Radiosonde Temperature Anomalies in the Troposphere and Lower Stratosphere for the Globe, Hemispheres, and Latitude Zones

Graphics

Digital Data

Alexander M. Sterin

Russian Research Institute for Hydrometeorological Information-World Data Center (RIHMI-WDC), 6, Korolyov str., Obninsk, Kaluga region, Russia, 249035

Period of Record

January, 1958 - September, 2005 (relative to the average for 1961-1975).

Methods

The observed radiosonde data from the Comprehensive Aerological Reference Data Set (CARDS) (Eskridge et al. 1995) were taken as the primary input for obtaining the series. These data were for the global radiosonde observational network through 2001. Since 2002, the AEROSTAB data (uper-air observations obtained through communication channels), collected at RIHMI-WDC in Obninsk, have been used. Both of these data sources were for the global radiosonde observational network. The CARDS data set is known as the most complete collection of radiosonde data.

The data from CARDS and from AEROSTAB passed through the Complex Quality Check (CQC). Only the observed temperature data that were flagged by CQC as correct were used for calculations.

The set of monthly statistics for each year of the period of record was calculated [MONADS Data Set; Sterin and Eskridge (1998)]. This set included, among others, monthly mean values of temperature at standard pressure levels. The processing of monthly mean temperatures at standard pressure levels included the following steps.

• Objective analysis of monthly means from the station points to latitude-longitude grid points. A special algorithm for objective analysis of climatic characteristics was applied. This algorithm was based on adaptive polynomial interpolation and sequential corrections.
• Mass-weighting calculation of temperatures for vertical layers (we considered the layers 850-300hPa (troposphere) and 100-50hPa (lower stratosphere)).
• Averaging over each latitudinal bands to obtain zonal mean temperatures.
• Calculation of anomalies relative to the reference period 1961-1975.
• Spatial averaging (square weighted) of anomalies for the globe, both hemispheres, and latitude zones. The following zones were considered: Southern Extratropics (-90°S to -20°S), Tropics (-20°S to +20°N), Northern Extratropics (+20°N to +90°N), and Northern Polar (+60°N to +90°N).

Series Comparisons

The time series for 1979-2005 are compared to those of the following investigators:

• University of Alabama at Huntsville (UAH) [Christy (1995); Christy et al. (2000)]
• Remote Sensing systems, Inc. (RSS) [Mears et al. (2003)]
• P. Jones (JON); surface temperature anomalies only

Correlations between the various series were computed. We considered both Pearson correlations and Spearman rank correlations. The latter are known to be less sensitive to outliers in some individual months.

Correlations are given in Table 2 (for the globe), Table 3 (for the Northern Hemisphere), and in Table 4 (for the Southern Hemisphere). Table 5 gives Pearson correlations between the tropospheric radiosonde series (RIHMI-WDC) and the surface series beginning in 1958.

Trend Estimates

Various techniques of trend estimation were applied for assessing the long period tendencies. Traditional regression techniques based on ordinary least squares (OLS) were applied, and four robust regression techniques were applied as well. The robust regression techniques were applied to reduce the sensitivity of trend values to possible outliers, especially near the beginning and end of a series. The effect of statistical trend estimation techniques and of series period of record choice is discussed in Sterin (2004a; 2004b).

The estimates of linear trends in these temperature anomaly time series indicate strong cooling in the lower stratosphere. This cooling occurs both in the full-length time series (beginning in 1958) and in shorter time series (beginning in 1979 for comparison with the MSU time series). Trend estimates obtained by various statistical techniques, as well as RMS values and lag 1 autocorrelations (for possible sub-periods of observation), are provided for the globe in Table 6, for the Northern Hemisphere in Table 7, and for the Southern Hemisphere in Table 8. This research was supported in part by the Russian Foundation for Basic Research (RFBR), RFBR Grant 04-05-64681, and is highly appreciated.

References

Christy, J.R. 1995. Temperature above the surface layer. Climate Change 31: 445-474.

Christy J.R., R.W. Spencer, and W.D. Braswell. 2000. MSU Tropospheric Temperatures: Dataset Construction and Radiosonde Comparisons. Journal of Atmospheric and Oceanic Technology 17: 1153-1170.

Eskridge, R.E., O. Alduchov, I.V. Chernykh, Zhai Panmao, A.C. Polansky, and S.R. Doty. 1995. A Comprehensive Aerological Reference Data Set (CARDS): Rough and systematic errors. Bulletin of the American Meteorological Society 76: 1759-1775.

Mears, C.A., M.C. Schabel, and F.J. Wentz. 2003. A reanalysis of the MSU channel 2 tropospheric temperature record. J. Climate 16: 3650-3664.

Sterin A.M., R.E. Eskridge. 1998. Monthly Aerological Data Set: Some features and comparison of upper-air temperature data to the NCAR/NCEP reanalysis monthly data. Proc. of the 22nd Annual Climate Diagnostics and Prediction Workshop, NOAA, pp. 210-213.

Sterin, A.M. 2004a. On the sensitivity of radiosonde data-derived estimates of temperature trends in the troposhere and lower stratosphere: 1. Choice of data set, length of series and estimation technique. Meteorology and Hydrology (Russian version) 5: 21-36.

Sterin A.M. 2004b. On the sensitivity of radiosonde data-derived estimates of temperature trends in the troposhere and lower stratosphere: 2. Detection of inhomogeneities in the series of monthly resolution. Meteorology and Hydrology (Russian version) 6: 5-21.

05/2007