NCDC / Climate Resources / Data / U.S. Normals / Historic Normals / Search / Help

---

Historic Normals

Documentation from the 1961-1990 Release

CLIM81: Climatogrophy of the U.S., No. 81
CLIM84: Climatogrophy of the U.S., No. 84
CLIM85: Climatogrophy of the U.S., No. 85
CLIM20: Climatogrophy of the U.S., No. 20
WMO Normals: World Meteorological Organization Standard Normals

CLIM81: Climatography of the U.S., No. 81

This publication presents normals of average monthly and annual maximum, minimum, and mean temperature (degrees F), monthly and annual total precipitation (inches), and heating and cooling degree days (base 65 degrees F) for individual locations for the 1961-90 period. There are temperature and degree day data for 4,775 stations and precipitation data for 6,662 stations. The locations represent cooperative weather observer sites, National Weather Service offices, and principal climatological stations in the 50 states, Puerto Rico, Virgin Islands, and Pacific Islands.

The monthly normals are published by state. The data are arranged in four tables representing temperature, precipitation, heating degree days, and cooling degree days. The locations are listed alphabetically within each table. A station locator map and cross reference index providing station name, number, type, location, and elevation are included in the publication for each state.

The monthly normals as well as the 30-year sequential temperature and precipitation data are available on microfiche and magnetic tape. The cross reference index is also available on magnetic tape and is designated as the "monthly 1961-90 normals name tape."

Computational Procedures

Monthly normals were computed for as many stations as practical. In order to be included, the station had to have at least 10 years of monthly temperature data and 10 years of monthly precipitation data from the period 1961-90.

A normal is the arithmetic mean of a climatological element computed over three consecutive decades (Guttman, 1989). The data record should be consistent (have no changes in location, instruments, observation practices, etc.; these are identified here as "exposure changes") and have no missing values so a normal will reflect the actual average climatic conditions. If any significant exposure changes have occurred, the data record is said to be "inhomogeneous" and the normal may not reflect a true climatic average. Such data need to be adjusted to remove the nonclimatic inhomogeneities. The resulting (adjusted) record is then said to be "homogeneous". If no exposure changes have occurred at a station, the normal was calculated simply by averaging the appropriate 30 values from the 1961-90 record.

To address the inhomogeneity problem, the normals methodology involved the following four steps:

1.estimating missing data;

2.adjusting First Order stations with inhomogeneous records;

3.calculating the average monthly values; then

4.converting the temperature averages to the station's official normal, which is valid for the current (as of 1990) observation time.

Neighboring stations were used to estimate missing data. For precipitation, missing values were estimated by averaging the precipitation values from the four nearest neighbors having data for the month in question. The neighboring stations included First Order and Cooperative stations that were within a 30-mile radius of the station being estimated.

For temperature, the nearest 40 neighboring stations were examined and their temperature variations were compared to the temperature variations at the station for which the normals were being calculated (the candidate station). Of these, a pool of 20 neighbors that had the highest correlation with the candidate station were used to estimate the candidate station's missing temperature value. The estimate was calculated using a weighted average of the values from these neighboring stations. The neighboring stations included stations that were part of the Historical Climatology Network (HCN; see Karl, et al., 1990).

The method used to adjust for inhomogeneities is based on the HCN methodology outlined by Karl and Williams (1987). This technique involves comparing the record of the candidate station to the records of neighboring stations. A neighboring station was not used if its record did not cover the same time period as the candidate station. The underlying assumption behind such a methodology is that variations in average weather have similar tendencies over a region. For example, cold winters at a candidate station usually occur simultaneously at its neighboring stations. If this assumption is violated, then there will be a systematic difference between the stations which will show up as temperature differences (or precipitation ratios) that do not follow the expected statistical pattern. Acceptance of this methodology allows the use of certain well-defined statistical techniques to make the adjustments. Inhomogeneities in the candidate station's record were determined by examining the location, instrument, and observation history of the station. After the periods of inhomogeneity were determined, adjustments were applied to remove the biases. The adjustments were determined using the following criteria. Neighboring stations were found which had homogeneous data records that covered the time period of the candidate station's inhomogeneous period. If the candidate station and a neighbor had a reasonably high correlation (r squared > 0.6) of monthly anomalies for the period in question, then the established homogeneous neighboring station was used to assess the impact of the candidate station's discontinuity. The part of the data record before the discontinuity was statistically compared to the part after the discontinuity. The Student's t-test was used for the temperature differences, while the nonparametric Wilcoxon rank-sum test was used for the precipitation ratios. If the statistical test indicated that the two parts of the candidate station's record were significantly different, then the earlier part of the record was adjusted (further details, with examples, can be found in Karl and Williams, 1987). After all exposure changes at the candidate station were corrected, the normal was estimated by averaging the appropriate 30 values from the 1961-90 adjusted record. If none of the neighboring stations had a sufficiently high correlation, then no adjustment was made. The climatological elements (maximum temperature, minimum temperature, and precipitation) were adjusted separately.

