Using Large Ensembles to Examine Historical and Projected Changes in Record‐Breaking Summertime Temperatures Over the Contiguous United States

McHugh, Colleen. E.; Delworth, Thomas L.; Cooke, William; Jia, Liwei

doi:10.1029/2023ef003954

Cited by 3 publications

(1 citation statement)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We also point out that there are only negligible differences across future forcing scenarios in the ensemble means of JJA TAVG until between 2030 and 2040, but by after 2080, there is very little overlap in their ensemble spreads due to the greater effects of projection scenario uncertainty. Additional comparisons between observations and SPEAR_MED for summertime temperatures across the CONUS can be found in McHugh et al (2023) and Eischeid et al (2023).…”

Section: Changes In United States Summertime Temperaturesmentioning

confidence: 99%

Changes in United States Summer Temperatures Revealed by Explainable Neural Networks

Labe,

Johnson,

Delworth

2024

Earth's Future

Self Cite

View full text Add to dashboard Cite

To better understand the regional changes in summertime temperatures across the conterminous United States (CONUS), we adopt a recently developed machine learning framework that can be used to reveal the timing of emergence of forced climate signals from the noise of internal climate variability. Specifically, we train an artificial neural network (ANN) on seasonally averaged temperatures across the CONUS and then task the ANN to output the year associated with an individual map. In order to correctly identify the year, the ANN must therefore learn time‐evolving patterns of climate change amidst the noise of internal climate variability. The ANNs are first trained and tested on data from large ensembles and then evaluated using observations from a station‐based data set. To understand how the ANN is making its predictions, we leverage a collection of ad hoc feature attribution methods from explainable artificial intelligence (XAI). We find that anthropogenic signals in seasonal mean minimum temperature have emerged by the early 2000s for the CONUS, which occurred earliest in the Eastern United States. While our observational timing of emergence estimates are not as sensitive to the spatial resolution of the training data, we find a notable improvement in ANN skill using a higher resolution climate model, especially for its early twentieth century predictions. Composites of XAI maps reveal that this improvement is linked to temperatures around higher topography. We find that increases in spatial resolution of the ANN training data may yield benefits for machine learning applications in climate science.

show abstract

Section: Changes In United States Summertime Temperaturesmentioning

confidence: 99%

Changes in United States Summer Temperatures Revealed by Explainable Neural Networks

Labe,

Johnson,

Delworth

2024

Earth's Future

Self Cite

View full text Add to dashboard Cite

show abstract

Escalating threat of human-perceived heatwaves in Brazil

Ballarin,

Oliveira,

Uchôa

et al. 2024

Environ. Res. Commun.

View full text Add to dashboard Cite

Heatwaves pose significant threats to socioeconomic and environmental systems, with their intensity and frequency expected to increase due to climate change. Despite their critical impacts, future heatwaves in Brazil remain underexplored, especially from a human-perceived perspective, which is crucial for assessing potential public health impacts. Here, we propose a method to assess heatwaves using the humidex (H) – a climate index that combines temperature and relative humidity to indicate human-perceived heat - alongside traditional temperature-based measures. Using bias-corrected simulations from 10 CMIP6 models under SSP2-4.5 and SSP5-8.5 scenarios, we quantified projected changes in heatwaves across Brazil. The results indicate that heatwaves will become more severe and prolonged, with greater changes under the SSP5-8.5 scenario by the end of the century, particularly in the North, Northeast, and Central regions. The magnitude of human-perceived heatwaves is expected to rise faster than temperature-based ones, underscoring the need for public health-focused assessments. CMIP6 models strongly agree on increased future heatwaves, potentially tripling population exposure in most Brazilian states, with the Southeast experiencing greater changes due to its larger population. These events are expected not only to affect more people but also to be more severe, exceeding over 60 days per year of serious danger (H > 45°C) by the end of the century under SSP5-8.5. Record-shattering events in the historical period are projected to become the norm by mid-century, highlighting the accelerating nature of these extreme events. Our findings emphasize the importance of considering human-perceived heat in climate impact studies and public health planning to mitigate potential impacts.

show abstract

A rapid response process for evaluating causes of extreme temperature events in the United States: The 2023 Texas/Louisiana heat wave as a prototype

Schreck III,

Easterling,

Barsugli

et al. 2024

Environ. Res.: Climate

View full text Add to dashboard Cite

As climate attribution studies have become more common, routine processes are now being established for attribution analysis following extreme events. This study describes the prototype process being developed through a collaboration across NOAA, including monitoring tools as well as observational and model-based analysis of causal factors. The prolonged period of extreme heat in summer 2023 over Texas, Louisiana and adjacent areas provided a proving ground for this emerging capability. This event posed unique challenges to the initial process. The extreme heat lasted for most of the summer while most heat wave metrics have been designed for 3–7 day events. The eastern portion of the affected area also occurred within the so-called summer-time daytime warming hole where the warming trend in maximum temperatures has been mitigated wholly or in part by increased precipitation. The extreme temperature coincided with a strong—but not record—precipitation deficit over the region. Both observations and climate model simulations illustrate that the temperatures for a given precipitation deficit have warmed in recent decades. In other words, meteorological droughts today are hotter than their historical analogs providing a stronger attribution to anthropogenic forcing than for temperature alone. These findings were summarized in a prototype plain language report that was distributed to key stakeholders. Based on their feedback, the monitoring and assessment tools will continue to be refined, and the project is exploring other climate model large ensembles to increase the robustness of attribution for future events.

show abstract

Using Large Ensembles to Examine Historical and Projected Changes in Record‐Breaking Summertime Temperatures Over the Contiguous United States

Cited by 3 publications

References 53 publications

Changes in United States Summer Temperatures Revealed by Explainable Neural Networks

Changes in United States Summer Temperatures Revealed by Explainable Neural Networks

Escalating threat of human-perceived heatwaves in Brazil

A rapid response process for evaluating causes of extreme temperature events in the United States: The 2023 Texas/Louisiana heat wave as a prototype

Contact Info

Product

Resources

About