United States. Congress. House. Committee on Scien.

Road from Kyoto : hearing before the Committee on Science, U.S. House of Representatives, One Hundred Fifth Congress, second session (Volume pt. 2) online

. (page 18 of 137)
Online LibraryUnited States. Congress. House. Committee on ScienRoad from Kyoto : hearing before the Committee on Science, U.S. House of Representatives, One Hundred Fifth Congress, second session (Volume pt. 2) → online text (page 18 of 137)
Font size
QR-code for this ebook

feed on the available moisture through storm-scale moisture convergence, are likely to produce
correspondingly enhanced precipitation rates. Increases in heavy rainfall at the expense of
more moderate rainfall are the consequence along with increased runoff and risk of flooding.
However, because of constraints in the surface energy budget, there are also implications for
the frequency and/or efficiency of precipitation. It follows that increased attention should be
given to trends in atmospheric moisture content, and datasets on hourly precipitation rates
and frequency need to be developed and analyzed as well as total accumulation.

1. Introduction

The character of precipitation, with highly variable rain rates and enormous
spatial variability, makes simply determining mean precipitation difficult let
alone how it will change as the climate changes. For instance, a detailed
examination of spatial structure of daily precipitation amounts by Osborne
and Hulme (1997) shows that in Europe the average separation distance
between climate stations v.Lere the correlation falls to 0.5 is about 150
km in summer and 200 km in winter — the more convective nature of
summer precipitation is responsible. for the difference. In addition, this
complexity also makes it difficult to model precipitation reliably, as many of
the processes of importance can not be resolved by the model grid (typically
200 km) and so sub-grid-scale processes have to be parameterized. Yet

' The National Center for Atmospheric Research is sponsored by the National Science

Climatic Change 36: pp-pp, 1998

©1998 Kluwer Acadamic Publishers. Printed in the Netherlands.


there are some overall aspects of precipitation related to the hydrological
cycle that can be clarified and for which expectations as to how they will
change are physically baised. Here the processes involved that influence
precipitation and link it to evaporation and heating are outlined along with
the importance of dealing not just with accumulated amounts, but also
precipitation rates (or intensity) and precipitation frequency. The relative
roles of moisture stored in the atmosphere, its advection, and resupply have
been examined in detail in Trenberth (1998), and only a brief summary of
those aspects are included here.

The term "global warming" is often taken to refer to global increases
in temperature accompanying the increases in greenhouse gases in the
atmosphere. In fact it should refer to the additional global heating
(sometimes referred to as radiative forcing, e.g., by the IPCC (1996))
arising from the increased concentrations of greenhouse gases, such as
carbon dioxide, in the atmosphere. Increases in greenhouse gases in the
atmosphere produce global warming through an increase in downwelling
infrared radiation, and thus not only increaise surface temperatures but
also enhance the hydrological cycle, as much of the heating at the surface
goes into evaporating surface moisture. This occurs in all climate models
regardless of feedbacks, although the magnitude varies substantially (see
section 3).

Temperature increases signify that the water-holding capacity of the
atmosphere increases and, together with enhanced evaporation, the actual
atmospheric moisture should increase, as is observed to be happening in
many places (Hense et al. 1988, Gaffen et al. 1991, Ross and Elliott 1996,
Zhai and Eskridge 1997). Of course, enhanced evaporation depends upon the
availabiUty of sufficient surface moisture and over land, this depends on the
existing climate. However, it follows that naturally-occurring droughts are
likely to be exacerbated by enhanced potential evapotranspiration. Further,
globally there must be an increase in precipitation to balance the enhanced
evaporation but the processes by which precipitation is altered locally are
not well understood.

It is shown that precipitating systems of all kinds feed mostly on the
moisture already 1:. the atmosphere at the time the system develops, and
precipitation occurs through convergence of available moisture on the scale
of the system. Hence, the atmospheric moisture content directly affects
rainfall and snowfall rates, but not so clearly the precipitation frequency
and thus total precipitation, at least locally. Thus, it is argued that global
warming leads to increased moisture content of the atmosphere which in
turn favors stronger rainfall events, as is observed to be happening in many
parts of the world (Karl et al. 1995), thus increaising risk of flooding. It is
further argued tnat one reason why increases in rainfall should be spotty
is because of mismatches in the rates of rainfall versus evaporation. The


arguments assembled here imply the need for new observations, datasets,
and ways of analyzing both model and observed data. Trenberth (1998)
discusses these aspects more fully.

