas River discharge (in millions of cubic meters) at Paso Robles from 1940-2004 (a). The monthly distribution reflects the winter-spring precipitation regime of California’s Mediterranean climate, where the water year is defined as the October in the previous year through September of the current year. The regional average blue oak chronology is significantly correlated with monthly (b) and water year (see Figure 15a) discharge at Paso Robles.

44

Figure 14. The time series of cumulative water year discharge at Paso Robles from 1940-2003 (data for 1966-1969 are missing). The record is skewed towards high flows (Figure 15) and lacks serial correlation. Note the dramatic inter-annual variability, extreme wet years (e.g.,1941, 1983), zero-flow years (e.g., 1976-1977), and periods of persistent deficit flow (e.g., late 1940’s, late 1980’s).

45

a

b

Figure 15. Observed water year discharge at Paso Robles is highly correlated with the regional average chronology (a). Simple bivariate regression was used to calibrate the tree-ring data with the flow data (equation 3, b). This reconstruction accounts for 70% of the variance in the instrumental record (adjusted for the loss of one degree of freedom). The reconstruction underestimates the peak flows of 1941 and 1983.

46

a b

Figure 16. The distribution of observed water year flow (a) and reconstructed water year flow (b) on the Salinas at Paso Robles. Both the observed and reconstructed values are skewed towards high flow.

47

Figure 17. Water year discharge on the Salinas River from 1409-2003, reconstructed from blue oak tree rings. Annual flow values are indicated in blue and a smoothing spline emphasizing 10-year variability is illustrated in black. This reconstruction illustrates the high inter-annual variability in Salinas River discharge, including notably wet years of 1441, 1524, 1656, 1825, 1868, 1884, 1914, 1941, 1978, 1983 and 1998. Low flows are much more common on the Salinas River. Sixteen percent (98 years) of values were reconstructed as zero flow years. The 10-year smoothing spline (in black) emphasizes the quasi-periodic nature of drought and wetness as described by Steinbeck (1952). Note decadal droughts in the 1470s, 1510s, 1630s, 1760’s, and 1920’s, and wet episodes during the 1480’s, 1600’s, 1740’s, and 1930’s-40’s. Steinbeck’s impression of Salinas River flow history was probably colored by the extraordinary decadal changes around average hydrologic conditions of the early 20th century. “Steinbeck’s Cycle” in the early 1900’s was more dramatic than any others witnessed in the last 600 years. Another multidecadal “cycle” is apparent in the mid-18th century when the most persistent wet period was followed by one of the most persistent dry periods in the last 600 years.

48

Figure 18. Spectral analysis (Jenkins and Watts 1968) of the 595-year discharge reconstruction indicates a significant (p < 0.01) peak at 23.6 years which accounts for 5.2% of the relative variance in the time series. Other spectral peaks significant at the 95% confidence limit are illustrated, although they do not account for much of the time series’ relative variance.

49

Figure 19. Events of surplus or deficit flow, respectively defined as 80% and 120% of the reconstructed median discharge, lasting 2 or more consecutive years. Mean, median, and maximum year intervals between respective events for the full 595-year period and the 20th century are listed below.

50

a b

Figure 20 . These composite maps of tree-ring reconstructed summer PDSI (Cook et al. 2004) illustrate average moisture conditions over North America during the 10 wettest years (a) and the 10 driest years (b) in the Salinas Valley since 1644, as indicated by the reconstruction (Tables 7 and 8).

51

a

b

Figure 21. Composite maps of observed spring sea surface temperature anomalies (Kaplan et al. 1998) illustrate average oceanic surface temperature divergences during the 10 wettest years (a) and the 10 driest years (b) in the Salinas Valley since 1856, as indicated by the reconstruction.

52

Table 1. Blue oak chronologies from the Central Coast Ranges of California, ordered north to south. Metadata provided for each chronology includes site name and code, University of Arkansas Museum accession number, geographic coordinates and elevation, series Intercorrelation (from COFECHA; Holmes 1983), number of dated blue oak series (typically one to two per tree), and length of the chronology.

Site Name Site Code

Museum Accession Number County

Lat. (n) Long. (w)

Elev. (m)

Series inter-correlation

Number of Dated Series

Chronology length

Mount Diablo DI2 95-123 Contra Costa

37.8792 121.9658 182 0.788 110 1582-2004

Marys Ranch B24 04-200 Santa Clara

37.3954 121.7775 475 0.648 42 1697-2003

Henry Coe Park COE 05-224 Stanislaus 37.1707

121.3968 421 0.770 41 1675-2005

Pacheco Pass PP2 95-159 Merced 37.0368

121.2091 434 0.792 140 1510-2003

Rock Springs Ranch B27 04-201 San Benito

36.4896 120.8776 1067 0.861 89 1379-2003

Pinnacles PN2 95-145 San Benito

36.4794 121.1792 487 0.797 100 1577-2003

Hastings Reserve HAS 04-204 Monterey 36.3947

121.5471 755 0.707 87 1460-2004

Wright Mountain B32 04-203 Fresno 36.3361

120.5153 1219 0.814 61 1409-2003

Indians IND 04-202 Monterey 36.1168

121.4618 646 0.662 50 1494-2003

Palo Prieto Canyon PPC 04-230 San Luis Obispo

35.6854 120.2557 558 0.847 47 1538-2004

American Canyon AC2 95-160 San Luis Obispo

35.2769 120.2689 561 0.769 147 1455-2004

Los Lobos LOB 04-229 Kern 34.9222

119.2432 1051 0.850 84 1333-2004

Figueroa Mountain FIG 04-205 Santa Barbara

34.7352 119.9970 1036 0.827 86 1293-2003

53

Table 2. Pearson correlation coefficients calculated among residual blue oak tree ring chronologies (three letter site codes and variable common periods are defined in Table 1). Generally high correlations suggest strong interannual moisture sensitivity at a regional scale. Note the correlations among the four longest tree-ring chronologies (B27, B32, LOB, and FIG, indicated with asterisks), which were used to compute the average chronology to maximize length of the Salinas River flow reconstruction.

DI2 B24 COE PP2 *B27 PN2 HAS *B32 IND PPC AC2 *LOB *FIG

DI2 0.50 0.58 0.75 0.73 0.67 0.51 0.67 0.58 0.57 0.64 0.65 0.47

B24 0.50 0.35 0.53 0.49 0.41 0.46 0.48 0.29 0.39 0.43 0.38 0.39

COE 0.58 0.35 0.66 0.59 0.62 0.61 0.61 0.64 0.54 0.60 0.66 0.44

PP2 0.75 0.53 0.66 0.70 0.71 0.49 0.65 0.55 0.59 0.61 0.65 0.51

*B27 0.73 0.49 0.59 0.70 0.80 0.60 0.85 0.54 0.76 0.71 0.71 0.65

PN2 0.67 0.41 0.62 0.71 0.80 0.63 0.73 0.63 0.68 0.78 0.75 0.60

HAS 0.51 0.46 0.61 0.49 0.60 0.63 0.55 0.45 0.43 0.52 0.51 0.49

*B32 0.67 0.48 0.61 0.65 0.85 0.73 0.55 0.50 0.72 0.68 0.69 0.64

IND 0.58 0.29 0.64 0.55 0.54 0.63 0.45 0.50 0.45 0.57 0.65 0.43

PPC 0.57 0.39 0.54 0.59 0.76 0.68 0.43 0.72 0.45 0.69 0.70 0.66

AC2 0.64 0.43 0.60 0.61 0.71 0.78 0.52 0.68 0.57 0.69 0.54 0.72

*LOB 0.65 0.38 0.66 0.65 0.71 0.75 0.51 0.69 0.65 0.70 0.54 0.62

*FIG 0.47 0.39 0.44 0.51 0.65 0.60 0.49 0.64 0.43 0.66 0.72 0.62

Average 0.61 0.43 0.58 0.62 0.68 0.67 0.52 0.65 0.53 0.60 0.63 0.63 0.55

54

Table 3. USGS continuous flow gauge records are ordered upstream to down, with metadata including gauge name, number, latitude, longitude, drainage area, period of record, mean annual discharge and remarks (USGS 2007). The Paso Robles record is the second longest discharge record, and is reported to be upstream from more human interference than any other on the Salinas River (Dettinger 2004).

