1) Electro-Magnetic Pulse. In 1859 a solar flare caused an EMP that drove our telegraph system crazy for three days.
Almost any small nation or NGO terrorist group could buy both a medium size atomic bomb and an ICBM missile to launch it to explode just below where satellites orbit. An EMP as big as the one in 1859, whether from solar flare or a terrorist attack, would knock out both our electric grid and most communication. Stocking spare parts (we have fallen behind on even normal maintenance), as well as protecting electric and computer equipment with Faraday cages which have substantial aluminum housing to carry EMP around equipment to ground and shield equipment, would much improve our chances to pick up whatever pieces are left and reassemble them before civilization dissolves.
2) Climate Change. At the rate world-wide civilization is spewing greenhouse gases into the atmosphere, if we fail to replace fossil fuel with renewable energy, we have a 1/6 chance of ending up with climate as hot as dinosaurs had. We ought not to play Russian roulette with world civilization.
Everything we need to fight catastrophic climate change is at least technically feasible. Liquid fuel for transportation needs more work to make it economically feasible. Carbon capture and storage good enough to actually clean up those greenhouse gas emissions is technically feasible. Global Thermostat does it on a small scale.
Politically it MUST be done by big government because there is no way for big business to profit and it is way too big for private philanthropy. Wind is already cost-competitive with coal, especially if you count externalized costs. Solar is in its infant industry stage and progressing rapidly. It is almost cost-competitive with natural gas when it comes to peak demand for summer air-conditioning.
Utility-size batteries are making progress; they will be needed to make good use of enough wind and solar (our two most abundant and cheapest-but-intermittent renewable energy sources) to meet between 80% and 90% of our electric power needs. Geothermal is in use in small amounts in areas with easily accessible geothermal resources. Right now, it looks as if we can meet 10% to 20% of our electric power needs with geothermal but as we get down to less easily accessible geothermal resources it will get more expensive.