Skip to content. | Skip to navigation

Personal tools
You are here: GES DISC Home Oceans Science Focus modis Science Focus:An assessment of the Indian Ocean, Monsoon, and Somali Current using NASA's AIRS, MODIS, and QuikSCAT data

Science Focus:An assessment of the Indian Ocean, Monsoon, and Somali Current using NASA's AIRS, MODIS, and QuikSCAT data

Andrey Savtchenko, NASA Goddard Earth Sciences Data and Information Services Center, Code 610.2, L-3 Communications Government Service Inc.

Introduction | Geography of the Indian Ocean | Winds and Currents | View from space | References | Useful Links


By all means Indian Ocean is quite different from all others on our planet.Especially in terms of impact on millions of lives in one of the most densely populated and poorest regions in the World.

What exactly is the difference, and why should we care?
We have just mentioned the first reason to care - the Indian Ocean has a profound impact on densely populated regions, Figure 1, which have little means for resource management and/or disaster mitigation.
Map of the world's population density
Figure 1.Map of the world's population density.

The second reason would be the unique annual cycle that drives the oceanic and atmospheric circulation that correspondingly impacts the region on the same temporal scales.

Indeed, oceanic currents and atmospheric winds reverse their patterns annually, causing dramatic seasonal changes, ranging from flooding rains to dreadful droughts.

+ back to top

Geography of the Indian Ocean

This would be a good point to start with our understanding of why the Indian Ocean is so different.First, its northern boundary does not extend beyond 25°N.Second, it is not bounded by a "solid" coastal eastern boundary, as the Atlantic and Pacific Oceans are.Next, it is split into two basins - the Arabian Sea and the Bay of Bengal.Thus the Indian Ocean doesn't have the currents to transport and discharge heat to higher latitudes, as the Gulf Stream and Kuroshio do in the Atlantic and Pacific, respectively. What's more, it is relatively well connected to the western part of the Pacific Ocean characterized with its "Warm Pool".The two-basin split of the ocean is a recipe for winds pattern unlike any other over the rest of the oceans where more stable trade winds patterns are observed.

+ back to top

Winds and Currents

It is important to understand and remember that winds and currents are particularly well coupled in the northernIndian Ocean.The latter is no doubt dominated by the Indian monsoon (Knauss, 1996).Any discussion of currents wouldn't be complete without considering winds, and vice versa, in that part of the ocean.

The Indian monsoon has an annual cycle within which the winds completely reverse their direction, and thus we may speak of summer and winter monsoon.To help memorize the relevant wind directions, one can think of the monsoon as similar to the familiar coastal circulation (breeze), but on a much larger scales.

During summer, the Indian subcontinent heats up, creates powerful convection, and large-scale airflow inrushes to replace the ascending air masses. This is known as the southwest monsoon, characterized with winds blowing from southwest, over the Arabian sea.These winds pick up tremendous amounts of moisture that is released over the land as drenching rains.

During winter, however, the continent cools down faster than the water, and thus the convection occurs over the warmer ocean.The replacement comes from northeast, and is known as the northeast monsoon.These are dry continental air masses that can cause prolonged droughts.

Both, the southwest and the northeast, monsoons have a profound effect on the ocean circulation, driving the remarkable Somali Current.This is the western boundary current for the northern Indian Ocean, the analogue to the Gulf Stream in the Atlantic, and Kuroshio in the Pacific.However, the strong air-sea coupling creates a unique distinction - unlike Gulf Stream and Kuroshio, the Somali Current is only present part of the year, June through September, during the time of the strongest southwest monsoon.It is absent from December through February, during the peak of the northeast monsoon.

Typical for western boundary currents is their high flow velocity.The Somali current is no exception - it can manifest velocities in excess of 3.5 m s-1.

The equatorial currents are also revealing an annual cycle that is coupled with the Indian monsoon.There are three major equatorial currents in the Indian Ocean - South Equatorial Current, Equatorial Countercurrent, and the North Equatorial Current.

The South Equatorial Current is always flowing westward, south of about 8°S, but it is stronger during the southwest than the northeast monsoon. This fact may not quite fit into your intuition, because the zonal component of the southwest monsoon has a direction opposite to that of the westward flowing South Equatorial Current.However, recall that the Indian monsoon has its direct impact in the northern Indian Ocean.

The Equatorial Countercurrent in the Indian Ocean is always flowing east.However, the interesting part is that it is south from the equator, unlike in the Atlantic and Pacific.

