THE UNIVERSITY OF MICHIGAN COLLEGE OF ENGINEERING Department of Civil Engineering Meteorological Laboratories Final Report METEOROLOGICAL ANALYSIS E. Wendell Hewson Professor of Meteorology Gerald C. Gill Associate Professor of Meteorology Eugene W. Bierly Assistant Research Meteorologist UMRI Project 2515 under contract with: POWER REACTOR DEVELOPMENT COMPANY DETROIT, MICHIGAN administered by: THE UNIVERSITY OF MICHIGAN RESEARCH INSTITUTE ANN ARBOR January 1961

6B9',, 4/rl = el q7I J/MK4-y alds~'I AOO 5

PREFACE This final report summarizes several projects sponsored by Power Reactor Development Company of Detroit, Michigan. All the projects are interrelated, concerning themselves with the meteorological aspects of the Enrico Fermi Atomic Power Plant. The first project was a climatological project begun in August, 1956. Five project reports were written covering the installation of the measuring equipment and seasonal and yearly summaries of the observations taken until 30 November 1959. The second project was a diffusion study which was made to determine diffusion characteristics at the plant site and in the surrounding environs, The diffusion studies were carried out under varying meteorological conditions including inversion conditions, Two progress reports describing the equipment and technique used in the study and qualitatively describing the days on which data were collected were written. A technical report described the diffusion in a quantitative manner. Several other smaller projects are also surmarized in the report. They are the computations made for the design and location of the waste-gas stack and graphs of maximum and minimum temperatures for use as design criteria. The authors wish to acknowledge again the help of the many individuals and agencies in making these several projects successful. In particular, thanks is given to Mr. Jal N. Kerawalla and Miss Ana Lucia Torres P. for aid in making some of the calculations, and to Mrs. Anne C. Rivette for typing the original manuscript. iii

TABLE OF CONTENTS Page LIST OF TABLES vii LIST OF FIGURES ix ABSTRACT xi I. WIND SUMMARY 1 1. Introduction 1 2. Wind Direction 1 3. Wind Speed 4 II. TEMPERATURE-LAPSE-RATE SUMMARY 7 1. Introduction 7 2. Distribution of Lapse Rates 7 3. Diurnal Variation of Inversion 11 4. Persistence of Inversions 15 III. PRECIPITATION SUMMARY 19 1. Introduction 19 2. Distribution and Frequency of Precipitation 21 IV. EXTREMES OF TEMPERATURE 25 1. Maximum Temperatures 25 2. Minimum Temperatures 31 V. SUMMARY OF DIFFUSION STUDIES 59 VI. STACK DESIGN 43 VII. GENERAL CONCLUSIONS 45 REFERENCES 47 v

LIST OF TABLES Table Page I Summary of Data Contained in Progress Reports. 1 II Percentage Frequency of Occurrence of Winds in Various Directions Grouped According to Wind Speeds: ThreeYear Summary, 1956-1959, Enrico Fermi Site, 2 III The Association of Temperature-Lapse Rates with Wind Direction at the Enrico Fermi Site: Three-Year Summary, 1956-1959. 8 IV Wind Direction and Mean Wind Speed Associated with Inversions and Noninversions at the Enrico Fermi Site: Three-Year Summary, 1956-1959. 10 V The Association of Inversion Periods and Wind Direction at the Enrico Fermi Site: Three-Year Summary, 1956-1959. 12 VI Hourly Percentage Frequency of Inversions at the Enrico Fermi Site: Three-Year Summary, 1956-1959. 13 VII Number of Occurrences and Percentage Frequency of Occurrence of Continuous Inversions at Enrico Fermi Plant Site: Three-Year Summary, 1956-1959. 16 VIII Periods of Stagnation, 4 Days or Longer, by Months in the Detroit Area, 18 IX The Association of Precipitation with Winds at the Enrico Fermi Site: Three-Year Summary, 1956-1959. 22 X Mean Computed Values of Cz and Cy at the Enrico Fermi Plant Site, 39 XI Diffusion Parameters Used in an Evaluation of a Hypothetical Contained Accident at the Enrico Fermi Reactor. 40 vii

LIST OF FIGURES Figure Page 1. A topographic map of the Enrico Fermi Site and surroundings. 3 2. Percentage frequency of occurrence of winds from 16 directions and corresponding wind speed in mph at Toledo Municipal Airport, Five-Year Summary, 1950-1954; Detroit City Airport, Toledo Express Airport, and Enrico Fermi Site, Three-Year Summary, 1956-1959. 5 3. Mean wind speed at the Enrico Fermi Site and Detroit-Toledo combined, for 16 directions, expressed as a percentage of the over-all mean three-year wind speed, 1956-1959. 6 4. Percentage frequency of inversions and noninversions associated with winds for 16 directions and corresponding wind speed in mph at the Enrico Fermi Site: Three-Year Summary, 1956-1959. 9 5. Diurnal variation of inversions at the Enrico Fermi Site and at WJBK-TV tower: Three-Year Summary, 1956-1959. 14 6. Percentage frequency of continuous inversions at Enrico Fermi Plant Site: Three-Year Summary, 1956-1959. 17 7* Percentage of particles, of a given diameter, scavenged from cloud by raindrops as function of rainfall rate, R, and time duration of the rate, t. 20 8. Percentage frequency of occurrence of winds from 16 directions and corresponding wind speed in mph with precipitation at Toledo Municipal Airport, Five-Year Summary, 1950-1954, and Toledo Express Airport, and the Enrico Fermi Site, Three-Year Summary, 1956-1959. 23 9. Probability and return period of maximum temperatures during May. 26 ix

LIST OF FIGURES (Concluded) Figure Page 10. Probability and return period of maximum temperatures during June. 27 11. Probability and return period of maximum temperatures during July. 28 12. Probability and return period of maximum temperatures during August. 29 135 Probability and return period of maximum temperatures during September. 30 14, Probability and return period of minimum temperatures during October, 32 15. Probability and return period of minimum temperatures during November. 33 16. Probability and return period of minimum temperatures during December, 34 17* Probability and return period of minimum temperatures during January. 35 18. Probability and return period of minimum temperatures during February. 36 19. Probability and return period of minimum temperatures during March, 37 x

ABSTRACT Wind, lapse-rate, and precipitation observations for the three-year period 1 December 1956 to 30 November 1959 are reviewed and summarized, emphasizing those factors which are of major importance to plant operation and diffusion characteristics. Extremes of temperature are included in this summary report. Diffusion characteristics based upon the observations from the diffusion project are discussed and summarized as well as the meteorological criteria used in the stack design. General conclusions summarizing the site meteorology are also presented. xi

Io WIND SUMMARY 1. INTRODUCTION The data referred to in this section are those from the three-year period 1 December 1956 to 30 November 1959. Some of these data are being reproduced in this report to aid in summarizing the material, Detailed data by seasons and by years are found in earlier progress reports 1-5 The following table may be used as a basis for such references. TABLE I SUMMARY OF DATA CONTAINED IN PROGRESS REPORTS Report Number Data 2515-1-P Details of instrumentation and observations prior to 12/1/56 2515-2-P Observations from 12/1/56 to 5/31/57 2515-3-P Observations from 12/1/56 to 11/30/57 plus 1957 summary and lake breeze discussion 2515-4-P Observations from 12/1/57 to 11/30/58 plus 1958 summary 2515-5-P Observations from 12/1/58 to 11/30/59 plus 1959 and 3-year summary 2. WIND DIRECTION Table II is the three-year summary of the wind-direction distribution grouped according to wind-speed categories. There are several features to be emphasized. The first is with reference to the prevailing wind direction and major population centers, Figure 1, a map of the area surrounding the plant site, will aid in the discussion. The prevailing wind directions are from SSW through WNW. Such winds occur about 45% of the time. Figure 1 shows that these wind directions are toward the lake and the Ontario shores area, a relatively distant and sparsely populated area, Except for SSW winds which after crossing several miles of marshy shore area would reach communities down river from Detroit, no significant population center is in the path of these winds. 1

