Multidecadal and centennial ENSO variability
Jorge Sánchez-Sesma1 and Arthur Miller2
1Mexican Institute of Water Technology, Jiutepec, Mexico; jsanchez@tlaloc.imta.mx
2Scripps Institution of Oceanography, University of California, San Diego, USA
San Diego, USA, 3-4 September 2010
ENSO is the largest signal of sub-annual climate variability in the Pacific Ocean, affecting not only coastal but also inland locations. Its torrential rains and severe droughts result in economical losses of several hundred millions of dollars in affected countries, from the USA and Mexico to southern South America and as far as Australia. The increase of ENSO frequency and intensities during the second part of the 20th century has affected various sectors, from agriculture to health, from fisheries to the economy. Thus understanding, and ultimately forecasting ENSO variability, has an enormous potential societal benefit.
 |
|
Figure 1: Comparison of three ENSO (N3 annual average) models. A) 21-year moving average smoothed series. The green line (N3s) is the 1000-1999 AD Zebiak-Cane (ZC) model by Mann et al. (2005). The pink line (N3SV) was obtained from an ensemble’s average of ZC simulation forced by volcanic and solar activity. The gray line (N3o) represents smoothed observation data for the last 150 years. B) Zoom of the annual (non-smoothed) N3SV same model, as A), over 2000-2040 AD and validation with independent observational data. Figure adapted from Sánchez-Sesma (2010). |
Although ENSO knowledge has been expanding enormously during last few years, scientists have not yet developed joint and interdisciplinary efforts to better reconstruct and model multidecadal and centennial climate variability (MCEV). The long-term trends that have been reconstructed through numerous proxy variables have neither been analyzed nor modeled well enough to understand the mechanisms required for long-term forecasts. The present trend towards more La Niña events, with more rainfall and tropical cyclones (as 2010 showed us) is highlighting the need to take into account different aspects of ENSO (such as multidecadal trends or ocean-atmosphere interactions) and merge the knowledge from proxy reconstructions, physical and mathematical simulations, non-linear climate analyses and socio-economical research to better understand, predict, and mitigate potential ENSO climate impacts in the future (see e.g., Fig. 1).
In order to discuss all these aspects, an international workshop on MCEV focusing on analysis, reconstruction and simulation of ENSO related climate oscillations was held in San Diego. The workshop drew more than 20 experts from Australia, Canada, Mexico, Spain, UK and the USA. The participants were mainly researchers from the fields of climate simulations and reconstructions or working on climate impacts and their related economical aspects. This workshop was a complementary meeting to the 10th International Conference on Paleoceanography (10th ICP) and was sponsored by the Institute for Mexico and the United States (UC-MEXUS) of the University of California.
Introductory presentations covered a wide range of topics including ENSO modeling, the linear and non-linear trends of climate variability, the differences between local and remote processes, the inherent noise behind the records and their biological feedbacks, GCM projections for the Pacific Decadal Oscillation in the 21st century, ENSO and PDO coincident contributions and trends, and new high-resolution geochemistry proxy records. Other aspects also discussed included: MCEV from simulations and proxies, non-linear decomposition of ENSO simulated records, and orbitally induced mean states of climate change in the Tropical Pacific during interglacials. Emphasis on ENSO climate reconstructions from geological records was provided through an overview of different aspects: (a) on the most adequate archives and the sensitivity of proxy variables linked to ENSO conditions, (b) on the influence of hemispheric climate processes on sea surface temperature (SST) during the Holocene, and (c) on isotopic and geochemical analyses from sedimentary records to determine the transmission of the ENSO signal from the western tropical Pacific and the relationship between ENSO and the California Current variability over the past millennium.
Another session focused on nonlinear mathematical analyses. For example, modeling the millennial synchronization between Greenland and Antarctic δ18O records and the non-linear behavior of ENSO on different timescales were discussed.
The concluding session was centered around the regional impacts of ENSO: analysis of inter-annual variability of precipitation and temperatures over Mexico, the social value of climate predictions in terms of public investments, and the reliance on information provided by institutions and organizations.
The next workshops will focus on centennial scale climate forcings (volcanic and solar). Reconstructions, models and their centennial scale forecasts will be discussed taking into account the associated global and regional climate variability.
references