HISTORY OF FLIGHTOn April 17, 2015, about 1615 Pacific daylight time, the pilot landed a Piper PA-18-150, N162SC, off airport near La Quinta, California. The pilot/owner was operating the airplane under the provisions of 14 Code of Federal Regulations (CFR) Part 91. The commercial pilot sustained minor injuries; the airplane sustained substantial damage. The local banner tow flight departed Bermuda Dunes (UDD), California, at 1531. Visual meteorological conditions prevailed, and no flight plan had been filed. The pilot stated that the airplane was in left orbit over the advertising destination at 1,200 ft agl when it began to shake violently, and lose altitude. The pilot immediately initiated a right turn away from the banner pattern, and reduced power to idle which alleviated the violent shaking. The pilot then made an emergency radio call on the Bermuda Dunes common traffic advisory frequency, began looking for a safe place to drop the banner, and maneuvered to release it over an empty fairway at a golf course at 600 ft agl. The pilot advanced the throttle, and noted that the severe vibration returned at the higher engine rpm. Neither a climb nor level flight was achievable. After pulling the mixture and checking that the seat belt was tight, the next empty fairway was selected as the landing site. The pilot engaged full flaps, and proceeded to forward slip the airplane in order to make a touchdown in the empty fairway. About 100 to 200 ft agl, the pilot noted that the landing area had significant elevation changes, but it was too late to select another landing area. The airplane touched down on a plateau at 60 knots indicated airspeed (KIAS), but could not be stopped prior to a 10-foot drop-off. The airplane became airborne, and bounced down at the next plateau, which was the green. The airplane did not stop despite full application of the brakes. The pilot noted a tree line about 20 yards away with residential housing behind it. The airplane became airborne again, and the pilot attempted a 90° turn to avoid crossing the tree line and impacting a house. The airplane contacted the ground sideways; it rolled at a 45° angle (to the 2-o'clock position) onto its back, and came to rest inverted. The pilot released the seatbelt, and exited the airplane. TESTS AND RESEARCH The National Transportation Safety Board (NTSB) investigator-in-charge and an inspector from the Federal Aviation Administration examined the wreckage. Several inches of one propeller blade tip was missing, and not located. The fracture surface was flat, and this blade tip was sent to the NTSB Research and Engineering Materials Laboratory for examination. No anomalies with the airframe or engine were noted that would have precluded normal operation. One magneto was sent to a shop for examination, and it passed functional checks. ADDITIONAL INFORMATION Blade Tip Examination The NTSB Materials Laboratory specialist reported that the fracture surface on the separated blade was generally flat with a finely textured appearance and a unified crack front consistent with fatigue crack progression. The fatigue cracking had propagated through approximately 83% of the blade cross-section. Overstress separation covered the remaining 17% of the blade fracture surface. The fatigue cracking appeared to initiate from one general area on the pressure side of the blade, which contained an anomaly. The anomaly appeared to consist of deformed material. Parallel lines consistent with hand finishing were on the external surface of the blade, and overlapped into the deformed material. There were also thinner circular lines on the surfaces of the blade that were consistent with machining marks from an orbital grinder that overlapped the hand finishing lines. Examination with a scanning electron microscope revealed feathery features and striations consistent with fatigue cracking on the fracture surface emanating from the pressure side of the blade in the area of the anomaly. No contamination was observed at the fatigue initiating site. Propeller Logbook A review of the propeller logbook revealed that it was overhauled on June 13, 2011; it was installed on the accident airplane on June 19, 2011. An annual inspection dated March 1, 2014, recorded 281.9 hours since overhaul. On October 23, 2014, the propeller was removed at 393.6 hours since overhaul; it was installed back onto the airplane on October 30. An entry dated February 23, 2015, was for an annual inspection, which indicated that the propeller had been dynamically balanced, inspected, and was in an airworthy condition. The time recorded since overhaul was 380.3 hours, but correlation with tachometer entries determined that it should have been 480.3 hours since overhaul. The tachometer read 948.68 at the wreckage examination. No entry, including the overhaul entry, indicated any work to a deformed area on the propeller.