Precipitation, Convection & Climate Change

An increase in tropical storm activity has long been predicted as a harmful side-effect of human induced global warming. Hurricanes are to become more frequent and more deadly. All of this is predicated on the notion that a hotter climate will result in more moisture in the atmosphere and more frequent tropical storms. As it turns out, this is only half true. There may be more precipitation in the temperate zone, but an increase in tropical storms is not predicted—even by the IPCC models.

According to basic atmospheric physics, if Earth's climate warms the upper atmosphere should heat up more than the surface. This is because of the way heat is transferred from the surface of the tropical ocean to the atmosphere and, eventually, back into space. An analysis of sea surface temperatures and rainfall appearing in Nature Geoscience suggests amplified warming in the upper atmosphere does occur in a way consistent with theory. This effect is even predicted by computerized climate models. Reporting in “Changes in the sea surface temperature threshold for tropical convection,” Nathaniel C. Johnson and Shang-Ping Xie counter claims that observed warming during the twentieth century did not conform to this prediction.

Tropical storms can be thought of as one of nature's mechanisms for moving heat energy from the surface of the ocean, through the warm air of the lower atmosphere to the base of the stratosphere. A tropical cyclone is a large convective cell driven by the latent heat carried aloft by warm moist air. The water vapor in the cloud condenses into water droplets releasing the latent heat which originally evaporated the water.

This latent heat provides the energy to drive the tropical cyclone circulation, though very little of the heat released is actually utilized by the storm to increase wind speeds and surface pressure. For such a convective cell to form conditions in the atmosphere and the temperature of the sea surface must be just right. To understand the needed conditions consider the structure of the lower atmosphere.

Temperature change in atmospheric layers.

The stratosphere is the second major layer of Earth's atmosphere, beginning around 8 km (5 miles) high, just above the troposphere. It is called the stratosphere because it has layers of different temperature, which are stratified with warmer layers higher up and cooler layers lower down. In contrast, the troposphere near the Earth's surface is cooler higher up and warmer nearest the surface. The border of the troposphere and stratosphere, the tropopause, is where cooling with increasing altitude changes to warming.

Most weather stops at the tropopause—running into the tropopause is what causes the flat-topped, anvil shape of tall storm clouds. The tropopause acts as a barrier between the troposphere and stratosphere because mixing and heat transport by convection can only occur when temperature decreases with height. In an accompanying news article, Adam Sobel provides this description of tropospheric convection:

To understand the vertical temperature structure of the tropical atmosphere, one should start by considering a saturated isolated air bubble, or 'parcel', ascending from near a tropical ocean surface with uniform sea surface temperature (SST). The air just above the ocean is close to thermodynamic equilibrium with the sea surface: its temperature is close to the SST, and its relative humidity is 80–90%. As the parcel rises in a cloud updraft, its water vapour condenses while the temperature drops as a function of altitude following a particular curve, a moist adiabat. The shape of the moist adiabat depends on the initial temperature and absolute humidity that the parcel starts with and is determined by basic thermodynamics. Because such convective clouds occur frequently over the tropical oceans, the vertical atmospheric temperature profile adjusts on large scales to match that of the rising parcels, and thus becomes moist adiabatic as well.

Now impose global warming. Both the temperature and absolute humidity of the surface air increase, so cloud parcels follow a warmer moist adiabat than before. These air parcels are not only warmer, but also contain more latent energy in the form of water vapour that is converted to sensible heat as it condenses. As these warmer and moister parcels ascend, condensation converts the additional latent energy into a smaller temperature decrease with altitude (Fig. 1). The new moist adiabat is warmer at all levels than the old one, but by a greater amount at high altitude, because by that point most of the water vapour has condensed.

Fig 1: a, In a climate before warming, convection and heavy tropical rain is restricted to a region where SSTs exceed a threshold value (dotted line), and temperatures decrease with altitude. b, Johnson and Xie show that this SST threshold has risen in tandem with mean SSTs over past decades, and that the area of surface ocean where convection occurs has remained constant. Adapted from Sobel.

Simply put, the warmer the sea surface is with respect to the temperature of the atmosphere the more moisture gets into the air and the more moisture, the more energy is available to drive convection. Further more, deep convection over tropical oceans—the type that leads to tropical storms—is observed generally above a threshold for sea surface temperatures, which falls in the vicinity of 26–28°C (79-82°F). As Johnson and Xie explained:

The physical explanation for the SST threshold relates to the strong dependence of atmospheric instability on local SST over the tropical oceans, which owes to two important characteristics of the tropical troposphere: the strong relationship between boundary-layer moist static energy and SST, and the weak horizontal gradients in free-tropospheric temperature.

