Aircraft Icing Weather Data Reporting and Dissemination System

Aircraft Icing Weather Data Reporting and Dissemination System

Ellen J. Bass and Brian Minsk Search Technology, Inc.

Norcross, Georgia 30071 Tenny Lindholm and Marcia Politovich National Center for Atmospheric Research Boulder, Colorado 80301

Abstract

The long-term operational concept of this research is to develop an onboard aircraft system that assesses and reports atmospheric icing conditions automatically and in a timely manner in order to improve aviation safety and the efficiency of aircraft operations via improved real-time and forecast weather products. The idea is to use current measurement capabilities on aircraft equipped with icing sensors and in-flight data communication technologies as a reporting source. Without requiring expensive avionics upgrades, aircraft data must be processed and available for downlink.

Ideally, the data from multiple aircraft can then be integrated (along with other real-time and modeled data) on the ground such that aviation-centered icing hazard metrics for volumes of airspace can be assessed. As the effect of icing on different aircraft types can vary, the information should be displayed in meaningful ways such that multiple types of users can understand the information. That is, information must be presented in a manner to allow users to understand the icing conditions with respect to individual concerns and aircraft capabilities.

This research provides progress toward this operational concept by: • Identifying an aircraft platform capable of digitally capturing, processing and downlinking icing data, • Identifying the required in situ icing data processing, • Investigating the requirements for routing the icing data for use by weather products, • Developing an icing case study in order to gain insight into major air carrier needs, • Developing and prototyping icing display concepts based on the National Center for Atmospheric Research’s existing diagnostic and forecast experimental icing products, and • Conducting a usability study for the prototyped icing display concepts.

1 NASA/CR—2002-211800

Acronyms and Abbreviations

A Airbus AWIN Aviation Weather Information AO1 Automated station without AZ Arizona precipitation discriminator AZM Azimuth AO2 Automated station with precipitation B Began discriminator Also Boeing ABQ Albuquerque B190 Beech 1900 Airliner A/C Aircraft BE Beech AC90 Rockwell 690 Turbo Commander BE19 Beech B-19 Musketeer ACARS Aircraft Communications Addressing BE20 Beech 200 Super King Air, C-12 A to and Reporting System F ACCUM Accumulation BE55 Beech E-55 Baron ACCUMN Accumulation BFG BFGoodrich ACFT Aircraft BKN Broken ACID Aircraft identifier BL Blowing ACMF Aircraft Condition Monitoring BLO Below Function BLW Below ACSL Altocumulus standing lenticular cloud BR Mist (visibility 5/8-6 SM) ACT Activity BRF Brief ACTL Actual BTN Between ACRS Across BTR Better AD Airworthiness Directive BTWN Between ADDS Aviation Digital Data Service BYD Beyond ADF Airline Dispatchers Federation C Celsius ADJ Adjacent Also Cessna ADNL Additional C177 Cessna 177, Cardinal AFD Airport/Facility Directory C206 Cessna 206 series AFTN Afternoon C550 Cessna 550/552 Citation 2 AGL Above ground level CA California ALG Along CARJ Canadair Regional Jet ALS Approach Light Systems CAS Calibrated airspeed ALT Altitude CAT Category ALTMR Altimeter Also clear air turbulence ALTN Alternate CAVOK Ceiling and visibility OK AIRMET Airman’s Meteorological Information CB Cumulonimbus ALG Along CDFNT Cold front AMDT Amendment CDL Configuration Deviation List AMS Air mass CHG Change ARINC Aeronautical Radio, Inc. CI Cost index ARVL Arrival CIG Ceiling ASA Aviation Supplies & Academics CLB Climb ASSOCD Associated CLD Cloud Also cold ATC Air traffic control CLDS Clouds ATIS Automatic Terminal Information System CLR Clear ATL Atlanta CLSD Closed AVG Average cm Centimeter AWABS Aircraft Weight and Balance (and CNTNGNCY Contingency Takeoff Performance) Data CNVTN Convection AWC Aviation Weather Center CO Colorado 2 NASA/CR—2002-211800 Also company ENRTE En route CONDS Conditions ER Extended range CONT Continuous ERN Eastern Also Continue ETA Estimated time of arrival CONTG Continuing ETE Estimated time en route CONUS Continental United States EWD Eastward COORD Coordinates EXP Expected COR Correction (to a previously EXPD Expected disseminated report) F2TH Falcon 2000 COS Colorado Springs FA Area forecast Also because Also final approach CPR Casper Also flight attendant CR Carriage return FAA Federal Aviation Administration CRZ Cruise FAP Final approach CSTL Coastal FAR Federal Aviation Regulation CTR Center FC Funnel cloud CTRLINE Center line FDC Flight Data Center CVG Cincinnati FG Fog (visibility < 5/8 SM) CYS Cheyenne FL Flight level CZ Cruise Also Florida D328 Fairchild Dornier 328 FLT Flight DAL Delta Air Lines FM From DCT Direct Also flight mode DEN Denver FMS Flight Management System DEP Departure FOB Fuel on board DEPT Departure FOD Foreign object debris DEST Destination FOM Flight Operations Manual DH Decision Height FP Flight plan DIR Direction FRMG Fuel remaining to destination DLAS Delays FT Feet DME Distance measuring equipment FU Smoke DOM Domestic G Gusts DRZL Drizzle GLF4 G-1159C Gulfstream 4 DS Dust storm FZ Freezing DSNT Descent GA Georgia Also Distant GJT Grand Junction DU Dust GMT Greenwich Mean Time DURC During climb GOES Global Orbiting Earth Satellite DURD During descent GPS Global Positioning System DURGC During climb GR Hail >= ¼ inch DURGD During descent GRD Ground DVER Database version GRTR Greater DVLPG Developing GS Ground speed Also small hail DZ Drizzle GST Gust E East Also ended HAT Height Above Touchdown ECN Economy HF High Frequency ECON Economy flight plan based on HVY Heavy minimum cost for FMS equipped HZ Haze aircraft IAF Initial approach fix EEC Engine Electronic Control IAS Indicated airspeed Eff Effective IC Icing ELEV Elevation Also ice crystals EMBDD Embedded Also in-cloud lightning 3 NASA/CR—2002-211800 ICAO International Civil Aviation MEL Minimum Equipment List Organization MEM Memphis ICGICIP Icing in clouds and in precipitation METAR Aviation Routine Weather Report ICGIP Icing in precipitation M/H Magnetic heading ID Identifier MI Mile IFR Instrument flight rules MIN Minimum IIDA Integrated Icing Diagnostic Algorithm MNLY Mainly IIFA Integrated Icing Forecast Algorithm MOD Moderate ILS Instrument Landing System MOGR Moderate or greater IMC Instrument meteorological conditions MOPS Minimum Operational Performance IN Inch Standards INCLDS Includes MOV Moving INCR Increasing Also movement INOP Inoperative MOVG Moving INT Intersection MPTW Maximum planned take off weight INTX Intersection MS Mail stop INTL International MSL Mean sea level JTSTR Jet stream MT Mountain K Kilo MTN Mountain KIAS Knots indicated airspeed MXD Mixed KM Kilometer NASA National Aeronautics and Space Administration KT Knots NAV Navigation L Left NC North Carolina Also local time Also TCAS-equipped B757 NCAR National Center for Atmospheric LAT Latitude Research Lb Pound N North LF Line feed NBR Number LGT Light NDB Non-directional beacon LGTS Lights NEG Negligible LJ35 Learjet 35/36 NM Nautical miles LMT Limit NMI Nautical miles LN Line NMRS Numerous LNDG Landing NOAA National Oceanic and Atmospheric Administration LOC Localizer NOSPECI No SPECI reports are taken at the LON Longitude station Long Longitude NOTAM Notice to Airmen LT Light NRP National Route Program LVL Level NTSB National Transportation Safety Board LWC Liquid water content NV Nevada LWR Lower NWP Numerical weather prediction LWT Planned landing weight NWS National Weather Service M Minus NXT Next Also Mach OCLN Occlusion Also million OCNL Occasional mA milliampere OCR Occur MAX Maximum OFFSHR Offshore MCH Mach OG On ground MD McDonnell Douglas OKC Oklahoma City MDA Minimum Descent Altitude OLCP Occasional light chop MDCRS Meteorological Data Commercial Reporting System OM Outer Marker MDT Moderate OMTNS Over mountains MEA Minimum En route Altitude OR Oregon 4 NASA/CR—2002-211800 OTS Out of service SKC Sky clear OV Over SKED Schedule OVC Overcast SLC Salt Lake City OVR Over SLD Supercooled Liquid Droplet P Precipitation SLP Sea level pressure Also plus SLPG Sloping PA31 Embraer 820 SM Statute mile PAY1 PA-31T1-500 Cheyenne 1 SMTH Smooth PD Period SN Snow PE Ice pellets SPD Speed PG Page SPECI Special PK Peak SQ Squall PIREP Pilot report SS Sand storm PLND Planned STG Strong POS Position STN Station PR Partial SW Southwest PREF Preferred T Temperature PRESS Pressure TA Temperature PRESRR Pressure rising rapidly TACAN Tactical Air Navigation PSBL Possible TAF Aerodrome Forecast PUB Pueblo TAS True airspeed o Also published TAT Total air temperature C PWR Target power setting TB Turbulence PY Spray TBD To be deterrmined R Reporting point (to company) TC True course Also right TCU Towering cumulus RA Rain TEM Temperature deviation from standard RAOB Radiosonde observation TEMP Temperature RCMND Recommend Also temporary RCV Receive TEMPO Occasionally RLS Release THLD Threshold RM Remark Turbulence Indicator TIND RMK Remark TM Time RMN Remain TMP Temperature RPT Report T/O T/O RPTD Reported TOC Top of climb RTE Route TOD Top of descent RUC Rapid update cycle TP Type RVR Runway Visual Range TRB Turbulence RWY Runway TRFC Traffic S South TRG Trigger SAT Static air temperature TRMG Time remaining to destination SC South Carolina TROF Trough Also South Central TROP Tropopause altitude SCAT Scattered TRW Thunderstorm SCT Scattered TS Thunderstorm SE South East TSHWR Thunder shower sec Second TSTM Thunderstorm SEV Severe TUL Tulsa SFC Surface TURB Turbulence SG Snow grains TURBC Turbulence SH Showers TWR Tower SIGMET Significant Meteorological UA Pilot report Information UKN Unknown 5 NASA/CR—2002-211800 UNKN Unknown UNUSBL Unusable UP Automated observation UPDT Update UPSLP Up slope UT Utah UTC Coordinated Universal Time V Variable VA Volcanic ash Var Variation VC Vicinity VFR Visual Flight Rules VHF Very High Frequency VIS Visibility VOR Very High Frequency Omni- directional Range VRBL Variable VV Vertical visibility (into a total obscuration listed in hundreds of feet AGL) W West WA AIRMET Also Washington WCP Average wind component (M = headwind; P = tailwind) WD Wind direction WDLY Widely WND Wind WNDSHR Wind shear WS Wind shear Also wind speed WST Convective Significant Meteorological Information WT Weight WTRS Waters WV Wave Also wind vector Wx Weather WY Wyoming XPCT Expect Z Zulu time ZD Zone distance ZDV Denver Air Route Traffic Control Center ZF Zone fuel ZT Zone time; Coordinated Universal Time 6 NASA/CR—2002-211800

Introduction

Types and Severity of Icing Accretion of ice on aircraft surfaces in flight is a result of cloud droplets remaining in a liquid state at temperatures below freezing. The severity of icing (categorized in terms of trace, light, moderate and severe [ASA, 1999]) is generally dependent upon the accretion rate. The amount, type and shape of ice accreted is dependent on several variables (Jeck, 1996; Pobanz, Marwitz & Politovich, 1994; Politovich, 1989; Ryerson, 2000) such as: • Aircraft airspeed (e.g., increased airspeed means the droplets have less time to flow around an object and the surface of the airfoil is heated by friction), • Aircraft type (e.g., size and shape of objects affect collection efficiency and accumulation), • Cloud phase (supercooled liquid water freezes on aircraft structures while ice crystals do not), • Droplet size (droplet size affects collection rates, ice shape and type, and runback), • Duration in the icing (given more time, more ice can form), • Liquid water content (LWC) (icing occurs at particular LWC depending on factors such as cloud type and altitude), • Temperature (temperature affects the type and location of ice on the airframe; the rate of ice accumulation is directly related to LWC for a given temperature), and • Wind shear (disturbances can cause icing to initiate).

Of all these variables, the more diagnostic ones are currently least accessible. LWC is probably the most important in determining the severity of the icing conditions (Politovich, 1989).

Unfortunately, without access to the output of specialized sensors aboard aircraft flying through the area, it is difficult to determine LWC. Another major factor in icing accretion is droplet size.

However, individual droplets capable of producing structural ice are too small to be seen through a cockpit window. Droplet size is also difficult to determine with the instrumentation currently installed on aircraft.

Problems Associated with Icing Problems associated with icing can be grouped into two main categories: 1.) safety and accident rates and 2.) increased operating cost.

Safety and Accident Rates Icing remains one of aviation's leading hazards (c.f., Boeing, 2001; NASA, 1998; NTSB, 1996, 1998). Weather conditions are never totally predictable and icing forecasts are not provided with the temporal and spatial accuracy and timeliness to help pilots avoid hazardous icing encounters.

The effects of icing on an aircraft are aircraft specific and have been found to affect various aircraft components and systems including: • Modification of the airflow pattern, leading to loss of lift or an increase in drag, • Loss in engine power, 7 NASA/CR—2002-211800 • Loss in propeller efficiency, • Increase in weight, • Unbalancing of the control surfaces, • Errors in the instruments if the pitot tube or static vent are blocked, • Degradation of radio communication, and • Degradation of visibility through the windshield.

All aircraft are susceptible to icing — even those with anti-icing equipment. Most aircraft involved in icing accidents are general aviation type aircraft, but there is a significant number of larger, commercial aircraft that have been involved in icing accidents (c.f., Aviation Safety Network, 1998; Boeing, 2001; NASA, 1998). Apparently, experienced pilots are not taking corrective action when icing conditions are encountered. This may suggest a lack of understanding of the seriousness of an icing encounter on the performance of the aircraft and/or a lack of understanding of the weather factors contributing to icing conditions. It may also suggest that pilots are unable to visually detect the ice accretion until aircraft performance noticeably degrades. These pilots most likely lacked adequate awareness of the nature and severity of the icing problem.

Complicating the icing safety issue is that both structural and engine icing must be considered. For example, an aircraft with an inoperative wing anti-ice valve must consider the potential for “hard ice” while one with an inoperative engine anti-ice valve must consider “soft ice” (Myszkowski & Rezsonya, 1996). “Hard” ice occurs between 0 and -40 ° C in visible moisture, and where “soft” ice (ice that can form once air is cooled) can occur if the temperature is between 10 and -20 ° C and when the humidity is high. Thus, the corrective action varies depending on the type of operative equipment.

Several factors suggest that the potential for hazardous icing encounters will continue.

Aircraft designs continue to include features that make aircraft susceptible to icing such as laminar flow airfoils and efficient engines intolerant of contamination (Ryerson, 2000). As air traffic increases and new air traffic route structures are created, aircraft may be increasingly exposed to icing conditions. For example, the use of lower en route cruise altitudes, necessary to accommodate the increase in air traffic, may expose aircraft to icing conditions for longer periods than previous higher altitude routings did (RTCA, 1995).

Increased Operating Costs Compensating for areas of anticipated or encountered icing yields disruptions in planned altitudes and/or routing, significantly decreasing aircraft efficiency and therefore increasing operating costs. Three areas in which icing significantly increases operating costs are: 1. Remaining at an altitude or on a course for too long, given the icing conditions : Because pilots may not have timely forecasts or may be relying on the inadequate subjective assessments of other pilots of icing conditions enroute, some may “ride out” an icing encounter for too long. This could be both dangerous and costly in terms of fuel reserves.

