Narrative:

This was the first of three legs planned for the day after an 11 hour layover. I occupied the pilot observer position in the left jumpseat. The crew experienced and responded to a GPWS terrain warning during approach to oaix runway 03 via the ILS procedure track from ginri (IAF). Departure and cruise were routine and uneventful. I occupied the crew rest bunk until shortly after entering afghan airspace as I anticipated flying the last leg of the day. The initial descent from airway M696; approximately point vuset; was via radar vectors to a location north of oaix and then a vector toward the southeast. ATC directed 300 KIAS during descent and we were in and out of clouds with occasional moderate turbulence and conditions conducive to icing. Turbulence subsided below FL180. During descent and shortly after being assigned the southeasterly vector; the pilots asked to fly the ILS runway 03 procedure from ginri. This was approved with a clearance to proceed direct to ginri; descend and maintain at or above 12;500 ft until ginri; cleared for the approach.soon after; the pilots asked for relief from the 300 KIAS and that was approved by approach control. Around the 16 DME arc intercept point; ATC directed 'final speed.' the pilots had already begun slowing and extending flaps and they continued that effort; ultimately selecting flaps 20 and 155 KIAS. I believe the entire descent had been accomplished with VNAV selected on the MCP but I didn't verify that and did not note the specific VNAV modes indicated on the fmas of either pfd. After approach clearance was received; the pilot not flying noted aloud the clearance and asked to select a lower altitude on the MCP. The pilot flying responded that he would wait until we were leveled off. I was familiar with the environment; and comfortable with the discussion based on our location; clearance and the mode of vertical automation that I assumed we were in; so I did not lean forward and visually back up the crew's actions during this portion of the approach.I failed to see or hear when; who or exactly how the altitude selector was set to 7;000 but that's where it was when; at a point between ginri and dapub (probably around the bgm 135-140 radial); the GPWS activated with a terrain; terrain; pull up warning. The pilot observer in the right jumpseat called 'go around.' the pilot flying initiated the GPWS response maneuver by disengaging the autopilot; raising the pitch and pushing the thrust levers forward. The airplane climbed immediately. The pilot flying didn't disconnect the autothrottles initially and the power came back when he removed his hand. The pilot not flying noted the speed decreasing in the climb and called 'speed' while he and the pilot flying pushed the thrust levers back up and disconnected the autothrottles. The airplane continued to climb and seemed to recover to an altitude above 12;500 ft expeditiously. At the moment of the warning; the aircraft was in a steady descent. I believe the lowest altitude reached was in the 10;000 - 10;500 ft range. If I had been deliberate enough in leaning forward to monitor the altitude that I could provide accurate details; I probably wouldn't need to write this report. I believe the pilot not flying announced to bagram approach that we were climbing due to a GPWS warning and the controller replied that the minimum vectoring altitude at our location was 11;500 ft; would we like a vector to join the localizer? With around 25 track miles to run; it seemed reasonable to continue for landing. The pilot not flying reset the CDU as necessary. The pilots re-established full automation; complied with ATC vectors and descents; and the airplane was configured for and accomplished a normal approach; landing and block in.specific to the pilot observer in the left jumpseat: (background) since observing anything on the right pfd; the only one visible; is difficult from the left jumpseat due to distance; I've always emphasized backing up the operating pilots by listening to radios and copying clearances and weather so I can quickly interject on interphone to clear up any uncertainty with minimum delay during the critical phases of flight. This is clearly not adequate. 1. While all the complications of flying (speed and altitude control; navigation; configuration; radio calls; traffic; etc.) are being handled by the operating pilots; I should be backing them up very deliberately with basic instrument procedure compliance; i.e. Stepping through our progress to each waypoint and the associated minimum altitude; to include verifying actual aircraft altitude. 2. I should be more attentive to backing up the operating pilots on company standard callouts. That is; understandably; one of the first things to suffer when things get busy; so that's a great place for an observer to help. Strict compliance helps prevent an altitude deviation or MCP misconfiguration from occurring without immediate identification and correction. Specific to the pilot crew in general: 1. In this case; I felt that the crew maintained a professional environment and complied with sterile cockpit protocol; but we could probably benefit from being more deliberate about establishing sterile cockpit at 10;000 ft above the ground; which would be 15;000 ft MSL for bagram; or higher if you consider the terrain we are flying over. Many of us just focus on 10;000 ft MSL regardless of where in the world we're flying. You still aren't going to get ATIS until you are in the descent due to the communications limitations with high terrain; but wrapping up as much as possible above 15;000 MSL would help keep all eyes forward during the approach procedure. 2. I honestly have no idea how the 12;500 ft altitude constraint was deleted during this approach. The flying pilots have not been known; in my experience; to deviate from SOP for manipulating and coordinating MCP changes. I don't know if the altitude selector button was pushed after 7;000 ft was set or whether the pitch mode was changed to flch speed at some point. I do know that they intended to be in VNAV; and that being the case; it would be good to have the VNAV descent page in view on one of the cdus so that whenever we set approach minimums in the altitude selector; we can immediately verify that the appropriate altitude constraint is still active. Obviously; the selector button should not be pushed after setting a new altitude unless clearing an existing altitude constraint is desired. As a qualified pilot in an observer seat; I have no excuse for allowing this incident to occur. I really don't have a single; all-encompassing revelation to submit that will prevent a similar incident. We obviously did a poor job of avoiding and managing the distractions that could allow all four of us to miss a glaring altitude deviation during the same critical moment. The lesson for me is simply to stay engaged at a higher level when I'm on the flight deck.

