By the time of this mishap, wind shear was a well recognized problem and the professional aviator class was well versed in how to recognize and how to deal with it. There was a thunderstorm overhead the airport and the airplane penetrated a microburst. The horizontal windshear was calculated to be as high as 86 knots. USAir 1016 was calculated to have gone through a 61 knot windshear in 15 seconds.

— James Albright






Still, reading through the NTSB Accident Report, one has to wonder if the crew could have anticipated the problem earlier, as the report alleges. They were preceded by other airplanes including one that reported a "smooth ride." The last thing they heard from tower was "windshear alert northeast boundary wind one nine zero at one three." They were headed for Runway 18R. I doubt a 13 knot wind would have prompted me to go around.

Once they decided to go around, I would agree, they made a few mistakes: they didn't set enough power, they allowed their pitch to fall below target, and they made a turn when flying straight ahead would have been okay. Simulations proved that they could have survived had the go around been flown properly.

So what to make of this? We have better radar and windshear detection today, to be sure. This crew should not have attempted the approach given the proximity of the weather. They didn't know what ground witnesses reported: "a wall of water" on top the airport. They didn't know what the weather service had just reported: "thunderstorm overhead the airport." But given the position of the cells to the airport, they should not have started the approach. The go around decision was smart, but the go around execution was not.

1 — Accident report

2 — Narrative

3 — Analysis

4 — Cause



Accident report

  • Date: 2 July 1994
  • Time: 18:43
  • Type: McDonnell Douglas DC-9-31
  • Operator: USAir
  • Registration: N954VJ
  • Fatalities: 0 of 5 crew, 37 of 52 passengers
  • Aircraft Fate: Destroyed
  • Phase: Approach
  • Airport: (Departure) Columbia Metropolitan Airport, SC (CAE/KCAE), United States of America
  • Airport: (Destination) Charlotte-Douglas Airport, NC (CLT/KCLT), United States of America



  • At 1834:57, the flight crew was issued a clearance to descend to 6,000 feet, and shortly thereafter contacted the Final Radar West (FRW) controller. The captain acknowledged the transmission.
  • At 1835:18, the FRW controller transmitted “USAir ten sixteen...maintain four thousand runway one eight right.” The captain acknowledged the radio transmission and then stated to the first officer “approach brief.” The first officer responded “visual back up ILS.” Following the first officer’s response, the controller issued a clearance to flight 1016 to “...turn ten degrees right descend and maintain two thousand three hundred vectors visual approach runway one eight right.” About this same time, the tower supervisor made the remark in the tower cab that it was “raining like he!l” at the south end of the airport, and the FRW controller observed on the airport surveillance radar (ASR-9) scope a VIP Level 34 cell “pop-up” near the airport.
  • At 1836:55, the FRW controller radioed flight 1016 and said “I’ll tell you what USAir ten sixteen they got some rain just south of the field might be a little bit coming off north just expect the ILS now amend your altitude maintain three thousand.” The captain acknowledged the transmission.
  • At 1837:33, the Charlotte Tower Local East Controller (LCE) transmitted to the flight crew of a DeHavilland DHC-8 that was landing on runway 23 “Piedmont thirty two eleven heavy heavy rain on the airport now wind one five zero at one four.” At 1837:4O, the FRW controller instructed flight 1016 to “turn right heading zero niner zero.” At 1838:24, the controller said “USAir ten sixteen turn right heading one seven zero four from SOPHE [the outer marker for runway 18R ILS]...cross SOPHE at or above three thousand cleared ILS one eight right approach.”
  • The captain testified at the Safety Board’s public hearing that as they were maneuvering the airplane from the base leg of the visual approach to final, they had visual contact with the airport.
  • At 1839:12, the flight crew of USAir flight 806 said to the LCW controller “And eight oh six looks like uh we’ve gotten a storm right on top of the field here.” The controller responded “affirmative.” The flight crew of USAir flight 806 elected to delay their departure.
  • At 1839:38, the captain of flight 1016 made initial contact with the LCW controller. The controller said “USAir ten sixteen... runway one eight right cleared to land following an F-K one hundred short final, previous arrival [USA3 677, a Fokker FK-28, that landed about 4 minutes earlier] reported a smooth ride all the way down the final.” The captain responded “USAir ten sixteen I appreciate a PIREP [pilot report] from that guy in front of us” [The airplane referenced by the captain of flight 1016 was USAir flight 984, a Fokker FK-100, that had circled from runway 23 to land on runway l8RJ. After receiving the flight conditions from USAir 984, the LCW controller relayed the report of a “smooth ride” to flight 1016.
  • About 1840:06, the first officer said “yep, laying right there this side of the airport, isn’t it...the edge of the rain is I’d say.” The captain responded “yeah.” In his testimony, the captain stated that he had been monitoring the weather conditions on the airborne radar and that while on final approach he had his navigational radio tuned to the Charlotte VOR for distance measuring information, although they had visually identified the runway during the initial portion of the final approach. The first officer testified that the “edge of the rain” that he observed was a “thin veil” through which he could see the runway and it was located “between us and the runway.”
  • At 1841:05, the CVR recorded the captain saying “stay heads up” and the LCW controller’s radio transmission of “windshear alert northeast boundary wind one nine zero at one three.” Meanwhile, at 1841:03, the LCE controller had transmitted “Attention all aircraft windshear alert the surface wind one zero zero at two zero northeast boundary wind one niner zero at one six.” However, this radio transmission occurred on a different radio frequency and was not heard by the crew of flight 1016.
  • At 1841:54 the CVR recorded the captain’s comment “here comes the wipers,” followed 3 seconds later by the sound of rain.
  • At 1841:58, the first officer commented “there’s, ooh, ten knots right there.” This was followed by the captain saying “OK, you're plus twenty [knots...take it around, go to the right.”
  • The following exchange of conversation and sounds were recorded by the CVR:
  • 1842:16 - Radio transmission by Captain - USAir ten sixteen’s on the go

