Meteorol. Zeitschrift, N.F. 1,122-138 (April 1992) @ by Gebriider Borntraeger 1992

Water vapour as an amplifier of the greenhouse effect: new aspects

H. FLOHN, A. KAPALA, H. R. KNOCHE and H. MXCHEL, Bonn

Summary. In view of the wide-spread negligence concerning water vapour as the most efficient greenhouse gas, its role in the recent climatic evolution is investigated. Increase of tropospheric water vapour content and rising evaporation (by about 15 %) from tropi­cal oceans have been found in recent decades. Evidence for an accelerated hydrologic cycle is given, leading simultaneously [Q a remarkable intensification of the tropospheric circulation in the Northern Hemisphere. Average geostrophic wind speed at the surface and in the troposphere have increased by 6-9 % between 1961 and 1989. This (natural) internal feedback through water vapour amplifies the "dry" greenhouse effect of CO2 and other trace gases by a factor of about 5, spreading poleward from tropical oceans.

Wasserdampf als ein Verstarker des Treibhaus-Effektes: Neue Aspekte

Zusammenfassung. Angesichts der weitverbreiteten Vernachlassi­gung des effektivsten Treibhaus-Gases, des Wasserdampfs, wird seine Rolle in der jiingsten Klimaentwicklung untersucht. Dber den tropischen Ozeanen wurde in den letzten Jahrzehnten ein Anstieg des tropospharischen Wasserdampfgehalts und eine Zu­nahme der Verdunstung (ungefahr + 15 %) beobachtet. Dies be­legt eine Beschleunigung des hydrologischen Zyklus, die zugleich zu einer deutlichen Verstarkung der tropospharischen Zirkulation auf der Nordhemisphare fiihrt. Die mittlere Geschwindigkeit des geostrophischen Windes an der ErdoberfHiche und in der Tropo­sphare ist zwischen 1967 und 1989 urn 6-9 % angestiegen. Diese (natiirliche) interne Riickkopplung iiber den Wasserdampf ver­starkt den "trockenen" Treibhauseffekt von CO2 und anderen Spurengasen um einen Faktor nahe 5 und breitet sich von den tropischen Ozeanen polwarts aus.

1. Introduction

In this paper, the authors intend to outline the evidence for a recent intensification of the hydrological cycle and of the tropospheric circulation of the Northern Hemisphere. It is an extension of two recent papers (FLOHN et al. 1990 a, b), based on marine COADS data for the extended period 1949-1989 and daily hemispheric maps from the German Weather Service between November 1966 and February 1989. The evaluation of this data has also been improved; investigations on the controversial positive trend of surface wind speed at the oceans using geostrophic data are present­ed in the Appendix.

Recent warnings concerning the expected global warm­ing and climatic change unprecedented since 104 (or 105

)

years had already been issued by leading U. S. scientists, based mainly on the unconvincing course of global surface temperature (ELLSAESSER 1986, 1990) as well as on the conflicting results of climate models. However, our main arguments are essentially different:

a) observed surface data alone are still controversial, in spite of the convincing review of SCHONWIESE & STAH­LER (1991). Stronger mid-tropospheric warming and stratospheric cooling are already in progress (Table 1),

Table 1. Difference (1911-89) - (196J-16) of mean temperature in °C for selected atmospheric layers and latitudinal belts (K. ANGELL 1990).

Tabelle 1. M ittlere T emperatur fur ausgewiihlte S chichten der A tmo­sphiire und Breitenzonen; Differenz (1911-89)-(1%3-76) in °C (K. ANGELL 1990).

Pressure (kPa) 85-30 30-10 10-5 Altitude (km) 1.5-9 9-16 16-20

N Polar (60-90 ON) to.16 +0.10 +0.22 Tropical (30 05-30 ON) +0.27 -0.32 -0.69 Equatorial (10 °5-10 ON) +0.36 -0.41 -0.99 5 Polar (60-90 °S) +0.40 -0.46 -1.07 Global +0.36 -0.20 -0.48

b) many more parameters are necessary (and data is availa­ble) for a conclusive assessment of on-going climatic change,

c) climati..: models which are not verified with regard to the complete actual surface climate with its seasonal varia­tions and vagaries of weather alone are not sufficiently reliable for recommending far-reaching and expensive strategies to the decision-maker,

d) the complete and critical evaluation of all kinds of obser­vations is of the same importance as the critical compari­son of comprehensive, truly interacting climate models. Here, only arguments a) and b) shaH be discussed in

detail; argument c) is shared by many experienced special­ists.

In preparation for the second World Climate Conference in Geneva (WCC II, October 1990), the Intergovernmental Panel on Climatic Change (IPCC) issued a comprehensive report "Assessment of Climatic Change" reviewing the physical background and discussing several scenarios for the future fate of the composition of the atmosphere and its effect on climate, together with impacts of climate on hu­man activities and strategies for its mitigation. This report (HOUGHTON et al. 1990) leads the reader to the conclusion that the "scientific community" has reached a consensus on the existence of a man-triggered global wanning and on the necessity of far-reaching measures to at least reduce the future emissions of greenhouse gases. In some respect, the reports of the German Parliament's Enquere Commission

0941-2948/92/0001-0122 $ 07.65 @ Gebriider Borntraeger, Berlin, Stuttgart 1992

121 N.F. 1.Jg. 1992, H.2 M. Dorninger et a1.: A thermodynamic diagnostic model for the atmosphere I

References

Bowen, I. S. (1926): The ratio of heat losses by conduction and by evaporation from any water surface. - Phys. Rev. 27,779­787.

Burkhardt, T. (1990): Subgrid-scale vertical heat fluxes over the African-Atlantic region. - Bonner Meteorol. Abh. 39, 105 pp.

(1991): Subgrid-scale vertical heat fluxes for the African region. Meteorol. Atmos. Phys. 46,155-168.

Cressmann, G. P. (1959): An operational objective analysis system. - Month. Weath. Rev. 87,367-374.

Dorninger, M. (1992): Diagnose subskaliger Vertikalfliisse uber Europa. Diploma thesis at the University of Vienna (in German), 96 pp.

Emeis, S. (1985a): Subsynoptic vertical heat fluxes in the atmos­phere over Europe. - Bonner Meteorol. Abh. 32, 106 pp.

- (1985b): Kleinskalige Energietransporte in der freien At­mosphare. - Geowiss. in unserer Zeit 3, 181-186.

- (1986): Subsynoptic vertical energy fluxes in mid-latitude cy­clones. Meteorol. Rdsch. 39,161-172.

Emeis, S. & Hantel, M. (1984): ALPEX-Diagnostics: Subsynoptic heat fluxes. - Beitr. Phys. Atmos. 57, 495-511.

Hantel, M. (1982): ALPEX: Subsynoptische Vertikaltransporte. - Ann. Meteorol. (N.E) 19, 79-80.

- (1986a): Variational modification of the 3D-wind field. - In: Variational Methods in the Geosciences, Y. Sasaki (ed.), El­sevier, 309 pp.

(1986b): Rain in the free atmosphere from synoptic data. -MeteoroL Rdsch. 39, 102-111.

(1987): Subsynoptic vertical heat fluxes from high-resolution synoptic budgets. Meteorol. Atmos. Phys. 36, 24-44.

- (1989): Eine neue Gleichung fiir die atmospharischen Energie­haushalte. - Wetteru. Leben 41,95-108.

Hantel, M. & Emeis, S. (1985): A diagnostic model for synoptic heat budgets. - Arch. Meteorol. Geophys. Bioklimat., Ser. A, 33,407-420.

Hantel, M., Reimer, E. & Speth, P. (1984): ALPEX-diagnostics: Quantitative synoptics over Europe. Beitr. Phys. Atmos. 57,477-494.

Hantel, M., Ehrendorfer, M. & Haimberger, L (1992): A thermo­dynamic diagnostic model for the atmosphere. Part II: The general theory and its consequences. - Manuscript in prep.

Holtsiag, A. A. M. & Ulden A. P. van (1983): A simple scheme for daytime estimates of the surface fluxes from routine weather data. - J. Clim. App!. MeteoroL 22, 517-529.

Reimer, E. (1980): A test of objective meteorological analysis with optimum utilization of the radiosonde network in Central Europe. - Beitr. Phys. Atmos. 53,311-335.

Sevruk, B. (1985): Correction of precipitation measurements. -Ziircher Geogr. Schr. 23, 288 pp.

Shao, Y. & Hantel, M. (1986): Sub synoptic vertical momentum flux in the atmosphere over Europe. - Bonner Meteorol. Abh. 33,159 pp.

Spangl, W. (1991): Skalige Budgets von Warme- und Feuchteglei­chung. Teil1: Das Diagnostische Modell. UnpubL manu­script (in German), available from the authors, 88 pp.

Stull, R. B. (1988): An introduction to boundary layer meteorol­ogy. Kiuwer, 666 pp.

Wang, Y., Ehrendorfer, M. & Hantel, M. (1992): 3D-mass flux modification: Theory and numerics. - Manuscript in prep.

Yanai, M., Esbensen. S. K. & Chu, J. H. (1973): Determination of bulk properties of tropical cloud clusters from large-scale heat and moisture budgets. - J. Atmos. Sci. 30,611-627.

M. DORNINGER

Dr. M. EHRENDORFER

Prof. Dr. M. HANTEL

E RUBEL

Y. WANG

Institut fUr Meteorologie und Geophysik Universitat Wien Hohe Warte 38 A-ll90Wien

Received 12.11.1991, accepted: 28.11.1991

123 N.F. I.Jg. 1992, H. 2 H. Flohn et al.: Water vapour as an amplifier of the greenhouse effect

"Vorsorge zum Schutz der Erdatmosphare" (Precautions for the Protection of the Earth's Atmosphere) - unani­mously agreed upon in early November 1990 - are more complete and well balanced (DEUTSCHER BUNDESTAG 1990).

