Search for: All records

The rapid loss of Arctic sea ice is a striking consequence of anthropogenic global warming. Its remote impacts on mid‐latitude weather and climate have attracted scientific and media attention. In this study, we use a hybrid (dynamical plus machine‐learning) atmospheric model—Google's NeuralGCM—to investigate the mid‐latitude atmospheric circulation responses to Arctic sea‐ice loss for the first time. We conduct experiments in which NeuralGCM is forced with pre‐industrial and future sea‐ice concentrations following the protocol of the Polar Amplification Model Intercomparisom Project. To assess the performance of NeuralGCM, we compare the results with those simulated by two physics‐based climate models. NeuralGCM produces a comparable response of near‐surface warming to sea‐ice loss and the subsequent weakened zonal wind in mid‐latitudes. However, there is a substantial discrepancy between the two models' stratospheric responses, where different temperature responses in these models are associated with different zonal wind and geopotential height responses. Further investigation of North Atlantic blocking shows that NeuralGCM produces stronger, more frequent, and more realistic blocking events. Our results demonstrate the capability of NeuralGCM in simulating the tropospheric responses to Arctic sea‐ice loss, but improvements may be needed for the stratospheric representation. 

Abstract Human-induced warming is amplified in the Arctic, but its causes and consequences are not precisely known. Here, we review scientific advances facilitated by the Polar Amplification Model Intercomparison Project. Surface heat flux changes and feedbacks triggered by sea-ice loss are critical to explain the magnitude and seasonality of Arctic amplification. Tropospheric responses to Arctic sea-ice loss that are robust across models and separable from internal variability have been revealed, including local warming and moistening, equatorward shifts of the jet stream and storm track in the North Atlantic, and fewer and milder cold extremes over North America. Whilst generally small compared to simulated internal variability, the response to Arctic sea-ice loss comprises a non-negligible contribution to projected climate change. For example, Arctic sea-ice loss is essential to explain projected North Atlantic jet trends and their uncertainty. Model diversity in the simulated responses has provided pathways to observationally constrain the real-world response. 

Abstract Deep convection in the Indo-Pacific warm pool is vital in driving global atmospheric overturning circulations. Year-to-year variations in the strength and location of warm pool precipitation can lead to significant local and downstream hydroclimatic impacts, including floods and droughts. While the El Niño-Southern Oscillation (ENSO) is recognized as a key factor in modulating interannual precipitation variations in this region, atmospheric internal variability is often as important. Here, through targeted atmospheric model experiments, we identify an intrinsic low-frequency atmospheric mode in the warm pool region during the austral summer, and show that its impact on seasonal rainfall is comparable to ENSO. This mode resembles the horizontal structure of the Madden-Julian Oscillation (MJO), and may play a role in initiating ENSO as stochastic forcing. We show that this mode is not merely an episodic manifestation of MJO events but primarily arises from barotropic energy conversion aided by positive feedback between convection and circulation. 

Abstract In contrast to surface greenhouse warming, surface greenhouse cooling has been less explored, especially on multi-century timescales. Here, we assess the processes controlling the pacing and magnitude of the multi-century surface temperature response to instantaneously doubling and halving atmospheric carbon dioxide concentrations in a modern global coupled climate model. Over the first decades, surface greenhouse warming is larger and faster than surface greenhouse cooling both globally and at high northern latitudes (45–90° N). Yet, this initial multi-decadal response difference does not persist. After year 150, additional surface warming is negligible, but surface cooling and sea ice expansion continues. Notably, the equilibration timescale for high northern latitude surface cooling (∼437 years) is more than double the equivalent timescale for warming. The high northern latitude responses differ most at the sea ice edge. Under greenhouse cooling, the sea ice edge slowly creeps southward into the mid-latitude oceans amplified by positive lapse rate and surface albedo feedbacks. While greenhouse warming and sea ice loss at high northern latitudes occurs on multi-decadal timescales, greenhouse cooling and sea ice expansion occurs on multi-century timescales. Overall, this work shows the importance of multi-century timescales and sea ice processes for understanding high northern latitude climate responses. 

Abstract Arctic amplification (AA), the greater Arctic surface warming compared to the global average, has been widely attributed to increasing concentrations of greenhouse gases (GHG). However, less is known about the impacts of other forcings - notably, anthropogenic aerosols (AER) - and how they may compare to the impacts of GHG. Here we analyze sets of climate model simulations, specifically designed to isolate the AER and GHG effects on global climate. Surprisingly, we find stronger AA produced by AER than by GHG during the 1955–1984 period, when the strongest global AER increase. This stronger AER-induced AA is due to a greater sensitivity of Arctic sea ice, and associated changes in ocean-to-atmosphere heat exchange, to AER forcing. Our findings highlight the asymmetric Arctic climate response to GHG and AER forcings, and show that clean air policies which have reduced aerosol emissions may have exacerbated the Arctic warming over the past few decades. 

A typical Madden–Julian Oscillation (MJO) generates a large region of enhanced rainfall over the equatorial Indian Ocean that moves slowly eastward into the western Pacific. Tropical cyclones often form on the poleward edges of the MJO moist-convective envelope, frequently impacting both southeast Asia and northern Australia, and on occasion Eastern Africa. This paper addresses the question of whether these MJO-induced tropical cyclones will become more numerous in the future as the oceans warm. The Lagrangian Atmosphere Model (LAM), which has been carefully tuned to simulate realistic MJO circulations, is used to study the sensitivity of MJO modulation of tropical cyclogenesis (TCG) to global warming. A control simulation for the current climate is compared with a simulation with enhanced radiative forcing consistent with that for the latter part of the 21st century under Shared Socioeconomic Pathway (SSP) 585. The LAM control run reproduces the observed MJO modulation of TCG, with about 70 percent more storms forming than monthly climatology predicts within the MJO’s convective envelope. The LAM SSP585 run suggests that TCG enhancement within the convective envelope could reach 170 percent of the background value under a high greenhouse gas emissions scenario, owing to a strengthening of Kelvin and Rossby wave components of the MJO’s circulation. 

Machine learning (ML) methods, particularly Reinforcement Learning (RL), have gained widespread attention for optimizing traffic signal control in intelligent transportation systems. However, existing ML approaches often exhibit limitations in scalability and adaptability, particularly within large traffic networks. This paper introduces an innovative solution by integrating decentralized graph-based multi-agent reinforcement learning (DGMARL) with a Digital Twin to enhance traffic signal optimization, targeting the reduction of traffic congestion and network-wide fuel consumption associated with vehicle stops and stop delays. In this approach, DGMARL agents are employed to learn traffic state patterns and make informed decisions regarding traffic signal control. The integration with a Digital Twin module further facilitates this process by simulating and replicating the real-time asymmetric traffic behaviors of a complex traffic network. The evaluation of this proposed methodology utilized PTV-Vissim, a traffic simulation software, which also serves as the simulation engine for the Digital Twin. The study focused on the Martin Luther King (MLK) Smart Corridor in Chattanooga, Tennessee, USA, by considering symmetric and asymmetric road layouts and traffic conditions. Comparative analysis against an actuated signal control baseline approach revealed significant improvements. Experiment results demonstrate a remarkable 55.38% reduction in Eco_PI, a developed performance measure capturing the cumulative impact of stops and penalized stop delays on fuel consumption, over a 24 h scenario. In a PM-peak-hour scenario, the average reduction in Eco_PI reached 38.94%, indicating the substantial improvement achieved in optimizing traffic flow and reducing fuel consumption during high-demand periods. These findings underscore the effectiveness of the integrated DGMARL and Digital Twin approach in optimizing traffic signals, contributing to a more sustainable and efficient traffic management system.