When it comes down to it, there is no such thing as a natural "disaster". There are massive events that can fundamentally change human society locally and/or globally or even threaten life on Earth. Yet, really, they are only disasters because they change the status quo. Just think about how the Chicxulub asteroid impact at the end of the Cretaceous opened the door for the rise of mammals.
Of course, humans have a vested interest in not going extinct. We might have the tendency to do things that might hasten our extinction: see also, climate change, pandemics, nuclear weapons. Yet, when it comes to singular "disastrous" events, humans are pretty keen on thinking we can prevent them from happening.
Diverting an Asteroid
Case in point: NASA's new DART (Double Asteroid Redirection Test) mission. This is the first real test of a "planetary defense" against an asteroid impact. The DART mission, launched November 24, will rendezvous with an asteroid named Didymos that is actually a double asteroid. The main asteroid is about 2,500 feet (780 meters) across while its companion is ~525 feet (160 meters) across.
The goal of mission is to see how much the mere impact of a spacecraft into an asteroid will change its trajectory. Now, Didymos is not a threat to Earth, but is a good candidate to see if this method could be feasible when we really do need to nudge an asteroid from a potential collision course with our planet. Think of it like a cosmic game of pool, sending the asteroid on a different path.
And herein lies the rub for our desire to prevent these types of massive and destructive events: the scale of processes. Didymos is a relatively small asteroid and if DART works, we can hope that it would on a larger scale with larger asteroids. However, maybe that would we make it worse -- maybe we break the big asteroid into smaller bits that hit more places? We also need to think about what we can do with existing technology and what the future might make feasible ... and at what cost.
Stopping Other "Disasters"
So what about other disasters? Could we conceive of ways to prevent them from happening? Let's ponder...
Earthquakes: These events, generated by the energy released with two large masses of rock grind against each other, occur on such massive scales that the idea of any remotely feasible technology stopping them from happening is ludicrous. We would, pretty much, have to stop plate tectonics on Earth. Maybe there could be a way to bleed off stress on a fault, but even that is beyond the wildest dreams of any seismologist. That being said, in places where we cause earthquakes -- think the pumping of waste fluids from fracking into the crust -- we can reduce the frequency of earthquakes.
Tsunamis: These massive ocean waves are formed mainly by earthquakes, volcanic eruptions or landslides. So, short of stopping those triggers, stopping the ocean from making these waves that can devastate coastlines, we can really hope to only reduce their impacts. However, even building massive seawalls to stop tsunamis was futile in many places in Japan during the 2011 Tohoku earthquake and tsunami.
Volcanic Eruptions: People love the notion of "stopping" a volcano from eruption. If volcanoes erupt due to pressure, then maybe we can reduce the pressure by drilling or even bombing the volcano. Sorry, no dice. Most bodies of magma reside deep in the Earth's crust (we're talking miles here), so making any impact on the pressure a magma body is feeling would be next to impossible as all that rock that surrounds it far outweighs whatever we might try.
The furthest we've drilled into the crust gets us to the depth of magma storage, but that was a borehole no bigger than a pie plate. And nuking a volcano? Sure, that's a great idea if you want a volcanic eruption mixed with radioactive fallout. We have diverted small lava flows, but that just moves the problem into someone else's backyard.
Landslides: Maybe the most feasible hazard to prevent, most landslides are fairly small. Stabilizing slopes with vegetation, netting or barriers could prevent some of these smaller events. But large scale landslides, like the one that destroyed Oso in Washington, aren't realistically preventable. Instead, we just need to avoid building where ancient landslide deposits reside or where slopes are unstable.
Hurricanes: Every time a large hurricane in barreling towards the eastern seaboard, somebody mentions the idea of nuking a hurricane to stop it. Well, much like nuking a volcano, we would likely just end up with a radioactive hurricane. Hurricanes gain their energy by sucking heat from the ocean surface, so if we want to "power down" hurricanes, we should focus on controlling climate change rather than thinking we can disrupt these massive weather systems. Remember, a single large hurricane packs enough energy to dwarf our planet's nuclear arsenal, so lobbing a lonely nuke into it would be futile.
Climate Change: Hey, so this disaster is one we brought on ourselves and still have a window to limit the damage. Unlike the above disasters, it is a slow rollout of destruction rather than a single moment or a brief window, so we seem to be content with not doing what is necessary to stop it: drop a carbon-driven world and switch to energy sources that don't upset the balance of carbon dioxide in the atmosphere and oceans.
We always like to think an ounce of prevent is worth a pound of the cure, but maybe the cost to develop technology to prevent some disasters is unreasonably high compared to preparation and mitigation of its consquences. This doesn't mean we can't strive to control our planet in these ways, but likely it is a fool's errand to think we can stop them from happening.