TABLE II PERCENTAGE FREQUENCY OF OCCURRENCE OF WINDS IN VARIOUS DIRECTIONS GROUPED ACCORDING TO WIND SPEEDS Enrico Fermi Site (Aerovane at height of 102 ft) 1 December 1956 - 30 November 1959 (3-Year Summary) Speed, mph Total Mean Speed Wind 32 Total Observations Direction 0-3 4-12 13-24 25-31 and 4 and N ~ o of OverOver Over mpN all Mean N 0.2 2.9 0.8 0.1 3.8 4.0 1015 10.1 81 NNE 0.2 2.4 1.4 0.1 3.9 4.1 1023 12.6 102 NE 0.2 1.6 3.7 0.2 0.0 5.5 5.7 1423 15.3 123 ENE 0.2 2.0 2.7 0.2 0.0 4.9 5.1 1271 14.2 115 E 0.2 1.9 1.9 0.2 0.1 4.1 4.3 1077 14.3 115 ESE 0.1 2.4 1.7 0.1 0.0 4.2 4.3 1070 12.4 100 SE 0.1 2.4 1.5 0.1 4.0 4.1 1033 12.0 97 SSE 0.1 3.4 1.3 0.0 4.7 4.8 1204 10.8 87 S 0.1 3.6 1.3 0.0 4.9 5.0 1264 10.7 86 SSW 0.2 4.6 3.2 0.1 0.0 7.9 8.1 2043 12.3 99 SW 0.2 4.9 4.5 0.1 0.0 9.5 9.7 2429 13.0 105 WSW 0.2 5.6 5.1 0.4 0.1 11.2 11.4 2851 13.4 108 W 0.2 4.7 3.5 0.2 0.0 8.4 8.6 2175 12.7 102 WNW 0.2 4.5 3.2 0.1 0.0 7.8 8.0 2013 12.3 99 NW 0.2 3.9 2.5 0.1 6.5 6.7 1690 11.9 96 NNW 0.2 4.0 1.6 0.1 5.7 5.9 1471 10.9 88 Calm 0.5 0.3 83 0.0 0 Totals 3.1 54.8 39.9 2.1 0.2 97.0 100.1 25155 Average 12.4 100 2

DETROIT ^^^ / CITY L LAKE P S -^ i UAIRPORTN, CLA IR WILLOW RUN E N AIRPORT \ i /"^y }^>^ ^s^^Tft-? C\A A \ A \\E)Y-ot i5/2 f TOLEDO S CTOD E ~~EXPRESS 3, g MUNICIPAL AIRPORT / AIRPORT Fig. 1. A topographic map of Enrico Fermi site and surroundings.

Figure 2 shows that, in general, the wind-direction distribution at the plant site is representative for the general area surrounding western Lake Erie, but with a higher occurrence of easterly winds than stations further inland such as Toledo and Detroit. It should also be noted from Table II that, when the wind speed is above 31 mph, winds blow only from the NE through E to ESE and from SSW through W to WNW. This bimodal distribution is a specific effect of the deep storms that traverse the mid-latitude regions, and is not a singularity associated with the plant site. 3. WIND SPEED The most important fact about the wind speed observed at the plant site is that it averaged 12,4 mph over the three-year period. This is 1-2 mph greater than at Detroit City Airport or Toledo Express Airport during the same period. A small part of the excess wind speed at the plant site may be due to the higher location of the wind-measuring equipment at the plant, but most of it must be attributed to the relatively frictionless flow from off Lake Erie. Figure 2 shows that winds with an easterly component average 3-5 mph higher than corresponding winds at Toledo Express or Detroit City Airports. This fact is brought out more clearly in Fig. 35 Table II shows that the average wind speed varies from 10,1 to 15.3 mph, which is quite a narrow range over all 16 directions, Other factors being equal, the concentration of any air pollutant at a point downstream varies inversely as the wind speed. Thus a high average wind speed regardless of direction improves diffusion. Both Table II and Fig. 3 show that the incidence of calm conditions at the plant site is low, 0.3%. The incidence of calm conditions at Detroit City is three times greater, while at Toledo Express it is five times greater,. The nearness of the plant site to Lake Erie allows a lake breeze to develop during periods that would normally be calm. Thus the lake-side location has a decided advantage over some inland sites, In addition to the indications of the observational data, which are favorable, as an additional precaution it is planned to release waste gases from a stack which is 200 ft high. This means that the average wind speed at the top of the stack will probably be near 1355 mph if a 1/7th power law is assumed. Any additional wind speed may be regarded as beneficial.

N N NW " > NE NW i NE 15% 8 mph /NW/ 15% mph W E W 0 S E S S TOLEDO MUNICIPAL AIRPORT DETROIT CITY AIRPORT TOLEDO, OHIO DETROIT, MICHIGAN Wind Instrument at Height of 47 ft. Wind Instrument at Height of 81 ft. Five year Summary 1950-1954 Three year Summary 1956-1959 N N NWW " NE NW NE 15/% o mph / X / 15% a mph 5,~ W E W E SW E S S TOLEDO EXPRESS AIRPORT ENRICO FERMI POWER PLANT SITE TOLEDO, OHIO LAGOONA BEACH, MICHIGAN Wind Instrument at Height of 72 ft. Aerovane at Height of 102 ft. Three year Summary 1956-1959 Three year Summary 1956-1959 Fig. 2. Percentage frequency of occurrence of winds from 16 directions (rectangles) and corresponding wind speed in mph (heavy lines) at Toledo Municipal Airport, Five-Year Summary, 1950-1954; Detroit City Airport, Toledo Express Airport, and Enrico Fermi Site, Three-Year Summary, 1956-1959. 5

_WATER TRAJECTORY AT'u140 "MONROE 130 _ I -120 W / \\ a. o //0 o / U 0 - 90 80 70 DETROIT TOLEDO 60 -— X- - MONROE N NNE NE ENE E ESE SE SSE S SSW SW WSW W WNW NW NNW Fig. 3. Mean wind speed at the Enrico Fermi Site and Detroit-Toledo combined, for 16 directions, expressed as a percentage of the over-all mean three-year wind speed, 1956-1959.

II. TEMPERATURE-LAPSE-RATE SUMMARY 1. INTRODUCTION Data are again reproduced from earlier reports for the summaryo Table I will aid in finding the references in earlier reports for specific details, Because the inversion condition is regarded as the poorest one for diffusion, more analysis has been made of this condition than of any of the other lapse-rate categories. Several factors need highlighting, The first is that the lapse-rate measurements at the plant site are made only in the vertical height interval from 25 ft to 100 ft0 This is a very thin layer and does not represent the lapse rate existing at the tope of the stack which is 200 ft high. It does permit a good analysis of the thin layer near the ground in which the reactor shell itself is located. The meteorological tower where the lapse-rate measurements are made is within 100 ft of the lake shore. Again, then, the observations at this point are not necessarily indicative of even the entire plant site, Wiresonde measurements made during the summer of 1960 indicate that the lake-breeze inversion is not only quite shallow but of limited horizontal extent. Therefore, the measurements made at the meteorological tower are characteristic of a thin and narrow layer of air near the lake shore, Diffusion conditions further inland and aloft may be different. The fact that conditions at the tower may not be representative of those at the stack mouth which is higher or of those on portions of the site farther from the lake does not invalidate the data for purposes of analysis0 Diffusion can, in general, be expected to be poorer at the tower due to its low height and its proximity to the lakea Accordingly, tower conditions are conservative and from the lapse rates and wind observations at the tower, it is possible to make conservative engineering estimates of the lapse rate and diffusion at various distances from the plant under a variety of atmospheric and lake conditions. 2. DISTRIBUTION OF LAPSE RATES The distribution of lapse rates with wind direction is shown in Table III. An important fact to note is that for the three-year period there is a frequency of strong lapse rate 4554% of the time, of weak lapse 26.7% of the time, and of inversion conditions 27I9%7 The association of the noninversion and inversion condition with wind direction and wind speed can be seen in Fig0 4 and in Table IV. The general shape of the distributions in Fig0 4 follows the basic wind pattern given in Fig, 2a There is no particular preference for one wind direction when an inversion occurs or when a noninversion condition occurs. 7

