[Where Basin-ward coal
seams are contiguous between producing CBM wells, the Fruitland outcrop and the
coal seams, the hydrologic system may reflect a “U-tube” concept. Consequently, a dynamic fresh water front could
migrate into the Basin, postulated as follows: Coal beds deposited in an
aqueous environment saturated with connate water following the coalification
process would have the equilibrated hydrostatic pressures disrupted by
subsidence of the basin relative to the basin ridges. Migration of connate
water would have been induced by the effects of relative topography and
perpetuated by atmospheric recharge events subsequent to the formation of the
basin. The hydrostatic driving force of recharge water entry into the system
through topographically elevated exposures would be anticipated to promulgate
groundwater flow through the coals preferentially seeking the most permeable, oxidized,
fractured and cleated zones, which in general reside relatively near the
outcrop. Fossil formation pressure
release augmented by surface recharge would be manifested by seeps at
topographic lows such as springs, and lend a contribution of sub-flow to
streams and rivers where coal outcrops/subcrops were intersected. With the onset of CBM water production,
water was withdrawn from coalbeds within the Basin, locally reducing reservoir
pressure around each gas well. This
pressure reduction was paramount in enabling gas desorption. The reduced pressure gradient emanating
outward from the gas well(s) as the water is produced would be largely
dependent on the permeability of the coal relative to water. The artificially induced pressure gradient
would tend to draw surrounding water toward the lower pressure. Dependant upon
migration pathways present and the permeability along these pathways,
freshwater recharge could be captured and historic pathways of groundwater flow
altered. If recharge were insufficient
to replenish water withdrawn by basin fringe gas wells near the outcrop,
reversals of groundwater flow might occur with subcrops being recharged by
surface flow from perennial rivers/streams. As pressure gradients from a host
of gas wells coalesce, a fresh-water
front would progress basin-ward. If
sufficient water were produced from shallow coalbed(s) proximate to the
outcrop, a major portion of the fresh
water available might be captured, thus pre-empting or slowing the basin-ward
migration of the fresh-water front.
(Basin-ward wells could only draw stored water available within their
respective pressure depletion zones of influence until the water surrounding
them was depleted.) Two-phase (water
and gas) flow dynamics can decrease relative permeability to water. Therefore, the zone of influence of some
interior basin wells may be very limited with respect to the ability to draw
water from the basin fringes. Considering
these scenarios, gas wells may provide a shield between basin-fringe wells and
interior basing wells starving basin interior wells from fresh water
access. (See next page for
illustration.)
Water
chemistry data is useful in understanding patterns of groundwater flow. In contrast to Basin-interior produced water
characterized by high TDS (total dissolved solids) associated with longer
residence time are produced-water chemistry data from Fruitland coalbed gas
wells on the northwestern edge of the Basin, the latter showing lower
concentrations of chlorides and bicarbonates-a signature of fresher water. Produced water data indicates the presence
of a relatively fresh water (low TDS) plume along the western end of the Indian
Creek drainage and also along the northern portion of the Basin within 4-5
miles of the Fruitland coal outcrop.
The occurrence of relatively fresh water in the coalbeds close to the
Basin fringes indicates connection to a fresh water source, presumably the
Fruitland outcrop. These fresh water zones, areas with moderate communication
with the outcrop, and areas with limited interaction with areas of recharge
need to be identified to calibration the current computer modeling effort so
that the model may reliably reflect the nuances of known basin hydrology.
Water
character including chemistry (age-dating and anion/cation relationships) and
TDS concentrations can be affected by mixing.
Age dating of water relies on assessing the amount of radioactive decay
within a dissolved mineral. Unfortunately,
water tends to take on the character of the host rock, slowly equilibrating
through time, and will ultimately reflect the age/character of the rock. Accurate age dating is complicated by the
introduction of uranium-laden volcanic ash deposited as thin layers within the
coal beds. Mixed waters tend to reflect an age proportional to the relative
quantities of water of different antiquity.
Neither could water with an initial high TDS be recognized, once diluted
by fresher water. Mixing obscures the individual
character of all waters mixed because the results of testing can only determine
the diluted concentrations of any given parameter. Once these limitations are recognized, different types of water
chemistry analyses taken together can provide useful information, but taken
singly may prove misleading. Iodine and
chloride ion content in produced water may yield patterns suggesting that areas
with high iodine and chloride concentrations correlate with older water that
has been in residence with the coals for extended periods of time. Higher TDS/older water indicates slower
migration rates, lower permeability of the coal locally, and diminished
likelihood of significant interaction with the coal outcrop. Low chloride/TDS values likely correlate
with progressively younger or mixed water.
