Environmental Engineering Reference
In-Depth Information
generation capacity. Identifying days when wind generation resulted in the
cycling of coal units allowed for a precise understanding of the emission
impacts. The gravity and frequency of these events increased as more wind
generation was introduced to the system. This mirrors the results found
on the PSCO system, supporting the theory that increased rates of cycling
arise from the incremental integration of wind generation. Furthermore,
these wind-driven, coal-cycling events resulted in significantly more SO
2
and NO
X
emissions than if wind generation had not been utilized. The same
results were found on the PSCO system. Not only does wind generation not
allow ERCOT utilities to decrease SO
2
, NO
X
,
and CO
2
emissions, it is directly
responsible for creating more SO
2
and NO
X
emissions and CO
2
emission sav-
ings are minimal at best.
GeneralConclusionsandFutureOutlook
Our studies detail the surprising conclusion that the use of wind energy in
the PSCO and ERCOT contexts results in increased emission rates for SO
2
and NO
X
and, in the case of PSCO, CO
2
. The mechanism driving increased
rates is the need to cycle coal facilities to accommodate wind—a must-
take resource under the respective states' RPS mandates. When wind gen-
eration comes online, generation from coal (and natural gas-fired plants
is curtailed until the wind subsides, then nonwind generation is again
ramped up to meet demand. Cycling coal units in this manner drives their
heat rates up and their operating efficiencies down, emitting more SO
2
,
NO
X
, and CO
2
than would have been emitted if the units had not been
cycled.
Two caveats must be understood when interpreting these results. First, we
found no instances in which PSCO violated any of its air permits as a result
of cycling coal. Neither PSCO case study indicated that PSCO's emissions
exceeded its permits. Furthermore, the study authors are not suggesting that
PSCO violated its permits in extrapolating the case study results to estimate
annual emissions. The second caveat pertains to the data. For the ERCOT
analysis, hourly generation data by plant and fuel type including wind was
available. Thus, it was possible to precisely identify wind events based on
a sudden decline in coal generation coupled with a simultaneous increase
in wind generation. For PSCO's territory, it was not possible to define wind
events with the same precision because PSCO does not release hourly gen-
eration data for its wind resources. Subsidiary conclusions based on our
analysis include:
Duration
— Cycling coal-fired power plants has short- and long-term
impacts. Studies of the interactions of coal and wind often mention the cycling
issue, but generally discuss the impacts in a very narrow context—the duration
