Analysis | Towards a better forecast of heat waves in the Chilean south-central zone

334

Never and nowhere do heat waves (HW) go unnoticed. That is particularly true during the summer in the south-central zone of Chile, which extends approximately from the Valparaíso Region to the Los Lagos Region.

Over the past few years, we have witnessed exceptional HWs. One of them resulted in the large forest fires in January and February 2017 (Bowman et al., 2018), which caused a very high social, ecological and material cost. In particular, the fires devastated more than 570,000 hectares, an area that exceeded the average of previous decades by more than ten times (González et al., 2019). In addition, several maximum temperature records were broken in central Chile in that year, including 42.2 ºC in Los Angeles (DMC, 2017).

Events of very intense magnitude also occurred in 2019. For example, the longest HW recorded in Curicó lasted 17 days since the end of January (Blunden and Arndt, 2020). Then another HW impacted from Los Ríos to Magallanes at the beginning of February, setting the record of 38.5 ºC in Valdivia (DMC, 2019) and promoting the development of the Cochrane fire, which affected more than 15,000 hectares of vegetation (González et al., 2019). Thus, the relevance of heat waves studies is evident, both from the perspective of the impacts they generate and from a meteorological approach focused on their causes. Therefore, thanks to the support of a research project [1], we have been committed to understanding and describing various aspects related to its occurrence, and this analysis synthesizes the findings of Jacques-Coper et al. (2021).

A heat wave occurs when the temperature is persistently warmer than usual for at least three days. To study HWs, we use temperature records from different meteorological stations and historical reconstructions of the atmospheric circulation generated by computer models. First, we wanted to know if the atmosphere over Chile exhibits a particular meteorological configuration during the occurrence of an HW. We found that these events typically occur up to four times a summer and last, on average, five days. In addition, we identified that they are the consequence of the passage of a migratory anticyclone (a high-pressure system) through southern Chile, which makes the sky over central Chile tend to be clear, which increases solar radiation and warms the lower atmosphere. In addition, a relatively dry and warm east wind called Puelche is promoted in pre-Andean and Andean sectors. Together, these factors can cause a sharp and sustained rise in temperature, as is typical during HWs.

Secondly, we set out to study whether this meteorological configuration could be caused, in turn, by some phenomenon that occurs before and in a place far from where the HW finally manifests itself. We call these types of climatic relationships teleconnections. Thus, we discovered that the frequency and intensity of migratory cyclones and anticyclones in southern Chile might be related to climatic phenomena in the tropics, particularly in the Indian Ocean, the Maritime Continent (north of Australia, where Indonesia is located), and the Pacific Ocean. There are alternately and prolonged periods of rain or clear skies over vast areas in these regions. When there is persistent precipitation, we know that large air masses rise and form clouds, which we call “deep convection.” This deep convection moves from west to east across the tropics regularly when it occurs. That is a phenomenon called the “Madden-Julian Oscillation” (MJO). The tropics are divided into eight longitudinal sectors called phases. A phase is “active” if it exhibits noticeable convection. In particular, if this convection is present over Indonesia and northeast Australia, we say that the MJO is in phase 4 active. Over the days, the active phases of the MJO advance to phase 5 (convection still over the Maritime Continent), then to 6, to 7 (convection over the Western Pacific), and, finally, to 8 (convection over the central Pacific and Africa). However, active convection in phase 4 matters to us the most since waves can be triggered from there in the atmosphere and propagate to South America within one to two weeks, reaching the south-central zone of Chile (when the convection is already in phase 6), favoring the intensification of an anticyclone over the south of the country. That is precisely the atmospheric configuration associated with high temperatures. In this way, our research showed that the summer temperature in central Chile is usually higher in the active MJO phases 6 to 8.

In this way, our work revealed that when there is intense or persistent convection in phase 4 of the MJO, it is very likely that a series of atmospheric phenomena will be triggered that ends in a heat wave in the country. Therefore, the active phase 4 of the MJO is a precursor of heat waves in central Chile. In other words, after approximately one or two weeks of observing this precursor, the probability of HW occurrence increases in the region so that certain heat waves could be predicted two weeks in advance.

However, that is not all. In addition to the tropics, we also focus on the extra-tropics (mid-latitudes) to look for HW precursors. Thanks to our research, we discovered another atmospheric signal in the Indian Ocean, southeast of South Africa, which usually precedes some HWs in central Chile by approximately two weeks. Therefore, this signal was identified as another precursor. In order to monitor, that is, observe and quantify this precursor, we define an index that we call SETI (the acronym for “standardized extra-tropical index” in English). This index can be used to study the past and the present and future. Today, every day, we calculate a SETI value from computational data. When SETI is very high, the extra-tropical precursor is activated. These two precursors, both the MJO and SETI (and especially both together), serve us as information to better forecast certain heat waves in Central Chile up to two weeks in advance.

This type of research aims to understand better the atmosphere’s behavior and how extreme meteorological events are triggered. In particular, according to ongoing research, heat waves can affect the development of forest fires, the thawing of mountain basins (which affects flows), and events of algal blooms. For this reason, advancing in the identification of precursors of heat waves is essential for the prevention and mitigation of their negative socio-environmental impacts. Specifically, anticipating their occurrence can contribute to extreme precautions regarding other controllable impact factors, such as in the case of forest fires. Additionally, this knowledge can serve to organize the necessary resources in a more timely and efficient way to mitigate the eventual consequences of HWs. For now, to refine our results, we will be closely monitoring the 2021/2022 summer season from meteorology.

Note

[1]This study was funded by the ANID / FONDECYT / 11170486 project “Heat waves in Central Chile and their predictability: our possible tropical connection”.

References

Blunden, J. and Arndt, D.S. (2020) State of the climate in 2019. Bulletin of the American Meteorological Society, 101, Si-S429.

Bowman DMJS, Moreira-Muñoz A, Kolden CA, et al (2019) Human–environmental drivers and impacts of the globally extreme 2017 Chilean fires. Ambio 48:350–362. doi: 10.1007/s13280-018-1084-1

DMC (2017) Informe Climático Especial: Enero 2017: un mes de récords. Santiago de Chile, Chile; https://www.yumpu.com/es/document/read/56849766/enero-2017-un-mes-de-records

DMC (2019) El calor sigue batiendo récord. Disponible en: https://blog.meteochile.gob.cl/2019/02/04/el-calor-sigue-batiendo-records/

González ME, Sapiains R, Gómez-González S, et al (2020) Incendios forestales en Chile: causas, impactos y resiliencia. Centro de Ciencia del Clima y la Resiliencia (CR)2, U. de Chile, U. Concepción y U. Austral Chile 3–80

Jacques-Coper, M, Veloso-Aguila, D, Segura, C, Valencia, A. Intraseasonal teleconnections leading to heat waves in central Chile. Int J Climatol. 2021; 41: 4712– 4731. https://doi.org/10.1002/joc.7096