Most explanations of sea level rise are either too vague to be useful or too technical to follow. This page is neither. It answers the questions people actually ask, defines every term the first time it appears, and corrects the three misunderstandings that come up most often.
Two things, in roughly equal measure so far.
Water gets bigger when it warms. Not a lot, but the ocean is enormous, so a small expansion across an enormous volume adds up. This is called thermal expansion.
Ice that sits on land melts and runs into the sea. That is new water arriving. This is the part that can get dramatic, because Greenland and Antarctica hold enough ice to raise the ocean by many metres if they were ever to go entirely.
And one thing that does not raise sea level at all, despite being the image everyone pictures: floating sea ice.
Most coastal cities can, in principle, build something. A wall, a barrier, a raised shoreline. Miami largely cannot, and the reason is geology rather than money or will.
South Florida sits on porous limestone. It is full of tiny connected holes, more like a hard sponge than like solid rock. Seawater moves through it. That has two consequences that most people find surprising.
First, a wall does not help much. If the barrier is on the shoreline but the water can travel underneath it through the rock, you have not stopped the water. You have only changed where it appears.
Second, the ground water underneath the city rises with the sea. That matters because the storm drains are already sitting in it. In this project's own measurements, more than half of the drainage structures with a known bottom depth already sit below the wet-season water table. The drains are not empty pipes waiting for rain. They are partly full before the first drop lands, which is why a heavy downpour on a normal sunny day can still flood a street.
El Niño is a recurring warming of the surface water in the middle of the tropical Pacific Ocean. It shows up every few years, lasts roughly a year, and shifts weather patterns worldwide. Its opposite, a cooling, is called La Niña.
Here is the part that catches people out, including people who build dashboards for a living. During a strong El Niño, the floating ice shelves around West Antarctica get taller and lighter at the same time.
Taller, because the weather pattern brings more snow, which piles on top. Lighter, because warmer water reaches in underneath and melts them from below. Height goes up. Mass goes down.
So if you measured only the height of that ice, from a satellite, during exactly the period it was losing the most mass, your instrument would tell you it was growing. Same ice, same moment, two honest measurements, opposite conclusions.
That is the single best reason to be careful with confident-looking numbers, and it is why the monitor on this site labels what each measurement can and cannot tell you.
Two numbers are shown because they answer different questions. The weekly figure is the ocean roughly a week ago. The official index averages three months and is centred about two months back, so during a strengthening event it always reads lower. Neither is wrong. They measure different things, and a page that showed only one would be quietly picking an answer for you.
Arctic sea ice is floating, so its melting adds essentially nothing to sea level. It matters enormously for other reasons: it reflects sunlight, it shapes weather, and its loss lets ocean waves reach places they could not reach before.
But if you want to know about sea level, watch Greenland and Antarctica, where the ice rests on rock.
This project searched for that specific evidence in six independent ways, including a US Senate investigation covering 249 million policy records. No insurer has ever given sea level rise as its reason for leaving a market. Not once.
That sounds like reassurance. It is not. Home insurance is sold one year at a time, so a slow-moving hazard never needs to be announced. It gets quietly priced in, renewal by renewal, or the policy simply is not offered again. A hazard that arrives over decades is invisible in a market that re-prices every twelve months. The silence is what the mechanism predicts.
Broadly correct, and routinely misused. Every major emissions scenario gives nearly the same sea level in 2050, because the ocean and ice respond slowly to what we do. The paths separate after about 2070.
What follows from that is the opposite of "so it does not matter." Choices made now determine the second half of the century, when the paths are far apart. And near-term uncertainty is not small, it just lives somewhere else: in how the ice sheets behave, which is exactly what is hardest to predict and what this site watches.
The ice monitor shows live numbers. Live is not the same as important, and the difference is worth ten seconds of your attention.
Check the lag before the value. Every tile says when the instrument actually observed the thing, and how far behind that is now. A number is not fresh because a web page loaded.
Check whether it raises sea level at all. Sea ice tiles are marked as context. They update daily and look urgent, and they contribute nothing directly.
Notice what is missing. The most important measurement, the actual mass of Greenland and Antarctica, is marked unavailable, because the data requires a login this site does not have. That gap is displayed rather than hidden, and the alarm that depends on it is shown as unarmed. A dashboard with no gaps is usually a dashboard that hides them.