This page has been archived and is no longer updated

 

Tropical Weather

By: Adam H. Sobel (Columbia University, New York, NY) © 2012 Nature Education 
Citation: Sobel, A. H. (2012) Tropical Weather. Nature Education Knowledge 3(12):2
Email
What makes tropical weather different from that at higher latitudes? Which is more predictable?
Aa Aa Aa

 

Sobel banner

There are two reasons why tropical weather is different from that at higher latitudes. The sun shines more directly on the tropics than on higher latitudes (at least in the average over a year), which makes the tropics warm (Stevens 2011). And, the vertical direction (up, as one stands on the Earth's surface) is perpendicular to the Earth's axis of rotation at the equator, while the axis of rotation and the vertical are the same at the pole; this causes the Earth's rotation to influence the atmospheric circulation more strongly at high latitudes than low. Because of these two factors, clouds and rain storms in the tropics can occur more spontaneously compared to those at higher latitudes, where they are more tightly controlled by larger-scale forces in the atmosphere. Because of these differences, clouds and rain are more difficult to forecast in the tropics than at higher latitudes. On the other hand, temperature is easily forecast in the tropics, because it doesn't change much.

Heat, Moisture, Clouds and Rain

The higher the temperature, the more water vapor can be in the air without condensing. As the sun shines strongly on the tropics — particularly on the warm oceans which have an effectively infinite amount of water to evaporate into the air — the overlying atmosphere becomes very humid.

Temperature and pressure both drop quickly with altitude, in the tropics as elsewhere on Earth. If air rises — as it can, for example, if it is a little warmer and thus lighter than the air around it — it will expand and cool, eventually causing some of the water vapor in it to condense into tiny liquid droplets, forming a cloud. The latent heat of the condensation warms the air, causing it to become still warmer, and allowing the updraft to rise further. If enough water condenses, the cloud droplets can become large enough to fall as rain.

Sometimes a tropical shower ends quickly, as the clouds and falling rain evaporate. The evaporating rain cools the air near the surface, so that it isn't warm enough to rise into a new cloud. Sometimes, though, the cooling effect as well as the weight of the rain itself can create a downdraft strong enough to create turbulence that in turn lifts nearby warm, humid air, making a new updraft (Figure 1). This process can feed on itself to produce a large complex of such storms that maintains rainy weather over a period of days and a region thousands of km in extent, sometimes moving coherently across the tropics and generating new storms as it moves (Mapes & Houze 1993). The circumstances that lead to one outcome vs. another — very rainy weather over a large region, completely clear skies, or anything in between — sometimes differ only in subtle ways.

Schematic of deep convective cloud showing an updraft, precipitation-driven downdraft, and turbulence below cloud base.
Figure 1: Schematic of deep convective cloud showing an updraft, precipitation-driven downdraft, and turbulence below cloud base.
© 2012 Nature Education All rights reserved. View Terms of Use

Rotation and the Coriolis Force

The orientation of the tropics relative to the Earth's rotation axis also complicates tropical weather prediction. At higher latitudes, where the rotation axis doesn't make too great an angle to the vertical, the Coriolis force plays a big role in determining which way the wind blows. The Coriolis force makes the wind blow approximately parallel to the isobars, or lines of constant pressure. In the northern hemisphere, the wind goes clockwise around highs and counterclockwise around lows; opposite in the southern. If there were no Coriolis force, the wind would simply blow from high to low pressure, across the isobars. If that were to happen, the highs and lows wouldn't last long. The wind would equalize the pressure field, flowing out of highs and into lows and making the pressure horizontally uniform. By preventing this, the Coriolis force allows the highs and lows to exist for a long time. These highs and lows — the weather systems of the extratropics — evolve in a somewhat predictable way. While being pushed along by the jet stream, they often arrange themselves into alternating high-low patterns in the east-west direction, sometimes referred to as "waves". These wave patterns go through cycles of growth and decay that we understand, and that computer models are able to simulate accurately. Strong temperature contrasts — fronts — form where the highs and lows push warm air up against cold air, and clouds and rain tend to form where they push air upwards.

