Hacker News new | past | comments | ask | show | jobs | submit login
Where America's climate migrants will go as sea level rises (citylab.com)
74 points by jdkee 20 days ago | hide | past | web | favorite | 86 comments

> In the worst-case scenario, in which sea levels rise by six feet by 2100, ...

Where does six feet come from?

It seems very suspicious. It contradicts NOAA Sea Level Trends data of tide gauge measurements, going back to 1930 and sometimes earlier. According to the map, in almost all places in US the projected sea level trend is no more than 1-1.5 feet by 2100:


What is more, the trend is clearly linear from the start of the measurement. The trend hasn't changed in recent years.

For example, see data from the gauge in Sandy Hook, New Jersey, from 1932: "The relative sea level trend is 4.09 millimeters/year with a 95% confidence interval of +/- 0.2 mm/yr based on monthly mean sea level data from 1932 to 2018 which is equivalent to a change of 1.34 feet in 100 years."


Gauge in Lewes, Delaware, data from 1919: "The relative sea level trend is 3.48 millimeters/year with a 95% confidence interval of +/- 0.23 mm/yr based on monthly mean sea level data from 1919 to 2018 which is equivalent to a change of 1.14 feet in 100 years."


There's a major research initiative taking place right now into Thwaites Glacier, as unexpectedly fast melting is occurring. It's thought this is due to warm saltwater penetration underneath the glacier but other factors are being considered.

The issue with Thwaites Glacier is it has a tapered edge that generally resists thawing, but that widens out to a huge contact surface area once erosion by salt water progresses.

This glacier alone will cause a global sea level rise of three feet.


And right next door is a nice mantle plume heating things up. [0] You may have the causality wrong - mantle heats water melts ice - might be the more accurate causal links.

[0] - https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/201...

> This glacier alone will cause a global sea level rise of three feet.

Where did you get that information? This source from Wikipedia says it would cause a rise of 4 inches:

At the beginning of 2020, researchers from the ITGC took measurements to develop scenarios for the future of the glacier and to predict the time frame for a possible collapse: The erosion of the glacier by warmed ocean water seems to be stronger than expected. The researchers noted with concern, that at the baseline of the glacier, the temperature of the water is already more than two degrees above freezing point. They confirm thawing of the Thwaites glacier contributes about four percent of global sea-level rise. The collapse of this glacier alone would raise the sea level by about 65 centimetres (25 inches).[25]


Your quote says 4% of global sea level rise from this one glacier, not 4" - it says 25" at the end there.

As a general observation, it's not that useful to look at single gauges. As glaciers melt, the land nearby rises because of the lower weight on it, which shows up as the sea level going down. This is besides all other kinds of geological activity that might be going on in a particular location.

[edit] Anyway to be more specific, there's a single Antarctic glacier with enough water to cause 0.5 meters of sea level rise, and a recent measurement shows it will melt in "decades, possibly more than a century." https://www.bbc.com/news/science-environment-51097309

Mate, you're trying to contradict NOAA government data with someone's blog.

No, I’m not. The NOAA site does not posit that the underlying trend is linear over the period - it simply displays a linear fit on the graphs. See here for more details https://tamino.wordpress.com/2018/04/16/recent-sea-level-cha...

> The long-term linear trend is also shown, including its 95% confidence interval.

I hope you'll forgive the reader for interpreting this as them positing that the trend is linear.

They’re forgiven - but if you’re saying NOAA are saying the trend is linear and will continue to be you’re wrong. All the site is saying is it displays the “average” trend over the period as a straight line. It’s a comment on how the data is presented, not on the underlying dynamics.

From NOAA web site, when you click on "Previous RSL Trends" button:

> Although the trend may vary with the end year, there is no statistically significant difference between the calculated trends if their 95% confidence intervals overlap. Therefore, the most recent calculated trend is not necessarily more accurate than the previous trends; it is merely a little more precise. If several recent years have anomalously high or low water levels, the values may actually move slightly away from the true long-term linear trend.

I was looking for the same, it appears to come from (or be reflected) at least here:


Where a SLR of 2m by 2100 is reported as an extreme scenario, presumably (but I can't find it in the text) with very low confidence. They give the following definition of the reasons for the low confidence rating, though: "Inconclusive evidence (limited sources, extrapolations, inconsistent findings, poor documentation and/or methods not tested, etc.), disagreement or lack of opinions among experts".

