Linkage between Indian Ocean Dipole Asymmetry and Southeastern Indian Ocean Upwelling

Rahaden Bagas Hatmaja¹, Ivonne M. Radjawane^{1, 2}, Agus Santoso^{3, 4}

1. Graduate Program in Earth Sciences, Faculty of Earth Sciences and Technology, Bandung Institute of Technology, Bandung, West Java, Indonesia
2. Oceanography Department, Faculty of Earth Sciences and Technology, Bandung Institute of Technology, Bandung, West Java, Indonesia
3. Australian Research Council (ARC) Centre of Excellence for Climate Extremes and Climate Change Research Centre, The University of New South Wales, Sydney, NSW, Australia
4. Centre for Southern Hemisphere Oceans Research (CSHOR), CSIRO Oceans and Atmosphere, Hobart, Tasmania, Australia

On interannual timescales, Southeastern Indian Ocean (SETIO) upwelling is closely related to the Indian Ocean Dipole (IOD) and El-Niño Southern Oscillation (ENSO). IOD events tend to be stronger when they co-occur with ENSO. However, the linkage between SETIO upwelling and IOD asymmetry that is independent and dependent on ENSO phases is not clear. This issue is investigated in this study based on the analysis of Kelvin wave propagation, subsurface ocean dynamics, and thermocline depth, as well as chlorophyll-a concentration as the upwelling parameters. The impact of IOD asymmetry related with ENSO on SETIO upwelling characteristics was assessed based on composite analysis conducted on two positive IOD events co-occurring with El-Niño events (pIOD–EN) ( 1997 and 2015) and three independent positive IOD (pIOD) events (1994, 2006 and 2012).  During pIOD event, easterly wind anomaly generates zonal sea surface height anomaly (SSHA) across the Indian Ocean, followed by upwelling Kelvin waves propagation along the equator to the coast off Sumatra-Java and also damped the Wyrtki Jet. Consequently, in the Southeastern Indian Ocean, the shallow thermocline depth strengthens the SETIO upwelling in this region as well. During pIOD-EN event, equatorial zonal wind anomaly blows longer and stronger, resulting in earlier and more persistent shallowing of the thermocline depth, starting in May until April in the following year, and also stronger upwelling with peak amplitude of up to 2.78 standard deviation. Even though the upwelling is stronger during pIOD–EN event due to shallower thermocline depth, the chlorophyll-a concentration during independent pIOD events is much higher with peak amplitude of up to 3.19 standard deviation. It is suggested that this is the result of widespread cold SST anomaly distribution north of 15°S due to the absence of El Niño.

Increased variability of Eastern Pacific El Niño under greenhouse warming

Guojian Wang^{1, 2}, Wenju Cai^{1, 2}, Agus Santoso^{3, 1}, Lixin Wu²

1. Centre for Southern Hemisphere Oceans Research (CSHOR), CSIRO Oceans and Atmosphere, Hobart
2. Key Laboratory of Physical Oceanography/Institute for Advanced Ocean Studies, Ocean University of China and Qingdao National Laboratory for Marine Science and Technology, Qingdao, China
3. Australian Research Council (ARC) Centre of Excellence for Climate System Science, The University of New South Wales, Sydney

The El Niño-Southern Oscillation (ENSO) is the dominant and most consequential climate variation on the planet.  ENSO varies from a type of events with a sea surface temperature (SST) anomaly centre in the equatorial central Pacific, to another with an anomaly centre in the equatorial eastern Pacific (Niño3, 5°S-5°N, 150°W-90°W), referred to as CP and EP regimes, respectively. How ENSO may change under greenhouse warming has been plagued by a persistent lack of inter-model agreement in the response of Niño3 SST. Here we find a robust future increase in EP El Niño variability in climate models participating in the fifth phase of Coupled Model Inter-comparison Project (CMIP5) that simulate the two distinct regimes and the associated nonlinear processes. We show that the EP El Niño pattern and its anomaly centre differ vastly from one model to another, therefore cannot be represented by the spatially fixed Niño3 SST index. The robust increase in EP El Niño SST variability occurs at the anomaly centre unique to each model.  Greenhouse warming-induced intensification in stratification of the equatorial Pacific upper ocean enhances ocean-atmosphere coupling, conducive to the growth of the EP SST anomalies that typically govern strong El Niño events.

