Supercells -- rotating thunderstorms which sometimes produce tornadoes -- are not only a U.S. phenomenon. They can happen anywhere the ingredients are right: moisture, instability, lift, and vertical shear (change in wind with height). The supercell in this picture formed after dark on the Serranias del Burro (a.k.a. Sierra del Huacha), a range of Mexican mountains west of Del Rio, Texas, and part of the northern Sierra Madre Oriental. [See the topographic image for geography of the area.]
The four needed ingredients for supercells came together rapidly on the night of March 22-23, 2000, in northeastern Mexico. As an upper level trough over Arizona moved eastward toward southwest Texas, winds aloft strengthened, increasing vertical shear. The trough also caused moisture-laden air from the Gulf of Mexico to move northwestward, up the valleys of the Rio Grande and the Mexican Rio Sabinas. Forced to rise thousands of feet by the sharp ascent up the east slopes of the Serranias del Burro, some of this moist and unstable air broke through a capping inversion and produced thunderstorms. One of these storms became a very intense supercell -- on radar, rivalling all but a few U.S. thunderstorms per year for strength and definition. A long, slow-loading javascript reflectivity loop (~850K) shows about 5 hours out of this storm's lifetime, from about half an hour after formation through about 2 hours before it dissipated.
There may have been a strong or violent tornado in this supercell, at the time of the image. The storm had an extremely well-defined hook echo -- a common signature of storms capable of producing tornadoes -- while moving northeast off the higher terrain toward the Rio Grande. The strongest rotation (mesocyclone) in the supercell was during this stage, while still in Mexico. A storm-relative velocity image from 10 minutes later showed gate-to-gate (side-by-side) speeds reaching the highest they can on the display scale -- 50 or more knots (at least 58 mph) of wind toward the radar, right beside 50 knots or more of wind away, in the low levels of the storm. This meant at least 100 knots (116 mph) of "gate-to-gate shear," along with a "tornado vortex signature," a radar term for a circulation strong enough to be a tornado if it were on the ground. A javascript loop of storm-relative velocity (~350K) shows this "shear couplet" became quite intense, and remained so while curving northwestward, before a newer and weaker one formed to its east. This process of forming repeat mesocyclones is common in supercells which spawn more than one major tornado. Supercells which do this are called cyclic -- in other words, they cycle through circulation after circulation.
The round "ball" of reflectivity in the hook is a rare feature, sometimes seen when a radar picks up tornado debris being lofted into a thunderstorm. [Tornado debris was the cause of a similar signature on May 3, 1999 in Oklahoma.] Unlike in the developed areas of central Oklahoma, there is apparently little which a tornado could hit in that part of Mexico to generate large debris -- unless the radar was detecting sagebrush leaves, broken mesquite branches and shredded chunks of prickly pear cactus. Maps of the area show no towns or villages, and only one dirt road winding west out of Ciudad Acuna into the Serranias del Burro mountains.
Because of the remote desert terrain west of Ciudad Acuna, the lack of population, and the lack of tornado verification in Mexico, we may never know if there was truly a tornado there. In the U.S., however, low level mesocyclones this strong are not usually seen even in many storms which do produce tornadoes. Based on all the radar signatures, the best we can say for now is that this supercell probably produced a significant tornado in Mexico. Later, as it crossed the border over Amistad Reservoir (a large man-made lake on the Rio Grande), it had two hook echoes, with a mesocyclone in each one, prompting this tornado warning for the Amistad Reservoir and Del Rio area:
