This WWW page presents maps of monthly surface air temperature and
precipitation anomalies regressed onto normalized
anomalies of the AO (definition and time
series). All of the
calculations are performed with monthly anomaly data for the calendar months November through April. The maps
should be interpreted as typical anomalies of temperature or rainfall associated
with a one standard deviation fluctuation of the AO. The sign of the map anomalies is consistent with the "high
index" polarity of the AO. Details of the data sets, indices, and how
to quickly obtain the analyses are described at the bottom of this page.
a) Global figures
land air temperature and precipitation PDF
| image
air temperature over the oceans PDF
| image
b) "North America" figures
land and marine temperature: PDF
| image (GIF) | PNG
land temperature: PDF
| image
precipitation: PDF
| image
air temperature over the oceans: PDF
| image
sea-level pressure: PDF
| image
c) "Atlantic" land and marine temperature:
GIF |
big GIF |
PDF |
PNG |
PS
d) "Europe" figures
land temperature: PDF
| image
precipitation:
PDF
| image
air temperature over the oceans: PDF
| image
e) Netcdf files of the analyses so that you can make your own plots. The fourth
map of the following netCDF files are the regression coefficients of
the named field onto normalized values of the AO.
land and marine temperature (map 1 of
this file.) | the data set
land air temperature
land precipitation
air temperature over the oceans
The sixth map in this netCDF file contains the SLP regressions.
f) "North America" summer (May through October) figures
land and marine temperature: PDF
| image (GIF) | PNG | netCdf file (map 2 is the regressions)
Calculation details
The land data sets are from the Univerisity of Delaware (UD), the
ocean data is COADS
(1-degree latitude-longitude resolution) air or sea-surface
temperature, and the atmospheric circulation fields are from the NCEP
NCAR reanalysis. The native resolution
of the UD data is 0.5-degrees latitude-longitude, and it has been
averaged into 1-degree latitude-longitude resolution so that the
resulting PDF files are of a manageable size. No other smoothing is performed.
The 0.5-degree UD precipitation data set in
netCDF. All errors in the netCDF file are due to Todd Mitchell. (~290 Mbytes)
There are many analyses and everything is in color. PDF files have been onto an ftp directory (
ftp://ftp.atmos.washington.edu/pub/jisao/mitchell/analyses0500/).
[ If using command lines, set the
prompt option to off (type "prompt"), and use "mget *".]
The figure titles and filename conventions employed:
"summer" for May through October analyses. If it doesn't say
"summer" then the analysis is for November through April.
"trends" for trends
"ao" for AO, "cti" for ENSO, "pdo" for PDO
Precipitation is always in cm/month, so precipitation trends
are in units of cm/month/decade (Unfortunately I omitted the "/month"
in the titles of some of the precipation plots.).
The land temperature and precipitation data are for 1950-96, the
COADS data for 1960-96, and the NCEP reanalysis for 1950-96.
