There is no rule that states how many instrument approaches a pilot can fly before going to another airport and there shouldn't be. I am not in favor of removing any of the pilot's authority to conduct operations the way he or she sees fit. However, every pilot must use common sense and determine when it is realistically time to find a different airport to head for if weather conditions are not conducive to landing at the intended destination.
On an early morning in December 2002, a Beech 58 Baron with a pilot and one passenger departed Rosecrans Memorial Airport in St. Joseph, Mo., (KSTJ) for Naples, Fla., (KAPF), located on the southwestern coast of the Sunshine State. The flight included a refueling stop in Centerville, Ala., where the 69-year-old pilot called the Aniston, Ala., AFSS to file an IFR flight plan to Naples and receive an abbreviated weather briefing for the route. The pilot told the FSS briefer that he had received a complete weather briefing a couple of hours before and asked if anything had changed, adding that he hoped things had improved.
The briefer advised the pilot that two AIRMETs has been issued for Florida, one calling for icing conditions above 15,000 feet and the other for IFR conditions until 4 p.m. local time. The briefer told the pilot that it looked like it would be "foggy down that way."
The pilot also asked if any rain was expected in that area. He said he had been told earlier to expect some in the afternoon. The briefer looked at the weather radar and said that there were scattered showers around the St. Petersburg area and south but that no Convective SIGMETS had been issued yet. He added that in the Tallahassee area and south it looked like the ceilings were 1,200 to 1,500 feet with patchy IFR conditions. At the time, the current Naples weather was calm winds, 10 miles visibility, broken clouds at 1,600 and 2,300 feet, and overcast clouds at 3,200 feet.
The pilot did not ask for any forecasts nor did the briefer provide anything more than the current conditions. No forecasts were provided in the accident report, so we don't know what the pilot had been told prior to his leaving St. Joseph.
The pilot's IFR flight plan called for a routing that would take him farther east but over land versus a more direct routing over the Gulf of Mexico. Before departing Missouri, the pilot had told family members that he did not want to fly directly over the Gulf because the aircraft was not equipped for extended operations over water.
Interestingly, the pilot did not file an alternate, even telling the briefer that one was not necessary. The accident report doesn't mention what weather information the pilot had received prior to leaving St. Joseph, so it's possible that the initial forecast indicated better conditions than what he eventually encountered. Even so, the Aniston briefer's comment about 1200- to 1500-foot ceilings and fog should have alerted him to look a little closer.
The flight proceeded to Naples without incident. The pilot checked in with the Fort Myers Approach controller at 2:40 p.m. while cruising at 7000 feet. The controller asked if he had ATIS Information Sierra for Naples and the pilot said he did.
Four minutes later the controller announced that Information Tango was now current. Naples was reporting visibility two miles in light rain and mist and a ceiling of 500 feet overcast. The wind was from 320 degrees at seven kts and the altimeter setting was 30.04. The approach in use was the VOR 23 with a circle to land on Runway 32. (There are no ILS approaches at Naples.) There is a 500-foot minimum descent altitude on that approach whether you fly straight-in or circle.
Shortly after that the controller cleared the Baron to fly direct to Naples and instructed the pilot to descend and maintain 5000 feet. At 2:47 p.m. the controller issued the pilot a vector of 190 degrees for the VOR 23 approach and told him to contact the next approach sector.
After switching to the next approach controller, the pilot was instructed to descend to 3000 feet and fly a 130-degree heading. Further vectors followed including an advisory that the aircraft would be vectored across the final approach course for sequencing. At 3:01 p.m. the controller issued the pilot a heading of 270 degrees and told him to join the final approach course. He also told him to reduce to his final approach speed and that he was 12 miles from the airport.
If the deteriorating weather wasn't already setting off alarm bells in the pilot's head, what he heard next surely must have: The controller told the Baron pilot that two other aircraft, a Comanche and a Westwind, went missed their first time around. On the second approach the Comanche was able to get in but the Westwind did not.
Recorded radar data indicated that the pilot flew through the final approach course -- the 055-degree radial from the Cyprus VOR -- then paralleled the course on the northwest side. It remained northwest of the course until passing the Cyprus VOR, which is located on the Naples Airport. The data suggests that the pilot descended to about 500 feet during the approach but was never established on the approach course. After passing the VOR he declared a missed approach and requested to divert to Fort Myers' Page Field (KFMY).
