The Case of the Golden Toad:Weather Patterns Lead to Decline

by Jennifer J. Neville
Undergraduate Student, Case Western Reserve University


The brightly colored golden toad of Costa Rica, Bufo periglenes, has drawn the attention of scientists, tourists, and local inhabitants since its discovery in 1966. The first scientific paper describing it was entitled, “An Extraordinary New Toad from Costa Rica,” and its discoverer, Jay Savage, descriptively wrote, “I must confess that my initial response when I saw them was one of disbelief and suspicion that someone had dipped the examples in enamel paint”(1966). Quickly news of the toad grew to make it seem more widespread than it actually was, as its image became a common feature on Costa Rican tourism posters promoting the biodiversity of the country (Phillips 1994). The golden toad was located near the continental divide in northern Costa Rica’s Monteverde Cloud Forest Preserve at elevations between 1480-1600 meters and was restricted to an area of approximately 10 km2.

These toads were known to be active above ground for only a few days when they emerged for explosive breeding at the end of the dry season. Most breeding sites were shallow pools of water formed by the branching roots of trees. The average clutch size was 228, and eggs took approximately two months to meta-morphose to toadlets (Jacobson and Vandenberg 1991). Mass numbers of golden toads were seen during the breeding season for 17 consecutive years (Pounds et al. 1997), and during April-July 1987, more than 1500 adult toads were observed. However, nearly all the deposited eggs in 1987 died before hatching because most pools dried, and a maximum of 29 tadpoles metamorphosed. During the 1988-1990 breeding seasons, only 11 toads were found. Since then, no golden toads have been seen (Crump et al. 1992).

Several hypotheses have been proposed to attempt to explain this drastic decline of Bufo periglenes. First, the decline may be due to the 1986-1987 El Niño Southern Oscillation, which led to warmer and drier conditions than normal during the breeding season (Phillips 1994; Pounds and Crump 1994). The decline could be part of a natural population fluctuation (Pechmann et al. 1991; Pechmann and Wilbur 1994), or golden toads may still be present in underground retreats waiting for better breeding conditions (Crump et al. 1992; Blaustein 1994; Pechmann and Wilbur 1994; Phillips 1994). Finally, an increase in UV-B radiation (Wake 1991) or an epidemic of parasites and fungus (Pechmann and Wilbur 1994; Phillips 1994; Young et al. 2001) could contribute to the decline, as could pollution or low pH levels (Phillips 1994; Pounds and Crump 1994).

The hypothesis dealing with abnormal weather patterns seems to be the best documented and the most widely accepted hypothesis concerning the decline of the golden toad among the scientific community; however, this paper will address the legitimacy and likelihood of each of the hypotheses.


Each of the various hypotheses will be analyzed by examining the results of several previously conducted studies. To examine the effect of weather patterns on the anurans of Monteverde, a 12-month amphibian moisture-temperature cycle consisting of four periods was defined: (1) late wet season (July-October); (2) transition into dry season (November-December); (3) dry season (January-April); and (4) post dry season (early wet season) recovery (May-June) (Pounds and Crump 1994). Twenty cycles were analyzed from 1970-1990 and the precipitation, stream discharge rate, and temperatures were plotted during these periods. To test the other hypotheses, previous experiments, current explanations, results from simple qualitative and quantitative tests, and the life history of the golden toad were examined.


Figure 1 clearly shows that during the 1986-1987 period, the total precipitation and the average stream flow in Monteverde were at record lows for both the Caribbean and Pacific slopes, while the temperatures in 1987 reached record highs. Furthermore, the 1986-1987 cycle was also the only one of the twenty analyzed with abnormally low rainfall during all four periods. Only during this year did Monteverde experience a weak late wet season, followed by an abrupt transition to a harsh dry season, followed by a delayed post season recovery. This climate disturbance was associated with the 1986-1987 El Niño Southern Oscillation, which was the strongest of the past century (Phillips 1994).

Pechmann et al. (1991) monitored the breeding population sizes of four amphibian species at one pristine site in South Carolina for 12 years. Each species was common during some years but uncommon or absent during others, but in the long run, populations of three of the four species stayed at a constant level, while one species increased. Pechmann and Wilbur (1994) further discussed examples when random fluctuations have produced apparent declines lasting several years that were really only parts of random cycles. A 1954 study by Bragg based on 18 seasons of observation of Bufo americanus reported a decline, but seven years later, the population had recovered to its original size. Similarly, a 1960 study by Bragg reported a decline of another species, Bufo cognatus, but thirty-two years later the area again supported large numbers of the toad (Pechmann and Wilber 1994). During the 1997 experiment of Pounds et al., the authors waited 5 years before publishing their data, and of the 25 species missing in 1990, 80% were still absent 5 years later (Pounds et al. 1997). Sightings of both golden toads and harlequin frogs at Monteverde decreased by about 99% in the same year (Pounds and Crump 1994).

