Stratospheric memory and skill of extended-range weather forecasts

Science

Volumn 301, Issue 5633, 2003, Pages 636-640

  Baldwin, Mark P ^a   Stephenson, David B ^b   Thompson, David W J ^c   Dunkerton, Timothy J ^a   Charlton, Andrew J ^b   O'Neill, Alan ^b  

^a NORTHWEST RESEARCH ASSOCIATES   (United States)

^b UNIVERSITY OF READING   (United Kingdom)

^c Department of Mechanical Engineering   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

ATMOSPHERIC PRESSURE; ATMOSPHERIC TURBULENCE; CLIMATOLOGY; MATHEMATICAL MODELS; STATISTICAL TESTS; TROPOSPHERE; UPPER ATMOSPHERE;

ATMOSPHERIC OSCILLATIONS; UPPER TROPOSPHERES;

WEATHER FORECASTING;

ARCTIC OSCILLATION; ATMOSPHERIC CIRCULATION; STRATOSPHERE; WEATHER FORECASTING;

ARCTIC; ARTICLE; ATMOSPHERIC PRESSURE; FORECASTING; GEOGRAPHY; OSCILLATION; PREDICTION; PRIORITY JOURNAL; STATISTICAL MODEL; STRATOSPHERE; SURFACE PROPERTY; TIME; WEATHER; WINTER;

EID: 0042768168     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.1087143     Document Type: Article

References (43)

