Interpreting differential temperature trends at the surface and in the lower troposphere

Science

Volumn 287, Issue 5456, 2000, Pages 1227-1232

  Santer, B D ^a   Wigley, T M L ^b   Gaffen, D J ^c   Bengtsson, L ^d   Doutriaux, C ^a   Boyle, J S ^a   Esch, M ^d   Hnilo, J J ^a   Jones, P D ^e   Meehl, G A ^b   Roeckner, E ^d   Taylor, K E ^a   Wehner, M F ^a  

^a LAWRENCE LIVERMORE NATIONAL LABORATORY   (United States)

^b NATIONAL CENTER FOR ATMOSPHERIC RESEARCH   (United States)

^c NOAA EARTH SYSTEM RESEARCH LABORATORY   (United States)

^d MAX PLANCK INSTITUTE FOR METEOROLOGY   (Germany)

^e UNIVERSITY OF EAST ANGLIA   (United Kingdom)

Author keywords

[No Author keywords available]

Indexed keywords

AIR-SEA INTERACTION; TEMPERATURE; TEMPORAL VARIATION; TROPOSPHERE;

AIR TEMPERATURE; ARTICLE; ATMOSPHERE; CLIMATE; PRIORITY JOURNAL; SAMPLING; TEMPERATURE MEASUREMENT; TROPOSPHERE;

EID: 0034681552     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.287.5456.1227     Document Type: Article

References (66)