Gauge Name, Number

Latitude, Longitude

Drainage (sq km)

Period (water year)

Mean Annual Discharge (M m3) Remarks

Salinas River at Paso Robles, 11147500

35°37'43" 120°41'00"

1010 1940-1965, 1970-

present

93.16 Records are fair except for estimated daily discharges, which are poor. Low flows regulated by Santa Margarita Lake, 32 mi upstream, beginning in December 1941, usable capacity, 23,000 acre-ft. Small diversions for irrigation upstream from station.

Salinas River near Bradley, 11150500

35°55'49" 120°52'04"

6566 1949-present

443.58 Records fair including estimated days. Flow regulated by Santa Margarita Lake beginning in December 1941, usable capacity, 23,000 acre-ft; Lake Nacimiento (formerly Nacimiento Reservoir) beginning in February 1957, usable capacity, 340,000 acre-ft; and Lake San Antonio beginning in December 1965, usable capacity, 330,000 acre-ft. Several small diversions upstream from station.

Salinas River at Soledad, 11151700

36°24'40" 121°19'06"

9228 1969-1978, 1984-

present

367.64 Records fair except January 1 to March 2 and estimated daily discharges, which are poor. Flow regulated by Santa Margarita Lake beginning in December 1941, usable capacity, 23,000 acre-ft; Lake Nacimiento (formerly Nacimiento Reservoir) beginning in February 1957, usable capacity, 340,000 acre-ft; and by Lake San Antonio beginning in December 1965, usable capacity, 330,000 acre-ft. Several small diversions for irrigation upstream from station.

Salinas River near Chualar, 11152300

36°33'13" 121°32'54"

10469 1977-present

431.68 Daily discharges prior to January 1979 determined by discharge measurements at this site correlated to streamflow for "Salinas River at Soledad" (station 11151700) and "Salinas River near Spreckels" (station 11152500). Flow regulated by Santa Margarita Lake beginning in December 1941, usable capacity, 23,000 acre-ft; Lake Nacimiento (formerly Nacimiento Reservoir) beginning in February 1957, usable capacity, 340,000 acre-ft; and Lake San Antonio beginning in December 1965, usable capacity, 330,000 acre-ft. Several small diversions upstream from station.

Salinas River near Spreckels, 11152500

36°37'52" 121°40'17"

10764 1930-present

381.84 Flow regulated by Santa Margarita Lake (formerly Salinas Reservoir) beginning in 1941, usable capacity, 23,000 acre-ft; Lake Nacimiento (formerly Nacimiento Reservoir) beginning in February 1957, usable capacity, 340,000 acre-ft; and by Lake San Antonio beginning in December 1965, usable capacity, 330,000 acre-ft. Large withdrawals fromground water and small surface-water diversions for municipal use and for irrigation upstream from station.

55

Table 4. Water year mean flows correlated (for common period, see Table 1) between the five USGS continuous flow records on the Salinas River. Watery year correlations between gauges on the Salinas River and its tributaries (not shown) are all highly significant, and indicate basin-wide homogeneity in inter-annual flow variability.

Bradley Soledad Chualar Spreckles

Paso Robles 0.95 0.96 0.95 0.93

Bradley 0.99 0.97 0.97

Soledad 0.98 0.98

Chualar 0.99

56

Table 5. Calibration statistics for the tree-ring estimation of Salinas River gauged discharge at Paso Robles for the water year (October – December), 1940 – 2003 (1966 – 1969 observed data missing). Coefficients Standard Error t-Statistic(H0: β=0) Regressionb

Equation Period R2Adj

a β0 β1 β0 β1 β0 β1 Residual Autocorr.

1 1940-1965 0.71 -151.38 232.43 31.16 29.31 -4.86*** 7.93*** -0.39$

2 1970-2003 0.69 -159.63 264.13 31.95 30.31 -5.00*** 8.72*** -0.03NS

3 1940-2003 0.70 -155.84 250.10 22.75 21.50 -6.85*** 11.63*** -0.13NS

NS Not significant, i.e., there is greater than a 5% probability that the result occurred by chance. * Significant, p < 0.05. ** Significant, p < 0.01. *** Significant, p < 0.001. aR2 adjusted downward for one lost of degree of freedom (Draper and Smith 1981). bAutocorrelation of the residuals from regression, tested with the Durbin-Watson statistic (Draper and Smith 1981). Failure to reject the null hypothesis indicates that the residuals occur randomly, an indication that the regression model is valid. $ Durbin-Watson statistic falls within the region of uncertainty, close to significant serial correlation in residuals.

Table 6. Verification statistics for the tree-ring reconstruction of Salinas River gauged discharge at Paso Robles for the water year (October – December), 1940 – 2003 (1966 – 1969 observed data missing).

Peroid Pearson Corr. 1st Dif. Corr.aPaired t-Test .

of MeanbSign Test

(Hit/Miss) c

Cross-Products t-

TestdReduction of

Error StatisticeCoefficient of

Efficiencyf

1970-2003 0.86 0.88 1.44NS 29/5*** -3.59** 0.69 0.68

1940-1965 0.86 0.80 -2.45* 22/4*** -2.34* 0.68 0.67

NS Not significant, i.e., there is greater than a 5% probability that the result occurred by chance. * Significant, p < 0.05. ** Significant, p < 0.01. *** Significant, p < 0.001. aObserved and reconstructed data first differenced (t – t-1). The transformation removes trends that may affect the Pearson correlation coefficient (Fritts 2001). bPaired comparison of observed and reconstructed data means (Steel and Torrie 1980). Note that no difference is the desired result. cSigns of departures from the mean of each series (Fritts 2001). Means are subtracted from each series and the residuals are multiplied. A positive product is a “hit”. If either observed or reconstructed data lie near the mean, the year is omitted from the test. dComparison of the relative magnitude of hits/misses in the sign test above. eThere is no formal test of significance for this statistic, but any positive result indicates that the reconstruction contributes unique paleoclimatic information (Fritts 2001) fA more stringent test than the Reduction of Error Statistic (RE). Will always be lower than RE..

57

Table 7. Mean discharge (Mean Q, in M m3) ranked for n-year drought periods ending with given year, so the rank 1 3-year drought dated from 1735-1737. Rows 1-30 are the lowest reconstructed flows and the last five rows are the lowest observed flows. Because 98 years were reconstructed as zero flow, the regional average tree-ring chronology (Figure 11; Appendix 1) was used to rank the 1-year reconstructed droughts. Reconstructed values were used in place of missing data (1966-1969) for the observed discharge rankings.

