Search NASA⌕ Search

Engineering topics

Kravitz, Benjamin S.

Publications and source records attributed to Kravitz, Benjamin S..

8 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

The Potential of Stratospheric Aerosol Injection to Reduce the Climatic Risks of Explosive Volcanic Eruptions

Sulfur-rich volcanic eruptions happen sporadically. If Stratospheric Aerosol Injection (SAI) were to be deployed, it is likely that explosive volcanic eruptions would happen during such a deployment. Here we use an ensemble of Earth System Model simulations to show how changing the injection strategy post-eruption could be used to reduce the climate risks of a large volcanic eruption; the risks are also modified even without any change to the strategy. For a medium-size eruption (10 Tg-SO 2 ) comparable to the SAI injection rate, the volcanic-induced cooling would be reduced if it occurs under SAI, especially if artificial sulfur dioxide injections were immediately suspended. Alternatively, suspending injection only in the eruption hemisphere and continuing injection in the opposite would reduce shifts in precipitation in the tropical belt and thus mitigate eruption-induced drought. Finally, we show that for eruptions much larger than the SAI deployment, changes in SAI strategy would have minimal effect.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Stratospheric Aerosol Injection Can Reduce Risks to Antarctic Ice Loss Depending on Injection Location and Amount

Owing to increasing greenhouse gas emissions, the Antarctic Ice Sheet is vulnerable to rapid ice loss in the upcoming decades and centuries. Here this study examines the effectiveness of using stratospheric aerosol injection (SAI) that minimizes global mean temperature (GMT) change to slow projected 21st century Antarctic ice loss. We simulate 11 different SAI cases which vary by the latitudinal location(s) and the amount(s) of the injection(s) to examine the climatic response near Antarctica in each case as compared to the reference climate at the turn of the last century. We demonstrate that injecting at a single latitude in the northern hemisphere or at the Equator increases Antarctic shelf ocean temperatures pertinent to ice shelf basal melt, while injecting only in the southern hemisphere minimizes this temperature change. We use these results to analyze the results of more complex multi-latitude injection strategies that maintain GMT at or below 1.5°C above the pre-industrial. All these multi-latitude cases will slow Antarctic ice loss relative to the mid-to-late 21st century SSP2-4.5 emissions pathway. Yet, to avoid a GMT threshold estimated by previous studies pertaining to rapid West Antarctic ice loss (1.5°C above the pre-industrial GMT, though large uncertainty), our study suggests SAI would need to cool about 1.0°C below this threshold and predominately inject at low southern hemisphere latitudes (~15°S - 30°S). These results highlight the complexity of factors impacting the Antarctic response to SAI and the critical role of the injection strategy in preventing future ice loss.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Harnessing stratospheric diffusion barriers for enhanced climate geoengineering

Stratospheric sulfate aerosol geoengineering is a proposed method to temporarily intervene in the climate system to increase the reflectance of shortwave radiation and reduce mean global temperature. In previous climate modeling studies, choosing injection locations for geoengineering aerosols has, thus far, only utilized the average dynamics of stratospheric wind fields instead of accounting for the essential role of time-varying material transport barriers in turbulent atmospheric flows. Here we conduct the first analysis of sulfate aerosol dispersion in the stratosphere, comparing what is now a standard fixed-injection scheme with time-varying injection locations that harness short-term stratospheric diffusion barriers. We show how diffusive transport barriers can quickly be identified, and we provide an automated injection location selection algorithm using short forecast and reanalysis data. Within the first 7 d days of transport, the dynamics-based approach is able to produce particle distributions with greater global coverage than fixed-site methods with fewer injections. Additionally, this enhanced dispersion slows aerosol microphysical growth and can reduce the effective radii of aerosols up to 200–300 d after injection. While the long-term dynamics of aerosol dispersion are accurately predicted with transport barriers calculated from short forecasts, the long-term influence on radiative forcing is more difficult to predict and warrants deeper investigation. Statistically significant changes in radiative forcing at timescales beyond the forecasting window showed mixed results, potentially increasing or decreasing forcing after 1 year when compared to fixed injections. We conclude that future feasibility studies of geoengineering should consider the cooling benefits possible by strategically injecting sulfate aerosols at optimized time-varying locations. Our method of utilizing time-varying attracting and repelling structures shows great promise for identifying optimal dispersion locations, and radiative forcing impacts can be improved by considering additional meteorological variables.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Response of the Indian Summer Monsoon to Global Warming, Solar Geoengineering and its Termination

The response of the Indian Summer Monsoon (ISM) to global warming, solar geoengineering and its termination is examined using the multi-model mean of eight global climate model simulations from G2 experiment of the Geoengineering Model Intercomparison Project. Under the global warming scenario, land-ocean temperature contrasts and low-level monsoon circulation progressively strengthen accompanied by enhanced precipitation over the Indian subcontinent. Notably, in the solar geoengineered scenario, marginal surface cooling is projected over the majority of the ISM region, and there is strengthening of both upper and lower level circulation. Upon the termination of the geoengineering, the climatic conditions temperature, precipitation, and winds — would abruptly change to what it would have been under the global warming scenario, and such abrupt change may cause devastating consequences.

