Remarks by Jim Hall, Chairman
National Transportation Safety Board
before the National Weather Service Workshop
on Weather Services for Aviation Operations
Anchorage, Alaska, April 9, 1998

Thank you for inviting me to address this important workshop dealing with an issue that has been on the "front burner" for my agency for many years, how to mitigate the effects of weather on aviation safety.

When I became Chairman of the National Transportation Safety Board almost 4 years ago, I of course had no inkling about the level of activity the agency was about to enter. July 1994 saw the end of an unprecedented 27-month period in which the major U.S. scheduled airlines incurred no passenger fatalities: a billion passengers carried safely to their destinations.

The USAir accident in Charlotte, North Carolina broke that string, and significant for this audience it was a weather-related accident. A crash later that year turned out to be another weather-related accident, this time icing. Crashes involving a Boeing 737 near Pittsburgh, a DC-9 in the Everglades and a Boeing 747 off Long Island in the succeeding years were among the more prominent in a string of accidents that comprised the extraordinary level of activity I was talking about.

This is my fourth visit to Alaska since I’ve become Chairman. One trip related to the Alaska pipeline, but the rest have dealt with Alaska’s aviation concerns. I was here in 1994 to chair a hearing that gathered information for a safety study the Board was conducting on air service to Alaska. I returned in early 1995 to present the results of that study to a gathering of Alaska’s air carriers. I am happy to return now to discuss our many initiatives in dealing with weather as a factor in aviation accidents.

It has been said that meteorology is one of the oldest sciences; it certainly is one we all think we know something about. And it is a science that has enough mystery for us that we can easily be impressed by those who appear to have a special insight.

There is a story about a movie being filmed on location in the desert Southwest. The entire company was amazed at the unfailing weather predictions made by an old Indian. He was consulted daily and his forecasts proved reliable. But one day he refused to predict the weather. "Is anything wrong?" he was asked.

"Yes," the old gentleman said, "the radio broke."

Despite the many technological advances in recent decades, it is humbling that there is much about the weather we still don’t know. It was only within the last 25 to 30 years that the windshear phenomenon was identified, only in the past decade that the El Nino phenomenon became more understood, and only within the last 4 years that extensive research has been conducted into the phenomenon of super cooled large droplet icing.

Throughout the Board’s history, it has made recommendations that resulted in research and regulatory action that increasingly has lifted the veil from the mystery of weather. This continues a tradition of accomplishment begun by the Civil Aeronautics Board in watershed investigations of weather-related accidents like the crash of the Northwest Airlines B-720 in the Everglades and the crash of the Pan American B-707 in Elkton, Maryland, both in 1963.

One of the major investigations inherited by the NTSB from the CAB was the 1966 destruction of a Braniff Airlines BAC-111 by severe convective weather in Nebraska. This and other investigations set the stage for landmark windshear research in the 1970s and 1980s. And even now, the industry continues to confront icing as a major concern to aviation safety.

In some ways, Alaska’s aviation environment is the most challenging in the nation. This is by far our largest state, our northernmost state, and our westernmost state, and contains the highest mountain in North America. At the same time, its population grew 10 percent in the last 7 years, a rate almost half again as much as experienced by the rest of the United States.

Unfortunately, the world thinks of aviation in Alaska only when high profile accidents occur, like Will Rogers 60 years ago, or Congressman Hale Boggs 25 years ago, or the Ryan Air commuter crash in Homer 10 years ago. But Alaska’s aviation system has grown to be a complicated network of air taxis that keep the State’s economy moving.

Alaska experiences all the aviation weather hazards found in the lower 48 with a greater risk of volcanic eruptions than almost anyplace else in the world. Flight operations and weather forecasting are challenged by the state’s geography – rough terrain, vast land area, exposure to ocean – extremely variable weather, and limited weather radar and satellite coverage as compared to that enjoyed by the rest of the country.