Rank 1-Year Mean Q 3-Year Mean Q 6-Year Mean Q 10-year Mean Q 20-year Mean Q 30-year Mean Q

1 1571 0.0 1737 0.0 1481 19.6 1931 34.5 1936 59.3 1783 71.3 2 1580 0.0 1845 0.0 1634 23.1 1930 39.0 1934 59.8 1780 73.5 3 1585 0.0 1900 0.0 1521 24.9 1934 41.7 1935 60.0 1777 73.6 4 1782 0.0 1824 1.5 1520 27.2 1933 41.7 1770 62.7 1655 73.9 5 1829 0.0 1931 3.8 1480 29.8 1929 43.3 1931 64.4 1782 74.1 6 1898 0.0 1796 3.9 1931 30.0 1522 43.8 1767 64.8 1781 74.9 7 1654 0.0 1858 4.6 1930 30.9 1932 45.2 1937 66.0 1658 75.7 8 1670 0.0 1529 6.3 1929 30.9 1523 46.6 1648 67.3 1778 76.1 9 1795 0.0 1930 8.7 1951 32.5 1761 48.2 1765 68.5 1659 76.4

10 1532 0.0 1667 9.4 1859 33.2 1521 48.7 1766 68.8 1779 77.2 11 1864 0.0 1708 10.5 1670 34.8 1481 49.0 1769 69.1 1647 77.5 12 1877 0.0 1655 11.8 1928 34.9 1928 49.1 1768 69.9 1648 77.7 13 1444 0.0 1846 13.1 1757 35.5 1760 49.9 1639 70.0 1654 77.7 14 1843 0.0 1925 13.6 1846 36.5 1638 50.6 1523 70.9 1784 77.7 15 1452 0.0 1634 13.8 1482 37.1 1926 51.6 1930 70.9 1523 78.0 16 1669 0.0 1633 13.8 1519 38.1 1639 51.7 1939 71.0 1650 78.1 17 1691 0.0 1961 14.1 1522 39.4 1757 52.1 1771 71.3 1521 78.3 18 1529 0.0 1859 16.2 1934 40.5 1763 52.6 1647 71.7 1776 78.5 19 1613 0.0 1926 18.2 1545 41.9 1762 53.0 1532 71.8 1657 78.7 20 1637 0.0 1481 18.9 1925 42.2 1927 54.0 1938 72.1 1653 79.7 21 1765 0.0 1948 20.0 1756 42.5 1482 54.6 1640 72.1 1533 79.8 22 1521 0.0 1478 20.3 1990 42.6 1759 54.8 1533 72.2 1481 80.2 23 1841 0.0 1480 20.5 1860 42.7 1484 55.3 1522 72.3 1977 80.4 24 1497 0.0 1872 21.0 1933 43.0 1480 55.5 1637 72.6 1976 80.4 25 1721 0.0 1972 22.7 1926 43.2 1864 55.9 1529 73.6 1660 80.5 26 1590 0.0 1521 23.7 1532 43.6 1637 56.0 1933 73.9 1785 80.6 27 1809 0.0 1520 23.7 1523 44.2 1765 57.0 1932 74.0 1646 81.0 28 2002 0.0 1857 24.0 1875 44.6 1634 58.0 1521 74.3 1972 81.0 29 1597 0.0 1543 24.0 1759 44.8 1483 58.2 1940 74.3 1786 81.0 30 1924 0.0 1479 24.1 1633 45.6 1519 58.7 1964 74.4 1522 81.2

1 1961 0.0 1990 4.9 1992 21.0 1955 44.3 1961 55.9 1977 63.4 2 1968 0.0 1989 6.5 1951 23.0 1968 47.1 1960 58.2 1973 63.8 3 1976 0.0 1961 6.6 1950 28.0 1954 48.6 1959 61.4 1976 63.9 4 1977 0.0 1949 13.9 1990 31.3 1957 50.3 1968 62.0 1975 65.0 5 1990 0.3 1991 14.0 1964 32.6 1956 50.7 1973 62.3 1974 65.5

58

Table 8. Mean discharge (Mean Q, in M m3) ranked for n-year surplus flow periods ending with the given year, so the rank 1 3-year surplus flow dated from 1867-1869. Rows 1-30 are the highest reconstructed flows and the last five rows are the lowest observed flows. Reconstructed values were used in place of missing data (1966-1969) for the observed discharge rankings.

Rank 1-Year Mean Q 3-Year Mean Q 6-Year Mean Q 10-year Mean Q 20-year Mean Q 30-year Mean Q

1 1825 334.9 1869 237.7 1745 189.9 1747 170.7 1749 132.2 1443 118.2 2 1868 330.9 1442 218.3 1441 179.7 1944 162.9 1568 130.9 1449 117.2 3 1884 328.1 1868 217.9 1442 178.8 1749 161.3 1617 129.0 1747 116.9 4 1941 327.9 1441 216.8 1942 176.2 1442 157.6 1620 127.2 1610 116.7 5 1524 319.6 1526 210.0 1443 175.0 1945 155.1 1750 126.1 1746 116.6 6 1978 304.3 1943 207.9 1747 174.1 1746 153.7 1618 125.7 1628 116.6 7 1441 299.6 1980 204.2 1746 173.3 1748 153.2 1748 125.4 1627 115.8 8 1952 296.3 1942 202.6 1983 171.5 1793 152.4 1747 125.3 1445 115.4 9 1656 295.3 1916 199.7 1941 169.4 1441 150.7 1442 123.2 1840 115.3

10 1998 287.3 1536 199.3 1743 165.8 1445 150.3 1833 122.7 1447 115.2 11 1420 283.3 1907 196.7 1910 165.8 1610 149.3 1616 122.7 1625 115.1 12 1914 281.8 1732 194.2 1792 163.7 1943 148.9 1443 122.5 1752 114.9 13 1433 266.6 1747 192.0 1539 163.4 1750 147.6 1834 122.5 1442 114.7 14 1723 266.6 1443 191.9 1606 162.0 1946 146.7 1619 122.3 1611 114.5 15 1740 266.6 1742 190.7 1945 162.0 1834 145.6 1751 122.0 1444 114.5 16 1467 260.6 1886 189.6 1943 161.0 1915 144.9 1569 122.0 1751 114.4 17 1958 258.8 1525 189.6 1744 159.1 1914 144.8 1746 121.9 1612 114.3 18 1828 250.8 1908 189.1 1568 159.1 1941 142.5 1570 121.5 1839 114.0 19 1730 250.3 1745 189.1 1911 159.0 1493 142.1 1451 121.3 1842 114.0 20 1993 250.3 1827 186.0 1793 156.5 1443 141.5 1441 121.2 1854 114.0 21 1589 246.6 1791 185.5 1794 156.5 1444 141.4 1567 121.1 1749 113.9 22 1462 244.6 1792 183.0 1998 155.7 1916 140.5 1835 121.1 1441 113.8 23 1509 244.1 1746 180.8 1538 155.2 1611 140.5 1615 120.4 1448 113.7 24 1789 243.1 1662 179.9 1491 154.7 1491 140.4 1832 119.7 1700 113.5 25 1890 242.3 1606 179.6 1489 154.5 1494 139.7 1446 119.2 1578 113.4 26 1832 241.6 1909 178.0 1444 154.4 1446 139.1 1621 117.4 1446 113.3 27 1867 241.1 1938 177.0 1944 153.7 1792 137.8 1565 117.4 1750 113.0 28 1838 240.8 1937 174.1 1830 153.5 1893 137.7 1503 117.1 1748 113.0 29 1680 239.6 1818 173.0 1471 153.4 1607 137.6 1753 117.0 1620 112.7 30 1745 238.8 1524 171.3 1821 152.7 1745 137.0 1445 117.0 1843 112.4

1 1983 469.5 1943 185.6 1998 212.8 2001 153.8 1998 126.2 1998 114.1 2 1941 410.0 1997 178.9 1983 198.0 2000 148.6 1999 125.4 2002 111.1 3 1995 331.9 1998 282.8 2000 186.0 2002 147.2 1997 123.7 2001 111.0 4 1998 323.4 1983 292.3 1999 175.1 1999 141.4 2001 119.3 2000 108.9 5 1958 302.5 1984 254.4 1997 170.6 1987 140.0 2000 117.0 1999 107.6

59

Appendix 1. This table provides annual ring-width index values (mean = 0.99, standard deviation = +/- 0.36) and sample size for the regional average blue oak chronology. TRI = regional average tree-ring index; 10yr = 10 year smoothing spline of TRI; n = number of tree-ring samples included in TRI.