Bhowmick, Mansi↗
DOE OSTI2021

Is turning down the sun a good proxy for stratospheric sulfate geoengineering?

Deliberately blocking out a small portion of the incoming solar radiation would cool the climate. One such approach would be injecting SO2 into the stratosphere, which would produce sulfate aerosols that would remain in the atmosphere for 1–3 years, reflecting part of the incoming shortwave radiation. The cooling produced by the aerosols can offset the warming produced by increased greenhouse gas (GHG) concentrations, but it would also affect the climate differently, leading to residual differences compared to a climate not affected by either. Many climate model simulations of geoengineering have used a uniform reduction of the incoming solar radiation as a proxy for stratospheric aerosols, both because many models are not designed to adequately capture relevant stratospheric aerosol processes, and because a solar reduction has often been assumed to capture the most important differences between how stratospheric aerosols and GHG would affect the climate. Here we show that dimming the sun does not produce the same surface climate effects as simulating aerosols in the stratosphere. By more closely matching the spatial pattern of solar reduction to that of the aerosols, some improvements in this idealized representation are possible, with further improvements if the stratospheric heating produced by the aerosols is included. This is relevant both for our understanding of the physical mechanisms driving the changes observed in stratospheric-sulfate geoengineering simulations, and in terms of the relevance of impact assessments that use a uniform solar dimming.

Visioni, Daniele↗
DOE OSTI2020

Loosely Conditioned Emulation of Global Climate Models With Generative Adversarial Networks

Climate models encapsulate our best understanding of the Earth system, allowing research to be conducted on its future under alternative assumptions of how human-driven climate forces are going to evolve. An important application of climate models is to provide metrics of mean and extreme climate changes, particularly under these alternative future scenarios, as these quantities drive the impacts of climate on society and natural systems. Because of the need to explore a wide range of alternative scenarios and other sources of uncertainties in a computationally efficient manner, climate models can only take us so far, as they require significant computational resources, especially when attempting to characterize extreme events, which are rare and thus demand long and numerous simulations in order to accurately represent their changing statistics. Here we use deep learning in a proof of concept that lays the foundation for emulating global climate model output for different scenarios. We train two "loosely conditioned" Generative Adversarial Networks (GANs) that emulate daily precipitation output from a fully coupled Earth system model: one GAN modeling Fall-Winter behavior and the other Spring-Summer. Our GANs are trained to produce spatiotemporal samples: 32 days of precipitation over a 64x128 regular grid discretizing the globe. We evaluate the generator with a set of related performance metrics based upon KL divergence, and find the generated samples to be nearly as well matched to the test data as the validation data is to test. We also find the generated samples to accurately estimate the mean number of dry days and mean longest dry spell in the 32 day samples. Our trained GANs can rapidly generate numerous realizations at a vastly reduced computational expense, compared to large ensembles of climate models, which greatly aids in estimating the statistics of extreme events.

climate emulation, extreme climate, impacts, machi↗
DOE OSTI2020

Technical note: Deep learning for creating surrogate models of precipitation in Earth system models

Abstract. We investigate techniques for using deep neural networks to produce surrogatemodels for short-term climate forecasts. A convolutional neural network istrained on 97 years of monthly precipitation output from the 1pctCO2 run (theCO 2 concentration increases by 1 % per year) simulated by the second-generation Canadian Earth System Model (CanESM2). The neural network clearly outperforms a persistence forecast anddoes not show substantially degraded performance even when the forecast lengthis extended to 120 months. The model is prone to underpredicting precipitationin areas characterized by intense precipitation events. Scheduled sampling(forcing the model to gradually use its own past predictions rather than groundtruth) is essential for avoiding amplification of early forecasting errors.However, the use of scheduled sampling also necessitates preforecasting(generating forecasts prior to the first forecast date) to obtain adequateperformance for the first few prediction time steps. We document the trainingprocedures and hyperparameter optimization process for researchers who wish toextend the use of neural networks in developing surrogate models.

54 ENVIRONMENTAL SCIENCES↗