Internal climate variability and impacts of greenhouse warming on the El Niño-Southern Oscillation

Benjamin Ng^{2, 1}, Wenju Cai^{2, 1}, Timothy Cowan³, Daohua Bi²

1. Centre for Southern Hemisphere Oceans Research, Hobart, TAS, Australia
2. CSIRO, Aspendale, VIC, Australia
3. University of Southern Queensland & Bureau of Meteorology, Melbourne, VIC, Australia

The El Niño-Southern Oscillation (ENSO) is the dominant mode of year-to-year climate fluctuations with wide-ranging socio-economic and environmental impacts. Understanding its response to a warmer climate is paramount, but the role of internal climate variability in modulating the response is not clear. Using 40 Community Earth System Model Large Ensemble (CESM-LE) simulations and 47 Max Planck Institute for Meteorology Grand Ensemble (MPI-GE) members, we find that internal variability generates a spread in ENSO standard deviation and skewness that is similar to the spread of 17 selected Coupled Model Intercomparison Project phase 5 (CMIP5) models. Based on CESM-LE and MPI-GE, internal variability can explain between 90 to 100% of the ENSO standard deviation and all (100%) of the ENSO skewness spread in the 17 CMIP5 models. Both CESM-LE and the selected CMIP5 models project that the standard deviation of eastern and central Pacific ENSO will increase in a warmer climate. However, MPI-GE shows no agreement, highlighting that different models have a different response of ENSO to a warmer climate and that there is large uncertainty within the CMIP5 ensemble which may be caused by internal climate variability.

Assessing Indo-Pacific climate variability and impact of model bias in CMIP5 models

Sebastian McKenna¹, Agus Santoso^{1, 2}, Andréa Taschetto¹, Alex Sen Gupta¹, Wenju Cai^{2, 3}

1. Australian Research Council (ARC) Centre of Excellence for Climate Extremes and Climate Change Research Centre, The University of New South Wales, Sydney, NSW, Australia
2. Centre for Southern Hemisphere Oceans Research (CSHOR), CSIRO Oceans and Atmosphere, Hobart, Tasmania, Australia
3. Key Laboratory of Physical Oceanography/Institute for Advanced Ocean Studies, Ocean University of China and Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

Accurately representing the variability of the tropical Indo-Pacific Ocean in climate models is important in both understanding and trusting future projections that climate models make, with regards to changes in variability and the associated impact. Using the historical simulation from 32 models in the Coupled Model Intercomparison Project phase 5 (CMIP5), we assess the fidelity of the simulated Indian Ocean Dipole (IOD) and El Nino Southern Oscillation (ENSO). We also evaluate the relationship between the IOD and ENSO, and how climatological biases impact on the strength, timing, and frequency of these climate events across the CMIP5 models.

Research has shown that ENSO impacts on both the evolution and strength of the IOD – usually increasing IOD magnitude or initiating IOD events. Realistic simulation of ENSO characteristics (e.g., pattern, timing, frequency, amplitude) is hampered by persistent model biases which vary in extent across CMIP5 models, in particular, in relation to the cold tongue being too cold, and the equatorial easterly wind being too strong. The Indian Ocean also features its own bias, especially in association with the Bjerknes feedback being too strong, which causes many  CMIP5 models to simulate overly strong IOD amplitude. However, there has been relatively little research into the impact of Pacific Ocean and ENSO biases on the IOD simulation across CMIP5 models.

This study aims to investigate the impact of mean state biases in both Indian and Pacific Oceans, and how these biases impact ENSO representation and in turn on the IOD. We have found that the Pacific Ocean cold tongue bias is significantly correlated with IOD strength. Across CMIP5 models we found no significant relationship between IOD event timing and ENSO strength or timing. However, mean state biases in zonal winds and sea surface temperature  in both ocean basins impact on the IOD.

Improving sea level fingerprints associated with future land ice melting

Shujing Zhang¹, Xuebin Zhang², Matt A. King³, Steven J. Phipps¹

1. Institute for Marine and Antarctic Studies, University of Tasmania, Hobart, TAS, Australia
2. Centre for Southern Hemisphere Oceans Research (CSHOR), CSIRO, Oceans and Atmosphere, Hobart, TAS, Australia
3. Surveying and Spatial Sciences, School of Land and Food, University of Tasmania, Hobart, TAS, Australia

Mass changes of land ice (e.g., glaciers and ice sheets) lead to a geographically variable pattern in regional sea level, also called “sea level fingerprints”. Sea level fingerprints associated with contemporary land ice mass changes can be derived by solving the sea level equation, in which changes in Earth’s gravity and rotation, and viscoelastic solid-earth deformation are included. In most existing studies, fingerprints are computed with limited spatial resolution, due to numerical methods and simplified Earth’s crustal structure in 1-D earth model, which might result in some inaccuracies or over-confidence in the fingerprints.

In our research, we use the sea level fingerprint module of the Ice Sheet System Model (ISSM), recently developed by NASA/Jet Propulsion Laboratory (JPL), to improve the sea level fingerprint in response to contemporary land ice mass changes in a rotating, elastic earth model with fixed shoreline geometry. We run the module to provide high-resolution sea level fingerprints driven by historical land ice mass changes from monthly Gravity Recovery and Climate Experiment (GRACE) and altimetry observations. Moreover, we produce future sea level fingerprints based on future projections of land ice mass changes from the Parallel Ice Sheet Model (PISM), and the ensemble of Ice Sheet Model Intercomparison Project (ISMIP6). We also explore the sensitivity of sea level fingerprints to the uncertainties in future land ice melting based on different Representative Concentration Pathway (RCP) scenarios, and to the uncertainties in Earth’s crustal structures parameterized in 1-D earth model.

