<p>Changes in the strength of the Pacific Walker circulation (PWC) can have a significant impact on global mean surface temperatures, as well as regional temperature, precipitation, and extreme weather events far beyond the tropical Pacific. Understanding PWC variability is therefore important for constraining future climate. But observational records of the PWC are short, and single-site proxy records for changes in the strength of the PWC during the last millennium offer contrasting interpretations. This leaves a critical gap in our understanding of PWC variability on the decadal to centennial timescales relevant to future climate change.</p><p>Falster et al. (in prep.) demonstrated that the PWC is strongly imprinted in modern global precipitation &#948;<sup>18</sup>O (&#948;<sup>18</sup>O<sub>P</sub>). This relationship arises via multiple complementary mechanisms, including but not limited to ENSO dynamics. We exploit this relationship to reconstruct changes in the strength of the PWC over the past millennium, using six different statistical and machine learning reconstruction methods in conjunction with a globally-distributed network of palaeo-&#948;<sup>18</sup>O<sub>P</sub> records (Konecky et al. 2020). Although &#948;<sup>18</sup>O<sub>P</sub> from a relatively small number of locations explains a large proportion of PWC variance in the calibration interval, we use a larger network of sites because larger networks are less susceptible to non-stationary teleconnections or non-signal biases than individual sites or smaller networks.&#160;</p><p>Preliminary results indicate that reconstructed PWC variability is coherent across methods, particularly for the past 400 years. Our reconstructions are also robust to both the calibration window used, and the particular palaeo-&#948;<sup>18</sup>O<sub>P</sub> records included in the reconstruction. This provides confidence that our network comprises sufficient proxy timeseries i.e. that we successfully extracted the common underlying climate signal (the PWC) from site-specific information inherent in individual palaeo-&#948;<sup>18</sup>O<sub>P</sub> records. Thus, we are confident that our reconstruction of changes in the strength of the PWC through the last millennium is robust, and it will therefore help to constrain the PWC&#8217;s long-term internal variability and sensitivity to external forcing.</p><p><br><strong>References:</strong></p><p>Falster, G. M., B. Konecky, M. Madhavan, S. Coats, S. Stevenson. 2021. &#8220;Imprint of the Pacific Walker circulation in global precipitation &#948;<sup>18</sup>O&#8221;. In preparation for <em>Journal of Climate</em>.&#160;</p><p>Konecky, B. L., N. P. McKay, O. V. Churakova (Sidorova), L. Comas-Bru, E. P. Dassi&#233;, K. L. DeLong, G. M. Falster, et al. 2020. &#8220;The Iso2k Database: A Global Compilation of Paleo-&#948;<sup>18</sup>O and &#948;<sup>2</sup>H Records to Aid Understanding of Common Era Climate.&#8221; <em>ESSD</em>. https://doi.org/10.5194/essd-2020-5.</p>