Body waves (P & S) travel through the Earth at velocities of 3 to 11 km/s. They are the first phases to be recorded at a station and can arrive tens of minutes before the surface waves at large distance. A key to interpreting body waves is to realize that these signals arc deep into the Earth and arrive at a steep angle of incidence at teleseismic distances. This conveniently orients raypaths close to the vertical, Z, component of motion. Compressional waves display most of their energy on the Z component of a seismogram. Shear waves dominate the two horizontal components.

This break down is a bit simplistic and the signal can couple into the other components but the concept is clear: If you want to look for P waves, check the Z component; for evidence of shear waves check the N and E components.

P phase
For events less than 95 degrees away, the P wave on the vertical record should be the first visible phase. It will have a clear onset, pulse-like shape, high frequency content and will typically be gone within a few tens of seconds. Secondary phases such as pP, arriving slightly later, can extend this packet to a couple minutes at most. The P wave arrival time can be measured as the first instant when the amplitude exceeds the background noise level. The angle of incidence of incoming rays is typically within 20 degrees of vertical. Most of the motion will show up on the Z component. Some energy however will be partitioned into the horizontal plane. This horizontal information can be used to determine the rough back azimuth to the event.

Z N E

The Z,N and E components (top to bottom) of a direct P wave from a magnitude 5.6 earthquake on the Kenai Peninsula, Alaska. The depth is 89 km and the distance from Palisades is 49.9 degrees (5545 km). 50 seconds of data is shown.

In this example from Alaska, the Z component shows a positive first pulse; the motion at Palisades began with a compression. On the N component the first pulse is negative; in the N/S plane this compression was pointed toward the south. Similarly the positive pulse on the E component points east. A phase "pointing" SE arrived from the NW. If the first vertical motion had been negative, then the other two components would have been inverted as well. Alaska is NW of Palisades so energy showed up on both the N and E components. Had the event been directly west of Palisades, very little energy would show up on the N component. If energy is seen on both the N and E component, then a little mental interpolation can often yeild a compass bearing to the event within 20 degrees. This method is particularly useful for distinguishing events in the Aleutian islands from Central and south America. Both regions are at similar distances and typically create very clear P waves. However, they differ in back azimuth by more than 90 degrees.

S phase
Primary S wave phases follow similar raypaths as their P wave counterparts. Many of the same rules as the P phase. Their velocities are slower so they arrive later on the record. Since shear motion is transverse to the near vertical raypath, most S wave motion is displayed on the horizontal components. On the N and E components a small P wave signature may be present, but the later arriving S phase will typically be larger. Similarly, this S phase should not be strong on the Z component. Compared to the P wave, the S wave will have lower frequency content and the exact onset of the signal may not be as clear. The direction to an event cannot be redily obtained from the first arriving S waves.

Z N E

In this shallow magnitude 6.1 example from Kodiak Island, Alaska, 15 minutes of data for the Z, N and E components is displayed. Note the clear P wave arrival and very weak S arrival on the Z component. On the N and E records the S phase is far stronger than P.

S-P time
Once the first P and S arrivals have been identified, the distance to the event can be determined. Since P and S phases travel at different velocities, the time seperating them increases steadily with distance. The amount of time seperating the first appearance of these phases can be translated into a distance in degrees. A traveltime table will yeild the most precise distance but the expression below is quite handy as well.

In the Kodiak, Alaska example above, 7.2 minutes seperate the P and S phases. The S-P time agrees nicely with the 51.4 degree distance to the epicenter. This analysis breaks down much beyond 100 degrees however. Beyond this point the core blocks direct P and S phases. In these cases you will need to be more creative. The same time differencing can be performed on any two phases you can positively indentify. A traveltime curve will be necessary to translate the time difference to a distance.

pP and depth phases
Deep events often display duplicate versions of the P wave pulse which arrive several seconds to a couple minutes behind the first arrival. These phases are the result of rays which depart upward from a deep source. After a reflection at the surface, their ray paths are nearly identical to thier counter parts which departed downward from the source. This short leg to the surface is denoted with a lowercase p. pP lags behind P by twice the time it takes to reach the surface. A rough depth can be calculated assuming an average velocity of 8 km/s.

Comparison of downward departing P and S phases and upward departing pP and sS. Note the near-parallel paths downward away from source. More body wave raypaths

This is your best shot at determining the depth of an earthquake. Usually P and pP are close enough that pP is contained in the decaying wave train following P. If they are too close pP may not be distinct. Because of its free surface reflection, pP is flipped. This upsidedown feature of pP can be a useful diagnostic though the focal mechanism of the earthquake can interfere with this simplification.

The same effect can happen with S waves departing up from the source. Phases which begin this way are prefaced with a lowercase s. After bouncing at the surface, a p or s phase is then presented with all the possible path choices of a regular downgoing phase. In addition to pP and sS, is it not uncommon to see pPP, sSS, pPKP, sSKS, etc. At the surface bounce point some P energy can be converted to S waves and vice versus. This permits phases such as pS, sP,sPP,etc. to exist. It is not uncommon to see such phases.

PP phase
A close cousin of pP, this phase departs the source downward but at a shallower angle than P. It arrives at the surface half way bewteen the source and receiver where it is reflected down for a second leg of its journey. The top few hundred kilometers of Earth are the most highly attenuating. High frequencies are preferentially absorbed. Thus PP, which passes through this zone two more times than P, is typically a lower frequency signal. For some focal mechanisms and at distances beyond 95 degrees, PP can be larger than P. A completely anologous shear wave exists. Not surprisingly it is labeled SS.

PS and SP phases
PS is similar to PP except the second leg is travelled as a shear wave. Since it arrives as a shear wave, it is predominately visible on the horizontal components. SP is just the reverse and shows up on the vertical record. For a shallow earthquake, the traveltime of PS and SP are the same. They are only distinguishable because they appear on different components. For deep earthquakes, look for separate PS and SP arrivals. Deeper events have a shorter first leg so SP arrives slightly ahead of PS.

PKP,PKiKP,PKIKP phases
These phases can present a baffling set of pulses arriving very close in time. Though they all exhibit vertical motion, judicious use of the traveltime tables often allows more than one to be indentified particularly at distances around 150 degrees.

SKS is analogous to PKP. Note however, that this phase must convert to a P wave to travel through the liquid outer core. With a conversion from S to P back to S, it is not surprising that SKP and PKS are often observed as well.

Back to the table of contents