Skip to main content
2016 Was The Hottest Year On Record

Big climate news out today — NASA and the National Oceanic and Atmospheric Administration announced that 2016 was the hottest year on record globally. To consider the implications of the record, we brought together FiveThirtyEight science writers Maggie Koerth-Baker and Anna Maria Barry-Jester, FiveThirtyEight science editor Blythe Terrell and Gavin Schmidt, a climate scientist at the Earth Institute at Columbia University and director of the NASA Goddard Institute for Space Studies.

blythe: So 2016 was officially the hottest year on record on Earth. Gavin, can you tell us what that means?

gavin: We have a dense enough network of weather stations and ocean records since about the mid-19th century to be able to estimate the global average surface temperature anomaly, and 2016 is the warmest we’ve seen in that entire period.

That comes on top of records in 2015, and before that, 2014.

blythe: And when we say “global average surface temperature anomaly,” what do we mean?

gavin: Good question. What we do is we take each of the weather stations and other records and calculate the difference from what you expect at each month and year — that’s the anomaly. It turns out that the anomalies from one station to another are highly correlated — much more so than the absolute temperature. There’s a good explainer of this at the Goddard Institute for Space Studies “Surface Temperature Analysis” website.

anna: And what you expect is based on averages from previous years, right?

gavin: Right. We generally calculate the average from the mid-century period, 1951-80, when we have good coverage and relatively stable climate.

maggie: Gavin, I want to shift gears just a bit and ask you about the data collection that these analyses rely on — the satellites, the land-based temperature stations, the ocean buoys. There’s been concern that under a president who has been highly critical of climate science — to the point of calling climate change a Chinese conspiracy — funding for those programs will be cut or data could simply disappear. There are enough scientists who are concerned about this that they’ve started backing up climate data. I know you can’t really speculate about what the government will or won’t do under Trump — who can? — but I’m curious, since you’re someone who works with this data and is intimately familiar with its collection and use … how easy would it actually be to inhibit climate science?

I’m asking because some scientists are really worried about this. Others seem to think that rather than their ability to make new findings being cut off entirely, what’s more likely is that their findings would be made inaccessible to the public.

gavin: The data collection that feeds these analyses is an international network of national weather services, the fleets of international shipping and an international floating buoy program. It would take a lot to disrupt this — and since we also rely on this network for inputs into weather forecasts and the like, it would be quite counterproductive. So I’m confident that this will continue. It’s worth pointing out that the data is all public domain, as is the code that pulls it together, so it is open science as much as we can make it.

Similarly, climate science is an international enterprise — with strong centers in Europe, Japan, China, Australia, etc., and indeed it has to be — you can’t study climate without that global context.

maggie: So it’s pretty unlikely that somebody would just take a climate satellite and point it toward space for the next four years?

gavin: Yeah — you can’t just change where the satellites are pointing.

maggie: What about the more likely possibility — public access to climate science being limited. We saw this during George W. Bush’s administration in the U.S. and Stephen Harper’s government in Canada, for instance. Reports that were finished but never published. Scientists who weren’t allowed to talk to reporters. Is that something that concerns you? And, if so, why? If the science is still getting done, why does it matter if the reports are published on time?

gavin: I have been quite concerned about political interference in science communications since we had attempts to restrict our ability to speak to reporters in 2005-06. That provoked a lot of action at NASA — for instance, the statement on scientific openness from Michael Griffin (the administrator at the time) — and the promulgation of scientific integrity standards across the executive branch by the Office of Science and Technology Policy in the first Obama term. I’m confident that these standards will be maintained in the future.

maggie: Are there policies you would like to see that would be crucial, from your perspective, to enhancing climate science or producing better projections? What does your field need right now?

I mean, I assume the answer is 💸. LOL

gavin: I think the science community can be doing more to help policymakers make better decisions. I’d like to see a Congressional Environmental Office that calculated the costs of the impacts on climate of specific legislation, like the Congressional Budget Office does for the economy, to better inform people of what the net effects are across a wide swath of problems (climate, water quality, air quality, pollution, public health). That’s more of long-term goal, though …

blythe: Right. And we’re heading into some unknown territory with the new administration, of course.

maggie: One thing I want to ask about is the interplay of normal, background variation and the ongoing upward climate change trend. So, for instance, it’s likely that temperatures will drop a bit now that we’re out of an El Niño period. How do you sort out the signal from the noise, to quote our esteemed leader?

I think it’s hard for people to hold in their heads the fact that both those things are happening at once — normal variation and climate change. I know it’s hard for me sometimes, and I get tangled up in the data!

gavin: This data includes both year-to-year weather variability and long-term trends, and that comes out clearly when you plot the temperatures. The last two years have been affected by the El Niño event that started in 2015 and went through to about May 2016. That’s a warm event in the tropical Pacific that has global impacts on temperatures (and rainfall).

