Search for: All records

Metaphor is crucial in human cognition and creativity, facilitating abstract thinking, analogical reasoning, and idea generation. Typically, human raters manually score the originality of responses to creative thinking tasks – a laborious and error-prone process. Previous research sought to remedy these risks by scoring creativity tasks automatically using semantic distance and large language models (LLMs). Here, we extend research on automatic creativity scoring to metaphor generation – the ability to creatively describe episodes and concepts using nonliteral language. Metaphor is arguably more abstract and naturalistic than prior targets of automated creativity assessment. We collected 4,589 responses from 1,546 participants to various metaphor prompts and corresponding human creativity ratings. We fine-tuned two open-source LLMs (RoBERTa and GPT-2) – effectively “teaching” them to score metaphors like humans – before testing their ability to accurately assess the creativity of new metaphors. Results showed both models reliably predicted new human creativity ratings (RoBERTa r = .72, GPT-2 r = .70), significantly more strongly than semantic distance (r = .42). Importantly, the fine-tuned models generalized accurately to metaphor prompts they had not been trained on (RoBERTa r = .68, GPT-2 r = .63). We provide open access to the fine-tuned models, allowing researchers to assess metaphor creativity in a reproducible and timely manner. 

Abstract. Carbonyl sulfide (COS) is the most abundant sulfur gas in the atmosphere with links to terrestrial and oceanic productivity. We measured COS in ice core air from an intermediate-depth ice core from the South Pole using both dry and wet extraction methods, recovering a 52 500-year record. We find evidence for COS production in the firn, altering the atmospheric signal preserved in the ice core. Mean sea salt aerosol concentrations from the same depth are a good proxy for the COS production, which disproportionately impacts the measurements from glacial period ice with high sea salt aerosol concentrations. The COS measurements are corrected using sea salt sodium (ssNa) as a proxy for the excess COS resulting from the production. The ssNa-corrected COS record displays substantially less COS in the glacial period atmosphere than the Holocene and a 2 to 4-fold COS rise during the deglaciation synchronous with the associated climate signal. The deglacial COS rise was primarily source driven. Oceanic emissions in the form of COS, carbon disulfide (CS2), and dimethylsulfide (DMS) are collectively the largest natural source of atmospheric COS. A large increase in ocean COS emissions during the deglaciation suggests enhancements in emissions of ocean sulfur gases via processes that involve ocean productivity, although we cannot quantify individual contributions from each gas. 

The atmospheric history of molecular hydrogen (H 2 ) from 1852 to 2003 was reconstructed from measurements of firn air collected at Megadunes, Antarctica. The reconstruction shows that H 2 levels in the southern hemisphere were roughly constant near 330 parts per billion (ppb; nmol H 2 mol −1 air) during the mid to late 1800s. Over the twentieth century, H 2 levels rose by about 70% to 550 ppb. The reconstruction shows good agreement with the H 2 atmospheric history based on firn air measurements from the South Pole. The broad trends in atmospheric H 2 over the twentieth century can be explained by increased methane oxidation and anthropogenic emissions. The H 2 rise shows no evidence of deceleration during the last quarter of the twentieth century despite an expected reduction in automotive emissions following more stringent regulations. During the late twentieth century, atmospheric CO levels decreased due to a reduction in automotive emissions. It is surprising that atmospheric H 2 did not respond similarly as automotive exhaust is thought to be the dominant source of anthropogenic H 2. The monotonic late twentieth century rise in H 2 levels is consistent with late twentieth-century flask air measurements from high southern latitudes. An additional unknown source of H 2 is needed to explain twentieth-century trends in atmospheric H 2 and to resolve the discrepancy between bottom-up and top-down estimates of the anthropogenic source term. The firn air–based atmospheric history of H 2 provides a baseline from which to assess human impact on the H 2 cycle over the last 150 y and validate models that will be used to project future trends in atmospheric composition as H 2 becomes a more common energy source. 

Abstract Reconstructions of paleoatmospheric H₂using polar firn air and ice cores would lead to a better understanding of the H₂biogeochemical cycle and how it is influenced by climate change and human activity. In this study, the permeability, diffusivity, and solubility of H₂are determined experimentally in ice Ih at temperatures relevant to polar ice sheets (199–253 K). The experimental data are used in conjunction with simplified diffusion models to assess the implications for: (a) Diffusion of H₂from pressurized closed bubbles to open pores in polar firn, (b) diffusive smoothing of H₂gradients in the ice sheet, and (c) post‐coring diffusive losses of H₂from ice core samples. The results indicate that diffusive equilibrium between open and closed pores is likely achieved in the firn lock‐in zone. Diffusive smoothing of atmospheric variations is significant and should be accounted for in atmospheric reconstructions on millennial time scales. Diffusive losses from a bubbly ice sample are sufficiently slow that samples may be meaningfully analyzed for H₂after storage on the order of a year. These results suggest that the mobility of H₂in ice should not preclude the reconstruction of paleoatmospheric H₂from firn air and ice cores.