Google
 

Original NASA ASRS Text

Title: Air carrier flight crew on approach selected the incorrect altitude for the existing autoflight mode and the aircraft descended too low in an area of mountainous terrain just south of the airport. The flying crew initiated an escape maneuver following receipt of a GPWS Terrain warning. Of particular note was the reported reminder that the sterile cockpit environment is more appropriately described as 10;000 AGL vice MSL.

Narrative: This was the first of three legs planned for the day after an 11 hour layover. I occupied the pilot observer position in the left jumpseat. The crew experienced and responded to a GPWS Terrain warning during approach to OAIX Runway 03 via the ILS procedure track from GINRI (IAF). Departure and cruise were routine and uneventful. I occupied the crew rest bunk until shortly after entering Afghan airspace as I anticipated flying the last leg of the day. The initial descent from Airway M696; approximately point VUSET; was via radar vectors to a location north of OAIX and then a vector toward the southeast. ATC directed 300 KIAS during descent and we were in and out of clouds with occasional moderate turbulence and conditions conducive to icing. Turbulence subsided below FL180. During descent and shortly after being assigned the Southeasterly vector; the pilots asked to fly the ILS Runway 03 procedure from GINRI. This was approved with a clearance to proceed direct to GINRI; descend and maintain at or above 12;500 FT until GINRI; cleared for the approach.Soon after; the pilots asked for relief from the 300 KIAS and that was approved by Approach Control. Around the 16 DME arc intercept point; ATC directed 'final speed.' The pilots had already begun slowing and extending flaps and they continued that effort; ultimately selecting flaps 20 and 155 KIAS. I believe the entire descent had been accomplished with VNAV selected on the MCP but I didn't verify that and did not note the specific VNAV modes indicated on the FMAs of either PFD. After approach clearance was received; the pilot not flying noted aloud the clearance and asked to select a lower altitude on the MCP. The pilot flying responded that he would wait until we were leveled off. I was familiar with the environment; and comfortable with the discussion based on our location; clearance and the mode of vertical automation that I assumed we were in; so I did not lean forward and visually back up the crew's actions during this portion of the approach.I failed to see or hear when; who or exactly how the altitude selector was set to 7;000 but that's where it was when; at a point between GINRI and DAPUB (probably around the BGM 135-140 radial); the GPWS activated with a TERRAIN; TERRAIN; PULL UP warning. The pilot observer in the right jumpseat called 'Go Around.' The pilot flying initiated the GPWS response maneuver by disengaging the Autopilot; raising the pitch and pushing the thrust levers forward. The airplane climbed immediately. The pilot flying didn't disconnect the autothrottles initially and the power came back when he removed his hand. The pilot not flying noted the speed decreasing in the climb and called 'Speed' while he and the pilot flying pushed the thrust levers back up and disconnected the autothrottles. The airplane continued to climb and seemed to recover to an altitude above 12;500 FT expeditiously. At the moment of the warning; the aircraft was in a steady descent. I believe the lowest altitude reached was in the 10;000 - 10;500 FT range. If I had been deliberate enough in leaning forward to monitor the altitude that I could provide accurate details; I probably wouldn't need to write this report. I believe the pilot not flying announced to Bagram Approach that we were climbing due to a GPWS warning and the Controller replied