    1842:17.7 - Captain - Max power

    1842:1x.5 - First Officer - Yeah max power

    1842:18.5 - Tower Controller - USAir ten sixteen understand you’re on the go sir, fly runway heading, climb and maintain three thousand.

    1842:19.4 - First Officer - flaps to fifteen

    1842:22.0 - Captain - Down, push it down

    1842:25.5 - Radio transmission - up to three we're taking a right turn

    1842:27.9 - Tower Controller - USAir ten sixteen, understand you’re turning right

    1842:28.4 - GPWS aural alert - whoop whoop terrain

    1842:28.5 - Unidentified voice on CVR - power

    1842:32.7 - Sound similar to stick shaker begins

    1842:32.7 - Sound similar to stick shaker ends

    1842:35.6 - Sound of ground impact

Source: NTSB AAR 95/03, ¶1.7



The flight crewmembers stated after the accident that they did not receive any aural or visual warnings in the cockpit from the windshear alert system prior to or during any portion of the flight. The CVR confirmed that the aural windshear alert did not activate.

Source: NTSB AAR 95/03, ¶1.6.2

The aircraft was equipped with a windshear detection system that used inputs from the airplane's AOA vanes, wing flap position sensors, engine N1 tachometers, and altitude sensors, with data from internal accelerometers.

[A study conducted by Honeywell] determined that a longitudinal shear that exceeded the computed threshold was encountered when the airplane was on the missed approach; thus, the flightcrew should have received both the red warning lights and the aural windshear warning. However, the warning would not have occurred until the airplane was at an altitude of between 100 and 150 feet above the ground, or approximately 3 to 4 seconds before ground impact.

The system was not evaluated with the flaps in transition; thus, the evaluation process neither revealed the system’s delayed activation feature when the flaps were moving, nor was it a requirement of the certification tests.