Year TRI 10yr n Year TRI 10yr n Year TRI 10yr n

1450 0.708 0.88 29 1500 0.874 0.93 41 1451 1.171 0.85 29 1501 1.288 0.99 41 1452 0.308 0.83 29 1502 1.134 1.01 42 1453 0.905 0.85 29 1503 1.165 0.98 41 1454 0.759 0.88 29 1504 0.875 0.94 41 1455 1.447 0.91 29 1505 0.529 0.91 41 1456 0.700 0.91 29 1506 0.867 0.93 41 1457 0.765 0.90 29 1507 0.872 0.99 41 1458 1.291 0.90 29 1508 1.276 1.06 41

1409 1.018 0.98 17 1459 0.475 0.89 29 1509 1.599 1.10 41 1410 0.821 0.98 17 1460 0.766 0.92 31 1510 0.506 1.09 42 1411 1.134 0.99 18 1461 0.974 0.97 32 1511 1.194 1.07 43 1412 1.192 0.98 18 1462 1.601 1.00 33 1512 1.168 1.04 45 1413 0.562 0.97 18 1463 0.998 0.99 33 1513 0.939 0.98 45 1414 1.074 0.97 19 1464 0.541 0.97 33 1514 1.058 0.90 45 1415 1.109 0.98 19 1465 0.626 1.01 33 1515 0.678 0.81 46 1416 1.121 0.98 19 1466 1.397 1.07 33 1516 0.605 0.74 44 1417 0.607 0.98 19 1467 1.665 1.11 35 1517 0.937 0.69 44 1418 1.156 1.02 19 1468 0.442 1.12 35 1518 0.468 0.67 44 1419 0.715 1.07 19 1469 1.247 1.13 36 1519 0.735 0.68 44 1420 1.756 1.12 20 1470 1.436 1.12 36 1520 0.795 0.74 45 1421 0.815 1.13 20 1471 1.051 1.08 38 1521 0.389 0.83 45 1422 1.148 1.13 20 1472 0.819 1.04 39 1522 0.972 0.98 45 1423 1.097 1.12 20 1473 0.816 1.00 39 1523 1.051 1.12 45 1424 1.406 1.09 20 1474 1.312 0.96 40 1524 1.901 1.21 47 1425 0.879 1.03 20 1475 1.036 0.89 40 1525 1.191 1.18 51 1426 0.701 0.99 20 1476 0.612 0.81 40 1526 1.296 1.08 50 1427 1.051 0.97 20 1477 0.679 0.75 40 1527 0.651 0.94 50 1428 1.115 0.95 20 1478 0.811 0.72 40 1528 0.672 0.83 53 1429 1.134 0.93 20 1479 0.668 0.72 40 1529 0.325 0.77 53 1430 0.586 0.92 20 1480 0.636 0.76 40 1530 1.364 0.78 53 1431 0.747 0.94 20 1481 0.792 0.83 39 1531 0.853 0.81 53 1432 1.026 0.99 20 1482 1.042 0.92 39 1532 0.262 0.88 53 1433 1.689 1.03 20 1483 0.958 1.01 39 1533 0.966 1.01 53 1434 0.628 1.04 21 1484 1.197 1.09 39 1534 1.503 1.14 53 1435 0.862 1.07 22 1485 1.199 1.15 39 1535 1.369 1.23 53 1436 1.325 1.13 22 1486 1.155 1.19 39 1536 1.388 1.25 53 1437 1.137 1.19 22 1487 1.025 1.23 39 1537 1.021 1.21 53 1438 1.118 1.25 22 1488 1.409 1.25 39 1538 1.214 1.13 54 1439 1.286 1.31 22 1489 1.459 1.24 39 1539 1.163 1.03 54 1440 1.363 1.35 24 1490 0.927 1.21 40 1540 0.932 0.91 54 1441 1.821 1.32 24 1491 1.474 1.16 40 1541 0.526 0.82 53 1442 1.304 1.23 24 1492 0.542 1.11 40 1542 0.604 0.78 53 1443 1.046 1.09 24 1493 1.445 1.08 40 1543 0.911 0.79 53 1444 0.304 0.98 24 1494 1.099 1.02 41 1544 0.829 0.83 54 1445 1.217 0.93 24 1495 1.055 0.95 41 1545 0.825 0.88 54 1446 0.876 0.91 27 1496 0.832 0.88 41 1546 1.309 0.92 56 1447 0.848 0.91 27 1497 0.390 0.83 41 1547 0.723 0.96 56 1448 0.979 0.91 29 1498 1.221 0.84 41 1548 0.646 1.01 56 1449 1.133 0.90 29 1499 0.529 0.87 41 1549 1.465 1.08 56