Meanwhile, a vivid discussion arose, particularly in the United States, questioning the role of greenhouse gases and stressing the view that the present global warming did not pass beyond the limits of natural fluctuations as observed during the last 100-150 years (ELSAESSER 1986, 1990). Unfortunately, this discussion, as well as Section 7 of the IPCC-Report, concentrate primarily on surface tempera­ture and omit many other aspects of the climatic evolution during the last 3-4 decades. In contrast to S. ARRHENIUS' classical paper (1896) and in spite of the well-founded esti­mates of F. MOLLER (1963) and V. RAMANATHAN (1981), the leading role of water vapour has widely been neglected, as stressed by K. KONDRATYEV and others during WCC II. As we shall see, the hydrological cycle, using about 88 % of the available net radiation at the ocean's surface, and clouds play a dominant role in the chain of climatogenetic processes.

Furthermore, climate, within the geophysical science, can only be understood in relation to the 3-dimensional circulation of atmosphere and oceans. Recent evolution of atmospheric 3D-circulation can be followed at the North­ern Hemisphere since about 1948. The first marked period of global warming, in the 1920's and 1930's, has been de­scribed by R. SCHERHAG (1936a, b, 1939) based on surface maps of temperature and pressure; now we have to look in detail at the entire troposphere and the lower stratosphere.

2. Comments on on-going climatogenetic processes

In addition to the emission of (dry) greenhouse gases, which is the main item of present discussion, several other pro­cesses are operating within the climate system. Among them, some comments shall be given to wide-spread changes of land use, in cloudiness, and within the oceanic mixing layer.

1) Changes of land use - mainly in tropical and subtrop­ical continents -lead to higher surface albedo (deforesta­tion, shifting cultivation, over-grazing) and lower soil mois­ture (soil erosion, biomass burning). These and other ex­amples are consequences of the increasing growth rate of human population, which was also stressed at WCC II.

2) Increasing cloudiness (HENDERSON-SELLERS 1986, 1989) and decreasing sunshine (WEBER 1990, ANGELL 1990) may lead d. the important results of the Earth Radiation Budget Satellite (RAVAL et al. 1989, RAMANATHAN 1989) ­to a decrease of incoming solar radiation at the surface and thus apparently counteract the greenhouse effect. However, this refers mainly to lower and middle clouds; cold, thin cirrus sheets as diverging from increasing tropical convec­tion (see Chapter 3) or produced by aircraft condensation trails indeed amplify the greenhouse effect.

3) Increasing ocean surface temperature (FLOHN & KA· PALA 1989, FLOHr-.; et aL 1990 a, b) causes - with constant

or sinking relative humidity and rising wind speed (see Chapters 3, 4 and Appendix) an exponential rise of eva­poration and consequently of the hydrological cycle. In ad­dition to the (radiational) greenhouse effect, this physically different process injects latent heat energy into the atmos­phere to be released by precipitation. At many coasts and in marginal seas, increasing inflow of nutrients intensify the production of planctonic algae and the turbidity of surface waters, thus enhance the rate of heating at the ocean's surface (SATHYENDRANATH et al. 1991).

The examples indicate that man's role for climate change is much more complicated than that caused by an increase of "greenhouse gases". The role of biogeochemical pro­cesses in the carbon cycle has been reviewed in the IPCC­Report, Section 1.

3. Role of water vapour in on-going climatic change

One of the most essential effects of water vapour occurs during its transport, together with latent heat, from its surface source (evaporation E) to its sink (precipitation P) in the cloud layer of the troposphere. In tropical latitudes, where strong convective clouds reach heights of up to 17 km, the bulk of latent heat is released at altitudes of 2-6 km, in the extratropics mainly at 1-3 km, thus warming the lower and middle troposphere. One millimetre of rainfall (equivalent to 1 liter/m2

) releases an energy of 28.6 W/m 2

(equivalent 57 cal/cm2/day); mean rainfall rates of 4-8 mm/d are normal in the humid tropics. Since the radiational cooling of air is only of the order 0.1 K/h, part ofthe released latent heat remains within the troposphere for several days. Typical area-averaged rainfall rates in convective cloud clus­ters and tropical disturbances are 20-30 mm/d; this releases an energy larger than the extraterrestrial solar constant at a rotating earth (340 W/m2). During large-scale disturbances the total area of such cloud clusters can reach 105-106 km2

•

The average atmospheric residence time of water vapour can be derived from the conventional figures of global pre­cipitation (100 cm/a) and precipitable water of the atmos­phere (25 mm) (PEIXOTO & OORT 1983). Their ratio indi­cates about 40 recyclings per year and thus an atmospheric residence time of 9 days. Following PEIXOTO & OORT again (and some of our own local investigations), the speed of vertically integrated transport of water vapour is in the order of 2-2.5 m/secorabout 200km/d, prevailing in zonal direction. The average distance between source (E) and sink (P) is then in the order of 2000 km, an estimate near the upper limit.

Release of latent heat follows the (relatively small-scale and noisy) rainfall pattern. This differential heating intensi­fies all types of ascending motion, including baroclinic waves and synoptic-scale cyclones. It should always be re­membered that, from the viewpoint ofphysics, this effect­due to phase changes - is different from the (radiational) greenhouse effect. It is indeed a natural (!) feedback within the climate system, using weak man-induced warming of the ocean's upper layers for intensifying the atmospheric circu­lation at local, regional, and hemispheric scales.

124 H. Flohn et al.: Water vapour as an amplifier of the greenhouse effect Meteoro!' Zeitsehrift

The hydrological cycle embraces evaporation (E), precip­itation (P), and continental runoff (Rf). Disregarding areas with internal drainage at the arid parts of the continents, the hydrological budget is written:

PL-EL-Rf = Ps+Rf-Es=O I ! i ; I !

Continents Oceans

Here all time-dependent storage terms have been neglected; in the oceans, increasing (decreasing) values of Es-Ps, i.e. of the fresh-water flux into the air, are indicated by rising (diminishing) salinity (S).

Table 2. Global water budget estimates after BUDYKO (1978) and German authors (ef text) in em water column/year.

Tabelle 2. Globaler Wasserhaushalt in em Wassers;"'ulelJahr, ge­sehatzt naeh BUDYKO (1978) und nach deutschen Autoren (vgi. Text).

Continents Oceans Globe PERf P E P E

BUDYKO, KORZUK 80 45 35 127 140 113 cmla German authors 78 44 34 104 118 97 cmla U.S. Nat. Res. 72 48 24 110 120 99 cmla

Council

P Precipitation, E Evaporation, Rf Runoff.

Unfortunately - but typical for our incomplete knowl­edge of the only planet inhabited by human beings all these basic figures are hitherto hardly more than educated guesses. Table 2 is primarily derived from HENNING (1989, Table 15), and demonstrates the deplorable gaps in our knowledge of the hydrologic balance, one of the essential parameters for the well-being of the fearfully and rapidly growing human population. After 30 years of expensive space research, geophysical scientists can only hope that the planned Global Energy and Water Budget Experiment (GE­WEX) may fill this scandalous gap in a foreseeable future.

At land, the number of available long records of precipi­tation (P) is rather small, except in the continental belt lat. 35-70 ON, where P has distinctly increased since about 1960 (BRADLEY et al. 1987, DIAZ et al. 1989). In Europe, an increase between 40 and 70 ON has been noted (SCHCNWIESE & BIRRONG 1990) since a century or more. In contrast to this, P diminishes in the subtropical belt between 35 and 5 ON extending also to northern Africa, here related to both the mediterranean winter rains and the tropical summer rains. Tropical rainfall data are hardly sufficiently representative for a reliable zonal analysis; since about 1962, the area­weighted Indian Summer Monsoon rains tend to remain slightly below average (e.g. GREGORY 1989).

Using global rainfall rank statistics from 2 °X2 ° boxes, the Climatic Diagnostic Center (HALPERT & ROPELEWSKI 1991) has shown a rise of the median rank from 0.44 (1951) to 0.56 (1990), which is statistically significant. U nfonu­nately, these statistics - well-adapted to the great spatial and temporal variance of precipitation - do not allow a quantitative estimate of the increase of precipitation.

In the Atlantic, data of the trend of salinity (S) between 1955-59 and 1970-74 (LEVITUS 1989) clearly indicate a freshening (i.e. more rainfall) at the belt 0-12 oN, here reinforced by rising Amazon discharge, as well as at the belt 40-60 ON, similar to the positive trend of land data. In equatorial areas, precipitation (P) depends on the height of towering cumulonimbi which penetrate the middle tropos­phere with its moist-static stability to altitudes up to 17 km, spreading large cirrus sheets outwards. The available latent energy controls the height of cloud tops - the latter can be estimated from satellites measuring outgoing in­frared radiation (OLR) decreasing with their temperature. First-order approximations of a conversion between P and OLR are available; these allow only a rough estimate of area-averaged P for areas of the order of 75000 km2

• Such data are evaluated e.g. by NITTA & YAMADA (1989), sug­gesting if homogeneous - a recent increase of P in the equatorial belt. DIAZ et al. (1989) has also found a significant increase of rainfall for the belt lat. 0-25 oS.

Applying the well-known bulk formula (for discussion see ISEMER & HASSE 1987,1991), FLOHN & KAPALA (1989) estimated evaporation (E) from the main shipping routes at the belt 10 °S-14 ON (Fig. 1) for the period 1949-79.

Table 3. Variations ofsea-air parameters 1949-1989, annual values derived from linear regressions.

Tabelle 3. A'nderung der Ozean-Atmosphiire-Parameter 1949­1989, Jahreswerte abgeleitet von linearer Regression.

