Michael Kelleher

Dr. Michael Kelleher


Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
31 December 2019 in Climate Dynamics
Verified
Journal Article
Open access
PDF file
PMPenelope Maher
Penelope Maher
MKMichael Kelleher
Michael Kelleher
PSPhilip G. Sansom
Philip G. Sansom
JMJohn Methven
John Methven
The tropics are expanding poleward at about $$0.5{^\circ }$$0.5∘ per decade in observations. This poleward expansion of the circulation is consistently reported using Hadley cell edge metrics and lower-atmospheric tropical edge metrics. However, some upper-atmospheric tropical metrics report smaller trends that are often not significant. One such upper-atmospheric metric is the subtropical jet latitude, which has smaller trends compared to the Hadley cell edge. In this study we investigate the robustness of the weak trends in the subtropical jet position by introducing a new method for locating the subtropical jet, and examining the trends and variability of the subtropical jet latitude. We introduce the tropopause gradient method based on the peak gradient in potential temperature along the dynamic tropopause. Using this method we find the trends in the subtropical jet latitude are indeed much smaller than $$0.5{^\circ }$$0.5∘ per decade, consistent with previous studies. We also find that natural variability within the subtropical jet latitude would not prevent trends from being detected if they were similar to the Hadley cell edge, as trends greater than 0.24 $${^\circ }$$∘ per decade could reliably be detected using monthly data or 0.09 $${^\circ }$$∘ per decade using daily data. Despite the poleward expansion of the tropics, there is no robust evidence to suggest the subtropical jet is shifting poleward in either hemisphere. Neither the current diagnostic methods nor natural variability can account for the small subtropical jet trends. The most likely explanation, which requires further investigation, is that the subtropical jet position is not tied dynamically to the Hadley cell edge.

27 May 2015 in Journal of the Atmospheric Sciences
Verified
Journal Article
Closed access
SCStephen J. Colucci
Stephen J. Colucci
MKMichael Kelleher
Michael Kelleher
Tropospheric blocking events over the Northern Hemisphere during 1980–2012 were composited and contrasted according to whether they coincided in time with a sudden stratospheric warming (SSW). Those that coincided with an SSW were associated with significantly greater poleward eddy heat fluxes in the upper troposphere near the block onset time than were those blocking events not coinciding with an SSW. Furthermore, the heat fluxes in the SSW–blocking composites were concentrated inside the stratospheric polar vortex (i.e., within an area enclosed by the outer edge of an objectively defined polar vortex). Thermally forced stratospheric geopotential height rises were also significantly larger near block onset time inside the stratospheric polar vortex in the SSW–blocking composites than in the non-SSW–blocking cases. Although all the SSW events during the investigated period coincided with tropospheric blocking, the reverse was not true since there were many more blocking events than SSWs. Therefore, blocking itself was not a sufficient condition for an SSW. It is conjectured that blocking may not be a necessary condition for an SSW if persistently anomalous tropospheric heat fluxes and thermally forced, stratospheric geopotential height rises, concentrated inside the stratospheric vortex, occur in the absence of blocking.

7 December 2017 in Advances in Atmospheric Sciences
Verified
Journal Article
Open access
PDF file
MKMichael Kelleher
Michael Kelleher
JSJames Screen
James Screen
This study examines pre-industrial control simulations from CMIP5 climate models in an effort to better understand the complex relationships between Arctic sea ice and the stratosphere, and between Arctic sea ice and cold winter temperatures over Eurasia. We present normalized regressions of Arctic sea-ice area against several atmospheric variables at extended lead and lag times. Statistically significant regressions are found at leads and lags, suggesting both atmospheric precursors of, and responses to, low sea ice; but generally, the regressions are stronger when the atmosphere leads sea ice, including a weaker polar stratospheric vortex indicated by positive polar cap height anomalies. Significant positive midlatitude eddy heat flux anomalies are also found to precede low sea ice. We argue that low sea ice and raised polar cap height are both a response to this enhanced midlatitude eddy heat flux. The so-called “warm Arctic, cold continents” anomaly pattern is present one to two months before low sea ice, but is absent in the months following low sea ice, suggesting that the Eurasian cooling and low sea ice are driven by similar processes. Lastly, our results suggest a dependence on the geographic region of low sea ice, with low Barents–Kara Sea ice correlated with a weakened polar stratospheric vortex, whilst low Sea of Okhotsk ice is correlated with a strengthened polar vortex. Overall, the results support a notion that the sea ice, polar stratospheric vortex and Eurasian surface temperatures collectively respond to large-scale changes in tropospheric circulation. 本文利用第五次耦合模式比较计划(CMIP5)工业革命前对照实验的模式模拟结果, 考察了北极海冰与对流层大气,及其与欧亚冷冬的联系. 对北极海冰与大气环流进行超前滞后回归分析发现, 大气环流既是海冰异常的前兆因子同时也存在大气对海冰的滞后响应. 总体来说, 大气环流(如平流层极涡)异常超前于海冰异常的回归信号相对更强. 对海冰减退而言, 其前期中纬度涡动热通量出现显著正异常. 海冰减少和极盖位势高度是对前期中纬度强涡动热通量的响应. “极地热大陆冷”异常型也超前海冰减退异常1-2个月, 而在其之后消失, 表明欧亚冷异常和低海冰异常一样, 也是由中纬度强涡动热通量所控制. 本文结果还指出与大气环流异常相关的海冰异常的地理依赖性, 比如巴伦之海-喀拉海海冰减少之前往往出现平流层极涡减弱, 而鄂霍次克海海冰减少则对应着极涡增强. 总之, 从本文对大气超前一面的讨论结果而言, 海冰、平流层极涡和欧亚表面温度等的异常是对对流层大尺度大气环流的响应.