1. P. O. Jones, M. New, D. E. Parker, S. Martín, I. G. Rigor, Rev. Geophys. 37, 173 (1999); P. D. Jones, J. Clim. 7, 1794 (1994); D. E. Parker, P. D. Jones, A. Bevan, C. K. Folland, J. Geophys. Res. 99, 14373 (1994).
2. P. O. Jones, M. New, D. E. Parker, S. Martín, I. G. Rigor, Rev. Geophys. 37, 173 (1999); P. D. Jones, J. Clim. 7, 1794 (1994); D. E. Parker, P. D. Jones, A. Bevan, C. K. Folland, J. Geophys. Res. 99, 14373 (1994).
3. P. O. Jones, M. New, D. E. Parker, S. Martín, I. G. Rigor, Rev. Geophys. 37, 173 (1999); P. D. Jones, J. Clim. 7, 1794 (1994); D. E. Parker, P. D. Jones, A. Bevan, C. K. Folland, J. Geophys. Res. 99, 14373 (1994).
4. D. J. Gaffen, J. Geophys. Res. 99, 3667 (1994); D. E. Parker et al., Geophys. Res. Lett. 24, 1499 (1997); D. E. Parker and D. I. Cox, Int. J. Clim. 15, 473 (1995).
5. D. J. Gaffen, J. Geophys. Res. 99, 3667 (1994); D. E. Parker et al., Geophys. Res. Lett. 24, 1499 (1997); D. E. Parker and D. I. Cox, Int. J. Clim. 15, 473 (1995).
6. D. J. Gaffen, J. Geophys. Res. 99, 3667 (1994); D. E. Parker et al., Geophys. Res. Lett. 24, 1499 (1997); D. E. Parker and D. I. Cox, Int. J. Clim. 15, 473 (1995).
7. P. D. Jones, T. J. Osborn, K. R. Briffa, J. Clim. 10, 2548 (1997).
8. B. D. Santer et al., J. Geophys. Res. 104, 6305 (1999).
9. R. W. Spencer and J. R. Christy, Science 247, 1558 (1990); J. Clim. 5, 847 (1992); J. Clim. 5, 858 (1992).
10. R. W. Spencer and J. R. Christy, Science 247, 1558 (1990); J. Clim. 5, 847 (1992); J. Clim. 5, 858 (1992).
11. R. W. Spencer and J. R. Christy, Science 247, 1558 (1990); J. Clim. 5, 847 (1992); J. Clim. 5, 858 (1992).
12. J. W. Hurrell and K. E. Trenberth, J. Clim. 9, 2222 (1996).
13. _, Nature 386, 164 (1997); P. D. Jones, T. J. Osborn, T. M. L. Wigley, P. M. Kelly, B. D. Santer, J. Geophys. Res. 102, 30135 (1997); J. K. Angell, Geophys. Res. Lett. 26, 2761 (1999). The satellite-estimated lower tropospheric temperature trend shows slight warming when data for 1998 are included.
14. _, Nature 386, 164 (1997); P. D. Jones, T. J. Osborn, T. M. L. Wigley, P. M. Kelly, B. D. Santer, J. Geophys. Res. 102, 30135 (1997); J. K. Angell, Geophys. Res. Lett. 26, 2761 (1999). The satellite-estimated lower tropospheric temperature trend shows slight warming when data for 1998 are included.
15. _, Nature 386, 164 (1997); P. D. Jones, T. J. Osborn, T. M. L. Wigley, P. M. Kelly, B. D. Santer, J. Geophys. Res. 102, 30135 (1997); J. K. Angell, Geophys. Res. Lett. 26, 2761 (1999). The satellite-estimated lower tropospheric temperature trend shows slight warming when data for 1998 are included.
16. Nonrandom temporal changes in the density of buoy SST measuremenls may have biased observed SST data [J. R. Christy, R. W. Spencer, E. S. Lobl, J. Clim. 11, 2016 (1998)].
17. P. D. Jones, P. M. Kelly, C. M. Goodess, T. R. Karl, J. Clim. 2, 285 (1989); P. D. Jones et al., Nature 347, 169 (1990).
18. P. D. Jones, P. M. Kelly, C. M. Goodess, T. R. Karl, J. Clim. 2, 285 (1989); P. D. Jones et al., Nature 347, 169 (1990).
19. F. J. Wentz and M. Schabel, Nature 394, 661 (1998).
20. J. E. Hansen et al., Clim. Change 30, 103 (1995).
21. L. Bengtsson, E. Roeckner, M. Stendel, J. Geophys. Res. 104, 3865 (1999).
22. J. E. Hansen et al., J. Geophys. Res. 102, 25679 (1997).
23. J. R. Christy, R. W. Spencer, W. D. Braswell, J. Atmos. Oceanic Tech., in press. This study finds that artificial cooling induced by the orbital decay effect identified in (70) is partially offset in MSUd 2LT by adjustments relating to instrument body temperature and diurnal drift effects. The MSUd 2LT data used here are monthly means on a 2.5° X 2.5° latitude-longitude grid, spanning January 1979 through December 1998. The maximum of the 2LT weighting function is at roughly 740 hPa.
24. N. Nicholls et al., in Climate Change 1995: The Science of Climate Change, J. T. Houghton et al., Eds. (Cambridge Univ. Press, Cambridge, 1996), pp. 133-192; see also (1). The IPCC surface data were available in the form of monthly mean anomalies (relative to climatological means over 1961-1990) on a 5° X 5° latitude-longitude grid, spanning the period January 1856 through December 1998. For comparison with MSU, we use data for 1979-1998 only.
