David Paynter

Changes in emissions of anthropogenic aerosols and their precursors made strong contributions to trends in global and regional climate over the historical era. Further strong and rapid changes are either underway or anticipated, and are expected to continue to influence global and regional climate in the coming decades. However, large uncertainties in both emission pathways and climate responses make anthropogenic aerosols a key factor in the spread of near-term climate projections, especially on regional scales. The regional aerosol model intercomparison project (RAMIP) is designed to quantify the climate and air quality impacts of feasible near-future changes in anthropogenic aerosol emissions, providing information that can directly link regional climate policies to regional climate change. We present high-level results from Phase 1 of RAMIP, documenting climate responses to regional aerosol perturbations in 10 CMIP6 generation Earth System Models. If aerosol emissions are reduced globally, following SSP1-2.6 instead of remaining high under SSP3-7.0, it results in effective radiative forcing of 0.71 ± 0.27 W m − 2 in 2050, and a 0.31 ± 0.12 K increase in global-mean annual-mean temperature. Warming occurs largely over the Northern Hemisphere, causing a northward shift of the Intertropical Convergence Zone and intensifying the Northern Hemisphere monsoons. RAMIP reveals distinct spatial patterns in the climate response to regional aerosol reductions. Reductions over North America and East Asia have hemisphere-wide impacts, accounting for most of the global-mean response. South Asian reductions have a predominantly local impact, while reductions over Africa and the Middle East have significant impacts on precipitation throughout the tropics. The regional experiments also reveal that regional climate can be as strongly influenced by remote emission reductions as those made locally. These results highlight a need to consider local and external policies when designing adaptation plans, and motivate work to better represent spatial responses aerosol in regional climate models, emulators, and impact models.

We implemented in GFDL's global atmospheric model a parameterization for convective gravity waves, which explicitly describes the wave source based on the latent heating rates from the parameterized deep convection. The characteristics of the simulated Quasi‐Biennial Oscillation (QBO) in the equatorial stratosphere depend on the tunable parameters in the convective gravity wave scheme. In particular, a larger efficiency factor or a smaller convective area leads to a shorter QBO period. A less‐than‐one scaling factor for the convective heating depth, though unphysical, is necessary for the model to simulate a realistic QBO over the lower stratosphere. Furthermore, we relax the assumption of constant convective area and replace it with the prognostic convective area diagnosed from the convection scheme. While the model with a constant convective area simulates a similar QBO to that with variable convective area at the present‐day condition, it simulates a stronger reduction of the QBO period and the QBO amplitude in response to surface warming. Our results highlight the sensitivity of the simulated QBO in general circulation models to details of the parameterization for convective gravity waves.

Effective climate sensitivity (E _ff CS) is a key metric for quantifying climate change, yet its uncertainty remains large, primarily due to intermodel differences in cloud feedback. Here, we assess E _ff CS and climate feedback to an abrupt doubling (2xCO2) and quadrupling (4xCO2) of CO ₂ over 450 years in GFDL CM4.0. Both runs have comparable E _ff CS estimates during the first 50 years (2xCO2: 2.99 K; 4xCO2: 3.25 K) with the 4xCO2 E _ff CS estimate still slightly stronger for Years 51–150. Extending the analysis to 450 years increases the E _ff CS estimate, but more so in 2xCO2 (5.8 K) than in 4xCO2 (4.79 K). In both runs, the E _ff CS increases due to a weakening of the net feedback with warming, largely through shortwave cloud feedback increase. A weaker E _ff CS during the slow response for 4xCO2 relative to 2xCO2 is associated with a weaker surface albedo feedback due to an earlier sea‐ice loss in both hemispheres.

Accurate and efficient radiative transfer computations are critical for numerical weather forecast and climate prediction, but current radiative transfer models are computationally expensive. We developed a neural network (NN) scheme that emulates a physics‐based radiative transfer scheme, achieving 4–5 times faster computation with high accuracy for well‐trained, targeted scenarios. However, NN models struggle with extrapolation to out‐of‐sample weather or climate conditions. To address this, we propose HRadNN, a hybrid framework that uses a NN scheme by default and switches to a physics‐based scheme whenever energy conservation is not maintained. Online simulations demonstrate HRadNN's accuracy in present‐day climate and stability in extreme warming/cooling scenarios (up to 9 K change of the sea surface temperature) even without training specific to these scenarios. This indicates that machine learning can accelerate radiative transfer computation while maintaining accuracy and efficiency, provided it is quality‐controlled by physical laws such as energy conservation.

Anthropogenic aerosol effective radiative forcing (ERF) remains one of the largest uncertainties in quantifying the human influence on climate. Estimates of aerosol ERF depend on the background meteorological state and composition, including factors such as cloud regimes, circulation, and pre‐existing aerosol and precursor levels. Here, we investigate the sensitivity of anthropogenic aerosol ERF to these background conditions using the GFDL‐ESM4.1 model. Following and extending the CMIP6 AerChemMIP protocol, we conducted pairs of simulations under different base states, by comparing present‐day versus pre‐industrial sea surface temperatures, sea‐ice, and atmospheric composition, to diagnose aerosol ERF. We performed experiments isolating individual aerosol species (sulfate, black carbon, organic carbon, and ammonia) to examine their ERF dependence on such base states. We find that the diagnosed global‐mean aerosol ERF is relatively robust but exhibits species‐specific sensitivity across approaches, especially for black carbon and ammonia, while regional aerosol ERF differs strongly depending on the background climate and composition, demonstrating sensitivity to meteorological, and in particular, background composition conditions. Specifically, black carbon ERF is much more positive in a present‐day background due to enhanced absorption from internal mixing with co‐emitted aerosols. We also find that aerosol ERFs are not strictly additive across species, reflecting nonlinear interactions linked to cloud and aerosol microphysical processes under different background states. Our findings highlight the importance of accounting for background climate and composition conditions and nonlinear multi‐species interactions when assessing aerosol ERF in climate models.

The availability of up-to-date forcing datasets is critical for reconciling Earth system model simulations with contemporary observations. With the newly released CMIP7 forcing dataset, which extends the historical record through 2021, this study provides the first systematic evaluation of how using updated forcings over the “middle-of-the-road” SSP2-4.5 scenario forcings influences the simulated radiation budget. The updated forcing dataset used in the atmosphere model does not improve the agreement between the simulated global-mean top-of-the-atmosphere radiation trends and the satellite-observed trends. Regionally, we identify pronounced shifts in radiative energy partitioning between atmosphere and surface. Specifically, during 2001 ~ 2021, accelerated reductions in sulfate and black carbon aerosols over East Asia drive a significant decline in atmospheric solar absorption that matches observations, while a substantial increase in organic aerosols at higher latitudes induces a localized increase in atmospheric absorption that does not match observations. The substantial influence of localized forcing changes on simulated regional climate underscores the necessity for routine updates of forcing datasets.

The Transient Climate Response (TCR) is a metric of climate sensitivity relevant to present‐day climate change, but has been less studied than the Equilibrium Climate Sensitivity. Here, we emphasize the utility of the TCR by demonstrating that temperature and radiative forcing in transient scenarios are linearly related, where the proportionality constant is essentially the TCR. This relationship has been invoked in previous studies but not extensively validated; here it is shown to hold across models as well as forcing scenarios. This linear scaling furthermore allows for a simple estimate of the time to cross a given temperature target, given only a forcing scenario and a value for TCR. Deviations from a strictly linear relationship between temperature and forcing are governed by the “slow” component of global warming driven by declining deep ocean heat uptake, but can only be fully accounted for by incorporating changing patterns of surface warming via ocean heat uptake “efficacy.”

The Transient Climate Response (TCR) is a metric of climate sensitivity relevant to present-day climate change, but has been less well studied than the Equilibrium Climate Sensitivity. Here, we emphasize the utility of the TCR by demonstrating that temperature and radiative forcing in transient scenarios are linearly related (to a good approximation), where the proportionality constant is essentially the TCR. This previously known linear relationship is shown here to hold across forcing scenarios as well as across models. This relationship furthermore allows for a simple estimate of the time to cross a given temperature target, given only a forcing scenario and a value for TCR. Deviations from a strictly linear relationship between temperature and forcing are influenced by the ‘slow’ component of global warming driven by declining deep ocean heat uptake, especially when changing patterns of surface warming are accounted for via ocean heat uptake ‘efficacy’.

Earth’s energy budget is modulated by climate forcings and feedbacks. Outgoing longwave radiation (OLR) is a critical component of the Earth’s energy budget, and the balance between absorbed solar radiation and OLR determines the net gain of energy of the entire climate system. Radiation computations using 43 years of ERA5 reanalysis profiles show an increase in clear-sky OLR in the polar regions and a decrease in the tropics. To understand this contrast, we analyze zonal- and annual-mean clear-sky OLR changes in the Northern Hemisphere from 1979 to 2021. We identify a transition in temporal clear-sky OLR changes from low to high latitudes, with a “crossover” point where the sign changes around 30°N. We improve upon previous studies by employing infrared spectrally-detailed line-by-line radiative transfer calculations, to decompose zonal-mean trends caused by forcing and feedback mechanisms. Our diagnostic investigation shows that the increase in water vapor exerts a dominant negative contribution to the OLR temporal change in the tropics due to a larger OLR sensitivity to water vapor increase there, while the increase in surface temperature (T _s) dominates the positive temporal change in the Arctic, both due to Arctic amplification and a larger OLR sensitivity to T _s increase. Greater greenhouse gases’ forcings and smaller OLR sensitivity to T _s change in the tropics, as well as an imperfect cancellation between atmospheric temperature and water vapor contribution, delineates the crossover point at around 30°N.

Atmospheric general circulation models are important tools to study climate change. A routine vindication for these models is to compare simulations with prescribed historical sea surface temperature (SST) and sea ice (SI) fraction against observations. Here, we report significant differences in the Earth's radiation budget in an Atmospheric general circulation model prescribed by different SST/SI data sets. This is caused by the distinct trends across different SST/SI data sets over the past decades despite all being based on observations. In particular, one data set reports a double rate of warming and an abrupt increase in SI coverage compared to the others. When using this data set as the boundary condition, the simulation predicts Earth's energy loss, in contrast to the observed energy gain. Our study calls for awareness of the impact of SST/SI data set uncertainty on radiation trends, especially for multimodel comparison projects, including CERESMIP and CMIP7.