Atlantic Warmest oceans

Upwelling Zonal regions averages

T,(oq Mean 27.04 28.45 24.15 27.37

(Sea surface) temperature)

1949 1989 .:linoC

26.94 27.14 +0.20

28.16 28.74

+0.58

23.83 24.46

+0.63

27.13 27.61 +0.48

T,-T,(°C)

(Sea surface minus air temperuture)

Mean

1949 1989 .:lin %

0.45

0.48 0.41

-14.6

0.47

0.39 0.54

+38.5

0.16

0.34 -0.02

-105.9

0.47

0.47 0.47

+0.0 .J q,-q. (g/kg)

(Spec. hu­midity at T, (saturated) minus at T, (air))

Mean

1949 1989 .:lin %

4.75

4.69 4.81

+2.6

5.32

5.01 5.62

+12.2

3.42

3.32 3.51

+5.7

4.90

4.71 5.09

+8.1

V''''l (corr.) Mean

(m/s) 1949 (Wind speed) 1989

.:lin %

7.34

6.80 7.87

+15.7

6.78

6.31 7.25

+14.9

6.68

6.19 7.17

+15.8

6.99

6.50 7,48

+15.1

Esc (mm/d)! (estimated) Mean

1949 1989 .:lin %

4.57 4.36 4.78

+9.6

4.66 4.21 5.10

+21.1

3.02 2.83 3.21

+13.4

4.43 4.10 4.75

+15.8

1 E is a function of q,-q, and of corrected wind speed (with 50 % V,cal-trend).

125 N. F. 1. ] g. 1992, H. 2 H. Flohn et al.: Water vapour as an amplifier of the greenhouse effect

IS

10

70 so so I/O '0 20 10 o -10

o ~------------------~~~--~~-'r-~--------~~--~------~~ 0

- 10 10

-IS IS eo 70 &0 50 I/O 30 20 10 0 -10

-30 IS

10

-110 -SO -60 -70

10

00

10- J0

• I IS -110 - I 0

-160 -170 -180 no IS

10 10

r------------------------r 0

. 10 10

-15~~~~~~~_T~~~~~~~~~~~~~~~~~~~~~~~~~+_15 . I

-I S..:j..,..................,...,.............................."T'I.........,...................,....~~~'T"""~~ •••• I"""'" I'" 170 160 150 I~o 130 120 I 10 100 90 60

t I""" "I'''''' 'I 15

Fig. 1. Selected COADS 2 oX 2 ° boxes, lat. 10 0 5-14°N, (black Abb. 1. Ausgewahlte COAD5-2° X 2° Felderim Bereich 10°5­warmest oceans, dotted areas with part-time upwelling, all boxes 14 oN (schwarz warmste Ozeane, gepunktet = Gebiete mit =zonal average). temporarem Auftrieb von kiihlem Tiefenwasser, alle Felder =

zonales Mittel).

••

126

5.01

H. Flohn et al.; Water vapour as an amplifier of the greenhouse effect Meteorol. Zeitschrift

These data have now been prolonged to 1949-89, with about 4 X 106 observations covering nearly 20 X 106 km2

• The two variable terms the saturation deficit q,-qa (q = specific humidity in g H 20/kg air) and the scalar wind speed Vocal rose nearly everywhere (Table 3). The positive trend of q,-q. is generally not questioned. It is correlated with a weak, but quite general reduction of relative humidity (RH), which may be due to increasing instability of the marine boundary layer (Fig. 2). The apparent rise of V,cal is controversial; representative comparisons between V.cal and geostrophic winds are given in the Appendix. Together with some other results, these independent parameters suggest that the real trend of V fL.1 since 1949 can be assumed to be at least 50 % of the (biased) observed trend. The largest chan­ges are observed in the warmest oceans with T ,.>28°

28,0

27,8

~27,6 . . -_ ......-...!-----;------.• u . ----- .,.::.... 27.4 . -- ------ .. .

-----~!~-----.- , . .,.!' 27,2 ___------ • II • 27,0 .. 26,8+n~~~~~~"T~~~~MT~~~~~~~~~~Try~nT~

48 50 52 54 56 58 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90

5.4

5.2 ..cr; ., .., .· -----_....... -_.- ........... ; ....<5.0 . . ......:--'!.---••-;---.'";.- ... .C1> ~4.8 ....... ----­-----,...... --.-- ..o ...f 4.6 <;

4.4

4.2+nnT~~~~~~~~~~~~~~~~~~~~~~~ 48 50 52 54 56 58 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90

6,2 j 6,0 J

~5.8 55,6

)54 5,2

5,0~~TnTNTn~~~~~MT~~~Tn~~~~~nTnTnTnTnTnT~n+

48 50 52 54 56 58 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90

4.8 ;

~4'61·E 4.4 ~

:il 4,2 w

4.0,

3.8+·~~~~~~Tn~~~~~~~~~~Trr~TM~~~~n+ 48 50 52 54 56 58 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90

81.8

81,5

~81,2 ~ ~80.9 J: Il:: 80.6

80,,3 ... 80,0 +n-TTT"'.,.,.,..CTT"~n-rr.,.."..fTTTTM"1T"'..,."."..,."T'1""MTTTT'J.,.,.....I"""",.,.,.........T'f'T"I"TTTT'1TrrTT.,......f.

48 50 52 54 56 58 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90

(eastern Indian Ocean and western Pacific) (Fig. 3): here E increased at least by 13 % (V,eal assumed to be detrended), at most by 30 % (assuming the trend of Vseal to be real and unbiased). Since the most probable values of V",al tend to lie between these extremes, we estimate a value near 21 % for the warmest oceans. In the zonal average (10 °S-14ON), the same assumption leads do an increase of E by 16 % (± 8 %) or about 18 W1m2

, which is surprisingly large and equivalent to a risefrom about 142 to 163 cmla. Representing the zonal average lat. lOCS-14°N of our data, Fig. 2 gives time series and trends of all important parameters.

This value for E may serve as a base for a global estimate (FLOHN et al. 1990a, b). BAUMGARTNER & REICHEL's (1975) data indicate an area-weighted ratio of E between the global ocean and the belt 10°5-15 ON of 0.74 per unit

. . !

Fig. 2. Sea-air parameters at tropical oceans (Lat. 10 cS-14°N, time-series 1949-1989). T, = sea surface tempera­ture, q,-q, difference of specific hu­midity g between sea (s) and air (a), V,cat

= scalar wind speed (uncorrected), Eso evaporation calculated with 50 % of the linear trend of V>ca! (corrected), RH = relative humidity.

Abb. 2. Ozean-Atmosphare-Parameter fUr die tropischen Ozeane (10 0 S­14°N), Zeitreihen (1949-1989). T, Wassenemperatur, g.-g, Differenz der spezifischen Feuchte q zwischen Wasseroberflache (s) undLuft (a), V,ca! = skalare Windgeschwindigkeit (unkorri­gien), E50 Verdunstung berechnet mit 50 % des linearen Trends von V,cal (kor­rigien), RH relative Feuchte.

127 N.f~ 1.Jg. 1992, H.2 H. Flohn et al.: Water vapour as an amplifier of the greenhouse effect

mm/dav

° 0.2 0.4 0.. 0 .• 1.2

i"..""""""" 1 .•" Zonal average 1~.•~

24.1'"

" ',:'" , " , '<":',,"~,,-, 12.•'" Warmeal oceana 21.1'" '

J ao.o",.

~'lS••"" Upwelling ...,llon. 1;'1.4'l'o

:, 22.o",

~1.5'" Atlanllc ...'" .

, 1.,.".

o 2 4 • II 10 12 14 1. 1. 20 22 24 28 2. ao a2 a4

W/m'

Evaporation e.tlmated with V""ai - trend:

Fig. 3. Changes of evaporation (linear trend 1949­1989) at tropical oceans, assuming three different values of the observed (biased) V.cal-trend (0, 50, 100%) for different sections (d. Fig. 1) in mm/d, Watt1m2 and % of the 1949 value.

Abb. 3. Anderungen der Verdunstung (linearer Trend 1949-1989) tiber den tropischen Ozeanen unter der Annahme von drei unterschiedlichen Werten des beobachteten Vocal-Trends (0, 50, 100 %) flir ausgewahlte Gebiete (vgl. Abb. 1) in mm/Tag, Watt/m2 und in % der Werte von 1949.

of area. Converted into energy units, we reach a global estimate of + 13 (±3) W/m2

• This indicates an additional energy input into the atmosphere from the ocean, i.e. within the climate system, triggered by the warming of the ocean's surface as a feedback of the "dry" greenhouse effect, but released through a physically different process, i.e. phase changes. This internal feedback is much stronger (RAMA­NATHAN 1981) than the original ("dry") greenhouse effect (2 W/m 2

: IPCC Report 1990). It explains why tropospheric warming (see later) is distinctly stronger than surface warm­ing and why it increases towards the equator, as a mere consequence of the Clausius-Clapeyron equation.

HENSE et at (1988) have shown that at the 50/70 kPa layer over the equatorial belt, the vertically integrated H 20­content ("precipitable water": W) has increased since 1965. GAFFEN et at (1991) essentially confirmed these results in the Tropics and found a regionally different increase of water vapour of up to 50 kPa at many stations in mid-lati­tudes. On the other hand, LINDZEN (1990) made a strong point of a hypothetical drier upper troposphere, without giving evidence. Extending HENSE's data to the entire layer 30/85 kPa - the changes below 85 kPa are smaller and less striking - Fig. 4 gives a nearly 30 % rise of W during 22 years for four stations in Micronesia representing the "warmest oceans". This is accompanied by a substantial warming of the same layer (30/85 kPa) in the order of about 0.8 K in 22 years, i.e. distinctly more than the global surface average. This strong warming of the tropical troposphere is observed at all stations - even above the southern Indian stations after a change in instrumentation. Table 1 indicates that it does not extend into the higher tropical troposphere

due to its very low H 20 content - and that the entire tropical troposphere between 1.5 and 16 km is destabilized, parallel to an intensification of penetrating convection.

Within the period 1949-79 (FLOHN et al. 1990b), the trend of the small difference between sea and air tempera­ture (T,-Ta) was - outside of areas with upwelling cool water significantly positive, thus indicating together

~ 0% •

26

24

E 22E.

20

822

821

E to ~ Co .820!El

819

50% 0 100%

Jan-Dec 30-85kPa Micronesia (4 Stat.)

o " Prec. Water o o

o

o

1965 1970 1975 1980 1985

c

Thickness 0

0

0

0 0

0

0....

c 0

1965 1970 1975 1980 1985

o I gpdam f------1 o 0.33 K

Fig. 4. Changes of precipitable water and thickness (= virtual temperature, d. conversion lower left corner) of tbe layer 30/85 kPa (1965-1986),4 stations of Micronesia (lat. 7-10oN, long. 135­180 0 E).