Similar water quality analyses for total dissolved solids and
cation-anion concentrations in produced water in the western portion of the
Basin also suggest freshwater plumes, and freshening of produced water over time
progressing basin-ward to some extent.
In specific areas, a line of demarcation seems apparent where produced
water is freshening toward the Basin fringes and gaining in salinity with
production basin-ward. Water character
boundaries may not necessarily represent static and impermeable barriers to
flow, but merely current conditions.
Water of antiquity may suggest areas not in effective dynamic
communication with the outcrop, or alternatively may depict the current
inter-face position of a progressive fresh-water-front moving basin-ward. Neither do water chemistry variances
necessarily signify that the coalbeds of the Fruitland Formation are not
laterally contiguous. Dynamic
relationships may become evident only after further coalbed methane/ formation
water production and a time factor allows equilibration throughout the
formation between respective coalbeds.
Gas wells drilled/producing in areas of fresh water are construed more
likely to affect the outcrop than gas wells in areas depicted by more saline
water. Monitoring over time may provide
useful parameters enabling computer modeling to forecast effects of continued
or increased CBM production.
There has been an effort to definitively correlate
water-bearing coalbeds in the respective gas well(s) to specific coal beds
expressed at the surface. The Southern
Ute Indian Tribe and the Colorado Geological Survey have mapped surface
expressions of the Fruitland coalbeds within the Southern Ute Indian Reservation
and have drawn preliminary correlations between the outcrop and coalbeds
penetrated by nearby CBM wells. The
Colorado Geological Survey has recently completed mapping the Fruitland
Formation coal outcrops north of the Southern Ute Indian Reservation with the
intention of correlating surface and sub-surface coal beds. Unfortunately, the
depositional environment of the coals may preclude extensive correlation of
more than a few miles. Recognizing the
depositional environment with transitional regressive shoreline advances and
retreats, the predecessor coalbeds tended to pinch out with later replacement
by another coal horizon. Which coals
are contiguous in three-dimensional relationships or communicate between layers
is debated. Current modeling assumes
that the coal packages act as a unit on a regional scale.]
The monitoring goal of the 3M Project includes the
establishment of eight water monitoring well clusters proposed north of the
Southern Ute Indian Reservation. Four
additional clusters - apart from the 3M Project - are being considered by the
Southern Ute Indian Tribe for installation within the exterior boundaries of
the Southern Ute Indian Reservation (See
Appendix B: Maps and Cross-Sections 15.)
Each cluster would contain 2-4 water-monitoring wells, each well
assessing one exclusive coal zone comprised of one to a few closely related
coal seams. These wells would be
monitored for temperature and water level/bottom hole pressure. Temperature may be an indirect indicator of
fresh water influx (decrease in temperature) or water withdrawal (the heat of
wetting). Water level/bottom hole
pressures will help evaluate the drawdown effects on the piezometric surface by
production of water from basin-ward coal wells.
The
modeling goal of the 3M Project will evaluate the hydrologic and gas reservoir
potential of the Fruitland coalbeds.
Pressure data gathered from producing gas wells and monitoring wells
will be used in a modified reservoir to determine past and future effects on
the outcrop and feasibility, both economic and environmental, on continued
development including infill drilling within the Ignacio-Blanco Field. The hydrologic portion will use the US
Geological Survey “MODFLOW” program to calculate recharge along the outcrop,
discharge out of the Basin from production activities and other influences that
would affect water quality and quantity in the San Juan Basin.
In
addition to water monitoring wells, other monitoring tools include thermal
infrared and near infrared aerial photography of the entire Colorado portion of
the San Juan Basin Fruitland outcrop along the Basin rim. These aerial photos document current
conditions, changes from prior photos, and may identify zones of aberrant heat
patterns. As a predictive tool, thermal
infrared photos may be used to distinguish areas of potential spontaneous
combustion in the near-surface coalbeds.
Additionally, temperature data loggers installed in the Fruitland
coalbeds may provide early warning of coalbed heating. Ongoing monitoring of soil vapor,
groundwater and produced water quality will augment the monitoring effort.
