The El Nino southern oscillation, variable fruit production, and famine in a tropical forest

Ecology, July, 1999 by S. Joseph Wright, Claudio Carrasco, Osvaldo Calderon, Steven Paton

INTRODUCTION

We hypothesize that the El Nino Southern Oscillation influences fruit production which, in turn, limits frugivorous and granivorous mammals on Barro Colorado Island (BCI), Panama. El Nino events bring warm, dry, sunny conditions to BCI and to a large portion of the wet tropics (see Introduction: El Nino Southern Oscillation). If these conditions generally favor fruit production, then similar effects may be expected wherever fruit production limits frugivores in wet tropical forests.

At least two attributes of El Nino events might favor fruit production. Drought affects most of the wet tropics during El Nino events (Ropelewski and Halpert 1987). Threshold levels of drought synchronize flowering and enhance seed set in selected tropical forest plants (Alvim 1960). If threshold drought responses were widespread, then drought would enhance community-level fruit production during El Nino events. Reduced cloudiness is the second attribute of El Nino events that might favor fruit production. Clouds absorb photosynthetically active radiation (PAR), and low PAR may limit many tropical forest plants, including trees and lianas (Van Schaik et al. 1993, Wright and Van Schaik 1994). El Nino events, high levels of PAR and mast fruiting may be associated in peninsular Malaysia. Here, members of 41 plant families reproduce in synchronized mast-fruiting events at irregular, multiyear intervals (Appanah 1985). These masting events coincide with unusually sunny years and also with El Nino events (Van Schaik 1986, Ashton et al. 1988). Associations between fruit production and El Nino events are unknown outside Malesia.

Seasonally variable fruit production limits frugivores and granivores in many tropical forests (reviewed by Van Schaik et al. 1993). Many frugivores reproduce, grow rapidly, store fat, and/or cache seeds during the season of greatest fruit production. Many of the same species lose body mass, suffer increased mortality, and/or emigrate during the season of scarcity. Interannual variation in fruit production may also limit frugivores. Janzen (1974), for example, hypothesized that prolonged scarcity limits frugivores and granivores during nonmast years in Malesian dipterocarp forests. This hypothesis remains untested, perhaps because few tropical studies span enough years to evaluate population fluctuations.

One of the longest ecological records from tropical forest concerns the mammals of BCI. Frugivorous mammals experienced famine in 1931-1932, 19581959, and 1970-1971. The famines all occurred between September and January. In 1931-1932, coatis (Nasua narica), peccaries, and brocket deer (Mazama americana) were emaciated, and collared and white-lipped peccaries (Tayassu tajacu and T. pecari) foraged in the laboratory clearing (Enders 1935, Chapman 1938). In 1958-1959, collared peccaries and deer were again emaciated, and more than half of the individuals in a marked population of coatis died (Kaufman 1962). At the height of the 1970-1971 famine, corpses of frugivores occurred, on average, every 300 m along forest trails (Foster 1982a). Foster (1982a) analyzed 18 years of BCI mammal observations and noted that the famines each followed a mild dry season. Foster also documented fruit production for two years, including the 1970-1971 famine. Many plant species failed to fruit after the mild 1970 dry season, suggesting a link between dry-season rains, failed fruit production, and famine. Foster hypothesized that dry-season rains prevent many species from attaining the threshold levels of drought required to initiate flowering. Recent events indicate that this hypothesis is, at best, partly correct. The two rainiest dry seasons yet recorded on BCI (1981 and 1996) were not followed by fruit failure or by famine.

We re-evaluate the relationships among climate, fruit production, and mammal populations. Observations of BCI mammals have been compiled for 49 years, and fruit production has been monitored for 10 years. A fourth famine in 1993-1994 is evident from an 11-yr record of natural deaths of mammals and a 2-yr record of population densities. The famines of 1930-1931, 1958-1959, 1970-1971, and 1993-1994 each followed one year after an El Nino event when the intervening dry season was mild. We evaluate the following hypotheses for 138 plant species: (1) El Nino conditions enhance fruit production; (2) high fruit production consumes stored reserves, limiting the next reproductive event; and (3) mild dry seasons reduce fruit production.

Terborgh (1992) hypothesized that large felids and raptors regulate populations of frugivorous and granivorous mammals in tropical forests. The three largest raptors and two largest felids recorded in central Panama are no longer resident on BCI. Terborgh (1992) hypothesized that reduced predation pressure has permitted the population densities of prey species to increase on BCI. As a corollary, he suggested that famines occur because the population densities of frugivores and granivores are not regulated by predation on BCI. We extended censuses to the nearby mainland, where poachers are active, to evaluate this hypothesis. Poachers reduce the abundances of game species below levels maintained in the presence of large felids and raptors (Freese et al. 1982, Peres 1990, 1996, Redford 1992). If there is evidence for famine in poached populations, we will reject the hypothesis that reduced predation pressure is a prerequisite for famine.