What would it take to predict climate change inundation well?

Last week I ranted about a poorly executed report claiming to show which properties around the US are at risk of climate change driven flooding. Peter Abrahamsen responded with:

Also, mean sea level, however it’s measured, doesn’t tell you who’s at risk of flooding.

and then pointed me at a couple of much better examples, most notably the Victoria, BC Capital Regional District’s climate change modeling project. I think it’s worth looking at all the other questions that would have to be answered to really get this right. Before diving in, I want to make clear that I’m not piling on Zillow any more: last week’s post was about flaws they should have figured out as a real estate data company, while most of what’s below will be details that they never claimed expertise in or took a position on.

0. What do we even mean by “sea level”?

Tides are big round here, big enough that you don’t have to spend long at a beach to notice the movement. They’re also complicated: each day we have two low tides and two high tides, but the difference between the lower and higher of the low (or high) tides can be several feet. Naturally, mapmakers have to pick a line somewhere in the range between highest and lowest ever, but the choice that makes sense for their purposes may not help us. Drawing a recognisable coastline calls for avoiding the extremes of high and low tide, but if we’re interested in flood risk it makes sense to start with the highest sea levels we’re already getting, which have come to be called King Tides.

Aside: even agreeing where “zero altitude” or “mean sea level” is turns out to be much more complicated than it looks. When a GPS device tells you your altitude, it’s giving you a height relative to a smooth model that massively simplifies the real shape of the planet. If you want to go down that rabbit hole, I recommend starting with this primer: Fundamentals of Mapping: Datums.

1. OK, then what’s the highest current sea level?

The difference between mean sea level and the highest tides also varies widely by place, even across fairly small distances. Victoria, BC’s sea level rise assessment included a thorough analysis of that, summarised in this map:

Variation in the height of highest tides around Southern Vancouver Island

The N-S distance covered by that map is roughly equivalent to the distance between Seattle’s Northern and Southern city limits. So the differences can be big enough to be worth taking into account even across a single city. The general trend in Puget Sound is that places further South have larger tidal ranges than those further North, but it’s not a perfectly smooth, simple relationship. We have to get this far to even understand the starting sea level before projecting decades or centuries of sea level rise.

2. How fast will the sea rise generally?

This may be the most obvious statement of this post, but here goes: how fast the sea will rise is a hugely contentious question even among the people who aren’t trying to pretend that climate change is not happening. Here’s a very rough sequence of some of the questions that have to be answered to come up with an estimate:

  1. How much greenhouse gases do we expect to be emitted in the future? (this isn’t even a geoscience question – it’s about economic and population forecasts, and how much faith we have in either technology or policy to help)
  2. What is the climate system’s sensitivity to greenhouse gases? (i.e. how much warming do we expect from a given addition of greenhouse gases)
  3. How much of the warming will be absorbed by the sea? (answering this lets us estimate how much the expansion of water will increase sea levels)
  4. How much of the warming will be observed in polar regions?
  5. How much of Greenland’s & Antarctica’s ice caps can we expect to melt at a given temperature, and how quickly?

It’s impossible to answer any of those questions with absolute certainty, and for most of them the closest thing that exists to a scientific consensus is a range of estimates. So we have to make assumptions and be honest that that’s what they’re doing. This is a part that the Zillow report handled sensibly: without going into much detail they cited a reasonable source for high and low estimates, and said they were simply going to pick a number midway between those. Even the Victoria report that I’m so impressed with picked 3 round numbers (half a metre, 1 metre & 2 metres) and ran scenarios framed as “what would happen if the overall sea level rise were X?”.

3. How fast will the sea rise here?

But of course, there’s a catch. Whatever assumptions we make about global sea level rise, we’re still only working out an average. But sea level rise is unlikely to be evenly distributed around the world. In fact some unevenness has already been measured, because of effects of prevailing winds, changing currents and the reduction in Greenland’s weight because of ice that’s already melted in recent years.

