How to improve climate predictions? McGill researchers turn to 19th century letters

A team led by McGill University researchers came up with a method they hope could improve climate models over Africa by combining them with 19th century missionary records, refashioning dubious documents in a bid to better inform projections of global warming’s impact.

Models are an important way for scientists and decision-makers to understand how human influence is changing the climate. To come up with those projections, climate models depend on historical baselines – temperature and precipitation, for example – to validate and refine their results.

But a lack of historical region-specific data across parts of Africa, plus a major deficit in weather stations compared to North America and Europe, has contributed to model uncertainty.

“We know very little about how global warming will affect regional patterns of rainfall in place in many regions of the Global South because the data that we’re basing the models on is very historically shallow,” said Philip Gooding, the lead author and project administrator with McGill Indian Ocean World Centre.

“They come from the last 20, 30, 40, 50 years. Something like that. Here we’ve got data in the form of missionary documents that go back a much longer period and these can be integrated. Into models to create a stronger, more rounded baseline on which you can project how global warming will affect the region’s climate moving forward.”

A study from last month’s peer-review journal, Climate of the Past, says, “Africa’s absence from the underlying data makes deploying these projections uncomfortable, as it partly represents the continent’s marginalization from global scientific discourse during and after colonialism.”

To fill that data gap, lead Gooding turned to a perhaps unlikely source – writings of 19th century Christian missionaries in Tanzania.

Those are problematic records, Gooding said, derived from people who contributed to Africa’s colonization and had their own reasons to misrepresent climate conditions.

But perhaps by analyzing the climate models and those missionary records together, both could be improved upon, he said.

“The hope moving forward (is) basically, we’ve made a better baseline on which global warming scenarios for Tanzania can now be projected,” said Gooding, a project administrator at McGill’s Indian Ocean World Centre.

Climate model uncertainty is one example of how decision-makers in Africa are left to predict and prepare for extreme weather with fewer resources than their counterparts in other parts of the world.

Researchers have been sounding the alarm over how a lack of weather radar stations, which also inform climate models, have left areas of the continent without lifesaving early-warning systems for extreme weather.

Europe and the United States have about 636 radar stations for a population of 1.1 billion people. Africa – with a comparable population and a greater land mass – has about 37 stations, says a 2023 article in the scientific journal Nature, outlining the continent’s undue exposure to climate risks.

“This is very much a small step in the, in the direction that could put, that could be a, a trigger for more research along these lines, but we definitely think to take this even further, particularly in Tanzania itself, the next stage of the research needs to be conducted in the region, with people on the ground,” said Gooding.

East Africa is one of the most vulnerable regions to climate change and yet one of the least understood, said Obed Ogega, a climate scientist and program manager with the African Academy of Sciences based in Nairobi.

To plan climate-resilient communities, whether that’s around drought-resistant agriculture or flood-resilient roads, “we have to be informed by climate information.”

“The more we understand our weather and climate systems, the better for the region, not just for the sake of building resilience of communities, but also for systems and ensuring that we minimize the vulnerability that is currently being experienced in the region,” said Ogega, who is also an adjunct professor at Halifax’s Mount Saint Vincent University.

Gooding, a historian, realized his doctoral research on Lake Tanganyika, the world’s second deepest freshwater lake that straddles Tanzania’s western border, may be helpful. His research included hundreds of written accounts from missionaries and imperialists who described conditions over a roughly 30-year period in Tanzania ending in the 1890s.

In those accounts, Gooding recognized region-specific data that’s so often missing from climate modelling of Africa.

Yet, relying on missionary accounts came with some obvious issues and limitations, the study notes.

“Nineteenth century meteorology, like cartography, was part of a wider practice that sought to impose European science, and thus Europeans’ ideas of ‘civilization’ on equatorial eastern Africa(ns), erasing Indigenous patterns of human–environment interaction and their understandings of climate and weather,” the study says.

In some cases, missionary reports may have also been exaggerated to provoke emotional responses from readers back in Europe who funded the missions. In others, reports may have minimized harsh conditions in a bid to establish the mission’s feasibility.

“In short, although highly valuable, the documents are highly subjective, and they both affected and were influenced by imperial knowledge-making,” and contributed to the further colonization of Africa in the 19th century, the study said.

Working with an international team of climate scientists and data analysts, Gooding standardized the documentary data into a seven-point scale so it could be combined with the climate models. Accounts of regular climate conditions were in the middle of the scale, with severe drought and severe rainfall at either end.

While there’s no simple way to measure whether climate models are more accurate when 19th century missionary accounts are included, “it can be safely assumed that by adding data from the region from the deeper past, it gives a more rounded picture of what we can think happened,” Gooding said.

Ogega, the Nairobi-based climate scientist, commended the researchers’ efforts, but was skeptical of the missionary records and the limited time series.

“Until I’m certain that the source I’m using is credible enough, is fit for purpose, I’ll be very hesitant to use it,” he said.

“If you make a conclusion based on wrong data, you can easily have a very serious policy implication.”

He preferred to focus on nature-based records of Tanzania’s historical climate, such as studies of lake sediment or tree rings. But he suggested the research was a good start and could inspire further study to bolster confidence in the missionary record.

Open Gallery 4 items

“Maybe this a resource that we haven’t thought about enough, but apart from that I wouldn’t say, as of now, it’s conclusive enough for me to say, ‘now, going forward, we look for missionary reports and see how we can put those into climate modelling,'” he said.

Gooding said he shared some of Ogega’s skepticism of the missionary records. The study included sorting through records and comparing them against climate models to determine which accounts appeared more or less confident.

“(They) are both inherently flawed, but I think reading them together enhances both their value,” he said.

The study suggests further challenges could include integrating evidence from local oral traditions into historical climate reconstructions.