Page 189 - Geothermal Energy Renewable Energy and The Environment

Generating Power Using Geothermal Resources 175

The loss of this energy is due to the hydrologic framework of the system. In essence, liquid water
cannot flow into the reservoir and be vaporized to steam at the same rate at which steam is being
extracted. Dry steam systems have low rates of natural recharge because liquid water must flow
around the impermeable zone that caps the system and then migrates into the reservoir region. In
order to maintain steam production at a constant level, the steam extraction rate must be managed so
that it occurs at the same rate as the natural system is recharged or liquid water must be injected into
the reservoir at a sufficient rate to produce extractable steam to balance that which is removed.
Predicting the rate at which steam can be removed during the early stages of development of
a dry steam field must be based on experience rather than theoretical models, at least for the cur-
rent state-of-the-art. This reflects the fact that the volume of the reservoir is usually unknown, the
permeability of the rock is usually insufficiently characterized to portray realistic steam flow paths,
and the actual natural recharge rate of the system is unknown. The historical record for The Geysers
documents that steam was extracted from the system faster than it was being replenished during its
early development, as indicated by the prolonged drop in per well steam production, even though the
overall steam production was increasing. These facts made it apparent that injection of water would
be a necessity if power production were to be maintained at The Geysers.
Beginning in 1969, injection of condensate from the generating complex was initiated. Wells that were
marginally productive were reconfigured as injection wells and the condensate was injected through
them into the reservoir. On a mass base, the injected fluid was equivalent to 25–30% of the extracted
steam (Brauner and Carlson 2002). Additional volumes of water were obtained from local streams to
supplement the condensate, increasing by a few percentage the total liquid mass injected into the reser-
voir. However, these fluid volumes were insufficient to prevent further decline in steam production.
In 1995 a project, called the South East Geysers Effluent Pipeline Project (SEGEP), was initiated
to inject treated wastewater from the city of Santa Rosa, 26 miles to the east, into the steam field. A
pipeline (see Figure 9.14 for location of the pipeline) was constructed that was capable of providing
29.5 million liters (7.8 million gallons) a day to the southern part of the field for injection into a com-
plex of 7–10 wells (Brauner and Carlson 2002). As is clear from Figure 9.16, wastewater injection
stabilized steam production. Additional wastewater injection capability, amounting to an additional
41.6 million liters (11 million gallons) per day is currently under development.
Water injection also had the benefit of reducing H S in the fluid. Prior to the injection program
2
in 1995, the concentration of H S in the steam was increasing at various wells in the area. This
2
reflected the impact of reduced dilution by steam in the reservoir. However, within a few months
of initiating the injection of wastewater, the H S concentration dropped by nearly 20% due to the
2
dilution of the noncondensable gas component in the reservoir.
However, it is important to note that there is not a simple relationship between the mass of water
injected and the mass of steam recovered. The total rate of water injection into The Geysers in 1999
was approximately 3.83 million kg/hr, most of which was distributed among a few wells in the
southeastern area. At the same time, approximately 6.6 million kg/hr of steam was produced from
the entire field. In other words, a mass equivalent to 56% of the total mass of steam produced in the
entire field was injected into an area accounting for less than 20% of the geothermal resource area.
Such a situation suggests that the subsurface flow paths of water and steam must be complex.
Figure 9.17 provides a graphic demonstration of the subsurface behavior of fluid injected into the
system. The figure shows the distribution of microseismic events around one of the injection wells.
Microseismic events are short releases of elastic energy by the rock, due to a number of possible
mechanisms. Cold water injected into hot rock can cause rapid cracking due to contraction, which
can be detected by sensitive seismometers. Other detectable events that can occur in these settings
are the forced opening of old fractures in the rock, formation of new fractures, and the release of
preexisting stresses in the rock that develop in response to tectonic forces. Whatever the specific
mechanism, the location of the microseismic events provides a picture of the possible migration
path of the injected water. As is clear in the figure, much of the microseismic activity occurs well
below and away from the production wells. This implies that at least some portion of the injected