3.2.4. Model 3: Contemporary Outflow and Salinity Reconstruction

Following the methods reported above for pre-development conditions, the annual 8RI time series tree-ring reconstructions from Model 1 were used as the basis for reconstructing annual Delta outflow volume and salinity in the estuary under contemporary conditions. The modeling logic employed for the reconstructions is presented as a flow chart in Figure 2. This logic was applied to both the long- and short-period reconstructions spanning WYs 1912–2008 and WYs 1912–2001, respectively.

Contemporary Delta outflow was estimated for both reconstruction periods assuming the power law relationship provided in (Equation (2)). This relationship, which was calibrated with historical annual runoff and Delta outflow data, does not represent the full contemporary period; rather, it is limited to a relatively stationary period prior to significant increases in water use in the Central Valley and Delta following construction of Shasta Dam in WY 1944. A residual analysis was conducted to address the observed time series trend. Tree-ring reconstructions of Central Valley runoff (from Model 1) were used in conjunction with Equation (2) to estimate contemporary annual Delta outflow volumes for the WYs 1912–1944 period; a modified form of Equation (2), presented later in this text, was used to estimate Delta outflow volumes for the post 1944 period.

Contemporary X2 position was estimated for the long- and short-period reconstruction periods assuming power law relationships between annual Delta outflow volume and seasonal (February–June and July–October) average X2 position. Equation (3) was calibrated for the contemporary period using a data subset spanning WYs 1920–2018. The initial years of the contemporary period (WYs 1912–1919) were excluded from model calibration due to lack of historical Delta outflow data at a daily resolution. Following the methodology used for pre-development model calibration, daily outflow was transformed into a daily antecedent outflow time series and this daily antecedent outflow time series was then transformed into a daily X2 time series using a contemporary flow–salinity relationship reported by Andrews et al. [26]. Finally, this daily X2 time series was averaged to develop seasonal average X2 time series for purposes of model calibration. The time series of annual contemporary outflow volume, estimated from the tree-ring reconstructions, were then used in conjunction with Equation (3) to estimate seasonal average X2 positions for each time series beginning in WY 1912.