25. In VARMASK, the MSU and IPCC areal coverage is identical and varies from a maximum of 77% in 1987 to a minimum of 70% in 1992. FIXMASK has smaller coverage (∼60%) than the minimum coverage in VARMASK. Although NOMASK involves no masking of observed data, it does entail masking of model surface data with observed surface coverage.
26. J. K. Gibson et al., ECMWF Re-Analysis Project Report Series. 1. ERA Description (European Centre for Medium Range Weather Forecasts, Reading, UK 1997).
27. E. Kalnay et al., Bull. Am. Meteorol. Soc. 77, 437 (1996).
28. J. W. Hurrell and K. E. Trenberth, Bull. Am. Meteorol. Soc. 80, 2661 (1999).
29. Because the strength of the surface-troposphere temperature coupling shows considerable spatial variability (6), subsampling tropospheric temperature data with surface coverage would be less useful if one considered limited geographical areas. Our analysis, however, deals with temperatures over at least 60 to 70% of Earth's surface.
30. The method for assessing statistical significance of trends and trend differences is described by B. D. Santer et al. (J. Geophys. Res., in press). It involves the standard parametric test of the null hypothesis of zero trend, modified to account for lag-1 autocorrelation of the regression residuals [see J. M. Mitchell Jr. et al., Climatic Change, World Meteorological Organization Tech. Note 79 (World Meteorological Organization, Geneva, 1966)]. The adjustments for autocorrelation effects are made both in computation of the standard error and in indexing of the critical t value.
31. The method for assessing statistical significance of trends and trend differences is described by B. D. Santer et al. (J. Geophys. Res., in press). It involves the standard parametric test of the null hypothesis of zero trend, modified to account for lag-1 autocorrelation of the regression residuals [see J. M. Mitchell Jr. et al., Climatic Change, World Meteorological Organization Tech. Note 79 (World Meteorological Organization, Geneva, 1966)]. The adjustments for autocorrelation effects are made both in computation of the standard error and in indexing of the critical t value.
32. This apparent contradiction has a simple explanation. The significance of trend differences is determined from d(t), the time series of paired differences between the surface and 2LT data. Use of d(t) markedly reduces noise levels by subtracting variability components common to the surface and 2LT data, thus facilitating the identification of relatively small trend differences embedded in noisy but covarying time series.
33. The acronym denotes version 4 of the atmospheric model jointly developed by the European Centre for Medium-Range Weather Forecasts and the Max Planck Institute for Meteorology in Hamburg, coupled to the Hamburg isopycnal ocean model.
34. E. Roeckner, L. Bengtsson, J. Feichter, J. Lelieveld, H. Rodhe, J. Clim. 12, 3004 (1999).
35. W. M. Washington et al., in preparation.
36. B. A. Boville and P. R. Cent, J. Clim. 11, 1115 (1998).
37. Version 3 of the NCAR Community Climate Model (CCM3) [B. A. Boville and J. W. Hurrell, J. Clim. 11, 1327 (1998)].
38. We calculated equivalent 2LT temperatures from the model monthly mean data using both a radiative transfer code and a static global mean weighting function (4, 5). The two methods yield highly similar global mean anomaly time series. Here we show results for the static weighting function approach only. Model SSTs were merged with air temperatures measured at 2 m over land to form surface temperature data sets directly comparable with the IPCC observations.