Golden toads were last seen in mass numbers in 1987; which means that if they are underground, they have been there for a minimum of 14 years. Although the average lifespan of the golden toad is unknown, some species in same genus as the golden toad can live for 10-12 years (Blaustein 1994). Temperature and precipitation levels generally returned to levels they were prior to the 1986-1987 El Niño (Pounds et al. 1997), starting in 1988.

Baringa reports that although ozone levels have fallen in the past decade, the predicted increase in UV radiation has not been observed (1990). The results from the only study of UV-B radiation in Latin America suggest no effects on the two species of anurans tested (Young et al. 2001). Golden toads spend 95% of the year in underground retreats and the heavy cloud cover typical of Monteverde would be expected to attenuate the UV wavelengths (Crump et al. 1994).

Chytrid fungal disease has been identified at sites in Costa Rica (Young et al. 2001), and this fungus, along with a variety of other microparasites, has been known to be difficult to detect in amphibians (Pechmann and Wilbuer 1994). These microparasites have been found to cause high adult mortality among anurans maintained under suboptimal conditions, including crowding and high temperatures (Phillips 1994).

Cloud water collected in Monteverde in 1987 contained abnormally high amounts of nitrates and phosphates (Phillips 1994). There are no data available for acidity levels prior to 1987, but in 1987 and subsequent years neither cloud water, precipitation, nor breeding pool water was strongly acidic (Crump et al. 1992). Furthermore, while acidity is known to kill aquatic embryos and affect the distribution of terrestrial adults, it has not been linked to high adult mortality (Pounds and Crump 1994).


This study found no support for the UV-B radiation hypothesis since not only was there no observed increase in UV radiation, but no study in Latin America even suggests any effect of UV-B on anurans. Furthermore, even if a relationship or increase had been detected, the likelihood of any effect on golden toads would still be quite small, since they are rarely exposed to UV-B radiation. Secondly, support for the underground hypothesis was lacking as well. If the golden toads had been underground for 14 years and are estimated to live 10-12 years, there is only a very small chance any currently survive underground. Also, favorable weather conditions have occurred during numerous breeding seasons since 1987, which would have been expected to draw the toads out from hiding. For these reasons, I express doubt that either the UV-B or the underground hypothesis holds true.

The next factor considered was whether the golden toad decline is truly a decline or if it is part of a natural population fluctuation. Short-term trends demonstrated in Pechmann et al. (1991) made it difficult to discern long-term trends, which demonstrates the necessity of having long-term studies to distinguish between natural population fluctuations and true declines. Pechmann and Wilbur (1994) claimed that since Bufo periglenes had been listed as endangered due to its restricted geographical location, it was a likely candidate for natural extinction. They suggested that the decline of amphibians in isolated, apparently pristine areas is due to natural population fluctuation. However, the authors also noted that natural extinctions of species have been rare or non-existent in historical times. Pechmann’s 1994 data on the recovery of Bufo cognatus and Rana pipiens was based solely on personal communication with a single source. Furthermore, his source was a supply house, which may not be reliable since with all the media attention regarding amphibian declines, supply houses may have a vested interest in saying conditions are satisfactory. Therefore, there is reason to believe that these data may not be credible. Pechmann’s 1991 and 1994 papers also only addressed his four studies, which showed no overall population declines, and did not mention four additional studies by other herpetologists in which the populations were found to be truly declining (Blaustein 1994). His study was undertaken at only one site and was not at the high altitudes of 500m or more, which are characteristic of amphibian declines (Young et al. 2001). Finally, sightings of both golden toads and harlequin frogs at Monteverde decreased by about 99% in the same year, and because annual adult survivorship in bufonids is generally much greater than 1%, the abrupt nature of the declines suggests high adult mortality, rather than just a lack of successful breeding and recruitment (Pounds and Crump 1994). While Pechann et al. make an important point regarding the importance of long-term census data, there are not sufficient data to indicate that the decline of the golden toad is purely a natural fluctuation.

