El Nino effects on the dynamics and control of an island ecosystem in the Gulf of California

Ecology, Sept, 1997 by Gary A. Polis, Stephen D. Hurd, C. Todd Jackson, Francisco Sanchez Pinero

INTRODUCTION

Primary productivity is an important determinant of community dynamics and structure in ecosystems worldwide. It strongly affects the abundance of primary consumers, the presence and relative success of consumer populations at higher trophic positions, and the importance of consumer-resource interactions (Oksanen et al. 1981, Oksanen 1988, Hunter and Price 1992, Power 1992, DeAngelis et al. 1996, Persson et al. 1996). Primary productivity is influenced by many stochastically varying factors (e.g., nutrients, light, temperature, species composition), but it is also strongly altered by large-scale climatic disturbances. One of the most prominent disturbances is the cyclical El Nino-Southern Oscillation (ENSO) oceanographic/meteorological phenomenon. The Southern Oscillation refers to the pressure differential between the South Pacific subtropical high-pressure region and the Indonesian equatorial low-pressure region. Occasionally, this pressure differential "flips"; in association with the relaxation of previously strong trade winds, this allows warm water from the western Pacific to move eastward to South America (Ramage 1986, Enfield 1987, Quinn 1987). This trans-Pacific movement and the subsequent spread of warm water far southward along the Peruvian coast is known as El Nino. El Nino occurs irregularly at intervals of 3-12 yr (Quinn 1987), and strongly affects marine ecosystems throughout the Pacific (Barber and Chavez 1983, Avaria and Munoz 1987, Huyer et al. 1987, Torres Moye and Alvarez Borrego 1987, Wilkerson et al. 1987, Glynn 1988, Silva Cota and Alvarez Borrego 1988, Barry and Dayton 1991, Karl et al. 1995).

Through its effects on precipitation, ENSO exerts a tremendous impact on productivity and community dynamics in many terrestrial systems. The Southern Oscillation can bring drought to Australia and Indonesia (Quinn et al. 1978, Canby 1984, Nicholls 1991); concurrently, El Ninos often bring large storms to the west coast of the Americas, greatly increasing precipitation (Markham and McLain 1977, Quinn and Neal 1983, Canby 1984, Cucalon 1987, Goldberg et al. 1987). Primary productivity on land is tightly correlated with precipitation (Rosenzweig 1968, Lieth 1973, Davidson 1977, Polis and Farley 1979, Noy-Meir 1981, 1985, Ludwig 1986, 1987, Polis 1991), and secondary productivity (consumer biomass) shows a high positive correlation with primary productivity and thus with precipitation (Davidson 1977, Orians and Solbrig 1977, Noy-Meir 1981, 1985, MacKay and MacKay 1984, Marsh 1986, Pianka 1986, Brown 1987, Vitt 1991). This is true at spatial scales across ecosystems (e.g., forests to deserts) and at temporal scales within ecosystems (e.g., "wet" vs. "dry" years).

The effects of ENSO on terrestrial productivity are particularly dramatic in arid regions, where many plants respond dramatically to pulses of precipitation. ENSO-related changes in terrestrial communities have been documented for the Atacama and Peruvian desert (Goldberg et al. 1987, Asa and Wallace 1990, Dillon and Rundel 1990, Meserve et al. 1993), the Chihuahuan desert (Brown and Heske 1990), the arid Australian interior (Canby 1984, Nicholls 1991), and the Galapagos Islands (Hamann 1985, Gibbs and Grant 1987, Trillmich 1991, Desender et al. 1992, Grant and Grant 1993).

Polis and Hurd (1995, 1996a, b) previously described an arid system of desert islands in the Midriff region of the Gulf of California, Baja California Norte, Mexico. Under normal arid conditions, this system is energetically driven by allochthonous input of marine biomass rather than by in situ primary productivity by terrestrial plants. Heavy El Nino precipitation could fundamentally affect the dynamics of this system by changing the energy base of the community; as such, it represents a potential switch in control from limitation by allochthonous productivity (dependent on the productivity of the surrounding waters and on such landscape variables as island perimeter/area ratio; Polis and Hurd 1996a) to limitation by autochthonous productivity (dependent on in situ characteristics such as plant species, soil quality, soil seed bank, and rain catchment). Here, we quantify the effects of the 1992 El Nino on terrestrial producers and primary consumers on these desert islands, and we delineate the immediate and longer-term effects of a change in the resource base. Elsewhere, we trace the direct and indirect effects of these changes up the food web to secondary consumers (G. A. Polis et al., unpublished manuscript).

MATERIALS AND METHODS

Study site

Field work was conducted from May-August in 1990-1994, October 1994, March 1994, and March 1995. The research involved islands in the Midriff region of the Gulf of California, near Bahia de los Angeles (28 [degrees] 55 [minutes] N latitude, 113[degrees] 30 [minutes] W longitude), Baja California Norte, Mexico [ILLUSTRATION FOR FIGURE 1 OMITTED]. The principal study site includes 16 islands between Bahia de los Angeles and Isla Angel de la Guarda, four islands at Bahia Animas (15 km ESE of Bahia de los Angeles), and the adjacent coast along the Baja California peninsula. This area is one of the driest in North America; mean annual precipitation at Bahia de los Angeles is only 59 mm (Reyes Coca et al. 1990). The study islands are within 20 km of the Baja California peninsula and lie in a region of the Gulf characterized by year-round upwelling due to tidal mixing and winds (Maluf 1983). Perimeter and area measurements of islands are from Due (1992).