Home

Purpose & Principles

Products & Projects

People & Partners

Contact

We are providing renewable energy consulting services to various clients.
As part of an effort to support sound decision-making by clients and consumers, we are writing a series of white-paper analyses to clarify (and possibly correct) some perceptions about energy generation, transmission, distribution, and consumption.  An example is given below.

Comparing Renewable Energy Environmental Benefits

It is difficult to put the environmental benefit of a single renewable energy generation facility (e.g., a PV system) into human terms.  We want to understand, “what good does it do?”  Such understanding can be aided by comparisons with familiar analogs in our daily lives, but such comparisons must be done with care.  It is common practice to use “conversion factors” from kWh of electricity produced to tonnes of CO₂ not emitted or to barrels of oil not imported.  Regrettably, this practice is often applied incorrectly, with the results ranging from simply misleading to completely misguided.

Summary

For the near-term (many years) and with few exceptions, renewable energy projects in the U.S. will displace natural-gas generated electricity, rather than coal or nuclear baseload, at an avoided emissions rate of 0.6 kg/kWh.

“Average” CO₂ Emissions Reduction

The relationship between electricity generation and CO₂ emissions is of obvious interest with respect to climate change.  Converting from kWh of electricity generated to kg of CO₂ emitted is not straightforward.  More importantly, converting from kWh of electricity not consumed to kg of CO₂ not emitted is much harder.

The heating value of fossil fuels covers a broad range.  Because the chemical composition of such fuels varies, the ratio of CO₂ emitted per unit mass burned depends on the makeup of the specific feedstock.  For example, the heating value of coal varies more than a factor of two across its many varieties, from well under 20 MJ/kg for lignite to over 32 MJ/kg for anthracite and most bituminous coals.  The lower heating value for natural gas is 50 – 55 MJ/kg depending on the percentages of its non-methane components.  There are various “standard” values used for converting from mass burned to energy and CO₂ released, but these must be used with caution.

Converting combustion heat to electricity incurs losses, depending on whether coal is pulverized or gasified before combustion or whether natural gas is burned in a turbine or a steam boiler.  The efficiency of the average U.S. coal-fired power station is about 33% and the resultant “average” amount of CO₂ such generators emit is ~0.9 kg/kWh (2.095 lb/kWh in 1999, neglecting year-to-year variation).  As noted in the previous paragraph, the emissions of a particular generator or of those in a particular region may be significantly different from this average.  Natural gas fired plants emit ~0.6 kg/kWh (1.321 lb/kWh in 1999, again neglecting year-to-year variation).

But the primary error in discussing “average” CO₂ emissions per kWh of electricity generated is that the end user does not consume “average” electricity.  CO₂ reductions that result from the displacement of utility-delivered power by on-site generated renewable power depend on the energy source mix of the specific user’s utility.  In the limiting case, a customer who has already signed up for the local utility’s 100% green power option would expect to provide no additional environmental benefit by installing a PV system on his building’s rooftop. 

California’s energy mix is very different from that of the U.S. as a whole, as illustrated by the following charts.  These mixes are not only functions of different available energy sources and power generation stations, but also vary year-to-year as hydroelectric water stocks and fossil fuel prices fluctuate. 

However, since a net-metered system offsets generation of a particular utility, even the statewide average isn’t the right metric.  The CO₂ content of Los Angeles DWP’s power content (2006) is worse than the U.S. average, while neighboring Southern California Edison’s power content (2006 projected) is better even than the California average.  The range is fairly pronounced, with PG&E coming in on the low end (of the various examples I’ve studied) at under 0.3 kg/kWh, and with Burbank’s and Pasadena’s municipal utilities each coming in at over 0.7 kg/kWh.  There are probably utilities in the Eastern U.S. that are so coal-dominant that they get close to 0.9 kg/kWh (or worse if they’re largely dependent on lignite).

For a wind farm that delivers power to the grid, the situation is even more complex than that of a single-site net-metered generator.  The CO₂ generation displaced by the wind farm will depend on which utilities buy its power and what other generation is not used as a result of buying the wind power.  Of course, this in turn depends on when the wind power is generated, and this leads to the next observation.

Incremental CO₂ Emissions Reduction

The problem with the preceding discussion is that statistics lie.  A kWh generated by a grid-tied PV system doesn’t displace an “average” kWh, regardless of whether it’s a national, state, or utility-specific “average” kWh.  The energy generated by the customer’s system is immediately consumed, either on-site or by some other relatively local consumer.  During the evening, the net-metered system owner still consumes fossil energy (which probably has a higher percentage of coal-derived content than the daily average since most coal is used for baseload generation).  Thus, the CO₂ offset is not that of the utility’s average emissions.  Instead, the emissions of a marginal kWh are eliminated – those emissions resulting from the generation of power required for incremental demand at the time that the PV power was injected into the grid.

The figure below shows how LADWP ramps generation from its various resources up and down during a typical (representative) summer day (LADWP Integrated Resource Plan 2006 (Draft),  p.22).  As can be clearly seen, coal and nuclear generation are constant during the day.  During the period of operation of a PV system, about 8:00 AM to 6:00 PM, marginal demand is served by hydro, pumped storage, and natural gas.

Clearly, when marginal demand is served by natural gas generation, the CO₂ emissions avoided by the PV system will be the ~0.6 kg/kWh associated with natural gas fired power stations.  When the marginal demand is served by pumped storage, this storage is replenished during the early morning hours (2:00 to 7:00 in the figure) with marginal generation at that time – which is based on natural gas.  Finally, if the marginal daytime demand is served by large hydro plants, CO₂ emissions are indirectly avoided by allowing the water saved to be used another day – generally with the effect of displacing natural gas operations.  Thus, in all these cases, CO₂ emissions are avoided at a rate of ~0.6 kg/kWh.

This analysis is by no means universal, but it is probably very representative of most utilities’ generation scheduling strategy.  Remarkably (and completely coincidentally), the value obtained by this approach is virtually identical to the U.S. average emissions rate – which is convenient, inasmuch as many people will compute the “right” answer for completely the wrong reasons…

(The similarity is almost spooky, differing by only one digit in the EIA report, “Carbon Dioxide Emissions from the Generation of Electric Power in the United States,July 2000.”   On page 5, it says, “The output rate for CO₂ from natural gas-fired plants in 1999 was 1.321 pounds CO₂ per kilowatthour [0.600 kg/kWh].”  On page 7, it says, “The national average output rate [was] 1.341 pounds CO₂ per kilowatthour [0.610 kg/kWh].”)