Tree-ring based streamflow reconstruction for Ashley Creek, northeastern Utah: implications for palaeohydrology of the southern Uinta Mountains

Two ring-width chronologies from Douglas fir (Pseudotsuga menzeisii) and one from pinyon pine (Pinus edulis) were used to reconstruct mean annual discharge of Ashley Creek in the Uinta Mountains, Utah, for the years AD 1637 � /1970. A backward-elimination multiple linear-regression model identified six statistically significant variables, including annual lags ranging from � / 1t o� /2 years. The final model explains 71% of variability (R 2 adjusted for degrees of freedom lost) when regressed against the annual discharges recorded for 1915 � /72. Statistical analysis of the model reconstruction using the reduction-of-error statistic, the Durbin � /Watson statistic and the hypergeometric distri- bution all indicate fidelity of the model for reconstructing discharge. The Lilliefors' test indicates that years of extreme discharge (greater than the 90th percentile or less than the 10th percentile) are non-randomly distributed between 1637 and 1970. Persistent years of above-median discharge as well as non-random clustering of years above the 90th percentile occurred from 1692 � /1704 and 1898 � /1945. Analysis of the modern gauge data set shows that mean annual discharge on Ashley Creek explains between 81% and 91% of the variations of mean annual discharge on the other main stream in the southern Uinta Mountains, indicating that reconstructed mean annual discharge on Ashley Creek is representative of palaeohydrologic conditions across the south flank of the range. These data suggest that the southern Uinta Mountains experienced persistent below-median discharge for 1741 � /1897, through the peak of the 'Little Ice Age' as identified in numerous surrounding mountain ranges in the Central Rockies. This interpretation agrees with lichenometric evidence for extremely limited 'Little Ice Age' glacial activity in the Uintas, indicating that the range experienced unusually warm/dry conditions at this time.