I think the article does a good job of identifying multiple sources influencing the flooding - current and future - in coastal communities. It also demonstrates that we as engineers have a variety of tools to help communities adapt to the changes. My primary objection to the approaches mentioned is that it is only forward-looking. Communities and engineers, generally imagine that the dataset they have - either their own experience or the period of record for instrumentation they rely on, plus some projections of future conditions based on models - is enough to tell them what to expect going forward. I think that in addition, we need to enlist our geologist friends to help us understand the range of conditions we have seen over centuries and, for coastal settings, millennia.
Let me explain a bit by way of examples. Devil's Lake, North Dakota is a landlocked lake, and in the 1930's had diminished to the point farmers were growing crops in what had been lake bed a few decades earlier. Roads, railroads, and dwellings were built based on the "common sense" approach of trusting current observation and the period of record. By the 2010's, a relatively wet multi-decade weather cycle was causing serious flooding of roads, threatened railroads and dwellings as the landlocked lake rose. It turns out that the lake was returning to it's historic extent - which was knowable if a person looked back to the mid-nineteenth century before systematic records were being kept. This tells me part of risk management is to look back at what we can know about natural ranges of conditions, but we must consider a much longer period than simply the years of record. Geological markers (e.g. of old shorelines) or other historic information needs to be considered.
To take it a step further, the cited ASCE article acknowledged some coastal areas are still in isostatic rebound from the last glaciation. When a person contemplates the radical changes in weather and climate, not to mention sea levels, represented by glaciation that probably ended a mere 12,000 years ago, you can see that our efforts to estimate return frequencies for extreme events (100-year floods, 1000-year floods) based on a period of record that is rarely over 100 years is severely lacking in historical perspective on what climate does over long durations. I'm not saying we need to be designing for a return of glaciation, I'm saying let's be more humble and diligent in using the insights available to us. Let's factor in what geologists and others who study the past can tell us about climate and sea level (or stream courses and lake levels) over at least the period of recorded human history. That would in some measure account for the Roman and Medieval warm periods and the Little Ice Age. We might find that even most of the predictions of climate warming models that are currently in vogue would fall within the range history says we should be thinking about.
America has a history of building boom towns and having them deserted when the boom fades, and in some measure that seems to orient us to thinking in terms of the lifespan of a given piece of infrastructure. What I trust we are now realizing is that most of our transportation corridors, infrastructure, and even housing tends to persist in the same or nearly the same location for not just decades, but centuries. That means it is valuable to think longer-term when we design and when we decide what to place where. Europe can offer some insights on the persistence of built features once they are in place.
If we as engineers can bring that sort of perspective to our work, we will have taken a major step forward in managing long-term risks to our built environment from natural events. Then somehow we need to convince planners and zoning authorities to factor our conclusions into their requirements. We also need to consider how this all gets paid for.
------------------------------
James Langseth
Retired
Bethesda MD
------------------------------