Spatio-temporal quantile regression analysis revealing more nuanced patterns of climate change: A study of long-term daily temperature in Australia

Many studies have considered temperature trends at the global scale, but the literature is commonly associated with an overall increase in mean temperature in a defined past time period and hence lacking in in-depth analysis of the latent trends. For example, in addition to heterogeneity in mean and median values, daily temperature data often exhibit quasi-periodic heterogeneity in variance, which has largely been overlooked in climate research. To this end, we propose a joint model of quantile regression and variability. By accounting appropriately for the heterogeneity in these types of data, our analysis using Australian data reveals that daily maximum temperature is warming by similar to 0.21 degrees C per decade and daily minimum temperature by similar to 0.13 degrees C per decade. More interestingly, our modeling also shows nuanced patterns of change over space and time depending on location, season, and the percentiles of the temperature series.

Automated summary

Many studies have examined global temperature trends, but there is a lack of in-depth analysis of latent trends. Using a joint model, this study analyzes Australian data and finds that daily maximum temperature is warming at a rate of about 0.21 degrees C per decade, while daily minimum temperature is warming at a rate of about 0.13 degrees C per decade. The study also reveals nuanced patterns of change based on location, season, and the percentiles of the temperature series.