2. Atmospheric moisture cycling

New estimates of the moistening of the atmosphere through evaporation

at the surface and of the drying of the atmosphere through precipitation

are given in Trenberth (1998). These are simple estimates based on the

precipitable water and average local evaporation and precipitation rates,

which ignore transport. Overall for the global annual mean, the e-folding

residence time (the time for amounts to fall by a factor e = 2.718) for

atmospheric moisture is just over 8 days. For precipitation, local values of

e-folding residence time of the atmospheric depletion rate of moisture are

less than a week in the tropical convergence zones but they exceed a month

in the dry zones in the subtropics and desert areas. Time constants for

depletion and restoration rates of atmospheric moisture are fairly similar

overall, but this conclusion does not take account of the fact that rain falls

only a small fraction of the time. In midlatitudes precipitation typically falls

from zero to 30% of the time, and so rainfall rates, conditional on when rain

is falling, are much larger than evaporation rates. The depletion rate time

scale is about 4 hours in the tropics when rain is falhng. In middle latitudes,

typical unconditional rainfall rates are 3 mm/day, but with rain faUing about

10% of the time and precipitable water amounts of 15 mm, the depletion

rate time scale of 5 days drops conditionally on rain falhng to about 12 hours

(Trenberth 1998). This inferred imbalance in the drying versus moistening

of the atmosphere implies that most of the moderate and heavy rain that

falls comes directly from the precipitable water already in the atmosphere at

the time the storm responsible for the precipitation developed, not directly

from evaporation, and so the lifetime of moisture in the atmosphere and

its availability to rain systems is a limiting factor. However, atmospheric

depletion of moisture by Hght rain could easily be restored by evaporation.

These above aspects do not take moisture transport into account.

Therefore new estimates have also been made of how much precipitated

moisture comes from evaporation from within versus transport from outside

a domain, called recycling. Approximate values of recychng are computed

following the approach of Brubaker et al. (1993), as detailed in Ttenberth

(1998). EquiHbrium conditions are assumed, so that there are no changes

in atmospheric moisture content but changes in moisture storage in the

atmosphere do not impac^^ the results for seasonal or longer averages. A

domain of length L ahgned along the trajectory of the air is considered. An

important assumption is that the atmosphere is well mixed so that the ratio


Length scale = 1000 km

Recycling (p)

Annual (1979 -1995)

-1 — I — I — 1 — I I I — I — I — I — I — 1 — I — I — I — I — I — I — 1 — I — 1 — 1 — I — 1 — I — I — 1 — 1 — I — 1 — r— I — p — I — t" 1 1 I I I ' ' ' I '
30E 60E 90E 120E 150E 180 150W 120W 90W 60W SOW o o 16 32

<■■■(:} '-i



4 8 12 16 20 24 28 32

Figure 1. The recycling, for annucil mean conditions, for length scales of 1,000 km, and using
E and moisture flux from the NCEP reanalyses (Kalnay et al. 1996) and P from CMAP
(Xie and Arkin 1997). Values are set to missing (white) where the surface pressure is less
than 800 mb.

of precipitation that falls arising from advection versus local evaporation
is equal to the ratio of average advected to evaporated moisture in the
air. While interest has often been on recycling estimates for large drainage
basins, the heterogeneity of the land surface is such that the recycHng clearly
varies substantially over the basins. The regions of mountains (where surface
pressures are less than 800 mb) are screened out from the calculation,
as those are regions where the moisture flux is small and there are huge
variations over short distances owing to orographic eff"ects on rainfall.

In Trenberth (1998) recycling results for annual means are presented for
L= 500 km. Here results presented for L= 1000 km (Fig. 1) reveal recycling
percentages of aLoL.t 8 to 20% over land typically. For 500 km scales the
global mean is 9.6%, consisting of 8.9% over land and 9.9% over the oceans
and for 1000 km scales the mean recycUng is 16.8% globally, 15.4% over land
and 17.3% over the oceans. Over the Amazon, the average is about 10% and
over the Mississippi basin about 12%. These values prove to be compatible
with most previous extimates (e.g., Brubaker et al. 1993) once the different
scales of the basins are taken into account. It is worth pointing out that the
larger values previously obtained for the Amazon versus the Mississippi are
mostly a result of the scale of the domain.

