Mr. WOODWARD. About 5 or 6-well, this covers a much longer period. This covers 10 or 12 days in a table on page 1. The first half of the table shows that the releases were the same as our suggestions, substantially as nearly as they could be regulated.

Mr. BIDDLE. And for the last 4 or 5 days what does it show?

Mr. WOODWARD. The 4 or 5 days at the last part of the table, there is a wide discrepancy between our suggested releases which were constant, and the figures in this column as being the actual ones.

Mr. BIDDLE. Presumably the power people wishing to have more water, to keep more water than you wished released; is that right?

Mr. WOODWARD. Yes; the releases were reduced. Now, I can say that I have been able to check this to this extent, I telephoned to Knoxville this morning this column of figures, but they said it would take them an hour or two to look it up, because there was no one place where they could get them; they had to look through various days. During the noon recess I got an answer back stating that they found certain discrepancies in 1 day particularly, but the others were about the same as these. The reasons for the discrepancies are these, they give the releases at a certain hour of the day, four times a day. Now, we keep track of the releases by automatic registers on the water wheels. We have a regular record, river measuring station, about a half a mile, or at this time it may have been a mile and a half below the dam, which another branch of our organization keeps the records on. There is a little time lag of about a half hour between those, and they don't always agree exactly. There is a little trouble in making that check. These probably are the powerhouse reports, while my staff was looking up our regular records at the reading station.

Mr. BIDDLE. I don't think that is important.

Mr. WOODWARD. But they told me they substantially agreed, with just slight discrepancies, except the 1 day, and we had to cut my telephone conversation off for me to get back here in time, but it has presumably been taken down in the office.

Mr. BIDDLE. Was there any flood at that time?

Mr. WOODWARD. No; this was after the big flood, 3 weeks after the big flood, when we had lowered Norris Reservoir down probably 10 or 15 feet.

Mr. BIDDLE. Did the withholding of that water have any effect. with respect to flood control at all?

Mr. WOODWARD. I don't know any way that this difference affected any subsequent flood control. The difference was this, that we recommended or suggested, as we call it, that Norris Reservoir ought to be drawn down lower, and I think the power people thought they would like to save the water, and they did save some of it.

Mr. BIDDLE. At that time you say it was a system of joint control. It is now a system in which you have complete control of the river?

Mr. WOODWARD. That is correct.

PREFERENCE GIVEN TO NAVIGATION AND FLOOD CONTROL

Mr. BIDDLE. Do you or do you not give preference with respect to one of these main purposes over the others?

Mr. WOODWARD. Well, we give preference to the navigation and flood control, entirely, but sometimes one is the one that is important, and sometimes the other one.

Mr. BIDDLE. As between flood control and navigation it would be a matter of judgment?

Mr. WOODWARD. Yes.

Mr. BIDDLE. As to which was most important?

Mr. WOODWARD. During flood times

Mr. BIDDLE. There is no question then.

Mr. WOODWARD. Then navigation is relatively unimportant.

Mr. BIDDLE. Except to keep the flow as steady as you can for navigation?

Mr. WOODWARD. Yes; all we do for flood control would automatically help navigation. Navigation has troubles during floods, as well as everybody else, but during the low-water part of the year, it is navigation that is controlling.

LACK OF CONFLICT BETWEEN FLOOD CONTROL AND POWER

Mr. BIDDLE. Mr. Woodward, it has been stated from time to time, and I think the current conception is somewhat puzzled, that if the needs of flood control are to keep a reservoir full and hold the water back, and if the needs of power are to let that water out and turn your water wheels, there must be an inconsistency, which a multiple operation cannot under all circumstances solve. I should like very much to have your detailed comments on that.

Mr. WOODWARD. There is a possibility in that respect that you mention of an inconsistency, but it is very minor, relatively rare. For example, at Norris Dam, the maximum flood flow on our records is about 110,000 cubic feet per second, so that if we had that kind of flood and stored all of the water, we would be holding out of the flow in the Tennessee this amount of 110,000 cubic feet per second.

Now, suppose during that same time there was a demand for power; our two water wheels at Norris can use about 8,000 cubic feet per second and that is all, so that if they ran the full amount they would let through the water wheels only 8,000 out of this 110,000 that would be coming there naturally. If they did run that way, then you would hold out only 102,000.

Now, during the most of the flood times, that amount of 8,000 is relatively so small compared with the flow in the rivers that it is rather trivial.