39. Sampling distributions were computed in the following way. For model surface and 2LT data, annual means were calculated and expressed as anomalies relative to the 300-year climatological annual mean. These anomalies were transformed to the IPCC grid. We then processed maximally overlapping 20-year data segments (i.e., model years 1 through 20, 2 through 21, etc.). For each 20-year segment, annual-mean anomalies were recomputed relative to the average over years 1 through 15 of each 20-year segment. This mimics the definition of anomalies in the observations (Fig. 1). For each control run, we used the NOMASK, VARMASK, and FIXMASK methods to sample the 2LT and surface data in each of the 281 20-year segments of model data. Finally, we computed area-weighted surface and 2LT spatial means and fitted least-squares linear trends to these and to the surface - 2LT difference time series. This yields nine sampling distributions of "unforced" 20-year trends and trend differences (3 control runs X 3 masking methods) for each of three "variables" (surface, 2LT, and surface - 2LT). From these distributions, we estimated the empirical probability of obtaining a model result that exceeds the observed trend or trend difference given in Table 1.
40. Although VARMASK has greater common coverage (between the surface and 2LT) than FIXMASK, VARMASK results are more difficult to interpret because part of the observed surface - 2LT trend difference in VARMASK is due to nonrandom changes in coverage over the period 1979-1998. This is not the case with FIXMASK.
41. Our terminology follows that used in (72).
42. J. A. Leggett, W. J. Pepper, R. J. Swart, in Climate Change 1992: The Supplementary Report to the IPCC Scientific Assessment, J. T. Houghton, B. A. Callander, S. K. Varney, Eds. (Cambridge Univ. Press, Cambridge, 1992), pp. 69-95.
43. This was done by linearizing observed monthly mean latitude-height profiles of stratospheric ozone change from November 1978 to April 1993, and then extrapolating the computed trends to December 1997 (72).
44. Bengtsson et al. (72) described further GSOP experiments that represent linear combinations of responses to various forcings. We considered the one GSOP experiment that was not a linear combination of individual responses. Although the GSOP linear combinations cover the period of the satellite record, their construction inflates the high-frequency variability, which hampers estimation of relatively small trends.
45. 1/2 has a Gaussian distribution, and that values of d > 1.96 (2.58) indicate trend differences significant at the 5% (1%) level.
46. See Web table 1 at Science Online (www.sciencemag. org/feature/data/1046025.shl).
47. 2 a Pollutant and Does EPA Have the Power to Regulate It?", Washington, DC, 6 October 1999.
48. 2 a Pollutant and Does EPA Have the Power to Regulate It?", Washington, DC, 6 October 1999.
49. J. R. Christy and R. T. McNider, Nature 367, 325 (1994); P. D. Jones, Geophys. Res. Lett. 21, 1149 (1994); S. F. B. Tett, J. F. B. Mitchell, D. E. Parker, M. R. Allen, Science 274, 1170 (1996).
50. J. R. Christy and R. T. McNider, Nature 367, 325 (1994); P. D. Jones, Geophys. Res. Lett. 21, 1149 (1994); S. F. B. Tett, J. F. B. Mitchell, D. E. Parker, M. R. Allen, Science 274, 1170 (1996).
51. J. R. Christy and R. T. McNider, Nature 367, 325 (1994); P. D. Jones, Geophys. Res. Lett. 21, 1149 (1994); S. F. B. Tett, J. F. B. Mitchell, D. E. Parker, M. R. Allen, Science 274, 1170 (1996).
52. See Web table 2 at Science Online (www.sciencemag. org/feature/data/1046025.shl).
53. Bengtsson et al. (12), in their analysis of various realizations (34) of the GSOP experiment, show slight warming (by roughly +0.01° to +0.03°C per decade) of the model surface temperature relative to temperatures at 850 and 500 hPa. These results are not inconsistent with our finding of a slight cooling (by -0.035°C per decade) of the surface relative to lower troposphere in the single GSOP experiment we consider (Fig. 5C). These small differences arise from our use of a vertically weighted lower tropospheric temperature (the 2LT retrieval) rather than discrete temperatures at 850 and 500 hPa, and from our masking procedure, which excludes areas of large surface warming at high latitudes in both hemispheres [see (12), plate 3].
