Resettlement due to climate change, past and future

The Gilbert Islands [Kiribati] are, what you call, now, dangerous, high tides and flood. It’s alright in the Solomon Islands, there are mountains.

As the sea-level rises, the fate of people in low-lying Pacific island nations like Kiribati, Tuvalu, and the Marshall Islands is uncertain. The growing cost and challenge of adaptation could eventually drive large-scale resettlement to countries safer from the rising sea like New Zealand and Australia.

How will people be received? What challenges will they face adjusting to a new environment? Will they be able to maintain their culture? There are no easy answers to these questions, but the past may be a guide.


Ghizo Town, Solomon Islands

The above quote is from an elderly man who was part of a mid-20th century British colonial program that relocated roughly 2,300 people from the Gilbert and Phoenix Islands to hilly Ghizo in the distant Solomon Islands because of fears of resource constraints and overcrowding. I interviewed him in 2011 during a visit to Ghizo. I wanted to learn about the Gilbertese* resettlement experience and see if there were any lessons for future climate change-driven resettlement of people from Kiribati and nearby atoll nations.

It was a fascinating and moving experience. Just a few years before, a tsunami had struck tiny Ghizo, taking a large physical and emotional toll on the small Gilbertese community. Another elder described the events to me:

When it [the lagoon] was drying up, they were surprised, they went to go see the sea, rather than running up the hill for their lives. They kept being surprised, then it was too late.

Of the fifty people killed, thirty were Gilbertese, including many children. Despite three generations in the earthquake- and tsunami-prone Solomon Islands, many people in the community did not know the signs of an impending tsunami (see the terrific 2009 paper by Brian McAdoo and colleagues). This gap in cultural knowledge and the challenges of rebuilding after the tsunami revealed some of the many unforeseen long-term consequences of resettling in a new environment and society.

The history of the Gilbertese resettlement and the lessons from my interviews are described in a new paper in the journal Natural Resources Forum. The paper reveals how uncertainty about land rights still persists, 60 years after the original resettlement, and is linked to a number of other present-day challenges the community faces.

The findings may help inform thinking about resettlement due to climate change – including the importance of establishing and funding permanent mechanisms for dealing with land and resource disputes. I also hope that describing the Gilbertese experience can, in general, help disavow the mistaken popular idea that there is “empty” land in other countries where the people of Kiribati and other low-lying islands can easily resettle.

A personal note: This research was made possible by many people who were willing to share often difficult thoughts and memories, but who must remain nameless due to research ethics. No thank you can be adequate. All I can say is that, for a decade now, I have traveled to the Pacific Islands and back to learn about climate change and people’s capacity to adapt. Each trip, I come home changed by the incredible generosity, openness, and overall decency of the people I meet.


Donner, S.D., 2015. The legacy of migration in response to climate stress: Learning from the Gilbertese resettlement in the Solomon Islands. Natural Resources Forum, 39: 191-201.

McAdoo, B.G., Moore, A., Baumwoll, J., 2009. Indigenous knowledge and the near field population response during the 2007 Solomon Islands tsunami. Natural Hazards 48: 73-82.

* The colonial term “Gilbertese” is used because the Solomon Islands community, established before Kiribati’s independence, self-refer as Gilbertese rather than as i-Kiribati. The word Kiribati itself is derived from the pronunciation of Gilberts (kee-ree-bas) in the local language.

Life as a climate change poster child: the new Scientific American article about Kiribati

???????????????????????????????I have a feature (“Fantasy Island”) in the latest issue of Scientific American and accompanying online slideshow about the reality of sea-level rise in Kiribati.

The article summarizes the complicated science of sea-level rise in coral islands and the even more complicated politics of being a poster child for the impacts of climate change on the developing world. In reflecting on years of research on the ground and on (and in) the water, I try to provide an antidote to all those well-meaning but generally inaccurate pieces of popular disaster porn written about remote island nations like Kiribati and Tuvalu.

If you are interested in what actually is happening in places like Kiribati, I encourage you to buy the issue. An excerpt:

A North American or European traveling to Kiribati may as
well be stepping through a wormhole into another universe. Combine
that naïveté with the reserved nature of the Kiribati people,
the custom of deferring to outsiders, the legacy of countless
past i-Matang asking about climate change and the lack of local scientific
capacity to verify claims, and a naturally flooding village
becomes a victim. Add in the geopolitics—the legitimate need for
a tiny country lacking agency on the world stage to raise awareness
of a threat to its existence—and the exaggeration about the
impacts of sea-level rise can look intentional, whether it is or not.
As my friend Claire Anterea of the Kiribati Climate Action Network
says, “This is not a story that you will just journalize in one
week or two weeks.”