that the Minimum Vectoring Altitude at our location was 11;500 FT; would we like a vector to join the localizer? With around 25 track miles to run; it seemed reasonable to continue for landing. The pilot not flying reset the CDU as necessary. The pilots re-established full automation; complied with ATC vectors and descents; and the airplane was configured for and accomplished a normal approach; landing and block in.Specific to the pilot observer in the left jumpseat: (Background) Since observing anything on the right PFD; the only one visible; is difficult from the left jumpseat due to distance; I've always emphasized backing up the operating pilots by listening to radios and copying clearances and weather so I can quickly interject on interphone to clear up any uncertainty with minimum delay during the critical phases of flight. This is clearly not adequate. 1. While all the complications of flying (speed and altitude control; navigation; configuration; radio calls; traffic; etc.) are being handled by the operating pilots; I should be backing them up very deliberately with basic instrument procedure compliance; i.e. stepping through our progress to each waypoint and the associated minimum altitude; to include verifying actual aircraft altitude. 2. I should be more attentive to backing up the operating pilots on company standard callouts. That is; understandably; one of the first things to suffer when things get busy; so that's a great place for an observer to help. Strict compliance helps prevent an altitude deviation or MCP misconfiguration from occurring without immediate identification and correction. Specific to the pilot crew in general: 1. In this case; I felt that the crew maintained a professional environment and complied with sterile cockpit protocol; but we could probably benefit from being more deliberate about establishing sterile cockpit at 10;000 FT above the ground; which would be 15;000 FT MSL for Bagram; or higher if you consider the terrain we are flying over. Many of us just focus on 10;000 FT MSL regardless of where in the world we're flying. You still aren't going to get ATIS until you are in the descent due to the communications limitations with high terrain; but wrapping up as much as possible above 15;000 MSL would help keep all eyes forward during the approach procedure. 2. I honestly have no idea how the 12;500 FT altitude constraint was deleted during this approach. The flying pilots have not been known; in my experience; to deviate from SOP for manipulating and coordinating MCP changes. I don't know if the altitude selector button was pushed after 7;000 FT was set or whether the pitch mode was changed to FLCH SPD at some point. I do know that they intended to be in VNAV; and that being the case; it would be good to have the VNAV descent page in view on one of the CDUs so that whenever we set approach minimums in the altitude selector; we can immediately verify that the appropriate altitude constraint is still active. Obviously; the selector button should not be pushed after setting a new altitude unless clearing an existing altitude constraint is desired. As a qualified pilot in an observer seat; I have no excuse for allowing this incident to occur. I really don't have a single; all-encompassing revelation to submit that will prevent a similar incident. We obviously did a poor job of avoiding and managing the distractions that could allow all four of us to miss a glaring altitude deviation during the same critical moment. The lesson for me is simply to stay engaged at a higher level when I'm on the flight deck.

Data retrieved from NASA's ASRS site and automatically converted to unabbreviated mixed upper/lowercase text. This report is for informational purposes with no guarantee of accuracy. See NASA's ASRS site for official report.