Chama Meeting 8/18/2023
Meeting Notes
Anne might need some Arduino support, short little office hours help at some point.
May Fly board with Arduino environment.
Conclusions section in our OneNote pages.
Anne's original thought of Ensemble was "multiple INP files".
- Ensemble of multiple SWMM runs results in a single Chama run.
- All plumes have the same probability.
Way to optimize memory to avoid the signal file generator issue.
Winter (sp) didn't have any issues.
Could be output timestep and run timestep difference? We've addressed that already.
We don't need to worry about DEM at this time and can assume flat ground.
Puff probably isn't a good fit. That's more for regular pulses of output?
They'll ask ___, a modeler, about what use case puff might be used for.
Plume makes the most sense.
Data visualization, such as showing on OL map, would be secondary priority. Anne is more interested in the end results of the opt, not necessarily looking at a map of the plume.
Office hours with Anne and Lifeng after I get an HPC test going on RHEL 8 (updated server).
Chama has only been run on HPC in the context of our work, when Lifeng ran it last batch. No one else in EPA is doing anything like this.
Winter hasn't been run on the HPC.
Writing SWMM INP file out with PySWMM or alternative Python package API
PySWMM doesn't currently have an export INP file option and they don't plan to add it to the new one. This addition would be a couple years down the road.