I think that if it were important for power to use some water through those water wheels in that kind of flood, we would say that is perfectly reasonable, because of the very complex nature of the flood control that we don't get 100 percent of efficiency. They only get certain parts. We are only trying to control a part of it. And ordinarily a power use of that sort would not be any serious matter. Now, there is another kind of conflict that is what is ordinarily spoken of, or thought of, I think, as the main conflict, which is that power companies ordinarily have no interest whatever in flood control, and everyone that is used to being in that business, and that is his only interest, is always anxious to have his reservoirs full. He has to draw them down during the low water season, but if any water comes along in July, he keeps it in the reservoir if he can. Power plant operators get that way, they get so thoroughly and firmly ingrained into doing that as a habit that they think it is a criminal offense not to save any water at the time that it comes along, flood season, or dry season, or any time.

Now, we are going to operate these reservoirs in such a way as to control the floods, and then will produce a lot of power, but we are

going to discharge water when reservoirs are not full, a large part of the time, because we are saving space for possible floods that come during the flood season.

Now, it is also true that at the end of the flood season, then such water as we have, and there usually is a tailing-out effect of the high water in the winter, so that for a month we have a considerable, or moderate flow that we can store in the reservoirs. Then we are going to use this storage space so far as we can to store up water that can be released during the summer for navigation value, and for use of power during the dry or low water season.

Mr. BIDDLE. Then I take it that in order to control any possible floods in the operation of your system, you will necessarily have to waste from your reservoirs water which could be used if flood control was not your first obligation, which would produce primary power. Mr. WOODWARD. Oh, yes, sir. Probably fairly frequently every year, or in half of the years we will waste water that if we had no interest in flood control, if it could be saved, would be useful for power. Mr. BIDDLE. For primary power?

Mr. WOODWARD. For primary power. That will not be true in the wet years because there will be such an excess of water, but perhaps half of the years.

Mr. BIDDLE. Now, are the primary power calculations of the T. V. A. based on the operation of the system which you have described, that is, is the amount of primary power calculated on the wasting or letting out from the reservoirs that water so that primary power is counted on your operating system on that only?

Mr. WOODWARD. Yes, sir; if we take the reservoirs above Wilson Dam, leaving out Gilbertsville, which has a tremendous storage value without much value for power, leave out Pickwick which is below Wilson Dam, but take the ones above, including Norris and Hiwassee, we have a storage in all of those put together that sums up to about 4,000,000 acre-feet, I think. I have not added that for this purpose. But out of that 4,000,000 acre-feet of storage that we expect to have available for flood control every year, in a dry year our power calculations are based on the assumption that we will use about 1,500,000 acre-feet out of that total amount of storage.

Mr. BIDDLE. Is it fair to put it this way that just as your floodcontrol figures are based on the worst possible flood, plus, which you testified to a little while ago, so also your power calculations are based on the worst possible drought?

Mr. WOODWARD. That is true for primary power.

Mr. BIDDLE. I am speaking only of primary power.

Mr. WOODWARD. Yes, sir.

Mr. BIDDLE. Dependable power.

Mr. WOODWARD. And that primary power will use about 1,500,000 acre-feet of stored water each year. But on an average year when we will get secondary power, we will use approximately-these are sort of mental figures, we will use say about 2,500,000 acre-feet of stored water. But we are going to store every year there is any flow at least 4,000,000 acre-feet above the Wilson Dam in the part that is important for power.

EXTENT OF FLOOD-CONTROL PROTECTION TO BE AFFORDED BY AUTHORITY 10-DAM SYSTEM

Mr. BIDDLE. Your calculations began from the year 1867, is that the first year in which you had figures you say on floods?

Mr. WOODWARD. Yes; that is the first one that we have a definite record on, apparently, as to a high-water level at Chattanooga. Mr. BIDDLE. Was the 1887 flood the worst flood?

Mr. WOODWard. No.

Mr. BIDDLE. Which was the worst?

Mr. WOODWARD. 1867 was the worst.

Mr. BIDDLE. 1867 was the worst?

Mr. WOODWARD. 1875 was second and 1886 the third.

Mr. BIDDLE. And I understand that you added to the 57 feet for the worst flood an additional number of feet?

Mr. WOODWARD. Yes; 14 feet more.

Mr. BIDDLE. 14 feet more?

Mr. WOODWARD. Yes.

Mr. BIDDLE. Was the 1937 flood on the Tennessee a bad flood? Mr. WOODWARD. Not on the upper half. It was a pretty good sized flood at

Mr. BIDDLE. At Chattanooga?