Upon switching back to Approach Control, the pilot was given the weather at KFMY. The winds were from 020 degrees at 13 kts, the visibility was three miles in light rain and mist with an overcast ceiling at 300 feet. The controller gave the pilot vectors and told him to expect the ILS Runway 5 approach.
At 3:22 p.m. the Approach controller advised the pilot that he was five miles from CALOO, the outer marker, and was instructed to fly a heading of 030 degrees and maintain 2000 feet until established on the localizer. Once cleared for the approach, the pilot was told to contact Fort Myers Tower.
Upon checking in with the Tower controller, the pilot was cleared to land. The controller then asked if the Baron was established on the final approach course. The pilot responded, "Yeah. I'm not quite established yet. I'm comin' up on CALOO, but I'm not actually on the glideslope yet. Looks like I should intercept it here in just about a minute." The controller then said that most of the other aircraft were breaking out of the cloud bases around 300 feet and they were reporting 3-sm visibility.
Radar data for that approach revealed that the aircraft intercepted the localizer about a mile southwest of CALOO. But the Baron flew through the course on the 030-degree track and continued for another mile before turning back toward the course. Then it made a teardrop turn to the right and rolled out on a westerly heading. All this occurred within 2-1/2 miles of CALOO, during which time the aircraft descended from 1100 feet while paralleling the course all the way to 200 feet as the aircraft turned to the west. At this point the controller called the pilot and asked if he had gone missed. The pilot responded, "Yeah. I missed it. I'm going back up to 2000 and can you bring me back around?"
The tower controller called the approach controller to explain that the pilot had turned to the west on his own and began to coordinate a second approach for the flight. The instruction was for the pilot to fly a 270-degree heading and climb to 2000 feet.
By this time, the controllers must have begun to suspect that all was not well aboard the Baron. On neither approaches so far had the Baron maintained the final approach course. The teardrop turn away from KFMY was also clearly improvised since the published missed approach procedure is to climb to 1000 feet on the runway heading and then make a climbing left turn to 2500 feet while intercepting the Lee County VOR 354-degree radial to SERFS intersection.
When the Baron pilot was switched back to Approach Control, he stated that he would like to try the approach again. At 3:30 p.m. the pilot was issued a vector of 080 degrees and told to intercept the localizer on that heading at 2000 feet. He was three miles from CALOO at that time. A short time later the controller called back and informed the pilot that he was going left of the localizer and asked if he was receiving the localizer signal. The pilot replied he had flown through it.
However, radar data indicates that the aircraft never intercepted the localizer but that it paralleled the left side of the localizer course. The controller instructed the pilot to turn right to a heading of 070 degrees and intercept the course, adding that he was in the vicinity of the outer marker at that time.
"I'm on it now," the pilot responded, and the aircraft turned toward the final approach course. However, before the aircraft intercepted the course it made another tear drop turn, this time to the left, and began reversing course. The controller asked the pilot to let him know when he was established on the final and received no response. Then the controller asked the pilot to confirm that he was on the localizer course. The pilot responded, "Sir, I'm having very big difficulties out here. I've gone through it again and I'm climbing back to 2000 feet."
This time, the aircraft descended no lower than 1,400 feet. That was when the pilot was making the teardrop turn. And once again, he didn't get very far past CALOO before initiating the unauthorized missed approach maneuver.
Any doubts by the controller that the Baron pilot was in trouble were now erased. The controller told the pilot that his approach clearance was cancelled and then said, "Just fly straight ... ahead, sir. Fly straight ahead." The controller had initially told the pilot to fly a 310-degree heading, and the pilot asked if he wanted him to fly that. The controller responded, "Fly straight ahead. Just whatever heading you have now. Maintain 2000 feet." He then instructed the Baron pilot to expect the ILS Runway 5 Approach.
At 3:33 p.m. the controller asked the pilot what his fuel status was. The pilot responded, "It's down in the yellow." The controller then asked how much that was in time. The response was, "Practically nil." The controller then instructed the pilot to turn to a 300-degree heading and asked what his flight conditions were like. The pilot acknowledged and told him, "I'm above the clouds at the moment."
The controller told the pilot that he could set him up for the NDB approach but the pilot said he wanted to fly the ILS. He told the controller he was trying to fly the approach "with my automatic" and that he was going to have to fly it manually. He added that he had some kind of instrument problem, but didn't elaborate.