THE ASSOCIATION OF TEMPERATURE-LAPSE RATES WITH WIND DIRECTION AT THE ENRICO FERMI SITF 1 December 1956 - 30 November 1959 (3-Year Summary) Hourl Laps RPercent Frequency of Lapse Rate Wind ourly apse RatesCompass Observations Within Total Direction Totals Categories Observations S W I S W I S W I N 462 167 201 830 5.1 3.1 3.6 2.3 0.8 1.0 NNE 444 144 206 794 4.9 2.7 3.7 2.2 0.7 1.0 NE 885 135 162 1182 9.7 2.5 2.9 4.4 0.7 0.8 ENE 754 120 115 989 8.2 2.2 2.0 3.7 0.6 0.6 E 529 148 128 805 5.8 2.7 2.3 2.6 0.7 0.6 ESE 418 233 192 843 4.6 4.3 3.4 2.1 1.1 0.9 SE 302 209 300 811 3.3 3.9 5.3 1.5 1.0 1.5 SSE 280 213 433 926 3.1 4.0 7.7 1.4 1.1 2.1 S 259 289 471 1019 2.8 5.4 8.4 1.3 1.4 2.3 SSW 494 519 664 1677 5.4 9.6 11.8 2.5 2.6 3.3 SW 487 685 761 1933 5.3 12.7 13.5 2.4 3.4 3.8 WSW 893 881 557 2331 9.8 16.3 9.9 4.4 4.4 2.8 W 703 575 465 1743 7.7 10.7 8.3 3.5 2.9 2.3 WNW 791 408 403 1602 8.7 7.6 7.2 3.9 2.0 2.0 NW 730 367 317 1414 8.0 6.8 5.6 3.6 1.8 1.6 NNW 700 282 222 1204 7.7 5.2 3.9 3.5 1.4 1.1 Calm 16 20 32 68 0.2 0.4 0.6 0.1 0.1 0.2 Totals 9147 5395 5629 20171 100. 3 100.1 100.1 45.4 26.7 27.9 Code: S = Strong lapse = a lapse rate in excess of the dry adiabatic lapse rate. W = Weak lapse = a positive lapse rate that is less than the dry adiabatic lapse rate. I = Inversion = temperature increase with height. 8

NONINVERSION INVERSION NW/ V-~- S`r NE NW/ \-^~ / W / 1% 8 mph / \^ \ I ^ 15% 8 mph WC -II 0.2 E - I r 0.22~ lll~ R W ^^ ^r^*W < / r^S S Fig. 4. Percentage frequency of inversions and noninversions associated with winds for 16 directions (rectangles) and corresponding wind speed in mph (heavy lines) at the Enrico Fermi Site: Three-Year Summary, 1956-1959.

TABLE IV WIND DIRECTION AND MEAN WIND SPEED ASSOCIATED WITH INVERSIONS AND NONINVERSIONS AT THE ENRICO FERMI SITE 1 December 1956 - 30 November 1959 (3-Year Summary) Win d 1 Inversion Noninversion Direction Occurrence, Mean Speed, Occurrence, Mean Speed, Direction mph lo mph mph.% mph N 1.0 7.1 3.1 10.1 NNE 1.0 8.1 2.9 12.4 NE 0.8 9.8 5.1 15.9 ENE 0.6 9.6 4.3 14.8 E o.6 10.2 3.3 15.3 ESE 0.9 11.8 3.2 11.8 SE 1.5 11.8 2.5 11.7 SSE 2.1 10.6 2.5 10.6 S 2.3 10.0 2.7 10.7 SSW 3.3 11.5 5.1 12.4 SW 3.8 12.0 5.8 13.2 WSW 2.8 10.6 8.8 14.0 W 2.3 10.1 6.4 12.6 VNW 2.0 10.3 5.- 13.0 NW 1.6 10.0 5.4 12.1 NNW 1.1 8.2 4.9 11.4 Calm 0.2 0.0 0.2 0.0 Totals 27.9 72.1 Average 10.4 12.8 10

It has been mentioned earlier in Section 1-5 that the average wind speed over the three-year period 1956-1959 was 12.4 mph. Table IV shows that the average speed with an inversion condition is 10,4 mph while with a noninversion it is 12.8 mph. Under the poorer diffusion condition, this wind speed is still considered moderately high, As pointed out earlier, the high wind speed is a factor that lowers the concentration at any point downstream, Figure 4 shows even more clearly that, when an inversion does occur, the wind speed regardless of direction is on the average near 10 mph, the wind speeds from the NNW through the NNE being lowest. 35 DIURNAL VARIATION OF INVERSIONS Table V shows a breakdown of inversions by wind directions for daytime and nighttime. As would be expected, the nighttime winds are slightly lower in speed than are the daytime ones. This table also shows that there is almost an equal occurrence of inversions by day and by nights Table VI gives a breakdown of inversions by hours of the day. Figure 5 is a plot of the same data plus a plot of similar data from the WJBK-TV tower located in northwestern suburban Detroit, Over the three-year period, the average number of hours of inversions at the plant site and WJBK-TV tower are nearly the same, The plot of the WJBK-TV tower data shows a typical diurnal pattern of inversions found at a continental station: a maximum frequency of inversions in the early morning hours followed by a minimum in the midafternoon and then a slow rise after sunset to the early morning maximum, This pattern is created by the well-known and often discussed nocturnal or radiational inversion, which is destroyed by solar radiation during the daylight hours, The plant site and WJBK-TV tower data in Fig. 5 show two distinct differences, The diurnal variation in frequency is much less at the plant site than at the WJBK-TV site and the minimum occurs in the late morning hours at the site rather than in the afternoon as in suburban Detroit, These differences can definitely be attributed to the proximity of the plant site to Lake Erie. The water has a modifying effect on the local climate. One result is that the advection of air from over water which is relatively warmer than the land does not allow the formation of as many nocturnal inversions at the plant site as a typical continental station would normally have, Thus the maximum inversion frequency at the plant site is less than at the WJBKI-TV tower, Counterbalancing this effect is the lake-breeze-induced inversion which occurs during the daylight hours, When the lake i-s cooler than the land and if the pressure gradient is weak enough, cool air from off the lake is advected over the land, causing an inversion to form over a narrow area paralleling the lake shore. So as the nocturnal inversion effects are being destroyed by solar radiation, the same solar radiation is causing a lake breeze inversion to form, The diurnal minimum frequency of inversions occurs when the nocturnal inversion has largely disappeared and before the lake breeze inversion has become pronounced, 11

TABLE V THE ASSOCIATION OF INVERSION PERIODS AND WIND DIRECTION AT THE ENRICO FERMI SITE 1 December 1956 - 30 November 1959 (3-Year Summary) _Daytime Nighttime Wind Mean Mean Wind No.. Occurrences, % Mean Occurrences, % Mean. Direction N W No. Wind Total Overall Speed Total Overall Speed N 58 2.3 1.0 7.2 143 4.6 2.5 7.1 NNE 63 2.5 1.1 8.1 143 4.6 2.5 8.2 NE 68 2.7 1.2 11.9 94 3.0 1.7 8.3 EKE 55 2.2 1.0 10.2 59 1.9 1.0 9.3 E 66 2.6 1.2 9.7 63 2.0 1.1 10.6 ESE 128 5.0 2.3 11.3 64 2.1 1.1 12.9 SE 228 9.0 4.1 11.3 77 2.5 1.4 12.2 SSE 324 12.8 5.6 10.6 104 3.4 1.8 10.9 S 232 9.1 4.1 10.0 239 7.7 4.2 10.0 SSW 290 11.4 5.2 11.5 374 12.1 6.6 11.6 SW 347 13.7 6.2 13.1 414 13.4 7.4 11.0 WSW 214 8.4 3.8 11.8 342 11.1 6.1 9.8 W 168 6.6 3.0 11.8 297 9.6 5.3 9.2 WNW 138 5.4 2.5 12.0 265 8.6 4.7 9.4 NW 93 3.7 1.7 12.2 224 7.3 4.0 8.9 NNW 54 2.1 1.0 9.1 168 5.4 3.0 7.9 Calm 14 0.6 0.2 0.0 18 0.6 0.3 0.0 Totals 2540 100.1 45.2 3088 99.9. 54.7 Average 11.2 9.8 12

