A Denver newspaper in 2016 reported that a new Colorado all-time record peak wind gust of 148 mph was recorded on 18 February 2016, on Monarch Pass in the Colorado Rockies near 11 000 ft above sea level. The article stated that this broke the previous record of 147 mph set on 25 January 1971 at the National Science Foundation (NSF) National Center for Atmospheric Research (NCAR) Mesa Laboratory, at an altitude of 6077 ft, on the western edge of Boulder, Colorado. Though there is no actual official peak gust record in Colorado, this raised the issue that Boulder had not recently experienced winds of the magnitude of the megadownslope windstorms that wracked the area in the 1960s, 1970s, and 1980s when extreme wind gusts recorded at the NSF NCAR Mesa Laboratory were not unusual. Due to Boulder's location at the eastern foot of a north-south mountain range (Earth), it is susceptible to destructive downslope winds (wind) often accompanied by fires (fire) such as the downslope wind-driven Marshall Fire just east of Boulder on 30 December 2021 that destroyed nearly 1100 homes. But after the 1990s, the weather station anemometer at NSF NCAR did not record a peak gust much over 100 mph. What changed? This detective story describes the search for causes of the apparent decrease in strength of extreme windstorms at NSF NCAR and their impacts in the Boulder area. The suspects in Boulder include a change in instrument location, changes in building codes, and increasing roughness length from tree growth. But climate change emerges as a chief culprit. SIGNIFICANCE STATEMENT: National Science Foundation (NSF) National Center for Atmospheric Research (NCAR) was at the epicenter of megadownslope windstorms that wracked Boulder in the 1960s, 1970s, and 1980s when extreme windstorms were not unusual. But after the 1990s, the weather station anemometer at NSF NCAR, which replaced the previous anemometer that recorded the huge gusts, did not record a peak gust much over 100 mph. What changed? This detective story describes the search for causes of the apparent decrease in strength of extreme winds at NSF NCAR and their impacts in the Boulder area. Changing instrument location is part of the story, but climate change emerges as a key culprit.

Global weather patterns and regimes are heavily influenced by the dominant modes of Pacific sea surface temperature (SST) variability, including the El Niño-Southern Oscillation (ENSO), Tropical Pacific Decadal Variability (TPDV), North Pacific Meridional Mode (NPMM), and the Pacific Decadal Oscillation (PDO). However, separating these modes of variability remains challenging due to their spatial overlap and possible nonlinear coupling, which violates the assumptions of traditional linear methods. We develop a Knowledge-Guided AutoEncoder (KGAE) that uses spectral constraints to identify physically interpretable modes, without the need for predefined temporal filters or thresholds. The KGAE separates ENSO-like modes on 2- and 3-7-year timescales and a decadal mode with characteristics reminiscent of the PDO and the NPMM, each with distinct spatial patterns. We demonstrate that the decadal mode modulates ENSO diversity (central Pacific versus eastern Pacific), and that a quasibiennial mode leads and follows the interannual mode, suggesting a role in ENSO onset and decay. When applied to climate model output, KGAEs reveal model-specific biases in ENSO diversity and seasonal timing. Finally, residual training isolates a primarily equatorial decadal mode, which may be a component of TPDV-related decadal variability, likely originating from advection linked to upwelling near the Galápagos Islands and the South Equatorial Current. Our results highlight how machine learning can uncover physically meaningful modes of Earth system variability and improve the representation and evaluation of variability across models and timescales.