60

Year TRI 10yr n Year TRI 10yr n Year TRI 10yr n

1550 1.040 1.12 59 1600 0.859 1.12 82 1650 0.820 1.01 92 1551 1.327 1.13 61 1601 1.406 1.18 82 1651 1.352 0.97 90 1552 0.988 1.11 61 1602 1.278 1.21 82 1652 0.845 0.89 92 1553 1.288 1.08 61 1603 0.917 1.23 82 1653 0.765 0.82 92 1554 0.805 1.06 61 1604 1.410 1.24 81 1654 0.247 0.80 92 1555 0.833 1.06 62 1605 1.419 1.23 82 1655 0.615 0.84 92 1556 1.086 1.09 62 1606 1.195 1.19 82 1656 1.804 0.92 92 1557 1.498 1.12 63 1607 0.887 1.16 82 1657 0.784 0.97 93 1558 1.103 1.12 63 1608 0.958 1.14 82 1658 0.805 1.01 94 1559 0.975 1.10 64 1609 1.356 1.13 82 1659 0.710 1.07 95 1560 1.412 1.07 65 1610 1.376 1.10 83 1660 1.497 1.13 95 1561 0.720 1.04 64 1611 1.051 1.03 83 1661 1.314 1.16 93 1562 0.831 1.05 64 1612 0.992 0.96 82 1662 1.216 1.13 93 1563 1.112 1.10 65 1613 0.326 0.93 82 1663 0.753 1.06 93 1564 1.123 1.17 65 1614 0.975 0.97 82 1664 1.373 0.98 93 1565 1.562 1.22 66 1615 1.107 1.03 82 1665 0.728 0.88 93 1566 1.096 1.22 69 1616 1.482 1.07 82 1666 0.631 0.80 93 1567 1.234 1.17 69 1617 1.131 1.07 83 1667 0.591 0.76 93 1568 1.429 1.08 69 1618 0.795 1.04 84 1668 1.344 0.75 93 1569 0.752 0.96 70 1619 1.028 1.00 86 1669 0.324 0.75 93 1570 1.005 0.84 71 1620 1.248 0.96 86 1670 0.252 0.81 95 1571 0.168 0.77 71 1621 0.491 0.93 82 1671 1.447 0.90 96 1572 0.640 0.78 71 1622 0.701 0.94 82 1672 1.328 0.96 96 1573 1.140 0.84 71 1623 1.452 0.97 82 1673 0.592 0.99 97 1574 1.109 0.89 71 1624 0.859 0.99 80 1674 1.184 1.00 96 1575 0.781 0.92 73 1625 1.082 1.00 81 1675 1.125 1.01 96 1576 0.736 0.94 73 1626 0.631 1.00 81 1676 0.460 1.02 96 1577 1.426 0.94 73 1627 1.379 0.98 81 1677 1.470 1.06 96 1578 1.149 0.90 75 1628 1.164 0.93 81 1678 1.082 1.09 95 1579 0.454 0.85 76 1629 0.502 0.85 85 1679 0.847 1.12 97 1580 0.191 0.85 77 1630 1.011 0.78 86 1680 1.581 1.15 99 1581 1.314 0.92 78 1631 0.499 0.72 86 1681 0.759 1.16 100 1582 1.019 0.98 80 1632 0.542 0.71 85 1682 1.318 1.16 100 1583 1.501 1.01 80 1633 0.789 0.74 86 1683 1.326 1.14 99 1584 0.988 0.99 80 1634 0.604 0.80 86 1684 0.972 1.10 99 1585 0.196 0.98 80 1635 1.448 0.84 87 1685 1.194 1.04 99 1586 1.333 1.00 80 1636 0.944 0.85 87 1686 0.632 0.98 99 1587 1.246 1.02 80 1637 0.331 0.85 86 1687 1.180 0.93 99 1588 0.650 1.02 80 1638 0.947 0.90 86 1688 0.967 0.89 99 1589 1.610 1.01 80 1639 0.666 0.97 86 1689 0.839 0.85 103 1590 0.397 0.97 81 1640 1.416 1.04 87 1690 0.794 0.84 105 1591 1.393 0.92 80 1641 1.169 1.08 87 1691 0.324 0.88 105 1592 0.544 0.87 80 1642 1.333 1.06 87 1692 1.291 0.97 105 1593 0.983 0.84 81 1643 0.592 1.02 88 1693 1.220 1.06 106 1594 0.817 0.82 82 1644 1.151 0.97 88 1694 1.016 1.13 108 1595 0.538 0.82 82 1645 1.008 0.93 88 1695 1.163 1.17 108 1596 1.303 0.85 82 1646 0.692 0.90 88 1696 1.507 1.18 108 1597 0.405 0.90 82 1647 0.719 0.92 89 1697 1.167 1.15 108 1598 1.064 0.97 82 1648 0.813 0.96 89 1698 0.711 1.11 108 1599 1.296 1.05 82 1649 1.509 1.01 89 1699 1.286 1.08 108

61

Year TRI 10yr n Year TRI 10yr n Year TRI 10yr n

1700 1.033 1.05 108 1750 1.139 0.95 102 1800 0.733 1.07 105 1701 0.639 1.04 104 1751 0.790 0.87 101 1801 1.384 1.05 101 1702 1.571 1.04 103 1752 0.730 0.81 101 1802 0.948 1.00 101 1703 0.681 1.01 103 1753 0.949 0.76 101 1803 0.467 0.96 101 1704 1.387 0.97 103 1754 0.429 0.73 101 1804 1.391 0.93 101 1705 0.983 0.90 102 1755 0.807 0.73 101 1805 0.810 0.89 101 1706 0.504 0.83 101 1756 0.859 0.76 101 1806 0.859 0.85 103 1707 0.599 0.81 101 1757 0.540 0.79 101 1807 0.744 0.82 103 1708 0.749 0.85 100 1758 1.116 0.83 102 1808 0.954 0.80 104 1709 1.262 0.91 101 1759 0.786 0.85 102 1809 0.397 0.81 106 1710 0.677 0.97 103 1760 0.944 0.86 104 1810 0.922 0.86 106 1711 1.399 1.00 101 1761 0.723 0.86 100 1811 1.375 0.90 105 1712 1.099 0.99 100 1762 0.921 0.87 100 1812 0.740 0.94 106 1713 0.787 0.96 100 1763 0.935 0.87 100 1813 0.815 0.98 107 1714 0.815 0.94 100 1764 0.981 0.87 100 1814 1.009 1.04 106 1715 0.937 0.94 100 1765 0.352 0.89 100 1815 0.916 1.12 106 1716 0.726 0.97 100 1766 1.315 0.94 100 1816 1.452 1.19 106 1717 1.266 1.00 100 1767 0.983 0.97 101 1817 1.491 1.21 106 1718 1.099 1.02 100 1768 1.041 0.99 102 1818 1.001 1.18 107 1719 1.055 1.01 100 1769 1.103 1.01 102 1819 1.270 1.12 107 1720 1.247 0.98 100 1770 0.467 1.02 102 1820 0.798 1.05 110 1721 0.392 0.95 98 1771 1.482 1.05 99 1821 1.389 0.98 106 1722 0.664 0.97 98 1772 1.126 1.06 99 1822 0.606 0.94 107 1723 1.689 1.00 98 1773 0.839 1.04 99 1823 0.618 0.95 105 1724 0.806 1.00 98 1774 1.229 1.01 100 1824 0.641 1.03 103 1725 1.159 0.97 99 1775 0.955 0.96 101 1825 1.962 1.12 101 1726 0.859 0.93 99 1776 0.771 0.92 101 1826 1.280 1.15 103 1727 1.031 0.91 100 1777 0.713 0.90 101 1827 0.858 1.13 103 1728 0.440 0.91 100 1778 0.929 0.88 101 1828 1.626 1.08 104 1729 0.620 0.97 100 1779 1.295 0.86 103 1829 0.221 1.03 105 1730 1.624 1.06 100 1780 0.697 0.81 103 1830 1.072 1.03 105 1731 1.118 1.11 100 1781 0.963 0.77 101 1831 0.996 1.06 103 1732 1.457 1.10 100 1782 0.196 0.75 101 1832 1.589 1.08 102 1733 0.594 1.03 100 1783 0.526 0.81 101 1833 1.056 1.06 103 1734 1.490 0.96 99 1784 1.388 0.92 101 1834 0.989 1.03 103 1735 0.615 0.88 99 1785 1.158 1.01 101 1835 0.804 1.01 105 1736 0.578 0.84 99 1786 0.904 1.09 101 1836 0.723 1.02 109 1737 0.549 0.88 99 1787 1.383 1.15 101 1837 1.013 1.06 108 1738 1.329 0.99 99 1788 0.611 1.21 101 1838 1.586 1.08 109 1739 0.840 1.11 99 1789 1.595 1.26 101 1839 1.130 1.05 109 1740 1.689 1.22 99 1790 1.488 1.28 106 1840 1.069 0.96 109 1741 1.283 1.29 98 1791 1.012 1.23 106 1841 0.389 0.85 108 1742 1.185 1.33 98 1792 1.565 1.15 108 1842 1.342 0.76 108 1743 1.390 1.35 98 1793 1.210 1.01 106 1843 0.305 0.69 108 1744 1.169 1.35 98 1794 0.576 0.88 106 1844 0.522 0.69 108 1745 1.578 1.32 98 1795 0.259 0.82 106 1845 0.590 0.75 107 1746 1.291 1.26 99 1796 0.670 0.86 106 1846 0.780 0.88 108 1747 1.303 1.18 101 1797 1.291 0.95 104 1847 1.370 1.01 109 1748 0.627 1.09 101 1798 1.247 1.04 103 1848 1.179 1.11 110 1749 1.167 1.02 101 1799 1.276 1.08 104 1849 1.197 1.16 113

