The University of Arizona Desert Laboratory Past Research Highlights

Dynamics of Holocene Plant Migration: Role of Landscape Structure and Climate Variability

Mark Lyford & Steve Jackson, University of Wyoming
Julio Betancourt, Desert Laboratory, USGS & University of Arizona
Steve Gray, University of Montana

Ecologists are devoting considerable effort to understanding and modeling mechanisms underlying biological invasions. Prehistoric plant migrations can serve as model systems for understanding ongoing and future natural invasions, as well as for studying how landscape structure and climatic variability influence both alien and natural invasions over timescales of centuries to millennia.

Paleoecological studies of plant migration require spatial networks of time-series data that record local establishment and expansion of populations. In reconstructing plant migrations using pollen stratigraphies, some pitfalls include a lack of taxonomic precision (e.g., most pollen grains can only be identified to family or genus), and the inability to discriminate between small, local populations and distant, regional ones. Packrat midden analysis, one of the principal paleoecological methods employed at the Desert Laboratory, overcomes these pitfalls. Plant macrofossils in middens usually can be identified to species, and originate from plants growing within 50 meters of the packrat den.

Beginning in 1998, the Desert Laboratory teamed up with Steve Jackson and his Quaternary Plant Ecology Laboratory at the University of Wyoming to reconstruct migrational histories at the northern distributions of key woodland trees (Utah juniper, pinyon pine, and ponderosa pine) in the Central Rockies. The effort involves synoptic development of packrat midden chronologies along presumed pathways for Holocene migration.

Map showing distribution of Utah juniper.
Map of Wyoming and adjoining state showing patchy distribution of Utah juniper (Juniperus osteosperma) in gray and bracketing dates (in thousands of calendar years before present) and sequence of arrival at each of 12 sites where midden series were developed. Open circles denote absence, closed ones presence of Utah juniper in midden series.

Holocene migration of Utah juniper into central and northern Wyoming and southern Montana from the south proceeded by a series of long-distance dispersal events, which were paced by climate variability and structured by the geographic distribution and connectivity of suitable habitats on the landscape. Its migration into the region involved multiple long-distance dispersal events, ranging from 30 to 135 kilometers. One of the earliest-established populations, on East Pryor Mountain in south-central Montana, is currently the northernmost population and among the most favorable sites for Utah juniper. Establishment by long-distance dispersal of that population and another in the Bighorn Basin occurred during a period of relatively dry climate between 7500 and 5400 years ago. Further expansion of these initial colonizing populations and backfilling to occupy suitable sites to the south was delayed during a wet period from 5400 to 2800 years ago. Development of dry conditions 2800 years ago led to a rapid expansion in which Utah juniper colonized sites throughout its current range. Recent encroachment of Wyoming grasslands by Utah juniper is probably due to the effects of livestock grazing, though increasing tree densities on rockier terrain may represent continued infilling of populations after late Holocene expansion. Our migration studies indicate that landscape structure and climate variability play important roles in governing the pattern and pace of natural invasions, and deserve close attention in studying and modeling plant invasions, whether alien or natural.