Packrat Midden Research in the Grand Canyon

Author: Kenneth Cole, USGS, FRESC Colorado Plateau Field Station

The zonation of vegetation types in elevational bands has been a subject for scientific analysis since C. Hart Merriam first described these zones ringing the San Francisco Peaks in 1890. In northern Arizona, this elevational zonation of plant communities extends from the arctic tundra (above 12,000 feet) down to arid deserts within the Grand Canyon (below 2000 feet). This extreme range of habitat types is rarely found within such a compact geographic area.

Until fairly recently, the fate of desert areas during the ice age (the Pleistocene), was poorly known. Deserts usually contained far less of the fossil evidence of past environments than more humid areas because most fossils are preserved in water-borne deposits such as lakes. There was speculation as to whether deserts were less affected by the ice ages, or whether the current elevational zones of vegetation were formerly located at much lower elevations. Some scientists believed that the equatorial regions were less affected by the ice ages than high latitudes, and it followed that perhaps deserts would be similar. Another important theoretical disagreement involved the nature of plant communities. Were assemblages of plants that grow together now part of larger interconnected units, or were they just an assortment of species that happened to grow in similar locations today? The former opinion suggested that the modern plant communities would migrate together through changing climates, while the latter suggested that each species would behave individualistically, forming new assemblages in each climate era.

It is worth remembering that our current non-glacial interval, the Holocene, is a climatic exception, not the rule. Perhaps as much as 90% of the last 1.8 million years have been glacial periods. So, the Pleistocene distribution of plants better represents the "normal" state of the environment during the time that North American plants and animals were evolving.

Outcrops of Tapeats Sandstone and Redwall Limestone in the Nankoweap Basin where fossil packrat middens are abundant.

In the 1960s it was discovered that deposits left in caves and crevices by packrats (packrat middens) contained assemblages of fossil plants that formerly grew nearby. Packrat middens turned out to be abundant within the Grand Canyon. Although the field work is difficult due to the rugged landscape, several rock formations within the canyon provide excellent shelters, preserving packrat middens from the elements for tens of thousands of years. Since numerous vegetation zones occur over relatively short distances in Grand Canyon, the newly discovered packrat middens became an ideal tool for comparing the Pleistocene vegetation zones with the modern ones.

In general, the packrat middens from the Grand Canyon revealed that many plant species are now growing 700 to 900 m (2300 to 3000 feet)  higher in elevation and 400 to 700 km (250 to 430 miles) further up-river than the were in the Pleistocene (Figure 1). However, most ice-age plant assemblages were unique, with their modern analog either unknown or having a restricted distribution. Similarly, some modern communities either had no analog in the Pleistocene, or they were too restricted in distribution to produce an abundant fossil record.

These data supported the notion that desert vegetation zones had been displaced an amount similar to higher elevational zones such as treeline and snowline (studied using fossil pollen in high lakes and glacial deposits on the San Francisco Peaks). More importantly, the data also demonstrated that individual plant species moved independently to each other through time. They tended to associate and disassociate with each other through time, making plant communities transitory. Although this conclusion has been firmly supported by other paleoecological and modern studies as well, many ecologists continue to assume that a plant community is a real unit rather than simply a temporary grouping of species with similar requirements in today's climate. In the greatly altered climates of the near future (the next 200 years), there is little doubt that these familiar associations will be replaced by new groupings.

The Grand Canyon midden record reveals the last ice age to be a time of relatively stable conditions when compared to the Holocene. The lowest portions of the Grand Canyon supported colder desert associations containing junipers. Today's Sonoran Desert areas of the canyon contained assemblages more typical of today's Mojave Desert while the higher deserts were more typical of the modern Great Basin deserts in Utah. Although juniper twigs were abundant within the middens, the packrat's preference for this food item makes determining the density of junipers difficult. Most researchers refer to this as a "juniper woodland". But, considering the other species present in the middens, these plant associations were likely more similar to the shrubby deserts within today's Mojave, where at low elevations juniper is a small shrub rather than a tree.

Higher elevations inside the canyon supported a mixed coniferous forest of Douglas fir, limber pine, and spruce at the higher elevations.  Many hot-desert species which are common in the canyon today -- such as creosote bush, brittle brush, ocotillo, and white bursage -- were not present during the last glaciation. Pinyon pine (Pinus edulis) is abundant in the canyon today, but during the Pleistocene limber pine (Pinus flexilis) was the abundant pine species. Today limber pine grows mostly at more northerly latitudes. Despite its large modern range across the Colorado Plateau, ponderosa pine (Pinus ponderosa) was either absent or too infrequent to be recorded in the Grand Canyon during the last ice age.  It arrives within the canyon in the early Holocene, about 9000 years ago. Its presence within the canyon at that time (along with other species) suggests that by 9000 years ago the summer monsoon rainfall was likely greater than today.

Cole, K. L. 1982. Late Quaternary zonation of vegetation in the eastern Grand Canyon. Science 217: 1142-1145.

Cole, K. L. 1985. Past rates of change, species richness, and a model of vegetational inertia in the Grand Canyon, Arizona. American Naturalist 125: 289-303.

Cole, K. L. 1990. Reconstruction of past desert vegetation along the Colorado River using packrat middens. Palaeogeography, Palaeoclimatology, and Palaeoecology 76: 349-366.

Cole, K.L. 1990. Late Quaternary vegetation gradients through the Grand Canyon.  Pp. 240-258 In: Betancourt, J.L., Van Devender, T.R., Martin, P.S., editors. Packrat Middens: the Last 40,000 years of Biotic Change. University of Arizona Press, Tucson.

Cole, K.L. 1999. The Holocene and Pleistocene vegetation history of the Grand Canyon. http://www.usgs.nau.edu/Global_Change/grdcany.html   5/10/99.