-From Oscar Bakke, Acting Administrator: This is in response to your September 15 denial of our request for a postponement of the public release of the Board accident report on the midair collision near Fairland, Indiana. You base your denial on the absence of any provision in the Board procedures for postponing the report.
Your reason for denying our request is surprising since your predecessor Board granted an earlier request by FAA of an identical nature under the same procedural regulations, with respect to the 8 February 1965 accident near Jones Beach, New York. Upon recalling that report prior to publication, the Board granted the FAA a hearing on the matter and then issued a revised report.
Your letter suggests that we analyze in depth the Fairland report and the "other Board reports referred to" in our letter, and then formally request reconsideration of the Board conclusions and probable cause. While we are pleased to accept your suggestion, we cannot resist noting that it conflicts with your denial of our initial request since the Board procedural regulations similarly lack any provision for the Board entertaining such a request for reconsideration.
Prior to making our postponement request, we had compared the Fairland report with the only other Board report referred to in our letter, the one on the Urbana, Ohio midair collision of 9 March 1967. Accepting the facts as they are stated in the two Board reports, we offer the following comparison accompanied by our comment:
Urbana: Air carrier DC-9 and a Beech Baron BE-55.
Fairland: Air carrier DC-9 and a Piper Cherokee PA-28.
(Comment: No substantial differences in the aircraft involved in the one accident as compared to the other.)
Urbana: Clear visibility six or seven miles, a haze layer, snow coverage of 80% to 90%.
Fairland: Visibility in excess of 15 miles, no haze, no Snow coverage.
(Comment: Weather conditions in the Fairland accident were more favorable for the detection of other aircraft.)
Urbana: The DC-9 was descending at 3500 feet per minute for 20 seconds, and overtaking and converging on the Beech Baron from the left. The Baron was level at approximately 4500 feet MEL.
Fairland: The DC-9 was descending at approximately 2400 feet per minute and converging from the left while the Cherokee was converging from the right. The
Cherokee was level at approximately 3500 feet MSL.
(Comment: The rate of descent of the DC-9 in the Fairland accident was substantially less than the rate of descent of the DC-9 in the Urbana accident.)
Urbana: The DC-9 was operating at 323 knots while the Baron was operating at approximately 193 knots.
Fairland: The DC-9 was operating at a gradually increasing airspeed from 236 knots to 256 knots with the Cherokee operating at 107 knots.
(Comment: The speed of the DC-9 in this accident was substantially less than the DC-9 in the Urbana accident.)
Urbana: The DC-9 had a white top, a red stripe on the fuselage, a black radome, the rest unpainted aluminum. The Baron's wings and lower half of the fuselage were painted red and the upper half painted white.
Fairland: Aircraft colors are not mentioned in the report.
(Comment: The absence of any reference to aircraft color
in the Fairland report indicates that no study was made nor consideration given to the conspicuity aspects of the aircraft colors in this accident.)
AIR CARRIER CREW
Urbana: The Copilot was flying the DC-9 from the right seat and the Captain was handling the communications in the left seat.
Fairland: The Captain was flying the aircraft from the left seat and the Copilot was handling communications.
(Comment: In the Fairland accident, the individual with the greatest freedom to observe was in the seat closer to the other air craft.)
Urbana: The Baron was approximately 300 to the right of the DC-9 path and closer to the Copilot flying the aircraft from the right seat. The DC-9 was approximately 950 to 1000 to the left of the Baron's path requiring a turn of the head beyond 900 to see the DC-9.
Fairland: The Cherokee was approximately 190 to the right of the DC-9 path and closer to the Copilot who was not flying the airplane. The DC-9 was approximately 550 to the left of the Cherokee's path.
(Comment: In the case of the Fairland accident the angles of convergence were far more favorable for observation of the other aircraft than in the Urbana accident where, however, the DC-9 pilots had received a traffic advisory on the Beech Baron.)
Urbana: The DC-9 could have been in the process of making a 100 left turn while the Baron was on a constant heading.
Fairland: Both aircraft were on constant headings for at least 22 seconds prior to impact.
(Comment: A turning situation was not involved in this accident but apparently was in the Urbana accident. The Board notes in the Fairland report that a turning situation will diminish a pilot's ability to locate other aircraft.)
