mental Protection Agency had proposed regulations to control nitrogen oxides emissions from stationary sources, the proposals will be withdrawn.

As for motor vehicle emissions, a 90-percent reduction in nitrogen oxides is, in my judgment at this time not necessary. We are undertaking further analysis to determine exactly what such a standard would be. Given the need for this further analysis, coupled with the ongoing studies of the health effects of nitrogen oxides, it is my judgment that a new nitrogen oxides emission standard should not be legislatively mandated, but rather that the Environmental Protection Agency should continue setting the standard under the provisions of section 202(a) of the Clean Air Act, in lieu of the present requirement for a 90-percent reduction in 1976 under section 202(b). We have drafted an amendment which would accomplish this purpose and request that it be considered by the committee.

This concludes my prepared remarks, Mr. Chairman. I would be happy to answer any questions you might have.

Mr. Chairman, I do not make a statement like this lightly because I believe it is very important before this committee to change any such legislative mandate standards as applies to nitrogen oxides, that there be a full public hearing of what we are recommending.

Starting last summer and intensifying in the late fall and early winter, I several times, as did the Deputy Administrator, Mr. Fri, made statements to the same effect as it made in this statement before congressional committees and in various public presentations.

The impact of those statements was I think to say the least largely ignored or missed by the public and it is for that reason that I would recommend and will carry out the recommendations myself that we publish not only this statement, but the analysis behind it which has led us to this conclusion, in the Federal Register requesting the public over a period of time, say 30 to 60 days, to comment on this analysis and give us the benefit of the best scientific advice we can get outside of the Agency and from the public at large as to whether our analysis in fact is correct.

At the end of that period of time, I think these comments should be forwarded to the committee along with this continued recommendation if our further analysis warrants that for whatever action the committee would want to take.

There is some degree of urgency that this matter be considered by the committee because where the nitrogen oxide levels are finally set has a tremendous impact on the kinds of technology that can be available to control hydrocarbons and carbon monoxide.

So I think we ought to set a schedule for consideration of this problem with some degree of urgency so that there can be a signal given to the automotive industry as to precisely what standards they have to meet, and what they should shoot at.

For that reason. I am recommending since we have not been able to generate very much public comment by any other method that we have adopted that we do publish this in the Federal Register and request public comments which, of course, will be made available to the committee.

Senator MUSKIE. I am sure there are a lot of questions on the part of the committee. Would you like to proceed with the presentation of the Doctor?

Mr. RUCKELSHAUS. Yes I think it would be beneficial to the committee to see what has led us to this tentative conclusion and the analysis that lies behind it.

Senator MUSKIE. Let me ask just two questions, then.

Maybe these will help lay the basis for their presentation. Once you spoke of an analysis that you would publish in the Federal Register. Is that analysis now available? Is this the analysis they are going to make?

Mr. RUCKELSHAUS. This is in summary form the analysis. We do not have a document that is presently ready to be published in the Federal Register. We do have an almost completed document on the measurement problem itself; that is obviously an integral part of this analysis.

That along with the conclusions that has led us to this is what we will publish and we will, of course, make that available to the committee also.

Senator MUSKIE. Will there be a document describing the analysis at some point?

Mr. RUCKELSHAUS. Yes. There will be.

Senator MUSKIE. When? Can you give us the time on that?

Dr. GREENFIELD. As a matter of fact, Mr. Chairman, I think a preliminary document which covers most of the points on the measurement analysis has already been made available to your staff.

Senator MUSKIE. Is this the one that you pointed out to me earlier this morning?

Mr. RUCKELSHAUS. No. That is not. That is a different document. Senator MUSKIE. This is the assessment of the analytical document available for the determination of nitrogen oxide and ambient air. Dr. GREENFIELD. That is right. That is the one prepared on February 24. It is being updated. It is the basis for the measurement portion of the analysis.

Senator MUSKIE. The second question I would like to ask refers to your statement in which you say that over the past year we have been measuring nitrogen dioxide levels by various measuring methods at some 200 sites across the country, including sites in the 47 air quality control regions where we originally believed that the national standard was being exceeded. A full report on this study will be available very shortly.