We show the results of a study investigating the predominant role of external forcing in steering Atlantic and Pacific ocean variability during the latter half of the 20th (and early 21st) century. By employing the PCMCI+ causal discovery method, we analyze reanalysis data, pacemaker simulations, and a CMIP6 pre-industrial control run. The results reveal a gradual (multi)decadal change in the interactions between major modes of Atlantic and Pacific interannual climate variability from 1950 to 2014. A sliding window analysis identifies a diminishing El Niño-Southern Oscillation (ENSO) effect on the adjacent Atlantic basin through the tropical route, coinciding with the North Atlantic trending toward and maintaining an anomalously warm state after the mid-1980s. In reanalysis, this is accompanied by the prevalence of an extra-tropical pathway connecting ENSO to the tropical Atlantic. Meanwhile, causal networks from reanalysis and pacemaker simulations indicate that increased external forcing might have contributed to strengthening ENSO’s opposite sign response to tropical Atlantic variability during the 1990s and early 21st century, where warming tropical Atlantic sea surface temperatures induced La Niña-like easterly winds in the equatorial Pacific. The analysis of the pre-industrial control run underscores that modes of natural climate variability in the Atlantic and Pacific influence each other also without anthropogenic forcing. Modulation of these interactions by the long-term states of both basins is observed. This work demonstrates the potential of causal discovery for a deeper understanding of mechanisms driving changes in regional and global climate variability. Karmouche, S., Galytska, E., Meehl, G.A., Runge, J.,Weigel, K.,& Eyring,V. (2023b, in review). Changing effects of external forcing on Atlantic-Pacific interactions. EGUsphere, 2023, 1–36. https://doi.org/10.5194/egusphere-2023-18

Abstract. The CMIP6 project was the most expansive and ambitious Model Intercomparison Project (MIP), the latest in a long history, extending back four decades. CMIP has captivated and engaged a broad, growing community focused on improving our climate understanding. It has anchored our ability to quantify and attribute the drivers and responses of the observed climate changes we are experiencing today. The project's profound impact has been achieved by combining the latest climate science and technology. This has enabled the production of latest-generation climate simulations and the dissemination of their output, which has seen increased community attention in every successive phase. The review emphasizes the pragmatics of progressively scaling up efforts, the evolution of how the MIPs were implemented, and the coordinated efforts to establish a minimal infrastructure to make that possible, most recently delivering CMIP6.

The onset of a La Niña event in 2020, with a major contribution from the huge amounts of smoke produced by the disastrous 2019–2020 Australian bushfires, resulted in that event persisting over the next several years with significant impacts worldwide. Here, we attempt to understand the processes and mechanisms related to the wildfire smoke that could have sustained this multi-year high-impact event by analyzing initialized Earth system predictions with E3SMv2 and CESM2 with and without the effects of the Australian bushfire smoke. We hypothesize that Bjerknes feedback sustains the La Niña conditions through an intensified anomalous Walker Circulation that connects strengthened precipitation and ascent in the western Pacific with anomalous subsidence, an invigorated South Pacific High, stronger Trades, and cooler SSTs across the tropical Pacific. Some ensemble members transition to El Niño after 2 years, driven by the development of a positive North Pacific Meridional Mode (NPMM) near Hawaii. Coupled processes in the off-equatorial western Pacific Ocean indicate a connection to the negative phase of the Interdecadal Pacific Oscillation with implications for understanding and predicting interannual and decadal Earth system fluctuations.

Recent studies have highlighted the increasingly dominant role of external forcing in driving Atlantic and Pacific Ocean variability during the second half of the 20th century. This paper provides insights into the underlying mechanisms driving interactions between modes of variability over the two basins. We define a set of possible drivers of these interactions and apply causal discovery to reanalysis data, two ensembles of pacemaker simulations where sea surface temperatures in either the tropical Pacific or the North Atlantic are nudged to observations, and a pre-industrial control run. We also utilize large-ensemble means of historical simulations from the Coupled Model Intercomparison Project Phase 6 (CMIP6) to quantify the effect of external forcing and improve the understanding of its impact. A causal analysis of the historical time series between 1950 and 2014 identifies a regime switch in the interactions between major modes of Atlantic and Pacific climate variability in both reanalysis and pacemaker simulations. A sliding window causal analysis reveals a decaying El Niño–Southern Oscillation (ENSO) effect on the Atlantic as the North Atlantic fluctuates towards an anomalously warm state. The causal networks also demonstrate that external forcing contributed to strengthening the Atlantic's negative-sign effect on ENSO since the mid-1980s, where warming tropical Atlantic sea surface temperatures induce a La Niña-like cooling in the equatorial Pacific during the following season through an intensification of the Pacific Walker circulation. The strengthening of this effect is not detected when the historical external forcing signal is removed in the Pacific pacemaker ensemble. The analysis of the pre-industrial control run supports the notion that the Atlantic and Pacific modes of natural climate variability exert contrasting impacts on each other even in the absence of anthropogenic forcing. The interactions are shown to be modulated by the (multi)decadal states of temperature anomalies of both basins with stronger connections when these states are “out of phase”. We show that causal discovery can detect previously documented connections and provides important potential for a deeper understanding of the mechanisms driving changes in regional and global climate variability.