Urbana: There was no evidence of attempted evasive action by either crew.
Fairland: There was no evidence of attempted evasive action by either crew.
Urbana: A period of 14 seconds was available for the observation of the other aircraft and the accomplishment of evasive action.
Fairland: A period of 14 seconds was available for the observation of the other aircraft and the accomplishment of evasive action.
(Comment: In the Urbana report, the Board concludes that approximately 5 seconds would have been sufficient for the DC-9 crew to detect the other aircraft and initiate a change in direction of the DC-9,and that the DC-9 response time would have been approximately 3 seconds.)
While the above comparison of the pertinent facts of the two midair collisions demonstrates that the pilots involved in the Fairland accident had more opportunity to detect and avoid the other aircraft than the air carrier pilots in the Urbana accident, the Board nevertheless concludes that no pilot was responsible for the Fairland accident, after having concluded the air carrier pilots were responsible for the Urbana accident. The reasoning presented in the Fairland report does not provide persuasive support for the absolution of the pilots from responsibility for this accident.
The Board report baldly states that the DC-9 Captain's ability to observe the other air craft was "virtually nil." This conclusion is apparently based on the cockpit visibility study which used the Douglas design eye position for the Captain and found the Cherokee obscured by the aircraft structural member between the Captain's front windshield and the center windshield. No evidence of the Captain's physical characteristics is provided to show that his eye position would correspond to the design eye position. The Board conclusion also assumes the continued maintenance of a dummy-like posture by the Captain throughout the 14-second period and is thus contrary to the normal reaction expected of a pilot upon emerging from clouds while descending. In such circumstances, we submit it is more realistic to conclude that the Captain should have immediately initiated a comprehensive scan of his area of operation, detected the
Cherokee within 5 seconds, and had the DC-9 responding to evasive action within another 3 seconds.
The Board's excuse for the first officer's failure to observe the Cherokee, i.e. the air carrier requirement that he call the altitude as the aircraft passed through 3500 feet, is based on the conclusion that he was thereby "required to monitor the altimeter for a few hundred feet prior to reaching the altitude in order to note passage" and that the airspace involved for high rates of descent in approach areas should, accordingly, be protected by "positive air traffic control procedures." This reasoning, and the conclusion it supports, does not give proper consideration to the surveillance that a reasonably prudent pilot should perform upon emerging from clouds. In addition, it means that any crew member charged with the responsibility of calling out an altitude is excused from maintaining a surveillance outside the aircraft and that the aircraft must be protected completely by air traffic control. I believe the mere statement of this proposition accurately demonstrates the extent to which the Board is reaching in this report to excuse pilots from the responsibilities placed on them by the Federal Aviation Regulations.
The Board's conclusion here with respect to the "see-and-avoid" responsibilities under the FARs would apparently also apply to instructions given the pilots by the employing air carrier. An excerpt from the Flight Operations Manual of the carrier was included in Exhibit 2A, the
Operations Group Chairman's Factual Report of Investigation. That excerpt emphasizes the responsibility of pilots to maintain surveillance for other traffic. It reads as follows:
"Prior to and during the takeoff and climb, as well as throughout flight operations, pilots shall maintain a sharp watch for other traffic. This is especially true when operating in VFR conditions. The fact that a flight is operating under Instrument Flight Rules does not alleviate pilots from the responsibility of observing Visual Flight Rules when in VFR conditions. VFR responsibility is geared to the type of condition, not the type of flight plan. As little cockpit paper work as practicable shall be done during climb or descent."
The Board's analysis of the operation of the Cherokee is somewhat scanty. It notes that the pilot, operating "only 500 feet below the clouds," would be unable to see the DC-9 until it emerged from the clouds. The Board then proceeds to the finding that there was insufficient opportunity for the Cherokee pilot "to reasonably be expected to see and avoid" the DC-9. This reasoning assumes that an operation conducted in marginal compliance with the FARs and without regard to other circumstances must be judged in every case to be a safe and proper operation. The reasoning also omits any consideration of the judgment of the pilot in selecting and pursuing an operation which would have his aircraft only 500 feet below the clouds in an area where other aircraft would be descending to a major airport.
