What a crater lake can resolve that a seafloor core cannot
The argument over Toba has always been an argument about resolution. Ocean and lake muds mix as they settle, so even a thinly sliced core smears several years together, and a volcanic winter lasting one to three years disappears into that smear. Lake Chala, a steep-walled crater lake on the flank of Kilimanjaro, does not mix at depth: its bottom water stays anoxic and still, and the sediment arrives as annual couplets, a pale diatom layer from the windy mixing season and a dark clay layer from the wet months. Jinheum Park and colleagues at Johannes Gutenberg University read the 450 years of mud bracketing the Toba ash, sampling every two to three years, with the ash itself showing up in X-ray scans as a sliver 0.3 millimetres thick.[1]
What the layers say is undramatic. For the 260 years before the ash the lake is calm and the region warm and wet. Then come two green films a few hundredths of a millimetre thick, read as a stress response by diatoms under a dimmed sky, and the following dry season leaves a pale layer 1.2 millimetres thick, a strong diatom bloom fed by deeper and longer mixing in a chilled lake. The dark layer above it is thin, meaning weak rains. By the third year the couplets look ordinary again, which puts the whole disturbance at roughly 18 months. To turn that into a temperature the team compared silicon-to-aluminium and manganese-to-iron ratios against the amplitude of the last ice age and found about a quarter of the threshold beyond which Chala stops keeping annual layers at all: approximately half a degree Celsius.[1]
Why the largest eruption did not produce the largest cooling
The mechanism Park offers for that smallness is physical, and it has a ceiling built into it. Sulfur dioxide reaching the stratosphere becomes a haze of droplets that reflect sunlight, and more sulfur means more droplets, up to a point. Past that point the droplets grow larger, and larger droplets are heavier: they settle out faster and scatter incoming light less well. So the forcing saturates while the magma volume keeps climbing, and the model runs that made Toba look like anything from a near-extinction event to a mild nuisance were sweeping the uncertainty in a sulfur estimate more than a climate response. The alternative reading deserves stating: eastern Africa may simply have sat in a favourable spot under a veil that drifted north, in which case the ceiling would be a regional artefact more than a general property of large eruptions.[1]
The position of the layer adds a second result that cuts both ways. Because pale layers form between June and September and dark ones from October through April, where the ash sits inside a couplet dates the eruption: near the end of a dark lamina, so January or February, Northern Hemisphere winter, the opposite season to what ash distribution across Asia had suggested. Sulfate travels more efficiently into the winter hemisphere, so the veil drifted north and thinned over the people living in eastern Africa. Most of the land is also in the north, and land cools faster and further than ocean, so the same timing points to a somewhat larger global cooling than a July eruption would have caused. A regional half degree does not earn a global half degree.[1]
One lake, and what would test it
Park is direct about the limit: Chala measures eastern Africa, not the planet, and one archive cannot carry a global claim. The background matters as much as the number. Across the same 450-year window the region was already sliding from warm and wet toward cool and dry, tracking northern high-latitude cooling seen in Greenland ice, so the eruption's mark sits inside the range of variation the last glacial cycles produced anyway. That is the honest form of the finding: it bounds what this eruption did in one region without clearing super-eruptions in general, and the paper offers the result as better parameterization for earth-system models rather than a closed question. The test is repetition, the same sub-annual method at other sites carrying Toba ash, at Los Chocoyos in Guatemala and the Oruanui eruption in New Zealand. If those sites are read the same way, the cooling they show should again come out closer to half a degree than to the several degrees older model spreads allowed, and shorter than two years; a second site returning two or three degrees would break the droplet-ceiling argument first.[1]