62

63

Year TRI 10yr n Year TRI 10yr n Year TRI 10yr n

1850 1.327 1.17 115 1900 0.605 0.76 146 1950 0.715 0.88 164 1851 0.633 1.15 113 1901 1.337 0.83 144 1951 0.664 0.95 163 1852 1.424 1.12 113 1902 0.708 0.89 144 1952 1.808 1.01 163 1853 1.249 1.06 115 1903 0.975 0.97 144 1953 0.843 1.03 162 1854 0.993 0.95 115 1904 0.708 1.06 145 1954 0.873 1.02 163 1855 0.857 0.84 116 1905 1.308 1.17 145 1955 0.834 1.01 163 1856 0.466 0.74 116 1906 1.380 1.25 149 1956 1.094 1.02 163 1857 0.678 0.71 116 1907 1.541 1.29 149 1957 0.854 1.02 163 1858 0.502 0.74 115 1908 1.217 1.26 150 1958 1.658 1.00 163 1859 0.762 0.81 115 1909 1.247 1.20 154 1959 0.792 0.93 163 1860 1.219 0.89 117 1910 1.023 1.13 158 1960 0.605 0.87 163 1861 1.005 0.93 115 1911 1.146 1.08 156 1961 0.433 0.85 162 1862 1.403 0.93 117 1912 0.779 1.06 156 1962 1.360 0.87 160 1863 0.672 0.91 118 1913 0.631 1.10 157 1963 0.894 0.90 160 1864 0.268 0.92 118 1914 1.750 1.15 155 1964 0.606 0.93 160 1865 1.036 1.01 118 1915 1.312 1.17 159 1965 1.186 0.96 160 1866 0.950 1.14 117 1916 1.203 1.13 162 1966 0.993 0.99 160 1867 1.587 1.24 117 1917 0.948 1.06 163 1967 1.206 0.99 160 1868 1.946 1.26 119 1918 0.877 0.98 163 1968 0.481 0.97 160 1869 1.187 1.16 119 1919 0.899 0.92 162 1969 1.532 0.96 160 1870 0.677 1.02 129 1920 0.796 0.87 163 1970 0.707 0.92 160 1871 0.618 0.91 128 1921 0.805 0.84 161 1971 0.812 0.90 160 1872 0.821 0.86 129 1922 1.117 0.81 161 1972 0.507 0.91 160 1873 0.848 0.85 128 1923 0.765 0.76 161 1973 1.497 0.93 160 1874 0.971 0.87 129 1924 0.427 0.73 161 1974 1.020 0.93 160 1875 0.868 0.89 133 1925 0.644 0.72 161 1975 0.922 0.92 160 1876 1.257 0.90 133 1926 0.820 0.74 160 1976 0.496 0.94 160 1877 0.268 0.92 135 1927 1.042 0.74 161 1977 0.582 1.01 160 1878 1.558 0.94 136 1928 0.682 0.73 161 1978 1.840 1.12 160 1879 0.515 0.96 138 1929 0.669 0.72 160 1979 1.127 1.18 159 1880 1.316 0.97 144 1930 0.569 0.73 162 1980 1.352 1.19 159 1881 0.987 0.99 142 1931 0.515 0.78 161 1981 0.822 1.18 156 1882 0.732 1.01 143 1932 1.546 0.84 161 1982 1.180 1.15 156 1883 0.671 1.07 144 1933 0.627 0.89 162 1983 1.531 1.10 154 1884 1.935 1.12 143 1934 0.488 0.96 162 1984 0.747 1.02 154 1885 0.906 1.14 143 1935 1.328 1.06 162 1985 0.834 0.93 153 1886 1.304 1.13 144 1936 1.142 1.15 163 1986 1.130 0.85 152 1887 0.824 1.12 144 1937 1.488 1.22 163 1987 0.528 0.78 148 1888 0.983 1.13 144 1938 1.363 1.26 164 1988 0.926 0.74 147 1889 1.122 1.15 146 1939 0.817 1.28 165 1989 0.588 0.75 147 1890 1.592 1.15 150 1940 1.060 1.31 165 1990 0.478 0.80 147 1891 1.130 1.11 143 1941 1.934 1.33 164 1991 0.959 0.92 143 1892 0.724 1.05 144 1942 1.305 1.30 164 1992 1.151 1.04 142 1893 1.216 1.00 143 1943 1.124 1.21 164 1993 1.624 1.13 142 1894 0.566 0.95 145 1944 1.186 1.10 164 1994 0.697 1.17 140 1895 1.238 0.90 145 1945 1.015 0.97 164 1995 1.564 1.20 140 1896 0.818 0.85 145 1946 0.803 0.86 164 1996 0.913 1.20 139 1897 1.166 0.78 144 1947 0.525 0.79 164 1997 0.904 1.20 137 1898 0.243 0.72 144 1948 0.683 0.78 164 1998 1.772 1.18 136 1899 0.614 0.71 144 1949 1.031 0.81 164 1999 1.020 1.10 132

2000 0.974 1.00 129 2001 0.918 0.89 120 2002 0.401 0.79 118 2003 0.958 0.72 109

Appendix 2. This table lists reconstructed and observed Salinas River water year discharge (Q) at Paso Robles in millions of cubic meters (M m3). Rec Q = reconstructed discharge; 10yr Q = 10-year smoothing spline of reconstructed discharge; Obs Q = observed discharge.

Year Rec Q 10yr Q Obs Q Year Rec Q 10yr Q Obs Q Year Rec Q 10yr Q Obs Q

1450 21.23 72.87 1500 62.75 86.23 1451 137.03 70.13 1501 166.29 97.16 1452 0.00 67.76 1502 127.77 99.76 1453 70.50 69.75 1503 135.53 93.09 1454 33.99 74.81 1504 63.00 82.03 1455 206.05 78.92 1505 0.00 76.08 1456 19.23 77.12 1506 61.00 81.89 1457 35.49 75.10 1507 62.25 97.19 1458 167.04 74.82 1508 163.29 114.63 1409 98.76 89.41 1459 0.00 74.98 1509 244.07 123.95 1410 49.49 90.79 1460 35.74 80.90 1510 0.00 122.17 1411 127.77 92.55 1461 87.76 91.34 1511 142.78 116.83 1412 142.28 91.70 1462 244.57 98.32 1512 136.28 106.62 1413 0.00 89.15 1463 93.76 96.10 1513 79.00 90.47 1414 112.77 89.63 1464 0.00 93.99 1514 108.77 70.76 1415 121.52 90.54 1465 0.72 102.39 1515 13.73 50.47 1416 124.52 90.14 1466 193.55 119.24 1516 0.00 35.22 1417 0.00 91.44 1467 260.58 131.43 1517 78.25 26.91 1418 133.28 99.40 1468 0.00 134.25 1518 0.00 24.16 1419 22.98 111.29 1469 156.03 134.84 1519 27.98 29.37 1420 283.34 123.20 1470 203.30 130.32 1520 42.99 43.75 1421 47.99 127.10 1471 107.02 118.38 1521 0.00 67.68 1422 131.27 127.22 1472 48.99 104.59 1522 87.26 99.71 1423 118.52 123.85 1473 48.24 93.74 1523 107.02 130.99 1424 195.80 115.51 1474 172.29 83.41 1524 319.60 148.99 1425 64.00 102.32 1475 103.26 67.30 1525 142.03 142.83 1426 19.48 91.63 1476 0.00 48.07 1526 168.29 118.66 1427 107.27 87.02 1477 13.98 33.19 1527 6.98 88.30 1428 123.02 84.10 1478 46.99 25.55 1528 11.98 66.56 1429 127.77 79.70 1479 11.23 25.35 1529 0.00 59.69 1430 0.00 76.48 1480 3.22 34.18 1530 185.30 64.58 1431 30.99 81.11 1481 42.24 51.88 1531 57.50 73.66 1432 100.76 92.25 1482 104.76 74.48 1532 0.00 89.92 1433 266.58 101.58 1483 83.76 96.98 1533 85.76 115.48 1434 1.22 104.65 1484 143.53 116.93 1534 220.06 141.92 1435 59.75 111.67 1485 144.03 131.96 1535 186.55 157.41 1436 175.54 125.06 1486 133.03 142.46 1536 191.30 158.23 1437 128.52 140.45 1487 100.51 150.49 1537 99.51 146.58 1438 123.77 157.03 1488 196.55 156.16 1538 147.78 127.60 1439 165.79 173.25 1489 209.06 155.17 1539 135.03 102.94 1440 185.05 183.53 1490 76.00 147.52 1540 77.25 75.96 1441 299.59 180.73 1491 212.81 138.00 1541 0.00 53.87 1442 170.29 160.71 1492 0.00 127.31 1542 0.00 43.47 1443 105.76 131.97 1493 205.55 118.73 1543 72.00 44.88 1444 0.00 106.41 1494 119.02 106.55 1544 51.49 52.99 1445 148.53 90.66 1495 108.02 91.06 1545 50.49 64.35 1446 63.25 81.13 1496 52.24 76.56 1546 171.54 75.68 1447 56.25 77.00 1497 0.00 68.78 1547 24.98 84.74 1448 89.01 76.57 1498 149.53 69.59 1548 5.72 97.31 1449 127.52 75.83 1499 0.00 75.17 1549 210.56 113.08