Abb. 4. Anderungen des Wassergehaltes und der Scbichtdicke (= virtuelle Temperatur, vgl. die Umrechnungsskala links unten) fur die Schicht 30/85 kPa (1965-1986), 4 Stationen in Mikronesien (im Bereich 7ON-10 ON und 135 °E-180 °E).

128

0

y 0.000+0.408x

2.5 r = 0.762

2.0

1.5 . . ,. 1.0

0.5

-4 -3 -2 -1 0 1 2 3 4 5

i 0.0

E -0.5-.~«l -1.0

g ~

-1.5

'" uS'

H. Flohn et al.: Water vapour as an amplifier of the greenhouse effect Meteorol. Zeitschri/t

with decreasing relative humidity (RH) - that the upper mixing ocean layer wanns faster than the air and decreases the stability of the marine boundary layer. This is probably to be interpreted as a consequence of the ocean heat storage, with a time-scale in the order of 20-30 years. During the 1980's, however, this difference decreased in several areas.

Last not least, the relationship between evaporation and EI Nino events in the equatorial eastern Pacific deserves attention. During the cold part of the EI Niiio cycle, equa­torial upwelling transports cool water (at 14-20°C) with high content of nutrients (nitrogen, phosphorus) from the thermocline (at 100-400 m depth) to the surface. In the euphotic zone (above 25 m depth) the growth of planctonic algae ("blooming") is consequently enforced, and photo­synthesis needs much of the CO2 within the water; in some cases, the CO2-flux can even be directed from the air into the water. CO2-flux measurements from research ships and satellite measurements ofchlorophyll confirm this hypothe­sis (WEBER & FLOHN 1984, ELLIOTT et aI. 1991); these "cold events" had been accompanied by minima of the CO2

growth rate in the atmosphere. In "warm events" (El Nino) the nutrient-poor surface water (27-29°C) suppresses the cool water below and releases a maximum of CO2 into the atmosphere.

A statistical evaluation of sea-air interface parameters in the Galapagos area (lat. 2°S_2°N, long 86-104°W), along the shipping route from Panama to Australia (about

70N

&ON

SON

40N

30N

20N

ION

IDS

20S ·1.5 ·1 .0.5 0 0.5 0

T, (OC) T."T. (OC) Fig. 6. Linear trends of sea-air parameters (same as Fig. 2, except V5<al) along a meridional cross-section (mainly Long 50-20 oW, d. text) in the Atlantic (Lat. 68°N-20 °S). Eso is given as a thick line, Eo and E100 - d. Fig. 3 - thin lines. Note the overall decrease of RH (here not shown. d. Fig. 2) and the change of sign at 32 ON (T•• T.-Ta, q,-q., E).

Fig. 5. Correlation Ts versus Esc in the upwelling region of the East Pacific near the Galapagos, deviations from monthly averages 1947-1989.

Abb. 5. Korrelation zwischen T, und Eso in einem Kaltwasser-Auf­triebsgebiet des Ostpazifiks in der Nme der Galapagos Inseln. Abweichungen von langjahrigen (1947-1989) Monatsmitteln.

T.-AnomaJies (0C)

70N 10N

&ON &ON

SON SON

40N 40N

30N 30N

20N 20N

ION ION

0 0

lOS lOS

20S 20S0.5 .(),S 0 0.5 .0.5 0 0.5 1.5

q,"q. (g/kg) E(mm/day) Abb. 6. Lineare Trends der Ozean-Atmosphare-Parameter (wie in Abb. 2, mit Ausnahme v<eal) flir den Adantik; ein Meridianschnitt 68°N-200S (hauptsachlich flir den Bereich 50 oW-20oW, vgl. Text). ist durch dicke Linie, Eo und ESQ durch dlinne Linien wiedergegeben (vgl. Abb. 3). Beachte den generellen Rlickgang von RH (hier nicht dargestellt. vgl. Abb. 2) und die Anderung der Trend-Vorzeichen ab 32 ON (T" T,-Ta. q,-ga. E).

129 N.F. I.Jg. 1992, H.2 H. Flohn et al.: Water vapour as an amplifier of the greenhouse effect

0.87 X 106 km2), shows a significant correlation between T,

(as indicating the vertical component of upwelling/down­welling) and E, which varies between 1 and 5 mm/d. The correlation coefficient between monthly values of E (calcu­lated with 50 % of the V,cal trend) and T, (Fig. 5) is above 0.7, slightly higher than the correlation between CO2 growth rate and Ts (ELLIOTT & ANGELL 1987). Thus, the sea-air exchanges of HP and CO2 run parallel with El Nino, causing a significant drop of the flux of both gases into the air in cold events, while in warm events, a likewise signifi­cant peak is realized.

Preliminary trends of sea-air parameters are presented for a meridional cross-section of the Atlantic along the meri­dional strip at long. 50-20oW, extending from lat. 60 0 N to the equator. At the northern end (lat. 60-68°), the strip had to be shifted to 30.:-0 oW (around Iceland), also at the southern end in lat. 0-200S. Fig. 6 gives the 41 year trends for four parameters, comparable to Fig. 2; Vseal is omitted because it is strongly positive everywhere, with little varia­tion in latitude. In spite of a reversal of the trend of T, (and very small values of q,-qa) north of lat. 32°N, E50 shows - if Vscal trends are at least partially positive (cf. Chapter 4 and Appendix) - still an increase, which has also been found at the North Sea.

4. Changes of tropical and extratropical circulations

The seasonal migrations and anomalies of climatic patterns as observed at the surface cannot be understood without regard to the 3-dimensional, seasonally varying planetary circulation of the atmosphere. The latter consists of four interacting parts:

a) tropical Hadley circulation, with two low-level trade wind belts between the Intertropical Convergence Zone (ITCZ) and the belt of subtropical anticyclones at both hemispheres, developed as a pair of bands spiralling against each other with ascending motions along the ITCZ and subsidence in the subtropics,

b) extratropical Ferrel (or Rossby) circulation, with a broad band of meandering westerlies poleward of the sub­tropical anticyclones, extending aloft towards the equator, centered within the deep cold tropospheric vortices near both poles with low-level polar anticyclones. Here the (nearly horizontal) meridional exchange processes are main­ly produced by travelling or quasistationary baroclinic waves and eddies, .

c) the seasonally reverting monsoon circulations, extend­ing in the Tropics from about long. 150 °E across Indonesia, the Indian Ocean and Africa to long. 25 oW, together with the northern summer Tropical Easterly Jet and its weaker equivalent during southern summer, are both embedded between the two Hadley spirals,

d) superposed on the equatorial belt of a) and c) is the weak zonally directed Walker circulation, with its ascending motions in the tropical heating centers (Indonesia, S. Ame­rica) and subsidence at the coolest sections with oceanic upwelling.

Considering the vast section between Asia, Africa, and Australia, including the "maritime continent" Indonesia, the monsoonal circulation distinguishes itself from the Hadley circulation by the occurrence of seasonally warmer belts on its poleward flanks, away from the cooler (!) equa­torial region, thus causing low-level westerlies together with easterlies in the upper troposphere. In contrast herewith, the meandering high-tropospheric westerlies (Ferrel belt) penetrate in some quasistationary areas diagonally deep into the equatorial region, sometimes even merging from both hemispheres. Situated here are the preferred areas of tropical­extra tropical interaction, together with the cloudy surface cols between the subtropical anticyclonic cells.

This short synopsis has been given here, because several comprehending summaries published during the 1970's should be updated on the base of the available satellite observations. Recent observational series shed light onto the occurrence and mechanism of ENSO events and indi­cate changes in the intensity and/or extension of these circulations. Their investigation needs different observa­tional sources: satellite wind and cloud observations and precipitation data representing dominant vertical motions.

Due to the large gaps in the aerological stations, evidence for a recent intensification of the Hadley cell is only repre­sented by the drop ofrainfall in the continental belt 5-35oN, accompanied by the rise of salinity (S) i.e. the freshwater flux E-P into the atmosphere,- over most of the Atlantic between 12 and 35 ON (LEVITUS 1989). Rising precipitation in the ITCZ-belt is suggested by the apparent decrease of OLR from the top of cumulonimbus towers in the period 1974-88 (NITTA & YAMADA 1989, Fig. 5). These hints ­each alone scarcely significant - are coherent, but need confirmation by verification of the reality of the positive trend of COADS trade winds (see Appendix).

Based on U.S. diagnostic maps of the 70 kPa level, the deviations ofthe decadalaverage 1981-90, from'a reference period 1951~80, are given by HALPERT & ROPELEWSKI (1991, Figs. 41-45) for the year and the four seasons, covering the area 20oN-Pole. Of particular interest is the northern winter (DJF) with a marked intensification of cyclonic cells above both the northern Atlantic and Pacific, together with an intensification of anticyclonic cells center­ed at Gibraltar and the American West Coast at 55°N. Also, the Siberian High, near Lake Baikal, is strengthened. In all seasons, positive anomalies are significantly rising, while the negative anomalies are relatively small, covering only 20 % of the 10° lat.llong. intersections between 20 and 70 0 N.

After World War II, sea-level pressure and 50 kPa anal­yses became available in the United States in 1946 (KNOXet al. 1988, SHABBAR et al. 1990, LAMBERT 1990), and by the German Weather Service in 1949 (FLOHN et al. 1990a, b), the latter without the Pacific section south of lat. 55°N before 1967. Similar diagnostic series certainly exist also in other meteorological centers, such as Moscow and Tokyo, as well as shorter series for the Southern Hemisphere in Melbourne. In all cases, changes of the analysis technique (hand analysis versus objective methods) and of the data

130 H. Flohn et at.: Water vapour as an amplifier of the greenhouse effect Meteorol. Zeitschrijt

(new radiosonde stations, temperature and wind data from satellites) leading to inhomogeneities must be taken into account (LAMBERT 1990). Thus our evaluation was first limited to the period 1967-86, now extended until winter 1988/89, also including the levels 85 and 20 kPa.

