Skip Navigation Links 
NOAA logo - Click to go to the NOAA home page National Weather Service   NWS logo - Click to go to the NWS home page
Climate Prediction Center

CPC Search
About Us
   Our Mission
   Who We Are

Contact Us
   CPC Information
   CPC Web Team

HOME > Stratosphere Home > Winter Bulletins > Northern Hemisphere Winter 2003 - 2004 Summary
Northern Hemisphere Winter Summary


National Oceanic and Atmospheric Administration

April 2004

National Weather Service

National Centers for Environmental Prediction



  • Angell, J.K. ERL/Air Resources Laboratory
  • Flynn, L.T. NESDIS/Climate Research and Applications Division
  • Gelman, M.E. NWS/Climate Prediction Center
  • Hofmann, D. ERL/Climate Monitoring and Diagnostic Lab.
  • Long, C.S. NWS/Climate Prediction Center
  • Miller, A.J. NWS/Climate Prediction Center
  • Nagatani, R.M. NWS/Climate Prediction Center
  • Oltmans, S. ERL/Climate Monitoring and Diagnostic Lab.
  • Zhou, S. RS Information Systems

Concerns of possible global ozone depletion (e.g., WMO/UNEP, 1994) have led to major international programs to monitor and explain the observed ozone variations in the stratosphere. In response to these, and other long-term climate concerns, NOAA has established routine monitoring programs using both ground-based and satellite measurement techniques (OFCM, 1988).

Selected indicators of stratospheric climate are presented in each Summary from information contributed by NOAA personnel. A Summary for the Northern Hemisphere is issued each April, and, for the Southern Hemisphere, each December. These Summaries are available on the World-Wide-Web at the site:

Further information may be obtained from:

Melvyn E. Gelman
NOAA, Climate Prediction Center
5200 Auth Road
Camp Springs, MD 20746-4304
Telephone: (301) 763-8000 ext.7558
Fax: (301) 763-8125


For the Northern Hemisphere winter of 2003-2004, anomalously high total ozone values predominated over the Arctic region. During December, January, and February of 2003-2004, there were portions of the Arctic region where average values of total ozone were greater than 45 percent higher than comparable values during the early 1980s. At the same time, total ozone values over middle latitudes had much lower than average values. Minimum temperatures observed in the lower stratosphere over the Arctic region were above average throughout the winter and only rarely fell below minus 78 C, the temperature below which polar stratospheric clouds form, allowing enhanced chemical destruction of ozone. Temperatures in the upper stratosphere rose dramatically in December, with subsequent significant warming throughout the Arctic stratosphere. These stratospheric warming conditions and a weak polar vortex were associated with the dominance of high ozone in the Arctic. From 1979 to the early 1990s, total ozone had generally decreased over the middle latitudes of the Northern Hemisphere at the rate of 2 to 4 percent per decade, but the downward trend has not continued in recent years. The amounts of chlorine and other ozone destroying chemicals in the stratosphere have been reported to have reached peak values around 1997-98, and have remained at high levels. Much of the recent year-to-year differences in north polar winter-spring stratospheric ozone destruction may be explained as being due to the varying conditions associated with interannual meteorological variability. The high total ozone values in the Arctic region in the winter of 2003-2004 are attributed to absence of very low stratospheric temperatures and meteorological conditions not favorable for ozone destruction, even with the continued presence of ozone destroying chemicals in the stratosphere.


The data available are listed below. This combination of complementary data, from different platforms and sensors, provides a strong capability to monitor global ozone and temperature.

Parameter Ground-Based Satellite/Instrument
Total Ozone Dobson NOAA/SBUV/2
Ozone Profiles Balloon - Ozonesonde NOAA/SBUV/2
Temperature Profiles Balloon - Radiosonde NOAA/TOVS

We have used total column ozone data from the NASA Nimbus-7 SBUV instrument from 1979 through February 1985; NOAA-9 SBUV/2 from March 1985 to December 1988; NOAA-11 SBUV/2 from January 1989 to December 1993; NOAA-9 SBUV/2 from January 1994 to December 1995; NOAA-14 SBUV/2 from January 1996 to June 1998; NOAA-11 SBUV/2 from July 1998 to September 2000; and NOAA-16 SBUV/2 from October 2000. Solar Backscatter Ultra-Violet (SBUV) instruments can produce data only for daylight-viewing conditions, so SBUV/2 data are not available at polar latitudes during winter darkness. Also, the NOAA-11 satellite orbit precessed, and SBUV/2 increasingly viewed high latitudes only in darkness, thus increasing data loss at sub-polar latitudes, especially in 1992 and 1993.


The four maps in Figure 1 show Northern Hemisphere monthly mean total ozone amounts for December 2003 to March 2004. High ozone extends over northern latitudes in the western hemisphere, while relatively low total ozone values prevail over the eastern hemisphere Arctic region. Figure 2a shows the monthly mean total ozone percent difference of the December 2003 monthly mean from the average of eight December monthly means, 1979-1986 (Nagatani et al., 1988). The base period is chosen because 1979 to1986 average values are indicative of the early data record. For December 2003, positive anomalies of greater than 10 percent are shown over most of North America, with maximum positive anomalies of more than 45 percent extending over the Arctic region. Persistence of the high total ozone anomaly in the Arctic region is shown in the maps for January (Figure 2b), and February (Figure 2c). Strong negative anomalies, predominant over much of the middle to high latitude region in December, January and February, are also shown in the Arctic region in March 2004 (Figure 2d).

Figure 3 shows the average area, during February and March for each year since 1979, of low Arctic ozone (lower than 300 DU). For 2004, the area of anomalously low total ozone was near zero. Lack of very low temperatures in the lower stratosphere (see discussion with Figure 7 and Figure 8) is associated with the absence in the Arctic of very low total ozone values in 2003-2004.

Figure 4 shows monthly anomalies of zonal mean total ozone, as a function of latitude and time, from January 1979 to March 2004. The percent anomalies are derived relative to each month's 1979-2004 average. SBUV/2 data (in this figure only) have been adjusted for long-term consistency (Miller et al., 2002). The largest anomalies occur in winter and spring months for the polar region of the Southern Hemisphere. In the high northern latitude region, positive anomalies prevailed in 1979 and the early 1980s, but mostly negative anomalies predominated in the 1990s. However, during the winter of 2003-2004, positive zonal mean total ozone anomalies are shown, with negative anomalies over middle latitudes. The Scientific Assessment of Ozone: 1998 (WMO, 1999) reported that total column ozone decreased at northern mid-latitudes between 1979 and 1991, with estimated linear trend downward of 4 percent per decade. However, since the recovery after 1993 from the 1991 Mt. Pinatubo volcanic eruption, the downward trend of total ozone has not continued.

The NOAA Climate Monitoring and Diagnostics Laboratory (CMDL) operates a 16-station global Dobson spectrophotometer network for total ozone trend studies. Figure 5 shows the total ozone data for four mid-latitude U.S. stations from1979 through 2003. The large annual variation is a result of ozone transport processes, which cause a winter-spring maximum and summer-fall minimum at northern mid-latitudes. Figure 6 shows the four-station average percent deviation from their long-term monthly means. These anomalies, derived from ground-based measurements, are consistent with the anomalies from SBUV/2 satellite ozone measurements, shown in Figure 4 . Middle latitude total ozone values in the years since 1993 have not continued to decline as they had declined from 1979 to 1993. However, ozone values have also not recovered to the higher 1980 values. The implication of these changes needs to be examined in the context of changes in amounts of ozone depleting gases in the atmosphere and varying meteorological conditions.

Figure 7 shows monthly-mean temperature anomalies at 50 hPa for three latitude regions. The temperature anomalies for high latitudes for December 2003 to February 2004 were positive, but for March they were negative. However, temperature anomalies were strongly below average for the middle latitudes and the tropical region.

Extremely low temperatures (lower than -78 C) over the Arctic region in the lower stratosphere are linked to depletion of ozone. Temperatures in the lower stratosphere are closely coupled to ozone through dynamics and photochemistry. Very low temperatures contribute to the presence of polar stratospheric clouds (PSCs). PSCs enhance the production and lifetime of reactive chlorine, leading to ozone depletion in the presence of sunlight (WMO, 1999). Daily minimum temperatures over the polar region, 65N to 90N at 50 hPa (approximately 19 km) are shown in Figure 8. During the winter of 2003-2004, daily minimum temperatures in the lower stratosphere were rarely lower than minus 78 C. They were mostly above average from December through February, and only reached below average in March.

