The worrying question about climate journalism right now

Recent data hint at a downtrend in coverage. But the matter requires more investigation before we can say for sure what's going on.

The press room at the April 2011 U.N. climate change conference in Bangkok. UNFCCC, via Flickr, CC BY 2.0.

Update 9/14/2025: When this article was first published, the figures below showed data through July 2025. The main chart has since been updated through November 2025 and will, hopefully, be updated continually.

We’re coming off the two hottest years on record. We’re going to blow past 1.5 degrees Celsius of warming. But in the U.S., there’s more reason than ever to question whether any substantial steps will be taken to avert this. It feels like Donald Trump’s re-election was a repudiation of all things climate. The administration is now reversing Biden era climate and clean energy policies, has fired many climate- and weather-focused federal workers (although some have since been rehired), and has discontinued or failed to release key climate and emissions related data.

How are the media responding?

This is a question to which I’ve been giving a lot of thought lately. Some things are obvious. We’re getting a lot more politics of science coverage in the climate sphere (what is the Trump administration doing to science and scientists?), as opposed to pure science coverage (what is the Earth doing?).

But there’s also something less apparent, and I’m not entirely sure what to make of it yet. So in this post, I’m going to do something new: I’m going to raise a question about a dataset I’ve been monitoring, without being fully able to resolve its answer. And I invite commentary from experts (and anyone) as to what they think might be going on.

To what am I referring? Well, the main dataset that tracks newspaper climate coverage around the world is showing that in many regions, including the U.S., coverage levels are noticeably lower lately. This dataset, the work of the Media and Climate Change Observatory, is hosted at the University of Colorado Boulder, but is produced by experts at multiple universities. The dataset (like all datasets) has a number of limitations, which I will discuss further below. However, it remains the best we have to go on.

“I don’t think there’s a better dataset, with the same coverage and longitudinality,” said science and climate communication researcher Mike Schäfer of University of Zurich, who is not one of the dataset’s authors, in an email.

So what is it showing?

U.S. trends

Let’s start with the U.S. Here is what we see in the 5 leading papers (and note by way of disclosure that, until about a year ago, I worked at one of them, The Washington Post):

[Note on 10/23/2025: This graphic will continue to be updated with monthly data going forward, so it may contain more data than it did when this article was first published.]

The volume of coverage of climate change at these major papers has clearly grown over time, with a particular ramp-up during the first Trump administration. As I have previously discussed, this in significant part reflects conscious investments by newsroom leaders.

Coverage then took a dive during the COVID-19 pandemic, not surprisingly (I and so many others switched beats at that time to cover the emergency). But it then recovered and reached a new peak in late 2021. Upon close investigation, that turns out to be the time of the United Nations global climate conference (or COP26) in Glasgow, Scotland in October-November of 2021, upon which the Biden administration focused heavily, and which the then-president himself attended. Here’s a little walk down memory lane for you:

Joe Biden speaks at the Glasgow climate change conference on November 1, 2021. Official White House photo by Adam Schultz, public domain. Via Wikimedia Commons.

After Glasgow, though, it’s downhill. Especially this year, although in July there was a bit of an uptick. In significant part, this reflects coverage of the Texas floods, according to Max Boykoff, a professor at the University of Colorado Boulder who is the lead researcher behind this dataset.

All of which leads to the big question: Is there something happening in the world right now (like, you know, the Trump presidency) that is leading to a decline in the volume of climate change coverage in these newspapers? Or, is this more something cyclical, where you need to be cautious interpreting a downturn over just a few years, from a previous high?

I’m uncertain, for reasons I will explain. Still, I think we can at least raise the question about what these data are telling us.

Trends in the UK

If the pattern above were only present here in the U.S., that would be one thing. I think it would be easier to dismiss.

But whatever this is, it is not just a U.S. phenomenon. If anything, the change in the UK appears even more dramatic:

Across six leading papers, recent coverage levels are actually around as low as they were during the COVID-19 pandemic (which is not the case in the U.S.).

James Painter, an expert on climate change coverage at the Reuters Institute for the Study of Journalism at the University of Oxford in the UK, told me he does think there has been a coverage downturn.

“Certainly in the UK, there’s a drop,” Painter said.

“It’s the news agenda more than anything,” he continued. “There’s just not much coming out which is newsworthy, and the government itself is sort of not paying as much attention because it is so contested and toxic in the UK political scene.”

Painter had a caution though, which gets to one of the limitations of the data that I’ll discuss further below. A lot of media coverage in the UK of late has been about the country’s contested “Net Zero” policy for the year 2050, Painter said, sometimes in articles that don’t mention the MECCO search terms “climate change” or “global warming.” If terminology is changing, the data above won’t be picking that up.

Is terminology changing across Western countries, though?

If you look at the data by global region, you see apparent declines from the Glasgow high point in both Europe and North America. You also see something similar in Oceania, although the coverage peak comes later (the data here only include Australia and New Zealand papers).