During the winter half-year, the linear trend of sea-level pressure (Fig. 7) shows a highly significant deepening of both quasistationary cyclones at the northern Atlantic and even stronger in the northern Pacific. This intensification leads, at the forward sections, to strong advection of warm air along with rising pressure at the western coasts of Eu­rope and North America. This is related to the increasing frequency of droughts in the winter-rain areas of the west­ern Mediterranean and of California/Oregon. Using all daily maps (nearly 3,500), the relative frequency of pressure values below 990 hPa had been evaluated. Instead of the linear trend of these frequencies, Fig. 8 indicates the percen­tage of these values in 1966/67 and 1988/89, derived from this trend. SW ofIceland, this frequency rises from 18 to 27, at the Aleutians, even more, from 17 to 33. This suggests a comparable rise in storminess, which extends, at the Atlan­tic section, over a very large area from Labrador beyond Novaya Zemlya, covering also the southern coasts of the

North Sea and the Baltic Sea. The occurrence of particularly excessive gales during the following winter 1989/90, marks the peak of this trend; in each of the three winter months (DJF), a negative anomaly of 20-25 hPa occurred in the Icelandic area.

This is accompanied by a significant annual warming of the lower troposphere. reaching about 1 °C120a (zonal aver­age) in the Tropics and Subtropics up to lat. 40oN, then decreasing northward to values close to O°C near the pole (Fig. 9). This warming is distinctly higher above the conti­nents, while above parts of the high-latitude ocean rather small areas with cooling are observed. This leads to an intensification of zonal tropospheric winds and of baroclin­icity in mid-latitudes. The annual warming trend of the extratropicallower troposphere (lat. 20-90oN, except the surface layer 851100 kPa) reaches about 0.7 K (virtual) in 22 years (Fig. 10). This warming also depends, for a zonal belt lat. 45-70oN, on longitude, with highest values above the Rockies and the Ural Mts.

All these observed facts (FLoHN et al. 1990a, b) coincide ....~ to an unexpected but distinct intensification of the extra- ... tropical (Ferrel) circulation. Fig. 11 demonstrates, at 20 kPa, marked longitudinal differences of the recent evolution

Oct-Mar Surface Pressure Mean 1967-89

70 70

60 60

50 50

40 40

30 30

120 -60 o 60 Contou~ inte~val: 2 hPa

,JOct-Mar Surface Pressure Trend 1967-89

70 70

60 60

50 50

40 40

30 30

120 180 120 -60 o 60 Contou~ inte~lIal : 1 hPa Datted: Significonce ~ 90%

Fig. 7. Surface pressure (October-March), average and linear Abb. 7. Bodendruck (Oktober-Marz), Mittelwert und linearer trend for 1966/7-1988/9 in hPa. Note the significant pressure rise Trend in hPa ftir die Periode 1966/7-1988/9. Beachte den signifi­in the Azores High and above West-Central Europe. kanten Druckanstieg im Bereich des Azoren-Hochs und tiber

West- und Zentral-Europa.

60

131 N.F I.Jg. 1992, H. 2 H. Flohn et al.: Water vapour as an amplifier of the greenhouse effect

Nov-Mar Sea-level Pressure 1967 p~990hPa Rel.freq. (X}

80 80

70 70

80 80

50 50

40 40

30 30

80 120 180 -120 -60 0 60 Contour intervall

5 "

Nov-Mar Sea-level Pressure 1989 p~990hPa ReI. freq. (Xl

-., 80 80

70 70

60 60

50 50

40 40

30 30

80 120 180 -120 -80 0 Contour interlfal:

5 " Fig. 8. Relative frequency (%) of sea-level pressure below 990 hPa, for November-March 1966/7 and 1988/9, derived from linear trends (see text).

1.0

w.

30 40 50 SO 70 80 90 lot i tude

Fig. 9. Meridional section of the virtual temperature trend (thick­ness) ofthe 50/85 kPa layer 1%7-1988, Year (solid line). Novem­ber-April (dashed line) and May-October (dotted line); see also Fig. 10 (above).

Abb. 9. Meridionaler Schnitr fiir den Trend der virtue lien Tempe­ratur (Schichtdicke) der Schicht 50/85 kPa 1967-1988. Jahreswer­te (durchgezogene Linie) , November-April (gestrichelte Linie) und Mai-Oktober (gepunktete Linie); siehe auch Abb. 10 (oben).

Fig. 10. Above: time series of annual averages of virtual tempera­ture (thickness) of the 50/85 kPa layer, 1967-1988, hemispheric averages lar. 20 ON-Pole. Below: linear trend of the 501100 kPa layer averaged over the zonal belt lat. 45-70 ON as a function of longitude between 1967 (solid line) and 1988 (dotted line). Note the shift of the cold troughs towards the east.

Abb. 8. Relative Haufigkeit (%) cler Tage mit Bodendruck unter 990 hPa flir November-Marz 1966/7 und 1988/9, abgeleitet yom linearen Trend (siehe Text).

Jon-Dec 50/85kPc 180W-180E 20N -90N

u -5.5

c: .~

I­ -B.D

Q

o

"

1970 1975 19BD 1985

50/100kPo Temperature Trend Jon-Dec 1967-88 45N-70N

-8

-. W. -10

-12

-14

Abb. 10. Oben: Zeitreihe der Jahresmittelwerte der virtuellen Temperatur (Schichtdicke) der 50/85 kPa Schieht, 1%7-1988, hemispharisches Mittel flir die Zone 20 °N-Nordpol. Unten: Li­nearer Trend der relativen Topographie 50/100 kPa, gemittelt iiber den Bereich 45 °N-70 ON als Funktion der geographischen Lange zwischen 1967 (durchgezogene Linie) und 1988 (gepunk tete Linie). Beachte die Verlagerung der kalten Troge naeh Osten.

100-100 o Long! tude

80

132 H. Flohn et at.: Water vapour as an amplifier of the greenhouse effect Meteorol. Zeitschrift

Jan-Dec 20kPa Geostr.Wind Mean 1967-88

80 ~---~ aD- . -­-­70 70

60 60

50 50

40 40

30 30

60 120 lBO -120 -60 0 Contour interval: 2 m/s

Jon-Dec 20kPa Geostr.Wind Trend 1967-88

BO BO

70 70

60 60

50 50

40 40

3D 30

60 120 lBO -120 -60 o Contour interval: 1 m/s Dotted: Significance ~ 95%

Fig. 11. Average annual geostrophic wind speed at 20 kPa and its linear trend 1967-1988. Note the three centers above Japan, the Gulf Stream region, and the SE coasts of the Mediterranean. Note the southward displacement of the East Asian center, in contrast to the northward shift of the others, and the weakening of the Medi­terranean center.

averaged over the year. The strong cyclonic intensification over the Pacific leads to a southward shift and intensifica­tion of the East Asian Jet, which extends towards the east. The Gulf Stream Jet also extends towards ENE, while the weakening Mediterranean Jet shifts towards the north, caused by the anticyclonic development over western and southern Europe.

The 50 kPa level represents the middle troposphere be­tween 3 and 7 km altitude quite well. Fig. 12 shows the annual average geostrophic flow at this level, indicating two broad quasistationary troughs over eastern Siberia and east­ernmost Canada and the baroclinic center of the westerlies in latitudes 35-45 ON (Pacific) and 45-55 ON (Atlantic). The trend of annual flow is governed by a marked cyclone above the Aleutian Low and a weaker one centered SW of Iceland. Anticyclonic flow trends are indicated above the Alps and in the Canadian Rockies. Fig. 13 demonstrates the seasonal differences in this trend of tropospheric flow: during summer, marked anticyclonic eddies above the west­ern Mediterranean, the mid-western U.S., the eastern Pacific

Abb. 11. Jahresmittelwert und linearer Trend der Geschwindigkeit des resultierenden geostrophischen Windes im 20 kPa Niveau 1967-1988. Beachte die drei Zentren tiber Japan, der Golfstrom-Region und der SE-Ktiste des Mittelmeers. Beachte auch die Stid-Verlagerung des ostasiatischen Zentrums im Gegensatz zu der Nord-Verlagerung der anderen Zentren und die Abschwachung des mediterranen Zentrurns.

(!), northwestern Siberia and - remarkable strong - above the Himalayas. Cyclonic eddies are developed above Scan­dinavia, the Chukchi Peninsula, (weakly) above Central Asia, and (somewhat uncertain) above Morocco. During winter, giant cyclonic cells centered south of Greenland and the Aleutians are superposed to the time-averaged flow, togeth­er with anticyclonic eddies near Lake Baikal, the Canadian Rockies, and west of Gibraltar. The intensification of the tropospheric circulation above the oceans is distinctly stronger in winter than in summer probably correlated with the seasonal variation of mid-latitude evaporation, in­dicating a strong feedback between hydrological cycle and cyclogenesis. This intensification is also stronger above the Pacific than above the Atlantic perhaps as a response to

the most powerful H 20 source at the warmest waters of the western Pacific.

Data for the monsoon circulation are hardly sufficiently reliable for a trend analysis. Unfortunately, some aerological series above tropical Africa and India show serious inho­mogeneities and errors within several parameters, even

133 N.F. I.Jg. 1992, H. 2 H. Flohn et al.: Water vapour as an amplifier of the greenhouse effect

Jan-Dec 50kPa Geastr.Wind Mean 1967-88

80 80

70 70

60 60

50

40 40

30 30

60 120 180 -120 -60 o 60

-+ 20 m/s

Jan-Dec 50kPa Geostr.Wind Trend 1967-88

80 i •

'" ... -t I \ , , - ,/

.//I . ~

\ \ .

'" ."

.~. I ,,--- I' .... . ............ .. ~+- .... . \; /"- ... ~ ....... -. .... ... ,

... .... ...... ...

