search
search

Press Releases

DATE2021.11.25 #Press Releases

The influence of vertical mixing in the ocean surface layer on the Indian Ocean Dipole Mode Phenomenon

Disclaimer: machine translated by DeepL which may contain errors.

Mai Nakazato (Master's course graduate student, Department of Earth and Planetary Science at the time of research)

Shoichiro Kido (Postdoctoral Researcher, Japan Agency for Marine-Earth Science and Technology)

Tomomi Higashizuka (Associate Professor, Department of Earth and Planetary Science / Invited Senior Staff, JAMSTEC)

Key Points of the Presentation

  • The causes of the different intensities of positive and negative Indian Ocean dipole mode phenomena (Note 1) occurring in the tropical Indian Ocean were identified by simulations using the domain ocean model (Note 2).
  • By accurately evaluating the heat budget in the upper ocean, it is revealed for the first time that vertical mixing processes in the ocean surface layer play an important role in the difference in the strength of positive and negative Indian Ocean Dipole Mode Phenomena.
  • The Indian Ocean Dipole Mode Phenomenon is a phenomenon that causes extreme weather events over a wide area of the world, including Japan, and the results of this study are expected to improve the accuracy of extreme weather forecasting for several months ahead.

Summary of the Announcement

Indian Ocean Dipole Mode (IOD) events occurring in the tropical Indian Ocean are known to have a significant impact on the global climate, including Japan as well as the coastal countries of the Indian Ocean. When a positive IOD occurs, SSTs in the tropical Indian Ocean are lower than normal in the eastern part of the region and higher in the western part, while when a negative IOD occurs, SSTs are higher than normal in the eastern part and lower in the western part (Figure 1). Previous studies have indicated that SST fluctuations associated with positive IODs have larger amplitudes than those associated with negative IODs, but the causes of these fluctuations are not fully understood.

Associate Professor Tomoki Higashizuka and Mai Nakazato (master's degree graduate student at the time of their research) of the Graduate School of Science, The University of Tokyo, and Postdoctoral Researcher Shoichiro Kido of the Japan Agency for Marine-Earth Science and Technology have successfully identified the cause of the difference in the strength of positive and negative IODs through realistic simulations using a domain ocean model. By examining the accurate heat budget of the ocean surface layer off the coast of Indonesia in the eastern part of the country, where SST anomalies are particularly pronounced, they have revealed for the first time that vertical mixing processes, in addition to the east-west heat transport noted in previous studies, play an important role in causing the differences in the amplitude of positive and negative IODs.

IODs are known to cause extreme weather events in Japan, but their prediction accuracy is not yet sufficient, and many issues remain. The physical knowledge of the IOD mechanism obtained through this achievement is expected to contribute to improved prediction of extreme weather events several months in advance.

Publication details

Previous studies have pointed out that SST variations associated with positive IODs have larger amplitudes than those associated with negative IODs (such differences in amplitude are hereinafter referred to as "asymmetry"), especially more pronounced in the eastern part of the country (Figure 1). Since this is one of the important characteristics of IODs and their associated modulation of atmospheric circulation, there has been active research on the mechanism of this asymmetry.

Figure 1: The figure shows how many degrees Celsius SSTs shift relative to the normal in September-November, when IODs grow the most, along with (left) positive IODs and (right) negative IODs. The warmer areas represent areas where SSTs are higher than normal, and the colder areas represent areas where SSTs are lower than normal.

In order to investigate the mechanism of SST fluctuations, it is effective to conduct a "mixed layer heat budget analysis," which examines how heat is exchanged in the region called the "mixed layer (Note 3) " near the ocean surface layer. However, it is not easy to accurately estimate the effects of various processes involved in the heat balance in the mixed layer, such as heat exchange with the atmosphere and horizontal heat transport by ocean currents, and the heat balance analysis conducted in previous studies contained large errors. It is also difficult to estimate the effects of vertical mixing processes by direct observation, and the effects of vertical mixing processes have not been fully evaluated in ocean model simulations. Because of these circumstances, it was difficult to precisely identify the mechanism causing the asymmetry in the IOD, and a complete understanding of the mechanism was not achieved.

