Capturing the frictional heat
The data obtained from the recovered temperature sensors contained detailed recordings of the temperatures for several depth levels around the fault, captured over a period of 9 months. Did they really manage to measure frictional heat? A start was made on analzying the data.
What we want to find out is if there is a difference in temperature between the fault and the surrounding area. As the underground temperature rises with increased depth at a constant rate, it can be calculated by subtraction. When the sensors were first placed, the inside of the borehole got cooler due to the effect of the water used at the time of drilling, and the overall temperature dropped. That was something that had to be taken into consideration too. If the measurements hadn’t been taken over such a long period of 9 months, accurate analysis would undoubtedly not have been possible.
Analysis results showed a temperature anomaly at depths exactly around the fault, which can be attributed to frictional heat. The temperature is 0.31°C higher than the surrounding temperature. Kano tells how frictional heat dissipates over time, and when the form of this temperature distribution is known, the amount of frictional heat can be estimated stretching back to the point of the earthquake. This is why 55 temperature sensors were installed over the various depths.
But time had already passed since the earthquake, and the thickness of the fault, or the duration of the slip cannot be estimated from these observational data. “Still, if we were to estimate," starts Kano, proceeding to explain that on the assumption that the minimum slip duration was 50 seconds and the minimum fault thickness 1 mm, they estimate the maximum temperature increase due to frictional heat at the time of the earthquake to be 1250°C, and 790°C for a thickness of 1 cm.
Observed data
 Estimated temperature rise at the time of main earthquake |
 Subseafloor temperature distribution |
 Observational data (enlarged for fault area) |
 Data interpretation and model calculation |
A slippery fault
When a fault slips rocks rub together, creating friction which in turn generates frictional heat. If there is a high level of friction, there will not be much fault slip. As the stresses are high there will also be a great deal of frictional heat. In contrast, if there is little friction the fault slips easily and there is also little frictional heat. Kano uses a familiar example to explain: “You know how when you rub your hands together, they feel warmer when you rub them together vigorously than when you only rub them lightly? It is the same as that." In other words, if you know the amount of frictional heat, you can find out how much force was applied and how much the fault moved at the time of the earthquake.
The friction coefficient (indicating the level of friction) estimated by Kano is 0.08. The friction for rocks in friction experiments is in general around 0.5-0.7, so the value that was found is very low in comparison. The amount of energy expended at the fault at the time of the earthquake was also extremely low at 27 ML/m2 (or in calories, 7665.3 kcal/m2), and the fault was found to have moved without the application of a great deal of force. In other words, friction levels for this fault are low making the situation extremely slippery.
In the meantime, analysis and verification of the geological samples that were successfully retrieved from the fault in the JFAST expedition continued. This found that strength is low for this fault, and that it contains large amounts of a clay called smectite which hardly lets any water through. The fact that it is almost impermeable means that water is contained inside the fault making it easier to slip.
From experiments using the geological samples it is estimated that this fault became even more slippery at the time of the earthquake. The water in the fault expanded due to the frictional heat generated when the fault slipped, and was trapped inside instantly increasing the pressure. It is thought that the force worked in the opposite direction to the restraining force, thus momentarily reducing the friction. It is just like the lid of a kettle lifting under the pressure of the steam produced when the water boils.
Says Kano confidently, “Because we got the same results using different approaches, I think these are more reliable results."
Boarding crew