'" ... ~ 80

-..., 70 70

80 60

50 50

4040

3030

80 120 180 -120 -60 0

-+ Ii mIl

Fig. 12. Geostrophic flow pattern at the 50 kPa level, annual mean (above) and linear trend (1967-1988, below). Note the different scale of arrows.

upper winds, which prejudice a numerically trustworthy month-to-month analysis. Assuming that their bias remains constant during the year, the trend of surface winds varies seasonally; available data suggest a relative weakening dur­ing summer and an intensification during winter.

lbe main branch of the Walker circulation extends be­tween the strongest heating center (Indonesia and West Pacific) and the upwelling region along the west coast of South America and the Galapagos. Here, as well as along other coastal regions with seasonal upwelling of cold water, a slight drop of sea surface temperature (until 1979) and the trends of other parameters at the sea-air interface (d. Table 3) indicate a general increase of upwelling (BAKUN 1990), which, at the same time, is a convincing argument for higher wind speed. Together with the greatest warming of the western Pacific, this indicates a strengthening of the Walker circulation.

A 3D-evaluation of the monthly averaged maps also allows an estimate of the time variations of meridional temperature gradients (in K/IOOO km) and vertical lapse rates (in K/km) for the hemispheric 50/100 kPa layer (lat. 20 oN-Pole). The results do not vary substantially with season. They indicate - over 66 % of the total hemisphere

Abb. 12. Vektor des mittleren geostrophischen Windes im 50 kPa Niveau, Jahresmittelwett (oben) und linearer Trend (1967-1988), (unten). Beachte unterschiedliche Skalen der pfeile.

- an increase of the meridional gradient and of the lapse rate each by about 4 % in 22 years. These values may serve as quantitative measures for the intensification of the 3D-cir­culation of the Nonhern Hemisphere.

Since our diagnostic data are available for the entire hemisphere since 196617, we omitted the analyses for the whole polar region, for the A tlantic and adjacent continents since 1949, in this chapter. These series indicate a warmer period in the 1950's and an abrupt change around 1962 (d. KNOX et al. 1988) to a cooler mode. Exactly at this time. data series from East Africa (FLOH:-'; 1987) and the Indian Ocean (REYERDI:-'; et al. 1986). along with subsurface ocean data from the Labrador and Greenland Sea, suggested a much larger extension of this event, and shall be investi­gated at a later date.

5. Conclusions

Some of the essential results of our purely empirical investi­gation of the recent climatic evolution can be summarized as follows:

60

134

60

H. Flohn et al.: Water vapour as an amplifier of the greenhouse effect Meteorol. Zei tschrift

Nov-Mar 50kPa Geastr.Wind Trend 1967-89

80 BO

70 70

......... 60 , I , , ." , • 60

, , \ I I50 50

40 40 I ...

... , ' , .......... , ..... ",.,.---_....... -... ­30 30

60 120 180 -'20 -60 o 60

-5 m/s

May-Sap 50kPa Gaostr.Wind Trend 1967-88 - , , ~ ..

\

12Q 180 -120 -60 0 60

-+ 5 II\/s

Fig. 13. Linear trends of the geostrophic flow patterns at the 50 kPa-level, for two seasons: (above) November-March 1966/7­1988/9, (below) May-September 1967-1988. Note the shift of the cyclonic trend patterns from winter towards summer above the Pacific (towards NNW) and the Atlantic (towards ENE); see text also.

1) In tropical oceans, a distinct increase ofT,since 1949 is in the order of 0.01 Kia. Assuming a mixed layer of 100 m depth, this warming needs only 0.13 W1m2, thus about 5 % of the"dry" greenhouse effect of CO2 and other trace gases.

2) Due to the heat storage capacity of the ocean, this small warming leads, 20-30 years later, to a substantial increase of evaporation of up to 15 % and of the H 20 content of the global atmosphere. It releases latent heat via clouds and precipitation with ascending motion and conse­quently causes increasing tropospheric cyclonic circulation and winds, notably above the oceans.

3) In middle and higher latitudes, the real increase of surface wind speed - in the order of 8-10 % at the ocean surface leads to cooling by evaporation; there, this process seems to counteract the radiational warming by the greenhouse effect and to maintain, to some extent, the cold polar vortex with its particular circulation.

4) Increasing cyclonic activity at mid-latitude oceans leads, at the forward flank of the cyclones above the western coasts of continents, to the formation of anticyclonic ridges

80

70 .J

60

50

40

30

Abb. 13. Lineare Trends des geostrophischen Windvektors im 50 kPa Niveau fur die extremen Jahreszeiten: (oben) November­Marz 1966/7-1988/9, (unten) Mai-September 1967-1988. Be­achte die Anderung der zyklonalen Trendmuster zwischen Winter und Sommer; uber dem Pazifik Verlagerung nach NNW und uber dem Atlantik nach ENE (siehe Text).

with warm air advection and subsiding motions, i.e. more frequent droughts, notably during the cool season.

5) The intensification of the hydrologic cycle injects, after a delay of some decades, an additional energy in the order of 12-15 W/m2 from the ocean to the atmosphere. This natural (internal) process is due to phase changes of water vapour, but triggered as amplifying feedback by the (radiational) anthropogenic greenhouse effect.

6) This evolution explains why the warming of the trop­osphere is larger at the level of low and middle clouds than at the surface and is higher in the Tropics than in the polar latitudes.

These changes - disregarding here all regional and local pecularities as well as the seasonal and interannual variabili­ty of the trends - are mainly produced by water vapour as an amplifier of the "dry" greenhouse effect of CO2, CH4!

N 20 and the CFM's. This certainly explains much of the anomalies which apparently contradict the greenhouse ef­fect on global warming. Indeed, the (positive) feedback mechanisms operating via water vapour, douds, and precip­

135 lV.F. J.J~ 1992, fl. 2 H. Flohn et al.: Water vapour as an amplifier of the greenhouse effect

itation are as indicated by HA:\SE:\, SCHLESINGER, and others (see IPCC-Report, Section 3) - much more effec­tive than the snow-ice-albedo feedback.

This explains the dominance of warming in the Tropics over the polar latitudes - in addition to the simple geo­graphical fact that the tropical oceans alone, between lat. 20 oN and 20 aS, cover nearly 133 X 106 km2, in contrast to the area of both polar caps between lat. 70 0 and the Pole with only 31 X 106 km2 together.

6. Comments (H.F.)

It is difficult to understand why this dominant feedback by water vapour and its phase changes has been nearly com­pletely neglected (a Case of infectious amnesia?), in spite of the convincing estimates by F. MOLLER (1963) and V. RA­MANATHAN (1981). There is nothing new or unexpected about this role one Can find it in most textbooks and monographs which line the bookshelves of every physically­oriented climatologist.

On the other hand, we should withstand the temptation to simply extrapolate the present evolution - this is cer­tainly no adequate solution. If the warming of the oceans and the intensificaton of their wind-driven circulation pro­ceeds - as one should expect for the next decades - the probability of a rapid retreat of the Arctic sea-ice should rise, even to its total (at least seasonal) disappearance. This may ultimately lead us, in the near future, into an amazing pattern of unipolar glaciation which controlled the earth's climate during the late Miocene/early Pliocene, between about 14 and 3.3 million years ago (FLOHN 1979, BUDYKO 1991: see IPCC-Report p. 203). "What has happened, may happen again"; for further discussion see RUDDIMAN & KUTZBACH (1989).

In this paper, the fascinating but at least regionally con­flicting conclusions from many climate models have hardly been mentioned - we may refer instead to the IPCC-Report (Sections 1-6). However, it should also be noted with satisfaction that at present, the most recent results with a coupled, interaCtive ocean-atmosphere model- developed at the Max-Planck Institute in Hamburg show, in its first decades, improved correspondence with the evolution of conventional observations. This growing coherency be­tween facts and models is most welcome (with a sigh of relief!) - its final appearance may be based mainly on a more realistic parameterization of sub-grid-scale effects, such as penetrating tropical convection and the broken pattern of polar sea-ice with its leads and polynyas.

In contrast to such a growing consensus, a working group planning to update the (mostly highly valuable) IPCC-Re­port had only found "intentionally limited" space for ob­servations. Are we still living in the scholastic era of the 13th century?

Regarding the reluctance of responsible decision-makers to accept - lacking convincing facts - the conflicting results of the majority of climate models as a base for expensive (and unpopular) preventive measures, there is no

reason for the community of model-designers to stay in their illuminated ivory tower. Indeed, more research is needed for improved, rigidly verified (!) models as well as for critical and complete (!) evaluations of all data sources, conventional or with sophisticated technology.

But, do we really have sufficient time for this research? The answer should be: No with the inclusion of water vapour as an overwhelmingly powerful natural agent, avail­able everywhere, processes leading to drastic and rather rapid climatic changes are already in full swing. They will ultimately lead to all kinds of extreme weather events beyond our imagination. We have to live with them while trying our best to decelerate their progress and to mitigate their impact, notably on the precarious water budget. Strategies towards that goal have been elaborated, but im­mense common efforts are desperately needed.

Acknowledgments

This research was supported (since 1984) by the Rheno-Westphal­ian Academy of Science, as a part of the long-tenn research pro­gram of the German Academies. The help of Mrs. W. RUBIN at the text computer is gratefully acknowledged.

Appendix: How realistic is the global rise of wind speed?

One of the basic questions in our research was the reality and magnitude of the rising wind speed as suggested by the maritime COADS data. According to RAMAGE (1987), this rise could be an artifact caused by the slow transition from subjective estimates based on the state of seaWaves to meas­urements. Since this paper, many local contributions with conflicting results have been presented, which cannot, at this time, be discussed nor quoted in detail. In this context, however, some comparisons between observed and geo­strophic winds (the latter based on pressure data from the COADS series) shall be given, along with some area-aver­aged estimates based on the daily hemispheric synoptic maps from the German Weather Service (see Chapter 4).