In the tropics, the Coriolis force is weak, the highs and lows flatten out, and global pressure maps show almost no structure — apart from the occasional tropical cyclone. With the flat pressure field comes a flat temperature field (see Figure 2, which shows global maps of temperature, pressure, and a measure of cloudiness, for a single day). There are no fronts in the tropics. Though temperature still drops quickly with height, at a given level the temperature is quite similar at any point within the tropics. So tropical temperature is quite predictable. Since it doesn't change much (see Figure 3, which compares a year's worth of surface temperature data from two Australian cities, one in the tropics and one in midlatitudes), it's easy to forecast, even far in advance. For a given upcoming date, just look up what the temperature was on that date in a few past years; it will probably have been not too different on that same date in different years, and will be within the same narrow range this year.

Three weather maps from May 15, 2005.
Figure 2: Three weather maps from May 15, 2005.
Top: mean sea level pressure, contour interval 5 hPa. Middle: surface temperature, contour 5ºC. Bottom: equivalent black body temperature, K; color scale shown below the plot, values greater than 270 are indicative of clear skies and are not shown, while lower values are indicative of clouds and rain. Note the much greater temperature and pressure variations (indicated by many more contours) in the higher latitudes compared to the tropics. Clouds and rain (as indicated by black body temperature) are nonetheless quite different from place to place within the tropics, although pressure and temperature are not.
© 2012 Nature Education All rights reserved. View Terms of Use

Daily maximum surface temperatures.
Figure 3: Daily maximum surface temperatures.
Recorded at the airports in Darwin (12ºS, 131ºE) and Hobart (43ºS, 147ºE), Australia, during the year 2005. Hobart — the higher-latitude city — besides being colder overall, has much greater variations in temperature, both in the seasonal cycle and the day-to-day fluctuations, than does Darwin, the tropical city.
© 2012 Nature Education All rights reserved. View Terms of Use

Winds and rain, on the other hand, are difficult to predict in the tropics. Without strong highs, lows and fronts pushing the air around and determining where it rises, the rain appears to form more from the spontaneous bubbling up of buoyant convective clouds. These convective clouds are what we know in many areas as thunderstorms — though over ocean in particular, they need not necessarily produce thunder and lightning. When these clouds become big and organized enough, they can generate their own large-scale weather systems. At any given moment, much of the tropics seems to have the potential for such systems to develop, but most of the time they do not, for reasons that are neither clear to scientists nor well-predicted by computer models. The humidity field may be part of the answer — regions of higher humidity may be more prone to disturbed weather than drier regions, and humidity varies more than temperature within the tropics, particularly in the upper atmosphere. However, this is not a completely satisfying or useful answer either, because humidity in turn is strongly influenced by the weather, and can change rapidly.

Types of Tropical Weather Systems

Once an organized tropical weather system does develop, it may move in a predictable way for a while. Several different types of tropical weather systems exist, each having its own typical characteristics, including its size, speed and direction of motion.

Some disturbances, for example, are known as "easterly waves" because they follow along in the direction the overall trade winds are blowing, from east to west — the opposite direction of the westerly winds at higher latitudes. As with the wave patterns of high and low pressure systems in the extratropics, the term wave here indicates a wavy pattern in a map of the wind field over a large area. Easterly waves have long been well known in the Caribbean, for example (Riehl 1954), where they arrive after forming over west Africa and moving across the tropical Atlantic ocean. Similar weather systems also occur over the tropical Pacific (Reed & Recker 1971) and Indian oceans. Some of them eventually strengthen into tropical cyclones — known as hurricanes in the Atlantic and eastern Pacific, typhoons in the western Pacific, and simply cyclones in the Indian ocean and southern hemisphere — which are the most organized and destructive of tropical weather systems.

Some other types of tropical weather systems were first predicted in the 1960s by theoretical meteorologists studying the partial differential equations that govern the motion of a thin fluid layer on a rotating sphere (Matsuno 1966). These "equatorial waves" were then discovered to exist first in the stratosphere (the layer above that in which weather occurs, the troposphere) (Yanai & Maruyama 1966, Wallace & Kousky 1968) and then later in moving cloud patterns seen from satellites, indicating their presence in the troposphere (Gruber 1974, Zangvil 1975, Takayabu 1994, Wheeler & Weickmann 2002). These tropospheric waves are now known as convectively coupled waves (Wheeler & Kiladis 1999), and come in several types including Kelvin waves (which move from west to east) and mixed Rossby-gravity waves (which move from east to west, like easterly waves). These various waves are somewhat predictable, lasting for periods of days or even a week or more. Computer models still don't capture these tropical disturbances as well as they do midlatitude weather systems, but are improving. The regularity of the waves' motion by itself allows for some predictability using statistical methods (Wheeler & Weickmann 2002; Figure 4). These statistical forecasts essentially just assume that a given wave will continue to move in the same direction and speed as it has thus far.