Exposure change adjustments were made to First Order stations in the Lower 48 States, but not to the stations in Alaska, Hawaii, or U.S. possessions because of the lack of a sufficient number of neighboring stations. The neighboring stations used in the adjusting procedure included stations from the Cooperation Station Network. No exposure change adjustments were made to the Cooperative Stations due partly to a lack of adequate computerized station history information, but also because a Cooperative Station's identity changes (according to National Weather Service standards) when significant moves occur (generally at least 5 miles horizontally or 100 feet in elevation, subject to the judgement of the National Weather Service Cooperative Program Manager).

Due to the adjustment techniques employed, the normals published in the Climatography of the United States No. 81 publication will not necessarily agree with values calculated by simply averaging the monthly observed values from 1961-90.

Comparison of temperature data between stations works best if all of the stations involved have the same observational schedule. This is generally true for First Order Stations which use the calendar day (midnight) observation time. Unfortunately, some Cooperative Stations have an observation time in the morning, some in the afternoon, some in the evening, and some at midnight, and this introduces a nonclimatic bias into the record. For an explanation of this bias, see Karl et al. (1986). To make the data reflect a consistent observational schedule, the adjustment technique developed by Karl et al. (1986) was used to determine midnight observation time adjustment factors to convert the maximum and minimum temperature data for all appropriate stations to a midnight-to-midnight schedule, thus removing the time of observation bias. No adjustments were made to stations in Alaska, Hawaii, or U.S. possessions because of the lack of a sufficient number of neighboring stations.

It should be emphasized that the official normal temperature values printed in the Series No. 81 publication are for the current (as of December 1990) observation time. The station's observation time and the adjustment necessary to convert the temperature values to a midnight-to-midnight observation time are also shown in the tables. The adjustment factors should be added to the official normals to approximate a "midnight observation time average." This helps a user determine if temperature differences between nearby stations are true climate differences or if they may be caused by different observing schedules. The precipitation data were not adjusted for observation time.

The monthly normals for maximum and minimum temperature were computed as described above. The monthly average temperature normals were computed by averaging the corresponding maximum and minimum normals. The annual temperature normals were calculated by taking the average of the 12 monthly normals. The annual precipitation normals were calculated by adding the 12 monthly normals.

Simple arithmetic procedures were not applied to obtain the heating and cooling degree day normals. Instead, the rational conversion formulae developed by Thom (1954, 1966) were used. These formulae allow the adjusted mean temperature normals and their standard deviations to be converted to degree day normals with uniform consistency. In some cases this procedure will yield a small number of degree days for months when degree days may not otherwise be expected. This results from statistical considerations of the formulae. The annual degree day normals were calculated by adding the corresponding monthly degree day normals.

CLIM84: Climatography of the U.S., No. 84

This publication presents daily 1961-90 normal maximum, minimum, and mean temperature (degrees F), heating and cooling degree days (base 65 degrees F), and precipitation (inches) for 422 National Weather Service offices and principal climatological stations. Monthly, seasonal, and annual normals of these elements are also presented. Monthly and annual precipitation probabilities and quintiles are included in the back of the publication. The data are published in a separate pamphlet for each location.

The daily normals were derived by statistically fitting smooth curves through monthly values; daily data were not used to compute daily normals. As a result, the published values reflect smooth transitions between seasons. The typical daily random patterns usually associated with precipitation are not exhibited; however, the precipitation normals may be used to compute average amounts accumulated over time intervals.

