Tech Talk
summary
Descriptions of long-running earth science data work: a decade-plus effort on AmeriFlux and a host of Department of Energy projects (NG-Tropics, watershed microbiomes, a central environmental data repository) that takes over a messy distributed data pipeline and standardizes identifiers, QA/QC and versioning. Progress is slow—14 out of 43 women in one data group and 14 years on AmeriFlux—but machine learning models to forecast groundwater under climate scenarios show early promise rather than ready solutions.
AI-written summary · about
chapters (16)
- 0:00Introduction by host
- 0:25Speaker introduction
- 1:07National Labs vs academia
- 2:14Department overview
- 3:15MEFluX carbon flux project
- 4:00Watershed nutrient cycle study
- 5:01Data repository for environmental science
- 5:43Earth science data challenges
- 7:10AmeriFlux project journey
- 8:55Data pipeline ownership impact
- 9:17Machine learning for groundwater prediction
- 10:22Reality of QA/QC in data
- 11:29Data producer vs user perspectives
- 12:15Data integration complexities
- 12:36Work under construction
- 13:19CRA-WP workshops plug
related talks
31mVirtual Laboratories Vision for an Earth Sciences
24mDigital Earth Pacific Cloud Native Geospatial Infrastructure for Earth Observation
46mData Discussion - AI for Good
02mThe US EPA: Digital transformation through data science
15mThe tech is there: society needs to step up
22mDevelopment and Implementation of Historical and Current Climate Mapping for Vanuatu
25mEmpowering weather-satellite data users with FOSS: the Pytroll case
38mThe Real World of Data Analytics