Resilient Communities are the Foundations of a Resilient America.

Measuring resilience – not for wimps!

The impossible we do immediately; miracles take a little longer.

There are several efforts currently underway aimed at “measuring” community resilience.  Too often, projects like these have started with great gusto – sails full and flags flying.  And then they ran aground on the sharp rocks of harsh reality.  For almost a decade, we at CARRI have tried to steer our course through the same rough shoals.  We’re not safely through – yet! – but we’ve learned a lot along the way.  In this post, I want to share with you a little bit of what we’ve discovered about the many difficulties in measuring resilience.  In a subsequent post, I’ll talk a bit about how we’re overcoming them.

Difficulty #1.  A community’s resilience is revealed through response to and recovery from a disruption.  In this sense, we can only be certain about a community’s resilience after we see it come through a storm.  This implies that resilience is displayed through the actions taken by the community.

Difficulty #2.  A community’s resilience depends on the type of disruption, the magnitude of the disruption, how the community is structured and how it behaves.  Carpenter and Walker a decade ago made a similar point.  I often talk about disasters having a direction:  a natural disaster’s first points of attack are the physical parts of a community; a recession first attacks a community’s economic infrastructure; a pandemic’s initial point of attack is the people who live in the community.  We have seen communities take very different paths – often with strikingly different results in recovering from Katrina and later Sandy.  Gulf Coast communities impacted by both Katrina and the BP oil spill followed very different paths in recovering from Huricane Katrina than they took in recovering from the oil spill.

Difficulty #3.  Vulnerability is not always the opposite of resilience.  It is well-recognized that resilience has many definitions, but a common thread is action – bouncing back or at least adapting to adversity.  Vulnerability may indicate susceptibility to a type of disruptive event – for example, coastal communities are susceptible to hurricanes.  Two coastal communities may be equally vulnerable and yet one may be much more resilient than the other.  Conversely, a poor community is generally more vulnerable than a rich one; in this case, vulnerability is a (negative) indicator of resilience.  Or, to put it another way, resilience is a manifestation of strength; resilience metrics need to focus on a community’s strengths (and weaknesses).

Difficulty #4.  The yardstick you use depends on what you want to do with the measurement.  If I’m an academic doing a case study of a community’s recovery from a disaster (e.g., Rockaway Beach recovering from Hurricane Sandy), then I will have direct evidence of the community’s resilience – we know the outcome.  However, if I’m a planner in Paramus, I’d like to be able to predict how resilient I might be.

Difficulty #5.  Some parts of a community recover more rapidly than others.  This implies that it may not even be possible to give a single answer to the question “How resilient is community X?”  Some (e.g., Betty Morrow) would say that a community’s resilience is determined by the last part to recover; others imply (e.g., Susan Cutter) that there is a sort of average resilience of a community that can be determined.

While these difficulties are not insuperable, they are traps for the unwary.  We’ve found that a good way to develop resilience metrics is to use the answers to the following questions to guide the development effort.

Who will use the metrics?
How will they use the metrics?
Who will collect the data for the metrics?
When, and how frequently, will data be collected?
How will data be collected?
Where (in what domains) will data be collected?
How will we know that our metrics are valid?

Measuring a community’s resilience is a difficult and demanding task.  It should not be taken on lightly.  But it can be done, as I’ll discuss in my next post on this.