aappling-usgs
commented
6 years ago - using fake "analysis" data (stand-in = forecasts with lag 0) for the 6 months preceding the forecast
- final flow data may be slightly different and/or longer
- full batch ran in just under 24 hours, no errors or retries