The North Equatorial Current is north from the equator, and from what was said about the coupling of winds and currents in the northern Indian Ocean one can expect strong dependencies. Indeed, the North Equatorial Current reverses its direction in concert with the Indian Monsoon.From November to March, during the northeast monsoon, it flows westward, and during the rest of the year it flows eastward.

+ back to top

View from space

A question may arise: Is it possible to observe oceanic and atmospheric circulation of the Indian Ocean from space?The answer is, it is not only possible, but the data from the earth observing satellites are an indispensable asset to better understand and forecast the oceanic and atmospheric variability in that uniquely dynamic system.

The NASA's Goddard Earth Sciences Distributed Active Archive Center (GES DAAC), and other centers like the Physical Oceanography DAAC (PODAAC) at the Jet Propulsion Laboratory (JPL), stores numerous relevant data that are not possible to describe here.Rather, we are going to present sample images that address the Indian monsoon and the Somali Current and will comment on data sources that include the Moderate-resolution Imaging Spectroradiometer (MODIS), (Masuoka et al., 1998), Atmospheric Infrared Sounder (AIRS), (Aumann et al., 2003) and SeaWinds.

Clearly, the surface temperature of the land and ocean is a parameter of utmost importance.GES DAAC archives data from two recent missions, MODIS and AIRS, that include this parameter.Another important parameter is the wind speed, that can be acquired from PODAAC.

We will look at the state at which the Indian monsoon and the Somali Current were at the end of April, 2004, by looking at the sea surface temperature data from MODIS, and winds from the SeaWinds instrument on board of QuikSCAT.April is a transition time for both, the Somali current and the Indian monsoon, and thus we shouldn't expect to see well pronouncedpatterns.We will contrast this April and the peak of the southwest monsoon from the summer of 2003.

Figure 2 shows an image of the MODIS/Terra sea surface temperature (SST) from the week beginning on April 22, 2004, where the colors represent the temperature, whereas the arrows indicate the surface winds (at 10 m) acquired by the SeaWinds.
Surface winds vector field from SeaWinds merged with MODIS 11 µm SST
Figure 2.Surface winds vector field from SeaWinds merged with MODIS 11 µm SST.Shown is "Night"-time SST.AIRS sunglint impact on the surface temperatures precludes from meaningful MODIS-AIRS SST comparisons during the "Day"-time overpasses.

The name of the MODIS SST parameter depicted in Figure 2 is "sst_mean".It contains the SST acquired from two long infrared (IR) 11 µm channels, and we have used a subset from the 8-days average 4.9km map data type "MO04MW" that actually covers the entire globe.

The names of the SeaWinds parameters used here are "des_avg_wind_vel_u", "des_avg_wind_vel_v", which are the zonal and meridional components of the surface wind, respectively.These are acquired during the descending orbit of QuikSCAT.The wind components are distributed as global daily 25-km grid product.We have aggregatedthree days of SeaWinds data to fill gaps between orbits and add more statistical significance to the wind components and thus wind direction.In addition, we applied spatial aggregation to arrive at 1-deg grid for the wind field.

Looking at the SST image off coast of Somalia, no apparent stream with pronounced temperature boundaries can be seen, indicating that the Somali current hasn't developed by that time.However, a sign of southerly winds off the coast of Somalia is clearly seen, possibly a precursor to the onset of the southwest Indian monsoon and Somali current.

Southwesterly winds are also clearly seen off coast of Oman.By means of Ekman transport, they create some upwelling and an eddy at (20N, 60E).Substantially colder SST manifest the upwelling.An enormous system of high atmospheric pressure creates anti-cyclonic atmospheric circulation with a center around (15N, 60E), Figure 2.It is likely that the southerlies off the coast of Oman are controlled by this system.However, they separate, intensify, and appear as well established southwesterly winds over the Gulf of Oman - another indication of the onset of the southwest Indian monsoon.These southwesterlies are seen in the SST image to create upwelling off coast of Pakistan as well, Figure 2.

Large pools of warm water exist in the ArabianSea (5N, 70E), and in the Bay of Bengal (10N, 95E).The convergence of the winds, and thus piling up the water under the wind stress, is the most likely cause for the pools.

To have a glimpse of the situation when the Somali current is fully developed, we present Figure 3.It is the same MODIS SST parameter as in Figure 2, which, however, is aggregated for the month of July, 2003.The difference with the transition period from Figure 2 is immediately obvious.Vast areas off the east coast of Africa clearly manifest strong upwelling which is recognizable by the cooler surface temperatures.A very large eddy is separating off the Somali coast at about (7N, 50E).These are all actually the surface signatures of the Somali current.
MODIS 11 µm SST from the month of July, 2003.
Figure 3.MODIS 11 µm SST from the month of July, 2003.This is the peak time for the southwest Indian monsoon.Somali current is well established and shows in the image as vast areas of cold waters off the east coast of Africa.

