1. MIT ICAT MIT ICAT

2. MIT ICAT MIT ICAT Motivation Adverse Weather Significantly Impacts Flight Operations Safety -- 22. 5% All US Accidents Efficiency -- 17% / $1.7B per year Avoidable Weather Delays (Source: FAA) Effervescence of Weather Information Technical Development Datalink There is a Need for a Coherent Approach for the Presentation of Hazardous Weather Information to Pilots and Other ATM Decision- Makers Turbulence & Clear Air Turbulence Thunderstorms & Microbursts In-Flight Icing (Courtesy of NASA)

3. MIT ICAT MIT ICAT Human-Centered Approach Closed Loop Feedback Process Aircraft Sensors Numerical Wx Models Flight Controls Direct Weather Observation Situation Dynamics Pilot Interaction Weather Aircraft State Displays Direct Aircraft State Observation Sensors Model Human Operator Displays Aviation Weather Information Remote Wx Sensors Sensor Readings Reports Fore- casts Voice Datalink In-Situ Wx Sensors Dissemination

4. MIT ICAT MIT ICAT New Weather Datalink Products ARNAV Vigyan Bendix/King FAA FISDL Avidyne Echo Flight Garmin Control Vision UPS – AirCell

5. MIT ICAT MIT ICAT Human-Centered Approach Closed Loop Feedback Process Aircraft Sensors Numerical Wx Models Flight Controls Direct Weather Observation Situation Dynamics Pilot Interaction Weather Aircraft State Displays Direct Aircraft State Observation Sensors Model Human Operator Displays Aviation Weather Information Remote Wx Sensors Sensor Readings Reports Fore- casts Voice Datalink In-Situ Wx Sensors Dissemination

6. MIT ICAT MIT ICAT Pilots’ Cognitive Tasks and Information Needs Pilots’ Cognitive Tasks  Identify a planned flight trajectory  Nominal  Alternative  Accept or reject a planned flight trajectory  Go/no-go  Continue/deviate  Fly to destination/alternate  Select resources for a flight trajectory (e.g., fuel)  Opt for legal fair-weather minimum or extra fuel How much extra fuel?  Manage systems along a flight trajectory  When to update weather information  When to use ice protection systems, seat-belt signs, flight attendant warnings Pilot weather information needs focus on the spatial distribution of hazardous weather conditions along 4-D aircraft trajectories  Reactive  Tactical  Strategic

7. MIT ICAT MIT ICAT TEMPORAL REGIMES OF WEATHER REPRESENTATION Historical Quasi-Constant Quasi-Deterministic Non-Deterministic 123 4 5 ? ReactiveTactical Strategic TIME TEMPORAL REGIMES OF FLIGHT PLANNING Future Past Temporal Regimes of Cognitive Processes

8. MIT ICAT MIT ICAT 4-D Trajectory Framework Aircraft Trajectory Modeled as a 4-D Aircraft Hypertube  3-D space + time  Specified variables: origin, destination, routing, ETD, ETE, cruising velocity Weather Field Modeled as Either: A Spatially Distributed and Temporally Varying Multi-Attribute Field  4D Gridded RUC-2 Data (20 km resolutions, 50 vertical levels) A Hazardous Weather Hypervolume  Assumption: hazardous weather condition thresholds identifiable Interaction Modeled as the 4-D Intersection of Hypervolumes Key Issues  How to represent time?  How to represent uncertainty? Protected Aircraft Hypertube Spatio-Temporal Multi-Attribute Field

9. MIT ICAT MIT ICAT Aircraft Hypertube Long. Latitude t Hazardous Weather Hypervolume Long. Latitude Altitude Hazardous Weather Space Propagated Over  t at Constant Altitude Illustration of 4-D Intersection Test (Quasi-Deterministic Regime) Aircraft Protected Airspace Propagated Over  at Constant Altitude

10. MIT ICAT MIT ICAT Pilots’ Perception of Forecast Accuracy Protected Aircraft Hypertube Hazardous Weather Hypervolume Prediction of 4-D Intersection* YN Occurrence of 4-D Intersection* YHitMiss N False Alarm Correct Rejection *Between Protected Aircraft Hypertube and Hazardous Weather Hypervolume CONTINGENCY MATRIX

11. MIT ICAT MIT ICAT Traditional Forecast Verification Methodology for  Hours Forecast Area Actual Adverse Weather Miss Hit False Alarm Correct Rejection Good forecast based on 4 contingencies above: Mean Square Error14% Critical Success Index73% Signal Detection Theory Hit Rate89% Signal Detection Theory False Alarm Rate17%

12. MIT ICAT MIT ICAT Sensitivity of 4-D Intersection to Estimated Time of Departure Actual Forecast Estimated Departure Time (Hrs) 4-D Intersection with Aircraft Hypertube No Yes 0 4 Correct Detection Correct Rejection False Alarm Forecast Area

13. MIT ICAT MIT ICAT Conclusions Limitations currently exist in the dissemination and representation of temporally varying weather information Weather datalink capabilities are improving the dissemination potential to pilots These limitations impair the ability to provide weather forecasts that can be perceived accurate by users of aviation weather information The 4-D trajectory framework serves as a basis for investigating across the flight deck, ATC and AOC environments: How to represent temporally varying information How to represent uncertainty in the adverse weather avoidance problem Continuing Work Advanced Visualization Techniques for 4-D Fields Trade-Offs in Representing Uncertainty in Quasi-Deterministic Regime Representation of Weather in Non-Deterministic Regime