Computational Procedures

Daily normals of maximum, minimum, and mean temperatures, heating and cooling degree days, and precipitation were prepared for 422 stations by interpolating between the monthly normal values. The interpolation scheme was a cubic spline fit through the monthly values. Each element was interpolated independently from the other elements. The procedure is described by Greville (1967).

The series of daily values of an element resulting from the cubic spline yields a smooth curve throughout the year without requiring the use of daily data. Another property of this technique is that the average of the daily temperatures in a month equals the monthly normal and that the total of the daily precipitation or degree days in a month equals the monthly normal. In order to eliminate discontinuities between December 31 and January 1, the spline interpolation was performed on a series of 24 monthly values. This extended series was created by appending July-December normals before January and January-June normals after December.

Since each element was interpolated independently, the daily series of temperatures and degree days were edited to remove spurious inflection points caused by rounding and to ensure adherence to functional relationships among the elements. Specifically:

1.All inflection points were examined for climatological reasonableness.
2.One-half of the sum of a daily maximum and minimum temperature, after rounding, was checked for equivalence with the daily mean temperature.
3.The relationship between a daily mean temperature T and the heating H and cooling C degree days for the day was checked to ensure that
T - 65 + H - C = 0


Daily precipitation normals were published as generated by the cubic spline interpolation. The smooth curve through a month does not represent a climatologically reasonable distribution. The spreading of the monthly precipitation by the spline over all the days in a month is useful for accumulating amounts over specified time intervals. A climatologically reasonable normal precipitation, based on daily data, for any one date would be much different from the published normals.

For some dates at most locations the published degree days are shown by an asterisk. The symbol represents a value of less than one degree day, but more than zero degree days. It is used to smooth through aperiodic oscillations of zeroes and ones that are climatologically unreasonable. For example, if a station has 17, 15, and 18 normal heating degree days in June, July, and August, respectively, it is not possible to distribute the 15 July degree days evenly throughout the month using integer values (zeroes and ones) without creating unrealistic oscillations through the 3-month period. The use of fractional degree days (asterisks) does allow for a smooth transition from June through July to August.

CLIM85: Climatography of the U.S., No. 85

This publication presents normals and standard deviations for the four 30-year periods and the 60-year period between 1931-90 for each division in a state. A division represents a region within a state that is, as nearly as possible, climatically homogeneous. Some areas, however, may experience rather extreme variations within a division (for example, the Rocky Mountain states). The divisions have been established to satisfy researchers in hydrology, agriculture, energy supply, etc., who require data averaged over an area of a state rather than for a point (station).

The divisional data are displayed by name and number for a state or island. The states and islands include the contiguous United States, Alaska, Puerto Rico, and the Virgin Islands, and are arranged alphabetically. Hawaii is not included because the varied topography and locations of the observing stations do not allow for the establishment of homogeneous divisions. The data include monthly and annual values of mean temperature (degrees F), precipitation (inches), and heating and cooling degree days (base 65 degrees F). Standard deviations of these values are also provided.

The divisional normals as well as the 60-year sequential monthly and annual data are also available on microfiche and magnetic tape.

Computational Procedures

Climatic divisions are regions within each state that have been determined to be reasonably climatically homogeneous. The maximum number of divisions in each state is 10. Monthly divisional average temperature and total precipitation data were derived using data from all stations reporting both temperature and precipitation within a climatological division. The number of reporting stations within a division varies from month to month and year to year. This variation was ignored in the computation of the normals.

Monthly temperature normals and 60-year averages for a division were computed by adding the yearly values for a given month and then dividing by the number of years in the period. The annual normal and 60-year average were computed by adding all of the monthly normal or long-term average values and then dividing by 12. Consequently, if an annual normal were computed by averaging annual values obtained for each year in the period (by adding the corresponding 12 monthly values and then dividing by 12), it may be slightly different from the average of the 12 monthly normals because of rounding differences. Precipitation normals and 60-year averages were computed in a similar manner, except that the annual values are the totals of the 12 monthly values.

Sequential monthly degree days were derived using procedures developed by Thom (1954, 1966). This technique utilizes the historical monthly average temperature and its corresponding standard deviation (over some "standardizing period") to compute degree days. The procedure for the computation of the divisional degree day normals involved the following three steps:

1.calculate the standard deviations of the temperatures for each of the 12 calendar months over the standardizing period;
2.use the Thom technique to compute the heating and cooling degree days for every month for every year in the period 1931-90; and
3.calculate the 30-year normals and 60-year (1931-90) averages of the degree days using the procedure discussed in the preceding paragraph.

