Other paleo proxy records

Tree rings are not the only environmental proxy that can be used to examine past streamflow and climate for the Colorado River basin and the southwest U.S. Fossil clams, packrat middens, pollen in lake and bog sediments, and changes in lake chemistry in lake sediments all record information about the past. These proxies can provide longer records than with the tree-ring data, but the information has coarser time-resolution than tree rings. (Most of these proxies are based on radiocarbon dating, which has a dating precision of decades to centuries, compared to the annual precision of tree-ring data.) Instead, these proxies record slower shifts in precipitation or temperature regimes, complementing the high-resolution information in the tree rings.


Oxygen isotopes in clam shells in the Colorado River delta in the Gulf of California closely match the salinity level of the delta estuary water, and salinity level is very closely tied to the amount of streamflow. Thus, after calibrating the isotope levels in modern clams with measured salinity and gaged streamflow (much like calibrating tree ring records), the isotope levels in long-dead clams can in principle be used to estimate past streamflow (Dettman et al. 2004). A clam-based streamflow reconstruction of the Colorado River at its mouth is currently being developed, and it will be interesting to compare these results to the tree-ring reconstruction.

For more information about the use of clams to study past environmental conditions, see the website for the Centro de Estudios de Almejas Muertas at the University of Arizona.

Packrat Middens

Packrats (genus Neotema) live throughout the interior western US. They bring large quantities of nearby plant material into their dens, inadvertently creating excellent records of the composition of the vegetation. The urine of the packrats dries and cements the plant material together, preserving these middens for thousands of years. Many packrat middens have been collected in the Colorado River basin (see map here). In a recent study, middens sampled in the Gunnison Basin in western Colorado indicated changes in vegetation over the past 4000 years, including evidence of warmer (and possibly drier) conditions from about AD 500-1050, and cooler (and possibly wetter) conditions from about 1350-1800 (Emslie et al. 2005).

For more information about packrat middens, see the USGS/NOAA North American Packrat Midden Database.


Sediments in lake bottoms and bogs serve as natural repositories for pollen grains that are released annually from nearby plants. Cores taken from lake bottoms are sampled at intervals along the core. The pollen present at each interval provides a "snapshot" of the species composition of local vegetation. Changes in the pollen record over time (for example, from domination by pine pollen to sagebrush pollen) can be taken to represent changes in climate, in the example, from wetter to drier conditions. As mentioned previously, the pollen record cannot be dated as precisely as tree rings. Nonetheless, pollen records can help confirm the general timing of major climatic shifts in a region.

Lake sediments

Lakes are a repository of sediments from their surrounding environments, and sediment cores taken from lakes can record environmental (including climatic) changes. Lake sediments can provide a number of different environmental proxies, including pollen, described above. Another useful proxy is the changes in isotopes of oxygen and carbon within carbonate-rich sediments. Oxygen isotopes can reflect the varying composition and sources of water coming into the lake, thus indicating precipitation variability and other climatic features. Work is currently underway to sample and analyze the sediment records from several lakes in western Colorado, with the goal of extracting long (up to 10,000 years) proxy climate records to compare with nearby tree-ring records. The tree-ring records can help define the nature of the climate information in the lake chemistry. The lake sediment records could be used to extend our knowledge about hydrologic variability in the upper Colorado River basin thousands of years into the past.