Migration from a Subtropical Climate to a Tropical Climate in Coastal Areas

INTRODUCTION

The expanding tropics is one of the main consequences of climate change in the atmospheric circulation (Lu et al., 2008; Lucas et al., 2014; Solomon et al., 2016). The tropics are defined by the Hadley cells, whose widening has been demonstrated in several studies (Lucas et al., 2014). Over the last decades, 1 to 2 degrees of expansion have been documented for both hemispheres (Lucas et al., 2014). El Niño-Southern Oscillation (ENSO) also modifies Hadley and Walker cells, generating significant fluctuations in tropical and global climate. Comparing the recent decades of expansion with this climate pattern is required.

Tropical and subtropical climate regions and their surrounding areas are subject to a wide range of natural variability and the effects of expanding tropics (Seidel et al., 2008). Many regions across the world (i.e., Australia, the Mediterranean region, North America, and South America) have been exposed to more frequent droughts related to changes in climate regions (tropical to subtropical). Coastal systems can also be affected by the displacement of these regions (Duarte et al., 2020). In fact, tropical cyclones displacing poleward is agreed upon by many recent studies (Studholme et al., 2022). Identifying which regions are shifting their climate is critical for understanding what areas are eexceeding the bounds of natural variability and assessing the implications on coastal regions.

This study identifies long-term trend and distribution of tropical regions in coastal areas, comparing with their natural distribution under El Niño-Southern Oscillation. We focus on the latitude position of the tropical cyclones and Sea Surface Temperature (SST) thermocline from 28-24ºC.

METHODS

For the purpose of evaluating migration of tropical climate in coastal areas, we define the tropical regions as the upper (Northern Hemisphere) and lower (Southern Hemisphere) latitude that delimited the tropical cyclones distribution and thermocline of SST thermocline between 24 and 28ºC. We analysed the spatial distribution of tropical cyclones under two climatology periods (1980-2000 and 2000-2020). We generated synthetic tracks using STORM model (Bloemendaal et al., 2020) for both climatologies, as well as for the phases of ENSO, El Niño and La Niña during the same period (1980 to 2020). To generate the synthetic tracks, we used SST and Sea Level Pressure of ERA5 (Hersbach et al., 2020) and historical tracks of tropical cyclones from International Best Track Archive for Climate Stewardship version 4 (IBTrACS v4) (Knapp et al., 2018; Knapp et al., 2010). The same dataset of SST used for synthetic tracks was used for the thermocline analysis. The time series of latitudinal position was obtained at every longitudinal 0.1º for tropical cyclones and 0.25º for SST.

Once the latitudinal limits of the climate variables were identifying, long-term trends of the latitudinal boundary were identified, only those trends statistically significant at 95 % confidence level by the Mandal Kendall approach were considered. The empirical distribution of their spatial position was also calculated.

RESULTS

The results reveal spatial changes in the tropical climate regions, considering long-term trends and natural variability induced by ENSO. We acknowledge that the use of unique reanalysis model with resolutions of 0.1 and 0.25º introduces a source of uncertainty. However, this study represents a first approach to identify potential areas globally that will be affected by expanding tropics in coastal regions. More efforts are needed to characterize future climate in these areas.