CLIM20: Climatography of the United States No. 20

The Climatography of the United States No. 20 (CLIM20) publication includes normals data that have been published in the Climatography of the United States No. 81 series, as well as statistics that have not been published elsewhere. The climatological data included in the CLIM20 make this publication the most appropriate summary for agricultural applications.

There have been several editions of the CLIM20 series. Each edition is based on a specific period-of-record of observations. There are no CLIM20 normals for the 1961-90 period. The 1951-80 edition comprises summaries for 1879 locations based on data from the indicated 30-year period. The previous edition was for 1063 sites, with each summary based on at least a 20-year period-of-record beginning in 1951, and from the overall period 1951 through 1975. Earlier issues, which were based on variable and sometimes very long periods-of-record, contain summaries for approximately 1800 locations (Station Table). Those summaries were prepared by Weather Bureau (and later, NOAA) State Climatologists, and contain narrative information along with the climatological data.

1951-1980 Climatography of the United States No. 20

This edition of the CLIM20 Station Climatological Summary provides climate data from 1879 selected observation sites in the National Weather Service (NWS) cooperative observation network. These sites are usually in or near cities which do not have airport weather stations or city weather offices. Because they are not major weather observation locations, their weather data are usually limited to daily maximum and minimum temperature readings, and daily precipitation and snowfall amounts and extremes.

The CLIM20 site/station list varies through the editions mainly because of changing user requirements for the summaries, and because of the opening and closing of observation sites in the cooperative observation network. The 1951-80 CLIM20 stations are locations that have a population of at least 5000 persons.

The 1951-80 CLIM20 summaries contain five tables:

1.a climatological summary which gives monthly and annual normals, means, and extremes of temperature, degree days, precipitation, and snowfall; and mean number of days with temperature and precipitation beyond various thresholds;

2.monthly and annual values of degree days (to base temperatures of 65, 60, 57, 55, and 50 degrees F for heating degree days, and 55, 57, 60, 65, and 70 degrees F for cooling degree days) which were derived from the 1951-80 temperature normals for the station;

3.monthly values of precipitation amounts which correspond to selected levels of probable occurrence;

4.dates of probable first and last occurrence, during the year beginning August 1st and ending July 31st, of freeze-related temperatures, along with probable durations (in days) where the temperature exceeds certain freeze-related values (36, 32, 28, 24, 20, and 16 degrees F); and

5.monthly and annual values of agriculturally related growing degree day units to selected base temperatures (40, 45, 50, 55, and 60 degrees F), with special values for corn.

Selected freeze occurrence statistics are summarized in a separate publication, Freeze/Frost Data (CLIM20 Supplement No. 1), for 3106 stations.