Longitude-time (Hovmoeller) diagram of outgoing longwave radiation (OLR, Watts per meter squared) averaged 15ºS-15ºN, for a period in the first half of 2011.
Figure 4: Longitude-time (Hovmoeller) diagram of outgoing longwave radiation (OLR, Watts per meter squared) averaged 15ºS-15ºN, for a period in the first half of 2011.
Shading shows the total OLR anomaly, or difference from the average ”climatological” conditions; blue indicates more disturbed, rainy weather while orange and red indicate drier conditions. Features tilted down to the right are moving eastward, features tilting down to the left are moving westward. The different colored contours indicate convectively coupled wave disturbances of different types: blue contours are the Madden-Julian oscillation, black are equatorial Rossby waves, green are Kelvin waves. The period after June 1 shows a forecast produced on that date by assuming propagation of these waves at their normal speeds.
© 2012 Nature Education Photo courtesy of Matthew Wheeler, Centre for Australian Weather and Climate Research. All rights reserved. View Terms of Use

The tropical climate also has slower variations that are coherent and, to some extent, predictable. The Madden-Julian oscillation (MJO) (Madden & Julian 1971, 1972, 1994) is manifest in periods of rainy weather and westerly winds, alternating with drier weather and easterly winds. The patterns move from west to east, taking a month or two to complete a cycle. The cycles are not entirely regular, and within a rainy or dry period the weather may differ from what is prevalent during that period (there may be some rain during a dry period or vice versa) so that knowing the state of the MJO by itself does not allow accurate daily weather prediction. It does, however, give some predictability to the broader state of the climate system over periods of 2–3 weeks or sometimes even longer. The MJO appears to be in a class by itself, as the largest and slowest of the tropical weather disturbances. Though it occurs in the tropics, the MJO also affects the weather in higher latitudes (Jones 2000, Bond & Vecchi 2003, Cassou 2008). The MJO was not predicted by the classic theory of equatorial waves — it was discovered first in observations, and still has not been adequately explained.

On still longer time scales, the El Niño/Southern Oscillation (ENSO) phenomenon (Philander 1990, Sarachik & Cane 2010, Stevens 2011) causes slow variations to develop in the climate of the eastern and central Pacific, with effects throughout the tropics and higher latitudes as well. This makes some aspects of the climate predictable on horizons of months to a year, though again it does not allow prediction of the specific weather on any given day that far in advance.

Outstanding Questions

Tropical meteorology and climatology are active research fields. The fundamental physical mechanisms behind important phenomena, such as the Madden-Julian oscillation, easterly waves, and the formation of tropical cyclones are subjects of active research. Understanding all these phenomena will require understanding why and how deep convective clouds "self-aggregate", or merge together to form larger weather systems (Mapes 1993, Liu & Moncrieff 2004, Bretherton et al. 2005). Along with observations and theory, computer models will play an important role in untangling these mechanisms, but the models are still limited by their ability to represent the physics of convective clouds. Numerical weather prediction, which uses the same kinds of models as are used in basic scientific research, appears also to be similarly limited. We don't know how far in advance tropical weather can be predicted, even with a perfect model. The answer almost certainly depends on how precise one wishes the answer to be. For day-to-day weather at a single location, it seems likely — but has not been proven — that the intrinsic range of predictability (Lorenz 1969) is shorter in the tropics than higher latitudes (Boer 1995, F. Vitart personal communication). On the other hand, if one is willing to accept forecasts of weather conditions averaged over a week, or a month, and over a larger area rather than a single location, the answer is likely to change for the better. The existence of slower tropical variations like the MJO and ENSO appears to make the tropics more predictable than the extratropics, when such averages are considered.

Acknowledgments

I thank Dan Kirk-Davidoff, Frederic Vitart, and Matthew Wheeler for discussions, Victoria Rompilla for producing Figure 1, Matthew Wheeler for providing Figure 4, and George Kiladis and an anonymous reviewer for helpful comments.

References and Recommended Reading


Boer, G. J. Analyzed and forecast large-scale tropical divergent flow. Monthly Weather Review 123, 3539-3553 (1995).

Bond, N. A. & Vecchi, G. A. On the Madden Julian oscillation and precipitation in Oregon and Washington. Weather Forecasting 18, 600-613 (2003).

Bretherton, C. S., Blossey, P. N. & Khairoutdinov, M. An energy-balance analysis of deep convective self-aggregation above uniform SST. Journal of the Atmospheric Sciences 62, 4273-4292 (2005).

Cassou, C. Intraseasonal interaction between the Madden-Julian oscillation and the North Atlantic oscillation. Nature 455, 523-527 (2008).

Gruber, A. The wavenumber-frequency spectra of satellite-measured brightness in the tropics. Journal of the Atmospheric Sciences 31, 1675-1680 (1974).

Jones, C. Occurrence of extreme precipitation events in California and relationships with the Madden Julian oscillation. Journal of Climate 13, 3576-3587 (2000).

Liu, C. & Moncrieff, M. W. Effects of convectively generated gravity waves and rotation on the organization of convection. Journal of the Atmospheric Sciences 61, 2218-2227 (2004).

Lorenz, E. N. The predictability of a flow which possesses many scales of motion. Tellus 21, 289-307 (1969).

Madden, R. A. & Julian, P. R. Detection of a 40-50 day oscillation in the zonal wind in the tropical Pacific. Journal of the Atmospheric Sciences 28, 702-708 (1971).

Madden, R. A. & Julian, P. R. Description of global-scale circulation cells in the tropics with a 40-50 day period. Journal of the Atmospheric Sciences 29, 1109-1123 (1972).

Madden, R. A. & Julian, P. R. Observations of the 40-50 day tropical oscillation: A review. Monthly Weather Review 122, 814-837 (1994).

Mapes, B. E. Gregarious tropical convection. Journal of the Atmospheric Sciences 50, 2026-2037 (1993).

Mapes, B. E. & Houze, R. Cloud clusters and superclusters over the oceanic warm pool. Monthly Weather Review 121, 1398-1415 (1993).

Matsuno, T. Quasi-geostrophic motions in the equatorial area. Journal of the Meteorological Society of Japan 44, 25-43 (1966).

Philander, S. G. H. El Niño, La Niña and the Southern Oscillation. Academic Press, 1990.

Reed, R. J. & Recker, E. E. Structure and properties of synoptic-scale wave disturbances in the equatorial western Pacific. Journal of the Atmospheric Sciences 28, 1117-1133 (1971).

Riehl, H. Tropical Meteorology. McGraw-Hill, 1954.

Sarachik, E. S. & Cane, M. A. The El Niño-Southern Oscillation Phenomenon. Cambridge, UK: Cambridge University Press, 2010.

Stevens, A. Introduction to the basic drivers of climate. Nature Education Knowledge 2, (2011).

Takayabu, Y. N. Large-scale cloud disturbances associated with equatorial waves. Part I: Spectral features of the cloud disturbances. Journal of the Meteorological Society of Japan 72, 433-448 (1994).

Wallace, J. M. & Kousky, V. M. Observational evidence of Kelvin waves in the tropical stratosphere. Journal of the Atmospheric Sciences 25, 280-292 (1968).

Wheeler, M. & Kiladis, G. N. Convectively coupled equatorial waves: Analysis of clouds and temperature in the wavenumber-frequency domain. Journal of the Atmospheric Sciences 56, 374-399 (1999).

Wheeler, M. & Weickmann, K. M. Real-time monitoring and prediction of modes of coherent synoptic to intraseasonal tropical variability. Monthly Weather Review 129, 2677-2694 (2002).

Yanai, M. & Maruyama, T. Stratospheric wave disturbances propagating over the equatorial Pacific. Journal of the Meteorological Society of Japan 44, 291-294 (1966).

Zangvil, A. Temporal and spatial behavior of large-scale disturbances in tropical cloudiness deduced from satellite brightness data. Monthly Weather Review 103, 904-920 (1975).

Email

Flag Inappropriate

This content is currently under construction.
Explore This Subject

Connect
Connect Send a message


Nature Education Home

Global and Regional Ecology

Visual Browse

Close