This Title All WIREs
How to cite this WIREs title:
WIREs Clim Change
Impact Factor: 5.124

Tropospheric temperature trends: history of an ongoing controversy

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

Changes in atmospheric temperature have a particular importance in climate research because climate models consistently predict a distinctive vertical profile of trends. With increasing greenhouse gas concentrations, the surface and troposphere are consistently projected to warm, with an enhancement of that warming in the tropical upper troposphere. Hence, attempts to detect this distinct ‘fingerprint’ have been a focus for observational studies. The topic acquired heightened importance following the 1990 publication of an analysis of satellite data which challenged the reality of the projected tropospheric warming. This review documents the evolution over the last four decades of understanding of tropospheric temperature trends and their likely causes. Particular focus is given to the difficulty of producing homogenized datasets, with which to derive trends, from both radiosonde and satellite observing systems, because of the many systematic changes over time. The value of multiple independent analyses is demonstrated. Paralleling developments in observational datasets, increased computer power and improved understanding of climate forcing mechanisms have led to refined estimates of temperature trends from a wide range of climate models and a better understanding of internal variability. It is concluded that there is no reasonable evidence of a fundamental disagreement between tropospheric temperature trends from models and observations when uncertainties in both are treated comprehensively. WIREs Clim Change 2011 2 66–88 DOI: 10.1002/wcc.80

This article is categorized under:

  • Climate, History, Society, Culture > Ideas and Knowledge
  • Paleoclimates and Current Trends > Detection and Attribution
  • Paleoclimates and Current Trends > Modern Climate Change
Figure 1.

Simulated 1979–1999 temperature trends from four modern‐day climate models with representations of human‐induced and natural forcings (see section on ‘Modeling Temperatures in the Atmosphere’). All exhibit a warming troposphere with a maximum in the tropical upper troposphere and a cooling stratosphere, but with differences in trend patterns and magnitudes. Adapted from Climate Change Science Program Synthesis and Assessment Product 1.1.2

[ Normal View | Magnified View ]
Figure 2.

Radiosonde station reporting performance, based on percentage of complete twice‐daily coverage in July 2009 by the European Centre for Medium‐Range Weather Forecasts. The poorer performance away from Northern Hemisphere mid‐latitudes is typical of both contemporary and historical patterns. Figure courtesy of Antonio Garcia‐Mendez and ©ECMWF.

[ Normal View | Magnified View ]
Figure 3.

Top 2 panels: monthly temperature anomalies (smoothed with a 13‐point running average) during 1958–2009 from radiosonde observations at Camborne, Cornwall, UK, at 200 hPa (near‐tropopause) and 700 hPa (lower‐troposphere), including both raw (black) and adjusted (green) HadAT data.7 The smoothed difference series between the two shows the adjustments (offset by 2.25 K). Bottom panel: the four radiosonde types used over this period (typical of UK‐managed stations) are (left to right, with typical periods of operation): Phillips Mark IIb (1950s–1970s); Phillips MK3 (mid 1970s to early 1990s); Vaisala RS‐80 (early 1990s to 2005–2006); and Vaisala RS‐92 (since 2005–2006). Dates of radiosonde changes (red dotted lines) are one sort of ‘metadata event’,5 others include: cross—radiation correction procedure change; star—data cut‐off change; diamond—change of pressure sensor; triangle—change of wind equipment; square—change of relative humidity sensor. Photos courtesy of Kevin Linklater, UK Met Office and background digitally enhanced for clarity by Sara Veasey NOAA NCDC.

[ Normal View | Magnified View ]
Figure 4.

Left: Vertical weighting functions for satellite products. Right: Schematic of atmospheric vertical structure and its latitudinal variation. The line at 30 hPa indicates the typical maximum height of historical global radiosondes data coverage. Because LT and *G involve combining data from other layers, they have negative weightings in parts of the atmospheric column. Adapted from Climate Change Science Program Synthesis and Assessment Product 1.1.2

[ Normal View | Magnified View ]
Figure 5.

