
model compressional and shear waves, use S-P arrival times to locate an earthquake, and explain how seismic evidence reveals a layered interior that cycles matter by convection.
Instruction
Seeing inside a planet. The deepest hole ever drilled reached about 12 km, less than 0.2% of the way to Earth’s center. Everything we know about the deep interior comes from indirect evidence, above all from seismic waves: the vibrations that earthquakes send through the planet.
Two kinds of body waves. P waves (primary) are compressional: rock is squeezed and stretched in the direction the wave travels, like a push along a spring toy. S waves (secondary) are shear waves: rock moves side to side, at right angles to the direction of travel, like a flick of the spring. In the crust, P waves travel at about 6 km/s and S waves at about 3.5 km/s. Liquids cannot be sheared, so S waves cannot travel through liquid. That single fact is one of the most powerful clues we have.
Worked example: how far away was the earthquake? Both waves leave the focus at the same moment, but the P wave arrives first. The gap between arrivals grows with distance. For distance d, the lag is d/3.5 − d/6.0, which works out to d = 8.4 × (S − P lag in seconds) km. A seismometer that records a 12-second lag is about 8.4 × 12 ≈ 101 km from the earthquake. One station gives only a distance, a circle on the map. Three stations give three circles, and the earthquake is where they cross. This is triangulation, and you will do it in the lab. When the magnitude 4.0 earthquake struck near Cooter, Missouri, on April 23, 2026, the seismometers of the New Madrid network recorded it at dozens of stations; 89 of them were used to locate it. The nearest stations, about 7 to 10 km away, recorded S-P lags of only a little over 2 seconds, and a station 71 km away in Kentucky recorded about 9 seconds. Your lesson reading lists the real arrival times so you can test the rule.
What the waves reveal. Seismometers around the world show that waves speed up with depth and bend (refract) as they cross layers, and that some boundaries reflect them. More than 104° of arc away from a large earthquake, S waves vanish: there is an S-wave shadow zone on the far side of the planet. In 1914 Beno Gutenberg used such evidence to place the core-mantle boundary at about 2,900 km depth, and the missing S waves show that the outer core is liquid. In 1936 Inge Lehmann found faint P waves inside the shadow that could be explained only by a solid inner core. The layers, from outside in: crust (5-70 km thick), mantle (to about 2,900 km), liquid outer core (to about 5,150 km) and solid inner core (to the center, 6,371 km).
Density evidence. Earth’s average density, calculated from its mass and volume, is 5.5 g/cm3. Surface rocks such as granite are only about 2.7 g/cm3. The interior must therefore hold much denser material: an iron-nickel core, around 10-13 g/cm3, consistent with iron meteorites.
A solid mantle that flows. The mantle lets S waves through, so it is solid. But over millions of years hot rock creeps like very stiff putty. Heat from the core and from radioactive decay warms rock at the bottom of the mantle; warm rock expands, becomes less dense and rises; cold ocean plates, denser than the mantle around them, sink at subduction zones. This thermal convection cycles matter between the surface and the deep Earth. Seismic images (tomography) show cold, fast-wave slabs of old sea floor sinking deep into the mantle. The plates are the cold top of the convecting system, and the pull of sinking slabs is thought to be the largest force moving them.
📖 Read for this lesson
The reading opens over the lesson; close it to come back to exactly where you were. The checks below draw on it.
Lab: Spring-Toy Seismic Waves and Locating an Earthquake
Question. How do compressional and shear waves differ in how they move and how fast they travel, and how can the difference in their arrival times locate an earthquake?
Materials
- A metal or plastic spring toy (such as a Slinky)
- A sturdy chair or table leg to anchor one end, and a string or rubber band to tie it
- Phone with a stopwatch or slow-motion video (or any stopwatch)
- Meter stick or tape measure and masking tape
- A small piece of ribbon or tape to mark one coil
- The triangulation grid from this lesson’s reading (printed or copied onto paper), a drawing compass and a ruler
- Calculator
- Wear safety glasses or sunglasses: a stretched spring can snap back toward your face if it slips from its anchor or your hand.
- Stretch the spring no more than 3 m along the floor, check that the anchored end is tied firmly before stretching, never let go of a stretched spring, and keep walkways clear so no one in your home trips.
