Visualizing Environmental Models

Project Summary

The Crossroad’s project: Visualizing Environmental Models (VEM) leveraged a unique combination of strengths at UCSB in environmental modeling, human-computer interface design, uncertainty quantification, and science-communication.  Environmental models are central tools for understanding the dynamic inter-relationship between people and nature - providing information that is critical for maintaining the health and well-being of both in an era of unprecedented global change. The complexity of these models however remains a barrier to their effectiveness, credibility and widespread use. The vision of the VEM team is to radically transform environmental models, though advances in human-computer interface design, uncertainty quantification, and science-communication. A key outcome of the Crossroads project was the formation of the VEM team and the development of new research proposals that will continue to evolve the ideas that emerged through year long weekly meetings with PhD students, faculty and researchers. Along with much cross disciplinary learning, two project ideas emerged – one seeks to transform coarse grained output from a forest ecosystem carbon cycling model into visually familiar forest structures, using high resolution remote sensing data and machine learning.  A second project developed a statistical emulator version of RHESSYs, an ecohydrologic model. This emulator radically expanded feasible RHESSys applications to a much larger study domains and scenario possibilities.  A working version of this emulator has been completed for Mission Creek, a chaparral dominated watershed near Santa Barbara, and Sagehen Creek, a snow dominated watershed in the California Sierra.  A research paper on the emulator is forthcoming and results integrated into two NSF proposals.

Faculty Investigators: Christina (Naomi) Tague (Ecohydrologist, Environmental Science & Management), James Frew (Ecohydrologist Information scientist, Environmental Science & Management), Tobias Hollerer (Computer Science), John O'Donovan (Computer Science)

Fellows: Willam Burke (Environmental Science & Management), Rachel Torres (Environmental Science & Management), Ehsan Sayyad (Computer Science), Yi Ding (Computer Science)

Faculty: Wendy Meiring (Statistics & Applied Probability)