Can we disentangle the climate influences on Colorado River water resources using a multi-proxy approach to tree-ring data?

English Abstract:

Snowpack in the mountains of Colorado is one of the most important resevoirs of water in the western US. Recent declines in snowpack, and its associated impact on streamflow in the Colorado river are increasingly being linked to climate influences other than total winter precipitation. Recent work indicates that spring temperatures are becoming an increasingly important driver of post-1980's snowpack declines in the Southern Rockies. Records of tree-ring widths have been used to generate long term reconstructions of Colorado River streamflow based on the sensitivity of tree growth to cool season precipitation. These long term proxy records of Colorado River streamflow have proven extremely valuable to water resource managers. Yet the influence of temperature and warm season precipitation, not captured by the ring-width proxy, have in the past served to ameliorate or exacerbate decadal-scale drought/pluvial events in a way that presents challenges to water resource managment. Until now, disentagling the relative roles of temperature and precipitation in driving major decadal-scale departures in upper Colorado River Basin (UCRB) water resources over the past centuries has not been possible because of a lack of proxy information for climate parameters asside from cool season precipitation. Here we have investigated δ¹⁸O records derived from tree-rings of Douglass-fir and piñon pine from the UCRB to determine whether multi-proxy studies of tree-rings could help us disentagle the various climatic drivers of snowpack and streamflow. Results from the low-elevation Douglass-fir site and the mid-elevation piñon pine site show a moderate correlation between δ¹⁸O and cool season temperatures (r = 0.54). The high-elevation Douglass-fir site and low-elevation piñon pine site correlated most strongly with summer precipiation (r = -0.49). Our results suggest 1) that site-to-site differences in the isotopic relationship are more important than species and 2) That tree-ring δ¹⁸O records do contain information to supplement ring widths. A better understanding of the causes of between site variability could help us to better select sites in the future. We are presently developing long term isotopic records to assess the role of these variable on streamflow beyond the instrumental period.