The recycling fraction depends greatly on the magnitude of the total


moisture flux. In the computations, this includes advection by the mean flow
as well as the transient eddies. Relatively high values (>30%) of recycling
occur either in the subtropical highs, where evaporation E is high and the
advective moisture flux is small, or in convergence zones where, again, the
advective moisture flux is small. Low values occur over the southern oceans,
the North Pacific, and the eastern equatorial Pacific, where the moisture flux
is at a maximum. All of these recycling values show that on average less
than 20% of the precipitation that falls comes from evaporation within a
distance of about 1000 km. Therefore the results reinforce the arguments
given above concerning the importance of transport of moisture and local
storage in feeding precipitating systems.

The dominant storm-scale process in both thunderstorms and
extratropical storms is the convergence of moisture by the storm-scale
circulation. The latter determines how much moisture is available to the
system and can vary in size from a few tens of kilometers to over 2000 km
spatial scales. The advected moisture may combine with the in situ moisture
to feed the storm but it is not all available as the relative humidity can not
be reduced to zero, except perhaps approximately in strong down drafts very
locally. The efficiency of thunderstorms is observed to vary from about 20%
to 50%. "Precipitation efficiency" is defined as the ratio of the water mass
precipitated to the mass of water vapor entering the storm through its base
(e.g., Fankhauser 1988) or the ratio of total rainfall to total condensation in
modeling studies (e.g., Ferrier et al. 1996).

In the United States, much of the moisture for precipitation, especially
in the winter half year, comes from moisture transported out of the
subtropics often in a southwesterly flow ahead of cold fronts. For storms
eaist of the Rockies, moisture flows northwards from the Gulf of Mexico or
subtropical Atlantic. At advection rates of 12 m s~^ (which is the standard
deviation of the northward velocity component at 850 mb just north of the
Gulf of Mexico in January), the moisture travels over 1,000 km in a day, so
that moisture from the Gulf can be readily precipitated out over the Great
Plains or Ohio Valley just a day or so later. In major storms, transient
northward advection rates often exceed 20 m s~^ at 850 mb. In the western
United States, the moisture comes from the subtropical Pacific. Therefore
much of the extratropical precipitation originates from moisture advected
from the Gulf of Mexico and subtropical Atlantic or Pacific a day or so
earlier and it is estimated that about 70% to 75% of the moisture in an
extratropical storm comes trom moisture that was stored in the atmosphere
at the beginning of the storm and brought into the region by the storm-
scale circulation. For thunderstorms, whose life is a few hours, nearly all
of the precipitated moistrre comes from moisture that was already in the
atmosphere at the time the storm began.


3. Relevance to climate change

The above discussion reveals the mismatch between precipitation rates and
evaporation, so that moderate and heavy precipitation, which contributes
most to the total accumulation, depends upon the moisture already in
the atmosphere and the advection and resupply of moisture by the storm
circulation. These points are pertinent to climate change experiments with
global climate models. However, most climate model studies have not
analyzed the results in a way that throws light on these aspects. The surface
heat budget is especially relevant.

There are many feedback processes in nature that can either amplify
or diminish the climate response to increases in greenhouse gases. The
net radiative forcing or "warming" at the surface depends critically on
these and the surface heat budget. In every case it seems that at the
surface there is an increase in downwelling infrared radiation associated
with both the greenhouse effect from carbon dioxide and other greenhouse
gases, as well as changes in water vapor and clouds. In some models,
changes in clouds produce an offset by reducing shortwave radiation, but
the net energy available from radiation at the surface is increased in spite
of the greater surface emissions associated with the higher temperatures.
Moreover, changes in the sensible heat flux also act to warm the surface
because of stabilization of the lower atmosphere (Boer 1993, Roads et al.