64

Year Rec Q 10yr Q Obs Q Year Rec Q 10yr Q Obs Q Year Rec Q 10yr Q Obs Q 1550 104.26 123.20 1600 59.00 127.74 1650 49.24 98.52 1551 176.04 126.08 1601 195.80 139.86 1651 182.30 92.51 1552 91.26 121.95 1602 163.79 147.20 1652 55.50 79.20 1553 166.54 114.97 1603 73.50 150.94 1653 35.49 67.36 1554 45.74 108.73 1604 196.80 153.42 1654 0.00 64.79 1555 52.49 109.74 1605 199.05 150.48 1655 0.00 73.67 1556 115.77 117.83 1606 143.03 141.73 1656 295.34 86.72 1557 218.81 125.18 1607 66.00 132.97 1657 40.24 92.76 1558 120.02 124.69 1608 83.76 129.65 1658 45.49 98.96 1559 88.01 118.82 1609 183.30 129.11 1659 21.73 111.09 1560 197.30 111.17 1610 188.30 123.58 1660 218.56 125.67 1561 24.23 104.24 1611 107.02 111.41 1661 172.79 130.96 1562 51.99 106.62 1612 92.26 98.98 1662 148.28 123.64 1563 122.27 119.38 1613 0.00 93.70 1663 32.49 107.77 1564 125.02 135.94 1614 88.01 99.74 1664 187.55 89.06 1565 234.82 148.29 1615 120.77 111.06 1665 26.23 68.64 1566 118.27 149.61 1616 214.81 118.23 1666 1.97 54.78 1567 152.78 140.97 1617 127.02 114.99 1667 0.00 52.53 1568 201.55 122.72 1618 42.99 105.70 1668 180.54 59.00 1569 32.24 97.68 1619 101.26 96.84 1669 0.00 67.63 1570 95.51 75.42 1620 156.28 88.82 1670 0.00 81.35 1571 0.00 61.32 1621 0.00 82.58 1671 206.05 97.10 1572 4.22 59.53 1622 19.48 84.07 1672 176.29 104.45 1573 129.27 66.93 1623 207.31 90.71 1673 0.00 104.05 1574 121.52 74.66 1624 59.00 94.12 1674 140.28 104.11 1575 39.49 80.13 1625 114.77 95.40 1675 125.52 104.83 1576 28.23 86.16 1626 1.97 95.58 1676 0.00 108.00 1577 200.80 91.08 1627 189.05 94.31 1677 211.81 115.69 1578 131.52 88.99 1628 135.28 84.49 1678 114.77 121.77 1579 0.00 84.96 1629 0.00 67.76 1679 56.00 127.14 1580 0.00 89.16 1630 97.01 51.70 1680 239.57 132.57 1581 172.54 101.70 1631 0.00 39.36 1681 33.99 134.06 1582 99.01 113.06 1632 0.00 35.72 1682 173.79 134.27 1583 219.56 118.40 1633 41.49 41.91 1683 175.79 129.25 1584 91.26 115.88 1634 0.00 54.32 1684 87.26 117.77 1585 0.00 113.18 1635 206.30 66.95 1685 142.78 103.63 1586 177.54 115.07 1636 80.25 71.77 1686 2.22 89.72 1587 155.78 115.58 1637 0.00 74.04 1687 139.28 79.89 1588 6.73 113.30 1638 81.01 81.60 1688 86.01 71.47 1589 246.57 109.86 1639 10.73 94.78 1689 53.99 66.05 1590 0.00 100.40 1640 198.30 109.78 1690 42.74 67.97 1591 192.55 88.88 1641 136.53 116.32 1691 0.00 80.02 1592 0.00 75.08 1642 177.54 111.66 1692 166.79 99.99 1593 90.01 64.97 1643 0.00 99.19 1693 149.28 118.43 1594 48.49 60.06 1644 131.77 87.05 1694 98.26 131.54 1595 0.00 62.24 1645 96.26 75.98 1695 135.03 139.59 1596 170.04 71.20 1646 17.23 69.28 1696 221.06 140.06 1597 0.00 82.44 1647 23.98 72.19 1697 136.03 131.20 1598 110.27 97.97 1648 47.49 83.80 1698 21.98 119.67 1599 168.29 114.12 1649 221.56 95.90 1699 165.79 112.21