TABLE VI HOURLY PERCENTAGE FREQUENCY OF INVERSIONS AT THE ENRICO FERMI SITE 1 December 1956 - 30 November 1959 Hour Annual 3-Year Ending 1957 1958 1959 Summary 0100 22.4 36.8 39.1 32.8 0200 23.4 37.1 42.9 34.5 0300 22.4 36.5 43.o 34.0 0400 22.4 35.0 43.4 33.6 0500 22.9 35.1 40.5 32.8 0600 23.5 35.3 41.1 33.3 0700 21.7 33.9 39.0 31.5 0800 17.2 29.9 27.8 25.0 0900 17.6 23.5 17.0 19.4 1000 12.2 22.1 15.7 16.7 1100 12.6 21.1 18.6 17.4 1200 17.1 18.5 16.5 17.4 1500 16.8 19.0 20.5 18.8 1400 20.2 23.9 24.8 23.0 1500 22.0 31.3 27.1 26.8 1600 22.7 33.2 35.0 30.3 1700 19.0 56.5 36.0 30.5 1800 17.2 34.4 38.3 30.0 1900 15.5 37.3 36.3 29.7 2000 18.0 36.7 35.2 30.0 2100 18.8 35.7 41.0 31.8 2200 19.4 355.9 42.6 32.6 2300 20.4 33.4 39.7 31.2 2400 19.0 34.0 38.7 30.6 Average 19.4 31.5 33.3 28.1 13

40 Enrico Fermi Power Plant z W - Mean 30 z 20 2 4 6 8 10 12 14 16 18 20 22 24 TIME IN HOURS atwe:Th Tower 50 LU z 60 z m z ^Mean W 30 0 20 U) z I 0 O — 2 4 6 8 10 12 14 16 18 20 22 24 TIME IN HOURS Fig. 5. Diurnal variation of inversions at the Enrico Fermi Site and at WJBK-TV tower: Three-Year Summary, 1956-1959. 14

On the average, this minimum frequency occurs between 0900 and 1000. Since there are two separate inversion types at the plant site, the minimum frequency there is greater than at the WJBK-TV tower, 4. PERSISTENCE OF INVERSIONS As noted in the previous section, the lake effect has a tendency both to create and also to destroy inversions, Another effect is that of sometimes maintaining inversion conditions once they have formed, Some evidence of this may be found in past reports by comparing the continuous inversions listed at the plant site with those listed at the, WJBKI-TV tower1 (see Appendices B of Refs. 4 and 5)~ In general, the periods of inversion at the plant site lasting 24 hr or longer coincide with either an inversion or a weak lapse rate condition at the WJBK-TV tower, The nearness to the lake and the lack of a heat source from a large population area are the primary factors which account for the differences in the persistence of inversions at the plant site and at the WJBK-TV tower. When warm air is advected into the general area over ground that is relatively cold, there are occasional long periods of continuous inversions observed at both the plant site and WJBK-TV tower (see Munn6 for a discussion of such an inversion). Table VII contains a compilation of the frequency of continuous inversions from 1 December 1956 to 30 No-!,ember 1959. Figure 6 is a plot of the same data. Such data may be looked at in two ways. The first way is to suppose an inversion exists, and then to ask what the probability is of having a 20-hr inversion, By looking at the relative frequency it is seen that a 20-hr inversion occurs 0.9% of the time. The next question is what the probability is of such an occurrence relative to the whole spectrum of meteorological occurrences, This is answered simply by multiplying the relative frequency of occurrence by the frequency of occurrence of inversions, 27.9%, so the result is the joint probability of having an inversion which lasts for 20 hr. This value is 0.25%. Such a circumstance then would be likely to occur one time in four hundred. Although the observations show that there have been some long periods of continuous inversions, the observational data from the diffusion studies and wiresonde flights during the summer of 1960 strongly Indicate that many of the situations observed as inversions at the plant site are actually quite shallow and narrow in their extent. In such cases diffusion is probably significantly better a short distance inland than at the site itself, One last point must be made clear concerning the persistence of inversion conditions. Long periods of inversion, that is those over 72 hr in duration, are caused by the large-scale circulation features. These are the inversion conditions that may blanket an entire area such as the whole Ohio River valley or all of the eastern United States, Recorded catastrophic fumnigations such as those at Donoia, the Meuse Valley, London, etc,, are some of the extreme cases. Stagnation of the pressure systems causes these conditions, As of this 15

TABLE VII NUMBER OF OCCURRENCES AND PERCENTAGE FREQUENCY OF OCCURRENCE OF CONTINUOUS INVERSIONS AT ENRICO FERMI PLANT SITE 1 December 1956 - 30 November 1959 (3-Year Summary) Length of Frequency of Occurrence Inversion, No. of No. of with Respect to hr Occurrences Hours Inversions, All Lapse Rates, 1-5 -- 1370 24.3 6.78 6 56 336 16.3 4.55 7 38 266 11.0 3.07 8 37 296 10.8 3.04 9 33 297 9.6 2.68 10 25 250 73. 2.04 11 10 110 2.9 0.81 12 26 312 7.6 2.06 153 14 182 4.1 1o14 14 24 336 7.0 1.95 15 8 120 2.3 0.64 16 11 176 3.2 0.89 17 10 170 2.9 0.81 18 12 216 3.5 0.98 19 7 133 2.0 0.56 20 3 60 0.9 0.25 21 3 63 0.9 0.25 22 1 22 0.3 0.08 23 4 92 1.2 0.33 25 2 50 o.6 0.17 26 2 52 0.6 0.17 27 1 27 0.3 0.08 28 2 56 0.6 0.17 29 2 58 0.6 0.17 33 1 33 0.3.o8 34 1 34 0.3 0.08 35 1 35 0.3 o08 57 1 57 0.3 0.08 59 1 39 0.3 0.08 40 1 40 0.3 0.08 42 1 42 0.3 o.o8 43 1 43 0.3 0.08 47 1 47 0.3 0.08 50 1 5o 0.3 o08 56 1 56 0.3 008 58 1 58 0.3 0.08 65 1 65 0.5 0.08 Totals 344 5629 100.4 27.91 16

NO. OF OCCURENCES -- W o oC o 0 O o o C 0 __ _ _ NCA. 0 - 4O D _ _ 0 _ _ 0 _ 0 0 Iz z 0 0 o 0 o p p 0 0 0000 - r CAcji N'- M 01'^.> b > -4 OD is ) r 0 o 0) N - tD -4 01 _ - C0. 0t0) FREQ. OF OCCURENCE RELATIVE TO ALL LAPSE RATE CATEGORIES IN PERCENT Fig. 6. Percentage frequency of continuous inversions at Enrico Fermi Plant Site: Three-Year Summary, 1956-1959. -~ z~~~l P O0 h) k ~~~~~~~~~~~CI in j bo in P io b s in~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.. a, O R) Il) 01 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~E, C~~~~~~~~~~~~~~~~~~~~~~~~~~~ P UI~~~~~~~~~~~~~~~~~~~~~~~~~~ Plant Site' Three-Year Summary, 1956-1959.

date, no observations of such a condition have been made at the Enrico Fermi Plant site. There is ample reason to believe that under such severe stagnation conditions, diffusion at and near the plant site is likely to be substantially better than in some other areas, Table VIII gives the number of stagnation occurrences that may be expected by months which last 4 days or longer in the Detroit area, These data are based upon 20 years of observation.7 TABLE VIII PERIODS OF STAGNATION BY MONTHS IN THE DETROIT AREA7..Month No, of Occurrences in Month 20 yr January 0 February 0 March 0 April 3 May 0 June 3 July 0 August 4 September 3 October 4 November 0 December 0 As indicated, the stagnant conditions just described are rare but, when they occur, they may apply over a widespread area. It is relevant for the nuclear industry generally, as well as for other industries, to consider precautions during these stagnant periods, such as imposing extra restrictions on the discharge of airborne pollutants. The U. S. Weather Bureau has recently initiated a warning service to alert interested parties about forthcoming periods of stagnant conditions. 18