Due to the state’s geography and weather, there are few ground transportation links. This makes the state extremely dependent on air transportation, unmatched anywhere else in the United States. Compared to the remainder of the U.S., the State has 6 times as many pilots, 14 times as many aircraft, and 76 times as many commuter airline flights per capita. Food, fuel, and school children are regularly transported by air. Medical evacuation flights from remote areas are commonplace. The need to use aviation to meet these transportation requirements places pressure on the air transportation system to perform on time, all year, and frequently in adverse weather.

This air system could not operate as successfully as it does without the work of the National Weather Service, in cooperation with the FAA, air carriers, pilots and other users. That is why we have worked so closely with all of you for so many years. I would like to describe some particular challenges we face in Alaska.

Volcanic ash: There are currently 42 active volcanoes in Alaska. We recognized that the NWS in Alaska has taken an active role in developing plans, procedures, and forecasting techniques for alerting flight operations of volcanic eruptions and ash cloud movement. Some years ago, a KLM Royal Dutch Airlines 747 encountered an ash cloud near here and all four engines lost power. Fortunately, the engines were restarted by the time the aircraft had descended to 13,000 feet, avoiding a catastrophe.

Turbulence has been of significant interest to the Board. Strong low level winds over mountains here, especially in the winter, frequently generate severe turbulence. A Japan Airlines 747 experienced severe turbulence shortly after takeoff from Anchorage 5 years ago, causing the number 2 engine pylon to separate from the airplane.

Icing would tend to be an obvious problem here in Alaska. This has been an issue on the Board’s Most Wanted list of safety improvements. The 1994 crash of a commuter airliner in Indiana caused the industry to launch a major initiative into airframe icing. This has resulted in a better understanding of the Supercooled Large Droplets (SLD) airframe icing hazard. Some of the recent activities as a result of the Board’s consequent recommendations include:

• a new FAA in-flight icing plan,

• weather forecasts issued by the NWS now contain specific mention of areas containing Supercooled Large Droplets,

• NWS and others are conducting scientific research to improve forecasting methods for airframe icing conditions, and

• NASA is conducting a series of flight tests in actual icing conditions to more accurately define the icing hazard posed by Supercooled Large Droplets.

But the icing problem persists. The Board is continuing its investigation of the January 1997 crash of a Comair commuter airliner near Detroit. Icing is a major issue in that investigation.

VFR into IMC: The problem of Visual Flight Rules flying into Instrument Meteorological Conditions is a nationwide problem, but even more so in Alaska. Flying weather can be quite variable in Alaska depending on the area of the state and the time of the year. While all parts of Alaska experience periods of IMC, the Aleutian Islands are recognized as having some of the worst flying conditions in the world. Other areas that frequently experience IMC include the Alaska Peninsula, southeastern Alaska and the Arctic coast during the summer and fall.

Alaska’s geography creates a myriad of flying problems. Innumerable small-scale climates exist near mountainous terrain, mountain passes and glaciers. Widely separated weather reporting stations and limited satellite imagery allow these localized conditions to go unreported. Also, mountain tops are frequently obscured by clouds, and the freezing level is rarely above 7,000 feet. These factors often cause significant problems for the VFR pilot.

Unlike in the lower 48, Alaska does not have an adequate system for flight under IFR at the low altitudes used by general aviation and the bulk of the commercial air carriers in the state. As a result, aviation in Alaska is more highly dependent on the weather facilities and products needed to make VFR flying safe in this difficult environment:

• Needed improvements include maintenance and expansion of the number of weather observation facilities;

• skilled weather observers who can augment the automated observations with the information needed by VFR pilots, such as conditions in nearby passes;

• graphical weather products;

• traditional information sources like flight service stations and private radio networks; and

• newer means of disseminating information, such as the Internet.

With these concerns in mind, the Board tackled the broad subject of safety issues concerning air service in Alaska in the 1995 study I mentioned earlier. Four of the 23 recommendations were addressed to the National Weather Service. The first dealt with automated observations. We recommended that the NWS and the FAA ensure that, where there are currently qualified weather observers on site, operationally significant information, including distant weather information, should be manually added to automated weather observations until technological progress eliminates the need for them.

