Jessamin A. Straub | Search USGS

USGS2023

The During Nearshore Event Experiment (DUNEX): A collaborative coastal community experiment to address coastal resilience

The During Nearshore Event Experiment (DUNEX) was a large-scale coastal field effort focused on improving understanding of during-storm nearshore processes to ultimately develop predictive technologies, engineering solutions, and actions to enhance coastal resilience. The experiments were conducted on the North Carolina coast by a multidisciplinary group of over 30 research scientists from 18 academic and federal institutions supporting over 30 graduate students and deploying over 300 instruments from 2019 to 2021. The overarching goal of DUNEX was to gather information collaboratively to improve understanding of the interactions of coastal water levels, waves, currents, beach and dune evolution, soil behavior, vegetation, and groundwater during major coastal storms that affect infrastructure, habitats, and communities. In the short term, these high-quality field measurements will lead to better understanding of during-storm processes and impacts and will enhance U.S. academic coastal research programs by providing opportunities for students to learn about field data collection and to potentially analyze data as part of their studies. Longer-term, DUNEX data and outcomes will improve the ability to predict extreme event physical processes and impacts, validate coastal processes numerical models, and improve coastal resilience strategies and communication methods for coastal communities impacted by storms. The purpose of this paper is to describe the motivation for and science goals of the experiment, how stakeholder needs led to these goals, collaborations amongst researchers, and the knowledge gained that will lead to tools to improve coastal resilience. Herein, we first describe how researchers worked with stakeholders to structure their community-driven needs into science-based requirements. Next, we summarize how federal, academic, and stakeholder researchers worked together to design and execute a multi-organizational experiment aligned with those requirements. Finally, we articulate early findings and lessons learned from the experiment. This paper does not summarize all the research findings from DUNEX, as analyses are still ongoing. An American Geophysical Union (AGU) Special Collection on Coastal Storm Research will be published in 2025 including outcomes from DUNEX research.

North Carolina↗

USGS2021

Labeling poststorm coastal imagery for machine learning: Measurement of interrater agreement

Classifying images using supervised machine learning (ML) relies on labeled training data—classes or text descriptions, for example, associated with each image. Data-driven models are only as good as the data used for training, and this points to the importance of high-quality labeled data for developing a ML model that has predictive skill. Labeling data is typically a time-consuming, manual process. Here, we investigate the process of labeling data, with a specific focus on coastal aerial imagery captured in the wake of hurricanes that affected the Atlantic and Gulf Coasts of the United States. The imagery data set is a rich observational record of storm impacts and coastal change, but the imagery requires labeling to render that information accessible. We created an online interface that served labelers a stream of images and a fixed set of questions. A total of 1,600 images were labeled by at least two or as many as seven coastal scientists. We used the resulting data set to investigate interrater agreement: the extent to which labelers labeled each image similarly. Interrater agreement scores, assessed with percent agreement and Krippendorff's alpha, are higher when the questions posed to labelers are relatively simple, when the labelers are provided with a user manual, and when images are smaller. Experiments in interrater agreement point toward the benefit of multiple labelers for understanding the uncertainty in labeling data for machine learning research.

Earth and Space Science↗