Sure, the trend is “clearly” linear until it’s not. Hell a hockey stick chart is mostly two linear trends stuck together. A sigmoid is too.

Most of your comment is flat-out false. That page shows current instantaneous sea level change; it doesn't show future predictions (except in, say, the paper linked at the bottom about predicted sea level change). Sea level rise is increasing each year and is projected to continue increasing faster each year, the rate dependent on how much CO2 we keep putting into the environment.

2m in sea level rise is roughly the current "intermediate" scenario, which would happen if the world made immediate and large reductions in CO2 emissions. That is, we're on track to be worse than that now.

Since I've seen this exact error on HN before (don't recall if it was you who made it or not): do you understand that giving the 2018 sea level rate change as "1.34 feet per 100 years" is NOT a prediction that the sea level will change 1.34 feet over the next hundred years?

> 2m in sea level rise is roughly the current "intermediate" scenario

What is the basis for 2 m prediction?

> giving the 2018 sea level rate change as "1.34 feet per 100 years" is NOT a prediction that the sea level will change 1.34 feet over the next hundred years?

Global 0.3 m rise is among the possible scenarios.¹ And this scenario is consistent with historical measurements. Also note that CO2 level has been rising for the last 60 years. Sea level trend stayed the same compared to the period before industrial CO2 rise.

¹ https://tidesandcurrents.noaa.gov/publications/techrpt83_Glo...

NOAA says "most probably x", you turn that into "no more than x". That's not a fair and honest represntation of that the NOAA says.

Well, considering that global CO2 emission has been steadily increasing [1], and sea level rise tends to follow CO2 level, it's very unlikely that it will continue to increase linearly...

[1] https://www.epa.gov/ghgemissions/global-greenhouse-gas-emiss...

"Dilkina and her team used migration data from the Internal Revenue Service to analyze how people moved across the U.S. between 2004 and 2014."

It looks like their analysis may have been skewed somewhat by the fracking boom in America's oil shale country. That may explain the otherwise odd migration pattern into sparsely populated areas running up from Texas through the center of the country.

>skewed somewhat by the fracking boom in America's oil shale country.

Isn't it extremely ironic, given the topic hand?

I find it hard to imagine 200k people leaving San Mateo County. most of the county is above 6ft elevation, and there's just so much money I would think dikes could be built to keep the flats dry. At the worst I would expect more housing to be built in the hills.

The West Washington State coast is EXTREMELY rocky and hilly. Pike's Market is on the 5th? floor, alongside the main seattle incline. There's no reason that 6 ft change would displace any significant number of people in that area. It will possibly damage the puget sound port, which has been migrating slowly, for years. Ultimately a rise in sea level would combat the silt buildup that has been a recent issue for the Puget Sound.

I would pretty much guarantee that wealthy areas like San Mateo will make the tax payer pay for dikes and stuff. Poor areas along the coast will be in trouble though.

Optimistic of you to expect the Bay Area to build more housing anywhere.

Look at the people displaced by fires in the North Bay who now can't afford to live anywhere nearby.

I think the OP meant more expensive housing in the hills. Not affordable or dense.

Possibly dense, but almost certainly not affordable. There isn't really any affordable housing in San Mateo County to begin with (source: my $2000/month one bedroom apartment with no dishwasher)

What's interesting is that many of the tech offices are actually adjacent to the bay (eg. Facebook's Menlo Park campus, Google's Mountain View and Sunnyvale campuses, Microsoft and Amazon buildings, NetApp, etc.) and in floodable areas [0].

It's not clear whether those companies would invest in dikes or just move somewhere else, but I'd assume the former.

[0] https://msc.fema.gov/portal/search and search for "1 Hacker Way, Menlo Park, CA", "1600 Amphitheatre Parkway, Mountain View, CA" or "1170 Bordeaux drive, Sunnyvale, CA"

I find it hilarious that even in a worse scenario with the climate crisis, Chicago somehow did not get any migrants .

Their map doesn't make any sense to me. They show no expected migration to really any of the cities in the midwest but somehow they expect a mass migration of people from Miami to rural western Kentucky?

