Weather, Storms, and Monarchs

On 13 February 2010 we downloaded the data from our electronic WeatherHawk weather station located in the Sierra Chincua, in the Monarch Butterfly Biosphere Reserve. This instrument is at nearly the same elevation as the Chincua butterfly colonies, and 2 – 4 km east of where the colonies form each year. We provide here a preliminary summary of the severe storm that caused extensive landslides, destruction of homes and tragic loss of human lives in Rosario, Angangueo, and other parts of eastern Michoacán.

Anecdotal accounts on the internet have suggested variable butterfly mortality, from 0% to 80% in different colonies. We have not made direct measures of butterfly mortality in this storm, but our weather records, combined with empirical data from Anderson and Brower (1996), Brower et al. (2004, 2009) and Fink et al. (in prep.) give insight into the interplay between local microclimate and butterfly survival. It is clear that butterflies suffered less than the local citizens …but only by a hair’s breadth.

As is typical of the region’s dry season weather pattern, from 17 to 30 January 2010 no rain fell in the area. Light rain began at about 11 pm on 31 January and fell steadily from 6 am to 11 pm on 1 February, accumulating 3.9 cm (1.5 inches)*. Rain began again at 7 pm on 2 February and was continuous and heavy until midnight on 4-5 February. The total precipitation on 4-5 February was 32.0 cm (12.6 inches); the entire storm produced 36.0 cm (14.2 inches).

We know that this heavy storm must have soaked the butterflies. We also know that in January 2002, a combination of heavy precipitation followed by an early morning temperature plunge resulted in 80% mortality of monarch butterflies (Brower et al. 2004). Butterfly survival and mortality depend not on low temperature and not on precipitation, but on the interaction of the two, because wet butterflies have significantly less freeze resistance than do dry butterflies.

During the recent storm, from 31 January through midnight on 4-5 February, the hourly temperature at our weather station ranged between 1.9 and 8.4 °C; at these temperatures no butterflies suffer freeze mortality. Because the lowest air temperature occurs close to sunrise, the morning temperatures on 5 February were crucial in determining the butterflies’ fate. On this morning, the minimum temperature at our weather station reached -3.02°C at 7:42 am. This was fortunate for the butterflies for two reasons. First, our experimental data tell us that wet, wintering monarchs have sufficient freeze resistance for few if any to be killed at this temperature. For wet butterflies, the crucial range of temperatures is -4 to -6°C: in mid-winter, the majority of butterflies can survive many hours of exposure to -4°C, but almost all will be killed by just a few minutes of exposure to -6°C (Fink et al. in prep). Second, because the oyamel fir canopy buffers the forest temperature, most butterflies in the colonies would have experienced a minimum temperature that we estimate to have been about 4°C warmer inside the forest (Brower et al., 2009).

During the day on 5 February the skies were clear. From late morning through mid-afternoon the temperature was 9 to 12 °C and the relative humidity was very low, between 9% and 11%. These conditions would give the butterflies the opportunity to dry their bodies and wings. This is fortunate, because the next night — 5-6 February — was clear and the temperature dropped to -5.96°C at the weather station. Depending on their specific locations within the forest, if they had not had the opportunity to dry, some monarchs would likely have frozen.

The weather data, therefore, tell us that this major storm very likely did not cause major butterfly mortality. If the skies had cleared earlier on the night of 4-5 February and a morning temperature plunge had occurred while the butterflies were still wet, the outcome might have been dramatically different.

It is well to remember that any thinning of the oyamel forest results in less temperature buffering, lowering temperatures in the forest and therefore increasing the probability of butterfly mortality during the clearing that follows storms. Vigilance is needed to prevent tree removal because the butterflies are overwintering on a microclimatic knife-edge.

Additional news of concern: the Sierra Chincua was reported to have had 15 cm of snow which was melted by rain on 18 February 2010. If conditions similar to those described above prevail, damage to butterflies will be minimal. However, where colonies have formed in and adjacent to thinned forests, the chances of butterfly mortality are much higher.

As of 20 February we are still receiving contradictory anecdotal statements on the extent of the mortality caused in several colonies by all these storms. We will keep you informed.

References:
Anderson, J.B. & Brower, L.P. (1996) Freeze-protection of overwintering monarch butterflies in Mexico: critical role of the forest as a blanket and an umbrella. Ecological Entomology, 21, 107-116.

Fink, L.S., Brower, L.P., Helton, T., and C. Kisiel. In prep. The ecology of cold-hardiness in overwintering monarch butterflies.

Brower, L.P., Kust, D.R., Rendon-Salinas, E., Serrano, E.G., Kust, K.R., Miller, J., Fernandez del Rey, C., & Pape, K. (2004). Catastrophic winter storm mortality of monarch butterflies in Mexico during January 2002. In The Monarch Butterfly. Biology and Conservation (eds K.S. Oberhauser & M.J. Solensky), pp. 151-166. Cornell University Press, Ithaca.

Brower, L.P., Williams, E.H., Slayback, D.A., Fink, L.S., Ramirez, M.I., Zubieta, R.R., Limon Garcia, M.I., Gier, P., Lear, J.A., & Van Hook, T. (2009) Oyamel fir tree trunks provide thermal advantages for overwintering monarch butterflies in Mexico. Insect Conservation and Diversity, 2, 163-175.

*This account does not mention the hailstorm that occurred from 5AM-8AM in Angangueo on 1 February 2010. Estela Romero, our colleague who lives in Angangueo, reported to us that 4-6 cm of hail accumulated during this period, an event that was also recorded on video by others. During this same time period 12 mm of precipitation was recorded at the Chincua Weather Station. We are not certain at this time whether it also hailed at Chincua, an event that might have knocked many butterflies from the trees. Nor is it clear whether hail would have been effectively recorded by our instrument. What we can say is that Angangueo received more than 12 mm of precipitation from 5-8 AM on 1 February. The amount of water in 4-6 cm of hail is uncertain but could have been as much as 2.8-4.3 cm.