Computational Procedures

The monthly normal values of heating and cooling degree days were computed from the monthly normal temperature and the standard deviation of the temperature using calculation-efficient approximation methods developed by Thom (1952, 1954, 1966). The daily temperature data used in the construction of the Freeze Data and Growing Degree Units tables were extracted from a validated serially-complete data base (Steurer, 1985) of maximum/minimum temperature observations (this data base also can be obtained from NCDC, on magnetic tape). Because of this, there are small differences between the base 55 and 60 growing degree units and cooling degree days which were estimated values. Freeze Data tables for previous CLIM20 editions were compiled from data extracted from the calendar year freeze-occurrence summary of the NCDC Climatological Data - Annual Summary publication. Also, for this edition the dates for the calculation of freeze data probabilities were changed to July 31 and August 1 because they better represent the seasonal transition for many northern and mountainous sections of the United States. The source for the serially-complete data base was the NCDC Summary of the Day (SOD) digital archive file (TD-3200). The SOD temperature data were put through extensive validation and interpolation procedures based upon the departure from the normal in conjunction with those from surrounding stations (Steurer, 1985). As a result, the freeze data and growing degree units were produced from high quality, serially-complete station records of daily maximum and minimum temperatures. This alleviated the many possible problems associated with developing freeze and growing degree unit statistics from an incomplete and poor quality data set. Station values of average daily growing degree units were computed for ten base temperatures (in degrees F): 40, 45, 50, 55, 57, 60, 65, 70, and the truncated bases 48/86 and 50/86. (Of these, 57, 65, 70, and 48/86 have not been published but are available on magnetic tape.) The bases correspond to many of the common phenological cycles in the United States. The truncated bases (48/86 and 50/86) represent adjustments of the daily maximum and minimum temperatures, which better describe specific growth patterns. Here, minimum temperatures below the lower bases are set to the lower bases (48 or 50) and maximum and/or minimum temperatures above the upper base are set to the upper base (86). Average daily station values of growing degree units were computed for each base temperature by an equation similar to that used for cooling degree days (compute the average daily temperature from the maximum and minimum, then sum the differences between the average daily temperature minus the base temperature for each day and each year, then divide by the number of years). The base temperature for 48/86 and 50/86 is 48 and 50, respectively, and the number of years is 30. In this process, when the average daily temperature was less than the base temperature, the value for growing degree units for that day was set to zero, and the average was always rounded up to the nearest degree. The values of daily average growing degree units for each base temperature were then summed to produce the monthly and accumulated monthly totals shown in the CLIM20 tables. Yearly station values of the last spring and first fall occurrences of selected low temperatures (36, 32, 28, 24, 20, and 16 degrees F) were chosen for the period 1951 through 1980. These spring and fall distributions were then used in producing freeze dates and growing season lengths associated with specified probability levels. All freeze dates were based upon the season August 1 through July 31 for each threshold temperature. Last spring dates of occurrence were chosen for the period August 1 of the previous year through July 31 of the selected year (e.g., spring season for 1961 runs from August 1, 1960 through July 31, 1961, except 1951 which begins on January 1, 1951). First fall dates of occurrence were chosen for the period August 1 of the selected year through July 31 of the next year (e.g., fall season for 1961 runs from August 1, 1961 through July 31, 1962, except 1980 which ends on December 31, 1980). This season definition is an improvement over that previously used (the period July 1 through June 30) because the new season definition coincides more closely with the annual march of temperature in which the warmest time of year occurs closer to August 1. The change of season definition produces more realistic dates in the extreme northern and mountainous regions of the United States where temperatures frequently are at the threshold temperatures near the June 30 date. However, it is important to note that the change of season definition has no effect on other stations where temperatures do not reach or exceed the preselected temperature during the summer. The estimation of freeze probabilities was based upon the work of Thom and Shaw (1958) and Thom (1959) which was later modified by Vestal (1971). The selected probabilities were 0.1 through 0.9 in increments of 0.1. A date associated with each of the preselected probability levels was computed for the last spring and first fall freeze seasons. Similarly, the number of days associated with the freeze-free period was computed for each probability level.

WMO Normals: World Meteorological Organization Standard Normals

Every 30 years the international meteorological community comes together to produce a document that summarizes the "normal" climate for all of the nations of the world. The effort was originated by the International Meteorological Committee in 1872 as an effort to assure comparability between data collected at various stations. International agreements eventually determined that the appropriate interval for computing a normal would be 30 years (Guttman, 1989). The World Meteorological Organization (WMO), which succeeded the International Meteorological Committee, defines normals as "period averages computed for a uniform and relatively long period comprising at least three consecutive 10-year periods" (WMO, 1984). The WMO defines climatological standard normals as "averages of climatological data computed for the following consecutive periods of 30 years: January 1, 1901 to December 31, 1930, January 1, 1931 to December 31, 1960, etc." (WMO, 1984). Normals are computed every decade by individual countries to keep up with any climatic changes that may take place, but a coordinated international effort to compile global standard normals is undertaken only once every 30 years (Guttman, 1989). The latest global standard normals period is 1961-1990.

The World Meteorological Organization

The WMO was founded in 1950 as part of the United Nations. As of April 26, 1995 the WMO consisted of 190 Members, which are defined as countries, territories, and groups of territories. The WMO, which is headquartered in Geneva, Switzerland, functions as a coordinator and facilitator in the atmospheric sciences on an international level by:

1.supporting international cooperation in establishing networks for meteorological observations and hydrological and geophysical observations related to meteorology;

2.promoting the creation of centers offering meteorological services and systems to facilitate the exchange of information;

3.encouraging uniform publication of observations and statistics and the application of meteorology as it involves aviation, shipping, water problems, and agriculture; and

4.fostering activities in operational hydrology and cooperation between meteorological and hydrological services (Irvin, 1993).