Above the SST threshold temperature deep convection can occur, but below this value the atmosphere is too stable and little or no thunderstorm activity takes place. This is also tied to the Convective Available Potential Energy (CAPE), a measure of the amount of energy available for convection. CAPE is directly related to the maximum potential vertical speed within an updraft, with higher values indicate greater potential for severe weather. In practical terms this means that if a rising packet of air cools to a temperature lower than the air above it convection is stopped. All of this implies that warming in the upper troposphere would put a damper on storm formation.

In the past, upper-atmospheric temperature records from both radiosondes and satellites have been called into question (see “Reconciling observations of global temperature change” and “Temperature Trends in the Lower Atmosphere: Steps for Understanding and Reconciling Differences”). Johnson and Xie use a new method based only on sea surface temperature and precipitation observations, eliminating the confusion caused by inconsistent historical radiosonde and satellite records.

The Moist-Adiabatic Lapse Rate

One more term needs to be explained before examining Johnson and Xie's work and that is the moist-adiabatic lapse rate (MALR). A lapse rate is defined as the rate of decrease with height for an atmospheric variable. The moist-adiabatic lapse rate, or saturation-adiabatic lapse rate, is the rate of decrease of temperature with height along a moist adiabat. In thermodynamics, an adiabatic process is a thermodynamic process in which no heat is transferred to or from the working fluid. A moist adiabat is the curve describing air temperature change inside a rising parcel of moisture saturated air assuming no heat transfer with the surrounding air.

As the parcel rises, pressure drops causing the parcel to cool adiabatically. The moist adiabatic lapse rate is less than the dry adiabatic lapse rate because as vapor condenses into water (or water freezes into ice) for a saturated parcel, latent heat is released into the parcel, mitigating the adiabatic cooling. As defined by the American Meteorological Society, the MALR is given approximately by Γm in the following equation:

Where g is gravitational acceleration, cpd is the specific heat at constant pressure of dry air, rv is the mixing ratio of water vapor, Lv is the latent heat of vaporization, R is the gas constant for dry air, ε is the ratio of the gas constants for dry air and water vapor, and T is temperature. Look here for more details. Johnson and Xie's hypotesis is that changes in the MALR can be tied to changes in the SST threshold and tropical sea surface temperatures:

This hypothesis of moist-adiabatic lapse rate (MALR) adjustment predicts a close covariability between the SST threshold and tropical mean SST. If true, the variability and long-term trend of the SST threshold may reveal important information about the variability and trends in the tropical troposphere. We test this hypothesis with the use of observations spanning multiple decades and with state-of-the-art global climate models. For the observational data, we use two widely used global precipitation products, the Global Precipitation Climatology Project16 (GPCP) and the Climate Prediction Center merged analysis of precipitation17 (CMAP), which provide precipitation estimates based on a mix of satellite and rain gauge measurements from 1979 until present.

Johnson and Xie found that the SST threshold for convection has risen in sync with the tropical mean SST in the past few decades. “Estimates show a remarkable correspondence between the tropical mean SST and the convective threshold,” they state. “Both the tropical mean SST and the SST threshold for convection have undergone an upward trend of approximately 0.1°C per decade.”

Since Earth did warm over the past half century, this finding is consistent with both observation and theory. It also implies that the upper troposphere warmed more than the surface. This is inferred from the fact that overall precipitation rates did not increase. If the temperature aloft had not increased more than surface temperatures, surface air parcels would have required less heat to rise and eventually condense into rain. What is more interesting is what happens when the researchers use climate models to project the impact of such tropospheric warming on future tropical storm activities.

Next, we examine SST threshold variability in ten global climate models of the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC AR4), through the use of the World Climate Research Program Coupled Model Intercomparison Project 3 (WCRP CMIP3) multi-model database archive. Despite differences in the simulated tropical climate among the models and consistent with ref. 10, all ten models exhibit a similarly strong correspondence between the tropical mean SST and the SST threshold on all timescales. In all cases but one, the correlation coefficient between the tropical mean SST and SST threshold equals or exceeds 0.90 for the climate of the twentieth century (20C3M) simulations and for the twenty-first-century emissions scenario A1B simulations.

As can be seen from the figure below, the 10 models gave quantitatively different answers but the general trend for all was consistent. “Despite the differences in the simulated tropical mean state, as evidenced by the scatter among the models, there is a remarkable correspondence between the SST threshold and the tropical mean 300 hPa temperature for each individual model,” report the authors. “On the basis of the regression of SST threshold on tropical mean 300 hPa temperature, the tropical mean 300 hPa temperature explains, in a statistical sense, more than 92% of the variance of the convective threshold in all of the climate models.”

Scatter plot of SST threshold versus tropical mean 300 hPa temperature with regression lines for each of ten CMIP3 models under emissions scenario A1B in simulations of the twenty-first century. Source Johnson and Xie.

After noting the “nearly perfect” correspondence between changes in SST threshold for convection and tropical mean SST, the authors analyzed two twenty year periods using the models and IPCC warming scenario A1B. Despite the assumption of a substantially warmer climate, and correspondingly increased moisture, the model results suggest that the portion of the tropical oceans that is convectively active may change little. The figure below reveals remarkable similarity in the curves between the two periods, which Johnson and Xie suggest means that, “relative to the tropical mean SST, the rainfall rate and SST frequency distributions are projected to change little under global warming.”

