Recent high-altitude observations (2013-2024) of extreme air temperatures and associated atmospheric circulation patterns in the tropical Andes

High-altitude mountains play a key role in modulating regional weather and climate. The tropical Andes in South America are characterized by strong climatic diversity and complex orography. In this region, identifying atmospheric circulation patterns (CPs) that control the meteorological extremes across different altitudinal and latitudinal gradients remains challenging. Using unique, quality-controlled hourly air temperature observations from four automatic weather stations located above 4700 m a.s.l. in the Peruvian Andes, this study links local extreme air temperature events to large-scale CPs during 2013-2024. CPs were identified using a k-means clustering algorithm applied to the standardized anomalies of the daily 200-hPa wind field from the ERA5 reanalysis over South America (10° N-30° S, 90°-30° W) for the 1980-2024 climatological period. Nine CPs were identified and classified into dry (D1-D4), wet (W1-W3), and transitional (T1-T2) circulation types, consistent with the regional seasonal cycle. Results show that warm nights (daily minimum air temperature exceeding the 90th percentile) are closely related to the occurrence of the transitional (dry-to-wet season) CP T1. This pattern is linked to warmer-than-normal conditions relative to the daily climatology, with a high frequency of warm nights observed from April to November. The 200-hPa circulation associated with T1 exhibits an upper-level ridge extending down to 500-hPa, resembling the Bolivian High. This circulation enhances easterly flow, favoring the advection of warm and moist air into the Andes and increasing nighttime and early-morning cloud cover. These conditions inhibit nocturnal radiative cooling and maintain elevated minimum air temperatures during a climatologically cold period in the Andes. During the 2023-2024 El Niño event, warm nights increased markedly compared to the previous years, while cold events became less frequent. This behavior appears to be primarily linked to an increased frequency of the T1 pattern, reaching up to 35%, particularly during July-October 2023 and April-July 2024. These findings provide a framework for future analyses of changes in this circulation regime under future climate scenarios and its role in modulating warm temperature extremes over the tropical glaciers.