65

Year Rec Q 10yr Q Obs Q Year Rec Q 10yr Q Obs Q Year Rec Q 10yr Q Obs Q 1700 102.51 106.90 1750 129.02 82.26 1800 27.48 115.09 1701 3.97 104.74 1751 41.74 65.96 1801 190.30 110.03 1702 237.07 105.04 1752 26.73 52.75 1802 81.26 100.46 1703 14.48 99.96 1753 81.51 44.17 1803 0.00 90.81 1704 191.05 90.52 1754 0.00 39.35 1804 192.05 83.24 1705 90.01 74.91 1755 45.99 39.63 1805 46.74 72.80 1706 0.00 59.91 1756 59.00 43.39 1806 59.00 62.80 1707 0.00 54.94 1757 0.00 49.05 1807 30.49 56.58 1708 31.49 62.15 1758 123.27 55.55 1808 82.76 55.72 1709 159.79 75.63 1759 40.74 58.88 1809 0.00 59.70 1710 13.23 87.05 1760 80.25 60.34 1810 74.75 68.57 1711 194.05 94.10 1761 24.73 61.59 1811 188.05 77.05 1712 119.02 91.56 1762 74.50 64.95 1812 29.23 81.82 1713 40.99 83.40 1763 78.00 69.66 1813 47.99 89.90 1714 47.99 77.93 1764 89.51 75.23 1814 96.51 104.62 1715 78.50 78.99 1765 0.00 82.63 1815 73.25 123.38 1716 25.73 86.20 1766 173.04 92.40 1816 207.31 140.36 1717 160.79 96.75 1767 90.01 98.96 1817 217.06 146.06 1718 119.02 103.25 1768 104.51 102.72 1818 94.51 138.47 1719 108.02 103.75 1769 120.02 105.30 1819 161.79 123.54 1720 156.03 99.71 1770 0.00 108.64 1820 43.74 105.50 1721 0.00 94.92 1771 214.81 113.47 1821 191.55 89.78 1722 10.23 96.66 1772 125.77 112.19 1822 0.00 79.04 1723 266.58 101.56 1773 53.99 105.31 1823 0.00 82.86 1724 45.74 99.09 1774 151.53 96.02 1824 4.47 103.06 1725 134.03 92.21 1775 83.01 83.95 1825 334.86 128.29 1726 59.00 83.73 1776 36.99 73.18 1826 164.29 140.14 1727 102.01 78.77 1777 22.48 68.15 1827 58.75 139.72 1728 0.00 81.56 1778 76.50 68.36 1828 250.82 133.90 1729 0.00 95.97 1779 168.04 67.82 1829 0.00 124.80 1730 250.32 115.54 1780 18.48 62.80 1830 112.27 121.34 1731 123.77 125.20 1781 85.01 59.05 1831 93.26 122.27 1732 208.56 121.35 1782 0.00 60.98 1832 241.57 121.41 1733 0.00 106.01 1783 0.00 72.97 1833 108.27 112.43 1734 216.81 87.44 1784 191.30 91.82 1834 91.51 100.63 1735 0.00 68.53 1785 133.78 107.78 1835 45.24 93.74 1736 0.00 61.07 1786 70.25 120.20 1836 24.98 97.17 1737 0.00 71.50 1787 190.05 132.41 1837 97.51 109.18 1738 176.54 96.28 1788 0.00 144.72 1838 240.82 118.97 1739 54.24 124.94 1789 243.07 158.35 1839 126.77 115.84 1740 266.58 151.64 1790 216.31 163.21 1840 111.52 101.68 1741 165.04 168.30 1791 97.26 155.60 1841 0.00 82.96 1742 140.53 176.80 1792 235.57 138.03 1842 179.79 65.29 1743 191.80 180.75 1793 146.78 110.87 1843 0.00 48.90 1744 136.53 180.18 1794 0.00 84.01 1844 0.00 42.44 1745 238.82 174.20 1795 0.00 71.41 1845 0.00 51.40 1746 167.04 159.32 1796 11.73 77.40 1846 39.24 74.22 1747 170.04 137.85 1797 167.04 95.00 1847 186.80 102.01 1748 0.97 115.40 1798 156.03 110.40 1848 139.03 123.11 1749 136.03 98.17 1799 163.04 116.78 1849 143.53 134.16

66

Year Rec Q 10yr Q Obs Q Year Rec Q 10yr Q Obs Q Year Rec Q 10yr Q Obs Q

1850 176.04 135.90 1900 0.00 46.94 1950 22.98 64.65 32.83 1851 2.47 131.49 1901 178.54 58.41 1951 10.23 82.40 29.44 1852 200.30 125.17 1902 21.23 70.40 1952 296.34 97.62 207.94 1853 156.53 110.75 1903 88.01 86.64 1953 54.99 100.21 47.31 1854 92.51 87.68 1904 21.23 108.93 1954 62.50 97.69 33.40 1855 58.25 62.24 1905 171.29 135.64 1955 52.74 97.12 16.80 1856 0.00 42.30 1906 189.30 157.00 1956 117.77 99.57 98.05 1857 13.73 34.36 1907 229.56 165.47 1957 57.75 101.05 8.82 1858 0.00 39.84 1908 148.53 159.48 1958 258.83 97.20 302.50 1859 34.99 55.93 1909 156.03 144.65 1959 42.24 83.80 9.19 1860 149.03 74.57 1910 100.01 127.43 1960 0.00 71.24 10.49 1861 95.51 86.44 1911 130.77 114.44 1961 0.00 67.92 0.00 1862 195.05 89.63 1912 38.99 110.16 1962 184.30 72.00 122.36 1863 11.98 87.87 1913 1.97 118.21 1963 67.75 75.66 50.17 1864 0.00 94.06 1914 281.83 132.23 1964 0.00 80.77 3.10 1865 103.26 113.53 1915 172.29 135.80 1965 140.78 89.18 37.29 1866 81.76 139.61 1916 145.03 125.97 1966 92.51 95.51 1867 241.07 160.59 1917 81.26 107.95 1967 145.78 97.53 1868 330.85 160.59 1918 63.50 89.19 1968 0.00 95.31 1869 141.03 134.98 1919 69.00 74.17 1969 227.31 91.36 1870 13.48 99.07 1920 43.24 63.90 1970 20.98 83.07 31.26 1871 0.00 70.99 1921 45.49 57.63 1971 47.24 77.85 26.18 1872 49.49 58.29 1922 123.77 52.01 1972 0.00 79.42 0.98 1873 56.25 58.68 1923 35.49 43.77 1973 218.56 84.64 178.30 1874 87.01 67.08 1924 0.00 36.54 1974 99.26 84.78 117.03 1875 61.25 78.42 1925 5.22 34.03 1975 74.75 83.36 45.21 1876 158.54 89.19 1926 49.24 35.15 1976 0.00 88.62 0.00 1877 0.00 96.43 1927 104.76 35.27 1977 0.00 106.03 0.00 1878 233.82 101.41 1928 14.73 32.26 1978 304.34 128.83 274.95 1879 0.00 100.64 1929 11.48 31.12 1979 126.02 141.52 45.09 1880 173.29 99.26 1930 0.00 36.32 1980 182.30 143.68 241.03 1881 91.01 97.29 1931 0.00 48.89 1981 49.74 138.83 42.20 1882 27.23 99.69 1932 230.81 64.29 1982 139.28 131.76 115.13 1883 11.98 110.78 1933 0.97 76.32 1983 227.06 119.25 469.48 1884 328.10 124.64 1934 0.00 92.82 1984 30.99 99.30 39.31 1885 70.75 128.47 1935 176.29 115.72 1985 52.74 79.53 10.79 1886 170.04 126.56 1936 129.77 136.93 1986 127.02 62.58 157.45 1887 50.24 123.59 1937 216.31 152.05 1987 0.00 48.19 4.92 1888 90.01 125.39 1938 185.05 159.15 1988 75.75 40.85 8.79 1889 124.77 130.33 1939 48.24 163.47 1989 0.00 42.62 5.69 1890 242.32 131.09 1940 109.27 171.61 87.84 1990 0.00 56.83 0.32 1891 126.77 121.72 1941 327.85 177.26 410.00 1991 84.01 81.82 36.02 1892 25.23 107.81 1942 170.54 168.53 105.84 1992 132.03 108.96 70.04 1893 148.28 96.20 1943 125.27 147.60 265.29 1993 250.32 128.99 257.01 1894 0.00 86.63 1944 140.78 120.19 66.81 1994 18.48 138.22 6.69 1895 153.78 79.88 1945 98.26 90.26 59.49 1995 235.32 143.08 331.94 1896 48.74 70.95 1946 44.99 63.55 34.08 1996 72.50 143.09 115.77 1897 135.78 59.72 1947 0.00 46.66 13.53 1997 70.25 142.41 242.10 1898 0.00 47.63 1948 14.98 43.28 9.35 1998 287.34 138.40 323.42 1899 0.00 42.22 1949 101.76 51.02 18.78 1999 99.26 122.94 30.65 2000 87.76 100.89 72.01 2001 73.75 78.18 87.89 2002 0.00 58.66 4.37 2003 83.76 43.84 24.83

67