Paper No. 3
Presentation Time: 2:00 PM
A 1500 YEAR RECONSTRUCTION OF ARIDITY PATTERNS IN THE INTERIOR PACIFIC NORTHWEST FROM LAKE SEDIMENT OXYGEN ISOTOPES AND PREDICTIVE MODELS
The Palmer Drought Severity Index (PDSI) reconstructions of Cook et al. (2004) are perhaps the most commonly invoked records of drought/pluvial history in western North America. These reconstructions suggest that in the Pacific Northwest (PNW) the Medieval Climate Anomaly (MCA) and Little Ice Age (LIA) were relatively drier and wetter, respectively. Interestingly, lake sediment oxygen isotope records from Castor and Scanlon lakes suggest the opposite, i.e., that the MCA was relatively wetter and the LIA was drier. The robustness of the tree-ring based PDSI reconstructions is defined by the number of available chronologies for a given time period and the strength of the verification correlation with the 20th century instrumental PDSI values. Many tree-ring records from the PNW span the last millennium; however the number of datasets decreases through time, such that during the MCA many of the records from the PNW are based on as few as one chronology. An additional limitation of tree-ring records lies in the techniques used to remove non-climatic variability, such as that caused by trends in the relationship between tree size and age and interactions with neighbors, which can potentially limit the faithful representation of climate variations on centennial and longer time scales. Lake sediment isotope records as archives of drought/pluvial history are not subject to the same set of restrictions that limit tree-ring chronologies. By analyzing oxygen isotope records from both open- and closed-basin lakes, the influence of variations in the source water (i.e., precipitation) isotopic composition can be subtracted from closed-basin lake records to produce a reconstruction of variations in precipitation-evaporation balance. Moreover, lake-catchment isotope mass balance models can be used to quantitatively interpret lake sediment isotope records and thereby relate measured sediment oxygen isotope ratios to specific, probabilistic rates of past precipitation or temperature. We therefore contend that reconstructions of long-term aridity developed using combined closed- and open-basin lake sediment records interpreted with lake-catchment mass balance models are potentially more robust than similar tree ring reconstructions, especially in locations such as the PNW for which few long-term tree-ring chronologies exist.