THE FREAK SUMMER STORM that laid waste to much of the mid-Atlantic on Friday night left chaos in its howling wake — and a mess of questions about the region’s capacity to cope with the unexpected.
The issue is framed as the “capacity to cope”. In this framing and throughout the editorial’s analysis there is a predisposition to an effective response that will quickly and fully restore the prior condition. This response-orientation is too narrow.
In Northern Virginia, where Verizon handles most 911 calls, emergency phone service simply did not exist for much of the weekend, even as residents scrambled to absorb a surge of bona fide emergencies. Suburban Maryland’s main power provider, Pepco, once again scrambled to restore electricity to hundreds of thousands of customers who have come almost to expect wildly inconvenient outages in extreme temperatures.
What these — and many other — examples point to is the increasingly interdependent character of the technological webs on which we have built our daily lives. On most days these interdependencies generate substantial benefits. But on bad days the same connections can be a collection of cascading vulnerabilities. The rush-to-blame service providers is too easy and — more importantly — obscures fundamental issues of real risk readiness.
In both cases, residents of the national capital region could only wince as they imagined what might befall them in more cataclysmic circumstances — a terrorist attack targeting not just population centers but critical infrastructure, for instance — and pondered the painfully evident lack of disaster preparedness.
I agree this was not a cataclysm. As bad as it was (still is for many), this was far short of a catastrophe. I agree there is good cause for the National Capital Region to anticipate a real catastrophe.
But what sort of “preparedness” is envisioned? Is it preparedness to put Humpty-Dumpty together again? The nursery rhyme has already warned us in this regard.
Malicious intent — criminal, terrorist or otherwise — brings with it a psycho-social multiplier effect that deserves our attention. But intentional threats often pale beside natural and accidental threats. Consider the potential implications of a New Madrid seismic event or an accidental collapse of the regional grid.
“We have emergencies,” said Sharon Bulova, chairman of Fairfax County’s Board of Supervisors. “Especially in the national capital region, we are susceptible to things happening, having public safety compromised.”
How, then, can the region be so ill-prepared?
I don’t expect to convince anyone who has been sweating out the power outage since Friday night, but the pace of restoration has seemed to me reasonably rapid.
When a hurricane or blizzard is forecast, the owners/operators of critical infrastructure have a day or more to prepare. This event-specific preparation often involves pre-deploying and enhancing response assets. If at all possible, additional electrical and telecommunications repair crews will be brought in from other regions outside the cone-of-uncertainty. The general population, famously, stocks up in advance and — in the case of hurricanes — may move out of the way.
On June 29, even if someone had gone to red alert as the derecho crossed the Ohio River, the realities of time and space eliminated this kind of preparation. That’s why no-notice — or minimum notice — events are so fundamentally different than hurricanes or blizzards or — with recent advances in weather prediction — even tornadoes.
That’s the question for leaders to contemplate as the cleanup continues. And not just elected leaders, but corporate ones too: Verizon and Pepco both owe the public a much more thorough accounting and, more to the point, explanation of why it is taking so long to set things right again.
It will always take “so long to set things right again” if we persist in the illusion that we can wait to respond or that our preparedness is mostly a matter of being ready to respond. Given the nature of our interdependent systems and their shared vulnerability to non-typical events, we are much better served to focus on prevention, mitigation, and resilience. We also ought to be more creative in conceiving and executing recovery operations. Failures will recur. Catastrophic failures of distributed interdependent engineered systems are infrequent… but practically inevitable.
Verizon, for its part, has been opaque about the 911 service crash in Northern Virginia, furnishing only vague answers to questions about why its primary and backup power sources were vulnerable and what can be done to avoid a repetition.
Then there’s Pepco. In the annals of corporate spin control, the company’s unabashed announcement Monday that it planned to restore electricity to 90 percent of its Maryland and District customers by late in the evening of July 6 — seven days after the storm — must qualify for a special mention in the Lowered Bar Category.
Or are these examples of honest uncertainty and worst-case realism? One self-described weather nerd told me, “A derecho is a 240-plus mile front of 80-plus simultaneous F-1 tornadoes.” Yet by Tuesday midnight telecommunications systems were — if still a bit unstable — mostly working. Electric utilities were reporting restoration of the network’s backbone and were turning to the very time consuming process of reconnecting individual customers. The number of National Capital Region outages had been reduced from about 1.5 million to less than 110,000 in less than four days for an uncommon, no-notice, very hard-hitting event. Despite extraordinary heat the public health consequences have been modest. The celebration of Independence Day on the National Mall proceeded. (Contrast this with the situation in West Virginia where late Wednesday 280,000 remained without power, down roughly 50 percent from the peak on Friday night.)
Should customers for whom power comes back midweek really be impressed that they suffered for just four or five days instead of for seven? And what of the 10 percent of customers whose service will still not be back by Friday night? Are they condemned to a second weekend with no air conditioning or refrigeration?
All of us might take a few moments to consider the connections — technological and human — on which we depend. What is the nature of these dependencies? What is the consequence of — unexpectedly — losing these connections? Is there anything we can do — now, today — to mitigate these consequences?
Consciously or not we typically make one of four choices regarding risk: 1) we transfer the risk to someone else, 2) we accept the risk, 3) we reduce the risk, or 4) we avoid the risk. The Washington Post editors seem to be trying to transfer all the risk responsibility to Verizon, Pepco, and other providers. Certainly these owners/operators should be held to high standards.
But any attempt to transfer all risk will only hide a high level of accepted risk. The level of risk accepted will be even higher because it is hidden.
It is delusional and dangerous if we — each and all of us — do not accept at least equal responsibility for the kinds of risk outlined above. What can each of us do to reduce the risk associated with the consequences of the most hard-hitting events?
It’s little consolation to imagine that some things might have been worse. Pepco, despite leaving hundreds of thousands of homeowners and businesses in the lurch, did manage to prioritize restoration of service to hospitals, nursing homes and, critically, Metro. Dominion Virginia Power was also able to restore electricity relatively quickly to hospitals in Northern Virginia as well as to the main jail in Fairfax County.
Damn with faint praise? Might this just be an indication of planning, preparedness, and a mitigation strategy in action?
The storm gave rise to massive inconveniences and discomforts across the Washington area. Usefully, it also exposed the region’s absence of reliable fail-safes, spotty preparedness and sluggish response times in the face of emergencies. Now it’s up to leaders to identify and act on those shortcomings.
Yes. We should treat this as a near-miss and learn every lesson possible.
But inconvenience and discomfort are the least of my concerns. Someday a no-notice, potentially catastrophic disaster will keep power off for more than a week. Telecommunications will be similarly disrupted. Fuel will be in short-supply. Delivery of water, food, and pharma will be uncertain. Our response may be further complicated by concern over biological, radiological, or some other potential contamination: natural, accidental, or intentional.
Leaders do have an important role to play. Part of that role is attending carefully to improving response capabilities. But even more important — and too often ignored — is identifying opportunities to prevent, mitigate, and improve resilience.
And it is not only a matter for political and corporate leaders. Organizing our economy and much of our lives around various interdependent distributed networks involves both risks and rewards. We tend to take the rewards for granted and deny the risks. This is irresponsible. It is unrealistic. It is a recipe for catastrophe.