Another remarkable feature during the week of April 25, 2004, was the tropical cyclone that can be recognized by the cyclonic wind pattern at (-10S, 62E).Because it is south from the Equator, the clock-wise rotation pattern can be clearly seen in the wind field of Figure 2. The blackened areas in the SST image are persistent clouds that can be seen in the true color MODIS image, Figure 4.

The cloud pattern in Figure 4 is consistent with cumulus towers - clouds of very strong convection and intense precipitation.Because of the strong convection these clouds can reach the top of the troposphere, and exhibit extremely low cloud top temperatures, Figure 5.
True color image of the tropical cyclone at (-10N, 62E), April 25, 2004.
Figure 4.True color image of the tropical cyclone at (-10N, 62E), April 25, 2004. The cloud tops here, seen like a bumpy relief at e.g. (-14S, 68E), can serve as visual indication of strong convective and precipitation activity.

The area from Figure 4 shown by means of the at-aperture brightness temperature from MODIS channel #31 (11 µm).
Figure 5.The area from Figure 4 shown by means of the at-aperture brightness temperature from MODIS channel #31 (11 µm).Cloud tops in the tropical cyclone reveal extremely low temperatures, which is an indication of strong convection very likely reaching the top of troposphere.

Going back to the winds in Figure 2, it is interesting to note that the anticyclone off the coast of Oman feeds the tropical cyclone with air flow that passes over warm equatorial waters.Thus, these air masses can pick up a lot of moisture and can provide enormous supply of energy for the cyclone.

Just for the sake of demonstration of AIRS capabilities we are presenting the AIRS standard retrieval of surface temperatures, parameter "TSurfStd" from the data type "AIRX2RET", Figure 5.The AIRS retrieval differs from MODIS in that it uses multiple IR and microwave bands to retrieve atmospheric profiles, and to execute declouding procedures.The surface temperature is actually deduced from the profiles.The spatial resolution of this and others AIRS standard retrievals is 40.5 km at nadir.Among all the bands the AIRS retrieval uses, are the mid-infrared IR bands at about 4 mm.These wavelengths, however, are short enough to be impacted by the sun glint over water surfaces.The latter is typical for instruments like MODIS and AIRS when they pass over the "Day"-side of the oceans of the summer hemisphere.This renders the day-time MODIS 4 µm SST, and AIRS surface temperatures over the oceans, of little use in that particular case.

It is worth noticing that AIRS surface temperatures are cooler than the MODIS SST, Figure 2.This fact is under investigation by MODIS and AIRS science teams, (Aumann and Straw, 2003; Hagan and Minnett, 2003)
AIRS retrieval of the nighttime surface temperatures.
Figure 5.AIRS retrieval of the nighttime surface temperatures.Shown is an aggregate of the three days corresponding to the SeaWinds data from Figure 2.AIRS data exist for the entire globe, continents are masked here only.This three-day aggregate is not a standard product and is produced for the purposes of this article only.

+ back to top


Aumann, H. H., et al., 2003: AIRS/AMSU/HSB on the Aqua Mission: Design, Science Objective, Data Products, and Processing Systems, IEEE Transactions on Geoscience and Remote Sensing, 41, 2, pp.253-264

Aumann, H. H., and Strow, L., 2003: Sea Surface Temperature Measurements with the Atmospheric Infrared Sounder (AIRS), International Geoscience and Remote Sensing Symposium, IGARSS'03, Toulouse, France

Hagan, D. E., and Minnett, P. J., 2003: AIRS Radiance Validation Over Ocean From Sea Surface Temperature Measurements, IEEE Transactions on Geoscience and Remote Sensing, 41, 2, pp. 432-441

Knauss, J. A., 1996: Introduction to Physical Oceanography, 2nd edition, Prentice-Hall, pp. 156 -162.

Masuoka, E., et al., 1998:Key Characteristics of MODIS Data Products, IEEE Transactions on Geoscience and Remote Sensing, 36, 4, pp1313-1323

+ back to top

Useful Links

AIRS information, data access and support:

Physical Oceanography DAAC (QuikSCAT data):

+ back to top


Document Actions
NASA Logo -
NASA Privacy Policy and Important Notices
Last updated: May 27, 2010 04:49 PM ET