15 April 2020 in Journal of Climate
Verified
Journal Article
Closed access
MKMichael Kelleher
Michael Kelleher
BABlanca Ayarzaguena
Blanca Ayarzaguena
JSJames A. Screen
James A. Screen
Connections across seasons in atmospheric circulation and sea ice have long been sought to advance seasonal prediction. This study presents a link between the springtime stratosphere and Arctic sea ice in summer through autumn. The polar stratospheric vortex dominates the winter stratosphere before breaking down each spring, which is called the stratospheric final warming, as solar radiation returns to the pole. Interannual variability of this breakdown is dynamically driven, leading to different springtime tropospheric and surface circulation patterns. To examine the different impacts of delayed and early final warmings, a multimodel composite was generated from selected CMIP5 models. Additionally, regressions were performed on JRA-55 against an index of springtime polar vortex strength. In both the multimodel composites and reanalysis regressions, significant anomalies in sea ice thickness persist several months following an anomalous timing of the final warming. A later final warming or stronger springtime polar stratospheric vortex leads to negative sea ice thickness anomalies in the East Siberian Sea and positive anomalies in the Beaufort Sea in comparison with an earlier final warming or weaker polar vortex. The spring polar stratospheric vortex is related to spring polar surface circulation patterns. The winds associated with this pattern induce anomalous sea ice motion, moving ice from the East Siberian Sea toward the Beaufort Sea. Reduced sea ice in the East Siberian Sea is linked to anomalous warmth over this region in autumn. Our results suggest that the timing of the stratospheric final warming exerts an influence on the tropospheric circulation and sea ice through autumn, which has implications for seasonal climate prediction.

28 May 2025 in EGUsphere
Verified
Preprint
Open access
PDF file
MKMichael Kelleher
Michael Kelleher
SMSalil Mahajan
Salil Mahajan
Simulating the Earth's climate is an important and complex problem, thus climate models are similarly complex, comprised of tens to hundreds of thousands of lines of code. In order to appropriately utilize the latest computational and software infrastructure advancements in Earth system models running on modern hybrid computing architectures to improve their performance, precision, accuracy, or all three; it is important to ensure that model simulations are repeatable and robust. This introduces the need for establishing statistical or non-bit-for-bit reproducibility, since bit-for-bit reproducibility may not always be achievable. Here, we propose a short-simulation ensemble-based test for an atmosphere model to evaluate the null hypothesis that modified model results are statistically equivalent to that of the original model. We implement this test in US Department of Energy's Energy Exascale Earth System Model (E3SM). The test evaluates a standard set of output variables across the two simulation ensembles and uses a false discovery rate correction to account for multiple testing. The false positive rates of the test are examined using re-sampling techniques on large simulation ensembles and are found to be lower than the currently implemented bootstrapping-based testing approach in E3SM. We also evaluate the statistical power of the test using perturbed simulation ensemble suites, each with a progressively larger magnitude of change to a tuning parameter. The new test is generally found to exhibit more statistical power than the current approach, being able to detect smaller changes in parameter values with higher confidence.

6 January 2026 in Earth System Dynamics
Verified
Journal Article
Open access
PDF file
MKMichael Kelleher
Michael Kelleher
SMSalil Mahajan
Salil Mahajan
Simulating the Earth's climate is an important and complex problem, thus climate models are similarly complex, comprised of millions of lines of code. In order to appropriately utilize the latest computational and software infrastructure advancements in Earth system models running on modern hybrid computing architectures to improve their performance, precision, accuracy, or all three; it is important to ensure that model simulations are repeatable and robust. This introduces the need for establishing statistical or non-bit-for-bit reproducibility, since bit-for-bit reproducibility may not always be achievable. Here, we propose a short-simulation ensemble-based test for an atmosphere model to evaluate the null hypothesis that modified model results are statistically equivalent to that of the original model. We implement this test in version 2 of the US Department of Energy's Energy Exascale Earth System Model (E3SM). The test evaluates a standard set of output variables across the two simulation ensembles and uses a false discovery rate correction to account for multiple testing. The false positive rates of the test are examined using re-sampling techniques on large simulation ensembles and are found to be lower than the currently implemented bootstrapping-based testing approach in E3SM. We also evaluate the statistical power of the test using perturbed simulation ensemble suites, each with a progressively larger magnitude of change to a tuning parameter. The new test is generally found to exhibit more statistical power than the current approach, being able to detect smaller changes in parameter values with higher confidence.