Is that full report essential at all to the conclusion which you have stated here in this statement this morning?

Dr. GREENFIELD. Yes. It is because if those measurements which lead you to what we might call a reclassification of the 47 air quality regions.

Senator MUSKIE. Does the fact that the Administrator has reached his firm conclusions as stated this morning indicate that he has had access to essentially all of that information?

Dr. GREENFIELD. Yes. He has.

Senator MUSKIE. So the preparation of the report is simply a matter of typing and putting it in form. The substance of that report is in part the basis of your conclusions this morning?

Mr. RUCKELSHAUS. Yes.

Senator MUSKIE. I will withhold any further questions until we have had the full presentation. Do other members of the committee have questions?

Dr. GREENFIELD. Because of my close quarters, I wonder if it might be better if I remained seated, and you can all see the charts? Senator MUSKIE. I think that would be fine.

Dr. GREENFIELD. The first chart described is the four basic methods that have been considered. The first is a Federal reference method, four columns, the sampling technique used there is a 24-hour bubbler. That means whether or not it happens in 24 hours or whether it is continuous measurement, which means you take measurements every few minutes during the day.

This third column is the question of whether or not this method had been field tested when the standard was set. The fourth column is a set of remarks which I will get to.

The first row is the current Federal reference method, the so-called Jacobs-Hachheiser technique. It is a 24-hour bubbler. It had been field tested when the standard had been set, but we did not know at that time of a problem that cropped up subsequently relating to the collection efficiency. At the time that method was set

Senator MUSKIE. When was that time?

Dr. GREENFIELD. This was the time just prior to the setting of the standard itself and the publishing of that method as a standard refer

ence method.

Senator MUSKIE. Was this before 1969 or 1970?

Dr. GREENFIELD. 1971; 1971 officially.

Senator MUSKIE. After the enactment of the legislation?

Dr. GREENFIELD. Yes; but in setting up this as the Federal reference method, it was assumed that it had a 35-percent collection efficiency which was constant across all concentrations.

The Griess-Saltsman method is a continuous sampler. It too was tested at the time. It was the one used in the camp stations and the Chattanooga abatement studies. It does not give reliable measurements at low levels of NO, and oxidants and others may be a problem. The third method is the so-called arsenite bubbler. There are several versions of that. They too are 24-hour bubblers. They were not field tested at the time the standard was set. However, they do appear to have a stable collection efficiency over a wide range of NO, concentrations. There may be some interferences but they are not yet completely contemplated. It appears to be relatable to the Saltsman observation. That is a very key point.

The fourth method is the so-called Chemluminescent method. It is a continuous method. It had not been tested at the time of the standards. It avoids the drawback of wet chemicals. There is an additional field testing now underway. It has the ability of providing short-term air quality standard measurements.

To get at the moment of the problems of the collection efficiencies, this is a graph which plots up the side collection efficiency or overall efficiency and percentage and along the bottom concentrations of nitrogen dioxide sampled. On the bottom it runs 30 micrograms per cubic meter to 750, certainly covering the range of our interest. The collection efficiency runs from zero up to 80 percent. The dotted line

running across the chart horizontally is the 35 percent assumed constant collection efficiency applied to the Federal reference standards. The curve starting at a little above 70-percent efficiency and running down to just above 10 percent efficiency at the high end of the NO concentration is the actual collection efficiency that we have now determined with this view.

This means that at the low end of the concentration where the curve goes above the dotted line, you are over estimating the concentrations of NO, in the atmosphere. Below the dotted line you are underestimating it.

Faced with that is a problem we decided we had to find out what we could do to recover or hold on to the data we had which underpins the standard that had been set.

Fortunately, at the time that the measurements were being made with the Federal reference standard in Chattanooga for which the health data was available, there was a small overlap period with the Saltsman measurement technique 1 month.

