Home

Annulus Monitoring - Closed or Open Annulus?

U.S. EPA Region IV allows rule authorized (existing) Class IIR injection wells to operate with either an open or closed annulus, at the operator's discretion, in Kentucky under their direct implementation (DI) underground injection control (UIC) program.  Said wells are not required to have their annular space, if present, closed, or monitored, other than for obvious leakage (casing injectors do not even have an annular space and are explicitly allowed by the UIC regulations).  Yet if a Class IIR rule authorized injection well must be permitted for some reason, Region IV imposes permit conditions whereby a closed annulus must be maintained and pressure monitoring conducted, though they appear to lack regulatory authority to do so.  Because thousands of rule authorized Class II injection wells are allowed to operate without a closed annulus, there is a substantive question of fairness and consistency at issue.  Accordingly, this questionable policy was challenged in a pair of UIC permit appeals before the Environmental Appeals Board (EAB) in UIC Appeal Nos. 98-3 & 98-5.

The UIC regulations neither support annulus monitoring, nor a closed annulus, for Class II injection wells; annulus pressure monitoring with a closed annulus is only mandated for Class I wells (see 40 CFR §§146.13(b)(2) and 146.13(a)(3)).  There is a monitoring of annulus pressure (MAP) [a/k/a SAMT] mechanical integrity test (MIT) that is a creature of 40 CFR §146.8(b)(1), but Region IV has never approved its use in Kentucky, and it should not be confused with monitoring requirements set out elsewhere in the regulations.  Region IV misled the EAB in its pleadings by citing 40 CFR §146.8(b)(1) as its authority to require a closed annulus and annulus monitoring.  While all of the above is troubling in its own right, there is a more important consideration: Is a closed annulus always preferable, especially for very shallow wells?

There is good cause to conclude an open annulus is preferable to a closed annulus for many, if not most, Class II injection wells as discussed below.  Not discussed below is the problem with the environmental effects on the pressure within a closed annulus.  While a sealed annulus with no leaks is closed to matter transfer, it is not closed to energy transfer, because it is free to exchange heat with its surroundings.  API hired Gruy Engineering to do heat transfer analyses, and they calculated a possible 1,100 psi annular pressure change for a mere 10ºF change in injection fluid temperature (there exists considerable literature on the effects of injectant temperature change on annulus pressure, both theoretical and empirical, but dramatic pressure changes with changes in temperature are universally reported).  Thus, a closed annulus pressure can vary to such an extent that periodic monitoring
will result in false indications of problems (the fact that Region IV does not acknowledge frequent difficulties with closed annulus monitoring speaks volumes about how seriously most operators take their obligations under UIC permits, but also see The 15 psi Vacuum at U.S. EPA, Region IV).  Further, potentially destructive pressures can theoretically occur in a closed annulus if you accept EPA's own research on the matter.

The EAB remanded the annulus status issue (and six others), and after over 1-1/2 years, Region IV finally addressed the remanded issues in a December 1, 2000 letter, and reissued the two UIC permits.  The Petitioners felt that EPA Region IV inadequately addressed the remanded annulus status issue, and filed a new Petition for Review with the EAB on January 15, 2001.  The EAB declined further review, and the matter was brought before the U.S. Court of Appeals for the Sixth Circuit.  Unfortunately, the court dismissed the pro se UIC permit appeal for lack of standing of the petitioner.

What follows is the full unaltered text (reformatted for this webpage) of the annulus status portion of the January 15, 2001 second Petition for Review:


2.  Annulus Status (8. Annulus Status, in EAB Opinion)

Section I.C.2. of the two new final UIC Permits states, in part, "The annulus pressure shall be maintained at 0 psig."  A cease injection and reporting requirement is triggered if annulus monitoring reveals a positive change in annulus pressure of 15 psig or when a vacuum of 5 psig occurs.  This provision remains unchanged from the previously appealed language.  Petitioners argued that neither a closed annulus, nor monitoring of annulus pressure, can be required under the UIC regulations.  Petitioners also asserted that for the shallow injection wells at issue herein, an open annulus is the preferred mode of operation.

The Board remanded the annulus status issue, instructing Region IV to "... provide a reasoned response to petitioners' concerns regarding the need for a closed annulus ...."  The Board stated, "Although we agree that the regulations authorize the Region to require monitoring of annulus pressure in appropriate circumstances, the Region's failure to adequately respond to petitioners' assertion regarding an open annulus requires that the permits be remanded on this issue."

