I've had several people comment on "what the heck is an RFD and why do we care!". Well, I'm going to do my best at explaining what they are and why they're important to storm and tornado research. First lets look at each term. Rear - the rear side with respect to storm motion. Flank - side or lateral. Downdraft - downward flowing air. So an RFD is an area of downward flowing air found near the rear and side of a supercell thunderstorm. More specifically, it is found from the rear to right side of the storm's main updraft with respect to motion. The picture shows an RFD. The view is looking north and the lower gray cloud straight ahead on the road is a wall cloud. Wall clouds form on the interface between the updraft and RFD. To the left is clearing behind the storm and the RFD scours away clouds wrapping counter-clockwise around the rotating updraft forming a clear slot or notch (visual evidence of the RFD). RFD winds hit the ground and spread out along the surface. Storm chasers look for this signature, which manifests as a horse shoe shaped cloud, as a precursor to tornado formation in supercell thunderstorms. Once the RFD develops, it will do one of two things... 1). focus the broad storm scale rotation near the surface into a tight circulation (i. e. tornado) or 2). cause the storm scale circulation to become less focused near the surface and essentially kill off low level rotation (no tornado). This is why the RFD is so important to tornado formation (at least we think that is true at the moment). It takes a special RFD to form a tornado... and a REALLY special RFD to form a long-lived damaging tornado. The basics are this: an RFD that is neutral or positively buoyant (think of a balloon that either hovers in place or moves upwards on its own) will readily rotate inward and upward into the storms main rotating updraft. These downdrafts are special because most downdrafts fall out of the sky because the air in the downdraft is heavier than it's surroundings due to evaporative cooling. However, a warm RFD is forced downward by something else... most likely vertical pressure differences. So once this warm, buoyant air is forced down to the surface, it spirals in to a common point under the updraft and is recycled. It had angular momentum as it rotated downward (or developed it due to other reasons). As the RFD air spirals inward to the center of the updraft... angular momentum is conserved. You can test the physics of this at home by sitting in your computer chair and start spinning with your legs out. While spinning, bring your legs inward. You will quickly begin to rotate faster. This is due to the conservation of angular momentum. It is theorized that the same thing happens with the warm RFD air. However, more commonly RFD air is colder than its surroundings and the opposite effect will occur. Evaporatively cooled air (from rain and dry air) can contribute to this. Cold RFDs will hit the ground and be too heavy for the updraft to recycle much of the air. The RFD will surge away from the updraft and the conservation of angular momentum will cause the rotation to slow down (do the computer chair experiment starting with your legs in and then push them out... you will start fast and then rotate slower). So RFDs are very important to the development of tornadoes and the demise of rotation. In fact, it is likely that the RFD that causes tornadoes to develop in time will cause the tornado to rope out and die. Warm RFD's that initially fuel the tornado turn colder in time and snuff off the circulation. This is where lots of current tornado research is being done today. Why are some RFDs cold and others warm? Why do they transition over time. Why can one storm produce multiple RFDs over time that are sometimes warm and other times cold. There are lots of questions to be answered.