However, if these pilots were to make a more timely decision to avoid the area of icing (based on more precise icing information), resources would be used more effectively.

Typically, though, most pilots aggressively seek to avoid icing encounters believed to be beyond the capability of the aircraft.

2. Prematurely vacating an optimum altitude or course based on reports of icing ahead : Once a flight receives reports of or encounters unexpected icing, a pilot (with the help of a flight dispatcher for airline operations) must assess the current severity, solicit additional route and 8 NASA/CR—2002-211800 altitude information regarding icing from air traffic control (ATC), and calculate fuel reserves and time constraints to determine the best course of action. Because reports of icing are subjective, based on the type of aircraft and the experience and priorities of the crew making the report, the variability of the reports is considerable. Pilots have little or no objective data on which to base a decision. They often err on the conservative side and request a change in altitude or course.

3. Avoiding areas of forecasted icing that actually would not adversely affect the flight : Forecasting icing is difficult. Further complicating this situation is that different aircraft experience the same icing conditions differently. What may be “moderate” for one aircraft type may only be “light” for another aircraft. Reports and forecasts should account for the specifics of the particular aircraft type, but this information may not always be available.

Despite oftentimes inaccurate forecasts, it is the responsibility of the Pilot-in-Command (and Flight Dispatcher for airline operations) to assess the planned route of flight and make adjustments as necessary in an attempt to avoid “significant” icing. In some cases, flights are delayed or are cancelled altogether. In other cases, these adjustments mean flying several thousand feet above or below the optimum altitude, or they may require flying a circuitous route around the area of forecast icing. These preflight planning adjustments may require increased fuel loads and/or longer flight times. These adjustments are expensive because they either add to the operating costs or lower the revenues.

Problems Associated with the Current Icing Reporting System As previously discussed, there are significant costs associated with icing encounters. Many of these problems are exacerbated by the sporadic, subjective, and imprecise way in which icing is currently measured and reported (ASA, 1999; Kelsch & Wharton, 1996). Many aircraft are not instrumented to provide the pilot with any more data than is visually detectable through a cockpit window. The current reporting system suffers from several shortcomings that are discussed next.

Pilot assessment is subjective Currently, icing is categorized and reported using a subjective system that requires the crew to assess the type of ice being accumulated, and then determine the aircraft’s reaction to it in terms of performance. The type and amount of ice are left to the “eyes of the beholder.” Each pilot makes his own judgment about the severity of weather events. New and low time pilots are known to overestimate the intensity of icing (Lankford, 1995). Additionally, current approved report terminology is too subjective to provide accurate descriptions of icing conditions. A related problem is that pilots are trained to report ice in terms of observable phenomena that are not perfectly diagnostic (Lankford, 2000) (Table 1 and Table 2). Ice types are more a function of ice accretion shape, rather than color or opacity, yet pilots are not trained accordingly.

Table 1. Types of icing Rime ice Rough, milky, opaque ice formed by the instantaneous freezing of small supercooled water droplets. This generally occurs in stratiform clouds at temperatures between 0 and -20 ° C.

Clear ice A glossy, clear or translucent ice formed by the relatively slow freezing of large supercooled water droplets. This generally occurs in cumulus clouds or freezing precipitation between 0 and -40 ° C.

Mixed ice A combination of rime and clear ice.

9 NASA/CR—2002-211800 Table 2. Intensity of icing and required actions Category Description Required action Trace Ice becomes perceptible. Rate of De/anti-icing equipment is not utilized unless accumulation is slightly greater than the rate encountered for an extended period of time (over 1 of sublimation. hour).

Light The rate of accumulation may create a Use of de/anti-icing equipment removes/prevents problem if the flight is over one hour in this accumulation. Without icing prevention equipment, environment.

one should consider a change of course or a 180 ° turn.

Moderate The rate of accumulation is such that even Use of de/anti-icing equipment or flight diversion is short encounters become potentially necessary. Light single and twin airplanes may not hazardous. be able to climb through this type of icing.

Severe The rate of accumulation is such that the Use de/anti-icing equipment. Immediate flight de/anti-icing equipment fails to reduce or diversion is necessary.

control the hazard.

Forecasts and reports are not aircraft-specific Different types of aircraft have different sensitivities to icing. Leading edge radius of curvature, wing surface area, wing sweep angle, and typical operating altitudes and airspeeds affect in-flight icing accumulation. Thus, a report of “moderate” icing by one aircraft may not be reported by another flying through the same area.

Reports are given voluntarily and are not generally available Pilots are urged to volunteer reports of icing conditions. Federal Aviation Administration (FAA) air traffic facilities are required to solicit reports under certain conditions (ASA, 1999).

However, the lack of PIREP availability is well known. PIREP availability is determined by factors such as pilot and controller attention and workload. PIREPs, when given, are not evenly distributed in either time or space (Kelsch & Wharton, 1996; Schwartz, 1996). In addition, there are very few PIREPs that report good conditions (Schwartz, 1996). Exacerbating the problem is that, even when PIREPs are made, they are not available to all airspace users. Icing conditions reported to busy air traffic controllers may only be passed along verbally to other pilots in the sector and may be delayed (Hansman & Wanke, 1989). Also, unless entered by someone, the data is lost.

Reports may be given to the company verbally or electronically but are not routinely available to others. Many U.S. carriers have aircraft equipped to downlink icing reports via the Aircraft Communications Addressing and Recording System (ACARS), a line-of-sight VHF telecommunication system. This data is maintained at each company and is not always shared. In order to facilitate the sharing of this information, the Meteorological Data Commercial Reporting System (MDCRS) Working Group (an informal advisory group) is dealing with issues like reporting frequency, data formats, and cost reimbursement. However, there are currently few participating airlines. In addition, airlines wish to keep down communication costs and to avoid expensive changes to their aircraft. These concerns are addressed more specifically later in this report.

Aircraft participating in sharing weather data provide latitude, longitude, altitude, time, temperature, and wind direction and speed. Some report vertical acceleration (an indirect measure of turbulence), and a handful experimentally report eddy dissipation rate (an aircraft-independent measure of turbulence) (Cornman & Sharman, 1999). In the future, a few aircraft may provide experimental dewpoint data. MDCRS is considering adding icing data (i.e., a binary ice detection parameter) depending on the associated costs, some of which are addressed later in this report.

10 NASA/CR—2002-211800 Pilots may not be able to see ice Ice accretions that affect the stability and control of the airplane may be very small and rather unspectacular in appearance (Green, 1998). Because of some aircraft designs, pilots may not physically be able to see enough of the wing surface to make an icing determination. There are also times when ice may form in such a manner that the pilots cannot visually detect its presence. Often the autopilot masks the disturbance, so the crew is unaware of the icing until the autopilot “gives up” and hands the aircraft back to the pilot with a serious control problem (Green, 1998).

The FAA has recognized the potential hazard of aircraft icing and has written regulations concerning aircraft operation in icing conditions as well as non-regulatory guidelines. For example, Airworthiness Directives (ADs) currently limit specific turbo-prop aircraft from flight in freezing rain or freezing drizzle based on pilot-observed visual cues. These visual cues include (FAA, 1996): • Unusually extensive ice accreted on the airframe in areas not normally observed to collect ice, • Accumulation of ice on the upper surface (for low wing aircraft) or lower surface (for high wing aircraft) of the wing aft of the protected area, and • Accumulation of ice on the propeller spinner farther back than normally observed.

Even with regulations and guidelines, pilots are still experiencing roll upsets, loss of control and accidents because they cannot identify these cues. Furthermore, pilots of aircraft operating at lower altitudes are more likely to encounter icing conditions. It is difficult, then, for these pilots who operate in such conditions on a regular basis to identify frequently experienced conditions as something “unusual.” New or low time pilots obviously have difficulty identifying situations as unusual. All pilots, especially those operating at lower altitudes, need more timely, objective information with regard to wing contamination and icing conditions.

In addition to visual cues on the surface of the aircraft, pilots sometimes use precipitation as a cue for icing conditions. Unfortunately though, the icing conditions typically occur in clouds that do not produce rain or snow on the ground. A particularly hazardous form of icing occurs when o aircraft encounter supercooled liquid water (temperatures below 0 C) in the drizzle drop size (30- 300 microns diameter) and high relative humidity (Pobanz, Marwitz & Politovich, 1994; Politovich, 1989). However, in order for these drizzle drops to form, the atmosphere must be undergoing upward vertical motion at slightly faster speeds than the large-scale lifting that forms large cloud masses (about 1 cm/sec). Too much upward vertical motion and water drops larger than drizzle-size quickly develop. The optimum vertical motion is on the order of 10 cm/sec. This condition occurs frequently at cloud tops, and is a good explanation of why significant icing is often observed there.

Weather system measurement is imprecise The density, frequency and resolution capability of today’s observation network is incompatible with the micro-scale nature of icing. Temperature, moisture, and wind data from radiosondes are taken twice a day at stations averaging hundreds of miles apart, with a vertical resolution around 2,000 feet. Thus, these radiosonde observations (RAOBs) provide a sparse sampling of the environment. Forecasting issues arise, as predictions have to span over twelve hours. The problem is made worse at times when an upper air reporting station is missing data forcing extrapolation over a “hole” in the sampling grid. Sparse data sampling requires the forecasters’ computer algorithms to smooth the prediction models vertically and horizontally to 11 NASA/CR—2002-211800 achieve a computational answer in a reasonable amount of time. In addition, there are some areas that do not have the equipment necessary to forecast icing (i.e., the far North) (Ryerson, 2000).

Existing Research Efforts to Solve These Problems Downlink of icing data The FAA is investigating methods for acquiring icing information from aircraft. The FAA’s icing sensor downlink approach is a method with an expected low implementation cost (Riley, Lindholm, Politovich, Brown, and Strapp, 1999).

Enhanced weather products A number of aircraft icing diagnostic algorithms have been developed in the past several years (none of which currently use quantitative in situ icing measurements from commercial aircraft). These algorithms have used various inputs such as: pilot reports, meteorological models, satellites, surface observations and radar mosaics (c.f., McDonough & Bernstein, 1999). These algorithms have used the input data in a variety of ways and have met with some success.

Unfortunately, algorithms based purely on models tend to overforecast icing by indicating it in locations where clouds do not exist. Algorithms based primarily on data from instruments (satellite, radar, surface observations) or pilot reports tend to be accurate in the locations where they indicate icing, but they underforecast icing because none of these instruments can identify all icing locations by themselves.

NCAR’s Integrated Icing Diagnostic Algorithm (IIDA) takes advantage of the abilities and minimizes the shortcomings of both the model-based and instrument-based approaches (McDonough & Bernstein, 1999). IIDA is run every hour to create a diagnostic based on relatively recent pilot reports, satellite data, surface observations, radar mosaics, and the RUC (rapid update cycle) model. Pilot reports less than an hour old are considered. The satellite data is generally less than 45 minutes old. The surface observations are 5-10 minutes old and the radar data is also very current. The RUC model is generated every three hours to create forecasts for the next three hours and therefore its currency depends on when the model was last generated. According to Ben Bernstein at NCAR, the age of the RUC model does not have a great effect with respect to icing diagnosis.

IIDA integrates information from the GOES-8 satellite, surface observations, and the RUC model to identify the three-dimensional extent of clouds. It then uses information from these resources plus pilot reports and national radar mosaic to identify the locations and likelihood of both conventional and supercooled large drop icing across the United States and Canada. A situational approach is used which applies information from the different data sources in different ways, depending upon the physics expected to be at work at each location within the domain. This approach minimizes the impact of bad data from any one source. Images of the resultant icing and SLD fields, as well as the ingredients from which they are derived are available as output.

A representative IIDA diagnostic map of icing potential is shown in Figure 1.

Denser/darker areas indicate regions of higher icing potential. Since icing is a three-dimensional phenomenon, the IIDA human-computer interface allows depiction of icing, SLD, and visible moisture in horizontal cross-sections at 3000 foot intervals. It also allows the user to view vertical cross-sections, by either selecting pre-defined routes or by defining a route (e.g., Denver to Milwaukee, Figure 2).

12 NASA/CR—2002-211800 Figure 1. Representative IIDA Diagnosis of Icing Potential 13 NASA/CR—2002-211800 Figure 2. Representative IIDA Vertical Cross-sections NCAR’s Integrated Icing Forecast Algorithm (IIFA) is a forecast version of IIDA and is meant to mimic to IIDA information integration techniques. Since many of the IIDA observations are not available for three to twelve hours in the future, IIFA creates surrogates for each input data field, based upon output from the RUC (Rapid Update Cycle) model (for a detailed explanation, see http://www.rap.ucar.edu/largedrop/iifa/iifa_concept.explain ).

Evaluations of the IIDA with regional airline flight dispatchers identified the tendency for the algorithms to be conservatively biased (i.e., to sometimes forecast icing where it does not materialize) (FAA, 2000). The evaluations also identified the need to add icing severity as an output. The addition of downlinked icing reports could improve IIDA’s (as well as IIFA’s) performance.

14 NASA/CR—2002-211800

Project Objectives

The goal of this research project was to investigate the use of downlinked icing data to improve the IIDA and the IIFA products. There exist both technical and economic barriers in the way of a successful implementation of this concept. Only aircraft with both ice detection equipment and the appropriate databus and communications infrastructure can provide icing data. The set of aircraft meeting both of these criteria tend to be modern commercial turbojet aircraft that are certified to operate in all but severe icing conditions and generally are exposed to significant levels of in-flight icing for only a short period of time. For the operators of these aircraft, it may be difficult to justify the costs to retrofit the aircraft with icing downlink capability and the recurring communication costs. Thus the focus of this project was on improvements to IIDA and IIFA that could benefit major airlines. This focus provided a side benefit to NCAR in that previous evaluations focused on the regional carriers (FAA, 2000) while this research focused on the needs of a major carrier.

The objectives for this research were three-fold: 1. Without overlapping with the FAA’s efforts along these lines, to provide information concerning in situ icing assessment and reporting based on the constraints of our airline participant, Delta Air Lines.

2. To investigate the integration of in situ data from multiple aircraft into the IIDA and IIFA products.

3. To determine the utility of the upgraded icing products for use by major air carriers, specifically DAL. The idea here was to investigate the utility of the improved icing products to DAL in order to identify an incentive to participate in the downlink program as well as to provide NCAR with feedback concerning their products from a new set of potential users.

This research provides progress toward these objectives by: • Selecting an aircraft platform capable of digitally capturing, processing and downlinking icing data, • Identifying the required in situ icing data processing, • Determining the requirements for routing the icing data to NCAR for use by the IIDA and the IIFA products, • Developing an icing case study in order to gain insight into major air carrier needs, • Developing and prototyping icing display concepts for major air carriers based on NCAR’s IIDA and IIFA, and • Conducting a usability study for the prototyped concepts at a major air carrier.

15 NASA/CR—2002-211800

Work Carried Out

We achieved the objectives by accomplishing the following tasks: • Selecting an aircraft platform capable of digitally capturing, processing and downlinking icing data, • Identifying the required in situ icing data processing, • Determining the requirements for routing the icing data to researchers, • Developing an icing case study in order to gain insight into major air carrier needs, • Developing and prototyping icing display concepts, and • Conducting a usability study for the prototyped concepts.

Platform selection The purpose of this task was to identify a candidate aircraft that has both ice detector data available on the aircraft data bus and downlink capability. Ideally the ice detector output is available on the aircraft data bus, accessible to non-safety critical avionics with modifiable software.

In this way, the non-safety critical avionics can serve as a host for the icing software.

At Delta Air Lines, most aircraft have downlink capabilities, although aircraft, such as the B727, with three person crews do not (Table 1). DAL aircraft are equipped with anti-ice protection systems that prevent the formation of ice on wings and engines, assuming the aircraft only encounters ice in the range specified by FAR Part 25, Appendix C.

