African Lake Study Leaves Some Questions

The headline from the University of Arizona News, and many other news outlets said, “Twentieth-Century Warming in Lake Tanganyika is Unprecedented.” The headline from Brown University press release (home of the lead author) said, “Brown Geologists Show Unprecedented Warming in Lake Tanganyika.”

Well, not exactly. The title of the study referred to is “Late-twentieth-century warming in Lake Tanganyika unprecedented since AD 500,” published in Nature Geoscience (16 May 2010). Even that more modest claim doesn’t tell the whole story.

First some background. Lake Tanganyika occurs within the East African Rift, which is a divergent tectonic plate boundary that is gradually separating East African countries from the main continent. The rift contains both active and dormant volcanoes. The lake is 418 miles long and 45 miles wide. Its average depth is 1,870 feet with a maximum depth of 4,820 feet. Portions of the lake are claimed by Burundi, Democratic Republic of the Congo, Tanzania, and Zambia. Fishing the lake provides a major food source for people in the surrounding lands. There is concern that lake warming will disrupt the fish supply.

The abstract of the paper concludes, “Our records indicate that changes in the temperature of Lake Tanganyika in the past few decades exceed previous natural variability. We conclude that these unprecedented temperatures and a corresponding decrease in productivity can be attributed to anthropogenic global warming, with potentially important implications for the Lake Tanganyika fishery.”

The questions I had upon reading this were: 1) Are the temperatures really unprecedented? 2) Do they exceed natural variability? 3) What is the evidence that the warming was caused by anthropogenic global warming? 4) Could there be some other cause of fish decline?

The researchers studied lake sediment cores going back 60,000 years and by using proxies deduced a temperature record for the lake surface temperature. In the current study, the researchers said that during the last 1,500 years, temperature varied between 22.5º C and 25.7º C, and that in the last 50 years the temperature rose by 1.6º C.

However, in 2008, these same researchers published a paper in Science (Vol. 322. no. 5899, pp. 252 – 255) which said the lake surface temperature fluctuated between 27° and 29°C over the last 60,000 years according to their interpretation of lake sediment cores.

I emailed a co-author of the paper, a UofA professor, asking for an explanation of this apparent discrepancy. He replied by referring me to the website of the lead author at Brown University. There, she explained that there was a problem in calibration of the temperature proxies. She presents a graph showing the records after recalibration. It is reproduced below. It should be noted that there are two separate core sample locations. The more recent core was taken closer to shore than the older, longer record. The more recent record initially shows cooler temperatures where the two records overlap. The researchers attribute this discrepancy to upwelling cold water from deeper in the lake. So which record is closer to the real surface temperature?


According to the lead author’s own data as shown on the graph, it is obvious that the current temperatures are not unprecedented, nor do they exceed natural variability. The title of their paper is technically correct only if one accepts cherry-picking start dates.

That leaves the question about the cause of the warming. The UofA scientist replied to my email, “our record only demonstrates a lake surface temperature history, not the cause of that history.” The allegation of an anthropogenic cause, a major conclusion of the paper, was made without any supporting evidence, just speculation.

I am wondering why the paper abstract contains the conclusions it does. Is it time for some scary scenarios to promote more study and more funding?

This whole study purports to be about lake surface temperatures, but it contains very few such measurements from the lake surface. From my reading, the researchers deduce surface temperatures from only two core sample locations. As the NOAA satellite graphic below shows, on any given day, at any given time, the variation in lake surface temperature can be as much as 4º C in different parts of the lake, and that equals or exceeds the entire range of temperatures found in the studies. It would seem, therefore, that any temperature record derived from sediment cores could vary greatly depending on location. Since this study had just two sample locations, it makes one wonder if it gives a true representation of actual conditions.


And about the fish. The current paper says that warming is causing a decline in fish abundance. Yet an earlier study, of which the UofA scientist was a co-author, says the fish decline is caused by land disturbance. “Watershed deforestation, road building, and other anthropogenic activities result in sediment inundation of lacustrine habitats.” “Our faunal analyses suggest that all three taxonomic groups are negatively affected by sediment inundation but may have varying response thresholds to disturbance.” (Citation: Conservation Biology, vol. 13, no. 5, Oct. 1999).