We want to validate the power/storm workflow against historic storms.
It's proposed to do this with nightlight imagery, where the diff between two overpasses before and after the event combined with population data might give an estimate as to the number of electricity customers affected by a given storm.
Yu will code a validation method as above, but requires certain inputs to proceed. For several historic storms:
1) Storm track (a series of storm eye locations and times)
2) Location and number of customers modelled as disconnected
These data should be produced in a format that's easy to process for Yu.
We want to validate the power/storm workflow against historic storms.
It's proposed to do this with nightlight imagery, where the diff between two overpasses before and after the event combined with population data might give an estimate as to the number of electricity customers affected by a given storm.
Yu will code a validation method as above, but requires certain inputs to proceed. For several historic storms: 1) Storm track (a series of storm eye locations and times) 2) Location and number of customers modelled as disconnected
These data should be produced in a format that's easy to process for Yu.