This page specifies how to interactively control interval velocity function parameterization using statistics, and converge towards your best estimate velocity and depth models.
Time to depth conversion velocity model will always generate well depth mismatches between measured depth and estimated depth trend.
Udomore® Depth prior vs. post parameterization aims to systematically minimize these mismatches. Note: depth mismatch statistics are relevant only if 10 wells or more are involved in your depth conversion scenario. With less than 10 wells, statistics can only be treated as informative. The well-depth mismatch table displays:
-
Experimental statistics computed on depth mismatches between well depth marker values and estimated depth trend values, to quantify the level of performance of your velocity model:
Experimental prior Depth mismatches between well depth marker values and prior depth trend values
(i.e. computed from prior parameterization values).Experimental post Depth mismatches between well depth marker values and post depth trend values
(i.e. computed from post parameterization values).Tips “Experimental Post” values should be lower that “Experimental Prior” values, demonstrating minimization
of well-depth mismatches. -
Experimental statistics may be displayed in terms of standard deviation (“RMSD”) of depth mismatches:
RMSD Standard deviation of depth mismatches between measured depth and estimated depth trend values. Tips RMSD values should be as low as possible to ensure optimal accuracy of your interval velocity model. -
Experimental statistics may be displayed in terms of mean value of depth mismatches:
Mean Average value of depth mismatches between measured depth and estimated depth trend values. Tips “Mean” values should as close as possible to 0 to ensure that interval velocity model doesn’t lead to
systematic over/under estimation of layer depth. -
Experimental statistics may be displayed in terms of minimum and maximum values of depth mismatches:
Minimum Minimum value of depth mismatches between measured depth and estimated depth trend values. Maximum Maximum value of depth mismatches between measured depth and estimated depth trend values. Tips The range between minimum and maximum depth mismatch values should be reduced to the maximum possible extent. -
Experimental statistics may be displayed in terms of mean value of depth mismatches:
Modeled RMSD Modeled standard deviation corresponds to the confidence range around the estimated depth trend,
derived (or “modeled”) from time uncertainty values and residual interval velocity uncertainty values
(i.e. without reference to measured well depth). For additional information, see Model Values on our Website.Tips Check that the modeled RMSD values (i.e. confidence range without reference to measured well depth)
are consistent with the experimental RMSD statistics (i.e. direct reference to measured well depth):
modeled RMSD values should be equal or higher than the experimental RMSD values.