Search for: All records

Abstract Artificial intelligence (AI) and machine learning (ML) pose a challenge for achieving science that is both reproducible and replicable. The challenge is compounded in supervised models that depend on manually labeled training data, as they introduce additional decision‐making and processes that require thorough documentation and reporting. We address these limitations by providing an approach to hand labeling training data for supervised ML that integrates quantitative content analysis (QCA)—a method from social science research. The QCA approach provides a rigorous and well‐documented hand labeling procedure to improve the replicability and reproducibility of supervised ML applications in Earth systems science (ESS), as well as the ability to evaluate them. Specifically, the approach requires (a) the articulation and documentation of the exact decision‐making process used for assigning hand labels in a “codebook” and (b) an empirical evaluation of the reliability” of the hand labelers. In this paper, we outline the contributions of QCA to the field, along with an overview of the general approach. We then provide a case study to further demonstrate how this framework has and can be applied when developing supervised ML models for applications in ESS. With this approach, we provide an actionable path forward for addressing ethical considerations and goals outlined by recent AGU work on ML ethics in ESS. 

Abstract While considerable attention has been given to how convectively coupled Kelvin waves (CCKWs) influence the genesis of tropical cyclones (TCs) in the Atlantic Ocean, less attention has been given to their direct influence on African easterly waves (AEWs). This study builds a climatology of AEW and CCKW passages from 1981 to 2019 using an AEW-following framework. Vertical and horizontal composites of these passages are developed and divided into categories based on AEW position and CCKW strength. Many of the relationships that have previously been found for TC genesis also hold true for non-developing AEWs. This includes an increase in convective coverage surrounding the AEW center in phase with the convectively enhanced (“active”) CCKW crest, as well as a buildup of relative vorticity from the lower to upper troposphere following this active crest. Additionally, a new finding is that CCKWs induce specific humidity anomalies around AEWs that are qualitatively similar to those of relative vorticity. These modifications to specific humidity are more pronounced when AEWs are at lower latitudes and interacting with stronger CCKWs. While the influence of CCKWs on AEWs is mostly transient and short lived, CCKWs do modify the AEW propagation speed and westward-filtered relative vorticity, indicating that they may have some longer-term influences on the AEW life cycle. Overall, this analysis provides a more comprehensive view of the AEW–CCKW relationship than has previously been established, and supports assertions by previous studies that CCKW-associated convection, specific humidity, and vorticity may modify the favorability of AEWs to TC genesis over the Atlantic. 

Abstract We introduce the National Science Foundation (NSF) AI Institute for Research on Trustworthy AI in Weather, Climate, and Coastal Oceanography (AI2ES). This AI institute was funded in 2020 as part of a new initiative from the NSF to advance foundational AI research across a wide variety of domains. To date AI2ES is the only NSF AI institute focusing on environmental science applications. Our institute focuses on developing trustworthy AI methods for weather, climate, and coastal hazards. The AI methods will revolutionize our understanding and prediction of high-impact atmospheric and ocean science phenomena and will be utilized by diverse, professional user groups to reduce risks to society. In addition, we are creating novel educational paths, including a new degree program at a community college serving underrepresented minorities, to improve workforce diversity for both AI and environmental science.