Data seem to support the El Niño hypothesis. Precipitation levels clearly have a direct affect on breeding, since toads cannot oviposit until after breeding pools retain water; and pools must remain filled to prevent desiccation of the eggs. Conversely, if precipitation levels are too high, eggs or tadpoles may be swept out of the pools and onto the forest floor. Subsequent days without rain would dry the substrate and leave the eggs or tadpoles stranded and subject to desiccation (Crump et al. 1992). Low precipitation levels may also affect mature toads since most amphibians readily lose water across their moist, permeable skin. At high temperatures, a dehydrating anuran may be faced with conflicting physiological demands. Increasing mucus secretions keep body temperatures below dangerous temperatures through evaporative cooling, but at the same time accelerates water loss. During the dry season, an anuran’s chances of avoiding dehydration may depend on more than just that season’s precipitation and temperature patterns. Other important factors may include how well the preceding wet season recharged groundwater, how quickly the dry season began, and how promptly the early wet season brings recovery (Pounds and Crump 1994). The data from Pounds and Crump (1994) and the theories behind them clearly support the 1986-1987 El Niño hypothesis.

Although the weather patterns of 1986-1987 seem to be largely responsible for the golden toad decline, several uncertainties remain, especially with regard to the effects of parasites, fungi, pH, and pollution levels. It seems unlikely that parasites and fungi would suddenly become highly lethal unless they were interacting with some other factor. Since parasites and bacteria have been found to cause high adult mortality among anurans maintained under sub-optimal conditions, including crowding and high temperatures, then perhaps parasites worked in conjunction with the weather conditions of 1986-1987 to cause the decline. Since the weather was warmer and drier during that year, then the toads may have been forced to be in closer proximity to each other in the smaller breeding pools, which could have facilitated an epidemic due to a decrease in inter-host distance. Secondly, although pH levels did not appear to be of abnormal acidity during 1987, there is no record of the pH values during years of successful breeding, so there is no way to know if the levels have changed and what pH level is optimal for the golden toad. Finally, the abnormally high levels of nitrates and phosphates – commonly found in fertilizers – found in cloud water in 1987 show that it is possible for contaminants to blow from other sources into Monteverde. Warm, dry conditions increase the rates at which agrochemicals volatilize into the atmosphere from exposed surfaces. During dry periods, there would be less precipitation to dilute pesticides and the chemicals could become more concentrated in the soil. A rehydrating anuran may absorb a volume of water equal to 60-70% of its normal body water content (Pounds and Crump 1994), so a partially dehydrated amphibian may have low volumes of body fluids to dilute toxic compounds. More research on the relationships between parasites, fungi, pH, and pollution and anurans is needed to determine if these factors work in conjunction with the weather patterns of 1986-1987 to cause the decline of the golden toad.

Literature Cited

Baringa, M. 1990. Where have all the froggies gone? Science 247:1033-1034.

Blaustein, A. B. 1994. Chicken little or nero’s fiddle? A perspective on declining amphibian populations. Herpetologica 50(1):85-97.

Crump, M. L., F. R. Hensley, and K. L. Clark. 1992. Apparent decline of the golden toad: Underground or extinct? Copeia 1992:413-420.

Jacobson, S. K. and J.J. Vandenberg. 1991. Reproductive ecology of the endangered golden toad (Bufo periglenes). Journal of Herpetology 25(3):321-327.

Pechmann, J. H. K., D. E. Scott, R. D. Semlitsch, J.P. Caldwell, L. J. Vitt, and J. W. Gibbons. 1991. Declining amphibian populations: The problem of separating human impacts from natural fluctuations. Science 253:892-895.

Pechmann, J. H. K., and H. M. Wilbur. 1994. Putting declining amphibian populations into perspective: Natural fluctuations and human impacts. Herpetologica 50:65-84.

Phillips, K. 1994. Tracking the vanishing frogs. New York: Penguin. 244 p.

Pounds, J. A. and M.L Crump. 1994. Amphibian declines and climate disturbance: The case of the golden toad and the harlequin frog. Conservation Biology 8(1):72-85.

Pounds, J. A., M. P. L. Fogden, J. M. Savage, and G. C. Gorman. 1997. Tests of null models for amphibian declines on a tropical mountain. Conservation Biology 111(6):1307-1322.

Young, B.E., K. R. Lips, J. K. Reaser, R. Ibáñez, A. W. Salas, J. R. Cedeño, L. A. Coloma, S. Ron, E. L. Marca, J. R. Meyer, A. Muñoz, F. Bolaños, G. Chaves, and D. Romo. 2001. Population declines and priorities for amphibian conservation in Latin America. Conservation Biology 15(5):1213-1223.

In cooperation with...

CloverLeaf Animal Hospital

Support NOAH