At 3:35 p.m. the controller vectored the pilot back to the localizer and told him to intercept the course. He was four miles outside of CALOO at the time. The controller contacted the Tower and asked how the weather was doing. The tower controller told him that the cloud bases were about 300 feet and that most inbound airplanes were breaking out at that altitude. He also said the visibility was three miles.
At 3:36 p.m. the controller advised the pilot that he was flying through the localizer again and suggested a 060-degree heading to intercept. Then he told the pilot he was over the outer marker and issued a descent to 1,500 feet saying he was going to initiate the pilot's descent early.
At 3:37 p.m. he instructed the pilot to fly a heading of 050 degrees and informed the pilot that he was going to convert his ILS approach to a surveillance approach to Runway 5. He instructed the pilot that the published missed approach point was one mile from the runway threshold and the minimum descent altitude was 500 feet.
A minute later, the controller told the Baron pilot to fly a 070-degree heading and to descend and maintain 900 feet. The aircraft was then 3-1/2 miles from the runway and "very slightly left of course."
The controller told the pilot to descend to 500 feet and cleared the aircraft to land if he could see the runway, but also issued missed approach instructions, which was to climb straight ahead to 2000 feet, if he couldn't.
At 3:39 p.m. the controller informed the pilot that he was less than a mile from the runway at an indicated altitude of 300 feet. He issued an instruction for the pilot to go around if he did not see the runway.
The recorded radar shows that the airplane was on course for the last three miles of the approach. The airplane tracked directly over the runway as it started to climb to 600 feet. At the end of the runway it began a left turn and descended again to 300 feet.
Just before 3:40 p.m. the controller told the pilot to maintain 1,500 feet if he could. When he did not get a response, he asked the pilot if he could hear him. The pilot said, "I hear ya."
The controller then instructed the pilot to climb to 1500 feet and fly a 180-degree heading, but the pilot flew a southeasterly heading at 1200 feet. The controller asked the pilot to fly a heading of 230 degrees, but he never did that. Radar showed the aircraft peaked at 1200 feet, then started a turn toward the east as it descended into a residential area, crashing into a garage attached to a home before impacting terrain. The pilot and his passenger were killed.
A witness at the accident site told investigators that he saw the airplane strike the garage roof and hit the ground as "pieces scattered." He indicated that the clouds were very low.
Another witness saw the airplane as it descended out of the clouds heading south at a low altitude. The landing gear was retracted, it sounded like the engines were at full power and the airplane was losing altitude and banking to the right.
The pilot held a private pilot certificate with single- and multiengine ratings. He also possessed an instrument rating, which he had received 4-1/2 years earlier. The pilot's logbook indicated that he had a total of 910.5 hours flying time with 211.8 hours of multi-engine time. All of the multi-engine time was in Barons. The log showed that he had accumulated 70.5 hours of actual instrument time and 65.9 hours of simulated instrument time.
The aircraft was well equipped with Garmin GNS 530 and 430 GPS units. It also had a three-axis autopilot with a yaw damper and an electronic HSI installed. Unfortunately, much of the airplane, including the instrument panel, was destroyed. While no signs of any mechanical or instrument failure were found, the damage was too extensive to rule them out.
In the end, the NTSB blamed the accident on the pilot's spatial disorientation during IMC, which resulted in the loss of control. The Board also added that the pilot's distraction by his diminishing fuel supply and low ceiling contributed to the accident.
While we'll never know for sure, there are a couple of things that may have had an effect on the flight.
First, the multiple approaches may have fatigued the pilot and contributed to his spatial disorientation. But it also seems that the pilot never considered looking for another airport at which to land. He did not file an alternate because he believed one was not necessary.
Once he discovered that he was having problems, whether they were autopilot related or due to a lack of currency and proficiency, the better course of action would have been to climb up into the better conditions above and then look around for an airport where the weather conditions were also better.
The question to ask yourself, though, is how many approaches will you make to the same runway before you decide that you are not going to land there? Flying an ILS approach to minimums with no visual sighting of the approach lighting system or the runway is a good indication of what you can expect on the next approach. If you had some kind of visual contact with the approach lighting system or the runway the first time it might warrant a second shot at it. But if the second approach proved to be no better than the first, it's time to move on.
More accident analyses are available in AVweb's Probable Cause Index. And for monthly articles about IFR flying including accident reports like this one, subscribe to AVweb's sister publication, IFR Refresher.