The article is  a testament to all the wonderful people in Kiribati that I have interviewed and worked with over the years, as well as to Mark Fischetti and the editors at Scientific American, who were willing to embrace a story about the incredibly important but less glamourous nuances of climate change.

Message from communications research: Climate change is real. Repeat. Repeat again.

There is an ongoing feud about value in communicating the scientific consensus on climate change to the public. One side argues that we need to talk about the consensus in order to raise public awareness about climate change. A new review article by John Cook and Peter Jacobs (also described in the Guardian) reviews the evidence for “consensus messaging”.  The counterargument, proposed by Dan Kahan and others, is that talking about consensus will increase political polarization about climate change.

A recent paper by Kahan called “Climate Science Communication and the Measurement Problem” suggests the disagreement among communications researchers is related to the “contamination of education and politics with forms of cultural status competition”. It is a fascinating paper with a lot of important findings. But I wonder if, deep in the data, there may be evidence that the drumbeat of climate change news and outreach campaigns has actually been effective.

The core result of the paper is described by the following figure:

Kahan 2014 - Fig 7 - no legendIf people were rationally assessing scientific information, higher science comprehension would translate into higher perceived risk from climate change (left panel). Kahan’s experiments find the opposite for people on the right of the political spectrum (red, right panel). That’s been the headline: for conservatives, better knowledge of climate science might mean less concern about climate change.

In other words, when people go beyond the basics, opinions become polarized. That is not very surprising, given that someone on the right of the political spectrum with greater interest and/or ability in science who looks for information about climate change may head to right-wing media and blogs, which often house an alternate universe of “facts” about climate change.

What is more surprising are the results for people with low “science comprehension”.

Why are people with low science comprehension on both the left and the right of the political spectrum perceiving moderate to high risk from climate change? If people’s views on climate change are defined more by their cultural identity than by the facts of the case, why would people on the right of the spectrum with low science comprehension have even moderate concern about climate change?

This opinion about the risk from climate change must derive from something. It isn’t a detailed knowledge of the science, or the problem. Otherwise, the people would fall elsewhere on the graph. It also isn’t their community. Their community, if defined correctly, generally believes the risk from climate change to be low.

What’s happening? Perhaps there has been enough mention of climate change in the public domain, whether on the news, in private conversations, etc., that even those who pay scant little attention to science have been able to develop some level of concern about climate change.

It may be that, current polarization aside, the much-maligned information deficit model has actually worked, at least with very basic information, and in the way political messaging works. Years of repeating the general facts of the case – climate change is real, caused by humans, and poses a risk to the future – appears to have created a basic public consciousness about climate change.

Finding the balance between science and values in public engagement about climate change

Researchers in the public eye have always struggled to find a balance between representing their results and being effective communicators. This was the subject of my recent essay in Climatic Change.875735-130928-kudelka

One solution, proposed by Roger Pielke Jr., is to be an “Honest Broker of Policy Alternatives” who provides all the available information and options. Thus, rather than pre-selecting policy options to present, which inevitably involves some value judgements, the honest broker presents all the options and allows the decision-makers to reduce the scope of options.

It is a very reasonable solution to avoid having the researcher’s personal values influence their public statements.  From my article:

Failure to consider the possible influence of normative judgements on one’s thinking can lead to “covert” or “stealth” issue advocacy, in which scientists do not perceive the implied advocacy position in their research or public statements (Lackey 2007; Pielke 2007). This can undermine one’s objective of informing the public and policymakers about science and policy options.

Yet many science communicators question whether it is possible to be an honest broker, since the process of filtering and synthesizing research for public consumption involves value judgements. This is especially the case if the data or knowledge is incomplete; the researcher must make a judgement on what information to include.

A recent Twitter exchange with Pielke about his USA Today op-ed piece about hurricane trends and storm protection demonstrated just how hard it can be to control for all personal judgements*.

My suggestion the op-ed was misleading led to some back and forth about the research methods and wound to a close with an honest, funny tweet from Pielke: “Guilty — Most aspects of hurricane mitigation/preparation did not appear in my 700-word op-ed”

The tweet rightly pointed out the absurdity of my complaining about what is and is not included in a short op-ed. How could anyone fully explain the difficulty of estimating for the influence of storm preparation on the costs of storm damage, let alone all of the other research studies on the subject, in 700 words? Especially if you want those 700 words to be interesting enough to be accepted by a major newspaper?

Inclimate-change-science-v-politics-cartoon presenting the findings to the public, subjective judgements had to made about which findings warrant mention. Yet this is what someone playing the honest broker would seek to avoid. A partial solution to this dilemma, suggested years ago by Stephen Schneider and discussed in my article, is to make backup materials available. Today, a blog is an easy, accessible way to present the detailed evidence behind a position taken in a shorter public statement. Many of us, including Pielke, maintain blogs for that reason.