Table 1. DAL fleet ACARS equipage and ice detection Aircraft type ACARS Ice detector model B-737-200 Yes To be determined (TBD) B-737-300 Yes TBD B-737-600/700/800 Yes TBD B-757-200 Yes BFG’s 0871BN3-4 or 0871BN3-10 B-767-200 Yes BFG’s 0871BN3-4/10 (advisory) or 0871DL6 (primary) B-767-300 Yes BFG’s 0871BN3-4/10 (advisory) or 0871DL6 (primary) B-767-300ER Yes BFG’s 0871BN3-4/10 (advisory) or 0871DL6 (primary) B-767-400ER Yes TBD B-777-200 Yes BFG’s 0871DL6 MD-11 Yes BFG’s 0871GD1 MD-88/90 Yes Vibrometer’s VS3960 Most of DAL’s aircraft use a variant of BFGoodrich Aircraft Sensors Division’s (BFG’s) model 871 (Table 1). The BFG model 871 outputs a digital “ice/no ice” signal. The ice signal output is a switched high impedance to ground capable of sinking a 100 mA load. The icing signal feedback circuit functions when the high impedance output state is pulled up to 10 volts minimum.

A low impedance output of 20 ohms or less signifies icing. Under non-icing conditions, the output is 100K ohms or greater. The output signal latches “on” for 60±10 seconds. The duration is reset to 60 seconds if an icing signal is encountered before the initial 60 seconds have expired.

According to the manufacturer, the sensor does not provide an indication of ice until approximately 0.020 inch of ice has accumulated on the sensor probe. As soon as the ice signal is activated, a heater comes on to melt the ice off the probe in less than 15 seconds. The sensor then 16 NASA/CR—2002-211800 begins sensing ice again. According to the manufacturer, for most icing conditions, the time to de- ice the probe and to start sensing a new icing encounter is 5-7 seconds. Once the ice is debonded from the probe, a timer in the software leaves the heater on for an additional 5 seconds. Typically o o the ice debonds in 1-2 seconds. When the temperature is –20 C to -30 C, the de-ice time can increase to 10 to 12 seconds. With model 0871BN8, if the heater is on longer than 20 seconds, the ice detector will indicate a fault.

The icing signal output determines a bound on the resolution of a sensed icing event. There is nothing to be gained by polling the output any sooner than it will update. Thus the expected horizontal resolution depending on current ice detection systems is on the order of 4 nautical miles at low altitudes (based on a detector latching on for one minute of flight at 250 knots below 10,000 feet) and about twice that figure at higher altitudes. The expected vertical resolution will be subject to small measurement errors in cruise. However, in climb or descent, the resolution decreases by approximately1000-2000 feet (assuming a 1000-2000 feet/minute climb or descent rate).

Given the positive results of the ice detector and ACARS capability analysis, the remaining issue surrounded identifying what aircraft could easily host the icing data processing and downlinking software. Initial discussions with DAL’s avionics engineers identified the fact that the B777 aircraft possesses a unique integrated system that has the capability of collecting data and downlinking reports. The icing reports could be created by updating the user modifiable software onboard the B777. Thus without requiring expensive avionics upgrades, icing data from Delta’s B777 aircraft could easily be sent to the ground.

To achieve the processing and downlink capability, DAL would have to modify the Aircraft Condition Monitoring Function (ACMF) software to report specified criteria. DAL maintenance would have to upload the new software on the B777 aircraft. At no cost to NASA or to Search Technology, Delta Air Lines created a schedule and budget that included software engineering, testing, and aircraft installation. Assuming that the requirements for the icing processing are well- defined, the schedule allowed for two cycles of development, a prototype build and a final build.

That effort resulted in a labor hour estimate of 400 hours and a total project cost of $40,000.00. The details of the schedule and budget are available upon request.

In situ icing data processing Even though we recognize the value of real-time reporting of atmospheric parameters by commercial airlines to improved weather diagnoses and forecasts, recurring communication costs via the ACARS network need to be considered before specifying what is downlinked and how often. Currently, airlines that participate in the Meteorological Data Collection and Reporting System (MDCRS) agree to pay the additional communication costs associated with weather reports. As a result, airlines limit the spatial and temporal resolution associated with reporting winds and temperature through the MDCRS network to save airborne and ground transmission costs. Although there are initiatives within the FAA and National Weather Service to have the government subsidize some of these costs, it is unlikely this will happen soon.

Therefore, reporting frequency and message length will continue to be driven by economic concerns and not value of the data to improving the quality of weather information and benefit to the industry.

Airlines negotiate a per message cost through the ARINC ACARS network which is generally not divulged to the public or other airlines. However, to give some perspective on the extent of the costs and how they are determined, the following general information is provided: 17 NASA/CR—2002-211800 • ARINC CONUS costs are about $.025 per block of 125 characters. International costs are on the order of $.59 per block of 220 characters.

• A three hour domestic flight will generate about 30 MDCRS reports, costing approximately $1.20. A 12 hour international flight will generate about 106 reports costing about $85. A major carrier with 2000 flights per day would spend (approximately) over $7M per year on MDCRS reports if all flights are turned on and report according to ARINC Specification 620 standards. In reality, airlines modify the reporting frequency from ARINC 620 and do not have all flights report to save ARINC communication costs.

The above costs assume the use of ARINC 620, which creates a report every 3 minutes (default), buffers 5-10 reports, and downlinks the bundle every 5-10 minutes. Most airlines do not use the ARINC 620 strictly, simply because it inefficiently uses the data blocks and increases communication costs. It also should be noted that these costs are approximate, as different airlines have varying agreements with ARINC relating to message volume.

Table 3 provides a printed report example and Table 4 provides an ACARS example of the modified ARINC 620 Specification (actually it is called ARINC 618, modified to incorporate a turbulence field used by United Airlines). A similar specification would have to be developed to include an icing parameter. The sample specification illustrates the problem associated with adding additional parameters without fully understanding the derived benefit and impact to message length. In the CONUS, the message block size is 125 characters and the cost per block is a flat $.025. Therefore, ACARS messages are formatted using multiple blocks. If a message is only one character over 125, the cost per message doubles. For the example ARINC 618 Specification, four weather reports are bundled in each actual downlink to maximize the use of two data blocks (250 characters). 210 characters, including flight identification and ARINC header, make up the two block downlink leaving 40 characters available for an icing field. This would suggest up to 10 characters are available for an icing field per weather report in the CONUS. However, considering a single data block for international reports is only 220 characters, only 10 characters are available to icing before exceeding the 220 block size. To quantify the impact on cost of exceeding block size, per message costs would double.

The ARINC 620 Specification is currently being modified to incorporate water vapor mixing ratio and turbulence, in accordance with the RTCA AUTOMET Minimum Operational Performance Standards (MOPS). The new standard will presumably also comply with ICAO Annex 3 standards for downlink weather reports. Field sizes for the various icing parameters being considered for downlink are: • Water vapor mixing ratio, 4 characters.

• Icing indicator, Boolean, 1 character.

• Peak and average liquid water content, 2 characters each.

• Humidity, 2 characters.

18 NASA/CR—2002-211800 Table 3. Print Format for the Weather Report 0000000001111111111222222222233333333334444444444555555555566666 1234567890123456789012345678901234567890123456789012345678901234 02 WEATHER REPORT <50> 04 RPT TRG DVER ACID FLT DEPT DEST DATE GMT FM *05 E22 C 99 999 9999 AAAA AAAA MMDDYY HHMM AA 07 LAT LON GMT ALT SAT WD WS TIND *08 SDDMM.M SDDDMM.M HHMM S99999 S99.9 999 999 0000 *09 SDDMM.M SDDDMM.M HHMM S99999 S99.9 999 999 0000 *10 SDDMM.M SDDDMM.M HHMM S99999 S99.9 999 999 0000 *11 SDDMM.M SDDDMM.M HHMM S99999 S99.9 999 999 0000 0000000001111111111222222222233333333334444444444555555555566666 1234567890123456789012345678901234567890123456789012345678901234 Notes: 1. Report Format Characters: “A” is an Alphanumeric, “9” is a numeric, and “S” is a sign indicator. 0 is a padded character ”0”.

2. Report lines, which have an asterisk (“*”) on the left hand side are to be downlinked when the Report’s Output destination is ACARS.

3. The header line (Line 05) is snapshot at the time of the report trigger.

4. It is possible that four weather points are not collected prior to the end of a Weather phase (e.g., Ascent). For this case, only the collected lines are downlinked.

5. The “S” sign for altitude and static air temperature (SAT) is defined below: • Negative Sign is “M” • Positive Sign is “P”.

6. The “S” sign for latitude (LAT) and longitude (LON) is defined below: • LAT negative value is SOUTH or “S” • LAT positive value is NORTH or “N” • LON negative value is WEST or “W” • LON positive value is EAST or “E”.

7. Other field names and formats appear in Table 4.

Table 4. ACARS Format for the Weather Report Data Sent Char Data Scale STANDARD ACARS 1-19 Reference: Honeywell Product Specification - Common HEADER Functions - 967-0212-601, Section 4.9.3.1.3.

E22 20-22 E22 E22 (Weather report ID) TRG 23 C C (computer generated) DVER 24-25 99 ACMS Database version ACID 26-28 999 Aircraft Tail Number FLT 29-32 9999 Flight Number DEPT 33-36 AAAA Departure DEST 37-40 AAAA Destination DATE 41-46 MMDDYY Date (Month, Day, and Year) 19 NASA/CR—2002-211800 Data Sent Char Data Scale GMT 47-50 HHMM GMT time (Hours and Minutes) FM 51-52 AA Flight Mode CR 53 <CR> Carriage Return LF 54 <LF> Line Feed LAT 1 55-60 SDDMMM Latitude S/N XX Degrees XX.X Minutes LON 1 61-67 SDDDMMM Longitude W/E XXX Degrees XX.X Minutes GMT 1 68-71 HHMM GMT time (Hours and Minutes) ALT 1 72-77 S99999 Altitude P/M XX,XXX Feet SAT 1 78-81 S999 Static Air Temperature P/M XX.X Degrees C WD 1 82-84 999 Wind Direction XXX Degrees WS 1 85-87 999 Wind Speed XXX Knots TIND 1 88-91 0000 Turbulence Indicator (Padded and sent as Zeros) “0000” CR 92 <CR> Carriage Return LF 93 <LF> Line Feed LAT 2 94-99 SDDMMM Latitude S/N XX Degrees XX.X Minutes LON 2 100-106 SDDDMMM Longitude W/E XXX Degrees XX.X Minutes GMT 2 107-110 HHMM GMT time (Hours and Minutes) ALT 2 111-116 S99999 Altitude P/M XX,XXX Feet SAT 2 117-120 S999 Static Air Temperature P/M XX.X Degrees C WD 2 121-123 999 Wind Direction XXX Degrees WS 2 124-126 999 Wind Speed XXX Knots TIND 2 127-130 0000 Turbulence Indicator (Padded and sent as Zeros) “0000” CR 131 <CR> Carriage Return LF 132 <LF> Line Feed LAT 3 133-138 SDDMMM Latitude S/N XX Degrees XX.X Minutes LON 3 139-145 SDDDMMM Longitude W/E XXX Degrees XX.X Minutes GMT 3 146-149 HHMM GMT time (Hours and Minutes) ALT 3 150-155 S99999 Altitude P/M XX,XXX Feet SAT 3 156-159 S999 Static Air Temperature P/M XX.X Degrees C WD 3 160-162 999 Wind Direction XXX Degrees WS 3 163-165 999 Wind Speed XXX Knots TIND 3 166-169 0000 Turbulence Indicator (Padded and sent as Zeros) “0000” CR 170 <CR> Carriage Return LF 171 <LF> Line Feed LAT 4 172-177 SDDMMM Latitude S/N XX Degrees XX.X Minutes LON 4 178-184 SDDDMMM Longitude W/E XXX Degrees XX.X Minutes GMT 4 185-188 HHMM GMT time (Hours and Minutes) ALT 4 189-194 S99999 Altitude P/M XX,XXX Feet SAT 4 195-198 S999 Static Air Temperature P/M XX.X Degrees C WD 4 199-201 999 Wind Direction XXX Degrees WS 4 202-204 999 Wind Speed XXX Knots TIND 4 205-208 0000 Turbulence Indicator (Padded and sent as Zeros) “0000” CR 209 <CR> Carriage Return LF 210 <LF> Line Feed 20 NASA/CR—2002-211800 Notes: 1. Report Format Characters: “A” is an Alphanumeric, “9” is a numeric, and “S” is a sign indicator. 0 is a padded character ”0”.

2. Decimal points are removed from ACARS message.

3. The “S” signs for Altitude and SAT is defined below: • Negative Minus Sign is “ M” • Positive Plus Sign is “P”.

4. The “S” signs for LAT and LON is defined below: • LAT negative value is SOUTH or “S” • LAT positive value is NORTH or “N” • LON negative value is WEST or “W” • LON positive value is EAST or “E”.

Considering the United Airlines ARINC 618 format, it would seem that any icing field could be included without exceeding two data blocks during CONUS flights. For international reports, only the Boolean indicator is possible. It is improbable that all airlines will comply with the new ARINC 620, so at this point it is difficult to make a judgment as to which icing parameters are both economically and technically feasible.

To summarize the preceding analysis: • The Boolean icing indicator appears to be technically and economically feasible for both CONUS and international weather reports. It has informational value to integrated in flight icing algorithms. Most likely there is no value to numerical weather prediction (NWP) models as such models are designed to handle physical units (atmospheric data) and would require significant modification to incorporate the icing data.

• Peak and average liquid water content, together, would be economically feasible for the CONUS. However, costs would double for international reports. Technically, there is some risk in the sensor development and attainable accuracy. These parameters should have value to both integrated in flight icing algorithms and NWP models.

• Operational humidity sensors have a limited life and may introduce quality control issues.

However, humidity (2 characters) might be feasible if reporting frequency is decreased.

• Water vapor sensors are being installed on a limited number of commercial aircraft. This program may expand to include many aircraft, and is government funded (including communication costs). The exact scope of the program that is approved and funded is unknown at this time. Since communication costs are not paid by the airlines, the block size issue goes away.

• The analysis suggests that an icing parameter field should be added to whatever ARINC Specification is being used by a particular carrier (for example, ARINC 618). If the block size is exceeded (say for an international report), one way to decrease the number of characters needed for any parameter is to use hexadecimal representation of a “bin” or range of values. This technique is being used for downlinking turbulence. Further analysis would be needed to optimize definition of the bin values so that accuracy of the data is not compromised.

21 NASA/CR—2002-211800 Routing of the icing data The operational concept assumes that in situ icing data from aircraft can be downlinked to the ground and can be available for use by NCAR’s IIDA and IIFA products. Search Technology contacted ARINC to investigate the routing of the icing data to NOAA (the Forecast Systems Lab in Boulder). NCAR already receives the MDCRS data from NOAA so this strategy eliminates creating a new data network just for NCAR.

ARINC responded that the creation of the routing mechanism for a new datalink message is very easy. Basically a new identifier is created and a new entry is added to a routing table. The cost for ARINC to create the identifier and to add it to the routing table is small (perhaps they would even do it at no cost to NASA, Search Technology, or to Delta Air Lines). This is not surprising, as ARINC would still make a profit from the new data stream.

Icing case study To help understand the icing data needs of major airlines, an icing case study was developed based on the March 20, 2000 icing event at Denver International Airport. On that day, hundreds of major air carrier and commuter flights were diverted or cancelled because of the lack of appropriate weather information. Cancellations were caused by the severity of the reported icing conditions. In addition, ground based deicing was overwhelmed by the amount of ice on aircraft that had landed.

