One system was constructed on Loyalsock Creek in Sullivan
County using an oxic limestone drain
(OLD) design. This is a buried limestone trench, similar to an ALD,
except that the influent AMD contains oxygen (Cravotta, 2000). Three
OLDs were constructed in parallel to treat a portion of a very high
flow (4,000 gpm (254 L/s)) discharge with very low acidity and metals
levels. Mass balance calculations showed that adequate treatment of
a portion of the discharge should provide enough excess alkalinity to
neutralize the entire discharge. A decision was made to construct three
drains in parallel so that flows could be individually adjusted going
into them to determine the effect flow has on alkalinity generation.
Construction of this system was completed in November 1999. Sampling
has shown that while the system is discharging net alkaline water, the
Olds are not generating the alkalinity needed to effectively neutralize
the entire discharge. Water discharging from the Olds has alkalinity
concentrations ranging from 20 to 60 mg/L, while 250 mg/L was the expected
alkalinity concentration based on empirical evidence from constructed
sites. This same condition is also being observed at a vertical flow
wetland (VFW) constructed on a nearby discharge into Loyalsock Creek
(see additional discussion under the Vertical Flow Wetland heading).
The limestone required for both systems was a minimum 90 percent CaCO3
equivalent, with a maximum of five percent MgO.
It was thought that inadequate retention time could be a factor, but
this doesn't appear to be the case. While the Olds are providing only
five hours of retention time, the nearby VFW is providing approximately
15 hours of retention time with the same results. There is some concern
that low acidity raw water (both discharges averaged less than 50 mg/L
acidity) may be unable to dissolve the limestone to the extent that
more highly acidic water does, even when influent pH is similar. One
interesting observation at this site is that increasing flows into the
Olds, thereby reducing contact time, has very little effect on the concentration
of alkalinity generated. It appears that a greater loading of alkalinity
can be introduced to the receiving stream simply by increasing the amount
of water being treated and decreasing the amount being bypassed. Flows
are gradually being increased to determine the optimal flow rate to
direct into the system. In the meantime, a post-construction macro invertebrate
survey has shown significant improvement when compared with pre-construction
data, so stream recovery is already occurring.
Return
to AMD Stakeholders' page