Application of Streamflow Reconstruction to Water Resources Management - Dendroclimatology - page 233

chronology development (Stokes and Smiley 1968 ; Fritts 1976 ) . Stockton ( 1975 )
incorporated these techniques in his tree growth/runoff analysis and reconstructions
for sub-basins of the Gila and Colorado Rivers, demonstrating the usefulness of
dendrochronological methods for reconstructing records of past runoff in the south-
western United States. Building on this work, Stockton and Jacoby ( 1976 ) went on
to develop a suite of annual runoff reconstructions for upper Colorado River gauges
with new tree-ring chronology collections and improved estimates of natural flow
for calibration. Stockton and Jacoby's 1976 report directly addressed the implica-
tion of the resulting reconstructions for water management in the Colorado River
basin. Specifically, the report identified the early decades of the twentieth century,
the portion of the gauge record upon which the 1922 Colorado River Compact was
based, as the wettest period in the past 450 years. Stockton and Jacoby ( 1976 ) con-
cluded that the apparent overallocation of water resources, based on this wet period,
could soon lead to water demands that exceeded water supplies.
The work of Stockton and Jacoby ( 1976 ) clearly demonstrated the value of
extended records of streamflow for water resource planning and management,
particularly for evaluating twentieth-century hydrology in a long-term context.
In the early 1980s, due largely to the efforts of Charles Stockton and W.R.
Boggess, inroads were made on communication of the potential value of stream-
flow record augmentation by tree rings to river-basin management (Stockton and
Boggess 1980 a , 1980 b , 1982 ) . A number of studies followed, many focusing on the
assessment of droughts and the potential applications to water resources manage-
ment. These studies included reconstructions of streamflow for the Potomac River,
Maryland (Cook and Jacoby 1983 ) ; the Occoquan River, Virginia (Phipps 1983 ) ;
and the White River, Arkansas (Cleaveland and Stahle 1989 ; Cleaveland 2000 ) .
In the western United States, streamflow reconstructions of the Sacramento River,
California, were made for the California Department of Water Resources (Earle and
Fritts 1986 ) , and reconstructions of the Salt and Verde Rivers, Arizona, were made
for the US Army Corps of Engineers (Smith and Stockton 1981 ) .
More recently, reconstructions have been generated for an array of rivers in
western North America, ranging from the Canadian prairie region (Saskatchewan
River; Case and MacDonald 2003 ) and the Pacific Northwest of the United
States (Columbia River; Gedalof et al. 2004 ) to the northern and central Rockies
(Yellowstone River in Montana, Graumlich et al. 2003 ; Boulder Creek in Colorado,
Woodhouse 2001 ; Jainetal. 2002 ) , the southwestern United States (Gila River in
Arizona; Meko and Graybill 1995 ) , and Gulf of California continental watersheds in
Mexico (Brito-Castillo et al. 2003 ) . Several efforts have recalculated upper Colorado
River flows (Michaelsen et al. 1990 ; Hidalgo et al. 2000 ) using the same or similar
data used by Stockton and Jacoby ( 1976 ) but different calibration approaches. Most
recently, a number of Stockton and Jacoby's ( 1976 ) Colorado River basin recon-
structions for gauges on the Green, Colorado, and San Juan Rivers, including Lees
Ferry, have been updated by using a new set of tree-ring chronologies and a longer
calibration period (Woodhouse et al. 2006 ) .
Several studies have more specifically addressed management and
decision-making issues with streamflow reconstructions. The first of these was