Mr. Chairman, as you know, the Board is holding a comprehensive factfinding hearing beginning November 27, 1978, to determine all of the facts, conditions, and circumstances surrounding the San Diego accident. The Board will analyze these facts and determine probable cause, and will submit recommendations to the FAA and others as to the steps that should be taken to prevent this type of accident from reoccurring.

The NTSB appreciates today's demonstration of your continued. dedication to safety. We are here today at your request to present our views on collision avoidance, a complex and critical subject on which we have recommended action over the past 11 years.

We have been asked to focus on the development and implementation of collision avoidance system technologies, the problems created by the mix of general aviation and commercial air carrier traffic at major U.S. airports, and the utilization and development of reliever airports to reduce congestion at major traffic centers. Lastly, Mr. Chairman, we would like to respond to your request for an account of the facts the Board has adduced to date concerning the accident which occurred in this city.

As you know, Mr. Chairman, only a few hours after the tragedy occurred our investigators were on the scene of this Nation's worst civil air disaster here in San Diego on September 25. The fact gathering which started then is continuing and will include a public factfinding hearing to be held in this city during the week of November 27. It will take us until then to assemble and prepare the necessary data needed to have a full and complete discussion of the facts and issues of the accident. After the hearing, the Board will complete a rigorous and exhaustive analysis of the facts, and then issue a full report which, in the language we use, will identify the probable cause or causes of this accident. Our recommendations for improving safety may be made at any time during the investigation, analysis, or preparation of probable cause.

With respect to the San Diego midair collision, these are the facts as we know them:

PSA flight 182 was a regularly scheduled flight on an instrumentflight-rules plan from Sacramento to San Diego with an intermediate stop at Los Angeles. The flight was making its approach to San Diego under the control of the Miramar radar air traffic control facility. After advising Miramar that they were descending out of 9,500 feet to 7,000 feet, and that they had the airport in sight, PSA 182 was cleared for a visual approach to runway 27 and given traffic advisories regarding a Cessna 172 making a practice instrument-landing-system approach, ILS, to runway 9. PSA 182 acknowledged the advisories and informed approach control that they had the traffic in sight.

The Cessna 172, N7711G, was operated by Gibbs Flite Center based at Montgomery Field for instrument qualification of a commercial pilot accompanied by a certified flight instructor. After completing the simulated instrument approach to runway 9, the Cessna 172, also under the control of Miramar, was cleared to climb to an altitude of below 3,500 feet on a northeasterly heading of 070 degrees. Twenty-two seconds after 9 a.m., the PSA flight was instructed by Miramar to maintain visual separation and contact the Lindbergh tower. Approximately 12 seconds later PSA did contact Lindbergh

tower and was then advised, "PSA 182, Lindbergh tower, traffic 12 o'clock, 1 mile, a Cessna."

Seven seconds later, PSA 182 replied, "OK, we had him there a minute ago." This transmission was acknowledged by the tower 3 seconds later. Two seconds later, at 49 seconds after 9 a.m., PSA again transmitted to the tower, "Think he's passed off to our right." The tower acknowledged this transmission 2 seconds later.

Two seconds after that, at 53 seconds after 9 o'clock, the tower asked PSA 182: "How far are you going to take your downwind, 182? Company traffic is waiting for departure." Four seconds later PSA replied, "Oh, about 3 to 4 miles." Six seconds past 9:01, Lindbergh tower cleared PSA 182 to land; 2 seconds later PSA acknowledged the landing clearance. Twenty seconds later-9:01:28-the conflict alert system at the Miramar approach control facility activated and remained activated until 9:01:59.

At 9:01:46, Miramar approach control transmitted to the Cessna, "Cessna 11 Golf, traffic's in your vicinity. A PSA jet has you in sight, descending for Lindbergh." The collision occurred at approximately 9:01:45. Since Miramar has no recording capability for its automated radar tracking system-ARTS III-it has been necessary for our investigative staff to use recorded radar data from the Los Angeles air route traffic control center located at Palmdale in order to determine if we can precisely identify what other traffic may have been in the area just before and at the time of the collision.

Mr. Chairman, I have with me charts and other materials from other accidents which if any of the committee is interested or would like to be updated on what we're trying to develop from this accident so they might understand fully the extent of time spent on our investigation, we would be more than glad to provide that for them either during this hearing or after or at their convenience, sir.

