2018 ESIP Winter Meeting

Introduction to the ESIP Lab followed by Tom Narock and Doug Fils presenting their funded Community PROV Challenge Fast Track work. After Tom and Doug present their results, they will  open the session to discuss next steps to improve provenance across agencies. 


Annie Burgess slides: https://docs.google.com/presentation/d/1-xHDgA42Mt_BFR0jt-RiyUTLkLsyja5f3Md47vXVt1w/edit?usp=sharing

The socioeconomic value and benefit of Earth science data, information, and knowledge (DIKW) could be better realized and more evident, when machine-facilitated methods are developed that allow for enhanced discovery and reuse of climate resilience solutions. These solutions are becoming more relevant, as the public becomes increasingly aware of the importance of “decreasing vulnerability by increasing resilience,” one of the objectives of NOAA’s Weather-Ready Nation initiative. The U.S. Climate Resilience Toolkit (US CRT) is an essential resource for documenting climate resilience use cases. The goal of the documentation is to develop a cohesive package of data, tools, and information, so that other communities can re-purpose the solutions contained in those use cases.


This session will explore a number of thought-experiments on “what-ifs,” applied to a hypothetical future version of the US CRT that is enabled by various technologies.
* What if we document traceable resilience solutions as knowledge graphs that can be used by automated reasoners to repurpose a resilience solution for other contexts (e.g., for another location, a related climate stressor, a related climate impact; using related observations, related models, related policy instrument, etc.)?
* What if we crowd-source the contribution of these knowledge graphs with the use of web-based editors that tap into existing ontologies to reduce duplication of concepts?
* What if we design a knowledge graph editor modeled after the Open Science Framework that implements essential open science concepts like attribution, citation, collaboration, unique identifiers, and project forking; from which we could generate metrics to assess the utility of data, information, and applications?
* What if we apply feature-extraction and machine learning algorithms to trawl the ecosystem of climate resilience knowledge graphs, to produce visualizations that could be used to answer user queries? Could the resultant metrics be used by federal agencies to prioritize and justify their investments?
To help guide these thought-experiments, the session will utilize products that have been generated out of the “Data to Decisions for Climate Resilience” and the “Agriculture and Climate” clusters.