The whole concept doesn't make sense; 80 years in the future we won't even know if the US exists as a singular entity anymore let alone where its centers of population will end up. Imagine if 1920 Russia tried to predict where its population would end up in 2000; nothing would match up.

Why rural Kentucky and not Northern Florida or Georgia?

Are the shades of red based on migration relative to that same region's previous numbers, or some national average?

Hilarious and wrong headed. Chicago is the largest city stateside (Toronto on the Canadian side) next to a Great Lake with the infrastructure and hinterland to support a large population. Fresh water supply is going to be critical in a climate crisis.

Yep the Great Lake region is probably the best situated to weather the storm of climate change.

The irony is that it used to be an ocean before climate change covered everything in glaciers.

On the map shades of red indicate above average immigration. White is unspecified so some meaning of avaerage; not none, but average.

That said I have basically the same question. Why are cities not getting above average immigration? You noticed Chicago, I noticed Buffalo and Rochester, which are the #2 and #3 largest cities in NY.

It appears that the colours have more to do with migration compared to historic trends:

Many may end up moving to the suburbs in the Midwest, where the model shows a larger-than-average influx of migrants compared to historic trends.

So, it is really showing places that have not historically had a lot of people moving in.

Right. The complaints seem largely about being unable to understand the map.

If rural western Kentucky gets 2 climate migrants and Chicago gets 100, this may represent a much greater increase in Kentucky migration rates than Chicago migration rates.

According to the map, Midwestern cities are immune. Only random rural areas will experience an influx of people. In Ohio for example Columbus, Cincinnati, Cleveland all are white while middle of nowhere is red

> Counties in shades of pink and red will see higher-than-average migration

It may be that the extra migration doesn't impact that average comparted to less populous counties.

The map is very rough and relies on counties affected, rather than topography (elevation and closeness to bodies of water and waterways) so some counties can have coastal exposure but also have shelter (high elevation). King county WA for example or Los Angeles co. CA

So it presents a more alarming picture than one would imagine.

Yes. That map shows entire counties in blue if any point in the county would be flooded.

For almost all of coastal California, you don't have to go very far inland until you're way above sea level. LA looks flat, but go three blocks inland from the beach in Venice and you're up a few meters. The California coast is the shelf of a mountain range.

Florida, though, does have real problems. Most of the state is barely above sea level. And Miami is on permeable rock; sea walls won't help.

Sea level rise will not be the problem. It is going to happen on multi-generational timescales.

Food insecurity, resource conflicts, extreme weather events, and a global migrant crisis (caused by the first two issues) will be. Unlike sea level rise, all it takes is one bad year to spark these.

Here in NZ, we already see issues where storms and 'king tides' cause significantly worse outcomes when they coincide. As those king tides rise and penetrate further inland, these increased weather events are going to be much worse at the coast.

Also, Miami and Venice are on borrowed time.

In 1903 humans first achieved heavier than air powered flight. Less than 60 years later, in 1959, man landed on the moon.

I feel confident that in the next 80 years we'll manage to figure out how to build some 6ft tall levees.

There will be little climate migration. People like living near the coast.

I'm in full agreement, but that idea isn't popular here, where I find folks a bit fatalistic. I understand why of course, assuming 'someone will solve it' takes the urgency out of the problem. We do need to take action, absolutely. But am I worried about sea level rise here in Florida? Not at all. There's just too much money at stake to throw our hands up and run. And I don't mean just here, or even the US... most of the world's wealth is on the coast.

A third of the country will be underwater sounds a lot more scary than we'll have to build some small walls, about 1ft per decade, in some areas.

About one third of the Netherlands lies below sea level, with the lowest point being 22 feet (6.7 meters) below sea level. Doesn't make for much of a post-apocalyptic visual - act now or end up like the Netherlands.

> Less than 60 years later, in 1959, man landed on the moon.

That was 1969, or less than 70 years.

"The first human-made object to touch the Moon was the Soviet Union's Luna 2, on 13 September 1959. The United States' Apollo 11 was the first crewed mission to land on the Moon, on 20 July 1969."


Poor reading comprehension on my part. You are, of course, correct.