Ensemble mean rainfall rate as a function of SST (a) and SST frequency distribution (b) for 2001–2020 (blue, solid) and 2081–2100 (red, dashed) for the ten CMIP3 models.

The primary conclusion of this investigation is that rainfall, even under an aggressive and unproven IPCC global warming scenario, will not change in any significant way. Furthermore, given the linkage between SST and upper tropospheric warming, there will be no increase in tropical storm activity. Even using the same wonky climate models as the IPCC AR4 report, there is no reason to expect crippling droughts or massive deluges because of global warming. As Sobel concluded:

From the findings reported by Johnson and Xie we should therefore not conclude that some rainy regions cannot become drier in a future, warmer climate (or vice versa), or that the tropical-cyclone-prone zones cannot shift. But the results suggest that, for climate change that might plausibly occur in the near future, there may be a conservation of total rainy area, such that losses in rain somewhere are compensated by gains elsewhere. If so, the reasons for this are unclear. There could be a fundamental principle of climate dynamics waiting to be uncovered.

This is consistent with other reports that historic bouts of real global warming did not harm tropical forests. Earth's tropical forests are not threatened by climate change—what threatens tropical forests are misguided efforts to produce biofuels and plan old human greed. Bottom line, even if there is significant global warming there will be no significant increase in tropical storms or change in tropical precipitation.

Be safe, enjoy the interglacial and stay skeptical.

The Twentieth Century Reanalysis Project;jsessionid=74...

The Twentieth Century Reanalysis (20CR) project is an international effort to produce a comprehensive global atmospheric circulation dataset spanning the twentieth century, assimilating only surface pressure reports and using observed monthly sea-surface temperature and sea-ice distributions as boundary conditions. It is chiefly motivated by a need to provide an observational dataset with quantified uncertainties for validations of climate model simulations of the twentieth century on all time-scales, with emphasis on the statistics of daily weather. It uses an Ensemble Kalman Filter data assimilation method with background ‘first guess’ fields supplied by an ensemble of forecasts from a global numerical weather prediction model. This directly yields a global analysis every 6 hours as the most likely state of the atmosphere, and also an uncertainty estimate of that analysis.

The 20CR dataset provides the first estimates of global tropospheric variability, and of the dataset's time-varying quality, from 1871 to the present at 6-hourly temporal and 2° spatial resolutions. Intercomparisons with independent radiosonde data indicate that the reanalyses are generally of high quality. The quality in the extratropical Northern Hemisphere throughout the century is similar to that of current three-day operational NWP forecasts. Intercomparisons over the second half-century of these surface-based reanalyses with other reanalyses that also make use of upper-air and satellite data are equally encouraging.

It is anticipated that the 20CR dataset will be a valuable resource to the climate research community for both model validations and diagnostic studies. Some surprising results are already evident. For instance, the long-term trends of indices representing the North Atlantic Oscillation, the tropical Pacific Walker Circulation, and the Pacific–North American pattern are weak or non-existent over the full period of record. The long-term trends of zonally averaged precipitation minus evaporation also differ in character from those in climate model simulations of the twentieth century.

just came across this today. Hopefully it will shed a more accurate light on past weather events. one thing people also seem to forget when reading media reports about how such and such storm has caused the most money in $$ damages is that 100 years ago there were a lot less people and buildings that could have been damaged here in the US and a whole lot less accurate record keeping. so of course any storm nowadays is going to do more damage as there is more damage to be done.

Wait and see

I'm afraid I'll have to take a wait and see attitude about this new, universal dataset. Since a number of experts on things like the Atlantic Multidecadal Oscillation claim they only have 15 years of decent data, I find it hard to trust a reconstruction going back to the 1800s. And the statement “It uses an Ensemble Kalman Filter data assimilation method with background ‘first guess’ fields supplied by an ensemble of forecasts from a global numerical weather prediction model,” does not fill me with confidence. You can filter and smooth your data all you want, if it was crap to begin with all you end up with is smooth filtered crap.

No More Storms

Thanks Doug, this was a very informative article. And it makes sense to a layman like me.

So now I have some armoury when alarmists claim the Australian and pakistan floods were due to AGW.

p.s. If the IPCC report supports the conclusion of this paper, how is it that the Synthesis Report and the Summary for Policy Makers can claim more droughts and floods for Australia?


The observations reinforced by the article I reviewed only came to light around 2008, so the original IPCC reports were not based on accurate information. What's going on in Australia was the result of a La Niña coupled with a tropical cyclone--a lot of rain on top of already saturated ground and you get a 100 year flood. Nasty, devastating but natural. I will have a follow on article about the recent wicked weather soon.