This is a larger scale question than the one about today’s highest sea levels, in that the difference between two points in one city or county won’t be enough to matter. But if we want to do an analysis for, say, the whole United States we definitely have to take it into account.

4. Will the land rise or fall?

17,000 years ago, Seattle was under 3,000 feet of ice. That weighed enough to push the land down significantly, and when the ice melted the land started to rise in response. It’s thought that the land may have risen by as much as 275 feet after the Ice Age ended, and while the majority of that happened a long time ago, the process has not finished. Victoria’s report estimates that parts of their region will rise by as much as 31cm (just over a foot) in the coming century. Most of their region is expected to rise by about a third as much, and some not at all. That’s a pretty significant modifier if we’re assuming 6 feet of sea level rise. In general it will reduce impacts on this region a bit, while the equivalent mechanism will make things a little worse for the US’s Atlantic coast.

5. How will the weather change?

Wind and air pressure both have noticeable effects on tide heights. This is the main reason that tide tables usually warn users that they are tide predictions: the weather on a given day can throw them off by as much as a foot. One of the reasons that this region tends to get its King Tides in winter is that that’s when we get our biggest storms, lowering air pressure and pushing water towards the shore as strong winds blow in from the open ocean. Climate change is generally expected to make big storms more frequent, and the biggest storms more intense. So weather will probably add to the problem, and we’d need to estimate how much to get this right.

6. How much flooding is enough to matter?

Having answered all of the questions above, we’re finally in a position to go back to a map of property parcels and see which ones will be touched by flooding. But we still need to look at individual properties to understand how big a problem this will be for each one. Consider a few example scenarios:

  • A small, flat parcel with a house on it, and the entire parcel is flooded. Obviously this one’s in serious trouble, and they’ll have a hard time moving once the problem becomes clear enough to wipe out the value of their land.
  • A long narrow parcel sloping towards the shore (pretty common round here), with the house built on the higher part so flooding will just shorten their lawn a bit (for now – see below).
  • A similar long and narrow parcel, but the house is built right against the beach. They’ll have to move the house, but they have much better options than the first one in the list.
  • A house built on stilts.
  • A houseboat that can rise and fall with the tides, but above a certain height it would break free of its moorings – sudden transition from no worries at all to total disaster.
  • A property that’s going to be comfortably above all flooding, but its only access is through a road with a low point that will be underwater every high tide.

I recently read a great article from the other side of Puget Sound that looks at how this works for individual properties and landowners.

It’s difficult—maybe impossible—to do this kind of analysis at large scale with aggregates. We’d need to know something about each individual parcel. And other questions about the owners will also matter: is this the only home of someone whose entire savings are the property itself, or a vacation home owned by a rich family whose primary home is safe?

7. How will changes in sea level move the shoreline over time?

After all that, there’s at least one more complication: shorelines are dynamic. Sometimes they change abruptly, like this section of Seattle’s Magnolia Bluffs which collapsed abruptly in 1996. Usually it’s less dramatic than that, but the sea moves sand and gravel around all the time, eroding the shoreline back in some places, and building wider beaches in others. Any change in sea level can be expected to drive more erosion, and if we do get the expected increases in storms those will contribute too. This won’t be a simple linear effect, either. Some layers of rock are much more resistant to this erosion than others, so if sea level rise lets the sea hit a softer layer, that location will see a big jump in erosion. Imagine the smug owners of that property I described with the sloping lot and the house on this highest ground: if they’re unlucky with the underlying geology they may suddenly find that each storm eats another foot of their lawn.

“All models are wrong, but some of them are useful”

So I’ve spelled out a lot of separate issues we’d have to make assumptions about to start capturing the real complexity of how sea level rise will affect a given property. I am not an expert on any of this, so I’ve probably missed as many subtleties again. But I don’t want to leave you with the impression that it’s not worth trying. For each of the questions, there is a body of research and there are experts who can give ranges of likely outcomes. And with those ranges, it’s perfectly reasonable to do what Victoria’s done and map out some what-if scenarios. It’s just important to acknowledge that it is a complicated picture with many important details and much local variation.