54. K. P. Shine, Y. Fouquart, V. Ramaswamy, S. Solomon, J. Srinivasan, in Climate Change 1995: The Science of Climate Change, J. T. Houghton et al., Eds. (Cambridge Univ. Press, Cambridge, 1996), pp. 108-131; J. E. Hansen, M. Sato, R. Ruedy, J. Geophys. Res. 102, 6831 (1997); T. K. Berntsen et al., J. Geophys. Res. 102, 28101 (1997).
55. K. P. Shine, Y. Fouquart, V. Ramaswamy, S. Solomon, J. Srinivasan, in Climate Change 1995: The Science of Climate Change, J. T. Houghton et al., Eds. (Cambridge Univ. Press, Cambridge, 1996), pp. 108-131; J. E. Hansen, M. Sato, R. Ruedy, J. Geophys. Res. 102, 6831 (1997); T. K. Berntsen et al., J. Geophys. Res. 102, 28101 (1997).
56. K. P. Shine, Y. Fouquart, V. Ramaswamy, S. Solomon, J. Srinivasan, in Climate Change 1995: The Science of Climate Change, J. T. Houghton et al., Eds. (Cambridge Univ. Press, Cambridge, 1996), pp. 108-131; J. E. Hansen, M. Sato, R. Ruedy, J. Geophys. Res. 102, 6831 (1997); T. K. Berntsen et al., J. Geophys. Res. 102, 28101 (1997).
57. Examples include the effects of El Chichón and other volcanic eruptions [N. G. Andronova, E. V. Rozanov, F. Yang, M. E. Schlesinger, G. L. Stenchikov, J. Geophys. Res. 104, 16807 (1999)], as well as aerosols resulting from biomass burning [J. E. Penner, R. Dickinson, C. O'Neill, Science 256, 1432 (1992)] and mineral dust [I. Tegen and I. Fung, J. Geophys. Res. 100, 18707 (1995)].
58. Examples include the effects of El Chichón and other volcanic eruptions [N. G. Andronova, E. V. Rozanov, F. Yang, M. E. Schlesinger, G. L. Stenchikov, J. Geophys. Res. 104, 16807 (1999)], as well as aerosols resulting from biomass burning [J. E. Penner, R. Dickinson, C. O'Neill, Science 256, 1432 (1992)] and mineral dust [I. Tegen and I. Fung, J. Geophys. Res. 100, 18707 (1995)].
59. Examples include the effects of El Chichón and other volcanic eruptions [N. G. Andronova, E. V. Rozanov, F. Yang, M. E. Schlesinger, G. L. Stenchikov, J. Geophys. Res. 104, 16807 (1999)], as well as aerosols resulting from biomass burning [J. E. Penner, R. Dickinson, C. O'Neill, Science 256, 1432 (1992)] and mineral dust [I. Tegen and I. Fung, J. Geophys. Res. 100, 18707 (1995)].
60. National Research Council, Board on Atmospheric Sciences and Climate, Commission on Geosciences, Environment and Resources, Reconciling Observations of Global Temperature Change (National Academy Press, Washington, DC, 2000).
61. D. J. Gaffen et al., Science 287, 1242 (2000).
62. B. D. Santer et al., Nature 384, 523 (1996); T. M. L. Wigley, P. J. Jaumann, B. D. Santer, K. E. Taylor, Clim. Dyn. 14, 781 (1998); S. F. B. Tett et al., Nature 399, 569 (1999); K. Hasselmann, J. Clim. 6, 1957 (1993).
63. B. D. Santer et al., Nature 384, 523 (1996); T. M. L. Wigley, P. J. Jaumann, B. D. Santer, K. E. Taylor, Clim. Dyn. 14, 781 (1998); S. F. B. Tett et al., Nature 399, 569 (1999); K. Hasselmann, J. Clim. 6, 1957 (1993).
64. B. D. Santer et al., Nature 384, 523 (1996); T. M. L. Wigley, P. J. Jaumann, B. D. Santer, K. E. Taylor, Clim. Dyn. 14, 781 (1998); S. F. B. Tett et al., Nature 399, 569 (1999); K. Hasselmann, J. Clim. 6, 1957 (1993).
65. B. D. Santer et al., Nature 384, 523 (1996); T. M. L. Wigley, P. J. Jaumann, B. D. Santer, K. E. Taylor, Clim. Dyn. 14, 781 (1998); S. F. B. Tett et al., Nature 399, 569 (1999); K. Hasselmann, J. Clim. 6, 1957 (1993).
66. Supported by NOAA Office of Global Programs ("Climate Change Data and Detection") grant NA87CP0105 (T.M.L.W.) and U.S. Department of Energy (DOE) grant DE-FG02-98ER62601 (P.D.J. and T.M.L.W.). Work at Lawrence Livermore National Laboratory was performed under the auspices of the Environmental Sciences Division of DOE (contract W-7405-ENG-48). We thank J. R. Christy for the MSU data and static MSU weighting functions, and C. F. Keller, G. C. Hegerl, J. M. Wallace, M. C. MacCracken, A. Robock, and two anonymous reviewers for useful comments and suggestions that substantially improved the manuscript.