Source: NTSB AAR 95/03, ¶1.6.2

  • The Charlotte/Douglas International Airport was equipped with a Phase II LLWAS, which was operational at the time of the accident.
  • The LLWAS system has several limitations: winds above the sensors are not detected; winds beyond the peripheral sensors are not detected; updrafts and downdrafts are not detected: and if a shear boundary happens to pass a particular peripheral sensor and the centerfield sensor simultaneously, an alarm will not occur. However, since the downward flow in macrobursts and microbursts turns horizontally as it approaches the ground, an outward flowing shear boundary is established which eventually affects one of the sensors and places the system on alert.

Source: NTSB AAR 95/03, ¶1.7.3

USAir flight 1016 encountered a microburst windshear while on a missed approach from runway 18R. The microburst was the result of convective activity that was centered near the east side of runway 18R and that had cloud tops measured to an altitude of 30,000 feet. The microburst was determined to be approximately 3.5 kilometers in diameter and was capable of producing a rainfall rate of about 10 inches per hour. The total wind change near the ground was determined to be about 75 knots (at approximately 300 feet the winds were 86 knots), with the strongest downward vertical winds below 300 feet agl calculated to be 10 to 20 fps. The outflow winds most likely exhibited asymmetry with stronger winds on the west side of the microburst.

Source: NTSB AAR 95/03, ¶2.2.1

  • Simulations revealed that given the NASA wind flow field the airplane could have escaped the windshear encounter if several crew actions had been performed: first, the power was advanced by the first officer to an EPR setting of approximately 1.82, however, the captain did not trim to the target EPR of 1.93; second, the FDR indicated that a positive rate of climb had been established; however, the landing gear was not retracted; and lastly, the pitch attitude of the airplane was not maintained at or near the target of 15” nose up.
  • The simulations indicate that lowering the nose to 5° below the horizon was the most significant factor that prevented the escape from the windshear encounter. Based on these simulations, the Safety Board concludes that flight 1016 could have successfully flown through the windshear encountered if the flightcrew had executed an optimum missed approach procedure, and if “firewall” thrust had been applied as the airspeed decreased below 120 knots. The combination of the crew’s failure to use maximum go-around thrust, and the reduction of pitch attitude at a critical phase of flight, resulted in the airplane descending to the ground.
  • About 8 seconds after the first officer pitched the airplane up to 15° (the maximum recommended pitch attitude specified in the pilot operating handbook for the normal go-around procedure) and rolled into a 17° banked turn to the right, the CVR recorded the captain as saying “down, push it down.” Although the captain and first officer testified that they did not recall making the comment or hearing the comment, the Safety Board examined the CVR and FDR data and correlated the timing of the statement to the control yoke position and pitch attitude. Comparing the captain’s statement of “down, push it down” which occurred at 1842:22.O, the FDR recorded movement of the control yoke that changed proportionally as the attitude of the airplane transitioned from 15° nose up to 5° down.
  • Examination of the circumstances during the last minute of flight strongly suggested that the captain, upon losing his visual cues instantaneously when the airplane encountered the heavy rain, could have experienced a form of spatial disorientation. The disorientation might have led him to believe that the aircraft was climbing at an excessively high rate and that the pitch attitude should be lowered to prevent an aerodynamic stall.

Source: NTSB AAR 95/03, ¶2.3



The National Transportation Safety Board determines that the probable causes of the accident were: 1) the flightcrew’s decision to continue an approach into severe convective activity that was conducive to a microburst; 2) the flightcrew’s failure to recognize a windshear situation in a timely manner; 3) the flightcrew’s failure to establish and maintain the proper airplane attitude and thrust setting necessary to escape the windshear; and 4) the lack of real-time adverse weather and windshear hazard information dissemination from air traffic control, all of which led to an encounter with and failure to escape from a mircroburst-induced windshear that was produced by a rapidly developing thunderstorm located at the approach end of runway 18R.

Source: NTSB AAR 95/03, ¶3.2


(Source material)

NTSB Aircraft Accident Report, AAR-95/03, Flight Into Terrain During Missed Approach, USAir Flight 1016, DC-9-31, N954VJ, Charlotte/Douglas International Airport, Charlotte, North Carolina, July 2, 1994

Shein, Karsten, "WX Brief: Better knowledge of down bursts can save you from dangerous flying experiences," Professional Pilot, July 2013