RAPID - Round 1 - Improving our ability to predict rapid changes in the El Nino Southern Oscillation climatic phenomenon Project

Introduction

The objective is to use a combination of palaeoclimate reconstruction from annually-banded corals and the fully coupled HadCM3 atmosphere-ocean general circulation model to develop an understanding of the controls on variability in the strength and frequency of ENSO, and to improve our ability to predict the likelihood of future rapid changes in this important element of the climate system. To achieve this, we target three periods: 0-2.5 ka: Representative of near-modern climate forcing; will reveal the internal variability in the system. 6-9 ka: a period of weak or absent ENSO, and different orbital forcing; a test of the model's ability to capture externally-forced change in ENSO. 200-2100 AD: by using the palaeo periods to test and optimise model parameterisation, we will produce a new, improved, prediction of ENSO variability in a warming world.

Project Duration: 1 Jan 2004 - 30 Sep 2007

This project is funded by NERC - Grant Ref. NER/T/S/2002/00443 - through the RAPID Climate Change NERC directed mode programme.

Availability of data and Results

The project will use a combination of palaeoclimate reconstruction from annually-banded corals and the fully coupled HadCM3 GCM to examine ENSO variability. Three periods are targeted: 0-2.5 ka; 6-9 ka; and 2000-2100 AD. The model experiments will result in one or two major integrations ~100 years long about 6000 years ago and 2-3000 years ago, depending on the palaeo data results. HadCM3 (1000 year) will be used with the standard set of diagnostics.

With respect to observational data, existing coral cores from a collaboration with an NSF-funded project (PI Pandolfi) and cores obtained in July 2002 will be dated and preliminary analyses carried out before a final fieldwork season targeting collection of material of specific ages. Optimal cores for further analyses will be selected using high precision U-series dating. Salinity and SST proxies will result from analyses involving δ¹⁸O, Sr/Ca, Mg/Ca and U/Ca. Proxy records external to the project will be compiled from published and new work on other coral sites, and deep sea sediments, tree ring, lake sediment and geoarchaeological data from relevant regions.

The anticipated final model output is several HadCM3 100 year runs and some shorter sensitivity studies. Other runs will be determined by the palaeo-observational data. It is likely that other RAPID PIs will be interested in the results.

Results

There should finally be 2400 atmosphere files (100 years x 12 months x 2 sets of files) for each run, and 1200 ocean files (100 years x 12 months x 1 set of files) for each run.

The run "xbhub" is the pre-industrial run.
The run "xbhuc" is the mid-Holocene run.

The setup for the runs is standard resolution HadCM3 with the only difference between the runs being a change in orbital configuration.

The main focus of the study was to compare ENSO amplitude and frequency in the two simulations. It was found that the amplitude of ENSO was reduced in the mid-Holocene run by around 12% as measured from the standard deviation of NINO3 monthly SST anomalies. There was no statistically significant change in ENSO frequency in the mid-Holocene simulation.

Figure 1: HadCM3 simulations of pre-industrial (0 ka) and mid-Holocene (6 ka) ENSO: NINO3 SST monthly anomaly for 100 year model runs. The amplitude of ENSO as measured from the standard deviation of the NINO3 timeseries is reduced in the mid-Holocene simulation,which is qualitatively consistent with coral records from Papua New Guinea (Tudhope et al., 2001). Understanding the mechanisms and impact of a reduced ENSO amplitude in the mid-Holocene may help us to understand the potential role of the tropics as a driver of rapid climate change in the future.

Access to data and information

Data and software supplied from this project to the BADC will be stored in the BADC RAPID archive.

Access to the archive is currently restricted to participants to this particular project. If you are a programme participant and would like to access this RAPID project data, then please apply for access^* to the RAPID data.

You will be requested to abide by the RAPID Climate Change Data Policy.
Data management, availability and use are ruled by the RAPID Climate Change Programme Data Management Plan (PDF document).
Access to Hadley Centre model data held at the BADC is subject to the signature of the NERC-Met Office agreement.

^* Note that to apply for access to this dataset, you should first register as a BADC user (anyone is allowed to register). If you are already a registered BADC user, skip this step. If you have forgotten your BADC user ID and/or password, please contact the BADC helpdesk at badc@rl.ac.uk. To register as a new BADC user, please follow the instructions online.

References

Brown, J., M. Collins and A. Tudhope (2006), Coupled model simulations of mid-Holocene ENSO and comparisons with coral oxygen isotope records, Advances in Geosciences, 6, 29-33.

Brown, J., M. Collins, A. Tudhope and T. Toniazzo (2006), Modelling mid-Holocene tropical climate and ENSO variability: Towards constraining predictions of future change with palaeo-data, submitted to Climate Dynamics.

Additional references are available from Dr Josephine Brown's website

Citation

et. al

Date of citation

Who to contact

This RAPID project is led by Prof Alexander Tudhope of the University of Edinburgh, with co-investigators at the Scottish Universities Env Research Cen, Bigelow Lab for Ocean Sciences, and the University of Reading

For further information about the results, please contact Dr Josephine Brown, University of Reading or Dr Matthew Collins, Met Office.

General queries about these pages or browsing the data should be directed to the BADC Support line.

Back to RAPID Home page

Home Contact Disclaimer
Last Modified: