Search

Solar radiation management (SRM) with injections of aerosols into the stratosphere has emerged as a research area of focus with the potential to cool the planet. However, the amount of SRM required to achieve a given level of cooling, and how this relationship evolves in response to increasing greenhouse gas emissions, remains uncertain. Here, we explore the evolution of solar dimming efficacy over time by defining and quantifying a new SRM feedback term, which is analogous to conventional radiative feedbacks. Using Earth system model simulations that dynamically adjust the amount of insolation to offset global mean warming from increasing CO2, we find that positive SRM feedbacks decrease global planetary albedo and diminish the efficacy of solar dimming. Physically, the decrease in albedo is primarily due to reductions in optically thick tropical cloud fraction in the boundary layer and midtroposphere, which is driven by a drying and destabilization of the tropical mid- to lower troposphere. These results offer an energetic explanation for reduced cloud fraction commonly observed in idealized SRM experiments, as well as reaffirm the need to understand the troposphere response, particularly from clouds, in realizable geoengineering experiments and their potential to feed back onto SRM efficacy. Significance Statement: The goal of this study is to understand how the effectiveness of solar geoengineering may evolve over time. Using a climate model with the ability to directly tune the amount of incoming sunlight, we explore the potential for feedback loops in the climate system to diminish or amplify the desired effect of solar tuning, which is to offset greenhouse gas–induced warming. For this climate model and this solar geoengineering proxy, in particular, we find that feedback loops reduce Earth's albedo and therefore diminish the desired effect of turning down the sun over time. This study lays the groundwork for understanding potential feedback loops in climate model simulations that represent solar geoengineering in a more realistic way. [ABSTRACT FROM AUTHOR]

The Canadian Earth System Model version 5.0 (CanESM5.0), the most recent major version of the global climate model developed at the Canadian Centre for Climate Modelling and Analysis (CCCma) at Environment and Climate Change Canada (ECCC), has been used extensively in climate research and for providing future climate projections in the context of climate services. Previous studies have shown that CanESM5.0 performs well compared to other models and have revealed several model biases. To address these biases, the CCCma has recently initiated the "Analysis for Development" (A4D) activity, a coordinated analysis activity in support of CanESM development. Here we describe the goals and organization of this effort and introduce two variants ("p1" and "p2") of a new CanESM version, CanESM5.1, which features important improvements as a result of the A4D activity. These improvements include the elimination of spurious stratospheric temperature spikes and an improved simulation of tropospheric dust. Other climate aspects of the p1 variant of CanESM5.1 are similar to those of CanESM5.0, while the p2 variant of CanESM5.1 features reduced equilibrium climate sensitivity and improved El Niño–Southern Oscillation (ENSO) variability as a result of intentional tuning of the atmospheric component. The A4D activity has also led to the improved understanding of other notable CanESM5.0 and CanESM5.1 biases, including the overestimation of North Atlantic sea ice, a cold bias over sea ice, biases in the stratospheric circulation and a cold bias over the Himalayas. It provides a potential framework for the broader climate community to contribute to CanESM development, which will facilitate further model improvements and ultimately lead to improved climate change information. [ABSTRACT FROM AUTHOR]

Projections of future climate change to support decision‐making require Earth system models (ESMs) running at high spatial resolution, but at present this is computationally prohibitive. A major challenge is the calibration (parameter tuning) during the development of ESMs, which requires running large numbers of simulations to identify optimal values for parameters that are poorly constrained by observations. Here, we train a convolutional neural network (CNN) to emulate perturbed parameter ensembles from two lower‐resolution (and thus much less expensive) versions of the same ESM, and a smaller number of higher‐resolution simulations. Cross‐validated results show that the CNN's skill exceeds that of a climatological baseline for most variables with as few as 5–10 examples of the higher‐resolution ESM, and for all variables (including precipitation) with at least 20 examples. This proof‐of‐concept study demonstrates a machine learning based approach that makes the process of constructing a higher‐resolution emulator 20%–40% more computationally efficient, and thus offers the prospect of significantly more efficient calibration of ESMs. Plain Language Summary: To determine how Earth's future climate will respond to greenhouse gas emissions requires building accurate computer models. Building these models requires a time‐consuming calibration process to find optimal values for uncertain constants (parameters) in the model equations that represent small‐scale processes. We took a machine learning method (called CNN) that is commonly used in image recognition applications and inverted it to replicate the calibration process of the climate model. The CNN reproduces all of the main features of the global simulation of the climate model in a fraction of the computational time, including for precipitation which varies a lot from place‐to‐place. The CNN also makes efficient use of information contained in outputs from simpler versions of the climate model, which are available at much lower cost. Our results suggest that inserting an artificial intelligence method, like CNN, in the calibration process for a climate model can significantly reduce the computational time required. Key Points: Calibration of poorly constrained parameters in higher‐resolution Earth system models (ESMs) is computationally expensiveA machine learning technique from computer vision can replace the ESM during calibration, even for complex variables like precipitationOur machine learning approach reduces the computational costs of emulating a higher‐resolution ESM by 20%–40% [ABSTRACT FROM AUTHOR]

The newest iteration of the Canadian Earth System Model (CanESM5.0.3) has an effective climate sensitivity (EffCS) of 5.65 K, which is a 54 % increase relative to the model's previous version (CanESM2 – 3.67 K), and the highest sensitivity of all current models participating in the sixth phase of the coupled model inter-comparison project (CMIP6). Here, we explore the underlying causes behind CanESM5's increased EffCS via comparison of forcing and feedbacks between CanESM2 and CanESM5. We find only modest differences in radiative forcing as a response to CO2 between model versions. We find small increases in the surface albedo and longwave cloud feedback, as well as a substantial increase in the SW cloud feedback in CanESM5. Through the use of cloud area fraction output and cloud radiative kernels, we find that more positive low and non-low shortwave cloud feedbacks – particularly with regards to low clouds across the equatorial Pacific, as well as subtropical and extratropical free troposphere cloud optical depth – are the dominant contributors to CanESM5's increased climate sensitivity. Additional simulations with prescribed sea surface temperatures reveal that the spatial pattern of surface temperature change exerts controls on the magnitude and spatial distribution of low-cloud fraction response but does not fully explain the increased EffCS in CanESM5. The results from CanESM5 are consistent with increased EffCS in several other CMIP6 models, which has been primarily attributed to changes in shortwave cloud feedbacks. [ABSTRACT FROM AUTHOR]