地表过程与资源生态国家重点实验室龚道溢等人，利用印度洋及西太平洋地区10个高时间分辨率珊瑚代用资料，重建了1660年以来的西印度洋越赤道低空急流（即索马里急流）。重建月份为北半球夏季(6-8月)，重建标定对象是ERA40资料，标定时段为1958－1989年。低频分量的重建技巧可以接受，高频分量的重建技巧比低频重建技巧更高。年际尺度上，ENSO信号在重建序列中有合理表征。主要低频重建分量信号包括年代际尺度的~12.2和~21.8年波动，可能与太阳活动影响有关。
研究表明，从18世纪后期到19世纪后期，索马里急流有增强趋势。但是，20世纪上半叶，急流强度存在轻度减弱趋势，越赤道气流的减弱在地面观测的风场资料中得到验证，也与IPCC20世纪气候变化多模式模拟结果 (20C3M)相吻合。这一发现与前人指出的20世纪气候变暖过程中伴随有南亚夏季风的增强趋势是明显不同的。气候变暖背景下季风如何响应的问题，有待更进一步的研究来澄清。
Variability of the low-level cross-equatorial jet of the western Indian Ocean since 1660
Dao-Yi Gong and Jurg Luterbacher
Using 10 month-season resolved proxies from a coral network in Indian Ocean basin, authors reconstructed the summer (JJA) low-level cross-equatorial jet in western Indian Ocean back to 1660AD. The skill of low-frequency reconstruction is acceptable, and the high-frequency reconstruction is of higher skill. On the interannual timescale, El Niño/Southern Oscillation (ENSO) signals are reasonably captured in the reconstructions. The raw reconstruction shows decadal variations of ~12.2 and ~21.8 years, likely suggesting the solar influence. Since the late 18th century to late 19th century, the jet has an increasing trend. However, the cross-equatorial jet slightly decreased in the first half of the 20th century, this decreasing can also found in surface wind observation, and consistent with the ensemble simulations of 20th century climate change (20C3M). This is apparently different from previous proxy studies which indicated an enhancement of Southern Asian summer monsoon in the 20th century in association with the rapid global warming. Elaborate analysis of month/season-resolved wind proxy is needed to clarify this puzzling question.
Fig.1 Upper: Location of coral proxies. ERA40 climate wind vectors of 850hPa level in June-July-August are plotted together. Meridional winds at 850hPa level locating within the box in western Indian Ocean are averaged to define the cross-equatorial jet flow (i.e., V850). Lower: Reconstructed cross-equatorial air flow. Shading indicates the range of 2 times standard error of estimation in calibration period. Smooth line is the low-frequency variations.
Geophysical Research Letters, 35, L01705, doi:10.1029/2007GL032409，2008 http://www.agu.org/pubs/crossref/2008/2007GL032409.shtml