There were also a few serious icing encounters.

th Reports of icing began with the first operations out of Denver on the 20 . NWS AIRMETs for moderate rime and mixed icing for Denver were issued as early as 0900Z. At least three reports went out between 1300Z-1400Z: one for light-moderate rime, one for moderate ice (no type), and one for moderate clear. Icing intensified as the morning went on, and peaked in the 1600Z-1700Z time frame.

The NWS issued SIGMET Oscar 1, calling for severe rime/mixed ice below FL150, at 1615Z. The SIGMET, valid from 16:15Z to 20:15Z (and beyond), was for occasional severe rime/mixed icing in clouds and in precipitation below 15,000 feet. Figure 1 depicts the area covered by the SIGMET. The text of the SIGMET follows: WSUS1 KSLC 201615 WS5O SLCO UWS 201615 SIGMET OSCAR 1 VALID UNTIL 202015 WY CO FROM BFF TO GLD TO PUB TO 50S HBU TO CHE TO 40NW LAR TO BFF OCNL SEV RIME/MXD ICGICIP BLW 150. CONDS CONTG BYD 2015Z.

PCF 22 NASA/CR—2002-211800 Figure 3. ATL-DEN Flight Plan and SIGMET Oscar 1 During the peak icing period, many landing and departing aircraft reported at least moderate ice, with some severe. Based on flight dispatcher comments, all of the arrival gates were experiencing icing. Some representative pilot reports include (aircraft identifiers removed to preserve anonymity): 1554Z IC LGT MIXED DURGC FL070-110 1629 IC LT ICING THRU FL080 BROKE OUT AT FL110 1647 IC MDT RIME DURGD FL110-100 MDT-SVR BELOW THAT 1655 IC MDT-SVR ICE FINAL RWY35R ALSO HAD TO KEEP ENGINE RUN UP ON APPROACH 1722 IC MDT MIXED DURGD FL095-070 R35R 1731 DEN-BOI NO PROBLEM COORDINATING EXPEDITIOUS CLIMB ABOVE ALL ICING PROBLEMS. CLIMBOUT WAS FINE WITH MINIMAL ICE BUILDUP.

1735 IC MDT MIXED FL080-070 -RA BELOW 070 IC MOD RIME BELOW 10000FT ON FINAL IC MDT MIXED DURGD FL110-070 R34R There were numerous voice reports of icing made directly to air traffic controllers and to airline flight dispatchers. Comments to flight dispatchers included the following: 5.5 inches of ice on the glare shield, could not retract flaps due to ice buildup, ice on all tail surfaces, and 2-3 inches of ice on the aircraft.

23 NASA/CR—2002-211800 To help develop a scenario for the case study, Delta Air Lines provided a set of flight plan, weather briefing and related data for aircraft flying into Denver before and during the time of interest. For the icing case study, we generated a flight plan and weather briefing data for a Boeing 757-200 flight, ACME 9999, from Atlanta to Denver. Figure 1 depicts the route of flight. See Appendix A for flight plan and weather briefing details and Appendix B for the locations for many of the referenced airports and navigational aids.

Weather related data for this event had been archived by NCAR and was available for analysis. NCAR created a web site with the archived data for the March 20, 2000 case study: www.rap.ucar.edu/largedrop/2000mar20case/case.html The web site included: • Satellite imagery, • Radar imagery, • METARs (See http://www.nws.noaa.gov/oso/oso1/oso12/fmh1/fmh1toc.htm for observing, reporting, and coding standards for surface based meteorological reports), • PIREPS (See http://www.ifis.airways.co.nz/shtml/planning/designators.shtml for aircraft type designators), • Sounding and profiler data, • AIRMETS and SIGMETS, and • IIDA plots.

Of particular interest to this project was the fact that the icing conditions were for altitudes well below typical cruise altitudes for all aircraft in a major carrier’s fleet. Data from such an event would only be relevant when aircraft are in the terminal area, (either departing or arriving), or on the ground.

Icing display prototype development and evaluation In this effort, the focus was to develop and to evaluate a display to help flight dispatchers with their icing related decisions. This section is divided into the main sections: the display concept and the usability study.

Display Concept Flight Dispatcher’s Icing Decision Making Before developing the display concept, several flight dispatchers were interviewed at Delta Air Lines in order to gain a better understanding of their icing related decisions. Each work day, a flight dispatcher is responsible for a set of flights. At the beginning of a shift, flight dispatchers are generally provided with a duty roster identifying the flights. Depending on the time of day that a shift begins, flights require different actions. Some of these flights may already be in progress and therefore require monitoring. Others may be at the gate with an initial flight plan on file and the dispatcher may need to modify the plan if conditions have changed. Other flights may still need to be planned for the first time. Some flights may be scheduled so far into the future that it is still too early to plan them.

There are several icing related decisions that flight dispatchers make: • Flight cancellation, • Flight plan routing based on aircraft equipment and forecast weather, 24 NASA/CR—2002-211800 • Alternate airport selection, • Fuel load planning, • Weight restriction considerations, and • Flight following As dispatchers plan flights well before scheduled pushback, they need good forecasts spanning the period from before the scheduled departure time to the scheduled arrival time (with cushions on either end for contingency planning). Dispatchers build in a margin of safety from icing conditions based on both the accuracy and recency of the forecast data. The more reliable the information, the less margin required. Due to the lack of complete current information, flight dispatchers may resort to contacting company aircraft that are in-flight for current weather conditions, particularly when forecast is old. For flight following, dispatchers need near real-time data. For a lot of these decisions, dispatchers preferred using PIREPs rather than forecast information because the PIREPs can be more reliable. However, when considering PIREPs, dispatchers must be concerned with the type of aircraft giving the report. For instance, they need to know if the airplane is a “hot wing” aircraft (i.e., an aircraft that uses engine bleed air to heat the leading edge) or an aircraft with inoperative equipment, unequipped to handle icing conditions.

With equipped aircraft, a major concern for dispatchers is inoperative equipment. The dispatcher must consider: • What is wrong with the aircraft?

• What types of icing conditions must this aircraft avoid?

• What route and at what altitude can I safely plan for this flight?

• If the altitude is restricted, what related changes do I have to make?

For example, for Delta Air Lines aircraft, inoperative equipment can be in the form of inoperative engine or wing anti-ice valves. The type of failure necessitates different actions. An inoperative wing anti-ice valve means that “hard ice” is a problem where an inoperative engine anti- ice valve means that “soft ice” must be considered. “Hard” ice occurs between 0 and -40 ° C in visible moisture, and where “soft” ice (ice that can form once air is cooled) can occur if the temperature is between 10 and -20 ° C and when the humidity is high (Myszkowski & Rezsonya, 1996). Thus, the available route and altitude selections vary depending on the type of inoperative equipment. In most cases though, the engine ice or “soft ice” is considered more serious.

In order to plan flights through and around icing conditions, dispatchers at Delta use both PIREPs and forecasts, similar to the ones used by pilots. Unfortunately, many of the products that are commercially available (including the web sites listed in the pilot needs section) only show the potential for hard ice. In order to determine the potential for soft ice, dispatchers tend to use temperature dewpoint spread heuristics. More data and tools for determining the potential for engine icing would be helpful.

25 NASA/CR—2002-211800 Flight Dispatcher’s Icing Display Needs Flight dispatchers at Delta Air Lines have a variety of computer support tools to help them do their job. Their workstation set up is comprised of three monitors with windows displaying different kinds of data. Figure 4 groups the available tools and data: Graphical Flight Following (e.g., Aircraft Weather Products(e.g., Situation Display) Company products; Airline Dispatchers Federation Weather Briefing Page) Access to company databases Flight Dispatcher Flight Tracking such as: • Flight planning status • ACARS messages • Flight status • Flight plans • Maintenance status • Flight schedules • … • Maintenance • Meteorology Alerts • Reservation system • Weather data (text) Figure 4. Typical Flight Dispatcher Support Tools At DAL, the flight dispatchers can access both company weather products as well as products available on the Internet (e.g., Airline Dispatchers Federation Weather Briefing Page).

The flight dispatchers can open multiple windows for the display of weather products.

The flight dispatchers can view a graphical flight following tool displaying the position of aircraft in near real-time. The DAL flight following tool, based on ATC data, is updated approximately every seven minutes.

The flight dispatchers have access to company databases that include a wide variety of information such as legacy flight planning tools, company communications, crew and flight schedules, maintenance information, weight and balance information, company weather data and meteorology alerts, and flight status. Company communications include messages sent via ACARS to and from the flight crew. A decision support tool displays the set of flights assigned to the flight dispatcher and annotates data fields with status items such as whether the flight plan has been filed, whether the aircraft has left the gate, what time the aircraft took off, and other related data.

Flight dispatchers need tools that help them make flight planning decisions with respect to the effect that icing has on both equipped aircraft and aircraft with inoperative equipment. Both structural and engine icing conditions should be supported. They would like forecasts to be more accurate so they do not have to build in margins that end up costing the company more operational expense.

The flight dispatcher interviews and the icing case study discussed previously revealed that major airlines’ flights generally spend a majority of their flight time above icing conditions. Based 26 NASA/CR—2002-211800 on interviews of flight dispatchers at DAL, it was determined that detailed icing information for the terminal area would be of the greatest value. At DAL, the flight dispatchers have relatively sophisticated workstations. They can display many weather products. However, they do not have good tools for viewing icing data in the terminal area. They can access IIDA on the ADDS web page and can hand draw routes in order to get the output tailored to a particular route. One positive aspect of ADDS is that it allows the integration of multiple sources of information and allows the ability to see PIREP text quite easily. However, the interface is slow and does not easily support viewing a single terminal area, let alone comparing several. For example, it is difficult to quickly zoom and add a new route. Also it is confusing to read the ADDS output as it is displayed linearly according to the order it is entered (rather than north up, for instance). Also, ADDS does not allow users to save the hand drawn routes. Flight dispatchers want tools that easily allow them to see if there is a path through the terminal area to the runway. Tools like ADDS, if they loaded quickly and were easy to use, would be of great value. Entering routes and waiting for them to load, however, can be laborious.

Displays tailored to Standard Instrument Departures (SIDs) and Standard Arrival Routes (STARs), the flight paths generally follow, could be of great value, especially if they allow the flight dispatchers to see if the icing is currently or is forecast to be along the route. However, flight dispatchers also need a regional view of the terminal area because aircraft do not always follow arrival and departure routes exactly. For instance, air traffic controllers can vector them around for traffic and other conditions. Information for a regional view should be available for altitude strata of concern as it is possible for aircraft to fly under or above the icing conditions.

To support flight planning for longer domestic flights, not only current information should be available but also 3, 6, 9 and 12 hours forecasts. To support international flights and decisions concerning whether an aircraft staying at a particular airport overnight can depart the next morning, even longer forecast periods are desirable.

Display Concept Example Based on these notions, a set of integrated displays was designed for icing data in the terminal area. The idea was to have a horizontal, regional view that could be filtered to show altitude strata of interest. The idea was also to allow flight dispatchers to easily view the possibility for icing along SIDS and STARs.

The following example uses the terminal area at Cincinnati. Figure 5 shows a display of the main screen: arrivals are depicted as the default. Each arrival is labeled with its entry fix and is color-coded according to the legend. The airport, CVG, is marked with a square. The background can be toggled from arrivals to departures at the user’s request. The bottom left portion of the display allows the flight dispatcher to request other views such as current and forecast icing or SLD data, either displayed in a composite view or by altitude strata. The bottom right portion allows selection of other icing related products tailored to the horizontal region of interest.

27 NASA/CR—2002-211800 Figure 5. Regional view with airport arrivals overlay The user selects the icing data to display from the arrivals display described above. Figure 6 illustrates the case where the flight dispatcher has selected the composite icing view. Although not depicted in the figure, a legend is available portraying the mapping of the color coding to the potential for icing. The darker the shading, the higher is the potential for icing. In this example, it is obvious that there is a potential for icing, especially in the northwest quadrant of the terminal area.

To gain a better idea of the vertical extent of the icing potential, the user can select the flight level filters and/or select a particular route.

28 NASA/CR—2002-211800 Figure 6. Regional view with airport arrivals overlay and example composite icing data As part of the display concept, flight dispatchers can easily access a vertical cross-section view for any of the routes depicted on the main screen. Figure 7 is an example cross-section view for the MOSEYS arrival route. To avoid confusion concerning the linear layout, the entry fix is displayed to the left and the destination to the right. On the vertical cross-section display each rectangle covers a 1000 foot by 40 km area. The top figure presents the icing potential data. The display makes it clear what altitude an aircraft would have to fly to avoid the icing.

29 NASA/CR—2002-211800 Figure 7. Example vertical cross-section display for the MOSEY5 arrival.

The middle of Figure 7 represents another vertical cross-section view (SLD) tailored to the arrival. The lower portion of the figure represents the location of visible moisture below 45º F.

30 NASA/CR—2002-211800 Similar figures would be available for each arrival route. Similarly, departure routes would be available. In addition to these icing diagnostic displays, forecasts would also be available.

Usability Study Although IIDA had been evaluated with regional carriers (FAA, 2000), the purpose of this study was to investigate the utility of IIDA and IIFA to flight dispatchers at major airlines. Based on the display concept described above, prototype displays were developed and a usability study was carried out. Our primary goal in the usability study was to find out if/how the displays should be refined.

Prototype displays Because of the limited resources for this study, the idea was to prototype IIDA and IIFA displays for one SID and one STAR at one airport. NCAR implemented prototype displays for the Cincinnati/Northern Kentucky International Airport in Covington, KY (CVG) since this is a hub for our airline partner, DAL, and there is a relatively high frequency of atmospheric icing conditions.

The prototype displays showed IIDA and IIFA output on plan view and vertical cross-sections. The displays operate on ‘live’ IIDA/IIFA data – that is, they were updated with current IIDA/IIFA output. The displays were implemented using HTML and were available over the Internet using a standard web browser.

Plan views (i.e. terminal scale views) were created of IIDA/IIFA output for an area approximately 840x1280 km centered on CVG. The main plan view included the composite IIDA output for current icing potential. The default main display showed stylized arrival routes.

Although the intention was to draw every arrival route to scale (as in Figure 6), resource constraints forced the design to include simplified, straight-line arrival routes overlaid on this view (Figure 8).

The composite icing view depicts a set of 40 km by 40 km areas where each area is color coded according to the highest icing potential for the vertical column associated with the area. To provide the ability for the flight dispatchers to gain an understanding of how the terminal area product could work, a single arrival route was linked to a vertical cross-section of that route (see the “+” at the bottom middle of Figure 8). The link was accessed by clicking on the “+” cross-hair at the end of the route shown on the display. PIREPs of icing were shown on each of these views as well. Other views (described below) were accessed through links at the bottom of these IIDA composite plan views.

A corresponding view was also created for the departures (Figure 9). As with the arrivals, a single route was implemented to provide the flight dispatchers with the opportunity to view an associated vertical cross-section (see the “+” on the middle right of Figure 9).

31 NASA/CR—2002-211800 Figure 8. Implemented regional view with airport arrivals overlay 32 NASA/CR—2002-211800 Figure 9. Implemented regional view with airport departures overlay The terminal scale plan views implemented for the usability test also included IIFA output for horizontal cross-sections at 2000 foot increments from 2000 to 20,000 feet MSL with 3, 6, 9, and 12 hour forecasts. An example plan view centered on CVG that includes a 3 hour forecast for 8000 feet is shown in Figure 10.

To provide the flight dispatchers the opportunity to comment on access to related icing data, icing cloud bases and tops, current precipitation, freezing precipitation, and snow were implemented for CVG. An example display of icing cloud tops is shown in Figure 11. An example display of current precipitation is show in Figure 12.

