Abstract: This study uses maximum covariance analysis (MCA) to investigate the linear covariance between anomalous soil moisture at southern Africa and atmospheric circulation in the Southern Hemisphere (SH) based on the ERA40 reanalysis dataset. In the Southern Hemisphere winter (JJA, Jun-Jul-Aug) and summer (JFM, Jan-Feb-Mar), the atmospheric signal resembling the positive phase of the Antarctic Oscillation (AAO) is significantly correlated with persistent soil moisture anomalies in southern Africa that can be up to five months ahead in the leading MCA modes. Regression analysis based on the soil moisture anomaly centers from the MCA covariance patterns confirms that early-season positive soil moisture anomaly in the northern part of southern Africa and negative soil moisture anomaly in the central and southern part of southern Africa later have significant effects on winter and summer atmospheric circulation. Significant squared covariance of JJA (JFM) AAO seems to be related to persistent soil moisture anomalies in the preceding summer and autumn (spring and early summer) and their continuous and cumulative impacts on the atmosphere. The leading time of southern Africa soil moisture anomaly provides an implication for skillful predictability of large-scale atmospheric variability in the Southern Hemisphere winter and summer.