As we've gotten more sophisticated in monitoring the physical world around us, we've acquired a better understanding of just how global some of those phenomena are. It's more of a real-world version of Douglas Adams' holistic detective Dirk Gently who explains his calling thus: "The term ‘holistic’ refers to my conviction that what we are concerned with here is the fundamental interconnectedness of all things. [...] I see the solution to each problem as being detectable in the pattern and web of the whole."
For instance, have a look at the lovely animation of modeled global aerosol movement courtesy of NASA's Global Modeling and Assimilation Office as described in Phil Plait's Bad Astronomy blog post, "Dust. Wind. Dude." You can watch dust from the Sahara make it's way around the globe and see where the soot from forest fires and the plumes of sulphates from fossil fuels and volcanoes ends up. It's jaw-dropping stuff and a beautiful display of large-scale systems at work.
Hurricane Sandy was a particularly nasty example of what happens when thermodynamics meets weather and the whole shebang hits a heavily populated area. Bad things happened, is what I'm saying. For all the chaos Hurricane Sandy caused, the accuracy of the forecast (in terms of the hurricane's track if not its intensity) illustrated just how much the National Hurricane Center's models have improved over time. Some of those improvements are detailed in Tekla Perry's IEEE Spectrum blog post, "Predicting Hurricane Sandy" which talks of how new data assimilation methods incorporated the masses of physical data collected into the computer models to improve forecast accuracy.
If our new normal does now include more severe weather, as it appears that it does, what do we do next? Monitoring is always going to be critical; we're always going to need early warning if Mother Nature plans to smack us down. Prevention is also key, and that's trickier because it means thinking big and formulating plans for disaster prevention and mitigation rather than just clean-up post event. Large-scale, long-term planning for which there must be funding, appropriate technologies, and political will. I'll leave you with two excellent articles to consider. The first, James Surowiecki's New Yorker article, "Disaster Economics" argues that the U.S. needs to develop its own Delta Plan—the name given to the Netherlands' epic flood control project begun in the wake of the disastrous 1953 flood that killed thousands and swamped a half million acres of land.
The Dutch, being a practical people who inhabit a country 1/5 of which was reclaimed from the sea (and lakes and swamps and marshes) are experts in flood control and prevention. Matthew Yglesias' Slate article, "How to Save New York" discusses some of the lessons New York City (and environs) could take from the Netherlands.