Satellite Local Equatorial Crossing Times (LECTs) (pm for ascending/northward and am for descending/southward satellite orbits) for MSU instruments (TIROS‐N to NOAA‐14) and subsequent AMSU instruments (all other satellite platforms). Changes in LECTs typically accompany changes in orbital height and viewing geometry.Courtesy of Carl Mears, Remote Sensing Systems.

[ Normal View | Magnified View ]
Figure 6.

Over several decades, increasing climate model complexity and increasingly realistic simulation of forcings (depicted at left) have led to little change in the expected pattern of atmospheric temperature change (right). Model representations are based on Treut et al.26 and courtesy Fiona Carroll. Zonal‐mean temperature responses from progressive versions of the GFDL (top three, trends in K/century) and Hadley Centre (bottom three, trends in K/decade) models are directly from the literature,27–32 digitally enhanced for clarity only by Deb Misch NOAA/NCDC.

[ Normal View | Magnified View ]
Figure 7.

Tropical temperature behavior in observations and models (including those in Figure 1). Top panel characterizes month‐to‐month surface and LT temperature variability; both observations and models show surface variability is amplified aloft (slope of the line fit to the model points >1—the black line). Lower panel repeats the analysis for temperature trends over the period 1979–1999; observations no longer agree with models, and all but RSS exhibit damping with height. Possible explanations are (1) a real mechanism that modulates long‐term behavior that all models miss, or (2) residual nonclimatic influences in some or all the observations that substantially impact their long‐term trend estimates. The Climate Change Science Program report concluded that the latter was the more likely explanation.25 Adapted from Climate Change Science Program Synthesis and Assessment Product 1.1.2

[ Normal View | Magnified View ]
Figure 8.

Left: tropical tropospheric temperature trend estimates for 1979–2005 (K/decade) from adjusted radiosonde datasets and from models (derived by scaling the range of model amplification in Figure 7 by the HadCRUT3 surface trend180) and from moist‐adiabatic lapse rate theory (MALR).162 Estimates from winds are from Allen and Sherwood.171 Right: same but for LT and MT layers and incorporating MSU time series estimates. The vertical displacement of the points in the right hand panels is for clarity only, and each point represents the average trend over the relevant MSU weighting function. The observations would fall within the model range if both the surface trend is accurate and the model amplification constraint applies in the real world.

[ Normal View | Magnified View ]
Figure 9.

Smoothed global‐mean temperature anomalies for 1958–2009 based on radiosonde and MSU datasets for two layers (top MT; bottom LT). In each panel, the bottom trace is the average of five radiosonde datasets (HadAT, RATPAC, IUK, RAOBCORE, and RICH), and above are differences for individual datasets. Modified from State of the Climate in 2008194 and courtesy of Carl Mears, Remote Sensing Systems.

[ Normal View | Magnified View ]
Figure 10.

Evolution of estimates of observed trends in global‐mean MT and surface temperatures during the satellite era (since 1979), based on satellite (blue), radiosonde (red) and land/SST (green) observations. Symbols show trends for 1979 to the year plotted, as reported in the literature, except for 1979–2008 trends which were calculated for this study (by Carl Mears or current authors). Blue line shows trends from the September 2009 version of UAH for each year. Differences between this line and the UAH published estimates (blue circles) illustrate the degree of change in the different versions of this dataset.

[ Normal View | Magnified View ]

Related Articles

Stratospheric temperature trends: our evolving understanding

Browse by Topic

Paleoclimates and Current Trends > Detection and Attribution
Paleoclimates and Current Trends > Modern Climate Change
Climate, History, Society, Culture > Ideas and Knowledge

Access to this WIREs title is by subscription only.

Recommend to Your
Librarian Now!

The latest WIREs articles in your inbox

Sign Up for Article Alerts