IIIo PRECIPITATION SUMMARY 1. INTRODUCTION The importance of precipitation as a scavenging or cleansing agent on the atmosphere has been referred to in several previous reports. To emphasize this statement, Fig. 7 is being included in this reporto8 Figure 7 shows the effect of the particle size of the particulate being scavenged, rainfall rate, and duration of rainfall on the scavenging effect of the precipitation, For a typical mid-latitude storm system, the warm frontal rain might occur with a rainfall rate of 10 mm/hr. Such a storm may give rain for a period of at least 2 hro If these values are entered into Fig3 7 using a 44 particle diameter, it is seen that the cleansing action is almost complete~ Another viewpoint may be taken, as follows. If, during such a rainy period, the wind speed averaged 10 mph, then according to Figo 7, at least 50% of the particles 4[t and larger in diameter would be scavenged within 1/4 of an hour or within an area 2-1/2 miles downwind from the point of injection into the atmosphere, If a release occurred in an area such as that which surrounds the plant site, that is, an area of sparse population, a large percentage of the particulate matter would be washed out of the atmosphere before such a particulate cloud would arrive over a large population centero But natural horizontal diffusion is active even under strong inversion conditions if a trajectory of some miles is considered; and if there were no washout immediately, this diffusion would lower the total contaminant over a unit area, so much so that, should washout take place over a population center, the amount of contaminant at and just above the ground would be substantially less than that at a location near the planto As is implied in the above paragraph, washout is an atmospheric process that can be considered as either favorable or unfavorable, depending upon where the washout takes place. Washout over the area surrounding the plant site may be viewed as good relative to a situation where polluted convective clouds of low concentration give showers over the Detroit areao Observations during the summer of 1960 indicate that, with a lake breeze and a light southwesterly gradient, a very typical summertime situation, convective clouds form parallel to the lake shore about 3-4 miles inland in an area of low level convergenceo This would be a preferred area for rain showers and is over a relatively unpopulated area, The maximum amounts of washout are likely to occur with warm frontal rain, which in turn occurs mainly with winds from E to S. Population is sparse in those directions, and even these maximum amounts will be relatively small a few miles from the plant because of the pronounced horizontal diffusion which occurs under all meteorological conditions, even strong inversions3 19

I I I' 1 ii i i i I I Example: To calculate for d IOAu, R= I mm/hr, -and t= * hr, proceed as indicated -0.9 by dashed line t-.7 _3 hr 0.8 — ~ —— f T —-------— ~0.7 0.6 R =0.5 mm/h 20 16 12 8 4 0 20 40 60 80 100 Particle Diameter (,j) Percentage Scavenged Fig. 7. Percentage of particles, of a given diameter, scavenged from cloud by raindrops as function of rainfall rate, R, and time duration ~~0of the rate, t. 20 0.3 0. 5' 0.1 I 20 16 12' 8 4 0 20 40 60 80 100 Particle Diameter (u) Percentage Scavenged Fig. 7. Percentage of particles, of a given diameter, scavenged from cloud by raindrops as function of rainfall rate, R, and time duration of the rate, t.

2, DISTRIBUTION AND FREQUENCY OF PRECIPITATION Table IX and Fig. 8 show the results of three years of precipitation data from the plant site, From Table IX it is seen that the wind speed is 2 mph higher when it rains than the average wind over the entire period. From the diffusion standpoint this is favorable because it causes greater dilution. Secondly, precipitation occurs 12.7% of the time during the entire three-year periods These three years are generally normal, in a climatological sense, so precipitation is not a major factor in the climatology of the plant site area, Figure 8 shows a bimodal distribution of the frequency of occurrence of the wind direction when rain occurs, The easterly mode is caused by prewarm frontal rains while the southwesterly mode is caused by precold frontal shower activity0 Wind speeds are always greater than 11 mph when it rains, regardless of direction. Table IX shows that precipitation with winds from the east to south, which might give warm frontal washout, occurs only a total of 2,7% of the total time. Calm air occurs only 0o2% of the time when it rains or 0,02% of the total time. Certainly this is an insignificant percentage, Thus the probability of widespread washout over the plant site is very small, 21

TABLE IX THE ASSOCIATION OF PRECIPITATION WITH WINDS AT THE ENRICO FERMI SITE 1 Decmeber 1959 - 30 November 1959 (3-Year Summary) Average Average Wind No. of Hours of Precipitation Wind Wind Speed During Observations as Percentage of Direction Speed, Precipitation, During Total Hours of Total mph mph Precipitation Precipitation Hours N 10.1 11.2 105 3.4 0.4 NNE 12.6 12.5 128 4.1 0.5 NE 15.3 16.0 191 6.1 0.8 ENE 14.2 16.8 167 5.3 0.7 E 14.3 17.9 165 5.3 0.7 ESE 12.4 15.3 108 3.4 0.4 SE 12.0 14.4 101 3.2 0.4 SSE 10.8 3.3 123 3.9 0.5 S 10.7 12.5 166 5.3 0.7 SSW 12.3 12.6 229 7.3 0.9 SW 13.0 14.1 302 9.6 1.2 WSW 13.4 14.7 432 13.8 1.7 W 12.7 16.6 349 11.1 1.4 WNW 12.3 14.0 260 8.3 1.0 NW 11.9 12.5 200 6.4 0.8 NNW 10.9 12.9 161 5.1 0.6 Calm 0.0 0.0 7 0.2 0.0 Totals 3194 101.8 12.7 Average 12.4 14.4 22

N N NW NE NW NE S S TOLEDO MUNICIPAL AIRPORT TOLEDO EXPRESS AIRPORT TOLEDO, OHIO TOLEDO, OHIO Annual (I Dec.-30 Nov.) 1950-54 3yr. sum. (I Dec.-30 Nov.) 1956-59 N NW NE / \ I- /y\ ~'15% k mph W -i f 0.2 - -E WW E SW SE S ENRICO FERMI POWER PLANT SITE LAGOONA BEACH, MICHIGAN 3yr. sum. (I Dec. -30 Nov.) 1956-59 C-25 Fig. 8. Percentage frequency of occurrence of winds from 16 directions (rectangles) and corresponding wind speed in mph (heavy lines) with precipitation at Toledo Minicipal Airport, Five-Year Summary, 1950-1954, and Toledo Express Airport, and the Enrico Fermi Site, Three-Year Summary, 1956-1959. 25

IVo EXTREMES OF TEMPERATURE During the course of this project, a deck of IBM data cards was obtained from the Uo SO Weather Bureau state climatologist for the Monroe climatological station which is located at the Monroe water works These cards contain data on maximum and minimum temperatures, occurrence of hail, rain, snow, thunderstorms, tornadoes, etc. Such information. is quite valuable in determining the climatological background of the local area. In addition, it is a source for design criteria concerning the probability of occurrence of the observed meteorological variableso The card deck has been used to determine the probability of extreme values of maximum and minimum temperature at the Monroe station and hence for the area surrounding the station, The extremes of minimum temperature are significant during the months which have the coldest weather, whereas during the summer months the minimum temperature is not as important except perhaps as a relief from a heat waveo Basing the work on such concepts, the months were divided into two categories, those in which a minimum temperature would be important for design or operational routines and those in which a maximum temperature would be important for similar reasonso The month of April was omitted because it is a transitional month when there are neither extremes of heat or coldo The April data added nothing to the computations, whereas September and October data did have significanceo The major value of these graphs is in design work where a certain temperature can or cannot be tolerated because it might cause equipment failure0 The probability of a given temperature and its return period as determined from these computations must be weighed in relation to the cost of a failure and the time lost in repairing such a failureo All the computations are'based upon the methods of extreme value statistics, pioneered by Professor E. Jo Gumbleo9 Details on the computations and analysis of the data may be found in either Gumble9 or Court0-10 1. MAXIMUM TEMPERATURES The months of May, June, July, August, and September are contained in the group for which extremes of maximum temperature were computed0 Figures 9 through 13 show the results of the computationso Figure 9 may be explained as follows, The heavy solid line is the line that expresses the expected extremes0 The probability of not equaling or exceeding a maximum temperature of 95~F in the month of May is 91%o It is seen, also, that a maximum temperature of 95~F is apt to return once in approximately 11 yearso The heavy solid line is drawn from computations, not drawn as a best fit of the plotted values, which are 25

RETURN PERIOD (Years) 1.001 1.01 1.1 1.2 1.3 1.4 1.5 2 3 4 5 10 25 50 100 200 300 400 500 1000 — 8~_-5____-___-_.__ —-____ —- TTT IMaximum Temperatures for May --- 80 85 I- I.1.5 1.0 5 10 20 30 =:40 50:60 70 80 90 95 96 97 9 99 99.5 99.7 99.8 99.9 95 PROBABI00 Fig. 9. Probability and return period of maximum temperatures during May. 105 2 WI|0PROBAB —ILIT —-- -..5 1.O 5 10 20 30 P b l 40 50 6 70 80 90 95 96 97 98 99 99.5 99.7 99.8 99.9,I,, PROBABLITY Doo-/f.,] Fig. 9. Probability and return period of maximum temperatures during May.

RETURN PERIOD (Years) 1.001 1.01 1.1 1.2 1.3 1.4 1.5 2 3 4 5 10 25 50 100 200 300 400 500 1000 -- -- -- -- ----- - - ------ --- -- -- Maximum Temperatures for June - T___ 85 e 5- v -—::4^::Z:::^::::: —-~....2 _ _ 90 ---- o 95- T ~-C i, —t.4 ---- 4 i t - - - -....4 ___W~~~~~~~~~.....L:~~~~~~ 1;~ 4~ - — Lt - i- 100^ i-~ -1 iii?T.C-i i r tt i-i t; —i -' i+'- -,-' I-*.....-.. I10: -;F/:'4;U |j t i;: * —-* *'-l^'^-t-l~~i-j-j'i j I Ii |~- —?!;h 4N 4 -,.- 4i1. ^.-t - - --- -4 -4 ^^^ —+ —i.i i.: 1- t-* t -1 t-. 4441 —m -t.-.. 14 ^r -t —|1~- ----- 1- - I _- -.1.5 10 5 10 20 30 40 50 o 60 70 80 90 95 96 97 98 99 995 9.7 99.8 99. PROBABILITY [?Om/~No+:)] Fig. 10. t~~~~~~~~~~~~~~~~ c —----- — ~-L —~~t- ---- - Fig. 10. Probability and return period of maximum temperatures during June.

RETURN PERIOD (Years) 1.001 1.01 1.1 1.2 1.3 1.4 1.5 2 3 4 5 10 25 50 100 200 300 400 500 1000 - Maximum Temperatures for July - 85 90 0 95- P [_rF 1 00 c[ E 105 I10 iio s~ i * - t ~^i-T -- -----.F.5 1.0 g. 10 20 30:i40 50 1 60 70 so 0o 95 96 97 989 99 995 99S 7 H

RETURN PERIOD (Years) 1.001 1.01 1.1 1.2 1.3 141.5 2 3 4 5 10 25 50 100 200 300 400 500 1000.......kit'i~L: i[~ I! [ r [~!- $" [ ]" i[-i -: ---—........ ~......i- I 4 4: i t 41 r 4i 4 i' l 44I —4- i -44.. 4 —-4 4 4 4 — t — - \- - ~4 144 4 f: j 44 —- — 4 Maximum Temperatures for August - --- - - - |.', - |4 I I. i,- - - i.- - 4 - - - - I I f o r 95~~~~ —— 4144 4 —4 — 4"l'V. r4 - -- -- - - _i i t C;ci* ii K'.. -, - C-1- C t-. 4. -4 4' - -..- S~ 105 -- I I | t —- L iL-cii t t lI-*-i -- -" - - I -- -..-.-. 4..., - -_I I W: fT 1 1-..:' -: Fig. 12 P b i.rpr o aim t: d i Au I t. w- -— i — T- -: t':7 - T — l "x,...:___,~__ 0:',_...:: wl PROBAB I [.1 TY Ft

RETURN PERIOD (Yeors) 1.001 1.01 1.1 1.2 1.3 1.4 1.5 2 3 4 5 10 25 50 100 200 300 400 500 1000 — I —__-.-_.4.- - - -.I- -.- -- I - - - - --- - -- - - - I IMaximum Temperatures for September —-- 85 %L 4_4I- -L^ " i -— i — - J 4 — ~' 90 - -_ I{ m^^^. — I 0 j - * - 5............. - I i. -..j^~-^:t'r:iKj|;. j I <7+-. l:!!::7'! "'t....................... i.....''" i l 3; i ]..iJ. 1 I: ~.14 - { 4 -.-.i iI.i.... IC. } \ - -I..i.......J.........L~~i...I i i j. t- -k4.J i.... 1 -- -~~ —ji ~._.^..,. i..4...4,11,, i::,}i;; ) ]. - -.. F -- -.-T: - - I- 1 i U- -i -,.4...,. -..... H.!. i........ i.....t..). — |...........t.1.O 0 30 50 40 50 I 60 70 80 90 95 96 97 98 99 99.5 99.7 99.8 99.9 I I, 3! PROBABILITY [oo -N +I I)] Fig. 15. Probability and return period of maximum temperatures during September..1.5 1.0 5 10 20 30 w:40 50 60 70 80 90 95 96 97 9 99 99.5 99.7 99. 1 —. —-- i -- lii -i — i i.i_! i 0i PROBABILITY E100, i i. j -\-i + -I;-i~- t i''!' i I I I T~ I j...r~..t.~ i tI —/ j -r~ j. N Fi. 3.Prbbiit adreur prodofmaiumtepeaurs urn Sptmbr

the encircled points, The dashed lines around the solid line are the onesigma confidence limits, That is to say, the probability is 0o68 that a maximum temperature of 95~F ~ 35F will have a return period of 11 years during the month of Mayo The confidence band flares out near the largest and next to largest extreme because there is less stability to these numbers than to other values which have occurred more often. Figures 9 through 13 are based upon 25 years of record, a time period usually quite adequate for most climatological worko 2o MINIMUM TEMPERATURES Figures 14 through 19 are plots of similar data for minimum temperatures during October, November, December, January, February, and Marcho Again the graphs are based on 25 years of observed data. The interpretation of the graphs is the same as with the maximum temperature. By looking at Fig. 19, it can be seen that a temperature of -50F or lower might be expected to occur in March once in 33 years with a probability of 99%; in other words, the value of -50F has a probability of 99% of not being equaled or exceeded in March once in 33 years. In all the plots, Figs. 9 through 19, the observed extremes usually fall within the 68% probability confidence band, indicating that the extreme values are adequately represented by the theory of extreme values, 31

RETURN PERIOD (Yeors) 1.001'Ol! 1.1 1.2 1.3 1.4 1.5 2 3 4 5 10 25 50 100 200 300 400 500 1000 Minimum Temperatures for October s 30-25-.-'.^^jg^"'^ I n 4+ i}Ii! i i i'''- ~~ DI I I 0 20 — _-...-_-Fig. 11k.i - rblity- an d -returnperio- o-4-: minm tep -eraure — -durng-ctoer 40 -..i -c-i-i -~-ii-t I~i:I I — — I- i —: —--- -— r —-'-4 i -— f-~ —T-l —- j I-f — -- -_ —--:4: ~ —.-iti]-.^ij-; ii |. -*f i* *^44- $~ —i -T~~~~~~4- 4 —.~ — 1- - 4- -'t - -- - - -~ — 1 - - ~+- ~- - - -- ----- i —-- - - 2 ~ ~ ~^ PROBABILITY [ON +] Fig. 14. Probability and return period of minimum temperatures during October.

1.001 l.oi 1.1 1.2 1.3 1.41.5 2 3 4 5 RETURN PERIOD (Years) 0 1 — -.,,1 I. 2 L 3 1.4 |5 2 3 4 | 2 5 50 t00 200 300 400 500 1000 Minimum Temperatures for November ^ ^ ii IZ~tZ_"-"~ - -T~~~~~~~~~~~~~~~~~~~~~~~~~~~~~---. --- --- __ - ___._.____ _,- -- --- ----.- - - - - ^^ — --- -4- ^._ —^ —— 1- c - -— ___ _ _ __ I^ 0 -- - I L 1 -+4^ - -^ 4 - - - - - _ _. _ _ _ _ _ ^ _ _ ~ _ " ~ -07^*-;*' *!- _ - - -l 5 _A-t ^ 1 ^^-^E11^~^^^!1^,^^*^^^^:~^^::^ —— ~~~~~~~~~~~~~~~~~~~~~~~~ t-ft-t -t -C —_ =:^ _ LL t -0- - 4J t IY: ~::: i-:t i< f<- f - ni*-l: — t- H —^^r ".t-p j.; --- - 20 - t — ^C fft-:i:~lll: 4 IIII r lxn 11II Y II^ - Tll >r —— t- ct —i- 4 -- t —-~ = = 5 I If 1 w o= — i — C-i tt ~ ~ ~ ~ j,:I-4 4-IJ Al — ------ - --- an - -+-r-+- - t - --- ___4_.__ - -t ti tx-nl 4 *? -i *- -- -—... - — _ --- - - -^ a> — ]-t —' t7 - 4-f t- -l - -- 4 - - --— C _ I I I - -! - -- - -I- -- i -. ___ ___ __ - -tTV-' tT-nti-1; — - ~ —-T tiF::!::;:,.; ^-!.- - T j -ti + _1 L- t —C - i*t[ - = -4-~~~~~~~~~~~~~~~~~~~~~~~~~~~-~ W I, I I -- - e — I —,- — d 15; _0~ir0 * 0 0' 0 00 —' - I I -- - _ - _______ — +_ __ 4- 4- -T - -4 - --- +~ z ~ 90 95 9 97 98 - T- +If- + -t-itt -ti-l — i~~~~~ — - 44 —-~ii- — tt i 71 ~ ~ ~ ~ ~ ~ ~ ~ — j-i -- _- - i I I ~ ~ 0 < s; WI PROBABILITYE [^ _(N + 1)]9' "Fi. 5. P ob bi it nd re u n er od o m ni mi t mp ra u es du in ov mb r Fig. 15. Probability and return period of minimum temperatures during Novembei

RETURN PERIOD (Years) 1001 101 i 1 1.2 1.3 1.4 15 2 3 4 5 10 25 50 100 200 300 400 500 1000 [i ii. | -I i' rlr Chi ti, I ti.I,...... -- - --- ~ --' I -'/': i f tJ;...i,-, "::.!,..-:;. i * M tM Mimimum Temperatures for December' t "'I -- - L- [ t: *' i!: - --;?'. - i * j, 1 -15 -- t I I W.:'-+X' it-ti t j i -__ /. j t it" -lor'T 7 " -t c I": di i L 1 I -5 —- --— J — ---- -r-. —-~-j —-- _ _ t-._ —-__ L H^: i:'; i:' uI |- j ~, t |I~,L4 --—.-. —-i-j --— iij —.. I — -- t i __ L -i -Jc t 0ii 4 _,i, -;tt -ST. j0 -, -,: j. i j I 0;:,:;.:... w<2 _0 I d S t. _'.... ~.......-; —:-i.............j i i _ - 1;t -*T ^*^^ Q\\\\^'''^.\'\\ ^i\\\\: \,\:.^.,.'.^ -: -, I r... I i! - t't' r't' t 1F l6 P b v, ]... -I I....-.... ~' 4" 5 j..- ~. 4 ~, - tI,.1 060 70 80 90 95 96 97 98 99 99.5 99.7 Fig 16 Prbblt adrtr pro miiuePRtBABu LidTuY D e m]b F ig. 16. L j 1t i i i t t c~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 70 0 9 9 97 9899 9. 997 5 1.0 5 1 0 20 30 40 50 60 90~~~~~~~~~~~~~~~~~~~ r- — Ct-~ —l — i-.~-c —t —c-i 10 — i-~ t --— "'-. —'t —-— ~~~~~~~~PROBBILIY 10- TIN Fi.16 roailt adreun eio fmiiumtmprtue drngDcebr

RETURN PERIOD (Years) 1.001 1.01 1.1 1.2 1.3 1.4 1.5 2 3 4 5 10 25 50 100 200 300 400 500 1000 _tt. t ij; - - _ _ _ _: _ inimum Temperatures for January 4 - 4 1 ~-1 < -2G-'' 1 -- - — I1' ---- -- -- -25 7 t I 4 4_____-__H+.-^fl - 4j.4. b.... LP. -....... Fl+ ig 17..I Prbblt and return, perio o minimum tempert i Ja r: 02r"~; i ^"^^^^in'^'n^'nT^ i — - U.[ _ __ ^ j H r 1 1,,,i t i tii i U t i i- [ * li ii ijii. -; " - "';.. "_ 0 i:': i-4 — Fig. 17. t

RETURN PERIOD (Years) 1.001 1.01 1.l 1.2 1.3 1.4 1 5 2 3 4 5 10 25 50 100 200 300 400 500 1000 + "4* Minimum Temperatures for February -- i ^s-^ —-^^-:!!^!!.!^^:^1'!^^'' L4'!/''" "-' ~: —-' -::. i i u.- 4 i4 l —. i; 1. - i <.^ —'- Iv3 ii- l — -_ii 4-li - ___ _ 4___ 25 _ _ i I!::^1: t- - - - -. -' *'*: * * - -4 ^^- ^; ^ - I i T A 1- --- 4 41 I E- "!:' f - ~ ~i. I j* *: * **^ r ^ 07. +... -~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~15~~~~~~~~~~ 4-' —' —'~-4 4 i! -j.hi ~!'lr iH-! iyi qi.t t B i I;: -::::.::}^ 1' — r' t' 1-..LL.54 10 5 10 20 34 i 40 1 6 \ 4- L 4L - T t — T4t t 4 I-i'1-5~~~~~2a. 4 PROBABILITY F m, 1 i i -- [ 5' J 7 7 J i t + 4 i i t~_~ P' BA... LIT -F - P b l a Ri~~~~~a.:18.~~~~~~'

RETURN PERIOD (Years) 1.001 1i I01 1.1 1.2 1.3 1.41.5 2 3 4 5 10 25 50 100 200 300 400 500 1000 iL Ole~~~~~~~~~~~~~~~~~~~~~fi i t' 1 IC I — — e ~ - >777P 7; -- Minimum Temperatures for March -4 —--------- -- - -—'11-![' -r -T — i- - - r- j,- -— i-^ --- -- - I~ I - | i.t —r __ ^,-... -.. -^ 1,,..., t -^ ___ /......__ _ _ d I — 7 —-4 ~ ~ ~ ~ ~ ~ I -101 _0 Q_:" " -f " "' "'~ Tt ^ --:t - ] -- {;-: -1-* * ^ ^ -; ^ ^ -., ^.........,...- ---._ _.... _~ __ I- -{ -~ + -4— 1- _ -ti+ 4-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~uj!`r. t-l^- -— ^ --- — ^ —!-^ -L^'-^'^ O - ^ ^' — I -c- *t -^...'-..'.-4 - J - I_. i- L.-K __-. -.-.- - -...,...._.__. —-- - r ^ T-t *~'"Ftt^"i ** * -|'~^ - 4*^ p^< *^<>- i'- **t —7 c ___ -_ __ ___ __ _-_ - L.T i lLL...Lj0,j _ ^ ^ " f- r-;^*, ^' ^ " - ":-'* " "-;r ~[ ~~ " ~; ~ r ~".t':\- -.- i —-i - \.. r-...L:HiMB^ ^ ^ ~i- -^' —W ~ ^-: ^^ — lji^' t ^ 1 -- ti- -L —it i- t - i — - i n IT -'- -^ " {''t " "....~ 4. - *i.5 1.0 5 10 20 30 ^40 50 60 70 80 90 95 96 97 98 99 99.5 99.7 99.8 99.9 LL 1 Fig. 19. Probability and return.period of minimum temperatures during March. 5! 0 -7~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~J D r _-i-.~jfI0 I~ ~ ~ ~ ~ ~~~~~~ -+ - l~~ f —i' t-r- -t -C- L.-C-I r ~~~~~~~~~~C1 f-~ W;il- i I iI I i t~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ w~~~~~~~~~~ rt f- /L-L. 3 o r C- — C —-l i_ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ —- I t c~ — ---------- w i: r- -t -I~~~~~~~~~~~~~~~~~~~~~~~~~t - 4- 4 —4 t;-t i` —- I L ii; I 1 f t 1 i- L __L: f;f:1;.::_I ItPROBABILTY [ 00 Fig. 19. Pobability nd returnreriod of mnimum tempratures duing March

V. SUMMARY OF DIFFUSION STUDIES A diffusion study was carried out at the plant site during 1959-1960. Details of instrumentation, days when measurements were made and computed values of SuttonIs parameters, Cz and Cy, are found in Refso 11-13o The emphasis was on measurements when diffusion was likely to be poor, and accordingly several experiments were carried out under inversion conditions, The method utilized in the analysis of data from the diffusion experiments was to assume a value of Sutton's n and then to compute Cy and Cz from the observed tracer particle counts. The value of n was adjusted until the values of Cy and Cz were consistent with theory and with other known observations. The results are summarized in Table Xo TABLE X MEAN COMPUTED VALUES OF Cz AND Cy AT THE ENRICO FERMI PLANT SITE Day n C Cy 6 August 1959 (run Noo 1) 0.25 0.15 o.38 27 November 1959 0.20 0,14 o 64 4 February 1960 0,30 0 08 0.54 3 April 1960 0 30 0o09 0o61 8 May 1960 0.20 0013 0.44 25 June 1960 0.23 0.14 0537 Mass balance studies in which the measured mass of tracer material emitted per unit time from the source was compared with the measured mass of tracer material passing per unit time through vertical planes normal to the wind at various distances downwind from the source confirmed the validity of the experimental techniques used. On the basis of the diffusion studies, several facts are apparentO The first and most important one is that a lake breeze induced inversion does not limit diffusion as much as one might anticipate at first sight, Both the experiments of 4 February 1960 and of 3 April 1960 were held under a lake breeze inversion condition, Using the relatively low value of 0~30 for Sutton's n, the vertical diffusion coefficient Cz is small, as anticipated, but the horizontal diffusion coefficient, Cy, is larger than would be expectedo Increas39

ing n to a more reasonable value for an inversion, say n = 0.50, gives values of Cz and Cy that are larger than any acceptable values of the Sutton parameterso The conclusion then is that the mean values of the computed Cy and C are showing the integrated effects of changes or transitions in the diffusion regime as the tracer is carried inland. The fact that there is usually either a heat source or a heat sink present near the plant site, i,e,, land and lake temperatures are usually different, resulting in density gradient which causes movement of the air, This movement ensures that adequate diffusion takes place most of the time. During the course of the diffusion study program, the contractor requested that the authors furnish diffusion coefficients to them for use in a study of the evaluation of a hypothetical contained accident at the Enrico Fermi reactor. Table XI is a table of the submitted values used in the hypothetical study, TABLE XI DIFFUSION PARAMETERS USED IN AN EVALUATION OF A HYPOTHETICAL CONTAINED ACCIDENT AT THE ENRICO FERMI REACTOR Diffusion Parameters Lapse Rate n Cz(mn/2) Cy(mn/2) Inversion 0.55 0.08 o040 Weak Lapse o025 0.35 0.40 Strong Lapse 0.20 0,40 Oo40 At first glance, these values do not seem to compare well with the values in Table X which were computed from the diffusion study observations. There are several reasons for this apparent discrepancy, It is known that the values of the diffusion parameters vary with sampling time. The recommended values from Table XI are to be used for sampling times on the order of one hour since the problem is that of atmospheric diffusion after leakage into the atmosphere from the containment vessel, In essence then, any problem for which the values from Table XI should be used is a problem concerning a continuous source. An instrument for measuring the concentration in the plume would be exposed to the plume passing over it for. a period of hours. On the other hand, the sampling periods during the diffusion study ranged from 15 sec to 1 min. The result is that the computed values of the diffusion coefficients from the diffusion study will be applicable to such a problem as when a puff passes over the 40

sampler in a matter of several minutes, Thus the recommended parameters from Table XI differ from the values in Table X because of sampling time. The values that were used in the computations of the leakage from the containment vessel had to be conservative, that is, the computed concentrations had to be higher than those actually observed or expected to be observed. At first glance, it can be said that as Cz and Cy decrease the computed concentration increases. If the values of Cz and Cy in Table XI are smaller than those in Table X, then the computed concentrations will be high and thus conservative. It is noted that Cy in Table XI is, in fact, lower than the values of Cy from Table X. This is not true of CZ, however. The difference is due to the sampling time; the method of airplane sampling used prevents corresponding differences in Cyo The values of Cz recommended in Table XI are in good agreement with hourly values based upon other field experiments.14 The value of n should also be looked at for conservatism. As n gets larger, the computed concentrations get larger, too, The values of n in Table XI are either the same or larger than the assumed values in Table Xo In summary, then, the recommended values of Suttonts parameters in Table XI are generally conservative. 41

VI. STACK DESIGN Meteorological criteria were used in determining the stack design factors of location, height, diameter, and effluent velocity.15 Effluent temperature and emission rate were fixed prior to the meteorological computations, Maximum surface concentrations were computed under various meteorological conditions. Those conditions were fumigation (Types I and III), looping of the plume, aerodynamic downwash, trapping, and deposition. The percentage of the time that such meteorological conditions were anticipated to occur was discussed, based upon the meteorological observations made at the plant site, 45

VII. GENERAL CONCLUSIONS The Enrico Fermi Atomic Power Plant is located on flat land at the western end of Lake Erie. The level of the land does not rise over 30 ft for a distance of 5 miles from the site or more than 100 ft for 20 miles from the site. Thus there is no physical trapping of the air as in a river valley. The presence of the lake and the flatness of the land contribute to above-average wind speeds and to relatively infrequent calm conditions. The yearly average wind speed at 102 ft above ground level is 12.4 mph. The wind speed averages 4 mph or greater 96% of the time. These winds provide good dilution potential for any airborne contaminant. There is a lake-:breeze-induced inversion of limited vertical extent and of short duration which reduces diffusion for 1 to 2 miles inland from the plant. Advective inversions as measured at the shore line tower occur mainly with on-shore winds. Most of the time, however, these inversions are broken up by mechanical or thermal turbulence before the air has advanced more than a mile or two inland. Diffusion improves rapidly as these inversions are destroyed. In addition, the wind speed during inversion conditions is higher than is normally found, being 10.2 mph on the average. In summary, the diffusion characteristics of the site are representative of those along a typical western shoreline of one of the Great Lakes. From a meteorological viewpoint there is no reason to disqualify the Enrico Fermi Site as a nuclear reactor location. In fact, there are several meteorological factors, such as prevailing winds from the SSW through the WNW and the lack of calm conditions, that tend to favor this location. 45

REFERENCES 1. Hewson, E. Wo, and Go C. Gill, 1957, Meteorological Installation and Analysis, UMRI Report 2515-1-P, Ann Arbor, Michigano 2. Hewson, E. W., Go C. Gill, and Jo J. Be Worth, 1958, Meteorological Analysis, UMRI Report 2515-2-P, Ann Arbor, Michigan. 53 Hewson, E. W., Go Co Gill, and H. Wo Baynton, 1959, Meteorological Analysis, UMRI Report 2515-3-P, Ann Arbor, Michigan. 4. Hewson, Eo Wo, G. C. Gill, Eo We Bierly, and Io Spickler, 1960, Meteorological Analysis, UMRI Report 2515-4-P, Ann Arbor, Michigano 5. Hewson, Eo Wo, Go C. Gill, Eo Wo Bierly, and I. Spickler, 1960, Meteorological Analysis, UMRI Report 2515-5-P, Ann Arbor, Michigan. 6. Munn, R. E,, and Mo Katz, 1960, "Air Pollution Levels Associated with a 49-Hr Inversion at Detroit-Windsor," Bulletin of the AMS, 41, 245-249. 7. Korshover, J., 1959, Synoptic Climatology of Stagnating Anticyclones East of the Rocky Mountains in the United States for the Period, 1936-56, unpublished manuscript, U. S. Weather Bureau, Washington, Do C. 8. Greenfield, SO Mo, 1957, "Rain Scavenging of Radioactive Particulate Matter from the Atmosphere," Journal of Meteorology, 14, 115-125. 9. Gumble, E, Jo, 1958, Statistics of Extremes, Columbia University Press, New York, 10. Court, A., 1952, "Some New Statistical Techniques in Geophysics," Advances in Geophysics, 1, 45-85~ 11. Hewson, Eo Wo, Go Co Gill, and E. W. Bierly, 1960, Atmospheric Diffusion Study at the Enrico Fermi Nuclear Reactor Site, Technical Installation and Preliminary Analysis, UMRI Report 2728-1-P, Ann Arbor, Michigan. 12. Hewson, E. W., G. Co Gill, and Eo Wo Bierly, 1960, Atmospheric Diffusion Study at the Enrico Fermi Nuclear Reactor Site, A Qualitative Analysis of Further Diffusion Experiments, UMRI Report 2728-2-P, Ann Arbor,, Michigan. 13. Hewson, Eo W., Go C. Gill, and E. Wo Bierly, 1960, Atmospheric Diffusion Study at the Enrico Fermi Nuclear Reactor Site, A Quantitative Analysis of Diffusion, UMRI Report 2728-3-T, Ann Arbor, Michigano 47

14. U. S. Atomic Energy Commission, 1957, Theoretical Possibilities and Consequences of Major Accidents in Lare Nuclear Power Plants, WASH-740. 15. Hewson, E. W., G. C. Gill, and E. W. Bierly, 1959, Maximum Ground Concentrations in Relation to Stack Heights under Various Meteorological Conditions at the Enrico Fermi Nuclear Power Plant near Monroe, Michigan, PRDC Special Report. 48