The most promising area for a representative verification is the tropical Atlantic, where the available data are much more plentiful than at other tropical oceans. We selected the belt lat. 14-26 oN, long. 20-50 oW, since the center of the trade wind belt is found at this latitude during the entire year. The data coverage is optimal- the number of 2 a X 2 a

boxes with less than 5 observations per month is marginal. Since the average resultant wind blows from 67 0 (ENE), we can limit the comparison to the zonal component (U). Fig. 14 shows the annual values (1949-1989) for the observed (above) and geostrophic U-component; the apparently downward linear trend represents in fact rising values due to the negative sign of the easterlies. Since the observed COADS-values are uncorrected - using an obsolete con­version table (ISEMER & HASSE 1987) - an (absolute) correction of +2 mls or at least of +1.4 m/s (FLOHN et aL 1990a) should be applied. The real observed easterlies should increase from about 6 to 7 mIs, thus reaching about 85 % of the geostrophic easterly component.

136 H. Flohn et al.: Water vapour as an amplifier of the greenhouse effect Meteorol. Zeitschrift

m/&-3 -3

-4

-6

* * *

* * *

1<

* * *• * III

* *

** * * *' *

, -4

* *.

-5

* -6 III III

-6

-7 II>

" til

III

III

" II>

" III III til

II>

II>

III " til

1lI,1I> .. III

II> -7

-8 -8

-9 -9 1948 1953 1958 1963 1968 1973 1978 1983 1988

• U-obs. "U-geostr.

Fig. 14. Time-series and trends of the annual U-component of observed surface winds (uncorrected, above) and geostrophic winds (below) at the Atlantic area (lat. 14-26 oN, 50-20 OW) based on COADS data 1949-1989. E-winds negative!

Abb. 14. Jahresmittelwerte und Trends der U-Komponente des beobachteten Bodenwindes (unkorrigiert, oben) und des geostro­phischen Windes (unten), ftir die Atlantik Region 14°N-26°N und 50 °W-20 oW, basierend auf den COADS-Daten 1949­1989. E-Winde negativ!

The linear trend of the observed V -components is -0.773 mls in 41 years; that of the geostrophic V-compo­nent (-0.778 m/s) is practically identical. The correlation between annual values and linear trend is -0.59 for the observed winds, -0.37 for the geostrophic winds (with a greater variance), significant at the 99.9 % resp. 95 % level. A comparison of the individual years indicates a coincidence of very high (1973) and very low (1968) values. This result suggests that our assumption that only 50 % of the trend is real is too conservative.

From the monthly resultant geostrophic winds, which can be calculated from German weather maps for each 5 0

(lat). X 100 (long.) box, one can derive monthly and annual weighted hemispheric averages for the latitudes 20-90 oN. Time series have been prepared, together with a linear trend, for the 22a -period 1967-1988, for sea-level pressure, 85, 50, and 20 kPa levels. As an example, Fig. 15 shows data from the 50 kPa level for the year (above) and the extreme seasons (middle and bottom); the significance of the trends are above 95 %. Averages and trends increase sharply from summer to winter. Table A1 gives averages and trends; they indicate that in the troposphere the linear trends are nearly equal or higher than those at the surface.

This apparent rise of tropospheric geostrophic winds coincides well with the similarly derived value of the hori­zontal gradient of the virtual temperature of the 50/100 kPa layer. Here the average gradient of 5.3 KIlOOO km increases during the 22a-period by +0.24 °C or 4.5 %.

Jon-Dec 50kPo 180W-180E 20N -90N 11.6

11.4til "­E 11.2 c .~

0 11.0

> 10.S

c 0

C ~,.fJ

a 0

c a"

1970 1975 1980 1985 1990

Nov-Mar 50kPo 180W-180E ZON -SON

a a

c a

" ...... c a .... D

1:1 D D

til 14.5 "­E c

>'" 14.0

1970 1975 19BO 1985 1990

Moy-Sep 50kPo 180W-180E ZON -SON D

D

c c

a "

8.2 ill "­E 8.0 c.... .. 7.S >

7.6

1970 1975 1960 1985 1990

Fig. 15. Resultant geostrophic winds based on 50 kPa maps of the German Weather Service. averaged for the Northern Hemisphere (20-90 ON); 1967-88. Annual values (above). winter and summer. Note the different scales.

Abb. 15. Resultate des geostrophischen Windes basierend auf den 50 kPa Analysen-Daten des Deutschen Wetterdienstes. gemittelt tiber die Nord-Hemisphare (200N-900N)j 1967-1988. Jahres­mittelwerte (oben), Winter und Sommer. Beachte die unterschied­lichen Skalen.

Table A 1.1 ncrease ofthe mean geostrophic resultant wind for the area 20 oN-Pole between 1967 and 1988.

Tabelle AI. Verstiirkung deT Resultante des geostrophischen Windes im Bereich 20 0 N -N ordpol zwischen 1967 und 1988.

Mean Trend/22a

20kPa 20.2m/s +0.37 mls - +1.8 % 50kPa 11.2 mls +0.70 mls - +6.2 % 85kPa 4.5 mls +0.41 mls - +9.0 % Sea level 4.5 mls +0.41 mls - +9.0 %

137 N.£. 1. Jg. 1992, H.2 H. Flohn et al.: Water vapour as an amplifier of the greenhouse effect

References

Angell, J.K. (1990): Variation in United States cloudiness and sunshine duration between 1950 and the drought year 1988. - J. Climate 3, 296-308.

Arrhenius, Sv. (1896): On the influence of carbonic acid upon temperature at the ground. Phil. Mag. 41,237-276.

Bacon, S. & Carter, D. T. (1991): Wave climate change in the North Atlantic and North Sea. - Internat. J. Climatol. 11, 545-558.

Bakun, A. (1990): Global climate change and intensification of coastal ocean upwelling. Science 247, 198-201.

Baumgartner, A. & Reichel, E. (1975): The World Weather Balance - Mean Annual Global, Continental and Maritime Precipi­tation, Evaporation and Runoff. - R. Oldenbourg, Munich, Vienna, 179 pp.

Bradley, R. S., Diaz, A. H, Eischeid, H. E, Jones, P. D., Kelly, P. M. & Goodness, C. M. (1987): Precipitation fluctuations over Northern Hemisphere land area since the mid-19th. century. - Science 237, 171-175.

Budyko, M. L (1978): The heat balance of the Earth. In: Grib­bin, J. Ed.), Climatic change, 85-113. Cambridge Univer­sity Press.

(1991): Climate prediction based on past and current analogues. In: Jager, J., Ferguson, H. L. (Eds.): Climate change:

Science, impacts and policy. Proceedings of the Second World Climate Conference. - Cambridge University Press, 225-229, d. also IPCC, Chapter 7.2.2.

Deutscher Bundestag (Ed.) (1990): Schutz der Erde. Eine Bestands­aufnahme mit Vorschlagen zu einer neuen Energiepolitik. Band I: 686 pp., Bonn.

Diaz, H. E, Bradley, R. S. & Eischeid, J. K. (1989): Precipitation fluctuations over global land areas since the late 1800's. - J. Geophys. Res. 94, 1195-1210.

Elliott, W. P. & Angell, J. K. (1987): On the relation between atmospheric CO, and equatorial sea-surface temperature.

Tellus 39B, 171-183. Elliott, W. P., Angell,J. K. & Thoning, K. W. (1991): Relation of

atmospheric CO2 to tropical sea and air temperatures and precipitation. Tellus 43B, 144-155.

Elsaesser, H. W., McCracken, M. C., Welton, J. J. & Grotch, S. L. (1986): Global climatic trends as revealed by recorded data. - Rev. Geophys. 24, 745-792.

Elsaesser, H. W. (1990): A different view of the climatic effect of CO2 - Updated. Atmosfera (Mexico) 3,3-29.

Hohn, H. (1979): Notre avenir climatique: Un Ocean Arctique libre de glaces? - La Meteorologie VI, Ser. 16, 35-51.

- (1987): East African rains of 1961/62 and the abrupt change of the White Nile discharge. Palaeoecology of Africa 18, 3-18.

Flohn, H. & Kapala, A. (1989): Changes of tropical sea-air interac­tion processes over a 30-year-period. - Nature 338, 244­246.

Flohn, H., Kapala, A., Knoche, H.-R. & Machel, H. (1990a): Recent changes of tropical water and energy budget and of midlatitude circulations. - Climate Dynamics 4, 237-252.

- - (1990b): Changes of hydrological cycle and North­ern Hemisphere circulation and their initation at the tropical air-sea-surface. - In: Internat. TOGA Scient. Conf. Pro­ceedings, Honolulu 1990, World Meteor. Org. WCRP 43 (WMO/TD 379): 47-57; also in: Houghton, ]. T. et al. (Eds.): Observed Climate Variations and Change: Contribu­tions in Support of Section 7 of the 1990 IPCC Scientific Assessment, Chapter VI: 1-11.

Gaffen, D. J., Barnett, T. B. & Elliott, W. P. (1991): Space and time scales of global tropospheric moisture. - J. Climate 4, 989­1008.

Grassl, H. (1991 ): Remote sensing and global climate change: water cycle and energy budget. - In: Jager, ]. & Ferguson, H. L. (Eds.): Climate change: Science, impacts and policy. Proceed­

ings of the Second World Climate Conference. - Cambridge UniversityPress,187-194.

Gregory, S. (1989): Macro-regional definition and characteristics of Indian Summer monsoon rainfall 1871-1985. - Internat. J. Climatol. 9,465-483.

Halpert, M. S. & Ropelewski, C. F. (Eds.) (1991): Climate Assess­ment. A Decadal Review 1981-1990. - U.S. Dept. Com­merce, NOAA, Climate Analysis Center VII, 109 pp.

Henderson-Sellers, A. (1986): Increasing cloud in a warmer world. Climatic Change 9, 267-309.

- (1989): North American total cloud amount variations in this century. - Global and Planetary Change 1,175-194.

Henning, D. (1989): Atlas of the Surface Heat Balance of the Continents. Components and Parameters Estimated from Climatological Data. Gebriider Borntraeger, Berlin, Stutt ­gart, 402 pp.

Hense, A., Krahe, P. & Flohn, H. (1988): Recent fluctuations of tropospheric temperature and water vapour content in the Tropics. - J. Meteorol. Atmos. Phys. 38,215-227.