33 NASA/CR—2002-211800 Figure 10. Implemented IIFA 34 NASA/CR—2002-211800 Figure 11. Implemented tops 35 NASA/CR—2002-211800 Figure 12. Implemented precipitation Vertical cross-sections, accessible by selecting the route, were created for a single STAR (HAGSS 2) and a single SID (JOBDU 2) for CVG (Figure 13 and Figure 14). These vertical cross- sections showed IIDA output (icing potential and visible moisture) along the route from surface to 18,000 feet MSL including surface topography. Due to time constraints, SLD was not implemented in the vertical cross-sections.

36 NASA/CR—2002-211800 Figure 13. Implemented STAR vertical cross-section 37 NASA/CR—2002-211800 Figure 14. Implemented SID vertical cross-section Text help is accessible from links on the plan view displays. The text help consists of text explanations on web pages separate from the displays of how to use the displays, what is shown on the displays, concepts associated with IIDA/IIFA, and limitations on their interpretation. The text help is fairly rudimentary (i.e. it is not interactive) and it was not a focus of this study.

Flight Dispatcher Observation/Interview Three flight dispatchers with different meteorological experience participated in the usability study. The most expert flight dispatcher had previously worked in the meteorology department and conducts meteorology training for the dispatchers. Another had spent a large 38 NASA/CR—2002-211800 portion of his career in the weight and balance area and had the least meteorological knowledge.

The third dispatcher had an intermediate level of meteorological knowledge.

These dispatchers work about 40 to 45 flight plans per day. Normally, each dispatcher is responsible for about 4 to 5 flights with as many as 8 to 10 in the air at a time. All three dispatchers had a relatively high concentration of flights to/from CVG. They were all observed and interviewed on February 7-8, 2002. The study was conducted while the dispatchers were working the regular shifts. The study was conducted on a non-interfering basis, meaning that operational activities took precedence over usability study concerns.

The procedure for the usability study was as follows: 1. Observe dispatchers without IIDA/IIFA (baseline).

2. Observe dispatchers using IIDA/IIFA without training.

3. Introduce and train dispatchers on IIDA/IIFA.

4. Observe dispatchers using IIDA/IIFA.

5. Debrief.

A complete dispatcher observation plan is included in Appendix C. Dispatcher Observation Plan.

The training conducted in step 3 of the procedure was not extensive: it consisted of a short PowerPoint presentation describing the history of the IIDA/IIFA development and a brief description of IIDA/IIFA concepts. The PowerPoint presentation is available upon request.

Needs for icing information The flight dispatchers’ need for icing information was similar to those interviewed during the display concept phase of the effort. One difference across the flight dispatchers, however, was their “personal minimums” with respect to planning alternates. For example, the most conservative flight dispatcher plans an alternate if the ceiling is forecast to be less than 3500 feet, especially if the prevailing winds appear to favor the movement of the weather system to be at the destination and if wet or slippery runways are indicated at the destination.

Weather Awareness without IIDA/IIFA The process of accessing information needed for the dispatchers’ tasks is laborious. Flight planning and weather awareness require access to many information products, both those that are relevant for icing and weather and those that are not. The flight dispatchers must slowly and methodically request and wade though a great deal of data on multiple displays. The 3 monitors used by the dispatchers were constantly full of overlapping windows. They scan through the text data provided by the company database in their terminal windows. They view graphical weather products in many formats: some integrated with their other decision tools but mostly not. Some of the products are hosted internally, some over the Internet, and some through a text terminal.

Moreover, the dispatchers might directly consult a staff meteorologist. For the task of evaluating alternate airports, the dispatchers might consider METARs for the airport, PIREPs near the airport, TAFs, and SIGMETs, among potentially many other products for weather information alone, for example. These individual weather products partially help address this task but in a piecemeal fashion. This laborious process is reflected in the choices the dispatchers have made about accessing products. Different dispatchers have different ‘favorite’ products that they view frequently while disregarding others: for instance, ‘bookmark’ lists for Internet weather products are 39 NASA/CR—2002-211800 popular for the dispatchers but their contents vary widely among them. The dispatchers have individually placed value on the weather products they use for a variety of reasons: usefulness for their tasks, understandability, resolution of the information, spatial specificity, reliability, and tendency not to overforecast are just a few of these reasons.

Each dispatcher is able to request text weather information, such as METARs, AIRMETs/SIGMETs, by airport (Figure 15). To maintain weather awareness, this text information is automatically sent to the console window when values on such variables as winds and visibility exceed thresholds. Dispatchers can pre-assign a color coding to these messages so that they can be distinguished from other, non-weather related messages.

Each flight dispatcher also has access to many graphic products. Some of the weather products he uses are available internally (e.g. as shown in Figure 16) and some are accessed over the Internet (e.g. Figure 17). To maintain icing awareness, the dispatchers check multiple products such as GOES satellite data, cloud levels, and freezing levels. They monitor the radar data at a low intensity level to look for snow and virga. One dispatcher mentioned that there are difficulties interpreting radar data with respect to icing due to differing reflectivity for different types of precipitation (e.g. light sleet can appear heavy, moderate to heavy snow can appear light). In general, the dispatchers prefer graphic depictions.

Figure 15. Console window 40 NASA/CR—2002-211800 Figure 16. Internal Delta Air Lines weather products 41 NASA/CR—2002-211800 Figure 17. Example weather product linked to the ADF weather briefing page Understanding IIDA and IIFA Without training, the flight dispatchers had conceptual difficulty interpreting IIDA and IIFA. One of the issues is particular to IIDA and not IIFA: misunderstanding the diagnostic information as actual occurrence of icing. IIDA does not determine that actual incidence of icing, but represents a ‘nowcast’, a diagnosis of current conditions. That is, it is an inference of the potential of icing for a given area, not an observation. Flight dispatchers are familiar with forecast data and concepts -- forecasts cannot, by their nature, be an actual determination of conditions since they are a prediction of future conditions – but maybe unfamiliar with the concept of nowcasts. One of the dispatchers without a meteorological background was unable to understand that IIDA represented a nowcast and not the actual, current conditions.

Another potentially problematic issue in interpreting both IIDA and IIFA output is confusing icing potential for icing severity (as also reported by the FAA [2000]). The IIDA/IIFA algorithms calculate a potential for icing, but not the severity of the icing. A feasible concern is that a flight dispatcher might interpret the color coded scale as representing a measure of icing severity, 42 NASA/CR—2002-211800 with dark red being the most severe, instead of a measure of the possibility that icing occurrence, regardless of severity. Furthermore, even if it is understood that the IIDA/IIFA scale represents potential of icing and not severity of icing, it is possible that a flight dispatcher might implicitly associate severity with potential. One of the dispatchers without a meteorological background was unsure how to interpret the icing potential scale without instruction. There was no evidence in this study that the flight dispatchers make inferences about icing severity even when they understand that the IIDA/IIFA scale represents icing potential.

An additional issue that emerged during this observational study is that the dispatchers had problems understanding what period of time an IIFA forecast represented. As with other forecast products, an IIFA forecast is relevant relative to the time when the forecast is issued and not the current time. IIFA forecasts are issued every 3 hours based on RUC model data. The dispatchers initially thought the forecasts were based on the current time: when the dispatchers accessed a display with a forecast and saw that the forecast was labeled with a time earlier than expected, they were confused about what time period the forecast actually represented.

When shown the ADDS version of the IIDA, one of the dispatchers commented that he thought he did not really need information about SLD, though it was not clear that he understood the significance of SLD. It is clear that more training with respect to SLD is warranted.

Human-computer interaction and display layout concerns Navigation to the arrivals and departures display with the boxes labeled “ARRIVALS” and “DEPARTURES” was confusing to the dispatchers. Initially, one of the dispatchers had problems navigating the displays. For instance, he did not understand where to click to get to the vertical cross-sections and he did not realize that he could get to the horizontal cross-sections by clicking in the grid on the bottom left of the terminal-area plan display. He indicated that he initially had problems navigating to the different displays, but that “after playing with it” he no longer had a problem. It is unclear if this still would have been an issue if the SIDs and STARs had been represented on the IIDA/IIFA displays as in the original display concept (i.e. Figure 5). One of the dispatchers also said he would like to be able to see multiple entry and exit points for these routes and to be able to access a vertical cross-section for the whole route from the entry/exit point to the airport by clicking on the entry/exit point on the display.

The dispatchers would like to be able to show more information to help understand where the icing is. Overlays of airways, routes, fixes, and county borders on the IIDA/IIFA plan views might help. More labels for routes and fixes may also help. More data on the vertical cross-sections such as weather stations may also be helpful.

The dispatchers did not initially understand that the text symbols that overlaid the maps in the terminal-area plan view represented PIREPs. The ADDS implementation for PIREPs is preferred. Also, a legend for PIREP symbol meaning and a scale for intensity may be helpful.

The color usage on the displays should be addressed. One dispatcher was unable to distinguish some of the colors on the display due to color blindness. He suggested that dispatchers should be able to set their own color scheme for the icing potential scale. The dispatchers thought that the various coloring used for the text on the displays were confusing. For example, “IFRARED” is in pink and “REGIONAL MOSAIC” is in blue.

The dispatchers had to scroll down frequently during our observation. There is a good deal of white space to the right of the map: perhaps many of the link boxes could be moved to this area.

43 NASA/CR—2002-211800 More suggested functionality The dispatchers would frequently ask questions about items on the displays or how to navigate the displays. While the help system was not a focus of this study, it is clear that a help system that is more specific to IIDA/IIFA controls and displays is needed. A reasonable next step is to make the help system more context sensitive by associating help text more directly to individual items on the displays or to provide a manual that shows example displays and text associated with items on the displays.

The filtering of icing potential data could be linked to its value. One of the dispatchers commented that he would like to be able to select an icing potential threshold to filter out display of low icing potentials, perhaps using a sliding bar control.

One dispatcher said he would like to see the freezing layer integrated into the IIDA/IIFA displays.

One of the dispatchers said he would like a zoom feature like in the ADDS Java tools. With the zoom feature the user defines a rectangle on a map by clicking on the upper left vertex and dragging to the lower right. The display updates with the area indicated by the rectangle zoomed to fill the display.

The flight dispatchers were concerned about adding yet another product. One of the dispatchers said he would like information about icing as an overlay on the aircraft situation display with an option to toggle on/off. Also he would like to be able to click on a route in the aircraft situation display to view a vertical cross section for that route. The notion of dynamically drawing a vertical cross-section for aircraft under a flight dispatcher’s control is worthy of investigation.

The IIDA/IIFA product is currently issued every hour. One dispatcher said that one hour is too long. He said that a fifteen minute time resolution would be good and that a thirty minute time resolution might be sufficient.

One dispatcher said he would like a long range forecast that would support a decision of whether or not an aircraft can leave an airport the next day.

One dispatcher emphasized the importance of not “over-forecasting” (i.e. indicating a high potential for icing in a place where there is actually a low potential. He believes that dispatchers will not use IIDA/IIFA displays if they notice a very few instances of overforecasting.

44 NASA/CR—2002-211800

Summary of Results Obtained

Platform selection At the major airlines, many aircraft types have ice detector and ACARS capability. Newer aircraft, such as the B777 aircraft, possess integrated information handling systems that can collect data and process downlink reports. Thus without requiring expensive avionics upgrades, icing data from such aircraft could easily be sent to the ground.

To achieve the processing and downlink capability for the B777, the airline would have to modify the Aircraft Condition Monitoring Function software to report the icing information.

Maintenance would have to upload the new software on the aircraft. Delta Air Lines estimated that for their seven B-777s, assuming that the requirements for the icing processing are well-defined, the effort resulted in a labor hour estimate of 400 hours and a total project cost of $40,000.00. The details of the schedule and budget are available upon request.

In situ icing data processing The dissemination of an indication of icing appears to be technically and economically feasible for both CONUS and international weather reports. It has informational value to integrated in flight icing algorithms. Most likely there is little current value to numerical weather prediction (NWP) models. Peak and average liquid water content, together, would be economically feasible for the CONUS. However, costs would double for international reports. Technically, there is some risk in the sensor development and attainable accuracy. These parameters should have value to both integrated in flight icing algorithms and NWP models. Operational humidity sensors have a limited life and may introduce quality control issues. However, humidity (2 characters) might be feasible if reporting frequency is decreased. Water vapor sensors are being installed on a limited number of commercial aircraft. This program may expand to include many aircraft, and is government funded (including communication costs). The exact scope of the program that is approved and funded is unknown at this time. Since communication costs are not paid by the airlines, the block size issue goes away.

The analysis suggests that an icing parameter field should be added to whatever ARINC Specification being used by a particular carrier (for example, ARINC 618). If block size is exceeded (say for an international report), one way to decrease the number of characters needed for any parameter is to use hexadecimal representation of a “bin” or range of values. This technique is being used for downlinking turbulence. Further analysis would be needed to optimize definition of the bin values so that accuracy of the data is not compromised.

Routing of the icing data The routing mechanism for a new datalink message is quite simple. Basically a new identifier is created and a new entry is added to a routing table. The cost for ARINC to create the identifier and to add it to the routing table is small (perhaps they would even do it at no cost).

45 NASA/CR—2002-211800 Icing case study The icing case study highlighted the notion that icing conditions generally occur for altitudes well below typical cruise altitudes for the aircraft in a major air carrier’s fleet. Thus products tailored for the terminal area may be of great benefit.

Icing display prototype development and evaluation This effort focused on developing and evaluating a concept for a display that helps major airline flight dispatchers with their icing-related decisions. The display concept included detailed IIDA/IIFA information for the terminal area with a focus on SIDs and STARs. A display prototype for a terminal area around CVG with one SID and one STAR was implemented and evaluated. The three Delta Air Lines flight dispatchers who participated in the evaluation liked the prototype displays and thought they would be useful.

The dispatchers in general liked the terminal area views. Arrivals and departures as well as ground operations are where the DAL dispatchers are most concerned with icing. The evaluation did uncover various usability issues such as usage of color, spatial layout of items, display navigation. The flight dispatchers were concerned about adding another product to their already large assemblage of weather information products, but expressed that the IIDA/IIFA information would be easier to embrace if it were easy to access, preferably integrated with their existing, commonly used tools. This balance between the high value of the information afforded by IIDA/IIFA and a desire for ease of access was expressed in the dispatchers comments about wanting to be able to access the IIDA/IIFA information through their existing, frequently used tools (e.g. overlays on their flight situation display, getting vertical cross-sections by clicking on routes in the flight situation display).

The dispatchers did have some conceptual problems with the displays -- unfamiliarity with the ‘nowcast’ concept, confusing icing potential for icing severity, confusion concerning IIFA time validity. These problems could be addressed through training and better display design. To help reduce confusion concerning the time validity of IIFA forecasts, IIFA displays should be amended to include the time the forecast was issued. Currently, the IIFA displays are implemented such that each forecast (i.e. 3 hour, 6 hour, 9 hour, 12 hour) for the same area must be accessed separately. An integrated display of all IIFA forecasts for a particular area might also help with the time confusion. Ideally, this integrated display might include an animation of the forecasts. The study also highlighted a lack of icing knowledge such as the importance of SLD. These misconceptions indicate a need for more extensive training.

46 NASA/CR—2002-211800

Future Research

This project should continue down the planned path--that is, equip an aircraft type (B777) to downlink an icing parameter and determine its value to NWP models and integrated algorithms like IIDA and IIFA.

High resolution information on in flight icing hazards has value to both commuter (FAA, 2000) and major air carrier operations. The concept display showing both plan view and vertical cross-sections is useful to the flight dispatcher function, particularly in the terminal area. Easy access to IIDA/IIFA information is a priority for adoption of an icing information tool by the flight dispatchers. IIDA/IIFA can augment the tools commonly used by dispatchers.

