What exactly is "liquefaction settlement"? I know soil can turn liquid-like during an earthquake, but why does the ground sink after the shaking stops?
🎓
Great question! Basically, during strong shaking, water pressure builds up in saturated sandy soil, making it lose strength and behave like a dense fluid. After the shaking, that high water pressure dissipates, and the soil particles settle into a denser, more compact arrangement. That's the permanent settlement you see. In this simulator, the "Liquefiable layer thickness (H)" and "SPT blow count (N)" are key controls for estimating how much that layer will compact.
🙋
Wait, really? So the settlement depends on how strong the earthquake was? How do you put an earthquake into a calculation?
🎓
Exactly. We characterize the shaking with two main parameters you see in the tool: the Peak Ground Acceleration (a_max) and the Moment Magnitude (Mw). `a_max` tells us the peak force, while `Mw` tells us how long the shaking lasts (more cycles). Try increasing `a_max` in the simulator—you'll see the Cyclic Stress Ratio (CSR) jump, making liquefaction and settlement more likely.
🙋
That makes sense. But what's the "safety factor" (FL) mentioned in the tool? Is it like a pass/fail grade for the soil?
🎓
In practice, yes! It's the ratio of the soil's strength (Cyclic Resistance Ratio, CRR) to the earthquake's demand (Cyclic Stress Ratio, CSR). If `FL < 1.0`, liquefaction is expected. For design, codes often require `FL > 1.5`. The simulator calculates this for you. A common case is a loose, saturated sand layer with a low SPT blow count (`N`)—try setting `N` to 5 and see how FL drops below 1.
Physical Model & Key Equations
The core of the Tokimatsu-Seed method is comparing the earthquake's shaking demand to the soil's inherent resistance. The demand is quantified by the Cyclic Stress Ratio (CSR), which scales the peak ground acceleration down to an equivalent cyclic shear stress.
Where:
$\sigma_v$ = total vertical stress at depth $z$ (from $\gamma$ and $\gamma_{sat}$).
$\sigma_v'$ = effective vertical stress (accounts for buoyancy below groundwater depth $z_w$).
$a_{max}/g$ = peak ground acceleration as a fraction of gravity.
$r_d$ = depth reduction factor (shear stress is less at greater depths).
The factor 0.65 converts the peak irregular shaking to an equivalent uniform cyclic stress.
The soil's resistance is the Cyclic Resistance Ratio (CRR), determined empirically from the Standard Penetration Test (SPT). The raw blow count ($N$) is first corrected for overburden pressure to get $N_{1,60}$.
$$N_{1,60}= N \cdot C_N, \quad C_N = \min\!\left(2.0,\, \sqrt{\frac{100}{\sigma_v'[\text{kPa}]}}\right)$$
Where:
$N$ = measured SPT blow count in the field.
$C_N$ = overburden correction factor. Higher confining pressure ($\sigma_v'$) makes soil seem stronger, so we normalize it to a standard pressure of 100 kPa.
The CRR is then found from charts linking $N_{1,60}$ and Fines Content (FC). The Liquefaction Safety Factor is simply $F_L = CRR / CSR$. Settlement is then estimated based on $F_L$ and the thickness of the liquefied layer ($H$).
Frequently Asked Questions
CSR (Cyclic Stress Ratio) represents the magnitude of shear stress acting on the ground during an earthquake, while CRR (Cyclic Resistance Ratio) represents the strength of the ground to resist liquefaction. The FL value is obtained by dividing CSR by CRR. If FL ≤ 1.0, the ground is evaluated as having a high potential for liquefaction.
Input the measured N-value obtained from the standard penetration test. For the groundwater level, enter the depth (in meters) from the ground surface. The shallower the groundwater level, the smaller the effective stress and the larger the CSR, increasing the liquefaction risk. If there are multiple layers, input the representative value for each layer.
The Tokimatsu-Seed method is primarily applicable to ground depths of up to about 20 meters. Beyond that depth, the applicability of the stress reduction coefficient rd decreases, and accuracy cannot be guaranteed. Additionally, as a prerequisite for N-value correction and effective stress calculation, apply this method to ground predominantly composed of sandy soil.
The settlement amount can be directly used in the design of structural differential settlement and ground improvement. For example, it is useful for selecting the bearing layer for pile foundations and for examining the required thickness of surface ground improvement. In the seismic design of buried pipes, it is also used to estimate ground displacement around the pipe based on the volumetric strain of the ground.
Real-World Applications
Seismic Foundation Design: Before building a bridge or high-rise in a seismic zone, engineers use this method to predict if the ground will settle. They might recommend deep foundations (piles) that bypass the liquefiable layer or ground improvement (like compaction) to increase the SPT `N` value.
Lifeline Protection (Water/Gas Pipelines): Buried pipelines can buckle or float if the surrounding soil liquefies and settles unevenly. This calculation helps route pipelines away from high-risk zones or design them with flexible joints to accommodate movement.
Port and Harbor Stability: Reclaimed land for ports is often built with hydraulically placed sand, which can be loose and prone to liquefaction. Assessing settlement is critical for ensuring quay walls don't tilt and cranes remain operational after an earthquake.
Verification of Advanced CAE Simulations: In sophisticated Finite Element Method (FEM) analyses using software like OpenSees, engineers model soil with complex constitutive models (e.g., PDMY). The result from this Tokimatsu-Seed calculator serves as a crucial benchmark to verify the accuracy of those much more computationally expensive simulations.
Common Misconceptions and Points to Note
When starting to use this tool, there are several pitfalls that beginners in particular often fall into. The first is the mindset of "just inputting the N-value is enough". While the N-value is certainly important, this calculation is fundamentally an evaluation "per layer". For instance, in ground where the top 5m is soft sand (N=5) underlain by hard sand (N=20), you shouldn't input a single representative N-value; you need to evaluate by separating the layers. The tool assumes a single homogeneous layer, so for complex ground profiles, you need the approach of calculating for each layer and summing the results.
The second point is the handling of the Fines Content (FC). It's easy to jump to the conclusion that "liquefaction won't occur" if FC exceeds 35%, but in reality, it changes significantly depending on whether the soil is "clayey" or "silty". Even with FC=40%, if the main component is fine silt, liquefaction risk remains. The tool's formula shows a tendency for the factor of safety to increase as FC rises, but this is merely an empirical formula based on "sand". On-site, it is essential to confirm the soil classification through soil tests.
The third point is the interpretation of the calculated "settlement". The output settlement is the subsidence caused by the "compression" of the liquefiable layer itself. However, in actual damage, the loss of bearing capacity due to liquefaction leads to problems like differential settlement of structures or large displacements from lateral spreading. The 30cm settlement calculated by the tool is strictly the "compression amount of the ground itself"; the impact on structures requires separate consideration.