Diagnosing Non‐Gaussian Temperature Distribution Tails Using Back‐Trajectory Analysis

Abstract

Anthropogenic warming has intensified drought conditions (Lehner et al., 2006), reduced frost day occurrence (Easterling, 2002), and altered ecosystems (Walther et al., 2002). Although continued and projected warming broadly increases warm and decreases cold extremes (Kharin et al., 2013), these changes will not occur at a uniform rate globally. For example, some regions have experienced changes in seasonal temperature extremes in excess of changes in the mean, including winter warm extremes in Europe (Gross et al., 2019). Previous work showed this region exhibits a shorter-than-Gaussian temperature distribution warm tail, meaning it would experience a more-rapid-than-Gaussian increase in exceedance frequency of a fixed extreme temperature threshold under a uniform warm shift, consistent with climate model projections (Loikith et al., 2018).The prevalence of non-Gaussian near-surface temperature distributions is well-documented (Cavanaugh & Shen, 2014;Perron & Sura, 2013). Recent work has investigated how large-scale dynamics within a background meridional temperature gradient produces large, spatially coherent regions of asymmetrical temperature distributions. For example, atmospheric stirring generated by Rossby wave propagation plays a role in mid-latitude distribution asymmetry (Garfinkel & Harnik, 2017;Linz et al., 2018). Tamarin-Brodsky et al. (2019) also highlighted the role of temperature advection in governing distribution skewness using a Lagrangian feature-tracking technique, with implications for changes in extreme temperatures under global warming (Tamarin-Brodsky et al., 2020).Identifying the fundamental dynamics governing broad temperature distribution symmetry is valuable for determining how extreme temperatures may change in a warming climate. However, non-Gaussian

More Frequent and Persistent Heatwaves Due To Increased Temperature Skewness Projected by a High‐Resolution Earth System Model

“…The result is more negatively skewed temperature (Figure S12 in Supporting Information ; from −0.27 to −0.42), reducing heatwave days from 1.93 historically to 1.59 in the future. In addition to the effect of soil moisture‐atmosphere interactions, dynamical impact such as the effect of storm tracks and meridional advection may play an important role in shaping the spatial structure of temperature skewness (Catalano et al., 2021; Tamarin‐Brodsky et al., 2019), which deserves further investigation.…”

Short Warm Distribution Tails Accelerate the Increase of Humid‐Heat Extremes Under Global Warming

“…A very high daily mean T W can also be suggestive of high nighttime T W , prolonging exposure to heat stress (Kravchenko et al., 2013). It is possible that in some cases, the prevailing meteorology that causes the short tails may also prevent T W from becoming too high, such as is seen with temperature distributions (Catalano et al., 2021).…”

“…Another complicating factor in the increase of T W extremes is the shape of the T W probability distribution. The non‐Gaussian nature of temperature climatology is well‐established (Catalano et al., 2021; Garfinkel & Harnik, 2017; Linz et al., 2018; Perron & Sura, 2013; Simolo et al., 2010, 2011; B. Zhang et al., 2022), and some physical explanations have been offered, generally related to underlying horizontal temperature gradients (Catalano et al., 2021; Garfinkel & Harnik, 2017; Linz et al., 2018; Perron & Sura, 2013), but non‐Gaussianity in T W distributions has not been explored to our knowledge. Tail shape relative to a Gaussian strongly influences the rate of changes in exceedances over a fixed extreme temperature threshold, and a shorter‐than‐Gaussian warm tail can lead to a greater increase in extreme warm exceedances under global warming than if the tail were Gaussian or longer (Loikith et al., 2018; Loikith and Neelin, 2015, 2019; Ruff & Neelin, 2012).…”

“…Another complicating factor in the increase of T W extremes is the shape of the T W probability distribution. The non-Gaussian nature of temperature climatology is well-established (Catalano et al, 2021;Garfinkel & Harnik, 2017;Linz et al, 2018;Perron & Sura, 2013;Simolo et al, 2010Simolo et al, , 2011B. Zhang et al, 2022), and some physical explanations have been offered, generally related to underlying horizontal temperature gradients (Catalano et al, 2021;Garfinkel & Harnik, 2017;Linz et al, 2018;Perron & Sura, 2013), but non-Gaussianity in T W distributions has not been explored to our knowledge.…”