Integrating IIDA/IIFA with other tools may allow for more rapid adoption. For example, Delta Air Lines flight dispatchers are currently evaluating a tool they call a “Duty Roster” (Figure 18). The Duty Roster displays various information on the flights being managed: flight status, origin and destination, departure and arrival time, payload, etc. Such data could be used to automatically tailor the IIDA/IIFA display to the terminal areas of concern. Ideally if there may be potential icing hazards, the Duty Roster could be augmented with this information as a prompt to remind the flight dispatcher to view the icing data.

Figure 18. Prototype Duty Roster While the conceptual problems in understanding IIDA/IIFA displays can be address to some extent with display design, it seems that some of these problems could be addressed with web-based training.

47 NASA/CR—2002-211800

Appendix A. Flight Plan and Weather Briefings

The following is a flight plan and pre-flight weather briefings for ACME airline flight 9999 on March 20, 2000.

Flight plan for ACME 9999 (scheduled 13:55Z – 17:08Z)

ATL/DEN ALTN SLC SKED PLND ACTL KATL 1355Z/0855L 1355Z ....

TAXI 0022 ....

OFF 1417Z ....

ETE 0244 ....

ON 1701Z ....

TAXI 0007 ....

KDEN 1720Z/1020L 1708Z ....

DOT ON-TIME ARVL LIMIT IS SKED PLUS 14 OR 1734 Z/1034L – APPLIES ONLY TO FLTS WITHIN 50 STATES / PUERTO RICO / U.S. VIRGIN ISLES SHIP11111 L/B752/E DGMR TYPE ECN FL 350 ROUTE NRP 1071 MI ELEV KATL 1026 FT KDEN 5431 FT REMARK- NRP KATL..WETWO..GAD..MEM..RZC..PER..GCK.J154.RYLIE.DANDD3.KDEN ETE-244 GAD-MEM AFTER GAD STEER GAD288 RADIAL UNTIL ABLE TO RCV MEM VOR 1 2 RAMP WT 218192 LWT 195076 PAYLOAD 182 /041517 MPTW 225422 FLIGHT PLAN INCLDS CARGO 003479 TARGET GATE ARVL FUEL 20.6 GATE C42 RAMP XXX FREQ 131.45 4 5 TRIP TIME/BURN ATL DEN 244/023120 -TAXI 22/00770 IFR/ALTN SLC FL350 59/008170 PLND CNTNGNCY FUEL 57/005940 SEE RMKS UNPLND CNTNGNCY FUEL 15/001560 RESERVE FUEL 005210 BLOCK FUEL 044000 MIN FUEL FOR T/O 041670 FLIGHT CONTROL AND MAINTENANCE REMARKS -FUEL 01 ATC/TRFC DLAS ... PSBL LWR FLT LVL ENRTE FOR RIDE -DISPATCHER NONE 1 Passenger count 2 Weight of passengers, baggage, and cargo 3 Scheduled destination gate/ramp 4 Burn includes taxi out and city maneuvering but not taxi in 5 Taxi out time and fuel 49 NASA/CR—2002-211800 -M.E.L. /C.D.L.

S32-00-01 THIS SHIP CARRIES A FLY AWAY KIT IN BIN 1 M25-20-01 PASSENGER CONVENIENCE ITEM/S/ -SHIP REMARKS 11111 ACARS AUTOMATIC ENGINE REPORTING NOT INSTALLED. PLEASE MANUALLY COMPLETE THE ENGINE PERFORMANCE REPORT SHEET AS REQUIRED PER FOM PG 7-43.

11111 EEC SWITCH IS LOCATED ON THE P-61 PANEL..ONLY ONE SWITCH FOR BOTH ENGINES.

11111 HF RADIO EQUIPPED 11111 ASSUMED TEMPERATURE DERATES ARE THE ONLY DERATED TAKEOFFS AUTHORIZED FOR THIS AIRCRAFT. TO1 AND TO2 SHOULD NOT BE USED PER 757/767 FLEET SPECIALIST.

11111 NEW MAX TAXI WT 241000 / MAX TAKE OFF WT 240000 EFF.

05DEC98 PER B757 FLEET SPECIALIST 11111 ***PLEASE NOTE COST INDEX - 02DEC99** DOMESTIC ECN-81 /SLI-54 / MNF-0 / OT1-244 / M83-400 INTL COST INDEX - INTERNATIONAL FLIGHTS SHOULD REFER TO FOB 99-13 RE FPS FLT PLANS FOR USE OF CI IN HOWGOZIT SECTION.

SUPPLEMENTARY ROUTE INFORMATION 01 ATC PREF--TIME/BURN 246/023300 RTE ATL N0468F350 DCT WETWO DCT VUZ J41 MEM DCT RZC DCT PER DCT GCK J154 RYLIE DANDD3 DEN 02 --FL 310--TIME/BURN 242/023820 TRMG FRMG ALERTS FIX FL TEM PWR IAS/M TAS WCP GS ZD ZT ZF FAT TAXI 0244 0432 SE1 ECN/CLB M029 041 010 032 WETWO 0234 0400 SE1 ECN/CLB M067 050 008 018 GAD 0226 0382 SE1 ECN/CLB M068 019 002 005 T-O-C 0224 0377 350 P07 216 272/800 468 M064 404 184 028 036 F302 MEM 0156 0341 SE1-SC1 350 P05 207 272/800 464 M058 406 215 032 042 F266 R RZC 0124 0299 SC1 350 P00 199 272/800 461 M052 409 150 022 028 F224 PER 0102 0271 SC1 350 M02 193 272/801 460 M044 416 184 026 033 F196 GCK 0036 0238 350 M02 187 273/801 459 M036 423 058 008 010 F163 RYLIE 0028 0228 350 M04 186 273/802 459 M033 426 043 006 008 F153 T-O-D 0022 0220 300/802 M003 026 004 000 F145 SELLS 0018 0220 008 001 001 F145 PRAGG 0017 0219 013 002 000 F144 KIPPY 0015 0219 028 004 002 F144 DANDD 0011 0217 6 Meteorological alerts 50 NASA/CR—2002-211800 052 011 008 F142 DEN 0000 0209 DESCENT XPCT 250KIAS AT OR BLO 170 VOR DVV DVV DVV DVV DVV DVV DVV RADIAL 118 122 122 122 122 122 122 DME 118 090 078 065 053 050 044 FL 350 310 270 230 190 180 170 COMPUTED ENROUTE WINDS FIX TROP CRZ FL TWO FLT LVLS LOWER TWO FLT LVLS HIGHER SAT TAT SAT TAT SAT TAT SAT TAT SAT TAT 10 11 T-O-C 33 35 27067 31 27074 33 27071 37 27061 39 W 27054 MEM 34 35 29063 31 29072 33 29068 37 29057 39 W28052 -47 -18 -45 -16 -47 -18 -48 -19 -48 -19 RZC 36 35 29054 31 29050 33 29053 37 29054 39 28053 -53 -25 -48 -19 -51 -23 -54 -26 -54 -26 PER 38 35 27052 31 26043 33 27047 37 27055 39 26057 -56 -28 -49 -20 -53 -25 -58 -31 -58 -31 GCK 39 35 24066 31 24058 33 24063 37 24067 39 24064 -57 -30 -48 -19 -53 -25 -60 -33 -61 -34 RYLIE 38 35 23072 31 23064 33 23068 37 23072 39 23069 -58 -31 -48 -19 -54 -26 -60 -33 -61 -34 T-O-D 38 35 23075 31 22066 33 22071 37 23076 39 23073 FIX LIST DATA AVG AVG AVG FIX FMS LAT LONG T/C VAR D/C M/H KATL N33 38.4 W084 25.6 279 W02 WETWO N33 43.7 W085 07.4 287 W01 GAD N33 58.6 W086 05.0 288 E01 MEM N35 00.9 W089 59.0 290 E04 286 RZC N36 14.8 W094 07.3 282 E06 R01 277 PER N36 44.8 W097 09.6 293 E08 L03 282 GCK N37 55.1 W100 43.5 296 E11 L07 278 RYLIE N38 20.2 W101 49.7 295 E12 L08 275 SELLS N38 49.6 W103 10.1 314 E12 PRAGG N38 55.0 W103 17.4 314 E12 KIPPY N39 04.3 W103 29.8 314 E12 DANDD N39 23.9 W103 56.3 309 E12 KDEN N39 51.5 W104 40.0 AIRPORT/NAVIGATIONAL REMARKS -AIRPORT PAIR REMARKS ATL-DEN **THIS IS A CITY PAIR IN WHICH WE ARE ALLOWED TO USE ANY ROUTE WHICH COMPLIES WITH NRP RULES..MUST INCLUDE AF/--NRP REMARK ON FLT PLAN..PREF RTE IS OPL//CB FLTCTL//29NOV99CL 7 Crossing restrictions 8 4000 foot descent checkpoints 9 Calculated for fix positions and planned fix crossings 10 First two digits are direction in tens of degrees; last three digits are velocity 11 W prefix indicates aircraft is too heavy for that altitude 51 NASA/CR—2002-211800 -AIRPORT REMARKS SLC 02 BEARR THREE ARRIVAL - AT BEARR INT ACFT LNDG SOUTH EXPECT CLEARANCE TO CROSS AT 16000 FT. ÄCHART NOTAM) SLC 03 BRIGHAM CITY ONE ARRIVAL - AT CARTR INT ACFT LNDG SOUTH EXPECT CLEARANCE TO CROSS AT 15000 FT. ÄCHART NOTAM) SLC 06 PARKING GATES D2 AND D6 USE 2L DOOR.

SLCCPO/28FEB00 -AIRPORT NOTAMS ATL 02/025 8L ILS DME CMSND WEF 0002240901 ATL 03/009 TOWER 1220 250 AGL 4.8 NE LGTS OTS TIL 0003210900 ATL 03/012 TOWER 1150 250 AGL 4.1 SE LGTS OTS TIL 0003250700 ATL 03/023 9R ILS CAT 2/3 NA WEF 0003151300 ATL 03/027 TOWER 1117 148 AGL 2.9 SE LGTS OTS TIL 0003292000 ATL 03/030 TOWER 1420 400 AGL 5.9 NE LGTS OTS TIL 0003310500 DEN 03/058 16/34 RWY LGTS OTS DEN 03/077 TOWER UKN 400 AGL 5 NW LGTS OTS TIL 0003300530 SLC 02/064 TOWER 5458 249 AGL 9E LGTS OTS SLC 03/050 TACAN AZM OTS SLC 03/053 14/32 CLSD SLC 03/052 ALL RWYS ALTNLY CLSD SNOW REMOVAL -JEPPESEN CHANGES/FDC NOTAMS/INTERNATIONAL NOTAMS ATL 0/0940 VOR OR GPS RWY 27L AMDT 4...

TERMINAL ROUTE FROM PANOL INT/ATL 10.1 DME IAF TO AMATE INT/ ATL 4.80 DME NOPT DELETE IAF AND NOPT. ADD FROM AMATE INT TO RWY 27L 2.94 DEGREES/ TCH 65 FT.

THIS IS VOR OR GPS RWY 27L AMDT 4A.

ATL 0/0944 ILS RWY 27R AMDT 3A...

TERMINAL ROUTE FROM HOKIE INT/ATL 10.32 DME/ RADAR IAF TO LIAMS OM/INT/ ATL 6.30 DME/RADAR NOPT DELETE IAF AND NOPT.

THIS IS ILS RWY 27R AMDT 3B.

ATL 0/0946 ILS RWY 26L AMDT 17B...

TERMINAL ROUTE FROM KINKY INT IAF TO PANOL INT NOPT DELETE IAF AND NOPT.

THIS IS ILS RWY 26L AMDT 17C.

ATL 0/0950 ILS RWY 8R CAT II AMDT 58A...

TERMINAL ROUTE FROM CHINN INT/ATL 13.70 DME IAF TO STUMP INT/ ATL 9.90 DME NOPT DELETE IAF AND NOPT.

THIS IS ILS RWY 8R CAT II AMDT 58B.

ATL 0/0953 ILS RWY 9R CAT II III AMDT 16...

TERMINAL ROUTE FROM TIZZY INT/I-FUN 10.70 DME IAF TO BURNY INT/OM/I-FUN 6.4 DME NOPT DELETE IAF AND NOPT.

THIS IS ILS RWY 9R CAT II III AMDT 16A.

ATL 0/0941 ILS RWY 27L AMDT 13...

TERMINAL FROM ANVAL INT/I-FSQ 10.64 DME/RA DAR IAF TO DEPOT 52 NASA/CR—2002-211800 INT NOPT DELETE IAF AND NOPT.

THIS IS ILS RWY 27L AMDT 13A.

ATL 0/1465 ILS RWY 26R AMDT 2A...

TERMINAL ROUTE FROM FREAL INT IAF TO BALLI INT NOPT DELETE IAF AND NOPT.

THIS IS ILS RWY 26R AMDT 2B.

DEN I 9/6089 ILS RWY 25 AMDT 1..

S-ILS DH 5668/HAT 316 RVR 4000 ALL CATS S-LOC MDA 5760/HAT 408 RVR 4000 ALL CATS FOR INOP MALSR TEMP CRANES..3..5455FT MSL 3704FT FROM RWY 25 THLD..954FT LEFT OF CTRLINE.

-ENROUTE NOTAMS MSL 001 09/003 TACAN AZM OTS AMG 001 03/003 VOR UNUSBL 330-078/095-116/141-149/169-193/210-215 BYD 10 BLW 5000/ 079-094/117-140/150-168/194-209/216-225/235-329 BYD 10 BLW 9000 PLUS SEE AFD FSM 001 03/017 VOR OTS WEF 0003201400-0003202200 IRW 001 EXPECT RADAR VECTORS FROM ATC DURING VOR OUTAGE//FEB02//FLTCTRL RLG 001 DEN 11/148 TACAN AZM OTS SNY 001 03/005 VORTAC OTS WEF 0003201600-0003202000 SLC 001 03/050 TACAN AZM OTS ACME 9999/20 RLS 1 ATL-DEN 20MAR1245RP 53 NASA/CR—2002-211800

Final weather briefing for ACME 9999 (12:48Z)