In this study, we aimed to identify the cause of the IOD asymmetry by performing a mixed-layer heat budget analysis based on a realistic reproduction of the observed tropical Indian Ocean conditions by using a regional ocean model. In particular, we made two innovations to solve the problems of the previous studies mentioned above: First, when simulating with the domain ocean model, we directly calculated all the effects of processes that affect the mixed layer heat budget and saved the data to enable more accurate calculation of the heat budget. The second is the introduction of an elaborate parameterization of vertical turbulent mixing proposed recently by Professor Noriyuki Hibiya and his colleagues at the Graduate School of Science, The University of Tokyo (Note 4) to reproduce as realistically as possible the vertical mixing processes near the ocean surface. As a result, in addition to successfully reproducing IOD more realistically than ever before within the domain ocean model, the mechanism can now be discussed based on the results of an accurate heat balance analysis.

Comparing the results of the heat balance analysis for positive and negative IODs (Figure 2), it was confirmed that the effect of east-west heat transport, the importance of which has been pointed out in many previous studies, is stronger during positive IODs, contributing to the asymmetry. In addition, as indicated in some previous studies, heat exchange with the atmosphere at the sea surface works in a direction that inhibits IOD growth, and the inhibitory effect is stronger during positive IOD.

Figure 2: (Left) Heat budget anomalies in the mixed layer during (left) positive IOD years and (right) negative IOD years. During positive IOD, the SST off the coast of Indonesia is 0.85°C lower than normal, whereas during negative IOD, the SST is 0.39°C higher than normal. Thus, the amplitude of the SST deviation from the normal (referred to as "SST deviation") is 0.46°C stronger at positive IOD. Vertical mixing, whose contribution is newly demonstrated in this study, contributes to a 0.67°C decrease in SST from the normal during positive IOD, whereas it contributes to a 0.12°C increase in SST from the normal during negative IOD. Thus, the effect of vertical mixing is trying to make the amplitude of SST anomalies 0.55°C stronger during positive IOD than during negative IOD. On the other hand, the east-west heat transport and other effects try to strengthen the amplitude of SST anomalies by 0.45°C and 0.07°C during positive IODs compared to negative IODs, respectively, while the heat exchange with the atmosphere at the sea surface tries to weaken the amplitude of SST anomalies by 0.59°C during positive IODs.

On the other hand, vertical mixing, whose contribution had not been discussed in any of the previous studies, was found for the first time to be the most dominant factor in the asymmetry, as it strongly contributes to cooling during positive IODs, while it does not contribute much to temperature increase during negative IODs. Therefore, a more detailed analysis of the causes of such vertical mixing asymmetry revealed that (1) differences in the intensity of vertical mixing and (2) differences in the depth of the cold water beneath the mixing layer play an important role (Figure 3).

(1) During positive IOD, the strengthening of the southeast trade winds overhead amplifies the effect of wind-induced stirring and intensifies vertical mixing in the vertical direction. On the other hand, during negative IOD, the vertical mixing is weakened because the southeast trade winds overhead weaken and the effect of wind stirring is reduced. Therefore, vertical mixing works more efficiently to decrease water temperature at positive IOD, while it does not contribute much to water temperature changes at negative IOD.
(2) In the tropics, the water temperature near the sea surface is high because it is heated by strong solar radiation, but this effect weakens with depth, so there is cold water below the mixing layer with relatively low water temperature. In the eastern part of the tropical Indian Ocean during normal years, this cold water is deeper than in other tropical regions. When the cold water is lifted by the upward flow associated with positive IOD, the mixed layer is efficiently cooled by vertical mixing processes. On the other hand, when the cold water is pushed down by a negative IOD, it has little effect on cooling by vertical mixing processes because of its original deep location.
The combination of these two differences reveals that the SST decrease in the eastern tropical Indian Ocean associated with positive IOD is stronger than the SST increase associated with negative IOD.

