Here I list unsolved problems in earthquake science that I am very interested in. If you know the answers, please let me know. When these problems are solved in the future, I will note "solved".

1. Can we distinguish between slow slip events preceding large earthquakes and the other slow slip events?

In recent years, slow slip events preceding megathrust earthquakes have been observed in subduction zones such as Tohoku, Chile, and Mexico . On the other hand, many slow slip events occur without large earthquakes. Are there any differences among these slow slip events? This is one of the most important problems for studies on earthquake predictability. Moreover, this problem is almost equivalent to the classical unsolved problem whether there is a difference between earthquake swarms and foreshocks.

Note: Nishikawa & Ide (2018, JGR) claimed that slow slip events preceding megathrust earthquakes can be differentiated from other slow slip events at the Ibaraki-Oki region in the Japan Trench.

2. Slow slip events preceding large earthquakes are earthquake nucleation processes?

As mentioned above, slow slip events are sometimes observed just before megathrust earthquakes, but what is the physical mechanism of this slow slip events? Is it an earthquake nucleation process in a large locked region, as observed in experiments and numerical simulations? Or is it just a slow slip event in the vicinity of a large locked area? If it is an earthquake nucleation process, the behavior of the slow slip event should strongly reflect the condition of the locked zone. That is, there is a possibility that the behavior of the slow slip event changes as unlocking of the locked region undergoes.

3. Are there physical and/or kinematic differences between deep slow slip, shallow slow slip, and seismogenic zone slow slip?

Slow slip process occurs under various temperature-pressure conditions. In the Nankai Trough, it occurs on the plate interface deeper and shallower than the seismogenic zone. In the Sagami Trough and the Japan Trench, it occurs in the seismogenic zone. We cannot find anything common in the temperature-pressure conditions of these locations. If these are the same physical phenomena, it means that the physics of slow slip process does not depend on temperature-pressure conditions. On the other hand, there is a possibility that different physical phenomena are observed as kinematically identical phenomena. Or it may also be kinematically different. It seems to me that earthquake nucleation process and deep slow slip events are obviously different physical phenomena. Is it true? Therefore, phenomena called "slow slip" may be mixing-up of multiple physical phenomena.

4. What are the geological entities of deep and shallow tectonic tremors?

Deep and shallow tectonic tremors are relatively high-speed slip among slow earthquakes, but why does not the slip accelerate to slip velocity of ordinary earthquakes (i.e., 1 m/sec)? In other words, what are geological differences between earthquake faults and tremor faults? It seems to me that there are also geological differences between deep-tremor faults and shallow-tremor faults. What are they?

Note: In 2018, seismogenic-zone tremors were discovered in the Japan Trench.

5. What percentage of large earthquakes occur via earthquake nucleation process? What percentage of large earthquakes occur via cascading process?

There are mainly two mechanisms of large earthquake occurrence. One is earthquake nucleation and the other is cascading process. The former can be a short-term precursor of a large earthquake. In contrast, in the latter case, earthquake prediction seems impossible. A rupture grows larger and larger like domino toppling, and the ultimate size of the rupture is unpredictable. What percentage of large earthquakes occurs via these mechanisms? If earthquake nucleation is extremely rare, the short-term prediction of large earthquakes is hopeless.

6. Are there slow slip events in "any" subduction zone?

In the past, slow slip was thought to occur only in subduction zones with high temperature (e.g., the Nanakai Trough and the Cascadia Trench). However, in recent years, this hypothesis has been rejected because of the discovery of slow slip in the Japan Trench, Ryukyu Trench, and Hikurangi Trench. So, is there slow slip process in any subduction zone? For example, are there also slow slip events in the Izu-Mariana Trench, the Kermadec Trench, and the Sumatra Trench? This is an important issue for elucidating the physical mechanism of slow slip.

Note: In 2018, a slow slip event was observed in the Izu-Mariana Trench. This problem was partially solved.

7. What tectonic properties cause variations in slow earthquake activity in subduction zones worldwide?

Slow earthquakes have been observed in many subduction zones, and they are not rare phenomena. However, their behaviours have large variations among subduction zones worldwide. For example, in the Nankai Trough, slow earthquakes occur on the plate interface shallower and deeper than the seismogenic zone. On the other hand, in the Sagami Trough and the Japan Trench, it occurs inside the seismogenic zone. Also, it seems that there are some differences between episodic tremor and slip (ETS) in Nankai and Cascadia. What is the mechanism that produces these differences? Is it possible to explain these variations in terms of tectonic properties in subduction zones (e.g., plate velocity, age, and curvature)?

8. Is velocity of stable aseismic slip on the plate interface fluctuating (slow slip ubiquity hypothesis)?

Based on analyses of small repeating earthquakes, it has been suggested that the velocity of stable aseismic slip is fluctuating periodically or aperiodically at each point on the plate interface in the Japan Trench. If we define acceleration of aseismic slip on the plate interface as "slow slip", we can call this idea "slow slip ubiquity hypothesis". Is this true? As in the Japan Trench, stable aseismic slip on the plate interface is also fluctuating in other subduction zones?

9. Horst-Graben structure has two roles, structural heterogeneity and facilitation of hydration of subducting plates, but which of them more strongly influences seismicity in subduction zones?

10. Differences between large inland earthquakes and subduction zone megathrust earthquakes and their causes (e.g., critical earthquake nucleation size)

11. Why do many of transform faults creep?

12. Why are many of slow slip events "characteristic"?

13. Does critical earthquake nucleation seize scale to seismic moment of earthquakes?