Drivers of Spatial Variation in Plant Diversity Across the Central Arizona-Phoenix Ecosystem

ABSTRACT

We examined how growth of the Phoenix urban landscape has changed spatial patterns in native Sonoran desert plant diversity. Combining data from the U.S. Census with a probability-based field inventory, we used spatial and multivariate statistics to show how plant diversity across the region is influenced by human actions. Spatial variations in plant diversity among sites were best explained by current and former land use, income, housing age, and elevation. Despite similar average diversity in perennial plant genera between desert and urban sites, numerous imported exotics have significantly increased variation in plant generic composition among urban sites, with a “luxury effect” of higher plant diversity at sites in wealthier neighborhoods. We conclude that controls on natural spatially autocorrelated desert plant diversity are replaced by a variable suite of site-specific human factors and legacy effects, which require an integration of ecology and social science to be fully understood.

Keywords:

• desert
• geomorphic template
• integrated field survey
• perennial plant genera
• socioeconomic factors
• spatial statistics
• urban ecosystem

We thank Steven S. Carroll for sampling design and statistical assistance, Amy Nelson for U.S. Census data work and input, and Ann Kinzig for comments on the manuscript; M. Myers, A. Budet, S. Paine, M. Clary, A. Stiles, L. Stabler, and S. Holland for field and lab assistance; Salt River Project for the donation of helicopter time; and the cities of Phoenix, Scottsdale, and Tempe, Maricopa County Parks, Tonto National Forest, Arizona State Lands Department, Sky Harbor Airport, and all the private property owners involved for giving us permission to access their land. This work was funded by National Science Foundation grants DEB-9714833 and DEB-0423704 (the Central Arizona–Phoenix Long-Term Ecological Research Project)

Notes

Note. Plant diversity is usually reported at the species rather than at the genus level. Individual plant genera often contain species that occupy entirely different habitats or niches that can render diversity at the generic level of little ecological meaning. However, across the Central Arizona–Phoenix (CAP) ecosystem perennial plant diversity on the genus level corresponds closely to diversity at the species level, except in the case of two desert genera (Ambrosia and Cylindropuntia), of which there were two or three species present in approximately 30 plots. Moreover, in this study we discuss only total plant diversity at the generic level, with no distinction being made between native and exotic genera.

Note. Perennial plant diversity is represented by number of woody plant genera per plot. Significant variables are listed in order of importance as judged by the level of significance denoted by asterisks (***p < .0001; **p < .001; *p < .01; no asterisk, p < .05). Test statistics are either t value, chi-square, or F value. In the case of plant diversity for the whole site, all of the F statistics are approximations and have degrees of freedom 1 and 196.