As I’m sure you know, the implementation of automated surface observing systems (ASOS and AWOS) and the level of human augmentation of those systems have been contentious issues not only here in Alaska but throughout the aviation industry. We understand that these systems are here to stay and we agree that their reliability and accuracy are improving. While there are many advantages to the automated systems, the fact that they essentially measure weather conditions at a point and cannot provide an indication of the weather in the surrounding area around an airport is a key limitation to the systems.

During the 1995 study, it was emphasized that the large majority of flight operations in Alaska are conducted under visual flight rules by single-engine aircraft and that pilots need as much information as possible about flight conditions between airports because of the variability of the weather and the vast distances between weather reporting stations. Information on matters like fog banks on the end of runways, mountain obscurement, or the presence of rotor clouds is vital and we believe that if there are certified observers present at an airport, they should be encouraged to report that information.

The FAA and NWS responses to this recommendation have been mixed so far. The FAA would not take any further action on augmentation, so the Safety Board was forced to close out its recommendation to the FAA as unacceptable action. However, some progress is being made by the NWS. Recently, the NWS said that it will issue a change to its handbook to specifically state that remarks deemed operationally significant by the observer or air traffic controller may be augmented at all locations. This is a very significant change in the NWS’s position, and we welcome it.

The second recommendation called for the provision of graphical weather products on the Direct User Access Terminal System (DUATS), on the Internet and to commercial vendors. At the time of our study, the NWS was in the process of implementing the Alaska Aviation Weather Unit. The unit was to produce weather graphics specially tailored for aviation in Alaska and disseminate the products to the NWS and AFSS offices. The graphics represented a major improvement over what was currently available and the Board felt that the graphics should receive wider dissemination to the flying community. Since then, the NWS has provided the graphics over the Internet and the FAA will incorporate them into the DUAT portion of the Operational and Supportability Implementation System (OASIS) next year.

The third recommendation asked the NWS and the FAA to evaluate the technical feasibility and aviation safety benefits of remote color video weather observing systems. There has been ongoing interest in video observing systems in Alaska for the last 20 years. The Canadians, however, have been far more successful in implementing video technology to provide supplementary information for automated weather observation sites or information about specific phenomena at non-airport locations. We have been informed by the NWS that a national test program is underway to evaluate the benefits of remote color video systems with one of the systems installed here in Alaska.

We are pleased to learn that Senator Stevens has been able to secure funding for four additional video systems and that the FAA is currently conducting site evaluations for their implementation. We also urge the FAA to continue to work with Alascom to make their remote video network available to the Alaskan aviation community.

Finally, the fourth recommendation asked the Weather Service to revise its policies to provide mike-in-hand radio service for aviation weather information at NWS and contract weather offices. We regret that in this age of budgetary difficulties, forcing many to have to do more with less, the NWS refused to consider allowing its personnel to provide this added service to pilots in Alaska.

In addition to the mike-in-hand recommendation addressed to the NWS, the safety study recommended that the state of Alaska establish a training program to enable mike-in-hand, also called near real-time, reports of airport conditions by designated state and contractual airport maintenance personnel. The state has taken several actions to date, including providing radios to operators of snow removal equipment at airports with the highest traffic volume, developing new radio procedures for these operators, and providing training in the new procedures. These are significant first steps.

The Board’s primary objective was to improve the timeliness and accuracy of airport condition reports at the far more numerous small village airports. We are pleased that the state is taking additional action to fulfill the intent of this recommendation through its participation in the "near NOTAM" program and to continue its program of radio acquisition.

In closing, I’m sure you will agree with me that Thomas Jefferson might have described our missions best when he said, "The care of human life and happiness is the first and only legitimate object of good government." This is how all of us responsible for the safety of our citizens should view our jobs. I commend the National Weather Service for all it has done to give the pilots of Alaska, and of the nation at large, the information they need to provide safe and efficient transportation. And I commend all of you for attending this conference and for your work in this important area.

I thank you for your attention. May we have a safe remainder of 1998.


Jim Hall's Speeches