RATE OF CONSUMPTION OF HIGH SERVICE WATER IN MANHATTAN.

The pump records taken at the end of each eight hour shift of the enginemen were copied and show the pumping of high service water to have been normal all of the time covered by these measurements except for the Saturday and Sunday already noted.

It was intended to also determine the curve of hourly variation in consumption for the high service, and arrangements, properly authorized by Mr. Birdsall, Chief Engineer of the Department Water Supply, were made for securing the reading of each engine counter at the beginning of every hour. These arrangements were broken up by a subordinate of the Water Department, who ordered counter readings discontinued, through a misunderstanding. The brief series of complete observations indicate such anomalous relations in day and night consumption of high service as to demand further investigation than I have found time to make. The complete investigation would require examination into slip of pumps, into the partial closing of gates in the high service distribution pipes, that has, according to my information, been executed for the purpose of lessening the pressure and so checking the consumption, and there should be made also an approximate census of consumers, number of taps and number of meters in the high service district. So far as I can learn, nothing of this kind now exists. This would have been a work beyond the scope of my instructions. Such roughly approximate figures as I have obtained regarding the high service consumption indicate a surprisingly high average consumption per capita, and an unusually high rate of draft at night; the rate of draft from 2 to 4 A. M. being not far from 90 per cent. of the mean draft for the 24 hours. These figures are so large as to suggest a more complete investigation before accepting them as giving the real draft of high service water at night, although without doubt the present severe shortage of water by day in the high service district necessitating the large use of house tanks, to be filled at night when the pressure is high, causes the night draft to continue larger than would otherwise be the case, until such hour as these tanks are refilled.

There is something unaccountable in the large high service draft. It would be profitable for the Water Department to not only follow up carefully this matter of consumption of high service water, but to make such experiments upon the rate of low service draft in different parts of the city as could be done by a manipulation of the main distributing gates, such as would throw first one district and then another on to the Old Reservoir in Central Park for its supply, and thus permit its fluctuation of draft to be noted.

CONCLUSIONS DRAWN FROM THESE CENTRAL PARK EXPERIMENTS.

Returning again to the measurement of rate of draft from hour to hour by the Borough of Manhattan as a whole, I can find in the observations themselves or in the arrangement of gates and conduits no ground for discrediting the indication of these experiments, that with an average rate of draft of 115 gallons per 24-hour day:

The rate of draft from 2 to 4 A. M., was never less than..
And averaged.

While the highest rate of draft reached during working
hours was....

90 gallons per inhabitant per 24 hours.

66

66

66 66

94

137

66

66

66 66

66

Thus the excess of consumption in the most active hour of the day was found to be only 43 gallons above that in the most quiet hour of the night.

Late on Sunday night, or rather, very early on Monday morning, we would expect the consumption to be at its minimum, for at that time domestic use will be almost nothing, factories will be idle, house tanks and factory tanks will nearly all have become refilled through the higher water pressure of Sunday, and almost the only legitimate uses that can be thought of are the comparatively few newspaper and telegraph offices, the filling of a few steam-heating boilers in office buildings, etc., a small quantity for electric-light stations and a little for hospitals, or for an occasional household in which there is sickness; but even in these instances the use will be much smaller than on the same premises by day.

From 2 to 4 A. M. on Monday, December 4, and in the same hours on Monday, December 11, 1899, we find the measurements agree most remarkably in showing a draft at the rate of 94 gallons per inhabitant per 24 hours. The fact that this rate is the same as on other nights indicates that the tanks in commercial buildings and factories do not fail to get full before 2 A. M. on other nights than Sunday, or that there is no special industrial or commercial use through the week at from 2 to 4 A. M. different from that on Sunday.

We would hardly expect on general principles or in comparison with other cities, taking the city as a whole, that the total rate of draft for real use at this quietest hour of the night would be more than 10 per cent. of the increase of the most extreme day use over the smallest rate of night use— 10 per cent. of 43 is 4 gallons, which, deducted from the 94, leaves 90 gallons. But the night use is without doubt larger in New York than in most other cities, and there are doubtless some tanks in the high service district especially which do not get refilled until after this hour.

There is, perhaps, a possibility that I was misinformed as to the connection being closed between the Croton high service and the Williamsbridge distribution pipes, or possibly the volume from High Service going back across the Harlem as given me by the Water Department officials may have been under-estimated, but on carefully reviewing all these chances the proof is positive that the error from these sources was not large.

To allow for all these and to further apply all margin of uncertainty from lack of precision of measurement in the direction of letting the use appear large rather than to run any risk of exaggerating the leakage, I will call the night use from 2 to 4 A. M. at the rate of 19 gallons per capita.