FL 350 ETE 0244 20MAR1248 ATL 1355Z 0855L-DEN 1720Z 1020L ALTN SLC ATL..WETWO..GAD..MEM..RZC..PER..GCK.J154.RYLIE.DANDD3.DEN DESTINATION WEATHER DEN 201000 METAR 200953Z 02016KT 10SM SCT060 BKN120 BKN250 03/M01 A2952 RMK AO2 SLP956 T00331011 DEN 201100 METAR 201053Z 36017KT 10SM SCT050 BKN110 BKN250 00/M02 A2959 RMK AO2 PK WND 01030/1030 PRESRR SLP989 T00001022 DEN 201200 METAR 201153Z 35019KT 10SM BKN014 OVC100 M01/M03 A2965 RMK AO2 PRESRR SLP015 T10061028 10106 21006 53044 DESTINATION FORECAST TERMINAL FORECAST DEN NWS 201142 KDEN 201138Z 201212 35015G28KT P6SM BKN015 TEMPO 1215 4SM -SHRASN FM1500 36012G22KT 5SM -SHRASN BKN008 TEMPO 1518 2SM -SHSN BR OVC005 FM1800 02015G28KT 2SM -SN BR BKN003 OVC010 TEMPO 1821 3/4SM SN BR VV002 FM2100 03017G30KT 1SM SN BLSN OVC003 TEMPO 2103 1/4SM SN BLSN VV001 FM0300 05016G35KT 1/2SM SN BLSN OVC003 WEATHER AROUND DESTINATION COS 201200 METAR 201154Z 01010KT 10SM FEW060 02/M06 A2952 RMK AO2 SLP968 T00171056 10106 20011 55006 CYS 201200 METAR 201156Z 36023G28KT 1SM -SN BR BKN015 OVC036 M05/M06 A2965 RMK AO2 PK WND 35030/1122 SLP033 P0002 60006 T10501061 10072 21050 51017 CYS 201200 METAR 201156Z COR 36023G28KT 1SM -SN BR BKN015 OVC036 M05/M06 A2965 RMK AO2 PK WND 35030/1122 SLP033 P0002 60006 70006 T10501061 10072 21050 51017 CYS 201222 SPECI 201219Z 35023G31KT 1SM -SN BR BKN013 BKN022 OVC036 M06/M07 A2966 RMK AO2 PK WND 35031/1219 P0000 CYS 201228 SPECI 201225Z 36024G31KT 3/4SM -SN BR SCT013 BKN022 OVC036 M06/M07 A2965 RMK AO2 PK WND 36031/1225 P0000 PUB 201200 METAR 201154Z 00000KT 10SM CLR 05/M03 A2947 RMK AO2 SLP943 T00501033 10050 21006 55014 GJT 201200 METAR 201156Z 28006KT 3/4SM -SN BR BKN004 OVC009 00/M01 A2960 RMK AO2 SLP005 P0012 60012 70012 T00001006 10133 20000 53018 GJT 201200 METAR 201156Z 28006KT 3/4SM -SN BR BKN004 OVC009 00/M01 A2960 RMK AO2 SLP005 P0012 60012 70012 T00001006 10133 20000 53018 CPR 201200 METAR 201155Z 03008KT 1 1/2SM -SN BR OVC017 M04/M06 A2983 RMK AO2 SLP106 P0002 60013 70022 T10391056 11033 21044 53016 ALTERNATE AIRPORT WEATHER SLC 201000 METAR 200956Z 09003KT 10SM SCT034 OVC110 M03/M04 A2979 RMK AO2 SNE09 SLP086 SNINCR 1/2 P0000 T10331039 SLC 201100 METAR 201056Z 00000KT 10SM FEW045 BKN060 BKN080 M03/M04 A2979 RMK AO2 SLP100 T10281039 SLC 201200 METAR 201156Z 34007KT 10SM FEW035 BKN080 M05/M06 54 NASA/CR—2002-211800 A2979 RMK AO2 SLP104 60019 70034 4/003 T10501061 11017 21067 56007 ALTERNATE FORECAST TERMINAL FORECAST SLC DL 201118 AMD 02 VALID 201118-210300 UTC 15 BKN 40 OVC 10 3510 OCNL 2SW- 15Z 15 BKN 40 OVC 10 3515G25 OCNL 2SW- 20Z 25 SCT 70 BKN 10 3618G28 02Z 30 SCT 150 BKN 10 3614 // NO ADNL ACCUMN SNOW // ORIGIN WEATHER ATL 201100 METAR 201053Z 22003KT 10SM FEW007 SCT017 BKN045 11/10 A2986 RMK AO2 SLP113 T01060100 ATL 201200 METAR 201153Z 24006KT 9SM SCT009 BKN012 BKN038 11/10 A2989 RMK AO2 SLP120 60002 70159 T01060100 10122 20100 53020 ATL 201245 SPECI 201241Z 27007KT 10SM FEW008 BKN026 BKN041 11/09 A2991 RMK AO2 VIRGA SW AND NW ORIGIN FORECAST TERMINAL FORECAST ATL DL 201107 AMD 01 VALID 201107-210300 UTC 10 OVC 7 2405 14Z 15 BKN 7 2610 16Z 25 BKN 7 2712 18Z 35 BKN 10 2812 20Z 45 SCT 10 2810 22Z CLR 10 3008 TAKEOFF ALTERNATE WEATHER NO REPORT ENROUTE SURFACE WEATHER ATL 201200 METAR 201153Z 24006KT 9SM SCT009 BKN012 BKN038 11/10 A2989 RMK AO2 SLP120 60002 70159 T01060100 10122 20100 53020 ATL 201245 SPECI 201241Z 27007KT 10SM FEW008 BKN026 BKN041 11/09 A2991 RMK AO2 VIRGA SW AND NW MEM 201200 METAR 201153Z 26005KT 10SM SCT021 OVC034 07/04 A2991 RMK AO2 CIG 024 RWY27 SLP129 70092 T00720044 10072 20067 53019 MEM 201219 SPECI 201218Z VRB05KT 10SM FEW020 BKN026 OVC033 07/04 A2992 RMK AO2 TUL 201200 METAR 201153Z 14003KT 6SM BR CLR 00/00 A2991 RMK AO2 SLP128 T00000000 10028 21006 53008 OKC 201200 METAR 201153Z 12012KT 10SM CLR 04/01 A2979 RMK AO2 SLP087 T00440011 10056 20039 56005 METRO ALERTS SE1 200754-202100 ***** SOUTHEAST REGION ***** OVER ERN GA/SC/NC/FL AREA WDLY SCT TRW PSBL....TOPS FL380 MOVG NE 20KT SC1 200755-202100 ***** SOUTH CENTRAL REGION ***** ***** NO MDT OR GRTR TURBC OR TRW FORECAST ***** GOVERNMENT WEATHER ALERTS Z22 201154-201355 CONVECTIVE SIGMET 15E VALID UNTIL 1355Z NC SC GA FROM 30NW 55 NASA/CR—2002-211800 CLT-20SE CAE-40WNW SAV LINE SEV TS 15 NM WIDE MOV FROM 25030KT. TOPS TO FL400. HAIL TO 1 IN...WIND GUSTS TO 50 KT POSS.

Z24 201154-201355 CONVECTIVE SIGMET 16E VALID UNTIL 1355Z SC GA AND CSTL WTRS FROM 30W CHS-60SSE CHS-50S SAV-30W CHS DVLPG AREA TS MOV FROM 23030KT.

TOPS TO FL320. OUTLOOK VALID 201355-201755 FROM ORF-170E PBI-70ENE PBI-130SE MIA-80WSW EYW-90W SRQ-CAE-HMV-ORF TS WILL CONT ALG/EAST OF CDFNT MOVG EWD ACRS THE AREA. ACT SHOULD BE MAINLY OVER THE CAROLINAS AND ADJ WATERS WITH WDLY SCT TS POSSIBLE IN MODERATELY UNSTABLE AMS OVER FL AND COASTAL WATERS.

OCNL WST ISSUANCES ARE LIKELY.

ACME PIREPS ACME0251/19 SHIP 0689 POS RZC OVR 0949 NXT CIM ETA 1106 ENS PGS ALT 310 FOB 0371 SAT 49 WND 287069 MCH 80 TRB LT CHOP SKY CLEAR ICE NONE ACME1448/19 SHIP 0665 POS RZC OVR 0956 NXT CIM ETA 1114 ENS PGS ALT 310 FOB 0383 SAT 48 WND 286066 MCH 80 TRB SMOOTH SKY CLEAR ICE NONE ACME0198/19 SHIP 0608 POS SGF OVR 0953 NXT ETA ENS ALT 370 FOB 0203 SAT 53 WND 275043 MCH 80 TRB SMOOTH SKY CLEAR ICE NONE ACME0548/19 SHIP 0127 POS PER OVR 1002 NXT KCVG ETA 1125 ENS ALT 370 FOB 0346 SAT 59 WND 281029 MCH 80 TRB SMOOTH SKY CLEAR ICE NONE ACME1244/20 SHIP 0611 POS GCK OVR 1029 NXT ENL ETA 1133 ENS KCVG ALT 330 FOB 0307 SAT 52 WND 263050 MCH 80 TRB SMOOTH SKY CLEAR ICE NONE ACME0384/19 SHIP 0639 POS LBL OVR 1020 NXT ETA ENS ALT 330 FOB 0317 SAT 52 WND 273045 MCH 80 TRB LT CHOP SKY CLEAR ICE NONE SUBSEQUENT STATIONS NONE AIRPORT ALERTS ATL NO REPORT DEN NO REPORT SLC 201245-210600 MDT TURBC DURGC/DURGD SFC-FL100 FIELD CONDITIONS ATL NO REPORT DEN NO REPORT SLC NO REPORT 56 NASA/CR—2002-211800

Updated weather briefing for ACME 9999 (13:48Z)

FL 350 ETE 0244 20MAR1358 ATL 1355Z 0855L-DEN 1720Z 1020L ALTN SLC ATL..WETWO..GAD..MEM..RZC..PER..GCK.J154.RYLIE.DANDD3.DEN DESTINATION WEATHER DEN 201300 METAR 201253Z 36013KT 10SM BKN012 OVC110 M01/M03 A2968 RMK AO2 SLP036 T10111033 DEN 201400 METAR 201353Z 34024G27KT 10SM OVC012 M01/M04 A2972 RMK AO2 PK WND 33027/1353 SLP054 SHSN VC SW-W T10111044 DESTINATION FORECAST TERMINAL FORECAST DEN NWS 201333 AMD KDEN 201320Z 201312 35012G20KT P6SM BKN012 TEMPO 1416 4SM -SHSN FM1600 01013G22KT 5SM -SHSN BKN008 TEMPO 1618 2SM -SHSN BR OVC005 FM1800 02015G25KT 2SM -SN BR BKN003 TEMPO 1923 3/4SM SN BR VV002 FM2200 04017G30KT 1SM SN BLSN OVC003 TEMPO 2304 1/4SM SN BLSN VV001 FM0400 05014G25KT 1/2SM SN BLSN OVC003 WEATHER AROUND DESTINATION COS 201300 METAR 201254Z 36025G35KT 10SM FEW005 SCT060 SCT110 01/M02 A2955 RMK AO2 PK WND 36035/1252 SLP991 T00061022 CYS 201300 METAR 201256Z 35015G22KT 1SM -SN BR BKN011 OVC036 M06/M07 A2970 RMK AO2 PK WND 36031/1225 SLP058 P0000 T10561067 CYS 201322 SPECI 201310Z 36017G22KT 3/4SM -SN BR VV009 M06/M07 A2971 RMK AO2 TWR VIS 1 P0000 CYS 201332 SPECI 201330Z 36015G24KT 1/4SM PSN FZFG VV005 M06/M07 A2972 RMK AO2 TWR VIS 1 P0000 PUB 201300 METAR 201254Z 02004KT 10SM CLR 01/M03 A2949 RMK AO2 SLP960 T00111028 GJT 201300 METAR 201256Z 27007KT 3/4SM -SN BR OVC003 M01/M01 A2960 RMK AO2 SLP009 P0006 T10061006 GJT 201308 SPECI 201305Z 27008KT 1 1/4SM -SN BR OVC003 M01/M01 A2959 RMK AO2 P0001 GJT 201322 SPECI 201305Z 27008KT 1 1/4SM -SN BR OVC003 M01/M01 A2959 RMK AO2 P0001 CPR 201300 METAR 201255Z 02014G18KT 1 3/4SM -SN BR FEW016 OVC021 M04/M06 A2985 RMK AO2 SLP117 P0001 T10441061 CPR 201309 SPECI 201305Z 03016KT 2 1/2SM -SN BR FEW014 OVC023 M04/M06 A2986 RMK AO2 P0000 CPR 201326 SPECI 201316Z 03014KT 2 1/2SM -SN BR OVC035 M04/M06 A2986 RMK AO2 P0000 CPR 201349 SPECI 201346Z 03013KT 3SM -SN BR FEW018 OVC033 M04/M06 A2988 RMK AO2 P0000 ALTERNATE AIRPORT WEATHER SLC 201300 METAR 201256Z 33005KT 10SM FEW020 SCT060 BKN080 M04/M05 A2981 RMK AO2 SLP110 T10391050 57 NASA/CR—2002-211800 ORIGIN WEATHER ATL 201300 METAR 201253Z 27007KT 10SM FEW008 BKN026 BKN041 11/08 A2992 RMK AO2 SLP131 VIRGA SW AND NW T01060083 ATL 201400 METAR 201353Z 29006KT 10SM FEW012 SCT040 11/08 A2993 RMK AO2 SLP135 MDT CU DSNT NE T01110083 ENROUTE SURFACE WEATHER ATL 201400 METAR 201353Z 29006KT 10 SM FEW012 SCT040 11/08 A2993 RMK AO2 SLP135 MDT CU DSNT NE T01110083 MEM 201400 METAR 201353Z 26006KT 10SM FEW020 OVC033 07/04 A2996 RMK AO2 SLP144 T00720039 TUL 201300 METAR 201253Z 00000KT 4SM BR CLR 00/00 A2991 RMK AO2 SLP130 T00000000 OKC 201400 METAR 201353Z 14016KT 10SM CLR 07/02 A2978 RMK AO2 SLP085 T00670022 GOVERNMENT WEATHER ALERTS Z28 201251-201455 CONVECTIVE SIGMET 17E VALID UNTIL 1455Z SC AND CSTL WTRS FROM 40SE CLT-20NNW CHS-50SE SAV LINE TS 25 NM WIDE MOV FROM 24035KT. TOPS TO FL380. OUTLOOK VALID 201455-201855 FROM ORF-170E PBI-70ENE PBI-130SE MIA-80WSW EYW-100WSW SRQ-CAE-30NE HMV-ORF TS WILL CONT ALG/EAST OF CDFNT MOVG EWD ACRS THE AREA. ACT SHOULD BE MAINLY OVER THE CAROLINAS AND ADJ WATERS. WDLY SCT TS POSSIBLE IN MODERATELY UNSTABLE AMS OVER FL AND COASTAL WATERS. OCNL WST ISSUANCES ARE LIKELY.

Z35 201349-201555 CONVECTIVE SIGMET 18E VALID UNTIL 1555Z SC AND CSTL WTRS FROM 40NNE CAE-20E FLO-110SSE CHS-30SE SAV-40NNE CAE AREA TS MOV FROM 25030KT. TOPS TO FL410. OUTLOOK VALID 201555-201955 FROM ORF-170E PBI- 70ENE PBI-130SE MIA-80WSW EYW-100WSW SRQ-CAE-30NE HMV-ORF TS CONTG ALG/ EAST OF CDFNT MOVG THRU SRN ATLC CST STATES. MOST ACTV CNVTN EXPD TO BE MNLY OVER THE CAROLINAS AND ADJ WATERS. WDLY SCT TS POSS THIS AFTN IN MODLY UNSTABLE AMS OVER FL AND CSTL WTRS. OCNL WST ISSUANCES ARE LIKELY THRU MUCH OF PD.

58 NASA/CR—2002-211800

Appendix B. Airports and Navigation Aids

This appendix lists the airports and navigation aids appearing in the case study information.

The format of each entry is the identifier, the longitude, the latitude, and the English name.

