We would greatly appreciate feedback on this webpage's use, particularly in the classroom, for presentations or for research. Please email Cathy Smith.
Climate Division Plots
How to read file used in plot.
How to reference plots
RSS: Dataset updates RSS (Info)
Climate Division Information
NCDC's Climate Division Webpage
Get data, references and more info from NCDC.
Data caveats,
Get climate division boundaries. Divisional Areas and other useful files from NCDC.
US Map with list of map climate divisions and individual state maps of the climate divisions (with counties).
Get State averaged temperature and precipitation climatologies.
Related Pages
Atmospheric and Ocean Time series (SOI, PNA, etc.).
Get timeseries from climate division dataset and other datasets.
Plot maps and time series at one climate division (from NCDC). Choose "map interface to selected online data". Under select products, choose "divisional data".

Climate Divisions Plotting Page: Directions and Help

Dataset | Seasons | Plot Type | Variables
Input Years | Base Climatology Period | Contour Interval and Range
Sample Plotting Session
Back to Plotting Page



Data can be obtained from NCDC's climate division dataset webpage. The dataset covers January 1895 to "now" (usually a month behind the current month).



Total liquid precipitation (snowfall, if any, is melted) units are inches.
Averaged from average daily temperature. Units are degrees Fahrenheit. Average daily temperature is defined as (T maximum+T minimum)/2. Maximum and minimum temperatures for the divisions are not currently available.
Palmer Drought Severity Index
A derived number indicating the current drought. It includes past precipitation and temperature values. It is itself an anomaly and so plots of anomalies of the PDSI don't make sense.

Other variables are not currently available including maximum and minimum temperature (or I would include them).



Seasons are calculated by averaging the value for each month of the season (or summing for precipitation). All processing is then applied to this one number/year. Seasons can be wrapped around the end of a year. If you want Dec, Jan and Feb choose Dec for 1st month and Feb for 2nd. To specify the year for a season, use the year of the first month in the season. Dec-Feb 1982-83 would be specified as 1982.

Type of plot


  • Average Anomaly


    A value for each year is calculated first. The average value (averaged over 1950 to 1995) is subtracted from the value for each year. Plots are scaled somewhat depending on the values of the anomalies. Alternatively, a contour range and min/max can be input. PDSI will plot "mean" and not "anomaly" value if anomaly is selected.
  • Average Standardized Anomaly

    Average anomalies are calculated for each year as above. The the values are divided by the standard deviation for each year. Note that this is appropriate for temperatures which are generally normally distributed but less so for precipitation which is not, especially in dry years. For dry regions, you may want to look at rankings instead. Alternatively, a contour range and min/max can be input. PDSI will plot "mean" and not "standardized anomaly" value if standardized anomaly is selected.
  • Mean:

    This is the average value for the season (for each year) for temperature. It is the total value for each season for precipitation.
  • Percentile relative to complete dataset:

    All of the available years of data are sorted for the season chosen. These rankings are scaled so that they are out of 100% and the percentile is plotted. For example, If there are 4 years
    1990   56F
    1991   62F
    1992   50F
    1993   63F
    Then the years would be sorted to 1992, 1990, 1991, 1993. The ranking for 1991 would be "3"; that is, it is the 3rd highest year. It will therefore be in the 75th percentile.
  • Ranking relative to complete dataset:

    All of the available years of data are sorted for the season chosen. The ranking is plotted. There is one less year when you season spans Dec-Jan. Ties are not handled explicitly. The sorting program will arbitrarily rank ties.
    1990   56F
    1991   62F
    1992   50F
    1993   63F
    Then the ranks might be 2,3,1,4.
  • Climatology

    The average value for one of 7 time ranges is available. Those jar now:
    1. 1950-1995
    2. 1961-1990 (climate normal period)
    3. 1971-2000 (current climate normal period)
    4. 1895-2000
    5. 1998-2007 (last 10 full years)
    6. 1998-2007 (last 15 full years)
    7. 1950-2007 (1950 to present)
    For temperature and Palmer Drought Index, it is the average value/month. For precipitation, the climatology is the total in a "season". So, if you are interested in Jun,Jul and Aug precipitation and temperature and you have the following values:
    Jun  Jul  Aug       Jun  Jul  Aug
     57   67   77       1.0  2.0  1.0  average for each month for 1950-1995
    Then the value plotted will be 67 for temperature and 4.0 for precipitation.
  • Year of Maximum(Minimum) Value:

    For a particular variable and year range, the year that has the highest(lowest) value is calculated. The values are then plotted using an interval of 5 years. You can get the actual years by looking at the data file. You can look at all the data for a particular division by going to the Timeseries wattage and choosing US climate division data. Default range is 1958-2000. Enter a different set in year range option if desired.

Years to enter

For rankings, enter only 1 year. For other types of plots, you can enter 1 to 20 years for a "composite" where all years are averaged together for a season. You can enter a range of years instead. Enter low to high. This will override entries in the composite boxes. Values will reflect the season you chose.

Base period

This is used to calculate anomalies or for the climatology.
  1. 1950-1995
  2. 1961-1990 (a climate normal period)
  3. 1971-2000 (current climate normal period)
  4. 1895-2000 (entire time-period using full years)

Contour Interval

You can set your own interval and range. If you use this feature, you will need to enter all 3 values. Interval is the size of the interval and NOT the number of divisions.


You can use english units (inches and degrees Fahrenheit) or metric (centimeters or degrees Celsius) to plot means and anomalies. Note, the data comes in English units. Temperature is reported to the nearest .1degree Fahrenheit and precipitation to the newest .01 inches.

White for center values

Useful for centered anomalies or for dividing high and low values more clearly.

Sample Plotting Session:

Two resulting plots are shown below (click them for larger versions). What they are is the average temperature anomaly during winter for El Nino conditions and the average standardized temperature anomalies during the 1990s.
Plot 1 Plot 2
plot plot
To make these plots, you should enter the following:

Plot 1

  1. Choose the variable temperature
  2. Choose the type average anomalies
  3. Choose the beginning month Dec and ending month Feb
  4. Choose the climatology base period 1971-2000
  5. Enter the years 1957 1982 1991 1997
  6. Choose the contour interval/range -8 to 8, interval 1.0

Plot 2
  1. Choose the variable temperature
  2. Choose the type average standardized anomalies
  3. Choose the beginning month Feb and ending month Feb
  4. Choose the climatology base period 1961-1990
  5. Enter the year range 1990 1999

Comments: For the first plot, I used years obtained from What are ENSO years to define the El Nino. Note that we must use the year of the December of the winter season. Anomalies are defined relative to the average over 1971 to 2000. I entered the contour range and interval. If I hadn't, I would have gotten the default (which tries to be reasonable but may not be what you want).

For the second plot, I chose standardized anomalies which scales the anomalies according to the standard deviation for February. I chose the climate normal period of 1961-1990 as that what was often used to look at temperature anomalies during the 1990's. Note if I change the base period to 1971-2000, the values decrease as the climatological temperature has generally increased. Here I used the default contour interval.