Proposal Description

Large and complex environmental datasets often have interesting geometric structures, which are often elicited quickly and efficiently using topological data analysis and related techniques in geometric statistics. In this session, three speakers will talk on some modern and path breaking advancements that couple topological data analysis with more classical statistical techniques like survival analysis and resampling, to perform accurate and precise inference on large environmental datasets. The applications range from public health and epidemiological issues rising from environmental issues to remote sensing. The session will also include a discussion of the talks by an expert on both climate sciences and machine learning, who will provide critical evaluation of the proposed methodologies contained in the talks, and propose open questions and challenges. The session has a balance of speakers from different races, ethnicities, gender and specializations, and will be of interest to scientists and statisticians interested in both classical approaches and modern data science techniques for analyzing complex spatio-temporal data that arise in environmental applications.