The other exercise of judgment by the Cherokee pilot which appears relevant and material to this accident but yet was not given any consideration
by the Board was his failure to utilize the expanded radar
service of Indianapolis Approach Control. The Board report contains a reference to other VFR aircraft utilizing this service at the altitude and within the time span involved in this accident but omits any discussion of the non-use of the service as a possible factor in the accident. If the Cherokee pilot had contacted Indianapolis Approach Control, reported his position and requested traffic information, the traffic advisories furnished would have undoubtedly included the DC-9. Also, this report would have enabled Approach Control to advise the DC-9 of his altitude, course and position.
No basis appears in the report for the Board conclusion that the Cherokee pilot was operating in full compliance with the FARs. These regulations impose the "see-and-avoid" responsibility. Assuming the shortest time span provided of 14 seconds to avoid the DC-9, a proper watch outside the Cherokee cockpit would have permitted at least partial, if not complete, evasive action and the physical evidence of the collision does not indicate any evasive action was ever initiated.
We note the use in the report of the term "positive radar control" and the introduction into aeronautical use, with-out definition, of the terms "active control" and "passive controll1 of the air traffic control system. The only positive radar control provided by the FAA is at the higher altitudes, flight level 180 and above in the Indianapolis area, where all aircraft are operated IFR and provided positive separation.
The references to ATC active and passive control are particularly misleading in view of the Board's conclusion that the ATC system, not the pilots, was responsible for this accident. While we are uninformed as to whether the Board would place our expanded radar service under the "active" or "passive" category, the fact remains that this part of the ATC system was not being utilized, accordingly, we do not believe this accident may form a basis for concluding the ATC system is the probable cause. In this respect, we should note that this accident represents the only midair collision involving an air carrier or commercial operator aircraft in which one of the aircraft had emerged from clouds shortly before the collision.
We have long been aware that there are circumstances which make difficult the compliance by pilots with their "see-and-avoid" responsibility. In the course of our presentation at your hearing of 6 November 1969, we listed 30 programs which FAA had accomplished to reduce the midair collision hazard. We also described 18 projects in which we were then engaged to reduce the midair collision hazard. Further, we summarized 17 programs we planned to undertake to reduce this hazard. We can now advise you that either progress is being made with these projects or they have been accomplished. The results are presently being employed either to lessen the potentiality of the collision hazard or to improve the ability of the pilots to cope with it.
We will not repeat here the statements of principle regarding the possible derogatory effect of the report on aircraft operations within the ATC system as it is now being operated and must continue to be operated for a substantial period into the future. It is our belief that it is incumbent upon the entire aviation community to devote its best efforts in ameliorating the midair collision hazard. With those best efforts, and the increased resources becoming available to us, we forecast a continued improvement in aviation safety.
In summary, we believe that the facts of this accident, as accepted by the Board in its report, do not provide a substantial basis for the Board determination that the probable cause was the ATC system but, rather, they more logically support a determination that the accident was due to deficiencies in the judgment and the performance of their surveillance responsibilities on the part of the pilots involved. The determination of probable cause and the relevant findings in the report should be revised accordingly.
In reaching its conclusion that the ATC system was responsible for this accident, the Board stated in its report that "it was recognition of the vast scope and far-reaching effects of this conclusion that prompted the Board to conduct a public hearing on the Midair Collision Problem." The announcement that the Board intended to conduct a public hearing on the Midair Collision Problem was made by the Presiding Officer of the Fairland accident hearing on 7 October 1969, prior to the commencement of that hearing. Thus, it follows that the conclusion expressed in the Fairland accident report on the culpability of the ATC system for the Fairland accident was reached by the Board prior to the public hearing on the Fairland accident. We suggest that in cases of accidents where the Board has made its determination regarding the probable cause of the accident prior to the hearing, the Board does the public and the participating parties a disservice in holding the hearing. Since a hearing is not mandatory in the accident investigative procedure, it appears desirable in future cases of this type that a hearing not be held and that the Board simply publish its report. This course of action would not only save the time and effort of the participating parties but should expedite publication of the report.
Thank you for the opportunity to record our views on this subject.