These goals are accomplished through a number of activities and publications, including a set of Technical Regulations (WMO Publication No. 49) and a Guide to Climatological Practices (WMO Publication No. 100). Members are not forced to follow the regulations and guidelines established by international agreement through the WMO. However, most countries do follow them because, as noted by the WMO in Publication No. 100, "it is in the interest of every country to apply consistent practices in the handling of climate records."

Procedure for Processing the 1961-1990 Global Standard Normals

NCDC received normals data from more than 135 WMO Member States and Territories (click here for map). Each Member was responsible for computing the normals for the stations within its territory and for providing the normals to a central collection site. The WMO Executive Council designated the World Data Center-A for Meteorology (which is collocated with the NOAA's National Climatic Data Center), along with the WMO Secretariat, as the collection site for the 1961-1990 standard climatic normals. The National Climatic Data Center (NCDC) was tasked with tracking the processing, implementation, and exchange of these normals and with assisting the WMO Secretariat with the preparation, publication, and distribution of these normals for the WMO.

The WMO Members provided their normals in a variety of formats, which ranged from printed tables to binary data files on diskette. NCDC's initial task was to convert all of the normals data and metadata (station location and related information) into a digitized ASCII common format. The data next were subjected to limited quality control to check for keying errors and to ensure that the data were internally consistent. The quality controlled data then were grouped into several products.

Normals Products

A magnetic tape has been created and added to NCDC's archives, and is available for processing orders. A CD-ROM is also available through our online store. A subset of the data was published by the WMO in printed format. The printed product contains data from all of the Members that participated, but the publication contains only selected elements due to page count limitations.

These products will include any narrative descriptions provided by the Members, as well as any necessary processing flags or footnotes. The magnetic tape can be ordered from:

National Climatic Data Center, NOAA
151 Patton Avenue, Room 120
Asheville, NC 28801-5001
USA
email:
orders@ncdc.noaa.gov
telephone:
(828) 271-4800
fax:
(828) 271-4876


Requests for the publication, Climatological Normals (CLINO) for the Period 1961-1990, WMO-No. 847, from residents in Canada and the United States should be sent to the American Meteorological Society:

American Meteorological Society
WMO Publications Center
45 Beacon Street
Boston, MA 02108
USA


All other requests for the publication should be sent to the WMO:

The Secretary-General
World Meteorological Organization
Case Postale 2300
CH-1211 Geneva 2, Switzerland
telephone:
(+41 22) 730 81 11
fax:
(+41 22) 734 23 26
telex:
41 41 99 OMM CH
WMO home page:
http://www.wmo.ch


Climatic Elements for the 1961-1990 Global Standard Normals

The WMO Secretariat provided guidance for the preparation of the normals, including station selection, reference period, and inhomogeneity adjustment techniques. Monthly and annual normals will be published by the WMO for the following climatic elements:

atmospheric pressure (reduced to mean sea level)
mean air temperature
maximum air temperature
minimum air temperature
vapor pressure
relative humidity
duration (hours) of sunshine
mean wind speed
amount of precipitation
number of days with precipitation greater than or equal to 1 mm


The following elements will also be published:

absolute (extreme) maximum and minimum temperatures
frequency groups of precipitation (quintiles)


Additional elements were requested by the WMO Secretariat, if available. The response varied from Member to Member. The following is a partial list of the additional elements that will be available in the digital data base:

mean dew point temperature
prevailing wind direction
frequency distribution of wind direction
vector wind direction and magnitude (mean)
wind steadiness
atmospheric pressure (at station level)
amount of snowfall
soil temperature (at various depths)
days with specified phenomenon (e.g., thunder, hail, fog, gale, blowing sand)
total cloud amount
pan evaporation
solar radiation


References

Guttman, N.B., 1989: Statistical descriptors of climate. Bulletin of the American Meteorological Society, vol. 70, no. 6, pp. 602-607.
Irvin, L. (ed.), 1993: Encyclopedia of Associations: International Organizations, Part I (27th Edition). Gale Research Inc., Detroit.
World Meteorological Organization, 1984: Technical Regulations, Vol. I. WMO Publication No. 49. Geneva, Switzerland.

Top of Page

NCDC / Climate Resources / Data / U.S. Normals / Historic Normals / Search / Help