Airports

ATL, -84.426944, 33.640444, “Atlanta” COS, -104.70025, 38.805806, “Colorado Springs” CPR, -106.464466, 42.908356, “Casper” CYS, -104.811838, 41.155723, “Cheyenne” DEN, -104.667, 39.85841, “Denver” GJT,-108.526735, 39.12241, “Grand Junction” MEM , -89.976667, 35.042417, “Memphis” OKC, -97.600734, 35.393088, “Oklahoma City” PUB, -104.496572, 38.289087 , “Pueblo” SLC, -111.977773, 40.788388, “Salt Lake City” TUL, -95.888242 , 36.198372 , “Tulsa”

Navigation Aid

AKO, -103.179740, 40.155578, “Akron” ALS, -105.815535, 37.349159, “Alamosa” BFF, -103.482022,41.894159, “Scottsbluff” BOY, -108.299712, 43.463152, “Boysen Reservoir” BTY, -116.747647, 36.800584, “Beatty” BZA, -114.60284, 32.768129, “Bard” CAE, -81.053904, 33.857249 , “Columbia” CHE -107.304893, 40.520084, “Hayden” CHS, -80.037811, 32.894313, “Charleston” CLT, -80.95175, 35.190289, “Charlotte” DIK, -102.773502, 46.859984, “Dickinson” DPR, -101.715071, 45.078175, “Dupree” DRK, -112.480349, 34.702556, “Drake” DVC, -108.931274, 37.80874, “Dove Creek” EED, -114.474104, 34.766004, “Needles” ELY, -91.830147, 47.821852, “Ely” EYW , -81.800476, 24.585878, “Key West” FMG, -119.656074, 39.531273, “Mustang” FMN, -108.098899, 36.748393, “Farmington” GCK, -100.725084 , 37.919067,“Garden City” GEG, -117.626889, 47.564944, “Spokane” GLD, -101.692306, 39.387861, “Goodland” HBU, -107.039792, 38.452153, “Blue Mesa” HLC, -100.22585, 39.258747, “Hill City” HMV, -82.129573, 36.437054, “Holston Mountain” ILC, -114.394226, 38.250193, “Wilson Creek” INW, 110.795027, 35.061602, “Winslow JNC, -108.792574, 39.059566, “Grand Junction” LAA, -102.687532, 38.197092, “Lamar” LAR, -105.720937, 41.337864, “Laramie” LBL, -100.9712, 37.0444, “Liberal” 59 NASA/CR—2002-211800 MIA, -80.278889 , 25.794722, “Miami” OCS, -109.015313 , 41.590214 ,”Rock Springs” ORF, -76.20033, 36.891897, “Norfolk” PBI, -80.0865, 26.680052, “Palm Beach” PIR, -100.162877, 44.394511, “Pierre” PUB , -104.429442, 38.294252, “Pueblo” SAV, -81.112505, 32.160554, “Savannah” SGF, -93.334052 , 37.355961, “Springfield” SHR, -107.061094, 44.842295, “Sheridan” SJN, 109.14352 , 34.424037, “St. Johns” SLC, -111.981913, 40.85025 , “Salt Lake City” SNY, 102.983, 41.09667, “Sidney” SRQ, -82.554264, 27.397765, “Sarasota” SSO, -109.263088, 32.269245,”San Simon” TBC, -111.269588, 36.121312, “Tuba City” TBE, -103.600056, 37.25866, “Tobe” TCS, -107.280542, 33.2825, “Truth or Consequences” 60 NASA/CR—2002-211800

Appendix C. Dispatcher Observation Plan

February 7-8, 2002 Delta Air Lines http://www.rap.ucar.edu/largedrop/integrated/terminal/ice_departures.html http://www.rap.ucar.edu/largedrop/iida http://www.rap.ucar.edu/largedrop/iifa http://adds.aviationweather.noaa.gov/projects/adds/icing/

Overview

We will be observing 2 dispatchers at two desks in the Delta OCC (one dispatcher at each desk). These desks have a high concentration of traffic at CVG. The observations will occur over two days.

This observational event supports a usability study using NCAR’s IIDA & IIFA. The observational study will occur in situ, while the dispatchers are doing their normal work.

Significant observations will be written down on paper. As appropriate, the observations will be supported by screen captures from the dispatcher’s monitors. If the dispatcher is in a high work-load condition doing the screen captures on his machine will be too disruptive, so as a fall-back we can capture the IIDA/IIFA screen the dispatcher is looking at with our laptops connected to the internet via modem. Screen captures can be accomplished by copying the screen to the clipboard using “Print Screen,” then copying the results into Paint (or the like), and saving to a file. The bitmap files will be rather larger (around 4Mb or so), but can be compressed down to < 100k using gzip (this program is gnu & easily contained on a floppy).

Sequence

• Observe dispatchers without IIDA/IIFA (baseline).

• Observe dispatchers using IIDA/IIFA without training.

• Introduce and train dispatchers on IIDA/IIFA.

• Observe dispatchers using IIDA/IIFA.

Assuming that icing decisions are not being made, we will pose icing related judgment questions to the dispatchers in order to observe if/how they use IIDA/IIFA.

Observation Goals

Our primary objectives will be to find out how the dispatchers understand the state of the weather (icing in particular), how IIDA/IIFA might impact how the dispatchers understand the state of the weather, how dispatcher use their understanding of icing to make job-relevant decisions, how IIDA/IIFA might impact these job-related decisions, usability issues with respect to IIDA/IIFA, what the dispatchers like/dislike about IIDA/IIFA, and what their suggestions are for the products.

61 NASA/CR—2002-211800

Observations

Without IIDA/IIFA -How do icing conditions impact your decision making? I.e. for what decisions do you need information about icing? What decision are most critical? What decisions are most difficult? What decisions are most easy?

-How do you determine if there are currently icing conditions? What tools do you use? Ask dispatcher to show how they do this.

-How do you determine if there will be icing conditions? What tools do you use? Ask dispatcher to show how they do this.

-How do you keep yourself aware of icing? Ask dispatcher to show how they do this.

-Do you try to discriminate between structural (hard) ice and engine (ice)? If so, what tools do you use to do this?

Ask dispatcher to show how they do this.

What do you like or dislike about the icing products you use? What other information would you like to have?

IIDA/IIFA without Intro/Training -Show dispatcher IIDA/IIFA and ask them what they think it show them. Allow dispatcher to use the tool. Look for misunderstandings, difficulties using, etc.

-Ask dispatcher how they think they might use such a tool for their job.

-Do you know what SLDs are? What is the importance of SLDs for aircraft?

-Visible moisture?

-What do the colors mean on the maps?

Train IIDA/IIFA IIDA/IIFA after training General Questions -Ask dispatcher again how they might use the product for their job. Any changes?

-Ask how he would use IIDA/IIFA, if at all, in concert with other products?

-Does the dispatcher understand the products? Is it clear that this is a likelihood value for potential icing (SLD if it is available)?

-What utility does potential icing have for them, if any? (As opposed to severity information, for example)?

-Utility of SLD? Utility of Visible Moisture? Icing/SLD/Cloud tops/bases?

- Is the SID/STAR concept useful or would another paradigm be preferable?

-Does the dispatcher view IIDA/IIFA when we don’t prompt him? What other icing relevant products does he use?

62 NASA/CR—2002-211800 -Does the dispatcher use IIDA/IIFA correctly? I.e. does the dispatcher make correct inferences based on IIDA/IIFA?

-What parts of the IIDA/IIFA tools does the dispatcher not use?

Terminal Scale Screen -Is the range of the map appropriate?

-Is the icing potential scale (i.e. 10% bins) appropriate?

-Are the dispatchers able to discriminate between colors?

-Are the dispatchers able to understand what the pirep symbols mean?

-Are the lines representing the routings on the display useful?

-What problems does the dispatcher have navigating the tool?

-Color of text?

-Position/layout of items on display?

-Size of items on display?

-Would you to be able to select a likelihood threshold?

-Performance of tool (does it load quickly, are the any problems displaying pages)?

-Are links to other products used?

Vertical Cross Sections -Is the range of the map appropriate?

-Is the icing potential scale (i.e. 10% bins) appropriate?

-Is the height scale appropriate?

-Are the dispatchers able to discriminate between colors?

-Are the dispatchers able to understand what the pirep symbols mean?

-Position/layout of items on display?

-Size of items on display?

-Would you to be able to select a likelihood threshold?

-Performance of tool (does it load quickly, are the any problems displaying pages)?

Horizontal Cross Sections -Are 2000 ft. increments appropriate?

-Is the range of the map appropriate?

-Is the icing potential scale (i.e. 10% bins) appropriate?

63 NASA/CR—2002-211800 -Are the dispatchers able to discriminate between colors?

-Are the dispatchers able to understand what the pirep symbols mean?

-Position/layout of items on display?

-Size of items on display?

-Would you to be able to select a likelihood threshold?

Icing Tops/Bases -Is scale appropriate?

-Is the range of the map appropriate?

-Are the dispatchers able to discriminate between colors?

-Are the dispatchers able to understand what the pirep symbols mean?

-Position/layout of items on display?

-Size of items on display?

Other Products -Show http://adds.aviationweather.noaa.gov/projects/adds/icing/ Debrief -IIDA/IIFA likes/dislikes?

-Dispatcher suggestions?

64 NASA/CR—2002-211800

References

ASA (1999). Aeronautical Information Manual . Newcastle, WA: Aviation Supplies & Academics, Inc.

Aviation Safety Network (1998). Ice Accumulation Abstracts .

www.geocities.com/CapeCanaveral/9575/wxi.htm Boeing (2001). Statistical Summary of Commercial Jet Airplane Accidents: Worldwide Operations 1959-2000. Seattle, Washington: Boeing Commercial Airplanes.

Cornman, L.B. and Sharman, B. (1999). In-flight turbulence. In Proceedings of Advances in Aviation Safety Conference , 45-49.

FAA (1996). Airworthiness Directive 96-09-25 . Washington DC: FAA.

FAA (2000). Integrated Icing Diagnostic Algorithm (IIDA) Assessment at Regional Airlines – Final Report . Atlantic City, NJ: William J. Hughes Technical Center.

Green, S. (1998). Inflight icing: The handling event. Air Line Pilot , 2, 10-15.

Hansman, R. J. & Wanke, C. (1989). Cockpit display of hazardous weather information. In the th Proceedings of the 27 Aerospace Sciences Meeting , 1-8.

Jeck, R.K. (1996). Representative Values of Icing-related Variables Aloft in Freezing Rain and Freezing Drizzle. DOT/FAA/AR-TN95/119.

Kelsch, M. and Wharton, L. (1996). Comparing PIREPs with NAWAU turbulence and icing forecasts – Issues and results. Weather and Forecasting , 11(3), 385-390.

Lankford, T. T. (1995). Pilot’s Guide to Weather Reports, Forecasts, and Flight Planning . 2nd Edition, New York, NY: McGraw Hill.

Lankford, T. T. (2000). Aircraft Icing . New York, NY: McGraw Hill.

McDonough, F. & Bernstein, B.C. (1999). Combining satellite, radar, and surface observations with model data to create a better aircraft icing diagnosis. The 8th Conference on Aviation, Range and Aerospace Meteorology , Dallas TX, 10-15 January. Amer. Meteor.

Soc., Boston, 467-471.

Myszkowski, B. & Rezsonya, L. (1996). Aircraft Icing Considerations . (Part of Dispatcher Training course). Atlanta, GA: Delta Air Lines.

NASA (1998). Commuter and GA icing incidents. Aviation Safety Reporting System Database Report Set. http://asrs.arc.nasa.gov/report_sets/icing.pdf 65 NASA/CR—2002-211800 NTSB (1996). In-flight Icing Encounter and Loss of Control, Simmons Airline, d.b.a. American Eagle Flight 4184, Avions de Transport Regional Model 72-212, N401AM, Roselawn, Indiana, October, 31, 1994 . Aircraft Accident Report NTSB/AAR-96/01.

NTSB (1998). In-flight Icing Encounter and Uncontrolled Collision with Terrain, Comair Flight 3272, Embraer EMB-120RT, N265CA, Monroe, Michigan, January 9, 1997 . Aircraft Accident Report NTSB/AAR-98/04.

Pobanz, B.M., Marwitz, J.D. and Politovich, M.K. (1994). Conditions associated with large-drop regions. Journal of Applied Meteorology , 33, 1366-1372.

Politovich, M.K. (1989). Aircraft icing caused by large supercooled droplets. Journal of Applied Meteorology , 28, 856-868.

Politovich, M.K. (1996) Response of a research aircraft to icing and evaluation of severity indices. Journal of Aircraft , 33 (2), 291-297.

Politovich, M.K., & Thompson, G. (1996) A meteorologically-based icing severity index. In Proceedings of the FAA International Conference on In-Flight Icing, Volume II: Working Group Papers . Springfield, VA, 6-8 May. 375-385.

Riley, J., Lindholm, T., Politovich, M., Brown, B. & Strapp, W. (1999). Prospects for the Acquisition of Icing Data from Operational Aircraft . DOT/FAA/AR-99/66. Washington DC: FAA.

RTCA (1995). Final Report of RTCA task Force 3: Free Flight Implementation . October 26, 1995.

Washington, DC: RTCA, Inc.

Ryerson, C. (2000). Remote Sensing of In-Flight Icing Conditions: Operational, Meteorological, and Technological Considerations. ERDC-CRREL M-00-1. US Army Cold Regions Research and Engineering Laboratory: Engineer Research and Development Center.

Schwartz, B. (1996). The qualitative use of PIREPs in developing aviation weather guidance products. Weather and Forecasting , 11(3), 372-384.

Vigeant-Langlois, L. (2000). Cockpit weather information system requirements for flight operations in icing conditions . Unpublished Master’s Thesis: Massachusetts Institute of Technology.

66 NASA/CR—2002-211800 Form Approved REPORT DOCUMENTATION PAGE OMB No. 0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503.

2. REPORT DATE 3. REPORT TYPE AND DATES COVERED 1. AGENCY USE ONLY ( Leave blank) Final Contractor Report August 2002 4. TITLE AND SUBTITLE 5. FUNDING NUMBERS Aircraft Icing Weather Data Reporting and Dissemination System WU–708–20–13–00 6. AUTHOR(S) C–74790–N Ellen J. Bass, Brian Minsk, Tenny Lindholm, and Marcia Politovich 8. PERFORMING ORGANIZATION 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) REPORT NUMBER Search Technology, Inc.

4960 Peachtree Industrial Boulevard E–13495 Suite 230 Norcross, Georgia 30071–1580 10. SPONSORING/MONITORING 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) AGENCY REPORT NUMBER National Aeronautics and Space Administration Washington, DC 20546– 0001 NASA CR—2002-211800 11. SUPPLEMENTARY NOTES Ellen J. Bass and Brian Minsk, Search Technology, Inc., 4960 Peachtree Industrial Boulevard, Suite 230, Norcross, Georgia 30071–1580; Tenny Lindholm and Marcia Politovich, National Center for Atmospheric Research, Research Applications Program, P.O. Box 3000, Boulder, Colorado 80301. Project Manager, Andrew Reehorst, Turbomachinery and Propulsion Systems Division, NASA Glenn Research Center, organization code 5840, 216–433–3938.

12b. DISTRIBUTION CODE 12a. DISTRIBUTION/AVAILABILITY STATEMENT Unclassified - Unlimited Subject Category: 03 Distribution: Nonstandard Available electronically at http://gltrs.grc.nasa.gov This publication is available from the NASA Center for AeroSpace Information, 301–621–0390.

13. ABSTRACT (Maximum 200 words) The long-term operational concept of this research is to develop an onboard aircraft system that assesses and reports atmospheric icing conditions automatically and in a timely manner in order to improve aviation safety and the efficiency of aircraft operations via improved real-time and forecast weather products. The idea is to use current measurement capabilities on aircraft equipped with icing sensors and in-flight data communication technologies as a reporting source. Without requiring expensive avionics upgrades, aircraft data must be processed and available for downlink. Ideally, the data from multiple aircraft can then be integrated (along with other real-time and modeled data) on the ground such that aviation-centered icing hazard metrics for volumes of airspace can be assessed. As the effect of icing on different aircraft types can vary, the information should be displayed in meaningful ways such that multiple types of users can understand the information. That is, information must be presented in a manner to allow users to understand the icing conditions with respect to individual concerns and aircraft capabilities. This research provides progress toward this operational concept by: identifying an aircraft platform capable of digitally capturing, processing, and downlinking icing data; identifying the required in situ icing data processing; investigating the requirements for routing the icing data for use by weather products; developing an icing case study in order to gain insight into major air carrier needs; developing and prototyping icing display concepts based on the National Center for Atmospheric Research’s existing diagnostic and forecast experimental icing products; and conducting a usability study for the prototyped icing display concepts.

14. SUBJECT TERMS 15. NUMBER OF PAGES Aviation meteorology; Aircraft icing; Airline operations; Commercial aircraft; Nowcasting; 16. PRICE CODE Numerical weather forecasting; Cockpit weather information systems 19. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION 17. SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACT OF ABSTRACT OF THIS PAGE OF REPORT Unclassified Unclassified Unclassified NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89) Prescribed by ANSI Std. Z39-18 298-102