Right now we are in neutral-to-La Niña territory in the tropical Pacific, and so we expect 2017 to be a little cooler than 2016, as I discussed in my FiveThirtyEight article that was published in September. The prediction would be a little different today with a few more months of temperature data in the tropical Pacific, but it’s still pretty solid.

anna: Going back to the question of variability, there’s also regional variability, right? It was the second-hottest year for the U.S. (in the 48 contiguous states), for example, but the hottest year in the Arctic. There’s some weird stuff going on in the Arctic right now — incredibly high temperatures there the last several months. What role does that play in the global temperatures?

gavin: That’s right. 2016 saw some crazy warm temperatures in the Arctic, but also pretty much the whole land surface has been warm. Here’s NASA’s map of the temperature anomaly in 2016, which shows that warmth.


blythe: Why have these high temperatures and ice issues in the Arctic come as such a surprise?

gavin: When we get an El Niño event, we expect warm temperatures in the Pacific, and that spreads globally, with a lag of a couple of months. But historically you don’t see a big Arctic response — and so this year is exceptional. We’ve seen tremendous records throughout the year for low sea ice extent in the Arctic and, even more surprisingly, in the Antarctic too, so it looks like something independent is going on there.

We know that there are a lot of factors in the Arctic that amplify warming — decreases in sea ice lead to reduced reflectivity (because the open ocean is darker than ice and snow, heat from the sun is absorbed rather than reflected back), and that can increase warming too. Alaska has had an exceptionally warm year as well.

blythe: To return for a moment to today’s report, it seems that three record years in a row (2014, 2015, 2016) make it easier for people to conceptualize a warming trend. But as Maggie and Gavin noted earlier, 2017 is expected to be cooler. Certainly there will be pushback from those who say that’s counter-evidence. Do any of you have strategies for interpreting and explaining complex results on this sort of thing?

gavin: Agreed. It’s certainly easier to be aware of the trend if the records keep falling. But we need to be careful here — the important things for the future and for impacts are that the trends are continuing and will accelerate in the absence of efforts to reduce emissions. Additionally, I think if we tell people about these ups and downs ahead of time, as we have been doing with the expectations for 2017, we can show that we do actually understand this system and build credibility for when we talk about the trends.

We do, however, keep harping on about the trends whether there is a record or not, and we’re happy to talk about the other factors that drive variations — volcanoes, El Niño, aerosols, etc. The fact is that the long-term trends are being driven by greenhouse gas increases, and that’s why we are able to predict longer-term future warming.

anna: So about that, Gavin, temperatures are increasing, which was expected. Can you say anything about the rate of acceleration of that increase? Is it steady, or is that increasing as well?

gavin: We do expect an acceleration in warming over the next few decades (assuming continued growth in carbon dioxide levels), but you really can’t detect it in short-term, noisy data. 2013 to 2016 looks like a crazy warming trend, while 1998 to 2012 looks much flatter — but we would be fooling ourselves if we thought short-term variability was predictive of the long-term future.

anna: That makes sense. Getting to what you said about long-term trends being driven by greenhouse gases, there are tipping-point events in the warming, like ice melt in the Arctic. Before the Paris climate talks, we were talking about the 2 degrees Celsius marker. What are you looking at in terms of long-term trends, Gavin?

gavin: There are lots of tipping points in the system at the local level — sea ice, ecosystem change, rainfall patterns, etc. — but I don’t think there is one at the global level, where everything will be fine before we get there and terrible afterward. Rather, as warming continues, we expect more and more local points to be tipped, increasing the impacts of warming at an ever greater rate. But we won’t ever be in a situation where we won’t have decisions to make that affect the climate, and if we want to avoid the worst impacts, we need to keep making more climate-positive decisions.

Temperatures in 2016 were about 1.2 degrees Celsius higher than average temperatures in the late 19th century, which is pretty close to what we consider “pre-industrial” — that’s a slightly mythical beast that the Paris agreement is tied to. Warming rates of a little under 0.2 degrees Celsius warming per decade mean that we are almost certainly committed to more than 1.5 degrees Celsius at this point, and maybe 2 degrees Celsius as well.

blythe: That overview seems like a good stopping point. Thanks so much for chatting, everyone.

maggie: Thank you, Gavin!

gavin: Thanks for having me!

Anna Maria Barry-Jester reports on public health, food and culture for FiveThirtyEight.

Maggie Koerth-Baker is a senior science writer for FiveThirtyEight.

Gavin Schmidt is a climate scientist at the Earth Institute at Columbia
University and the director of the NASA Goddard Institute for Space

Blythe Terrell is a former senior editor at FiveThirtyEight.