Unexplained Deletion of Permit Condition

In footnote 19 to the Opinion, the Board points out some peculiar language in the original UIC Permit KYA0362, where condition I.C.2. stated, "For those well [sic] completed with an open annulus ...", concluding that "... the Region considers it possible that wells with an open annulus can nevertheless maintain an annulus pressure of 0 psig."  Of course any well with an annulus, open or closed, can have an annulus pressure of 0 psig (essentially atmospheric pressure).  The "open annulus" language has been removed from the current version of UIC Permit KYA0362, without explanation for the change.  Petitioners are puzzled, and believe Region IV should have explained the deletion of said language, especially since the Board commented on same.

Closed Annulus Monitoring

In its Response, Region IV cited 40 CFR §146.8(b)(1) to support its position that it has legal authority to require a closed annulus and monitoring of the annulus pressure.  However, 40 CFR §146.8(b)(1) describes a particular MIT, not a monitoring requirement.  The subject injection wells are required to use the SAPT MIT (a creation of 40 CFR §146.8(b)(2)) by Section I.B.3. of the two new final UIC Permits, not the monitoring of annulus pressure ("MAP") MIT of 40 CFR §146.8(b)(1).  Even if dual internal MITs could be justified, 40 CFR §146.8(b)(1) was amended in 1993 (Reg-Fix) such that the 0 psig well head annulus pressure required in the subject UIC Permits is now prohibited by the regulations (the post Reg-Fix 40 CFR §146.8(b)(1) requires "… maintaining an annulus pressure different from atmospheric ...").  Ironically, FOIA requests to Region IV reveal that the Region has never allowed an operator in Kentucky to use the 40 CFR §146.8(b)(1) MAP MIT, but at the same time would here argue that said section authorizes the use of annulus monitoring for Class IIR wells.

The regulations distinguish between an MIT and monitoring requirements.  The criteria and standards for MITs are contained in 40 CFR §146.8.  The criteria and standards for monitoring of Class II wells are contained in 40 CFR §146.23(b).  Compare 40 CFR §146.23(b) to the parallel monitoring requirements for Class I wells at 40 CFR §146.13(b).  Class I wells have a requirement to monitor "... the pressure on the annulus between the tubing and the long string of casing ...."  40 CFR §146.13(b)(2).  No such monitoring requirement exists in the equivalent Class II regulations at 40 CFR §146.23(b).  Even more telling is the Class I requirement at 40 CFR §146.13(a)(3) that "... the annulus between the tubing and the long string of casings shall be filled with a fluid approved by the Director and a pressure, also approved by the Director, shall be maintained on the annulus." There is no such requirement at 40 CFR §146.23(a) for Class II wells.

Annulus monitoring is not required for hundreds of rule authorized wells that surround the subject facilities.  Your Petitioners know of not one single rule authorized injection well for miles around the herein permitted wells that is required to monitor annulus pressure, and many, if not most, have open annuli.  It should be noted that Region IV concedes that the subject existing injection wells are rule authorized Class IIR wells, hence they are indistinguishable from surrounding rule authorized wells (except that Region IV forced the wells herein to be permitted).

There is no regulatory requirement for annulus monitoring for Class II injection wells.  There is no regulatory requirement for Class II wells that the annulus be closed. Region IV, by its own actions with regard to thousands of rule authorized Class II wells, has demonstrated annular monitoring is not required, nor is it necessary for Class II wells.  The Board is correct that Region IV may "... require monitoring of annulus pressure in appropriate circumstances ...", but only as an MIT, not as a monitoring requirement for Class II wells.  Region IV chose the SAPT MIT and has not explained why the MAP MIT is also necessary, nor how the MAP may be used in conflict with the explicit language of 40 CFR §146.8(b)(1).

Open Annulus Preferred

Petitioners have stated that a closed annulus is not necessarily more protective than an open annulus, and that an open annulus can often provide an indication of a problem faster and more conclusively than a closed annulus.  Petitioners have further stated that an open annulus is often the preferred mode of operation for very shallow injection wells.  In the December 1, 2000 response letter, Region IV states, "To maintain this pressure [0 psig] the Region contends the annulus must be closed."  As discussed above, a closed annulus is not necessary to maintain a 0 psig pressure, but clearly Region IV means to say that a closed annulus is necessary to enable detection of changes from 0 psig.