This seems to only be factoring in oceans. Won't the sea level rising have a huge impact on the rivers and floodplains in the middle of the country that this predicts will see a big influx of people? There is plenty of high elevation coastal land that I would expect to be safer than land on the banks of the Mississippi River.

No. Why should it? If the sea level rises two feet, how does it affect rivers that are, say, 100 feet above sea level? Or 500 feet? What's the cause and effect here?

The sea level rise is due to more water in the ocean. That doesn't translate into more water in a river, because water doesn't flow uphill.

Rivers and floodplains wouldn't be effected by the rising sea level. River systems get their water from precipitation not the sea. The only effect a sea level rise would have on a river system is the delta of the river would move further inland with the coast line.

However, one could argue there would be an increase in precipitation as less water is trapped in ice but that doesn't seem to be what many people are predicting. I say all this with the caveat that I am not very well read up on the science of all of this, I'm just going by what I have read in articles and what not.

It seems really unintuitive to me that there is enough ice on the globe to raise the sea level by 6 feet (let alone 230). But then if I had to guess the land area of Antarctica, I would have estimated something the size of Oklahoma or Texas (when Antarctica is 40% larger than the USA)

Much of the Antarctic landmass is below sea level, if the Antarctic ice shield melted the continent would look like an archipelago. There is much water tied up in the Antarctic ice shield(well, it took 40 myr to accumulate, you get this much even in a cold desert), on average it's 2 km thick. The IPCC estimates that if it melts fully it would result in 50 m of sea level rise.

Any ice below sea-level will melt into water that will take up exactly the same amount of space. So any ice below sea-level that melts will not cause sea levels to rise.

Could you please elaborate on your claim? I feel confused.

I learned that ice has a (slightly) lower density than water, hence it usually floats. I was convinced that submerged ice is doing so because of an additional weight on top of it (above-water ice pack).

Maybe there is also ice that is submerged without this extra weight, but then its density should be lower than water.

Either way, I don't see how submerged ice could take up exactly the same space when melted. If it has a different density, it should change in volume (elementary physics).

Ice that is floating displaces an amount of water whose weight equals its weight. That's why a little bit floats above the water and most is under the water.

When that ice melts, it still weighs the same amount and fills exactly the amount of space that it was displacing. In other words, it fits perfectly in the space in the water that it was already occupying.

So floating ice that melts (above or below the water) doesn't change sea levels at all.

Only melting ice that is currently supported by land will increase sea level.

A reference for anyone unfamiliar with this explanation: this is Archimedes' principle.


I've heard most of the rising sea levels will come from thermal expansion of water, not melting ice.

Liquid water is notorious for having almost no thermal expansion, but the little bit it does have will be multiplied by the volume of the ocean, then divided by the surface area of the ocean, yielding the rough vertical rise.

Millions flee from Miami to Buffalo, and yet Venice and the Netherlands plod on.

I had this wonky idea that maybe buildings in these area should be adapted to have their first floors underwater yet still usable in some ways.

No because you'll get black mold growing that will spread throughout the structure.

Maybe in otherwise dry areas, but I doubt it.

Depends on how you go about it.

In the 1800s, Chicago was lifted up from nearly being at the same level as Lake Michigan: https://en.m.wikipedia.org/wiki/Raising_of_Chicago

On a small scale sure, there's always exceptions.

I don't see this happening across the entire coastline of the US or (even less likely) along the shores of affected tributaries.

I live in Atlanta and I've been joking that if things get really bad, at least I'll have beachfront property.

There's going to be good money to be made in climate change-related real estate trends!

In the free market of ideas, no one actually believes this will happen.

Just look at coastal real estate. Buyers and lenders would have to be colossal fools to pay such premiums to be waterfront. Even Obama just plunked down 12M on a waterfront mansion on the low lying island of MV.

Real estate prices are priced by mostly by current demand, not future oncoming disasters.

Much waterfront property is _already_ vulnerable (and expensive to insure) today, regardless of whether you believe in consensus climate science, due to storms and flooding in current climatic conditions. That house by the ocean is going to destroyed at some point, the only question is when and how.

I have seen ample consensus amonsgt reputable scientists that the earth is getting warmer, and this fact is due to human activity (e.g https://climate.nasa.gov/faq/17/do-scientists-agree-on-clima...)