Procedure
- Tie one end of the spring firmly to a chair or table leg. Stretch it to 3.0 m along the floor, measured with tape marks, and tape the ribbon to a coil near the middle.
- Make a P wave: push your end sharply forward and back once, along the spring’s length. Watch the ribbon and record the direction it moves compared with the direction the pulse travels.
- Make an S wave: flick your end sharply sideways once, along the floor. Record how the ribbon moves.
- Time each kind of pulse traveling the 3.0 m length to the anchored end (start when you send it, stop when it reaches the chair leg; a slow-motion phone video propped to show the whole spring gives the most accurate times). Take three trials each and compute the average speed = 3.0 m / average time.
- Triangulate: use the S-P lags given for three stations in the data table, compute each distance with d = 8.4 × lag, draw a circle of that radius around each station on the reading’s grid, and mark where the three circles meet. Then test the rule against the real Cooter arrival times in the reading.
Data
| Wave | Trial 1 time (s) | Trial 2 (s) | Trial 3 (s) | Average (s) | Speed (m/s) | Ribbon motion |
|---|---|---|---|---|---|---|
| Compressional (P) | ||||||
| Shear (S) |
Triangulation data (illustrative, for the reading’s grid): Station A S-P lag 5.0 s; Station B 8.0 s; Station C 12.0 s. Distances: 8.4 × lag.
Analysis
- Claim: how do P and S waves differ in particle motion? Use your ribbon observations as evidence, and explain why a liquid could carry a P wave but not an S wave.
- Did your P pulse travel faster than your S pulse on the spring? Compare with real Earth, where P waves travel about 6 km/s and S waves about 3.5 km/s in the crust, and say whether the spring is a good model of speed as well as motion.
- Where is the epicenter on your grid, and how close did your three circles come to crossing at one point? Explain why a single station cannot locate an earthquake. Then compare d = 8.4 × lag with the real distances for three Cooter stations in the reading: does the rule overestimate or underestimate, and why?
- Sources of error: how do reaction time, uneven stretching, a slipping anchor and the size of your circles’ pencil lines affect your speeds and the location? How would you reduce each?
If you cannot do this lab in person: Use the PhET Wave on a String simulation for transverse (S-like) pulses and the recorded spring-toy run in this lesson’s reading for compressional pulses, time them with the sim’s stopwatch, and complete the same triangulation on the reading’s grid.
Formative check
Work through these before moving on. They are not graded — they tell you whether the standard below has landed yet.
Simulation & tools
The April 2026 Cooter, Missouri, M4.0 earthquake
Open the USGS event page for the April 23, 2026 earthquake near Cooter, Missouri. Record its magnitude, depth and location. Open the Did You Feel It? map and note the farthest Missouri city that reported shaking. Then open the Origin tab and find how many seismic stations were used to locate it. Explain in two sentences why so many stations give a better location than three.
No device? Use the printed event summary in this lesson’s reading: M4.0, depth about 11.8 km, 1 km west-northwest of Cooter in Pemiscot County, located with 89 stations.
Open U.S. Geological Survey ↗ · U.S. Geological Survey · Public domain (U.S. federal)
A seismometer records the S wave 20 seconds after the P wave. Using d = 8.4 × lag, how far away was the earthquake?
The S-wave shadow zone is evidence that Earth’s outer core is liquid.
In mantle convection, why does hot rock rise?
Draw a labeled cross-section of Earth with the crust, mantle, outer core and inner core at their depths. Add paths of a P wave and an S wave from one earthquake, and show the S-wave shadow zone. Add convection arrows in the mantle, with a sinking slab at a subduction zone. Label the evidence for each layer.
How confident are you that you can use seismic waves to locate an earthquake and explain how they reveal Earth’s layers and mantle convection?
Practice
Work these on paper or in your notebook, then open Check your answer. Aim for all of Fluency and Application; try at least one Challenge.
Printable version: this unit’s practice workbook (PDF)
Fluency
Build speed and accuracy with the core skill.
- A sound in air (speed 343 m/s) has wavelength 1.56 m. What is its frequency?
Check your answer
Answer: 219.9 Hzf = v/λ. - An S-P lag of 6.0 s is recorded. Using d = 8.4 × lag, how far away is the earthquake?
Check your answer
Answer: 50.4 km - An S-P lag of 15 s: how far?
Check your answer
Answer: 126 km - An earthquake is 84 km from a station. What S-P lag would the station record?