Climate modelling and analysis are facing new demands to enhance projections and climate information. Here we argue that now is the time to push the frontiers of machine learning beyond state-of-the-art approaches, not only by developing machine-learning-based Earth system models with greater fidelity, but also by providing new capabilities through emulators for extreme event projections with large ensembles, enhanced detection and attribution methods for extreme events, and advanced climate model analysis and benchmarking. Utilizing this potential requires key machine learning challenges to be addressed, in particular generalization, uncertainty quantification, explainable artificial intelligence and causality. This interdisciplinary effort requires bringing together machine learning and climate scientists, while also leveraging the private sector, to accelerate progress towards actionable climate science. Machine learning methods allow for advances in many aspects of climate research. In this Perspective, the authors give an overview of recent progress and remaining challenges to harvest the full potential of machine learning methods.

An adequate characterization of internal modes of climate variability (MoV) is a prerequisite for both accurate seasonal predictions and climate change detection and attribution. Assessing the fidelity of climate models in simulating MoV is therefore essential; however, doing so is complicated by the large intrinsic variations in MoV and the limited span of the observational record. Large ensembles (LEs) provide a unique opportunity to assess model fidelity in simulating MoV and quantify intermodel contrasts. In this work, these goals are pursued in four recently produced LEs: the versions 1 and 2 LEs. In general, the representation of global coupled modes is found to improve across successive E3SM and CESM versions in conjunction with the fidelity of the base state climate while the patterns of extratropical modes are well simulated across the ensembles. Various persistent shortcomings for all MoV are however identified and discussed. The results both demonstrate the successes of these recent model versions and suggest the potential for continued improvement in the representation of MoV with advances in model physics.

Abstract Two Earth system models are analyzed to gain insight into the processes that govern projected changes in the South Asian monsoon. Warmer present‐day base state tropical SSTs contribute to coupled processes that produce greater future tropical Pacific warming in CESM2 with less of an increase in season‐mean monsoon precipitation compared to E3SMv2. This is attributed to changes in the large‐scale east‐west atmospheric Walker circulation, with relatively larger increases in precipitation and upper‐level divergence over the tropical Pacific and increases in upper‐level convergence over South Asia in CESM2. The stronger El Niño‐like response in CESM2, which increases Pacific precipitation and upper‐level divergence farther to the east, and larger future ENSO amplitude in E3SMv2, produce a greater relative increase in future monsoon‐ENSO connections in E3SMv2 compared to CESM2. This analysis indicates that the key processes that affect future monsoon‐ENSO connections are ENSO amplitude and size of the future tropical Pacific El Niño‐like response.

This work assesses a recently produced 21-member climate model large ensemble (LE) based on the U.S. Department of Energy's Energy Exascale Earth System Model (E3SM) version 2 (E3SM2). The ensemble spans the historical era (1850 to 2014) and 21st century (2015 to 2100), using the SSP370 pathway, allowing for an evaluation of the model's forced response. A companion 500-year preindustrial control simulation is used to initialize the ensemble and estimate drift. Characteristics of the LE are documented and compared against other recently produced ensembles using the E3SM version 1 (E3SM1) and Community Earth System Model (CESM) versions 1 and 2. Simulation drift is found to be smaller, and model agreement with observations is higher in versions 2 of E3SM and CESM versus their version 1 counterparts. Shortcomings in E3SM2 include a lack of warming from the mid to late 20th century, likely due to excessive cooling influence of anthropogenic sulfate aerosols, an issue also evident in E3SM1. Associated impacts on the water cycle and energy budgets are also identified. Considerable model dependence in the response to both aerosols and greenhouse gases is documented and E3SM2's sensitivity to variable prescribed biomass burning emissions is demonstrated. Various E3SM2 and CESM2 model benchmarks are found to be on par with the highest-performing recent generation of climate models, establishing the E3SM2 LE as an important resource for estimating climate variability and responses, though with various caveats as discussed herein. As an illustration of the usefulness of LEs in estimating the potential influence of internal variability, the observed CERES-era trend in net top-of-atmosphere flux is compared to simulated trends and found to be much larger than the forced response in all LEs, with only a few members exhibiting trends as large as observed, thus motivating further study.

The effects of differences in climate base state are related to processes associated with the present-day South Asian monsoon simulations in the Energy Exascale Earth System Model version 2 (E3SMv2) and the Community Earth System Model version 2 (CESM2). Though tropical Pacific and Indian Ocean base state sea surface temperatures (SSTs) are over 1 degrees C cooler in E3SMv2 compared to CESM2, and there is an overall reduction of Indian sector precipitation, the pattern of South Asian monsoon precipitation is similar in the two models. Monsoon-ENSO teleconnections, dynamically linked by the large-scale east-west atmospheric circulation, are reduced in E3SMv2 compared to CESM2. In E3SMv2, this is related to cooler tropical SSTs and ENSO amplitude that is less than half that in CESM2. Comparison to a tropical Pacific pacemaker experiment shows, to a first order, that the base state SSTs and ENSO amplitude contribute roughly equally to lower amplitude monsoon-ENSO teleconnections in E3SMv2. Two different Earth system models are analyzed to investigate how differences in simulated base state tropical sea surface temperatures (SSTs) and El Nino/Southern Oscillation (ENSO) amplitude affect the processes associated with the South Asian monsoon. Though tropical SSTs are over 1 degrees C cooler in the Energy Exascale Earth System Model version 2 (E3SMv2) and there is overall reduced Indian sector precipitation, the regional pattern of South Asian monsoon precipitation is similar in the two models. More significantly, monsoon-ENSO teleconnections are reduced in E3SMv2 compared to Community Earth System Model version 2 (CESM2) due to cooler mean tropical SSTs combined with ENSO amplitude in E3SMv2 that is less than half that in CESM2. Base state differences in Energy Exascale Earth System Model version 2 (E3SMv2) compared to Community Earth System Model version 2 (CESM2) include cooler tropical Indian and Pacific sea surface temperatures (SSTs) and reduced ENSO amplitudeBase state differences in the two models do not appreciably affect simulations of the regional patterns of South Asian monsoon precipitationCooler SSTs and lower amplitude ENSO in E3SMv2 combine to contribute about equally to weaker monsoon-ENSO teleconnections compared to CESM2

The Energy Exascale Earth System Model version 2 (E3SMv2) was publicly released by the Department of Energy (DoE) in September 2021. An important component of the model validation were historical climate simulations that followed the protocols of the Coupled Model Intercomparison Project Phase 6 (CMIP6). In particular, five historical ensemble members were recently released, and 16 additional ensemble members will soon become available as part of an E3SMv2 Large Ensemble (LE). The paper sheds light on the characteristics of the E3SMv2 CMIP6 stratospheric circulation which has not yet been documented in the literature. Particular attention is paid to the tropical stratosphere which includes the so-called water vapor tape recorder and the Quasi-Biennial Oscillation. In addition, the general circulation and its variability are briefly described to reveal E3SMv2’s strengths and weaknesses. We compare E3SMv2’s circulation to observations and ERA5 reanalysis data. Furthermore, selected comparisons to the predecessor version E3SMv1 as well as other CMIP6 models are provided to put the results into context. The analysis informs the DoE Sandia National Laboratories project CLDERA which uses the Mt. Pinatubo volcanic eruption for climate attribution studies.

Antarctic shelf ocean warming affects melt of ice shelf/sheets and sea ice but projected changes vary vastly across climate models. A projected increase in El Niño variability has been found to slow future mid-latitude Southern Ocean warming but how this impacts the Antarctic shelf ocean is unknown. Here we show that a projected increase in El Niño variability accelerates Antarctic shelf ocean warming, hastening ice shelf/sheet melt but slowing sea ice reduction.

El Nino-Southern Oscillation (ENSO) can effectively modulate global tropical cyclone (TC) activity, but the role TCs may play in determining ENSO characteristics remains unclear. Here we investigate the impact of TC winds on ENSO using a suite of Earth system model experiments where we insert TC winds, extracted from a TC-permitting high-resolution simulation, into a low-resolution model configuration with nearly no intrinsic TCs. The presence of TC winds in the model increases ENSO power and shifts ENSO frequency closer to what we observe. TCs lead to an increase of strong to extreme El Nino events seen in observations and not simulated in the low-resolution model without intrinsic TCs, mainly through enhanced zonal advection feedback and thermocline feedback. Our results indicate that TCs play a fundamental role in producing the ENSO characteristics we experience today in the climate system and point to a two-way climatological interaction between TCs and ENSO.

Regime-oriented causal model evaluation of Atlantic-Pacific teleconnections in CMIP6Abstract:The Pacific Decadal Variability (PDV) and the Atlantic Multidecadal Variability (AMV) are two important modes of long-term internal variability that significantly impact the climate system and its spatio-temporal changes. In this study, we use a regime-oriented causal discovery method (Karmouche et al, 2022) to examine the changing interactions between the PDV and AMV. The results of this analysis are used to evaluate the ability of models participating in the Coupled Model Intercomparison Project Phase 6 (CMIP6) to represent the observed changing interactions between the PDV, AMV, and their extra-tropical teleconnections.Applying the regime-oriented causal discovery method to reanalysis time series revealed that the interactions between AMV and PDV differ from one regime to the other. The results also show that there are both direct and indirect connections between the Atlantic and Pacific oceans, which are established through various teleconnection patterns.In order to evaluate the ability of climate models to represent these observed interactions, we applied the same regime-oriented causal discovery method to the CMIP6 Large Ensemble historical simulations. We show that several models performed well in simulating the observed causal patterns when AMV and PDV are "out-of-phase", and that the two models with the largest number of members generally outperformed other models in simulating observed causal patterns during longer regimes. This work shows how causal discovery on LEs complements the available diagnostics and statistics metrics of climate variability to provide a powerful tool for climate model evaluation.Karmouche, S., Galytska, E., Runge, J., Meehl, G. A., Phillips, A. S., Weigel, K., and Eyring, V.: Regime-oriented causal model evaluation of Atlantic-Pacific teleconnections in CMIP6, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2022-1013, 2022.

The climate system and its spatio-temporal changes are strongly affected by modes of long-term internal variability, like the Pacific decadal variability (PDV) and the Atlantic multidecadal variability (AMV). As they alternate between warm and cold phases, the interplay between PDV and AMV varies over decadal to multidecadal timescales. Here, we use a causal discovery method to derive fingerprints in the Atlantic–Pacific interactions and to investigate their phase-dependent changes. Dependent on the phases of PDV and AMV, different regimes with characteristic causal fingerprints are identified in reanalyses in a first step. In a second step, a regime-oriented causal model evaluation is performed to evaluate the ability of models participating in the Coupled Model Intercomparison Project Phase 6 (CMIP6) in representing the observed changing interactions between PDV, AMV and their extra-tropical teleconnections. The causal graphs obtained from reanalyses detect a direct opposite-sign response from AMV to PDV when analyzing the complete 1900–2014 period and during several defined regimes within that period, for example, when AMV is going through its negative (cold) phase. Reanalyses also demonstrate a same-sign response from PDV to AMV during the cold phase of PDV. Historical CMIP6 simulations exhibit varying skill in simulating the observed causal patterns. Generally, large-ensemble (LE) simulations showed better network similarity when PDV and AMV were out of phase compared to other regimes. Also, the two largest ensembles (in terms of number of members) were found to contain realizations with similar causal fingerprints to observations. For most regimes, these same models showed higher network similarity when compared to each other. This work shows how causal discovery on LEs complements the available diagnostics and statistical metrics of climate variability to provide a powerful tool for climate model evaluation.

The subpolar North Atlantic (SPNA) shows contrasting responses in two sensitivity experiments with increased stratospheric aerosols, offering insight into the physical processes that may impact the Atlantic meridional overturning circulation (AMOC) in a warmer climate. In one, the upper ocean becomes warm and salty, but in the other it becomes cold and fresh. The changes are accompanied by diverging AMOC responses. The first experiment strengthens the AMOC, opposing the weakening trend in the reference simulation. The second experiment shows a much smaller impact. Both simulations use the Community Earth System Model with the Whole Atmosphere Community Climate Model component (CESM-WACCM) but differ in model versions and stratospheric aerosol specifications. Despite both experiments using similar approaches to increase stratospheric aerosols to counteract the rising global temperature, the contrasting SPNA and AMOC responses indicate a considerable dependency on model physics, climate states, and model responses to forcings. This study focuses on examining the physical processes involved with the impact of stratospheric aerosols on the SPNA salinity changes and their potential connections with the AMOC and the Arctic. We find that in both cases, increased stratospheric aerosols act to enhance the SPNA upper-ocean salinity by reducing freshwater export from the Arctic, which is closely tied to the Arctic sea ice changes. The impact on AMOC is primarily through the thermal component of the surface buoyancy fluxes, with negligible contributions from the freshwater component. These experiments shed light on the physical processes that dictate the important connections between the SPNA, the Arctic, the AMOC, and their subsequent feedbacks on the climate system.

Tropical cyclones (TCs) alter upper-ocean temperature and influence ocean heat content via enhanced turbulent mixing. A better understanding of the role of TCs within the climate system requires a fully coupled modeling framework, where TC-induced ocean responses feed back to the atmosphere and subsequently to the climate mean state and variability. Here, we investigate the impacts of TC wind forcing on the global ocean and the associated feedbacks within the climate system using the fully coupled Community Earth System Model version 1.3 (CESM1.3). Using the low-resolution version of CESM1.3 (1 degrees atmosphere and ocean grid spacing) with no intrinsic TCs, we conduct a suite of sensitivity experiments by inserting TC winds extracted from a high-resolution (0.25 degrees atmosphere grid spacing) TC-permitting simulation into the low-resolution model. Results from the low-resolution TC experiment are compared to a low-resolution control simulation to diagnose TCs' impact. We found that the added TC winds can increase ocean heat content by affecting ocean vertical mixing, air-sea enthalpy fluxes, and cloud amount. The added TCs can influence mean SST, precipitation, ocean subsurface temperature, and ocean mixed layer depth. We found a strengthening of the wind-driven subtropical cells and a weakening of the Atlantic meridional overturning circulation due to the changes of surface buoyancy fluxes. TCs in the model cause anomalous equatorward ocean heat convergence in the deep tropics and an increase of poleward ocean heat transport out of the subtropics. Our modeling results provide new insights into the multiscale interactions between TCs and the coupled climate system.

This study focuses on assessing the representation and predictability of North American weather regimes, which are persistent large-scale atmospheric patterns, in a set of initialized subseasonal reforecasts created using the Community Earth System Model, version 2 (CESM2). The k-means clustering was used to extract four key North American (10 degrees-70 degrees N, 150 degrees-40 degrees W) weather regimes within ERA5 reanalysis, which were used to interpret CESM2 subseasonal forecast performance. Results show that CESM2 can recreate the climatology of the four main North American weather regimes with skill but exhibits biases during later lead times with overoccurrence of the West Coast high regime and underoccurrence of the Greenland high and Alaskan ridge regimes. Overall, the West Coast high and Pacific trough regimes exhibited higher predictability within CESM2, partly related to El Ni & ntilde;o. Despite biases, several reforecasts were skillful and exhibited high predictability during later lead times, which could be partly attributed to skillful representation of the atmosphere from the tropics to extratropics upstream of North America. The high predictability at the subseasonal time scale of these case-study examples was manifested as an "ensemble realignment," in which most ensemble members agreed on a prediction despite ensemble trajectory dispersion during earlier lead times. Weather regimes were also shown to project distinct temperature and precipitation anomalies across North America that largely agree with observational products. This study further demonstrates that unsupervised learning methods can be used to uncover sources and limits of subseasonal predictability, along with systematic biases present in numerical prediction systems.