Yet you do not see the same pattern in Africa, Asia, or Central and South America. In Africa and Central and South America, Glasgow itself was not even the high coverage moment, with the actual peak instead occurring the next winter (2022), timed to the next U.N. climate conference, which was in Sharm el-Sheikh, Egypt (or COP 27). (In the Middle East, coverage does appear to be down from a high moment in mid 2023).

So again, what is going on here?

Understanding the data, and the role of media change

It isn’t hard to spin out possible reasons why climate change could be garnering less media attention in the last few years in some countries or world regions.

Many events during the 2020s have potentially outcompeted it for media attention. There are the two wars — Ukraine and Gaza — and the U.S. 2024 presidential election, in which climate change did not feature as a major campaign issue. And then in the U.S., the second Trump administration has come in with a “flood the zone” approach that makes it hard for media to focus on any one topic, stretching journalists thin and potentially forcing them off of this story.

“The hypothesis you've outlined here is quite probable: media outlets may very well be reallocating coverage as they're getting caught up in other events/ developments/ stories in a finite news hole,” wrote Boykoff, the lead researcher behind the dataset shown above, in an email. “That is certainly a strong possibility so worth considering further.”

However, there are also reasons for skepticism. The issue lies in what we can safely infer from the data.

“Personally, I would not overestimate the decline after 2021,” said the University of Zurich’s Schäfer.

“There have been declines before, coverage is still above pre-2020 levels, and the effects of the ongoing economic crisis of legacy media (shrinking numbers of articles overall) are not visible here as well,” Schäfer continued.

The point is well taken. Again, the dataset shown here, MECCO, is the most prominent publicly available dataset for tracking such things. It produces monthly data for newspapers all over the world and also for world radio and TV in the U.S. (which also show a decline). And as noted, the general method is to use databases of media articles to count mentions of either “climate change” or “global warming” in various papers (although in some countries and regions, slightly different terms are used). This means that an “article” could just as easily be an op-ed mentioning climate change as a news story.

As Schäfer notes, there are limitations to these data. First, these raw or absolute article numbers do not necessarily reflect what may be happening to the denominator: how many articles there are of any kind being published in the papers. That total amount of coverage has likely changed over time during an era of convulsion in the news industry.

Indeed, the ongoing era of media convulsion has meant that more and more influential content is online and in social media, or being produced by creators rather than traditional journalists. But this content is not being measured by this method.

As I have previously written, the growth areas for climate coverage right now tend to be more specialized or locally-focused, and often nonprofit endeavors. In other words, sources that would not be picked up in this search:

And then there’s that issue of terminology. It may well be changing, notes Lucy McAllister, a researcher at Denison University who works with Boykoff on the MECCO database.

“Our searches only include two key words and I wonder if how we talk about climate change has significantly changed since Glasgow,” McAllister said in an email. “For example, under the Biden administration as you know there was a huge push for clean energy, just transitions, environmental and climate justice etc so I wonder if stories were focusing on these terms and renewable energy projects and leaving out mentions of climate change/ global warming?”

I discussed the three issues raised above with Boykoff before publishing this article. He agreed that the data are not adjusted for how the overall volume of coverage may be changing in different papers. He also agreed that it is possible that the terminology used in the climate debate may be changing.

Still, Boykoff said he believes newspapers remain a very relevant indicator. They still set the agenda much of the time, or reflect it.

“It’s still considered to be a way to track what are the news stories of the day,” Boykoff said.

Finally, there’s just a statistical issue. As we know in the climate change debate itself, peaks can lead to incorrect conclusions. When it comes to global temperature data, El Niño causes peaks, and if you begin or end a trend analysis on one of those, you can reach questionable results.

Here, we’ve got a peak in 2021 in many regions and countries, and interpreting a decline after that is complex. We should think carefully about what that peak represents, and in particular, how much it might be influenced by the COVID-19 pandemic.

After all, the Glasgow climate meeting was supposed to happen in 2020, but was postponed to 2021 due to the pandemic. By the time it rolled around, there might have just been a lot of pent up journalistic energy that was unleashed, leading to the new high. In the U.S., this would have been only amplified by the incoming Biden administration’s heavy focus on climate change and the summit.

Given all of this, how sure are we, based on these data, as to what is happening out there?

Schäfer suggested one would have to design a considerably more complex analysis to address the question more deeply. Such an analysis would do things like take apart the different drivers of these data, such as seasonality: There are clear and recurrent peaks associated with late-year international climate meetings (it will be key to see what happens with coverage of COP 30 in Brazil this November). Then that analysis would come up with the right statistical model for assessing change. This post certainly is not that.

And just to underscore my level of uncertainty: I can easily imagine that in many newsrooms, the general hair-on-fire mentality, supercharged by the Trump administration, may be leading to climate beat reporters getting constantly pulled onto other things. But I can also imagine a situation in which these reporters are still on the beat, but stories are harder and take longer to do, as there’s more need for careful investigation of what the administration is doing, and more pushback when the press tries to take that on.

For now, then, this post is simply about raising a question. I’m looking at the data, wondering what’s going on. And it would be great to hear some feedback from anyone who has any thoughts.