Tropical cloud forest climate variability and thedemise of the Monteverde golden toadKevin J. Anchukaitisa,b,1and Michael N. Evansa,b,caLamont-Doherty Earth Observatory of Columbia University, Palisades, NY 10964;bLaboratory of Tree-Ring Research, University of Arizona, Tucson, AZ85721; andcDepartment of Geology & Earth System Science Interdisciplinary Center, University of Maryland, College Park, MD 20740Edited by Lisa Graumlich, University of Arizona, Tucson, AZ, and accepted by the Editorial Board February 1, 2010 (received for review July 29, 2009)Widespread amphibian extinctions in the mountains of the Amer-ican tropics have been blamed on the interaction of anthropogenicclimate change and a lethal pathogen. However, limited meteoro-logical records make it difficult to conclude whether current climateconditions at these sites are actually exceptional in the context ofnatural variability. We use stable oxygen isotope measurementsfrom trees without annual rings to reconstruct a century ofhydroclimatology in the Monteverde Cloud Forest of Costa Rica.High-resolution measurements reveal coherent isotope cycles thatprovide annual chronological control and paleoclimate informa-tion. Climate variability is dominated by interannual variance indry season moisture associated with El Niño Southern Oscillationevents. There is no evidence of a trend associated with globalwarming. Rather, the extinction of the Monteverde golden toad(Bufo periglenes) appears to have coincided with an exceptionallydry interval caused by the 1986–1987 El Niño event.Costa Rica ∣ ENSO ∣ extinction ∣ stable isotopesThe demise of the Monteverde golden toad (Bufo periglenes)inthe montane cloud forest of Costa Rica in 1987–1988, as wellas subsequent amphibian extinctions throughout the Americantropics, has been believed to be a consequence of the interactionof global warming (1, 2) and the introduced chytrid fungusBatrachochytrium dendrobatidis (3–5). Analysis of the limitedavailable weather data from Monteverde has suggested that atrend toward decreasing immersion cover since the late 1970s re-flects the influence of increasing tropical air and sea surface tem-peratures (SST) (1, 2). Pounds and coauthors (1) argued that thispattern of increasing dry days implicated rising global tempera-tures due to anthropogenic greenhouse gas emissions. Nair et al.(6) and Lawton et al. (7) used a regional atmospheric model todemonstrate that deforestation in the tropical lowland forests up-wind from the Monteverde Cloud Forest could also potentiallyresult in elevated cloud base height and drier conditions.Subsequent examination of a larger dataset of the last year ofobservation (LYO) of extinct neotropical amphibians in the genusAtelopus found a positive, lagged correlation between meanannual tropical (30 °N to 30 °S) temperatures and the timingof Atelopus extinctions (2). The authors hypothesized thattemperatures at many montane locations in the Americas wereshifting towards an “optimum range” for the growth of the chytridfungus. This “chytrid-thermal-optimum hypothesis” suggestedthat tropical temperature trends associated with anthropogenicglobal warming were responsible for widespread amphibian ex-tinctions due to chytridiomycosis.This “climate-linked epidemic hypothesis” (2) postulates thatglobal warming contributes to the development of optimalenvironmental conditions for the fungus. Upward trends in globaltemperatures could therefore already be fundamentally alteringthe suite of climatic conditions that maintain mountain forestecosystems in Central America. General circulation model simu-lations of climate under doubled CO2conditions predict higherlifting condensation levels and reduced cloud contact for tropicalmontane cloud forests as a result of increasing temperatures (8).Imprinted on the apparent instrumental drying trend at Monte-verde, El Niño events also cause local increases in temperatureand reductions in cloud cover and moisture (9). However, therole of climate change in neotropical anuran extinctions has beenquestioned on statistical grounds (10, 11). Without the contextprovided by long-term climate records, it is difficult to confidentlyconclude whether the extinction of the golden toad and otherchanges in cloud forest ecology are the consequence of anthro-pogenic climate forcing (2), land-surface feedbacks due to defor-estation (6, 7), or natural variability in tropical climate (9).ApproachWe developed annual proxy records of climate variability fromtropical montane cloud forest trees without annual rings atMonteverde (10.2 °N, 85.35 °W, 1500–1800 m) over the lastcentury using stable isotopes (12, 13). The advantage of thisapproach is that it does not rely on the formation of yearly mor-phological growth hiatuses in trees, which in the tropics can oftenbe absent (Fig. 1A), but rather takes advantage of the annual var-iation in the stable oxygen isotope ratio (δ18O) of water used byplants over the course of a year. This technique can also be ap-plied where posited “growth rings” cannot be shown to be annual.Annual oxygen isotope cycles have been identified in lowlandtropical forests (12, 14), and at Monteverde we have previouslydemonstrated that variation between the summer wet season andthe cloud-dominated winter dry season is sufficient to induce anannual δ18O cycle along the radial xylem growth of cloud foresttrees (13). The δ18O of cellulose in cloud forest trees reflects theseasonal change in the δ18O of source water as determined by theamount of rainfall and the18O-enrichment of cloud water. Oninterannual time scales, departures from the mean annual cycleamplitude will result from anomalous rainfall and changes in tem-perature, relative humidity, and evapotranspiration (13, 15). AtMonteverde, these are further related to changes in the intensityof moisture advection over the continental divide into the Pacificslope forests and the amount of cloud cover. Interannual changesin the annual δ18O cycle maxima can therefore be interpreted asmoisture changes during the boreal winter dry season (approxi-mately January to April).We made high-resolution (200 μm increment) stable oxygenisotope ratio measurements along the radial growth axis oftwo Pouteria, mature canopy trees in the Sapotaceae family grow-ing in the cloud forest on the Pacific slope below the continentaldivide at Monteverde, using a unique online induction pyrolysissystem (16). The resulting δ18O data