This leaves only the latent heat flux through increased evaporation to
compensate and balance the surface heat budget. The latent heat flux, which
ranges from 3 to 10 W m"'^ for CO2 doubling for the four models considered
by Boer (1993), determines the global enhancement of the hydrological
cycle and average precipitation rate (of about 3 to 10%). However, the
atmospheric moisture content increases by about 20% (Mitchell et al. 1987)
or more (in the case of the CCM2, Roads et al. 1996) although with
very little change in model relative humidity. With other things kept
constant, moisture convergence would be enhanced by the same amount
and should lead to similarly enhanced precipitation. But a 20% increase
in precipitation cannot occur because of the hmitations associated with the
surface energy budget. Nevertheless such mechanisms should take place for
individual storms, whether thunderstorms, or extratropical cyclones, leading
to increased rainfall rates. If this is the case, however, there are implications
for the frequency of storms or other factors that must come into play to
restrict the total precipitation.

One factor clearly of importance is that the moisture increases are not
uniform. Generally, evaporative cooHng is more important in the tropics
and subtropics. Bigger increases occur in lower latitudes because of the non-
linear nature of the Clausius-Clapeyron equation in spite of larger increases


in surface temperatures at high latitudes. Thus much of this moisture may
not be within reach of many extratropical storms. Another factor is the
precipitation efficiency, discussed above. How precipitation efficiency might
change with cUmate change is not known and this is not a factor that can
be dealt with by current climate models. Warmer conditions could imply
that more moisture might remain in the atmosphere if this is determined
by relative humidity, as is likely. Therefore the rainfall may not increase in
direct proportion to the moisture convergence, because more moisture is left
in the atmosphere.

In most models, surface temperature increases with increased
greenhouse geises are greatest in the Arctic, in part because of ice-
albedo feedback, so that the meridional surface temperature gradient and
baroclinicity is reduced, although this may not be the case above the
surface. Therefore another factor relates to extratropical storms and
the overall baroclinicity, as argued by Held (1993). Held notes that
one effect of increased moisture content in the atmosphere is to enhance
the latent heating in such storms and thereby increase their intensity.
But he also notes that more moist air would be transported polewards,
reducing the required poleward energy transports normally accomplished
by baroclinically unstable eddies and the associated poleward down-gradient
heat transports. He therefore argues that this would contribute to "smaller
eddies" and a decrease in eddy ampUtudes. While recognizing that both
effects are important. Held suspects that the latter is dominant. There are
other possibilities not considered by Held. In particular, individual storms
could be more intense from the latent heat enhancement, but fewer and
farther between. Changes in the vertical temperature structure (the lapse
rate) will also play a role in such storms.

Therefore the other major factor worth considering in more detail is the
frequency of precipitation events. The above discussion suggests that for
rain rates to increase faster than rain amounts, then the frequency of rain
should decrease. However, this would only apply globally. A preliminary
examination of trends in frequency of precipitation events for the United
States computed over the period 1963 to 1994 in Trenberth (1998) shows
that the most notable staiisi-ically significant trends are for increases in the
southern United States in winter and decreases in the Pacific Northwest from
November through January, which may be related to changes in atmospheric
circulation and storm tracks associated with the trend toward more El Nino
events (Trenberth and Koar 1996). For instance, an example has been
the 1997-98 El Niiio winter which featured heavy rains across the southern
states from California to Florida, while somewhat drier conditions generally
prevailed across the northern states.

These aspects have been explored only to a limited extent in climate
models. None deal with true intensity of rainfall, which requires hourly (or


higher resolution) data, as the analysis is of daily rainfall amounts. Cubasch
et al. (1995) and Hennessy et al. (1997) have analyzed changes in intensity
and frequency in coarse resolution models with increased 002- Cubasch
et al. note that while precipitation change does not display a clear signal,
increases in rain intensity and dry periods are simulated in the ECHAM3
model. The UKHI and CSIR09 models (Hennessy et al. 1997) are consistent
in showing heavier rainfall events with doubled CO2, a general decrease in
the probability of moderate precipitation, and an increase in no or light
precipitation. Return periods for extreme events whose period is greater
than one year decrease by factors of 2 to 5. Hennessy et al. further argue
that the frequency of precipitation should be expected to decrease with
increases in intensity, and find this to be true in the model simulations for
the most part.