The problem is that we are still making choices in preparing that backup material. And, regardless, few people will ever read the material. We’re lucky if they read the whole op-ed, rather than just the headline! One outcome of this is that even if a researcher perceives herself or himself to be close to the honest broker ideal, others may not.

It is for these reasons that I encourage students to understand their motivations and personal biases, rather than strive to be objective automatons. We can’t remove all personal judgements. But we can strive to be honest with ourselves about our values and motivations and take them into account when engaging with the public. From my essay:

Finding a comfortable and effective position on the science-advocacy continuum requires analyzing ourselves with the same rigour we would use to analyze our data.We are scientists, but we are also citizens, voters, taxpayers, parents, children and homeowners. Research on science communication suggests that we must consider our knowledge, our motivation, our cultural values and our ability to reach different audiences in order to be effective at public engagement… This self-analysis is critical to making personal biases explicit and separate from scientific findings, to recognizing the limits of your communication abilities, and to protecting the hard-won public trust in science.

Global warming or Climate Change? It depends when and where

For years, people have argued over the best term to describe what is happening to the climate: global warming or climate change.

Market-savvy environmental groups have tried re-branding the issue entirely, producing terms like global heating, global weirding, global melting, climate crisis, climate disruption, climate chaos, springtime for CO2, climate farklempt, I’ve got some coastline in Florida to sell, etc. etc. The two originals, global warming and climate change, continue to corner the market like the Coke and Pepsi of the climate world.

A recent study by the Yale Project on Climate Change Communication found that global warming is more “effective” from a communications standpoint than climate change. Detailed public opinion analysis showed that “global warming” instills more concern about the issue and generates more motivation to take action.

The study also points out that the American public has for years used global warming far more than climate change, at least according to Google search data. However, if you dig further into the search data, you find some changing patterns.

Here is the U.S. search volume going back to 2004:

US search volume2Global warming certainly dominated the “aughts”. Since then, its relative use has declined. If we look at the ratio of searches for global warming and climate change in the above data, you see that word preferences have been changing:

US vs. Canada ratio in searchesIn the US (blue), the ratio has declined towards the 1:1 line (black), meaning that today there is close to an equal proportion of searches for global warming and climate change. That’s a big change since the mid-2000s, when global warming vs. climate change was like the Harlem Globetrotters vs. the Washington Generals.

In Canada, global warming was never nearly as dominant term north of the border. Climate change is now searched more often (52% of combined searches in 2014). This is no surprise to Canadians; the term climate change was adopted here years ago by most political parties and the media. That decision probably reflects the more moderate Canadian political climate, where people are more likely to trust authority, in this case, scientists and their preferred term. Granted, it also may reflect the actual climate; the term “global warming” may be scary if you live in Phoenix, but dreamy if you live in Winnipeg.

Narrowing the U.S. search data by region shows a huge geographic divide in the preferred term and the interest in the subject. For example, the next graph compares the search ratio for Texas and for the Bay Area. I tried using individual metropolitan areas in Texas and other conservative states, like Dallas-Ft. Worth or Houston, but none featured consistently high enough search volume for the past ten years. As it is, the Texas data had large gaps before 2007 (and one in 2008), when search volume was too low for the Google metric.

Texas vs. Bay Area ratio in searchesThe ratio is dropping in both Texas and the Bay Area. Texans, however, still prefer global warming by 2:1 or more over climate change. People in the Bay Area, however, are almost evenly split (52% global warming in 2014).

If you examine the overall data state by state, a clear pattern emerges. Republican-leaning states in the southeast and central U.S. have the highest global warming to climate change search ratio, whereas Democratic-leaning states have the lowest ratio. Texas is actually not even among the top ten in global warming preference. tableWashington DC is the only “state” where climate change is the preferred term, no doubt related at least in part to its use in government.

Harry Enten at fivethirtyeight came to a somewhat similar conclusion using data from U.S. Congress and cable news channels. Enten found that democrats prefer “climate change” and Republicans prefer “global warming”. He then argues that the Yale research would recommend the exact opposite:

If the polling is to be believed, Democrats, Republicans and the news channels they watch are actually having the opposite effect they are intending. We’ll have to see whether the Yale study makes them reverse course.

There are many unresolved questions, including whether the influence noted in the Yale Study is changing over time, and what that means for language choices. Does the choice of wording matter among people who are already motivated, like many in the Bay Area, or those whose attitudes are unlikely to change?  And would re-labeling by either party be so obvious to the public as to undermine the objective?

Regardless of the answers, one thing is clear from the search data. We are not paying nearly enough attention to ocean acidification:acidification search