Around 66 million years ago, an object some six miles in diameter impacted Earth in what is now the Yucatan. Not only did the kinetic energy of the impact itself cause great fires in southern North America and northern South America, but the blowback from that event in the form of debris kicked up by the impact falling back to Earth liberated even more kinetic energy transformed into thermal energy started firestorms all over the world, fires that could sweep whole continents. This was the K-T event, the event that closed out the Cretaceous Period and the Mesozoic Era of Earthly life.
As Daniel James Brown points out in his
Under a Flaming Sky: The Great Hinckley Firestorm of 1894, a firestorm is not like other fires, even relatively large ones, and does not fit into zny of the three wildfire classifications of ground, surface, and crown fires. It likely has been all of these in the course of its evolution, but it now becomes capable of growing into something very different from them, and very unusual: a mass fire, to use the term used by Stephen J. Pyne in his landmark study,
Fire in America: A Cultural History of Wildland and Rural Fire (Weyerhaeuser Environmental Book (see Brown, op. cit., pp. 63-65).
True firestorms generated their own weather. The radiant heat they give off can cause anything organic within half a mile or so to burst into flame without being touched by blazing firebrands thrown ahead of the fire front. The heat of the fire will turn all the carbon, all organic material in the soils over which it rages down to a depth of 15 feet into white ash and sand, the latter blowing away in the wind. Anything living in those soils, say, in burrows, will only be fragile fuel for that horrendous fire. Afterwards, little or nothing can grow in what's left of that soil.
More: during the K-T event, the heat released by that event turned the atmosphere immediately above Earth's surface to a scorching 500○ F -- broiler oven temperatures. Any living organism not protected from that heat would have burst into flame and gas within a couple of seconds. Yet we know that some life on land did survive, up to 10% or so. So how did they survive?
They could have been underground. Problem: since the heat of the fires could penetrate deep into soil and rock, they would have to have found or made shelter for themselves 40-50 feet down or more. This explains -- maybe -- why mammals survived and big dinosaurs didn't, but it doesn't explain the survival of birds, testudines, amphibians, and non-burrowing arthropods. And since firestorms suck virtually all the oxygen out of the atmosphere for many miles around, those burrowing underground could easily have died of asphyxiation, in a one-two hit of roasting heat and extreme lack of oxygen.
So where else could animals have survived the worst of it?
Deep and extensive cave systems, perhaps. The air in caves is often significantly cooler than air outside caves. The deepest chambers of deep, extensive, large caves and cave systems may even be downright cold. They take in air through a myriad of cracks in cave walls, and cool it down quickly to the cave's ambient temperature. And there is water, coming from the same rivers and streams responsible for carving out the cave in the first place. They would make the perfect redoubts for amphibians, who need to keep their skins moist at all times. And the protective coolth and availability of water would also provide a perfect shelter for any land animals small enough to reach the cave and. go far enough back into it.
So paleontologists should explore what's left of ancient caves and cave systems for evidence of occupation of those caves at the time of the K-T event.