Trinity R. at Lewiston, CA

https://www.treeflow.info/sites/default/files/Trinity.xls
Background 

The Trinity River is one of the main tributaries of the lower Klamath River. The headwaters of the Trinity River are in northeastern California. California diverted much of the Trinity River’s water into the Sacramento River basin in the 1960s. Trinity River water is an important contribution to California’s water supply via the Central Valley Project. The reconstruction of the Trinity River was generated as part of a project supported by the California Department of Water Resources (CDWR) and the Bureau of Reclamation WaterSmart program. In 2012, Connie Woodhouse (University of Arizona Laboratory of Tree-Ring Research) with graduate student Brewster Malevich recollected tree-ring sites originally sampled by David Meko and others, and made several new collections in central Oregon and northern California. Reconstructions of water year streamflow were developed for the Trinity at Lewiston CA and the Klamath at Keno OR. Water year precipitation reconstructions for Yreka and Weaverville, CA and Klamath Falls, OR were also generated as part of this CDWR project, which included updated and new flow reconstructions for the Sacramento and San Joaquin basins.

References:

Klamath/San Joaquin/Sacramento Hydroclimatic Reconstructions, Final Report to CADWR from Tree Rings


Calibration & Validation 

Methods

The Trinity River at Lewiston CA streamflow reconstruction was developed using stepwise regression, calibrated on the estimated natural flows developed by the California Department of Water Resources, 1912-2003. The pool of predictor candidate chronologies for these reconstructions included a total of 20 chronologies developed for this study or from the International Tree-Ring Databank (ITRDB). Selection criteria for the pool included a start date of 1650 or before, ended in 1996 or later, and significant correlation with the gage record (p < 0.5). Leave-one-out cross-validation was used to validate these reconstructions.

Predictors: 

Statistic Calibration Validation
Explained variance (R2) 0.68  
Reduction of Error (RE)   0.64
Standard Error of the Estimate    
Root Mean Square Error (RMSE)   353 KAF

(For explanations of these statistics, see this document (PDF), and also the Reconstruction Case Study page.)

 

Figure 1.  Scatterplot of observed and reconstructed Trinity River annual flow, 1912-2003.

Figure 2. Observed (black) and reconstructed (blue) annual Trinity River annual flow, 1912-2003. The observed mean is illustrated by the dashed line.


Long-Term Reconstruction 

Figure 3. Reconstructed annual flow for the Trinity River flow (1584-2003) is shown in blue. Observed flow is shown in gray and the long-term reconstructed mean is shown by the dashed line.

 

Figure 4. The 10-year running mean (plotted on final year) of reconstructed Trinity River flow, 1584-2003. Reconstructed values are shown in blue and observed values are shown in gray. The long-term reconstructed mean is shown by the dashed line.