Page 202 - The Geological Interpretation of Well Logs
P. 202

-  THE  GEOLOGICAL  INTERPRETATION  OF  WELL  LOGS  ~


        on  a  time  scale  rather  than  depth  scale,  correcting  for   it  must  be  angular:  disconformities  or  paraconformities
        interva]  velocity,  plotting  the  dip  results  in  the  plane  of  the   will  not  be  seen.  The  first  example  (Figure  ]2.33)  shows
        seismic  section  (i.e.  with  apparent  dip)  and  with  a  com-   a  strong,  angular  unconformity;  the  beds  above  being
        pressed  horizontal  scale.  The  results  are  best  presented  in   nearly  horizontal,  those  below  dipping  at  approximately
        the  form  of  a  stick  plot.  Necessarily,  this  conversion  is   22°  to  the  west.  The  example  also  shows,  which  is  very
        achieved  by  computer.  The  example  (Figure  12.32)  shows   typical,  that  the  dipmeter  cannot  be  used  to  pinpoint
        a  listric  fault,  with  a  marked  rollover,  crossing  the  well  just   the  actual  unconformity  level;  the  dips  are  too  scattered
        above  TD  (Werner  et  ai.,  1987).  The  fact  that  it  is  a  listric   due  to  burrowing,  weathering  or  diagenesis.  Using  the
        fault  is  clear  on  the  seismic  but  not  on  the  dipmeter;  but   neutron-density  logs  allows  the  precise  level  to  be  locat-
        the  location  of  the  faule  is  clear  on  the  dipmeter,  difficult   ed  at  a  probable  hardground  (Figure  12.33),  but  it  is  only
        to  pick  on  the  seismic.  The  integration  of  both  sets  of  data,   the  dipmeter  that  indicates  the  angularity  of  the  break.
        dipmeter  and  seismic,  gives  confidence  in  an  interpreta-   The  second  example  (Figure  12.34)  shows  a  more  typical
        tion  beyond  either  set  individually.  In  areas  for  example,   unconformity,  where  the  angular  change  is  very  small
        of  difficult  multiples,  side-swipe  or  complex  tectonics,  the   and  the  break  indicated  by  a  moderate  azimuth  change,
        seismic  dipmeter  plots  are  very  useful.       not  a  dip  change.  For  such  subtle  breaks  to  be  brought
                                                          out,  it  is  necessary  to  use  a  compressed  scale  log  (1:2000
        Unconformities  and  disconformities               to  1:5000)  and  to  zone  the  dipmeter  data  carefully,  the
        One  of  the  most  frequently  perceived  uses  of  the  struc-   actual  break  being  selected  on  the  standard  logs  rather
        tural  dipmeter  is  in  the  identification  of  unconformities.   than  the  dipmeter  itself.  The  necessity  for  combining  the
        However,  for  an  unconformity  to  be  seen  on  the  dipmeter   dipmeter  with  other  open  hole  logs  is  clear.

                                                           SHDT  0.6  x_0.3m_35"  x  2


                              45      NPHI       -15             DIP  —i-        o
                              1.95    RHROB     2.96    O°   yo   20°   30°   .   s0°   ,   3



                          1760  ~   >
                                 en
                                 _=
                                 =                                                  S
                                 =                                                  ¢
                                 S
                                                                                    a
                                a                                                   o
                                 en                                                 £
                                 on                                                 2
                                 go                                                 oO
                                 Tre                                                 .
                                a                                                   >
                                 em»
                             i
                            nen  ney
                                                                                  a
                                        UNCONFOANITY |  |:
                                    thardground,   :
                                         gap  ~100Ma)
                                                                                    Rm
                                                                                    w
                                                                       e   ~~       w
                                                                                    o
                          1800  -                                                   5  3
                                                                                    2
                                                                                    UO  ug

                            E                                                       =
                            £
                            $
                            a
                            o
                            =a


                                  WwW
                                                                      U.Cretaceous
                                  hardground
                                                                       M.Jurassic


        Figure  12.33  An  angular  unconformity  on  the  dipmeter.  Jurassic  sands  dip  at  approximately  22°  to  the  west  and  are  overlain  by
        horizontal,  Upper  Cretaceous  shales.  The  unconformity  is  picked  from  the  standard  logs  and  a  hardground  is  interpreted  from  the
        density  log  at  the  unconformity  level.  It  is  the  dipmeter,  however,  that  shows  that  the  unconformity  is  angular.

                                                       192
   197   198   199   200   201   202   203   204   205   206   207