Gilberto Pastorello

Biographical Sketch

Gilberto Pastorello is a research scientist in the Usable Data Systems Group at the Lawrence Berkeley National Laboratory. His research is focused on managing the life-cycle of scientific data, encompassing data and metadata structures and linkages, data quality and data uncertainty quantification, and end-to-end data systems. He is part of the data teams for the AmeriFlux and FLUXNET networks (with the AmeriFlux Management Project) and the data team for the NGEE-Tropics project. He implemented new methods for data integration at multiple stages of processing, data quality assurance for heterogeneous data sources, and algorithms for execution and evaluation of data processing pipelines for environmental data -- in particular carbon, water, and energy fluxes and micro-meteorological data. He is part of the team creating the FLUXNET datasets, which are used in research ranging from soil microbiology to climate change. He has also developed data models and data behavior models for observational data, allowing more complete characterization of data quality and improved evaluation of data processing algorithms. He has also contributed to the Deduce project, investigating data change dynamics, and to the Tigres project, investigating workflow specification APIs and workflow patterns for data intensive pipelines. Before joining LBNL as a Post Doctoral Fellow in March 2013, he held post doctoral positions at the University of Alberta, in Edmonton, AB, Canada, where he implemented end-to-end data management solutions for biodiversity data and micro-meteorological data from traditional and wireless sensor networks systems; created data processing pipelines for optical data acquired from a variety of platforms (including, for instance, ground-based observations and airborne imaging spectrometry); and, created Web-based data exploration portals. He holds PhD, MSc, and BSc degrees in Computer Science from the University of Campinas, in Campinas, SP, Brazil.