HARM RATING SYSTEM

The relative harm caused by a specific pollutant is difficult to assess. There are health effects but also qualitative aspects such as visibility and beauty (forest damage by acid rain). Ironically a major wind project was just rejected based on the degradation to the beauty of the area. Mercury is the present pollutant de jour. But who is to say that concerns about eating fish will be at the top of the environmental priority list forever. With the coming PM_2.5 regulations and the claims that fine particulate kills 50,000 Americans per year, this pollutant may shoot to the top.

There are many instances where reducing one pollutant increases another. Some air pollution control processes require up to five percent of the electrical output of the plant. This means a five percent potential increase in pollutants not reduced by the process. One highly effective mercury control process increases carbon fine particulate. NO_x control can easily increase sulfuric acid emissions.

Every choice these days is a multi-pollutant decision. Very few options provide the same degree of reduction of all the pollutants. So the decision maker must decide which pollutants are more important than others. Anticipating future allowance prices is valuable. But it is not certain that allowance trading in certain pollutants will even be an option. We contend that one factor that should go into the decision making is the total pollution quotient of each option.

We are providing two sets of analyses. The second set deals just with air toxics. This set deals with the entire range of pollutants. Figure l shows that U.S. emissions of pollutants from power plants are 27 billion pounds. Mercury is only 96,000 pounds. So it is included under “other toxic chemicals.” In Figure 2 we have created a pollution potential factor using the Lesser Quantity Emission Rate (LQER) numbers from an EPA draft. Where toxic chemicals did not have an LQER we assumed that the 10 tons/yr threshold was valid. Since criteria pollutants such as NO_x and SO_x are considered major sources when the emissions reach 100 tons/yr, we utilized this number for those two pollutants. We reasoned that, because of the associated deaths, fine particulate rated at least a 10 and maybe it should be even lower.

We then determined the average emissions of a 300 MW boiler by simply dividing the 300,000 MW of capacity by 1,000. We then divided the yearly emissions for this average boiler by the LQER and obtained the “Average plant Pollution Equivalent” shown in Figure 3. We used two different LQERS for mercury (these are explained in the later analysis) to compare the different impacts on the total. Then in Figure 4 we reduced the pollution equivalents to percentages.

Immediately we see that NO_x and SO₂ are the largest contributors. Mercury is third but fine particulate is almost as much using mercury LQER of 0.001. It is much more significant than mercury at the higher mercury LQER. Most people do not realize that a high percentage of power plant particulate emissions are calcium and sodium and not SiO₂. These will react in the atmosphere to form PM_2.5 sulfates and nitrates. Lead, chromium, arsenic and nickel combined are either nearly as important as or four times more important than mercury.

There is the question of 95 percent SO₂ removal (required in some consent decrees) versus the lower efficiencies required typically. From Figure 4 it is seen that an additional five percent SO₂ removal has the same benefit as a 10 percent mercury reduction in the worst case scenario.

Some are advocating for a mercury cap of five tons at an early date. This will require 90 percent mercury removal. It will be much less costly to achieve 80 percent removal. For a fraction of the cost of achieving the additional 10 percent mercury removal one can reduce the total pollution potential.

Why are we even talking about all the utility HAPS?

Analysis of Air Toxics in Figure 5, it is plain that only a few air toxics account for most of the weight of toxics emitted. However, in Figure 6 it is shown that there is a huge difference in toxicity between HAPS. Figure 7 shows toxicity based on 2 LQERs for mercury. The average plant has a toxic quotient of 242,000. It should be pointed out that this quotient already reflects 90 percent removal of the metals other than mercury. So the quotient with no air pollution control equipment could easily be 800,000. If one were to require 90 percent reduction from this uncontrolled level then the toxicity quotient or toxic equivalent would be 80,000. If mercury were treated separately, the raw quotient would be 700,000. Meeting a 70,000 quotient limit could easily be accomplished with the normal particulate equipment and SO₂ scrubber.

The original EPA draft listed mercury with an LQER of 0.1. We arbitrarily changed this to 0.001 to reflect the recent focus on mercury as the highest priority pollutant. In Figure 8 we compare the mercury impact with the LQER both at the 0.001 level and at 0.01 (the toxicity for lead and chromium). As shown in Figure 8, mercury is either 7 percent of the toxics to only 5 percent.

These comparisons should give legislators pause before requiring a 90 percent mercury reduction. Reducing mercury the extra 10 percent could add as much as 10 mils/kWh to electricity cost. In the base case, an extra 10 percent reduction of HCl, chromium and lead would accomplish the same toxicity reduction but at a far lower cost (less than 0.5 mil/kWh). In the case where mercury is assigned a lower toxicity, the extra 10 percent mercury reduction only reduces total toxicity by 0.5 percent.

Figure 9 provides the quotient factor and total quotient for each of the many HAPs emitted by power plants. These are based on annual emissions nationwide as reported in the TRI inventory.

Figure 9.  U.S. Utility Air Toxics Analysis