next up previous print clean
Next: PS-DMO in the frequency-wavenumber Up: Kinematics of PS-DMO Previous: Kinematics of PS-DMO

PS-DMO smile derivation

The trajectory equation of conversion points is  
 43#43 (14)
By introducing the ratio, 44#44, and by following Huub Den Rooijen's derivations, as well as work by (), we have the following standard form of elliptical equation:  
 45#45 (15)
where D is the element responsible for transforming the data from CMP coordinates to CCP coordinates. D comes from solving equation ([*]) for z2. The difference among the existing PS-DMO operators is in their definition of D. In this paper we use the definition presented by ():
46#46 (16)
where tn is the NMO-corrected time.

The spatial shift distance needed to convert CMP gathers to common-reflection point gathers is  
 47#47 (17)

The one-way normal-incidence distance R is given by  
 48#48 (18)
Using equations ([*]) and ([*]) we eliminate the z and x dependencies. This result is substituted in the relationship for the one-way normal-incidence distance R [equation ([*])] in order to get the equation:  
 49#49 (19)
The two-way normal-incidence time is  
 50#50 (20)
and the traveltime equation, in terms of the normal moveout time, is  
 51#51 (21)
Combining equations ([*]), ([*]) and ([*]) produces the PS-DMO smile equation:  
 52#52 (22)
where
53#53


next up previous print clean
Next: PS-DMO in the frequency-wavenumber Up: Kinematics of PS-DMO Previous: Kinematics of PS-DMO
Stanford Exploration Project
6/7/2002