March 2005
Influence of Climatic Variability on Local Population Dynamics of a Sonoran Desert Platyopuntia
Janice E. Bowers, USGS, Desert Laboratory, Tucson AZ
Proliferation of Opuntia (Cactaceae), especially the platyopuntias or prickly-pears, has been noted at many times and locations, as on the Great Plains in the 1930s and in Australia in the 1910s and 1920s. Populations of certain species can reach densities > 4,000 plants/ha. Proliferation or invasion of platyopuntias has been variously attributed to grazing, drought, efficient dispersal of seeds, disturbance, absence of predators, released competitive pressure, or a combination of these and other factors. Some ranchers and land managers consider platyopuntias to be pest species that require biological or chemical control, but in their native range, platyopuntias are an integral component of food webs, providing a reliable source of food and moisture for insects, birds, and mammals at a time of year when little else is available. In addition, platyopuntias can be an important element of community structure in arid and semi-arid North America.
Whether platyopuntias are native or introduced, pulses of establishment are necessarily demographic phenomena. With a few exceptions, however, little attention has been paid to their population age structure. Size-class distributions have been used as a substitute for population age structure but cannot be conveniently correlated with dated events in the past such as disturbance or seasonal rainfall. Age-based analysis makes it possible to correlate peaks and troughs in establishment with climatic trends over the lifetime of a population. Although platyopuntia wood does not form annual growth rings, age can be estimated from the relation between annual growth and plant size, that is, number of cladodes. Height-growth models have long been used to estimate ages of columnar cacti. Such age estimations, along with repeated censuses, have been powerful tools for understanding cycles of population growth and decline.
I investigated age-based population dynamics of Opuntia engelmannii Salm-Dyck. (Engelmann prickly-pear) at the Desert Laboratory, which has been protected from grazing by domestic livestock since 1907. Opuntia engelmannii, a shrubby cactus that reaches about one meter in height, is widely distributed in the southwestern United States at elevations of 300 to 2000 m above sea level. Fecundity is highly variable from year to year, depending on winter rain, predispersal seed predation, and meristem allocation. Germination requires ample summer rains. Establishment is infrequent. Plants reach reproductive maturity at nine to 11 years of age. Few individuals live longer than 25 to 30 years. Plants can reproduce vegetatively when detached cladodes become rooted in the ground. My ultimate objective was to determine how a species with moderate life span, variable fecundity, and episodic establishment can persist, much less increase, in a climatically variable environment where severe drought can last a decade or longer.
In 1996, I determined plant size (number of cladodes) for all living plants in two large plots, then used the relation between size and age to estimate year of establishment for each plant. I repeated the census seven years later. After grouping plants into annual cohorts, I plotted population age distribution as a function of estimated age.
Age-frequency distribution for Opuntia engelmannii at Tumamoc Hill.
Bars represent number of plants in one-year age classes. A) North Slope
Plot, 1996; B) Northwest Corner Plot, 1996; C) North Slope Plot, 2003;
D) Northwest Corner Plot 2003.
The two plots presented strikingly different age structures in both censuses. In 1996, the Northwest Corner sample was young, with an average age of 12.8 ± 4.3 years. The North Slope sample was older on average, 18.1 ± 4.7 years. During the eight years between censuses, the North Slope subpopulation declined in size, while the Northwest Corner subpopulation remained stable. Despite the loss of many individuals, average age of the North Slope sample in 2003 remained steady at 17.7 ± 6.2 years. The Northwest Corner subpopulation reached an average age of 17.0 ± 4.2 years in 2003, about the same as for plants on the North Slope.
I used age-frequency distribution to calculate residual regeneration in 1996 and 2003. Residual regeneration is an index of the difference between observed cohort size and idealized survivorship. From residual regeneration, it was apparent that establishment peaks occurred in the late 1970s, the mid-1980s, and the early to mid-1990s.
Residual regeneration of Opuntia engelmannii at Tumamoc Hill.
Residual regeneration is an index of the difference between observed cohort
size and idealized survivorship. A) North Slope Plot, 1996; B) Northwest
Corner Plot, 1996; C) North Slope Plot, 2003; D) Northwest Corner Plot
2003.
The influence of climatic variability on establishment was examined using principal components regression, a procedure that combines principal components analysis (PCA) with stepwise multiple regression. PCA was used to extract major climatic components from 17 original climatic variables, then the relation between principal climatic components and residual regeneration was examined with stepwise multiple regression. Plots were pooled within censuses for analysis. The results indicated that establishment peaks during the 1970s, 1980s, and 1990s coincided with increased winter moisture in the years before germination, ample summer rain in the year of germination, and decreased drought in the years after germination, reflecting favorable conditions for fruit production, seed germination, and seedling survival. At the Desert Laboratory, rapid population growth occurred during periods of favorable climate for regeneration, and declines were precipitated as large cohorts reached the end of their natural lifespan. Drought-related mortality probably played a role in declines as well.
In the last half of the twentieth century, populations of O. engelmannii in the northern Sonoran Desert increased markedly at many locations. As with other patterns showing regional synchronization, this resurgence can probably be ascribed to large-scale climatic phenomena. One major source of interannual variability in winter rain in the region is the El Niño-Southern Oscillation phenomenon (ENSO). Peaks of O. engelmannii establishment at the Desert Laboratory since 1975 coincided with periods of frequent El Niño conditions: the late 1970s peak with the El Niños of 1976-78 and 1979-80; the mid-1980s peak with the El Niños of 1982-83 and 1986-87; and the early to mid-1990s peak with prolonged El Niño conditions from 1991 to 1995.
A concurrent source of climatic variability in the region is the Pacific Decadal Oscillation (PDO). Like ENSO, PDO exerts a strong influence on cycles of drought and wetness, but over periods lasting two to three decades. Because PDO can influence both winter and summer rain, prolonged negative PDO should have a deleterious effect on all aspects of regeneration, including fecundity, germination, and seedling survival, resulting in long periods with little regeneration. Opuntia engelmannii populations in southern Arizona tended to be stable and rather small between 1890 and 1985, overlapping with mostly negative PDO from 1890 to 1924 and again from 1947 to 1976. In the late 1990s, PDO shifted to negative values, and at the Desert Laboratory there has been little regeneration since then. If PDO stays negative, as some observers think likely, O. engelmannii establishment will remain low for two to three decades.
Opuntia engelmannii, like other Sonoran Desert cacti, experiences natural cycles of population growth and decline. Regionally, pulses and gaps in establishment should coincide with the frequency and amplitude of large-scale climatic phenomena that affect cycles of moisture and drought on decadal and interdecadal scales. Because of local factors, however, subpopulations within a few km of one another can experience virtually identical climates yet differ strikingly in age structure and density. For example, the North Slope sample in 1996 had experienced little recruitment in recent years and was much older on average than the Northwest Corner sample, where recent recruitment had been strong. The patchiness of increase and decline at the Desert Laboratory suggests that population dynamics on a larger scale should form a mosaic in which some populations are increasing in response to favorable climatic signals while others remain out of step, perhaps because deaths from disease or old age have diminished recruitment to the point where favorable climate cannot reverse a decline.
Bowers, J. E., 2005. Influence of climatic variability on local population dynamics of a Sonoran Desert platyopuntia. Journal of Arid Environments 6:193-210.
For more information contact:
Janice E. Bowers (jebowers@usgs.gov)