An analysis by Mearns et al. (1995) used a nested regional model
with 60 km resolution for regions of the United States for control and
doubled-carbon dioxide results. They explored the frequency and intensity
of modeled precipitation but only for daily values, not the true precipitation
rates. Results revealed increased daily rainfall variability under doubled
CO2. There are some areas where frequency of precipitation decreases but
precipitation mean daily amounts increase. Overall, however, they find both
increases and decreases of both precipitation frequency and intensity. Jones
et al. (1997) produced results over Europe using a similar technique and
a nested model with 50 km resolution. They find a substantial increase in
precipitation intensity in extreme events, and were able to trace most of that
increase simply to the increased atmospheric moisture concentrations in the
models. While moderate precipitation decreased, the frequency of dry days
also increased along with an increase in evaporation, and so these were all
symptoms of an increased hydrological cycle.

4. Conclusions and recommendations

The arguments on how cHmate change can influence moisture content of
the atmosphere, and its sources and sinks are assembled in the schematic in
Fig. 2. This provides the sequence described earlier. The sequence given is
simplified by omitting some of the feedbacks that can interfere. For example,
an increase m atmospheric moisture may lead to increased relative humidity
and increased clouds, which could cut down on solar radiation (enhance
shortwave cloud forcing) and reduce the energy available at the surface
for evaporation. Those feedbacks are included in the climate models and
alter the magnitude of the surface heat available for evaporation in different
models but not its sign. Figure 2 provides the rationale for why rainfall rates
and frequencies as well as accumulations are important in understanding


Greenhouse Gases


Radiative Forcing
(Global Warming)

Specific Humidity/Not RH


Cold Fronts
Warm Fronts
Tropical Cyclones

Extratroplcal Cyclones

Enhanced Precipitation

Increased Runoff
Increase'' Flooding

Enhanced Storm



Systems Feed






Precipitation amounts?
•Duration and size of systems
•Competition among systems
•Vertical heat transport
stabilizes systems

Figure 2. Schematic outline of the sequence of processes involved in climate change and
how they alter moi-iure content of the atmosphere, evaporation, and precipitation rates. All
precipitating systems feed on the available moisture leading to increases in precipitation rates
and feedbacks.



what is going on with precipitation locally. The accumulations depend
greatly on the frequency, size and duration of individual storms, ais well as
the rate (Byers 1948) and these depend on static stability and other factors
as well. In particular, the need to vertically transport heat absorbed at
the surface is a factor in convection and baroclinic instability both of which
act to stabilize the atmosphere. Increased greenhouse gases also stabilize
the atmosphere. Those are additional considerations in interpreting model
responses to increased greenhouse gais simulations.

Another clearly important factor in interpreting observed and modeled
changes, not explored here, is the changes in atmospheric circulation which
can alter the location and intensity of storm tracks and thereby lead to
dipole structures in precipitation changes, with decreases in rainfall in some
areas and increases in others. For example, Trenberth and Guillemot (1996)
show how storm tracks changed across North America to help bring about
the spring-summer 1988 drought and 1993 floods.

There is firm evidence that moisture in the atmosphere is increeising. In
the Western Hemisphere north of the equator, annual mean precipitable
water amounts below 500 mb are increasing over the United States,
Caribbean and Hawaii by about 5% per decade as a statistically significant
trend from 1973 to 1993 (Ross and Elliott 1996), and these correspond
to significant increases in relative humidities of 2 to 3% per decade over
the Southeast, Caribbean and subtropical Pacific. Precipitable water and
relative humidities are not increasing significantly over much of Canada,
however, and are decreasing sHghtly in some areas. In China, recent analysis
by Zhai and Eskridge (1997) also reveals upward trends in precipitable water
in all seasons and for the annual mean from 1970 to 1990. Earlier, Hense et
al. (1988) revealed increases in moisture in the western Pacific. A claim for
recent drying in the tropics by Schroeder and McGuirk (1998) using TOVS
data is questionable owing to the changes in instruments and satellites.
Clearly, there is a need to obtain more reliable atmospheric moisture trends

Online LibraryUnited States. Congress. House. Committee on ScienRoad from Kyoto : hearing before the Committee on Science, U.S. House of Representatives, One Hundred Fifth Congress, second session (Volume pt. 2) → online text (page 18 of 137)