Published in Ecological Informatics, 2022
This paper is a study of the landscape of interactive visual tools built to compare and analyze climate data based on research published from 1990-2021.
Published in Computing within Limits, 2023
This paper attends to five key tensions in this design space by demonstrating how design theories can reorient the design process for computing systems used in both top-down decision-making and bottom-up community effort. More specifically, we investigate how four design approaches (Social Design, Speculative Design, Feminist Data Practices, Design Justice) can be applied to a tool for empowering community efforts and informing decision-making for climate change.
Published in Environmental Modelling & Software, 2023
In this study, an analytical framework is presented that compares model-predicted values of climate parameters with their observed historic records. The resulting framework is a standardized methodology for operating climate data that can aid researchers and practitioners in performing comparison analysis.
Published in ACM Conference on Fairness, Accountability, and Transparency (FAccT), 2024
This paper examines data practices in machine learning dataset development through the lens of data curation. We develop a framework for evaluating machine learning datasets using data curation concepts and principles through a rubric. Through a mixed-methods analysis of evaluation results for 25 ML datasets, we study the feasibility of data curation principles to be adopted for machine learning data work in practice and explore how data curation is currently performed.
Published in Computing within Limits, 2024
This paper explores the role of alternative design approaches in developing visualizations used for climate policymaking. We conduct a literature review and synthesis which bridges psychological distance with speculative design and data visceralization by illustrating the value of affective design methods via examples from previous research.
Published in Advances in Neural Information Processing Systems (NeurIPS), 2024
This work provides an analysis of dataset development practices at NeurIPS through the lens of data curation. We present an evaluation framework for dataset documentation, consisting of a rubric and toolkit developed through a literature review of data curation principles. We use the framework to assess the strengths and weaknesses in current dataset development practices of 60 datasets published in the NeurIPS Datasets and Benchmarks track from 2021-2023.
Published in Eleventh Computing within Limits, 2025
This paper explores whether the World3-03 model is a feasible method to quantitatively simulate the impact of data centers on limits to growth. We reflect on these dynamics by adding new variables to the model in order to simulate a new AI-augmented scenario. We find that through our addition of AI-related variables (such as increasing data cen- ter development) impacting pollution in the World3-03 model, we can observe the expected changes to dynamics, demonstrating the viability of the World3-03 model for examining AI’s impact on planetary boundaries.