In my last few posts, Down
by the Water and Take
Me to the River, I mentioned that my office looks onto the Merrimack River
and the upstream Amoskeag hydroelectric operation that has been producing
electricity for the past ninety years. From
this view, I often take note of river flows and whether the water is spilling
over the top of the dam wall, as in the photograph below. In this post, I
discuss the variability of river flows and how hydro plant electricity outputs
are very dependent on these. I also look at the capacity factors of the
Merrimack River hydro operations and compare them to national averages.
Photo: PSNH Flckr Photostream
Weather and precipitation are the most important variables in hydro electricity operations because these determine river flow. I dug up some relevant information about Merrimack River flows near my office from the United States Geological Services (USGS). The chart below shows the river flows at the Goffs Falls monitoring point, which is just downstream of the Amoskeag Dam. The jagged blue line shows the data for 2013 and it is surprising how variable the flow is from day to day. The orange dots show the average flowrates over the past 76 years. This historical data shows that river flows generally rise during April to June due to snow melt, reaching flows that are four times the average value, and then drop off considerably during the dry August to October period, to about one quarter of the average. Interestingly, the summer of 2013 was a wet one, as indicated by the higher-than-average river flows during this period.
Source: USGS
A simple relationship dictates the generating capacity of a
hydro operation:
Power = Constant x Flow x Height
In an operation such as Amoskeag, the height (or head) is essentially fixed because this is a constant-level run-of-river operation. However, since Merrimack River flows do vary, I took a look at the 2013 monthly river flows and compared them to monthly electricity production (measured in MWh) for the two larger operations on the Merrimack River. These are plotted in the figure below on the left, with river flows in the green bars, and Amoskeag and Garvin Falls electricity generation in blue and red, respectively.
As expected, high river flows, particularly during the April
snowmelt or the wet July of last year, generated higher amounts of electricity.
Low river flows, such as in the dry
months of August to October, were associated with lower generation rates.
The chart on the right plots energy generation against river flows for the Amoskeag plant. I was somewhat expecting a 1:1 linear relationship and was initially surprised to note how generation tended to start leveling off at high flowrates. However, the PSNH hydro folks pointed out that the maximum flowrate through the Amoskeag turbines is 5000 cubic feet per second (cu. ft/sec), so one would expect to see generation level off above this flowrate. Moreover, an average monthly flowrate of greater than 5000 cu. ft/sec does not necessarily mean that flow rates are higher than 5000 cu. ft/sec for 24 hours a day – there may be periods when it is substantially higher and then there are periods of lower flows.
The chart on the right plots energy generation against river flows for the Amoskeag plant. I was somewhat expecting a 1:1 linear relationship and was initially surprised to note how generation tended to start leveling off at high flowrates. However, the PSNH hydro folks pointed out that the maximum flowrate through the Amoskeag turbines is 5000 cubic feet per second (cu. ft/sec), so one would expect to see generation level off above this flowrate. Moreover, an average monthly flowrate of greater than 5000 cu. ft/sec does not necessarily mean that flow rates are higher than 5000 cu. ft/sec for 24 hours a day – there may be periods when it is substantially higher and then there are periods of lower flows.
In one of my recent posts, Down
by the Water, we noted the simple mathematical relationship between energy
and power:
Energy =
Power x time.
Applying this relationship to the Amoskeag operation, which has a nameplate capacity of 16 MW, and assuming 30 days per month and 24 hours per day of operation, the maximum monthly generation from the Amoskeag Dam can be calculated as
Energy = 16 MW x
30 days x 24 hr/day = 11,520 MWh
This is pretty close to the maximum monthly generation output on the chart above.
I was also surprised to note that the power generation of Garvin
Falls was half that of Amoskeag although its generation capacity (12 MW) is 75%
that of Amoskeag (16 MW). Calculating the total generation for both operations
for 2013, I noted that Amoskeag produced 66% of its maximum electricity output (also
termed its capacity factor), whereas
Garvin Falls produced only 44%. It turns out that Garvin Falls is a more
troublesome operation because it has a gatehouse and a narrow channel for a head
race that is used to direct water into the power houses. River-borne debris,
such as leaves, branches, trees, etc., build up behind the gatehouse and
restrict flow to the channel, which significantly compromises the steady
generation of electricity from this operation. Regular maintenance, involving
the removal of debris from the screens where the water enters the power house,
is required.
The Energy Information Agency, EIA, produces an average
annual capacity factor for all hydro operations across the US. In 2013 it
was 38.1%, which is much lower than I would have anticipated. I was expecting
capacity factors for hydro operation to be of the order of 80% or so but the annual
data from 2008 to 2013 shows US capacity factors ranging from 37.2% to 45.9%. The
national data does indicate that the Garvin Falls and particularly the Amoskeag
operation have capacity factors greater than the average US hydro operation.
There are several reasons for these lower-than-expected capacity
factors for hydro operation:
- Precipitation and river flows are variable and the maximum flow of water that the generators can handle is not always available.
- River-borne debris and winter ice can at times significantly compromise water flows into the generating units.
- The generating units need to be slowed or shutdown for periodic maintenance.
- Even though hydro plants are generally the lowest cost producers of electricity when selling into the wholesale markets, they can be underbid by other generators, particularly heavily subsidized wind operations which will sometimes even pay to produce electricity. At times like these, there might not be any call for hydro power and the units are shutdown.
PSNH owns and operates several hydro operations in NH. Those listed below are owned and operated by Northeast Utilities, the parent corporation of PSNH.
Based on recent documentation submitted to the NH Public Utilities Commission, hydro operations will be responsible for about 11% of the ~3,016,000 MWh of electricity that PSNH is planning to generate from its own facilities in 2014. These are PSNH’s lowest cost electricity generators and are thus an important asset to keep in operation and perhaps even consider expanding, even though the impacts of new or expanded hydro operations could be considerable and permitting could be extremely difficult. We have to recognize that there is a price to be paid for every energy source we use but, unlike fossil fuels where every ton of carbon dioxide we dump into the atmosphere intensifies the green-house effect, hydro plants will be still generating electricity one hundred years from now and will still not be pumping carbon dioxide into the atmosphere.
An important debate regarding these facilities is presently underway
in NH (discussed in Should
I Stay or Should I Go). In order to
complete the electricity deregulation process in New Hampshire, it has been
proposed that PSNH should be compelled to sell these hydro generating
operations, along with their wood- and gas-fired operations and the large coal-burning
plant on the Merrimack River in Bow. However, with electricity prices shooting up
this winter and with PSNH customers, for the time being at least, somewhat
shielded from these increases, this does give one pause for thought and to consider that ownership of generating operations may perhaps have some
benefits. This is certainly a topic I will return to in a future post.
Until next time, remember to turn off the lights when you
leave the room and, if it is raining, contemplate that the river’s gonna rise*
and the hydroelectricity output will increase.
Mike Mooiman
Franklin Pierce University
mooimanm@franklinpierce.edu
(*River’s Gonna Rise
– An instrumental tune by Patrick O’ Hearn, a LA bass player and electronic
musician who has had a long and varied career, including stints playing bass in
Frank Zappa’s band, the New Wave group, Missing Persons, as well as releasing
over a dozen solo albums featuring electronic and ambient music. He is well
known for his film scores and in the past few years has been playing bass in
John Hiatt’s band. Here is
the title track from Patrick O’ Hearn’s 1988 album River’s Gonna Rise.)