Check your answer
Answer: 10 slag = 84 / 8.4.
Application
Use the skill in context. Show your reasoning.
- Earth’s mass is 5.972 × 1024 kg and its radius 6,371 km. Find its average density in g/cm3 and compare with granite (2.7). What does the difference imply?
Check your answer
Answer: about 5.51 g/cm3; the interior must be much denser than surface rock, consistent with an iron-nickel coreρ = M / (4/3 πR3), converted from kg/m3. - The core-mantle boundary is at 2,890 km depth. What fraction of Earth’s radius and of Earth’s volume does the core occupy?
Check your answer
Answer: radius fraction 0.546; volume fraction about 16%Core radius 3,481 km; cube the ratio for volume. - Three stations record S-P lags of 5.0 s, 8.0 s and 12.0 s. Find the three distances and explain how they locate the epicenter.
Check your answer
Answer: 42 km, 67.2 km and 100.8 km; draw circles of those radii around the stations; the epicenter is where all three cross.
Challenge
Stretch problems. Expect to think before you write.
- Derive the S-P factor: if P waves travel 6.0 km/s and S waves 3.5 km/s, show that d = 8.4 × lag. Then find the factor for deeper rock where P = 7.0 km/s and S = 4.0 km/s.
Check your answer
Answer: lag = d/3.5 − d/6.0, so d = lag / (1/3.5 − 1/6.0) = 8.4 × lag; for 7.0 and 4.0 km/s the factor is 9.33
Review
Keep earlier skills sharp.
- Review (Unit 2): a mineral has 1 daughter atom per 1 parent atom. How many half-lives have passed?
Check your answer
Answer: 11 + 1 = 2 = 21.
Real P and S Arrivals from the 2026 Cooter, Missouri, Earthquake, a Grid for the Triangulation, and a Recorded Spring Run
USGS ComCat event nm60620376, origin and phase-data products (QuakeML arrivals, network nm) and Did You Feel It? summary, retrieved November 6, 2026; SLS original illustrative spring-toy record and triangulation grid · SLS original (cites USGS public-domain catalog and phase data; lab record and computations by SLS)
Why this reading. The lab’s triangulation needs a map grid, its spring-toy table is blank until you time the pulses, and the d = 8.4 × lag rule is stated but never tested against a real earthquake. This reading supplies the grid, a recorded spring run, and the actual P and S arrival times that the New Madrid seismic network used to locate the April 23, 2026 earthquake near Cooter, Missouri, so you can test the rule yourself.
Table 1. The Cooter M4.0 as located
| Quantity | Value from the USGS origin product (network nm) |
|---|---|
| Origin time | 2026-04-23 17:59:04.21 UTC (12:59 p.m. Central Daylight Time) |
| Epicenter | 36.0543°N, 89.8255°W, 1 km west-northwest of Cooter, Pemiscot County, Missouri |
| Depth | 11.8 km |
| Magnitude | 4.0 (moment magnitude, Mw) |
| Stations used to locate it | 89 stations, 121 phase arrivals |
| Nearest station | 0.066°, about 7 km from the epicenter |
| Root-mean-square error of the fit | 0.14 s |
| Felt reports | 913 (USGS Did You Feel It?); farthest Missouri reports from Monroe City and Joplin, 432 km |
Table 2. Stations that reported both a P and an S arrival (a selection)
From the phase-data file for the event. 32 stations have both picks; 14 are listed, from nearest to farthest. Times are seconds after the origin time; the epicentral distance is the catalog’s, converted from degrees at 111.19 km per degree. The lag is S minus P.