What we did was go back and take the sampling by the Saltsman method and compare it with the samples or the concentrations that had been determined by the Jacobs-Hochheiser method. At the levels we were measuring in Chattanooga at that time, namely about 100 micrograms per cubic meter, we found a fair amount of agreement between the Saltsman measurements and the Jacobs-Hochheiser which gave us some degree of assurance that the level, approximately the level we had set for the observable health effects, was about right, namely about 100 micrograms per cubic meter.

I can't emphasize enough the fact that in no way does this remove the feeling and assurance that we have that there is a health-associated effect due to NOx. What we are talking about now is exactly where you set that standard and whether or not the data we had was usable in setting the standard and what it allows us to do today.

Senator MUSKIE. Could I ask a question or two to clarify in my own mind what that means?

Are you saying that the levels of concentration of nitrogen oxides that relate to healthy effects is the same now in your judgment as it was in 1970 or different?

Dr. GREENFIELD. No; I am saying that the level that we set the standard at, approximately 100 micrograms per cubic meter, which were related to health effects noted in Chattanooga, has not changed. There may be a degree of uncertainty as to exactly what that number should be, whether it should be 90, 100, or 110 or what have you. But the approximate level has not changed and the degree of uncertainty is primary in the degree of uncertainty associated with the various instruments that we have and how they relate one to the other. At the level of about 100 micrograms per cubic meter, at about the level we set the standard, we know that the Saltsman measurement would have a degree of confidence at that level that just about matches the JacobsHochheiser. So we know that at that point at least the two instruments were reading about the same.

The next question is, that is the one I will get to now, is can I go out into the field where I have a large number of measurements and find a measurement system that I can use which matches with the Saltsman?

The method that we have in the field right now is the arsenite bubbler. Its collection efficiency is shown on this chart. We have a fair degree of confidence now that over the range of NOx concentrations that we are interested in, its collection efficiency remains essentially constant.

Senator MUSKIE. It is about 85 percent?

Dr. GREENFIELD. About 85 percent. The difference between the circumferential and the triangle is the difference between the type of orifice or frit that you put on the instrument. The question was how well was the Saltsman measurements with the arsenite bubbler measurements. That is the next chart.

We have a set of monitoring stations, Chicago, Cincinnati, Denver, and so forth. The second column is the measurement at these stations, annual averages, measured with the Saltsman continuous instrument. The third column is the same type of measurement, measured with the arsenite bubbler corrected for that 85 percent collection efficiency and the fourth column is the ratio of the Saltsman to the arsenite corrected. As you can see in the vicinity of about 100 micrograms per cubic meter, 75 to 100, the Saltsman and the arsenite give about the same answer. The ratio is between 1 and 0.9 or thereabouts.

The one measurement that is bad is in the California 841 where we know we had a bad arsenite bubbler. But when you are talking about measurements, in the vicinity of the standard we know the arsenite bubbler and the Saltsman are giving approximately the same

answer.

What we are looking for is where do we approximately draw the line and has that changed when we look at our air quality control regions?

We also wanted to look at what the relationship might be between the Saltsman continuous and the chemiluminescent method. Once again looking at a set of stations and we have now drawn the second column, Saltsman continuous, chemiluminescent, and now the ratio of Saltsman to chemiluminescent, we find once again in the range of the standard, the current standard, the Saltsman and the chemiluminescent gives us approximately the same answer.

If we go now and look at the levels of measurements or the values that are associated with the various measuring techniques at the 47 air quality control regions, we have the following chart. I don't know whether it is completely readable, but the AQCR are listed on the far left column. The next column are the measurements done by the Jacobs-Hochheiser instruments. The third column are those from the arsenite. The fourth column are the chemiluminescent and the fifth column are the number of days, and as you see, there are fewer than a years' worth of data because we have not had the chemiluminescent in the field.

The fifth column is the suggested priority classification. Priority one being those regions that require control, priority three those that do not require control.

As you can see, around 100 micrograms per cubic meter, we find that there are two areas under the measurements and realizing uncertainties that exist, there are two regions, Los Angeles and Chicago, both of which with the Saltsman, chemiluminescent and the arsenite