In the December 1, 2000 response letter, Region IV goes on to state, "Your argument that an open annulus is the preferred mode [sic] operation and can provide an indication of a problem faster and more conclusive [sic] than a closed annulus is not supportable."  They do concede that in the case of a tubing leak where the casing remains intact, "It is true that, where an injection well is shallow with an open annulus between the tubing injector [sic?] and the outer well casing, it is possible to easily and readily visually detect a leak because any leak from the injection tubing will flow into the open and visible annulus and will be visible to the naked eye."  Petitioners agree, and would point out that any such leak would likely be detected sooner than the 30 day interval for annulus pressure monitoring measurements specified in the subject UIC Permits.  So in the case of a tubing leak, with casing remaining sound, an open annulus is as good, and probably better than a closed annulus for fast leak detection.

But we must consider other mechanical integrity failure modes.  What if the tubing remains sound and a leak develops in the casing?  With an open annulus, a casing leak opposite an overpressured zone (a formation that will flow to surface) would flow to the surface and be readily detected, but with a closed annulus, the pressure would have to rise to 15 psig to trigger action under the subject UIC Permits, and your Petitioners have never seen an overpressured zone as high as 15 psig at surface on the subject facilities (a positive hydrostatic head in excess of 34.5 feet).  If the casing hole occurs opposite a zone of no significant porosity/permeability, then said casing leak would not likely be detected with either an open or closed annulus, but would probably be detected with a subsequent MIT.  If the casing leak occurs opposite a zone of porosity/permeability that is not overpressured, then said casing leak would not be detected with an open annulus (until the next MIT), but it might possibly be detected with a closed annulus.  Closed casing annulus pressure monitoring can only detect a leak under this scenario if the fluid drops a sufficient distance in the annular space to produce a 5 psig vacuum (in excess of 12 feet).  It should be noted here that the subject UIC Permits are the only two ever issued by Region IV with a 5 psig vacuum action trigger; all others have been 15 psig, and more recently 13 psig, numbers that guarantee no leak resulting in a vacuum could ever trigger a cessation of injection or reporting.  Included is a printout of a web page at www.logwell.com entitled, "The 15 psi Vacuum at U.S. EPA, Region IV", http://www.logwell.com/tech/uic/15psi_vacuum.html (attached as Exhibit 5).  Though only peripherally relevant to this Petition, it speaks to every other UIC Permit ever issued by Region IV (and to some other things as well).

The final mechanical integrity failure mode we must consider is a leak in the tubing (or a packer leak) combined with a casing leak.  If the casing leak is opposite an overpressured zone, fluid would flow to the surface and be readily detected with an open annulus, but with a closed annulus, the pressure may or may not attain the 15 psig trigger pressure (depending on the size of the tubing leak since no 15 psig surface pressure overpressured zones are known under the subject facilities).  If the casing leak occurs opposite a zone of no significant porosity/permeability, then said casing leak would be readily detected with either an open or closed annulus, but likely faster with an open annulus.  If the casing hole occurs opposite a zone of some porosity/permeability that is not overpressured, fluid would almost certainly flow to surface and be readily detected with an open annulus, but with a closed annulus, the pressure may or may not attain the 15 psig trigger pressure, and very likely would not (depending on the size of the tubing leak and the ability of the zone behind the casing to take fluid).

Finally, what if our casing leak is opposite very high effective porosity and permeability?  This could be a vugular formation, a karst-like feature, or even an underground source of drinking water ("USDW") (Petitioners have long asserted no USDW is present beneath the facilities at issue, but a hypothetical USDW type aquifer capable of serving a public water system ("PWS") would yield considerable water and thus would be capable of receiving same).  Any casing leak can be thought of as an orifice, where the pressure drop across said orifice determines the size of the leak (the gallons per minute or barrels per day).  A hole in casing considerably smaller than one-quarter (1/4) inch diameter, with no more than 15 psi pressure drop across said casing hole, would leak more water than any of the subject injection wells can inject.  Under this last scenario, a well with an open annulus will almost certainly flow to surface and be readily detected (except in the case of a fairly large hole opposite a zone of infinite effective permeability and a hydrostatic head below surface, something never seen on the subject facilities).  But with a closed annulus, such a leak might not be detected, and in fact the allowed pressure up to 15 psig could actually result in a higher leak rate through the casing defect!  Think in terms of the fluid that would have escaped at the surface having no place to go but into a receiving formation, with up to 15 psig additional pressure differential across the casing leak.