| Station | Epicentral distance (km) | P arrival (s) | S arrival (s) | S − P lag (s) | 8.4 × lag (km) | Distance / lag (km per s of lag) |
|---|---|---|---|---|---|---|
| NM.PEBM | 7.3 | 2.58 | 4.70 | 2.12 | 17.8 | 3.4 |
| NM.CHNM | 9.4 | 2.81 | 5.18 | 2.37 | 19.9 | 4.0 |
| NM.GSAR | 16.5 | 3.81 | 6.80 | 2.99 | 25.1 | 5.5 |
| NM.GNAR | 20.0 | 4.37 | 7.94 | 3.57 | 30.0 | 5.6 |
| NM.LNXT | 30.6 | 5.95 | 10.43 | 4.48 | 37.6 | 6.8 |
| NM.MIST | 31.9 | 6.05 | 10.44 | 4.39 | 36.9 | 7.3 |
| NM.ARPT | 35.8 | 6.60 | 11.86 | 5.26 | 44.2 | 6.8 |
| NM.PVMO | 41.5 | 7.64 | 13.57 | 5.93 | 49.8 | 7.0 |
| NM.MADT | 42.4 | 7.83 | 14.06 | 6.23 | 52.3 | 6.8 |
| NM.NWCT | 51.9 | 9.37 | 16.32 | 6.95 | 58.4 | 7.5 |
| NM.RVAR | 58.0 | 10.24 | 18.05 | 7.81 | 65.6 | 7.4 |
| NM.NMDM | 64.1 | 11.32 | 20.34 | 9.02 | 75.8 | 7.1 |
| NM.HCAR | 67.0 | 11.70 | 20.55 | 8.85 | 74.3 | 7.6 |
| KY.CUSSO | 71.0 | 12.39 | 21.53 | 9.14 | 76.8 | 7.8 |
What the real data say about d = 8.4 × lag
The rule comes from P at 6.0 km/s and S at 3.5 km/s: 1/(1/3.5 − 1/6.0) = 8.4. Against Table 2 it overestimates: for KY.CUSSO the rule gives 76.8 km and the station is 71.0 km away. Beyond 30 km, distance divided by lag runs from 6.7 to 7.8 km per second of lag, not 8.4. Three reasons, all worth naming in an analysis: (1) the Bootheel sits on hundreds of meters of soft Mississippi Embayment sediment, and waves near the surface travel slower than the textbook crustal speeds; (2) the focus is 11.8 km down, so the true path is the hypocentral distance √(d2 + 11.82): NM.PEBM is 7.3 km from the epicenter but 13.9 km from the focus, which is why its distance/lag of 3.4 is so low; (3) arrival times are picked on noisy records. The method still works: 89 stations locate the earthquake with an RMS error of 0.14 s, and seismologists use a measured velocity model instead of one constant.
The triangulation grid
Use this grid (print it, or copy it onto paper). One grid square is 20 km. The stations’ positions, in kilometers east and north of the grid’s lower-left corner: A at (92, 100), B at (10, 154), C at (80, 4). From the lab’s illustrative lags, distances are A: 8.4 × 5.0 = 42.0 km; B: 8.4 × 8.0 = 67.2 km; C: 8.4 × 12.0 = 100.8 km. Draw each circle with a compass set to the scale and mark where they meet.
Check after you draw: the three circles should meet close to grid point (50, 100), one square east and five squares north of the corner. If your circles miss by more than half a square, measure your compass settings again.
Table 3. Illustrative spring-toy record
Constructed teaching data, not measurements of a particular run. Metal spring stretched to 3.0 m on a tiled floor with the far end tied to a chair leg; timing from a phone stopwatch, started at the push and stopped when the pulse reached the chair leg.
| Wave | Trial 1 (s) | Trial 2 (s) | Trial 3 (s) | Mean (s) | Speed = 3.0 m / mean (m/s) | Ribbon motion |
|---|---|---|---|---|---|---|
| Compressional (P) | 0.86 | 0.91 | 0.88 | 0.883 | 3.4 | Back and forth along the spring, then still |
| Shear (S) | 1.32 | 1.28 | 1.35 | 1.317 | 2.3 | Side to side across the spring’s length |
The P pulse was faster by a factor of 1.5; in Earth’s crust the factor is about 1.7 (6.0/3.5). The spring is a fair model of the order and of the two motions, a rough one of the ratio, and no model at all of the absolute speeds, which are a thousand times slower than rock. Reaction time of roughly 0.2 s on a 0.9 s measurement is a 20% uncertainty, which is why three trials are the minimum and a slow-motion video helps.
Sources: USGS event page, M 4.0 – 1 km WNW of Cooter, Missouri (nm60620376) (earthquake.usgs.gov); USGS phase-data product for nm60620376 (QuakeML with arrivals) (earthquake.usgs.gov); USGS FDSN event web service, event nm60620376 (GeoJSON summary) (earthquake.usgs.gov); USGS Did You Feel It? ZIP-code table for nm60620376 (earthquake.usgs.gov)