Mr. GOLDWATER. Would the gentleman explain what that has to do with your testimony.

Mr. KING. There has been from time to time question as to the tying the gap in time between the accident and the completion of our investigation. That has come up. What I want to do is to anticipate that if possible, and say that we do have materials we would be willing to share with you if you would find it useful.

Mr. GOLDWATER. You mean the question on why it appears you're dragging your feet?

Mr. KING. I couldn't imagine that it appears that way. But if there were a question as to what the procedures are that we go through, we would like to be responsive to you.

A total of 144 persons died, 135 persons in the 727, 2 in the Cessna, and 7 on the ground. In addition, 16 persons were injured on the ground.

During the 22 years since the Grand Canyon midair collision between two air carriers, U.S. civil aviation has had 537 midair collisions of which 291 have resulted in fatal injuries to the occupants of one or both aircraft. Midair collisions have accounted for 1,331 fatalities, or 4.3 percent of the total fatalities from all U.S. civil aviation accidents during this period.

Twenty-five of the accidents involved an aircraft operated by a major air carrier; 18 of these were collisions with a light airplane, 5 involved collision with a military airplane, and 2 were collisions of an

air carrier aircraft; 12 of the 25 accidents resulted in fatalities to occupants of the air carrier aircraft, a total of 574.

Thus, while the vast majority of midair collision accidents and fatalities occur between two general aviation aircraft, the high occupancy rate of air carrier aircraft means that a large number of the fatalities have occurred in those relatively few accidents involving air carrier aircraft. During this 22-year period, midair collision accidents have accounted for 9.9 percent of the total number of fatalities resulting from air carrier accidents.

There may be a tendency for some to take these statistics as evidence that the midair collision is not a serious problem. This contention might find even more support since when we look closer we find that over half of the midair accidents involving air carrier aircraft happened between the years 1965 and 1972, a period characterized by rapid growth in air carrier as well as general aviation with perhaps a corresponding lag in the introduction of new air traffic control concepts.

The San Diego accident is the first such accident since June 29, 1972. We of the Safety Board certainly recognize that this is no consolation to those who lost loved ones in that accident. Nor is anyone, particularly those of us whose business is transportation safety, satisfied based on the statistics that the midair collision is no longer a potential threat to air carrier aircraft. The PSA accident proved otherwise.

Too, the statistics would have told a much different story if two wide-bodied aircraft had collided with several hundred passengers aboard. We may remember that this almost happened over Carleton, Mich., when several passengers in a climbing DC-10 were injured as it took evasive action to avoid colliding with an L-1011 which was cruising at 35,000 feet. These two aircraft carried 319 passengers and

crew.

Nonetheless, we do believe that the avoidance of major midair accidents over the past 7 years, a period during which the miles flown by air carrier aircraft has remained nearly constant and the miles flown by general aviation aircraft has risen from less than 3%1⁄2 billion miles to nearly 5 billion miles, is not purely coincidental. Certainly, the expansion of radar coverage throughout the national airspace system, the installation of automated radar-tracking services at major terminal facilities, the further addition of systems to alert controllers to potential conflicts, the expansion of positive control airspace have all had an effect on the air carrier aircraft exposure to the collision threat.

Whether or not more positive actions might be taken during this period, actions which would have prevented the PSA collision with the general aviation aircraft on September 25 is one of the questions which the Board will explore in depth at the hearing beginning on November 27, and will attempt to answer in its analysis of the facts of the accident following this hearing.

Several of the accidents involved one aircraft operating under an instrument flight rules flight plan, mostly the air carrier aircraft, which collided with an aircraft operating under visual flight rules, not under air traffic control.

The Safety Board has been concerned for many years that the "see-and-avoid" philosophy still followed in aviation, although it works most of the time, cannot be considered anywhere near 100

percent reliable to assure traffic separation, particularly in an area where there is a mix of controlled and uncontrolled traffic and a wide difference in aircraft performance. The human vision capacity, cockpit window restriction, visual illusions, atmospheric conditions, and aircraft noticeability are all factors affecting the pilot's ability to see another airplane. We know that the largest percentage of midair collisions involving general aviation aircraft occur in clear weather during daylight hours and at or near an uncontrolled airport.

