Assessing Bias and Uncertainty in the HadAT-Adjusted Radiosonde Climate Record

Abstract

Uncertainties in observed records of atmospheric temperature aloft remain poorly quantified. This has resulted in considerable controversy regarding signals of climate change over recent decades from temperature records of radiosondes and satellites. This work revisits the problems associated with the removal of inhomogeneities from the historical radiosonde temperature records, and provides a method for quantifying uncertainty in an adjusted radiosonde climate record due to the subjective choices made during the data homogenization.This paper presents an automated homogenization method designed to replicate the decisions made by manual judgment in the generation of an earlier radiosonde dataset [i.e., the Hadley Centre radiosonde temperature dataset (HadAT)]. A number of validation experiments have been conducted to test the system performance and impact on linear trends.Using climate model data to simulate biased radiosonde data, the authors show that limitations in the homogenization method are sufficiently large to explain much of the tropical trend discrepancy between HadAT and estimates from satellite platforms and climate models. This situation arises from the combination of systematic (unknown magnitude) and random uncertainties (of order 0.05 K decade Ϫ1 ) in the radiosonde data. Previous assessment of trends and uncertainty in HadAT is likely to have underestimated the systematic bias in tropical mean temperature trends. This objective assessment of radiosonde homogenization supports the conclusions of the synthesis report of the U.S. Climate Change Science Program (CCSP), and associated research, regarding potential bias in tropospheric temperature records from radiosondes.

Toward Elimination of the Warm Bias in Historic Radiosonde Temperature Records—Some New Results from a Comprehensive Intercomparison of Upper-Air Data

“…• The updated HadAT2 dataset (Thorne et al 2005;McCarthy et al 2008). It is the most complete homogeneity-adjusted pure radiosonde dataset published so far and consists of 676 station series, ranging from 1958 to the present.…”

“…These pieces can be rather short and such a restrictive criterion could be applied only because daily data have been used, which allowed break-size estimation for time series as short as 130 days. This is much shorter than has been possible in the HadAT2 adjustment process (Thorne et al 2005;McCarthy et al 2008), which used monthly and seasonal data.…”

The historical pathway towards more accurate homogenisation

“…Another important step in the methodological development is the introduction of perturbed parameter experiments to the test-process of automatic homogenisation methods (McCarthy et al, 2008;Domonkos, 2008;Titchner et al, 2009;Williams et al, 2012). In these examinations some parameters of the homogenisation methods are varied randomly (in ensemble tests) or systematically (in sensitivity tests) and from the synthesis of the results a more complete picture can be obtained about the connections between the datasetproperties and the performance of the applied homogenisation method.…”

Detecting breakpoints in artificially modified- and real-life time series using three state-of-the-art methods

“…Scientists have long been aware that time series often cannot be described adequately by a single linear trend, since discontinuities may arise [4,5]. Recently there has been a boom in papers on both detecting the changepoints and revealing the phenomena behind them [6][7][8][9][10][11][12].…”