Overview of the 2017 PHI Prototype Experiment with NWS Forecasters
Christopher Karstens, Cooperative Institute for Mesoscale Meteorological Studies (CIMMS) and National Severe Storms Laboratory (NSSL), Norman, OK
Alan Gerard, National Severe Storms Laboratory (NSSL)
Daphne LaDue, Center for Analysis and Prediction of Storms (CAPS)
James Correia, Jr., Cooperative Institute for Mesoscale Meteorological Studies (CIMMS) and Storm Prediction Center (SPC)
Kristin Calhoun, Cooperative Institute for Mesoscale Meteorological Studies (CIMMS) and National Severe Storms Laboratory (NSSL)

Abstract
This presentation summarizes the 2017 Hazardous Weather Testbed (HWT) Probabilistic Hazard Information (PHI) Prototype experiment with National Weather Service (NWS) forecasters, an effort that is part of Forecasting a Continuum of Environmental Threats (FACETs).  Forecasters worked 3 displaced real-time and 3 real-time severe weather events each week of the 3 week experiment occurring this past May and June.  Each day, forecasters rotated responsibility for issuing and updating PHI for a single hazard type (tornado, severe, or lightning) for which various forms of rapidly-updating automated object-based guidance were utilized. 

Automated guidance for severe included NOAA/CIMSS ProbSevere objects, a real-time statistical model for calculating the probability of a storm producing any severe hazard in the next 60 minutes, with embedded machine learning algorithms for 1) predicting storm duration, 2) predicting storm classification, 3) predicting straight-line damaging winds, and 4) correcting unjustified automated tracking breakages.  Automated guidance for tornado included the NSSL Experimental Warn-on-Forecast System for Ensembles (NEWS-e) and NOAA/CIMSS ProbTor, a sub-component of ProbSevere.  Automated guidance for lightning included early-stage algorithm development occurring at CIMMS/NSSL.

The over-arching objective of this experiment focused on understanding the mechanisms and situations involved a continuous flow of information (CFOI).  This concept was developed out of HWT experimentation involving forecasters and users (Emergency Managers and Broadcasters).  In addition to summarizing experimentation logistics, this presentation highlights situations when CFOI was apparent and the relative meaning of CFOI in the context of evolving the NWS warning paradigm.