Figure 3: Schematic representation of the variation in vertical mixing associated with (left) positive IOD and (right) negative IOD. The stronger the southeast trade winds blowing off the coast of Indonesia, the stronger the vertical mixing. In addition, due to the effect of the earth's rotation, warmer near-surface seawater is transported toward the left side of the wind direction in the southern hemisphere. Therefore, when the southeast trade winds strengthen with positive IOD, more near-surface seawater is transported from the coastal areas of Indonesia to the offshore areas. As a result, the relatively cooler cold water below the mixing layer moves upward and closer to the mixing layer to compensate. On the other hand, when the southeast trade winds weaken with negative IOD, less near-surface seawater is transported offshore from the Indonesian coastal areas, and the upward flow also weakens, moving the cold water away from the mixing layer.

IODs are known to bring extreme weather conditions to Japan and are thought to have recently contributed to Japan's record warm winter in 2019-2020. Therefore, if the occurrence of IODs can be accurately predicted in advance, including their intensity, it will be possible to take measures to mitigate the effects of extreme weather events. However, the accuracy of such predictions is still far less than that of El Niño events in the Pacific Ocean, and many issues remain to be addressed. Furthermore, in recent years, abnormal weather events caused by climate change phenomena such as IODs and El Niño events have become more apparent as a result of global warming, making the improvement of forecast accuracy an urgent issue. The physical knowledge on the mechanism of IOD obtained through this achievement is expected to contribute to the improvement of extreme weather forecasting several months ahead through the improvement of IOD forecasting accuracy.

Published Journals

Journal name
Scientific Reports
Title of paper
Mechanisms of asymmetry in sea surface temperature anomalies associated with the Indian Ocean Dipole revealed by closed heat budget
Author(s)
Mai Nakazato, Shoichiro Kido, Tomoki Tozuka*, Mai Nakazato, Shoichiro Kido, Tomoki Tozuka
DOI Number
10.1038/s41598-021-01619-2
Abstract URL

Terminological Explanation

Note 1: Indian Ocean dipole mode phenomenon

A phenomenon that occurs in the tropical Indian Ocean. When a positive Indian Ocean Dipole Mode event occurs, sea surface temperatures in the tropical Indian Ocean are lower than normal in the eastern part and higher in the western part, while when a negative Indian Ocean Dipole Mode event occurs, sea surface temperatures are higher than normal in the eastern part and lower in the western part. ↑up

Note 2: Regional ocean model

Ocean temperature, salinity, and current strength are affected by heating due to solar radiation and wind, and change according to the laws of dynamics and thermodynamics. A model that expresses these various physical processes in mathematical equations and calculates the state of the ocean under given atmospheric conditions is called an "ocean general circulation model. Although ocean general circulation models generally cover the entire global ocean, it is also possible to take a specific region of interest and solve it while considering influences from the outer boundary, and such a model is called a "regional ocean model. In this study, a regional ocean model was constructed for the Indian Ocean, and the model successfully reproduced the structure of water temperature and ocean currents in a realistic manner given the observed atmospheric conditions. ↑up

Note 3 Mixed layer

In the surface layer of the ocean, there is a layer called the "mixed layer" in which the water temperature and other factors are uniform in the vertical direction due to vertical stirring. This mixing layer plays an important role in the interaction between the atmosphere and the ocean because it exchanges heat with the atmosphere. The thickness of the mixed layer varies greatly depending on the season and ocean area. For example, when the ocean surface cools to form cooler, denser water, vertical convection occurs and the mixed layer becomes thicker. The mixing layer also becomes thicker as winds increase and become more forcefully stirred. ↑up

Note 4 Parameterization

Even with today's improved computer processing power, there are still challenges in ocean-scale simulations with ocean models, for example, fine turbulence cannot be resolved in the model. However, studies have shown that turbulence plays an important role in the ocean, and it is necessary to incorporate some form of its effects in realistic simulations. We call that method parameterization.