Timko, Lindahl & Schweikhardt comment on monitoring annulus pressure with the annulus closed as follows:  "The danger arises in that a leak in the casing to a permeable zone could inhibit pressure buildup which would otherwise be caused by a leak in tubing.  In the extreme, flow of injected fluid through both leaks into an USDW could occur undetected."  (Dewan, John T. 1983. Mechanical Integrity Tests - Class II Wells, Review and Recommendation, Final Report. Prepared for EPA Region II and III. Page 9.)  Similar observations appear throughout the literature.

The following table summarizes the above discussion:

OPEN VERSUS CLOSED ANNULUS COMPARISON TABLE

                             OPEN ANNULUS                  CLOSED ANNULUS
                         
Visible Flow Trigger           +15, -5 psig Trigger

Group I
Tubing Leak
                    Detected                       Detected
(Casing Intact)               (Probably Faster)


Group II*
Casing Leak
(Tubing Intact)
Leak Opposite:

Overpressured Zone             Detected                       Not Detected

No Permeability                Not Detected                   Not Detected

Some Permeability              Not Detected                   Perhaps Detected

High Permeability              Not Detected                   Perhaps Detected


Group III (Highest Risk)
Casing Leak and
Tubing Leak
Casing Leak Opposite:

Overpressured Zone             Detected                       Probably Detected
                              (Probably Faster)

No Permeability                Detected                       Detected
                              (Probably Faster)

Some Permeability              Detected                       Maybe Detected**
                              (Probably Faster)

High Permeability              Detected                       Maybe Detected**
                              (Probably Faster)              (Maybe a worse leak!)

Notes:
The above Table contemplates only the injection wells located on the subject facilities, and assumes that any overpressured zones will not exceed 15 psig at surface.

*  Region IV concedes that at most there can only be one USDW under the subject facilities.  Therefore, Group II mechanical integrity failures cannot allow communications between USDWs (Petitioners have long asserted no USDW is present).

** Group III mechanical integrity failures with casing leaks to permeability will not be detected if the equilibrium pressure is less than 15 psig or greater than 5 psig vacuum.


In the December 1, 2000 response letter, Region IV states, "However, this premise assumes there is no leak in the well's outer casing; for, if there were such a leak, it is quite possible that all liquids leaking from the injection tubing into an open annulus could leak through the outer casing and not accumulate in the annulus.  Thus a leak would not necessarily be visible to the naked eye.  This problem would not occur with a closed annulus which is maintained and monitored at 0 psig."  The likelihood that all liquids leaking into an open annulus could leak through the outer casing and not accumulate in the annulus is incredibly remote.  Any porous formations above the injection zone underlying the subject facilities have hydrostatic heads at least closely approaching the surface.  Any tubing leak will cause fluid to accumulate in the annulus, and to nearly always overflow.  It is more likely that a situation as described by Dewan for closed annuli would occur than for an open annuli to fail to overflow in the manner described by Region IV, but of course either is possible.  The lengthy foregoing discussion and table should make it clear that the above statement from Region IV is grossly oversimplified and misleading.

As noted above, there is no regulatory requirement for Class II wells that the annulus must be closed; that requirement only applies to Class I wells.  Region IV, with its own actions with regard to thousands of rule authorized Class II wells, has demonstrated that open annuli is an accepted practice in Region IV.  There exists a real possibility that a serious leak involving failure of mechanical integrity in both the tubing and the casing will not be detected with a closed annulus (and the leak rate to permeability could actually be made worse with a closed annulus).  Petitioners maintain their position that an open annulus would be the preferred mode of operation for the subject very shallow injection wells.  The possibility that a closed annulus could mask detection of a leak in the tubing and casing should be sufficient to recommend against a closed annulus.

Annulus Status Summary

The UIC regulations neither support annulus monitoring, nor a closed annulus, for Class II injection wells; annulus pressure monitoring with a closed annulus is only mandated for Class I wells.  40 CFR §§146.13(b)(2) and 146.13(a)(3).  There is a monitoring of annular pressure (MAP) MIT that is a creature of 40 CFR §146.8(b)(1), but Region IV has never approved its use in Kentucky, and it should not be confused with monitoring requirements set out elsewhere in the regulations.  Thousands of rule authorized Class II injection wells are allowed to operate without a closed annulus; thus, there is a substantive question of fairness and consistency at issue.  For the reasons set out above, Petitioners remain firm in their assertion that the subject injection wells should be allowed to operate with an open annulus.  EPA Region IV has made findings of fact and conclusions of law which are clearly erroneous and/or has exercised discretion which the Administrator should further review.

See also Pro Se UIC Permit Appeal.

| Home | Tech & Tips | UIC |

03-17-01
Last 10-20-10