Updating the update: After further testing I’m now splitting the circle radius down the middle at 35 mi. I don’t know, I’ll probably change it again. The sensitivity seems to be about right based on cells that went through this morning. Email recipients can expect 1-3 emails depending on the sequence of cells moving through the zone. Inter-email time is set at 60 minutes, meaning once you get an email, you won’t get another for an hour, and not at all if conditions downgrade, so check the radar page if you have concerns about passing storms once you’ve been emailed. The timing is set in consideration of those who are not in the path of active weather when the emails hit.

Update: After extensive testing and calibration, I’m back to 40 mi radius. The sensitivity trip point has been raised to a 60 dBZ level.

Last night’s cold front passage was a trial by fire of sorts for EWR. If you were on the receiving end of a blizzard of alerts, I apologize for that. We are currently testing some very early new pre-releases of the next version of the program software, as well as fine-tuning the sensitivities of the system, and the combination is turning out to be a handful. Presently there are 3 computers running 4 versions of the software (no, not all at the same time!). Ephemerata Weather Radar currently uses two computers to generate the web updates when both scans are up. The goal eventually is to do that from one machine.

One thing I’ve done is to reduce the scan area from 40 mi. radius to a 30 mi. radius. With every new data update from the NWS, an alert is generated if conditions within the scan zone are met. Since it can take some time for a system to move through the zone, the potential exists for multiple alerts from the same storm pattern. The software has two different time loops to modulate the alert frequency and it will take some time to get these in sync. The target is for a maximum of 3 alerts over the course of a storm event moving through the zone.