Recent hemispheric asymmetry in global ocean warming induced by climate change and internal variability

Saurabh Rathore¹, Nathaniel L Bindoff¹, Helen E Phillips¹, Ming Feng^{2, 3}

1. Institute for Marine and Antarctic Studies, University of Tasmania, Hobart, Australia
2. CSIRO Oceans and Atmosphere, Crawley, WA, Australia
3. Centre for Southern Hemisphere Oceans Research, CSIRO, Hobart, WA, Australia

Recent research shows that 90% of the increase in global ocean heat content during 2005-2015 was confined to the southern hemisphere. Using a collection of Argo-based data products and data assimilating model results, we show that this heating pattern of the ocean is most likely driven by anthropogenic climate change and an asymmetric climate variation between the two hemispheres. This asymmetric variation in the coupled ocean-atmosphere system is identified in 11 climate models with 6000 years of simulations of pre-industrial variability. The observed hemispheric asymmetry is more pronounced in the 0-700 m ocean layer than in the 700-2000 m layer. While both layers experience steady anthropogenic warming, 0-700 m layer also has prominent internal-decadal variability that we show is the primary contributor to the observed hemispheric asymmetry.  The rate of warming of our ocean observed from the Argo data is tracking the projected rate from climate model simulations. The changes in ocean heat content below 700 m have the higher signal to noise ratio for tracking anthropogenic warming, however, both layers indicate the robust detection of anthropogenic climate change. There appears to be no need for other hypotheses, such as aerosols, to explain the asymmetry in ocean warming over the 2005-2015 period.

Intermodel relationships between Southern Ocean surface temperature and global climate

Jules B Kajtar^{3, 2, 1}, Agus Santoso^{4, 2}, Matthew Collins³, Matthew H England^{4, 2}, Leela M Frankcombe^{4, 2}, Andréa S Taschetto^{4, 2}

1. University of Tasmania, Battery Point, Tasmania
2. ARC Centre for Climate Extremes, Sydney, Australia
3. University of Exeter, Exeter, UK
4. University of New South Wales, Sydney

CMIP models tend to overestimate the absorbed solar radiation over the Southern Ocean^1,2, with implications for the representation of the Southern Hemisphere climate system and beyond. Specifically, a strong anti-correlation between equilibrium climate sensitivity (the global mean temperature response to doubled atmospheric CO₂) and net top-of-atmosphere radiation in the Southern Hemisphere was found in CMIP3¹. That anti-correlation does not manifest to the same extent in CMIP5, and it has been argued that it is not a robust physical relationship, but a symptom of unrealistic processes in a subset of models². Here we revisit the issue by examining intermodel relationships between surface temperature and various climate variables such as top-of-atmosphere radiation, cloud fraction, sea-ice extent, and atmospheric circulation, with compelling connections among them. Our analysis highlights the dominance of Southern Ocean processes on global mean temperature change, in that models that exhibit cooler sea surface over Southern Ocean tend to project stronger global warming. Our findings suggest that while a physical constraint on equilibrium climate sensitivity may yet be plausible, understanding model biases over the Southern Ocean is of primary importance. Forthcoming CMIP6 analysis could provide further evidence.

1. Trenberth KE, Fasullo JT. Simulation of present-day and twenty-first-century energy budgets of the southern oceans. J Clim 23, 440–454 (2010)
2. Grise KM, Polvani LM, Fasullo JT. Reexamining the relationship between climate sensitivity and the Southern Hemisphere radiation budget in CMIP models. J Clim 28, 9298–9312 (2015)

Separating roles of momentum, heat, and freshwater fluxes in the Southern Ocean warming and water mass changes

Kewei Lyu¹, Xuebin Zhang¹, John A. Church²

1. Centre for Southern Hemisphere Oceans Research (CSHOR), CSIRO Oceans and Atmosphere, Hobart, TAS, Australia
2. Climate Change Research Centre, University of New South Wales, Sydney, NSW, Australia

The Southern Ocean absorbs most of the heat uptake in the climate system due to the anthropogenic warming. The fast ocean warming is found in the mid-latitude Southern Ocean accompanied by poleward expansion of the subtropical gyres. Using model experiments forced by the wind stress, heat flux, and freshwater flux perturbations for doubled CO₂ concentration provided by the Flux-Anomaly-Forced Model Intercomparison Project (FAFMIP), here we separate the roles of different types of air-sea fluxes in driving the Southern Ocean warming and water mass changes. Several key findings include: (1) The surface heat flux changes account for most of the ocean warming by adding heat into the ocean; (2) In terms of the enhancement of warming at middle latitudes (centred at ~45ºS), the wind-driven heat convergence and the accumulation of surface heat uptake by the background circulation have nearly equal contributions; (3) The poleward expansion of the subtropical gyres is primarily attributed to the wind forcing. The wind forcing also drives clockwise shift of the isopycnals in the Southern Ocean which resembles a spin-up of the meridional overturning circulation; (4) The surface heat flux changes dominate the spiciness changes on density surfaces including cooling and freshening within the Subantarctic Mode Water, whereas the surface freshwater flux changes and wind forcing contribute to a lesser extent.