Monthly mean temperature anomalies for December 2003 to March 2004 are shown in the latitude versus pressure cross sections in Figure 9. Positive temperature anomalies are shown in the upper to middle stratosphere in December, with downward progression of the Arctic stratospheric warming shown in January and February. The negative temperature anomalies overlaying the positive anomalies, associated with strong stratospheric warming, also appear to progress downward from the upper stratosphere Arctic in January to lower levels in February and March.

A time series, from December 2003 to March 2004, of normalized height anomalies from 1000 to 30 hPa, for the north-polar region 65-90N, is shown in Figure 10 . The strong positive anomalies in December, January and the first part of February, associated with strong stratospheric warming, appear to propagate from the stratosphere downward to the surface, associated with a negative phase of the Arctic Oscillation. In February and March the negative height anomalies also appear to progress downward (Baldwin and Dunkerton, 1999; Zhou et al., 2002).

Figure 11 compares the average 100-hPa temperatures in the polar region for each March of the last 25 years with the date the stratospheric polar vortex diminished below a specific threshold size. The apparent relationship in previous years of March temperature and vortex size did not hold in 2004. An explanation can be found in the fact that the duration of the polar vortex was very limited in 2004, because of warming conditions earlier in the winter. However the warming subsided and by March north polar temperatures were not especially high.

Figure 12 shows the relationship between the persistence of the polar vortex and the persistence of high latitude total ozone values of less than 300 DU for December through March. This relationship holds also for this year. In the winter of 2003-2004, there was close to minimum in persistence of both the Arctic polar vortex and region of anomalously low ozone.


In the winter of 2003-2004, positive anomalies of total ozone were prevalent in the high latitudes of the Northern Hemisphere. The positive anomalies in total ozone were associated with the meteorological conditions of positive anomalies of lower stratosphere temperature. Arctic temperatures were not sufficiently low for the formation of polar stratospheric clouds and consequent chemical ozone depletion within the polar vortex. Chlorine and other ozone destroying chemicals in the lower stratosphere reached peak values around 1997-98, and have remained at high levels. Lower stratosphere ozone destruction is strong when meteorological conditions of a strong polar vortex and cold polar temperatures prevail. Those cold conditions were not present in the lower stratosphere in the winter of 2003-2004.

Total ozone declined over mid-latitudes of the Northern Hemisphere at the rate of about 2 to 4 percent per decade from 1979 to 1993. In recent years the strong rate of decline of Northern Hemisphere total ozone has not continued, but current stratospheric ozone amounts continue to be below the amounts measured before the early 1980s. A full explanation of ozone and temperature anomalies must include all aspects of ozone photochemistry and meteorological dynamics. Continued monitoring and measurements are essential toward this end.


Baldwin, M.P. and T.J. Dunkerton, 1999: Propagation of the Arctic Oscillation from the stratosphere to the troposphere. .J. Geophys. Res., 107, 30937-30946.

Miller, A.J., R.M. Nagatani, L.E. Flynn, S. Kondragunta, E. Beach, R. Stolarsky, R. McPeters, P.K. Bhartia, M. Deland, C.H. Jackman, D.J.Wuebbles, K.O.Putten, and R.P. Cebula., 2002, A cohesive total ozone data set from SBUV/(2) satellite system, press, J. Geophys. Res., 107(0), doi:10.1029/200,D000853.

OFCM, 1988: National Plan for Stratospheric Monitoring 1988-1997. FCM-P17-1988. Federal Coordinator for Meteorological Services and Supporting Research, U.S. Dept. Commerce, 124pp.

WMO, 1999: Scientific assessment of ozone depletion: 1998. World Meteorological Organization Global Ozone Research and Monitoring Project - Report No. 44.

Zhou, S., A.J. Miller, J. Wang, and J.K. Angell, 2002: Downward-propagating temperature anomalies in the preconditioned polar stratosphere. J. Climate, 15, 781-792.

NOAA/ National Weather Service
National Centers for Environmental Prediction
Climate Prediction Center
5200 Auth Road
Camp Springs, Maryland 20746
Climate Prediction Center Web Team
Page last modified: May 11, 2004
Disclaimer Privacy Notice