1. J. G. Charney, J. Shukla, in Proceedings of the Symposium on Monsoon Dynamics, New Delhi, December 1977, J. Lighthill, R. P. Pearce, Eds. (Cambridge University Press, Cambridge, 1981), pp. 99-109.
2. World Meteorological Organization, Manual on Global Data Processing Systems (World Meteorological Organization, Geneva, 1992), vol. 1.
3. M. Kanamitsu et al., Bull. Am. Meteor. Soc. 83, 1019 (2002).
4. M. Hoerling A. Kumar, Science 299, 691 (2003).
5. R. A. Madden, P. R. Julian, J. Atmos. Sci. 28, 702 (1971).
6. M. P. Baldwin, T. J. Dunkerton, J. Geophys. Res. 104, 30937 (1999).
7. M. P. Baldwin, T. J. Dunkerton, Science 294, 581 (2001).
8. D. W. J. Thompson, M. P. Baldwin, J. M. Wallace, J. Climate 15, 1421 (2002).
9. D. W. J. Thompson, J. M. Wallace, Geophys. Res. Lett. 25, 1297 (1998).
10. J. W. Hurrell, Science 269, 676 (1995).
11. J. M. Wallace, Q. J. R. Meteorol. Soc. 126, 791 (2000).
12. "Zonal" means in the east-west direction, with positive values for winds blowing from the west.
13. D. W. J. Thompson, J. M. Wallace, Science 293, 85 (2001).
14. The NAM is defined as the leading empirical orthogonal function (EOF) of slowly varying (for instance, month-to-month), wintertime, hemispheric geopotential at each isobaric level and is the spatial pattern that accounts for the greatest fraction of geopotential variance. Daily indices of the annular modes ate are calculated for each level by projecting daily geopotential anomalies onto the leading EOF patterns. For details of the calculation, see (7).
15. During December through February, the daily correlation between the NAM at 10 hPa and the zonal-mean wind at 10 hPa, 60°N, is 0.96.
16. Wave-induced, angular-momentum transport, driven by upward-propagating waves, leads to downward phase propagation of wind anomalies. The downward phase propagation creates what may be an illusion of downward influence, especially when the data are smoothed in time. Downward phase propagation does not in itself imply that anomalies at lower levels originate at upper levels. The stratosphere is modified by waves originating in the troposphere, altering the conditions for planetary-wave propagation in such a way as to draw mean-flow anomalies downward.
17. We used the National Centers for Environmental Prediction (NCEP) reanalysis data for 1000 to 10 hPa during 1958 to 2002 on a 2.5° longitude by 2.5° latitude grid. The NCEP reanalysis data were obtained from the National Oceanic and Atmospheric Administration-Cooperative Institute for Research in Environmental Sciences (NOAA-CIRES) Climate Diagnostics Center.
18. For the Southern Hemisphere, we used NCEP data from 1979 to 2001. Stratospheric data before 1979 are considered unreliable and we did not use the highly unusual winter-spring events of 2002, which included the only observed major stratospheric warming in the Southern Hemisphere.
19. D. W. J. Thompson, J. M. Wallace, J. Climate 13, 1000 (2000).
20. The stratospheric NAM variance peaks in midwinter (6).
21. W. A. Norton, Geophys. Res. Lett. 30, 1627 (2003).
22. The question of causality, i.e., whether the stratosphere causes changes to the troposphere, is irrelevant for the forecasting problem.
23. A. J. Charlton, A. O'Neill, D. B. Stephenson, W. A. Lahoz, M. P. Baldwin, Q. J. R. Meteorol. Soc., in press.
24. We use the square of the anomaly correlation to measure skill.
25. The AO accounts for 24% of the variance of monthly-mean 1000-hPa geopotential during December through February.
26. Artificial skill is an overestimate of the real skill of a forecasting system caused by the inclusion of the same data to evaluate the skill that was used to develop or train the forecasting system. Artificial skill can be avoided by the use of independent training and assessment data sets. Artificial skill often occurs in practice because of the presence of long-term trends in the data set (42).
27. Because it is persistent, the AO adds predictive skill on shorter time scales.
28. MCA is also known as singular value decomposition analysis (41).
29. T. G. Shepherd, J. Meteorol. Soc. Japan 80, 769 (2002).
30. V. Limpasuvan, D. H. Hartmann, J. Climate 13, 4414 (2000).
31. P. H. Haynes, T. G. Shepherd, Q. J. R. Meteorol. Soc. 115, 1181 (1989).
32. During December through February, the daily correlation between 300-hPa momentum-flux anomalies (latitudinally averaged north of 20°N) and the rate of change of the AO index is 0.46.
33. D. Lorenz, D. Hartmann, J. Climate, 16, 1212 (2003).
34. They found similar lag correlations between the leading EOF of 1000- to 100-hPa zonal wind and eddy momentum-flux convergences.
35. When the NAM strengthened with height, the mean was 0.028 and the standard deviation was 0.90. When the NAM weakened with height, the mean was -0.018 and the standard deviation was 1.06.
36. P. Chen, W. A. Robinson, J. Atmos. Sci. 49, 2533 (1992).
37. P. H. Haynes, C. J. Marks, M. E. McIntyre, T. G. Shepherd, K. P. Shine, J. Atmos. Sci. 48, 651 (1991).
38. J. Perlwitz, H.-F. Graf, Geophys. Res. Lett. 28, 271 (2001).
39. J. Perlwitz, N. Harnik, J. Climate, in press.
40. R. X. Black, J. Climate, 15, 268 (2002).
41. H. von Storch, F. W. Zwiers, Statistical Analysis in Climate Research (Cambridge Univ. Press, Cambridge, 1999), p. 321.
42. I. T. Jolliffe, D. B. Stephenson, Eds. Forecast Verification: A Practitioner's Guide in Atmospheric Science (Wiley, New York, 2003), p. 203.
43. We thank P. H. Haynes, D. A. Ortland, W. A. Robinson, S. Schubert, and T. G. Shepherd for discussions and A. Worsham for assistance with Fig. 1C. Supported by NSF's Climate Dynamics Program, NOAA's Office of Global Programs, NASA's Supporting Research and Technology Program for Geospace Sciences, NASA's Living With a Star Program, and NASA's Oceans, Ice, & Climate Program (M.P.B.); NSF's CAREER program (D.W.J.T.); and NOAA's Office of Global Programs (T.J.D.).