I have not yet, however, seen any such research on consensus for the effects of climate change. Is there a large, documented consensus on any estimate of sea level raise?

Sorry, perhaps the emphasis in the way I wrote that comment made it unclear what I was saying.

My point was that real estate market prices are much more heavily influenced by current conditions, and current demand, rather than hypothetical future conditions. The price elasticity of demand has little to do with whether the buyer believes that the property will flood in future decades due to climate change. Someone who wants a beachfront mansion is probably going to buy it for close to the same price whether or not s/he believe it will still be there in 50 years, just as s/he might buy an expensive car that will similarly degrade and need expensive maintenance exceeding the original purchase price over time.

> Much waterfront property is _already_ vulnerable (and expensive to insure)

If we are talking about the US then there's government flood insurance.


Well, I don't believe in "consensus" [insert field] science because that's an oxymoron.

If you still believed what was once the consensus, you would believe in some crazy stuff. Science is skeptism of the consensus, not the other way around.

>Science is skeptism of the consensus

As long as the skepticism is evidence based, of course.

Buyers and Lenders are in fact, colossal fools. People build in floodplains all the time.

Obama only ran the country for 8 years, but hfdh434535 knows better.

Real estate is about the closest thing to putting your money where your mouth is as far as climate change goes and the people have spoken - the coasts are still the most valuable and desirable places to live.

>Real estate is about the closest thing to putting your money where your mouth is as far as climate change goes and the people have spoken

I would definitely buy coast property today because this won't be a significant problem in my lifetime. I'd buy from coast again when it becomes a significant problem, just that the coast location would be different of course.

You're re missing the point. Look at the map, the blue region is the extent in 100 years.

What do you think happens in that 100 years? I assume this would be a near linear progression. That means current waterfront has about 10-20 years, not 100.

The blue region will not be totally underwater. The most pessimistic estimate is a 6 feet rise by 2100. A coast property, if you are inclined, can deal with that.

The note under map says: Blue indicates counties where flooding will displace residents if sea levels rise by six feet by 2100. Counties in shades of pink and red will see higher-than-average migration, with the darker shades representing larger population increases.

So it only means people will likely migrate to other areas, not that the blue places will be totally underwater by 2100.

Got me wondering though, are you really a climate change skeptic? I mean climate change is something as obvious as gravity, we have all the data to support it, we can directly observe it etc. I find it hard for a sane people to question it, really. I mean I understand how someone can be skeptical about "man made climate change" which is something open to debate, but climate change itself isn't really open to debate.

Does the climate change? Of course. Are we doing a good job predicting it and understanding how it works? No. Not even close.

I have a background in financial modeling, using exponentially more observations, in a less chaotic system, with participants betting on the models with real dollars. It makes climate change models look like simplistic toys, which most are.

There is NO consequence of being wrong with a climate model, no shame, and there's a high degree of survivorship bias. Climate models that have been right, have been right for the wrong reasons. Climate models that have been wrong (most) are long forgotten and never discussed. New models come out all the time to hype fears for events that prior models predicted would be happening now (and aren't).

The question isn't does the climate change, the question is do we understand it at all.

That is true, always has been and always will be true, regardless of where the coastline happens to be at the time :)

That place of Obama's will probably be fine for his and his wife's remaining life, and the life of his children, and at least a good part of the lives of any grandchildren. He'll get his money's worth out of it.

Most people don't buy real estate considering an 80-year time horizon. Even 40 years is often a bit of a stretch. I would guess that a lot of people only look 15-20 years into the future. Perhaps not a majority, but enough to keep waterfront property attractive and expensive.

Next you will tell me it's impossible for a major earthquake to strike San Francisco (or Tokyo), because then nobody would want to live there.

crazy how coastal real estate isn't tanking

I think it's because insurance policies are still supporting real estate in areas at risk.

This is what I take away:

- Everyone already knows rising sea levels will cause more flooding and property damage.

- What's new here is that they make attempts to estimate displaced populations and where they will migrate.

- The reception of American climate migrants will not be evenly distributed. Cities nowhere near the coast will be impacted by climate change.

Applications are open for YC Summer 2020

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact