Regulatory Guide 1.135
| ML003740143 | |
| Person / Time | |
|---|---|
| Issue date: | 09/30/1977 |
| From: | Office of Nuclear Regulatory Research |
| To: | |
| References | |
| RG-1.135 | |
| Download: ML003740143 (5) | |
U.S. NUCLEAR REGULATORY COMMISSION
September 1977
4 REGULATORY GUIDE
, OFFICE OF STANDARDS DEVELOPMENT
REGULATORY GUIDE 1.135 NORMAL WATER LEVEL AND DISCHARGE AT NUCLEAR POWER PLANTS
A. INTRODUCTION
accidental events on structures and systems at nuclear power plants. This guide also suggests data General Design Criterion 2, "Design Bases for sources and acceptable methods for determining such Protection Against Natural Phenomena," of Appen water levels and discharges.
dix A, "General Design Criteria for Nuclear Power Plants," to 10 CFR Part 50, "Licensing of Produc tion and Utilization Facilities," requires that struc
B. DISCUSSION
tures, systems, and components important to safety Some structures and systems .. 'Mnost nuclear be designed to withstand the effects of natural power plants are continually suif phenomena such as floods, tsunami, and seiches o uctuating without loss of capability to perform their safety live loads that result from variions i ound water levels, discharges -ld a Is isn'treams, or functions. Criterion 2 also requires that design bases water below levels in lakes anid for these structures, systems, and components reflect: tend grousd surfac~nto theuctures that ex local ground water aquifer*-(o1ý` 1 A # ifers must withstand the
1. Appropriate consideration of the most severe of resulting hydftaticýressti'e. Structures that are af the natural phenomena that have been historically fected b1:.zurfhceý*atetdischarges and levels include reported for the site and surrounding region, with damss, 6*-.,*a; aters, canals, and intake and dis sufficient margin for the limited accuracy and quan tity of the historical data and the period of time in cl*4e s*c*.cids. In considering the effects of design bas eientsiuch as earthquakes, tornadoes, and hur which the data have been accumulated, ricwsnt&;and the effects of design basis accidental
2. Appropriate combinations of the effects,,of n*6 e*ffVw, 'ts such as plane crashes, transportation acci
"."dents, explosions, fires, and loss-of-coolant accidents mal and accident conditions with the effectsof the'.
(LOCAs), the designer must include in his calcula natural phenomena, and ... tions the loads caused by the water level (or dis
3. The importance of the safety furfons i6 be per charge), as well as the load on the structure or system formed. caused by the design basis event being considered.
This water level (or discharge) is referred to in this Paragraph 100.1 r bftOCFR Part 100, "Reactor guide as the normal water level (or discharge).
Site Criteria." requi t'* [k., ysical characteristics of the site, Inc] I ei lo&g' meteorology, geology, Since the design basis events have a very low and hydro y, b k e-into account in determining probability of occurrence, the water level (or dis the ac. ta 'ty o ',site for a nuclear power reactor. charge) used in combination with the design basis event need not represent a condition with low Th uide sets forth methods acceptable to the Design basis events are those natural events that there is virtually NRC f for determining normal ground and sur no risk of exceeding, such as the Probable Maximum Hurricane, face wat r levels and surface water discharges for use Safe Shutdown Earthquake, and other natural events of similar in analyses of the. effect of design basis natural and low probability of occurrence, and accidental events, such as ex plosions, plane crashes, and fires.
USNRC REGULATORY GUIDES Comments should be sent to the Secretary of the Commission. US. Nuclear Regu Regulatory Guides are issued to describe and make available to the public methods latory Commission, Washington. D.C. 20555, Attention.
acceptable to the NRC staff of implementing specific parts of the Docketing and Service Commission's Branch.
regulations, to delineate techniques used by the staff in evaluating specific problems or postulated accidents, or to provide guidance to applicants. Regulatory Guides The guides are issued n the following ten broad divisions are not substitutes for regulations, and compliance with them is riot required 1. Power Reactors 6. Products Methods and solutions different from those set out in the guides will be accept- 2. Research and Test Reactors 7. Transportation able if they provide a basis for the findings requisite to the issuance or continuance 3. Fuels and Materials Facilities 8. Occupational Health of a permit or license by the Commission.
4. Environmental and Siting 9. Antitrust Review Comments and suggestions for improvements in these guides are encouraged B. Materials and Plant Protection tO. Generat times, and guides will be revised, as appropriate, to accommodate commentsat all and Requests for single copies of issued guides (which may be reproduced)
to reflect new information or experience. However, comments on this guide,if or for place.
ment on an automatic distribution list for single copies uf future guides received within about two months after its issuance, will be particularly in specific useful in divisions should be made in writing to the US. Nuclear Regulatory evaluating the need for an early revision. Commision.
Washington, D.C. 20555, Attention: Director. Division of Document Control.
is quite possible that the influence of flood levels on probability of occurrence. As used in this guide, the normal ground water levels must be considered.
term normal water level (or discharge) means that water level (or discharge) that has a probability of ap The following discussions of various water bodies proximately 0.5 of occurrence at the time of interest. assume that the seasonality of the design events are properly taken into account.
Some design basis events may actually consist of a series of events such as an earthluake followed by 2. Nontidal Streams one or more aftershocks. This fact is important if the earthquake could affect the normal water level (or Data on stream discharge rates are readily discharge) and result in a different water level (or dis available for thousands of stream-gaging stations charge) when the aftershocks occur. For example, the operated by the U.S. Geological Survey; U.S. Army initial earthquake could affect an underdrain system Corps of Engineers; Bureau of Reclamation; other to lower ground water levels at tjhe time of the Federal, State, and local agencies; power companies;
aftershocks. Similarly, a dam failure caused by the in and other organizations. Most such data are in the itial earthquake could substantially change the dis form of mean daily discharges for varying periods of charge in a river at the time of the aftershocks. Under record. Hydrology textbooks contain acceptable such conditions, the normal water level (or discharge) standard methods for transposing these data from the should be assumed only at the time of the initial gaging stations to the plant site. Thus, relatively event, and the water level (or discharge) during the reliable information on stream discharge may be ob aftershocks should be determined by analysis of the tained for a plant site located on a nontidal stream.
effect of the initial event. Short-term discharge data are also frequently col lected at the sites and should be used to verify the transposed data.
1. Seasonal Occurrence of Design Basis Events A period of record of at least 50 years is desirable Earthquakes, plane crashes, transportation acci for use in determining the normal water level and dis dents, explosions, fires, and LOCAs may occur at any charge at sites on nontidal streams. For many loca time. However, water levels (or discharges) may vary tions, however, such long-term records do not exist.
considerably during any year. Accordingly, when In such cases, periods of record of at least 12 years determining normal water levels (or discharges) for may be acceptable if it can be demonstrated that the these events by the methods recommended in this period used includes major wet and dry periods.
guide, water levels (or discharges) for the entire year Where shorter-term data or no data exist, records should be considered. should be estimated on the basis of weather records and records for other streams in the region. In all On the other hand, floods, hurricanes, and tor cases, watershed development conditions anticipated nadoes are seasonal in most parts of the United during the plant lifetime should be considered.
States. In determining normal water levels (or dis charges) for these seasonal events, such seasonality For nontidal streams, the median discharge will be should be taken into account. For example, tor considered the normal discharge. The median dis nadoes do not generally occur during the winter charge may be determined as the 50% point on a months in the northern states. Therefore, the normal flow-duration curve (discharge v
s. percent of days
2 water level (or discharge) to be considered with tor when indicated discharge was equalled or exceeded).
nado effects should be derived only from data for Acceptable methods of constructing a flow-duration those months in which tornadoes may be expected to curve are outlined in many publications, e.g., U.S.
occur. Geological Survey Water-Supply Paper 1542-A,
"Flow Duration Curves," 1959.1 Similarly, in Florida hurricanes have been recorded in virtually every month, but they are more It would be desirable to determine the normal likely to occur during the late summer and early fall, water level on nontidal streams by similar means, i.e.,
which is coincidentally the period of the year in which use of a stage-duration curve (water level vs. percent high water levels (or discharges) are most likely to oc of days when indicated level was equalled or ex cur. Accordingly, for this type of seasonal variation, ceeded). However, stages (water levels) at gaging sta the normal water level (or discharge) should be deter tions are not as readily available as are the cor mined only for that period of the year during which
2 The U. S. Geological Survey computes data for most of their the event is most likely to occur or for the ap propriate period as indicated by studies of the area. gaging stations from which a flow-duration curve may be plotted.
It should be noted, however, that such a curve will cover only the period of record available for the gaging station.
Floods are seasonal in many parts of the United
' Copies are available from the Branch of Distribution, U.S.
States. While the surface water levels (or discharges) Geological Survey, 1200 South Eads Street, Arlington, VA 22202.
addressed in this guide are not applicable to floods, it
1.135-2
responding discharge data. Therefore, the NRC staff will accept, as the normal water level, the water level mean monthly records is sufficiently accurate or corresponding to the normal discharge. Computation whether mean daily data should be used. Such a of the normal water level by this method requires the study could consist of computing medians for a short establishment of the relationship between river level period (e.g., two or three years) from both monthly and discharge (the rating curve) and the application and daily information and comparing the results.
of that relationship to the normal discharge. If the 5. Regulated Lakes and Reservoirs stream has a movable bed, allowance should be made for shifting of the rating curve. At their outlets many lakes have dams or other regulating structures, which in effect convert the lake
3. The Great Lakes into a reservoir. Regulated lakes are therefore addres sed as reservoirs in this guide.
The levels of some of the Great Lakes have shown upward or downward trends over long periods. For Most reservoirs have at least one level designated example, Lake Michigan-Huron has shown a slight as "normal pool" elevation. For some reservoirs, downward trend over a period of more than 100 more than one elevation, usually varying with the years. Such trends should be taken into account in time of the year, may be considered normal pool.
determining the normal level. An acceptable Those reservoirs that have reserved flood control procedure is to adjust the last 50 years of data for the space are frequently of this type. For example, a indicated trend before determining the median level.
reservoir in the Sierra Nevada of California might This adjusted 50-year median level should be con have a given normal pool level for the July-to sidered as representing the normal level. November period, followed by a November-to December period when the level must be drawn down Lunar tides in the Great Lakes may be ignored to accommodate potential flood runoff, and a because their size is negligible. December-to-June period when the level must be Because of the size of the Great Lakes, their water kept as closely as possible to the drawndown level but levels change very slowly. Therefore, it is not neces may rise to as high as the top of the flood control sary to determine the normal water level from daily pool.
data. Mean monthly values are of sufficient accuracy.
The Detroit District, U.S. Army Corps of Engineers, Normal water level should be considered the nor mal pool level of the reservoir, with due considera collects and publishes monthly data on the levels of tion for variations in normal pool level during the the Great Lakes. year. If the reservoir includes flood control storage, the median of the flood control pool levels for the life
4. Unregulated Lakes of the reservoir should be considered normal for As used in this guide, the term unregulatedlakes events that are likely to occur during the flood means those natural lakes whose outlets are virtually season.
uncontrolled by dams or other structures and those natural lakes that are not appreciably affected by Records of levels of regulated lakes and reservoirs diversions into or out of the lake. Regulatedlakes are are usually collected by the operator of the reservoir.
discussed as reservoirs in this guide. In some cases, these data are published by the U. S.
Geological Survey, U.S. Army Corps of Engineers, A median level based on 50 years of data is accep or other Federal, State, or local agencies.
table as the normal water level. This level can be determined from long-term records of the lake level. 6. Oceans If only short-term records are available, they may be extended by estimating longer-term data on the basis A complete tide cycle averages 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />
25 of records for other lakes, stream discharge and at most points on the shores of the .Unitedminutes States.
ground water data, and weather records. As with Because of this relatively rapid rate of change of ocean levels, the duration of the design basis event nontidal streams, a shorter record may be acceptable if it can be demonstrated that major wet and dry should be considered. Most such events are of short duration, ranging from essentially instantaneous periods are included. As with the Great Lakes, if a trend can be identified, the data should be adjusted (e.g., plane crashes, explosions, and most transporta for this trend. Consideration should also be given to tion accidents) to a few seconds or minutes (e.g.,
earthquakes, tornadoes, and some transportation ac potential changes in water use that could significantly cidents). However, hurricanes and fires may last suf alter this median.
ficiently long to span the entire tidal range from low In many cases, mean monthly data may be used to to high.
compute the median level. However, a study should be made to determine whether the median based on For short duration events, the median tide level may be considered as the normal water level. For
5-3
longer duration events, however, it is probable that A period of record of 20 years (to account for the tide will be above (or below) the median level for cyclic tidal variations) is acceptable if it can be at least part of the time of the occurrence of the event. demonstrated that the period includes major wet and For these events, therefore, median higher high tide dry periods.
or median lower low tide (see discussion below)
should be used as the normal. The choice of median 8. Ground Water higher high tide or median lower low tide will depend on the event being considered. For example, hur Normal ground wa'ýýr level should be the median ricane winds blowing seaward from land will lower level over a 20-year period. This length of record will the level of the ocean (setdown). If this will have an not be available at most sites for nuclear power effect on a safety-related water supply, the normal plants. However, most sites will have one or more water level should be the median lower low tide. short-term records of a year or two. The staff will ac cept extension of these short-term records by correla High and low tides do not consistently reach the tion with longer records in the region. The effect of same levels. The two high tides during a complete 24 future changes in water use should be considered.
hour 50-minute cycle at some locations are of dif The U.S. Geological Survey collects and publishes ferent heights, as are the two low tides. The higher of records for thousands of observation wells the two high tides is designated as higher high tide throughout the United States and may advise where and the lower of the two low tides as lower low tide. other data in the region may be obtained.
Other cyclic variations have also been identified.
For those sites at which the applicant plans to Tidal heights vary during each lunar month from spring (high) to neap (low) range. The tidal range also lower the ground water level permanently by use of has a cyclic variation of somewhat more than 19 an underdrain system, either pumped or gravity flow, years. In addition, tidal heights are affected by varia the staff will accept, as normal, the anticipated lower tions in barometric pressure and by wind. These water level for which the system is designed.
meteorological phenomena cause the principal dif Ground water levels to be used for soil liquefaction ferences between predicted and recorded tidal analyses will be discussed in a regulatory guide now heights.
under preparation on liquefaction.
In order to ensure that these variations are con sidered, normal tide levels should be computed from
C. REGULATORY POSITION
a continuous record of at least 20 years.
Analyses of the capability of safety-related struc A long-term tide record is not available at the exact tures, systems, and components to resist combina location of many nuclear power plant sites. Usually, tions of loading conditions often require identifica however, a short-term record will have been collected tion of normal water levels or discharges. For the at the site. In this case, the short-term record may be purposes of this guide, normal water level (or dis extended by correlation with one or more records for charge) means the water level (or discharge) that can nearby long-term tide stations. be expected at the time of occurrence of design basis natural or accidental events.' Subject to the excep Records of tidal heights are collected and main tions below, the normal water level (or discharge) will tained by several agencies but principally by the generally be the median water level (or discharge),
National Oceanic and Atmospheric Administration i.e., that water level (or discharge) that will be equal (NOAA) in the Department of Commerce. This led or exceeded 50% of the time.
agency regularly estimates mean high tide or mean higher high tide, primarily for coastal locations. Data 1. Seasonal Occurrence of Design Basis Events for other locations may be estimated, using the rela tions given in tide tables, by interpolation. If the design basis event under consideration is seasonal, the normal water level (or discharge) should
7.- Estuaries be determined for that part of the year in which the event is likely to occur.
Both freshwater runoff and lunar tides affect water levels in estuaries. Most estuaries are gaged by the 2. Nontidal Streams U.S. Geological Survey, National Oceanic and At mospheric Administration, U.S. Army Corps of The normal discharge should be the median dis Engineers, or others. A stage-duration curve should charge based on a period of mean daily discharge be used to estimate the median water level as the nor record, preferably at least 50 years in length. The nor mal water level. This procedure may be used for loca mal water level should be either (a) the median mean tions upstream and downstream from the site with daily water level based on a record of preferably at the normal water level at the site estimated by inter least 50 years or (b) the water level corresponding to polation. ' See Footnote 1.
1.135-4
normal discharge, including consideration of shifting rating curves for streams with movable beds. A 7. Estuaries shorter period of record (not less than 12 years) is ac ceptable if longer records do not exist and it can, be Normal water level should be the median level based on a 20-year period of record. For longer demonstrated that the period used includes major wet duration events, median higher high or lower low tide and dry periods. based on a 20-year period of record should be sub stituted for median tide level.
3. The Great Lakes
8. Ground Water Normal water level should be the median mean monthly level based on the latest 50 years of record, with suitable adjustment for demonstrated long-term Normal ground water level should be the median level based on 20 years of data. If permanent lower trends. ing of the ground water level by using underdrains is
4. Unregulated Lakes (Other Than the Great Lakes) planned, normal level should be the design lowered level.
Normal water level should be the median based preferably on 50 years of data. Unless it can be 9. Availability of Data demonstrated that the median based on mean monthly levels is sufficiently accurate, mean daily If the minimum periods of record specified in the foregoing positions are not available, standard data should be used. If long-term trends can be iden methods may be used to transpose or extend the tified, the 50-year median should be adjusted for available data. Full explanations of the methods used these trends. A shorter period of record (not less than for such procedures should be furnished.
12 years) is acceptable if longer records do not exist and it can be demonstrated that the period used in cludes major wet and dry periods.
5. Regulated Lakes and Reservoirs
D. IMPLEMENTATION
Normal water level should be the designated nor mal pool elevation. There may be different The purpose of this section is to provide informa designated normal pool elevations at different times tion to applicants regarding the NRC staffs plans for of the year. If the reservoir contains flood control using this regulatory guide.
space, the normal water level during the flood season should be the median of the flood levels attained dur Except in those cases in which the applicant proposes an acceptable alternative method for com ing the life of the reservoir.
plying with specified portions of the Commission's
6. Oceans regulations, the method described herein will be used in the evaluation of submittals for construction per For events of short duration (a half hour or less), mit applications docketed after May 1, 1978, unless normal water level should be the median tide based this guide is revised as a result of suggestions from the on a 20-year period of record. If the event may have a public or additional staff review.
longer duration, normal water level should be the me dian higher high tide or median lower low tide based If an applicant wishes to use this regulatory guide on a 20-year period of record. Choice of higher high in developing submittals for applications docketed on or lower low tide should be the most conservative for or before May 1, 1978, the pertinent portions of the the event being considered and its consequences. application will be evaluated on the basis of this guide.
1.135-5