Mr. WOODWARD. No, not at Chattanooga, but below Wilson Dam. The heaviest part of the rainfall that caused the 1937 flood was a strip across western Tennessee and across the Tennessee River, but that was just the lower quarter of the river that had the big rainfalls. The upper part of the river didn't have any big flood.

Mr. BIDDLE. Did your control of that flood, or whatever control you exercised, have any effect on the river at Paducah?

Mr. WOODWARD. Yes, sir; we think it did, a measurable effect.
Mr. BIDDLE. How much?

Mr. WOODWARD. I always hate to state figures because we were controlling it at Norris Dam, 650 miles away, and at the best we cannot claim great precision for these figures. I studied over this for months, and I compared with engineers working for the Army Engineers, and other engineers who discussed it, and made assumptions and tentative calculations and finally used what I thought was the most reasonable general judgment that I could, and no engineer has ever criticized-that is at least none that I am friendly with-has ever criticized it to me as being unreasonable. The nearest I could determine was that during the crest of the flood at Paducah, we lowered the level at Paducah about 5 inches. That could be an inch wrong either way. I didn't try to favor it, I just tried to compute it, with the best judgment that I could.

Mr. BIDDLE. Was there any effect at Cairo?

Mr. WOODWARD. The effect at Cairo would have been less, about an inch less.

Mr. BIDDLE. About 4 inches?

Mr. WOODWARD. About 4 inches at Cairo.

Mr. BIDDLE. Did you make any calculations below Cairo?

Mr. WOODWARD. No. But ordinarily at the crest of the big floods, the crest is very uniform from Cairo down the river for about 300 miles.

Mr. BIDDLE. How far is Cairo from Paducah?

Mr. WOODWARD. Forty-some-odd miles.

Mr. BIDDLE. Forty-some-odd miles?

Mr. WOODWARD. They are both on the Ohio.

Mr. BIDDLE. I understand.

Mr. WOODWARD. And from Cairo down it is the Mississippi.
Mr. BIDDLE. I understand.

Mr. WOODWARD. It is a bigger channel, a bigger valley. Mr. BIDDLE. Have you had any other serious floods since the dam system has been in operation on which you could give any estimates?

Mr. WOODWARD. No, sir. I think the only other one that was serious at all was in the spring of 1936, and that was the one I described this morning. It is supposed to have done a half-million dollars or three-quarters of a million dollars damage, I believe, at Chattanooga. No one made a very careful survey, but it was not a very big flood.

Mr. BIDDLE. You spoke of Gilbertsville when complete with relation to flood control on the Ohio and Mississippi. Would you venture any opinion as to the extent of that flood control?

Mr. WOODWARD. We think it will be able to reduce the crests of all big floods on the Ohio River about 22 feet from Paducah down to Cairo, and a corresponding amount down the Mississippi River. Mr. BIDDLE. You mean that from Cairo down 300 miles, the amount probably would correspond to the amount between Cairo and Paducah.

Mr. WOODWARD. It will be a little bit less. It might be 2 feet from Cairo down, if we say maybe 22 feet from Paducah to Cairo, because it is a wider valley, and there is accession to the flow from the Mississippi which comes down to Cairo, and it is a little smaller fraction of the total.

But all of the big floods are very uniform from Cairo down to Helena, about Helena, Ark., where the St. Francis River comes in, which at times carries very big floods, and then 100 miles below that the White and Arkansas come in, and they carry floods half as big as the flood coming down the Mississippi at bad times, so the effect will be less if you go below that point. But for the first 300 miles it is very uniform.

All of the Mississippi River records show that the flood stays very uniform for about that 300 miles.

ADDITIONAL STEPS NEEDED TO AFFORD COMPLETE FLOOD-CONTROL

PROTECTION

Mr. BIDDLE. I understand you have testified Chattanooga has in the past been seriously damaged by floods and is still in a precarious situation with respect to floods.

Do you believe that the completion of your dan system will entirely control floods? I am now speaking of a 10-dam system, that is, the 10-dam system authorized, at Chattanooga, or do you believe additional steps will have to be taken to control floods at Chattanooga completely?

Mr. WOODWARD. Yes; a great deal of additional steps.

Mr. BIDDLE. Would those steps consist both of the use of levees locally and of dams constructed at other places?

Mr. WOODWARD. Yes, sir; I think the most economical system will be to use both means; the levees and some protection works around the city, up to a considerable extent, and then in addition more reservoir storage above to help in the protection.

Mr. BIDDLE. Are there levees at Chattanooga now?

Mr. WOODWARD. Not of any size. There may be little ones around small areas.