Pacific and Indian Ocean climate signals in a tree-ring record of Java monsoon drought



Extreme climate conditions have dramatic socio-economic impacts on human populations across the tropics. In Indonesia, severe drought and floods have been associated with El Niño-Southern Oscillation (ENSO) events that originate in the tropical Indo-Pacific region. Recently, an Indian Ocean dipole mode (IOD) in sea surface temperature (SST) has been proposed as another potential cause of drought and flood extremes in western Indonesia and elsewhere around the Indian Ocean rim. The nature of such variability and its degree of independence from the ENSO system are topics of recent debate, but understanding is hampered by the scarcity of long instrumental records for the tropics. Here, we describe a tree-ring reconstruction of the Palmer Drought Severity Index (PDSI) for Java, Indonesia, that preserves a history of ENSO and IOD-related extremes over the past 217 years. Extreme Javan droughts correspond well to known ENSO and IOD events in recent decades, and most extreme droughts before this recent period can be explained by known ENSO episodes. Coral proxies from regions near or within the two poles of the IOD show good agreement with Javan PDSI extremes over the past ∼150 years. The El Niño of 1877, in conjunction with a positive IOD, was one of the most intense and widespread episodes of the past two centuries, based on instrumental and proxy data from across the tropical Indo-Pacific and Asian monsoon regions. Although Java droughts typically show the expected association with El Niño-like conditions and failed Indian monsoons, others (mainly linked to positive IOD conditions) co-occur with a strengthened Indian monsoon, suggesting linkages between the Indian monsoon, Indonesian drought and Indian Ocean climatic variability. The close associations between the Java PDSI, ENSO and Indian Ocean climate are consistent with the hypothesis that interannual ENSO to decadal ENSO-like modes interact to generate dipole-like Indian Ocean variability. Copyright © 2008 Royal Meteorological Society