Dataset: Detection and attribution of trends of meteorological extremes in Central America [2025]

Abstract

We present an analysis to determine whether historical trends in extreme precipitation and temperature indices, as well as in yearly averages of several climate variables show (or not) statistical indication that they contain a strong enough climate signal associated with anthropogenic climate change; and therefore, to determine if those trends can (or cannot) be explained solely by natural causes. We used three methodologies: a) climate model-based, b) a hybrid method that combines models and observations (1979-2019) and c) climate observations-based (1983-2016). For each methodology the signal of climate change, represented by the historical trends, was compared to the noise which is the distribution of trends from simulated climate constructions (using models or statistics) without human influence. Overall, the model-based method is less conservative and suggests possible detection of the human influence in most temperature extreme indices and in precipitation-related indices in the northern countries. The hybrid method showed detection in much fewer variables and in many cases consistently with the model-based method. The hybrid and the observation-based method showed similar noise variability as the model-based method. Notably, due to data availability limitations, the analysis excludes the most recent five years, during which substantial warming and increased intensity of extreme events has been observed worldwide.