Houghton, J. T., Jenkins, G.]. & Ephraums, J.J. (Eds.) (1990): !PCC, Climate Change. The IPCC Scientific Assessment. Cambridge University Press, 365 pp.

Isemer, H. J. & Hasse, L. (1987): The Bunker Climatic Atlas ofthe North Atlantic Ocean. Air-Sea Interactions. - Vol. 2: Springer, Berlin, 252 pp.

(1991): The scientific Beaufort equivalent scale: Effects on wind statistics and climatological air-sea flux estimates in the North Atlantic Ocean. - J. Climate 4, 819-836.

Knox, J. L., Higuchi, K., Shabbar, A. & Sargent, N. E. (1988): Secular variation of Northern Hemisphere 50 kPa geopoten­tial height. - J. Climate 1,500-511.

Lambert, St. J. (1990): Discontinuities in the long-term Northern Hemisphere 500-millibar heights dataset. - J. Climate 3, 1479-1484.

Levitus, S. (1989): Interpentadal variability of salinity in the upper 150 m of the North Atlantic Ocean, 1970-1974 versus 1955-1959. - J. Geophys. Res. 94, 9679-9685.

Lindzen, R. (1990): Some coolness concerning global warming. Bull. Amer. Meteor. Soc. 71, 288-299.

Moller, E (1963): On the influence of changes in the CO2 concen­tration in air on the radiation balance of the earth's surface and on the climate. - J. Geophys. Res. 68,3877-3886.

Nitta, T. S. & Yamada, S. H. (1989): Recent warming of tropical sea surface temperature and its relationship to the Northern Hemisphere circulation. - J. Meteor. Soc. Japan 67,375-383.

Peixoto,]. P. & Oort, A. H. (1983): The atmospheric branch ofthe hydrological cycle and climate. - In: Street-Perrott, A., Beran, M., Ratcliffe, R. (Eds.); Variations in the Global Wa­ter Budget. - D. Reidel, Dordrecht, Boston, Lancaster, 5-65.

Ramage, C. S. (1987): Secular change in reported surface wind speeds over the ocean. - J. Clim. Appl. Meteor. 26, 525­528.

Ramanathan, V. (1981): The role of ocean-atmosphere interactions in the CO 2 climate problem. -]. Atmos. Sci. 38,918-930. (1989): Cloud-radiative forcing and climate: Results from the Earth Radiation Budget Experiment. - Science 243, 57­63.

Raval, A. & Ramanathan, V. (1989): Observational determination of the greenhouse effect. - Nature 342, 758-761.

Reverdin. G. et al. (1986): Interannual displacement of convection and surface circulation over equatorial Indian Ocean. ­Quart. J. Roy. Meteor. Soc. 112,43-67.

Ruddiman, W. E & Kutzbach, J. E. (1989): Forcing of the Late Cenozoic Northern Hemisphere climate by plateau uplift in Southern Asia and in the American West. - J. Geophys. Res. 94,18,409-18,427.

Sathyendranath, Sh. et al. (1991): Biological control of surface temperature in the Arabian Sea. - Nature 349, 54-56.

138 Referate Meteorol. Zeitschrift

Scherhag, R. (1936a): Eine bemerkenswerte Klimalinderung iiber Nordeuropa. - Ann. Hydrographie 64, 153-160.

(1936b): Die Zunahmeder atmospharischen Zirkulation in den letzten 25 Jahren. - Ann. Hydrographic 64, 397-407.

(1939): Die gegenwartige Milderung der Winter und ihre Ur­sachen. - Ann. Hydrographie 67,292-303.

Schonwiese, C.-D. & Strahler, U. (1991): Multiforced statistical assessments of greenhouse-gas-induced surface air tempera­ture change 1890-1985. - Climate Dynamics 6,21-33.

Schonwiese, C.-D. & Birrong, W. (1990): European precipitation trend statistics 1851-1980 including multivariate assess­ments of the anthropogenic CO 2 signal. - Z. f. Meteor. 40, 92-98.

Shabbar, A., Higuchi, K. & Knox, J. L. (1990): Regional analysis of Northern Hemisphere 50 kPa geopotential heights from 1946 to 1985. - J. Climate 3, 543-557.

Weber, K. H. & Flohn, H. (1984): Oceanic upwelling and air-sea exchange of carbon dioxide and water vapour as a key for

Referate

Bartels, H., Albrecht, F., Guttenberger, J.: Starkniederschlags­hohen fiir die Bundesrepublik Deutschland. - Selbstverlag Deut­scher Wetterdienst, Offenbach am Main, 1990. 13 S., 40 Karten und zahlreiche TaheIlen, DIN A3. ISBN 3-88148-255-5. Teil 1: NiederschHige Iangerer Dauerstufen (D>24 h). TeiJ 2: Niederschla­ge kiirzerer Dauerstufen (D<24 h) Sommer. Zeitraum 1951­1980. Nur zusammen lieferbar, DM 349,-.

Als ein Ergebnis des Projektes KOSTRA wird mit der Untersu­chung iiber Starkniederschlagshohen ein zwischen Meteorologen und Hydrologen abgestimmtes einheitliches Niederschlagsregel­werk vorgelegt.

Das Karten- und Tahellenwerk umfaBt zunachst in zwei TeiJen die regionalisierten Starkniederschlage langerer Dauerstufen (D ;c 24 h) und die Niederschtige kiirzerer Dauerstufen Sommer.

In einem drittem Teil, der die Niederschlage kiirzerer Dauerstu­fen (D < 24 h) - Winter und Jahr erfaBt, wurden Kartendarstel­lungen und Tabellen von Starkniederschlagshohen fur die Win­termonate sowie fiir das Jahr mit 18 Dauerstufen und 8 Wieder­kehrzeiten veroffentlicht.

Der Teil 1 hehandelt die Flachenverteilung der Niederschlags­hohen der Langzeitniederschlage von 24, 48 und 72 Stunden fiir Wiederhohlungszeitspannen von 1, 10 und 100 Jahren der Som­mermonate (Mai bis September), der Wintermonate (Oktober bis April) und dem Jahr Uanuar bis Dezember). Erganzt werden diese Karten durch ein Tabellenwerk mit Einzelauswertungen von 1065 ausgewahlten Stationen, sowie einem Stationslexikon, einem er­lauternden Text und Intetpretationshilfen.

Der Teil2 enthalt fiir die Sommermonate die Flachenverteilung der Niederschlagshohen fiir die Wiederkehrzeiten 1, 10 und 100 Jahre der Dauerstufe 12 Monate sowie fiir die Wiederkehrzeiten 1, 2,5,10 und 100 Jahre die Dauerstufe 15 und 60 Minuten. Erganzt werden die Karten durch Einzelauswertung von 125 Stationen.

Das iibersichtlich aufgebaute Werk erfordert nur eine kurze Einarbeitungszeit fiir den praktischen Gebrauch. Es ermoglicht miihelos die Berechnung der Starkniederschlagshohen bei vorge­gebener Dauer und Wiederkehrzeit oder die Bestimmung der Wie­derkehrzeit, wenn die beiden anderen GroBen vorgegeben sind. Es . ist fUr viele Bemessungsaufgaben in der Wasserwirtschaft eine un­ersetzbare Grundlage. Das Kartenwerk ist eine gelungene Neube­

large-scale climatic change? - Bonner Meteor. Abhandl. 31, 73-99.

Weber, G.-R. (1990): Spatial and temporal variation of sunshine in the Federal Republic of Germany. Theor. App!. Climato!' 41,1-9.

Prof. Dr. H. FLOH:-': Dr. Alice KAPALA Dr. R. KNOCHE DipL-Met. H. MACHEL Meteorologisches lnstitut der Universitat Bonn Auf dem Hiigel20 D-53oo Bonn 1

Accepted: 29.10.1991

arheitung und Erweiterung der 1940 von Reinhold herausgegebenen Publikation Uber Starkregen in Deutschland. Da die Karten nur die alten Bundesliinder umfassen, ware eine Erweiterung auf die neuen Bundeslander dringend erwiinscht. U. Maniak, Braunschweig

Gotz, G., Meszaros, E. & Vali, G.: Atmospheric Particles and Nuclei. Akademiai Kiad6 Budapest, Hungary, 1991. 274 pages, DM57,-.

Die Autoren beschreiben als Ziel ihres Buches zum einen die DarsteIJung der physikalischen und chemischen Eigenschaften der atmospharischen Aerosolteilchen und -kerne und zum anderen in einer koharenten Form ilire Rolle bei der Bildung von Wolken und Niederschlag sowie heim Klima. In einem klein und anspre­chend gehaltenen Buch sollen Studenten und Anfanger angespro­chen, aber auch aktive Wissenschaftler durch die Literaturiiber­sicht interessiert werden. In vier Kapiteln werden die folgenden Themenkreise angesprochen: Das atmospharische Aerosol, Wol­kenkondensationskerne, Eiskeime, Aerosol und Klima. In Anhan­gen wird naher eingegangen auf: Grundlegende Konzepte der Kembildung, Wachstum von Wolkentropfen, Strahlungstransfer, Terminologie. Die einzelnen Kapitel sind von verschiedenen Auto­ren geschrieben.

Die Zusammenstellung der behandelten Themen war dringend notwendig. Wurden sic doch in der Vergangenheit versteckt in anderen Biichem behandelt (Twomey, 1977: Atmospheric Aero­sols; Pruppacher and Klett, 1978: Microphysics of Clouds and Precipitation; Warneck, 1988: Chemistry of the Natural Atmos­phere). Dadurch wird nun ein recht breiter Oberblick moglich und Anfanger solI ten prinzipiell die Chance eines Oberblickes erhalten. Die Autoren sind bestens in den jeweiligen Arbeitsgebie­ten ausgewiesen und prasentieren die Informationen in einer soli­den, ausgewogenen und padagogischen Weise. Die Verwendung von Gleichungen unterstreicht den physikalischen und chemischen Verstandnisauftrag und verliert sich nicht in der unsinnigen An­haufung von empirischen Gleichungen, deren Lebenszeit nur bis zur nachsten Veroffendichung reicht. Hervorzuheben sind die