Most of the Board recommendations based on midair collisions have recognized the limitations and reiterated our continuing concern about the see-and-avoid philosophy. We have constantly recommended actions directed toward supplementing the see-and-avoid concept with more positive separation methods. These have included the reduction of traffic mix through increased positive control; the development of collision avoidance hardware; the enhancement of aircraft noticeability, review, and improvement of air traffic control procedures; and other pertinent actions. Throughout this period, the Safety Board has recognized that the see-and-avoid concept was not likely to be completely replaced by positive control at all locations. Therefore, many of our recommendations dealt with calling for increased pilot awareness and positive training in visual scan techniques. In any event, the soundness of the see-and-avoid concept will also be an issue that will be addressed in some depth during the Board's hearing of the facts of this accident.

We have monitored the development of collision avoidance systems for the last 10 years. First the compatible airborne system, then the beacon-based system. We have urged technological development of a system which would automatically detect an impending collision and command the pilot to execute an evasive maneuver without need for human interpretation.

Such collision avoidance systems can make very positive contributions to safety and should be pursued, but even they will not provide a panacea for midair collisions. Other positive programs are also needed.

Regardless of the factors determined causal in the San Diego accident, we must take a closer look at present and predicted airport congestion and particularly the compatibility of general aviation and air carrier aircraft. As the environment becomes more saturated, complexities caused by a mix of users must be understood and managed.

The substantial growth of traffic into the hub areas will require constant surveillance to assure that air traffic control equipment and procedures are upgraded as necessary to keep pace with the problem. San Diego's Miramar facility last year handled almost 1 million flights. The facilities and equipment to handle increases in traffic must be provided as needed if we are to avoid an increase in risk to the traveling public.

Thus far I have addressed the midair collision as it involves air carrier aircraft. I believe that I would be remiss if I did not speak to the Safety Board's concern about collisions on the ground. Everyone, I am sure, is very much aware that the worst collision involving two air carrier aircraft occurred, not in the air, but on the ground at Tenerife in the Canary Islands on March 27, 1977. In this accident, 583 persons were killed.

We of NTSB participated in that investigation. And as you're aware, the Spanish Government released the results of that investigation just last week.

A summary of that, very briefly: A KLM B747 began its takeoff roll in restricted visibility and struck a Pan American B747 which was taxiing down the active runway. The Safety Board investigated an accident having similar circumstances at the Chicago O'Hare Airport in December 1972 in which there were 10 fatalities. We are currently investigating an incident which occurred at La Guardia Airport on June 21, 1978, involving a DC-9 with 93 occupants and a Cessna Citation. The local controller had cleared the Cessna for takeoff while ground control had cleared the DC-9 to backtrack on the same

runway.

And, I think as recently as yesterday, we had a DC-8 and a Sabreliner at LAX in yet another incident.

Although solutions to the ground control problem are not directly related to the airborne-collision problem, we believe that any collision may present problems of great magnitude. Unfortunately, the risk of both will grow as the number of operations into the hub areas continues to increase.

The Vice Chairman of the Board, Mr. Elwood T. Driver, testified before the Subcommittee on Oversight of the House Ways and Means Committee on two separate occasions regarding the airport and airway trust fund and its impact on airport safety. The Safety Board's testimony was given on May 19 and September 21 of this year. Similar views were presented earlier by the Safety Board at hearings on January 24, 1978, before that same committee of the House. And I reiterated those views in testimony before your subcommittee, Mr. Anderson, on May 16 of this year.

One of the major points of the Safety Board's testimony on these occasions was the projected growth of aviation and the strain which has been placed upon the existing airports and airway systems. We strongly suggested that FAA work with the Congress to reexamine the funding limitations of the airport development aid program to assure proper emphasis on a major program to increase the capacity of existing airports and to build new ones if the need be. We advocated the development of reliever airports which are equipped with allweather capabilities.

Mr. Chairman, we will provide that testimony for the record of this hearing with your permission, sir.

The CHAIRMAN. It will be made a part of the record by reference. Mr. KING. Thank you.

As I have mentioned, the vast majority of midair collisions, about 30 per year, occur between two general aviation aircraft on a clear day in the vicinity of an uncontrolled airport. We can well envision that as traffic into the major hubs increases, so too will commuter and air taxi operations into these small, ill-equipped airports which offer needed service to smaller communities. To our knowledge, there is no program effort to insure that these smaller airports have adequate facilities to serve their increasing traffic.

In summary, the Safety Board recognizes that collision avoidance. both in the air and on the ground is a complex problem. We recommend action in five areas.