Recent Developments: Stratigraphy

The relationship between fracturing, asymmetric folding, and normal faulting in Lisburne Group carbonates: West Porcupine Lake Valley, northeastern Brooks Range, Alaska

This page is meant to be a potpourri of interesting figures and explanations of what I'm currently working on.  I'll try to keep it up to date.


"Fracture Stratigraphy"
I've started to look at how fractures are behaving within the un-folded stratigraphy.  Here's a sample of a stratigraphic section from the Lower Lisburne showing how lithology, bed thickness, and chert content affect fracture spacing (HUGE thanks Michelle McGee for so graciously giving me her strat columns).  The fracture spacing data were collected by measuring the number of fractures in a 0.5m line perpendicular to the fracture set.  This is a "quick and dirty" way to get at fracture density.
Note the general increase in fracturing in the interval between 22-29m.  The interval contains mudstones with mostly lenticular and nodular chert.  The relative competency contrast between mudstone and chert likely created conditions that were favorable for fracturing.  Also note the interval between 63-75m which consists of 5-15cm thick shales interbedded with mudstones.  This interval has multiple thin beds, but not enough competency contrast to create conditions favorable for fracturing.
More on Fracture Stratigraphy
*Note that in this graph I have simply summed all the fracture data to make one line for simplicity.  The summed values are shown in green.

I began to suspect that what affects fracture spacing is not one particular lithology or bed thickness, but changes in either one.  So, to take this one step further, I calculated the "rate of change" (ROC) in the lithology by assigning a number to each lithology (mudstone=1, wackestone=2..etc) and then subtracting each value from the one stratigraphically above it.  I did the same for bed thickness.  So, if my theory was correct, the peaks on the rate of change diagram should correspond with higher fracture densities. (sorry for the poor quality of this figure, I'll try to get a better copy where you can see the rate of change peaks a bit better)  The theory seems to hold up to some degree, although it's clear we haven't explained everything.

Here's where the real arm waving starts.  Once I calculated values for the ROC lithology, ROC bed thickness, and chert (no chert=0, bedded=1,lenticular=2....) I could come up with a very simple "model" of fracture density in the stratigraphy by summing the values corresponding to the three attributes.  The blue graph corresponds to these values and it seems to be a reasonable fit, except for a few places where it deviates from the observed values.  These may be due to factors I haven't accounted for, which include dolomite content, "swarming" of fractures, and various scales of bedding (mechanical compartmentalization/amalgamation of bed contacts).

Taking the "models" one step further:

Below is a figure of three relatively simple predictive models for fracture density in the "Forks Canyon" stratigraphy.  The first three columns show stratigraphic data as shown above (lithology, bed thickness, and chert).  The fourth column from the left shows a purple line that represents a model that assumes that chert, lithology and bed thickness have equal influence on fracture density.  In this model, mudstones are assumed to have the highest fracture density, and grainstones are assumed to have the lowest fracture density.  In the same manner nodular cherts are assumed to create the most favorable conditions for fracturing, followed by lenticular and bedded chert. The next column (red line) shows a model that is similar to the previous model, but assumes that wackestones and packstones will have higher fracture density than mudstones, grainstones, or floatstones, as was suggested by Joe Brinton (UAF M.S. Thesis and DOE reports).  The last model ignores bed thickness altogether and assumes that fracturing is a function of the chert content and the difference in lithology (or changes in lithology) throughout the section.  The column on the far right shows the deviation between the model and observed values.

By comparing which models most accurately fit the data, I hope to be able to determine which factors (bed thickness, lithology, or chert) have the most influence on fracture density in the stratigraphy.  I am currently improving on these models, coming up with better ways to compare them, and getting a better fit to the data.  I will also include dolomite content in the models, which will likely have a significant effect on fracture density, and hopefully explain some of deviation between the model and actual data.


 

Go to: Recent Developments: Fractures and Folds

Contact

Ryan's Home Page

UAF Tectonics and Sedimentation Research Group Home Page

UAF Department of Geology and Geophysics
Geophysical Institute