Narrative:

I was working local control 1 with runways 2R/2L active. A cirrus was inbound with approach control at 4;000 MSL from the northwest transitioning north of the field to enter a right downwind for runway 2R. A meridian was inbound with approach control from the northeast at 4;000 MSL approaching the right downwind for runway 2R. It appeared that the meridian was being sequenced in front of the cirrus. A VFR beech baron called me six miles southeast of the field inbound. Initially I told the baron to reduce his speed so that I could sequence him behind the meridian; but noticed that the meridian was being vectored a bit further east of the airport. Therefore I told the baron to keep his speed up and that he was number one for runway 2R and cleared to land. After clearing the baron; I noticed the meridian take a turn to the north; I assumed approach control was spinning him for some unknown reason. A helicopter called and requested a west departure from the helipad; and I issued him a take off clearance. That's when I looked back at the racd and noticed the meridian had not made a complete 360 degree right turn but instead was continuing on a northerly heading. The meridian called me at this moment stating he was entering a left base for runway 20R. The meridian was pointed directly at the cirrus inbound from the northwest about to turn a 4 mile right downwind for runway 2R. I immediately issued a traffic alert to the meridian regarding the cirrus at his 11 o'clock; one and a half miles. I believe the meridian was only 300 to 700 ft below the cirrus; thereby having lost standard separation. The two aircraft targets appeared to merge as the meridian passed under the cirrus. Realizing the pilot of the meridian was headed to the wrong runway; my ground controller called approach to obtain control of the meridian for me and the request was approved. I provided suggested headings to the meridian to route him back over the north end of the field to enter a left downwind for runway 2L. About 2 minutes later; the pilot of the meridian lined up on final for runway 20R. I told him to break off his final for 20R and enter a left downwind for runway 2L. The pilot replied he was lining up on final for runway 20L. I immediately instructed the pilot to climb and maintain 2;500 MSL and turn right heading 270 so as to guide him to the left downwind for runway 2L. Finally; the pilot of the meridian entered the left downwind for 2L and landed. After asking approach control to review the tapes; the pilot was cleared for; and read back; the visual approach for runway 2R. Sometime between then time approach switched the meridian to tower frequency and the time he checked on with me; an excessive amount of time to switch frequencies in my opinion; the pilot must have forgotten he was headed for runway 2R and turned left downwind to base for runway 20R. I can provide no recommendations because this was clearly a pilot deviation that resulted in less than standard separation. Perhaps if the approach controller had ensured that the meridian had leveled at 3;000 ft as assigned before switching the aircraft to tower frequency; then when he made the erroneous northerly turn he would have at least been 1;000 ft below the cirrus when the targets merged and separation may not have been lost.

Related reports:

Google
 

Original NASA ASRS Text

Title: Tower Controller described a loss of separation event when traffic inbound to the pattern entered an opposite direction downwind.

Narrative: I was working Local Control 1 with Runways 2R/2L active. A Cirrus was inbound with Approach Control at 4;000 MSL from the northwest transitioning north of the field to enter a right downwind for Runway 2R. A Meridian was inbound with Approach Control from the northeast at 4;000 MSL approaching the right downwind for Runway 2R. It appeared that the Meridian was being sequenced in front of the Cirrus. A VFR Beech Baron called me six miles southeast of the field inbound. Initially I told the Baron to reduce his speed so that I could sequence him behind the Meridian; but noticed that the Meridian was being vectored a bit further east of the airport. Therefore I told the Baron to keep his speed up and that he was number one for Runway 2R and cleared to land. After clearing the Baron; I noticed the Meridian take a turn to the north; I assumed Approach Control was spinning him for some unknown reason. A helicopter called and requested a west departure from the helipad; and I issued him a take off clearance. That's when I looked back at the RACD and noticed the Meridian had not made a complete 360 degree right turn but instead was continuing on a northerly heading. The Meridian called me at this moment stating he was entering a left base for Runway 20R. The Meridian was pointed directly at the Cirrus inbound from the northwest about to turn a 4 mile right downwind for Runway 2R. I immediately issued a Traffic Alert to the Meridian regarding the Cirrus at his 11 o'clock; one and a half miles. I believe the Meridian was only 300 to 700 FT below the Cirrus; thereby having lost standard separation. The two aircraft targets appeared to merge as the Meridian passed under the Cirrus. Realizing the pilot of the Meridian was headed to the wrong runway; my Ground Controller called Approach to obtain control of the Meridian for me and the request was approved. I provided suggested headings to the Meridian to route him back over the north end of the field to enter a left downwind for Runway 2L. About 2 minutes later; the pilot of the Meridian lined up on final for Runway 20R. I told him to break off his final for 20R and enter a left downwind for Runway 2L. The pilot replied he was lining up on final for Runway 20L. I immediately instructed the pilot to climb and maintain 2;500 MSL and turn right heading 270 so as to guide him to the left downwind for Runway 2L. Finally; the pilot of the Meridian entered the left downwind for 2L and landed. After asking Approach Control to review the tapes; the pilot was cleared for; and read back; the Visual Approach for Runway 2R. Sometime between then time Approach switched the Meridian to Tower frequency and the time he checked on with me; an excessive amount of time to switch frequencies in my opinion; the pilot must have forgotten he was headed for Runway 2R and turned left downwind to base for Runway 20R. I can provide no recommendations because this was clearly a pilot deviation that resulted in less than standard separation. Perhaps if the Approach Controller had ensured that the Meridian had leveled at 3;000 FT as assigned before switching the aircraft to Tower frequency; then when he made the erroneous northerly turn he would have at least been 1;000 FT below the Cirrus when the targets merged and separation may not have been lost.

Data retrieved from NASA's ASRS site as of July 2013 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.