Home

In Western Kentucky (and perhaps elsewhere), there exist a small number of enhanced recovery water injection wells with very low injection rates, a few under five (5) barrels per day (BPD).  Some of these wells are located in areas with very limited groundwater resources, and in some cases the injectant is fresh potable water.  Back in 1992, it occurred to us that the very lowest risk of these wells should be able to take advantage of the explicit language of the US EPA underground injection control (UIC) regulations with regard to demonstration of internal mechanical integrity (MI), and thus was created the Self Evident mechanical integrity test (MIT).

The standard at 40 CFR §146.8(a)(1) for internal mechanical integrity is that there be "...no significant leak in the casing, tubing, or packer..."  Unfortunately, EPA has conspicuously avoided issuing much guidance on what constitutes a "significant leak", but it is clearly something more than no leak at all.  In certain parts of the US, a 5 or 10 BPD leak would be considered insignificant, and some accepted MIT procedures cannot detect leaks of this size.  Further, the UIC regulations do not prohibit all leakage, only that leakage capable of causing a violation of the primary drinking water standards.

The basic concept behind the Self Evident MIT is that a well taking little injectant, say 10 BPD of fresh potable water, for many years at full normal injection pressure, can hardly have a "significant leak".  In our hypothetical well, if the injection zone is taking anything at all , there would be little left to leak.  Importantly, it is well established that casing and tubing leaks get bigger with time (oilfield workers have known this for decades, but EPA, Ashland, and others have done actual scientific research on the topic).  Hence, our hypothetical injection well cannot have a "significant leak" for several reasons:

1. Even if all the injectant was leaking, the leak would still not be significant because of the very low injection rate,
   
2. The behavior of the well is inconsistent with that of leaking wells where leak rates increase with time, and
   
3. The potable water injectant cannot comprise a "significant leak", especially when its quality is  higher than the aquifer that could be affected.
   

Our hypothetical well is actually undergoing a continuous pressure test, having a known maximum leakage rate equal to , and almost certainly less than, the very low long term injection rate, where said leakage must be deemed insignificant not only due to the small size, but also because of the character of the injectant.  What is described here properly falls under 40 CFR §146.8(b)(2) a "...Pressure test with liquid or gas…" where said test is of a continuous nature.  It is not the test contemplated at 40 CFR §146.8(b)(3), a records of monitoring test (monitoring records are only important here as theoretical underpinning, i.e., casing and tubing leaks always get bigger with time).

A true tale about a client's injection well may help clarify the Self Evident MIT.  This client had an injection well (badly in need of acidizing and/or a clean out) where injectivity had gradually declined until it was next to nothing.  A supposed standard annular pressure test (SAPT) was performed on said well with an EPA witness present, and the well passed.  The novel part is that the well is a casing injector with no tubing and packer, bridge plug, or anything else to isolate the injection zone from the casing.  Obviously, said well was losing so little fluid to the injection zone and any possible casing leaks combined, that it was able to pass the MIT.  In other words, the well had passed a Self Evident MIT that looked like an SAPT!  Ironically, this is probably the only SAPT ever conducted in Kentucky that actually tested casing shoe primary cementation integrity.

We have thus illustrated an actual EPA witnessed Self Evident MIT at one end of a spectrum, a well that could not have a "significant leak" because it passed a Self Evident MIT well enough to look like an SAPT.  We need only define the other end of the spectrum, and that can only be done if EPA fairly addresses what constitutes a "significant leak".  It is our contention that the quality of the injectant must also enter into the equation (a position consistent with the practice in at least one EPA region).

If there is any uncertainty inherent in the Self Evident MIT, it should be remembered that EPA's favorite MIT, the SAPT, fails to address leakage by the casing shoe in any form or fashion.  The cement records MIT is almost always used to demonstrate external integrity when the SAPT is used to demonstrate internal integrity.  As is discussed in The SAPT / Cement Records MITs, the cement records check external MIT is an abomination in that it reveals nothing about primary cement integrity (other than cement may be present).  Stated otherwise, a well that has passed the SAPT and cement records check MIT combination could very easily be leaking a significant percentage of its injectant past the casing shoe (applicable to perforated wells also).  We live with a great deal of uncertainty with regard to mechanical integrity under the most commonly used Class II external integrity MIT; the Self Evident MIT is far more dependable than the horrid cement records check (actually wells subjected to the Self Evident MIT would also be subject to the cement records check because the regulatory language compels us to regard the Self Evident MIT as an internal pressure test only).

In order to dismiss the Self Evident MIT, US EPA persists in characterizing it as a 40 CFR §146.8(b)(3) records of monitoring test; it is not.  They have also said it cannot be a pressure test covered by 40 CFR §146.8(b)(2) because said test "...would take place over a short duration of time.", but in fact 40 CFR §146.8(b)(2) is completely silent on the time duration of pressure tests (aside from said argument making no sense).  Region IV approved the "ADA" Pressure MIT as an acceptable variant of the pressure test contemplated at 40 CFR §146.8(b)(2), not as an alternative MIT pursuant to 40 CFR §146.8(d); and so should they embrace the Self Evident MIT.

09-30-00
Last 10-20-10