CORRESPONDING AUTHOR AUTHOR QUERIES

Journal … Isotopes in Environmental and Health Studies

Article ID GIEH 288889

TO: CORRESPONDING AUTHOR AUTHOR QUERIES - TO BE ANSWERED BY THE AUTHOR Dear Author Please address all the numbered queries on this page which are clearly identified on the proof for your convenience. Thank you for your cooperation

Q1

Ref. [20] is cited in the text but not provided in the list. Please provide complete reference details or delete the Production Editorial Department, Taylor & Francis 4 Park Square, Milton Park, Abingdon OX14 4RN Telephone: +44 (0) 1235 828600 Facsimile: +44 (0) 1235 829000

Isotopes in Environmental and Health Studies Vol. 44, No. 1, March 2008, 1–10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

Deuterium excess in precipitation of Alpine regions – moisture recycling KLAUS FROEHLICH*†, MARTIN KRALIK‡, WOLFGANG PAPESCH§, DIETER RANK¶, HELFRIED SCHEIFINGER|| and WILLIBALD STICHLER# †Viktor-Wittner-Gasse 36/7, Vienna, Austria (IAEA Vienna) ‡Environment Agency, Vienna, Austria §ARC Seibersdorf Research, Seibersdorf, Austria ¶Center for Earth Sciences, University of Vienna, Vienna, Austria ||Central Institute for Meteorology and Geodynamics, Vienna, Austria #GSF-Research Centre for Environment and Health, Neuherberg, Germany (Received 03 July 2007; in final form 19 October 2007) The paper evaluates long-term seasonal variations of the deuterium excess (D-excess = δ 2 H − 8 · δ 18 O) in precipitation of stations located north and south of the main ridge of the Austrian Alps. It demonstrates that sub-cloud evaporation during precipitation and continental moisture recycling are local, respectively, regional processes controlling these variations. In general, sub-cloud evaporation decreases and moisture recycling increases the D-excess. Therefore, evaluation of D-excess variations in terms of moisture recycling, the main aim of this paper, includes determination of the effect of sub-cloud evaporation. Since sub-cloud evaporation is governed by saturation deficit and distance between cloud base and the ground, its effect on the D-excess is expected to be lower at mountain than at lowland/valley stations. To determine quantitatively this difference, we examined long-term seasonal D-excess variations measured at three selected mountain and adjoining valley stations. The altitude differences between mountain and valley stations ranged from 470 to 1665 m. Adapting the ‘falling water drop’ model by Stewart [M.K. Stewart. Stable isotope fractionation due to evaporation and isotopic exchange of falling water drops: applications to atmospheric processes and evaporation of lakes. J. Geophys. Res., 80(9), 1133–1146 (1975).], we estimated that the long-term average of subcloud evaporation at the selected mountain stations (altitudes between about 1600 and 2250 m.a.s.l.) is less than 1 % of the precipitation and causes a decrease of the D-excess of less than 2 ‰. For the selected valley stations, the corresponding evaporated fraction is at maximum 7 % and the difference in D-excess ranges up to 8 ‰. The estimated D-excess differences have been used to correct the measured long-term D-excess values at the selected stations. Finally, the corresponding fraction of water vapour has been estimated that recycled by evaporation of surface water including soil water from the ground. For the two mountain stations Patscherkofel and Feuerkogel, which are located north of the main ridge of the Alps, the maximum seasonal change of the corrected D-excess (July/August) has been estimated to be between 5 and 6 ‰, and the corresponding recycled fraction between 2.5–3 % of the local precipitation. It has been found that the estimated recycled fractions are in good agreement with values derived from other approaches. Keywords: Alpine regions; Craig–Gordon model; Deuterium excess; Evaporation; Hydrogen-2; Precipitation

*Corresponding author. Email: [email protected]

Isotopes in Environmental and Health Studies ISSN 1025-6016 print/ISSN 1477-2639 online © 2008 Taylor & Francis http://www.tandf.co.uk/journals DOI: 10.1080/10256010801887208

Techset Composition Ltd, Salisbury

GIEH288889.TeX

Page#:

10

Printed: 16/1/2008

2

51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100

1.

Klaus Froehlich et al.

Introduction

Atmospheric moisture recycling plays an important role in the formation of precipitation. The sources for precipitation that falls out of the atmosphere in a given region include moisture evaporated from the surface within the region and moisture transported by winds through it. The fraction of precipitation that comes from evaporation is defined as recycling rate, which depends on the size of the selected region (domain). As a global average, recycling over land is for a 500 km domain just less than 10 % [2]; higher values have been found for larger domains, for example, about 34 % over the Amazon and more than 21 % for the Mississippi Basin [2, 3]. Given the uncertainties of the approaches based on global and regional climate models and back-trajectory algorithm, a combination with alternative approaches is desirable. The use of the stable isotope composition in precipitation has shown specific potential in this regard [4–8]. Moisture evaporated from the land surface is formed by plant transpiration and evaporation of water from soils and lakes. The latter component is high in D-excess because of kinetic isotope fractionation during evaporation. Recycling of such moisture to the atmosphere increases the D-excess of the atmospheric vapour and consequently of the precipitation formed by condensation of this vapour. Thus, the D-excess of precipitation appears to be an indicator of evaporated moisture recycling. However, the D-excess can also decrease due to the local effect of sub-cloud evaporation of precipitation [1], which consequently hampers the use of this parameter as recycling indicator. To eliminate this effect, Peng et al. [9] suggested a selective precipitation sampling (snow and heavy rain samples), anticipating that in these samples the effect of sub-cloud evaporation is small or negligible. This paper presents, in section 3, an alternative approach in reducing or even eliminating the effect of sub-cloud evaporation. Given, that the distance between cloud base and ground and the saturation deficit are lower at high-altitude than at low-altitude stations, it is assumed that sub-cloud evaporation is remarkably reduced in precipitation of mountain stations. Such stations have been selected from the Austrian Network of Isotopes in Precipitation (ANIP). The isotope data of mountain stations have been compared with corresponding data of nearby valley stations. It has been found that the difference between the D-excess of adjoining stations is in good agreement with the corresponding values derived from a sub-cloud evaporation model that is based on the ‘falling water drop’ model by Stewart [1]. In section 4, the measured D-excess values are corrected for this sub-cloud evaporation effect. In general, the corrected values tend to increase during summer months. Assuming that the corrected D-excess represents a mixture of vapour entering a given domain and vapour recycled within this domain and applying a simple two-component mixing model, the recycled fraction can be estimated.

2.

Deuterium excess variations in Alpine precipitation

Data from long-term combined measurements of oxygen-18 (δ 18 O) and deuterium (δ 2 H) in precipitation of more than 30 sampling stations are available from ANIP, which is operational since 1972 [10]. In this paper, isotope data for the period 1973–1994 have been used to evaluate precipitation-weighted monthly averages of δ 18 O and D-excess of selected ANIP stations. During this period no significant temporal change of the D-excess has been observed. Three mountain stations and three nearby valley stations have been identified at which the isotopic composition of precipitation, precipitation rate, air temperature and relative humidity have been available for the selected observation period. The difference between the altitudes of the

Deuterium excess in precipitation

101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150

Table 1.

Location of selected stations, mean annual temperature, precipitation and its isotopic composition.

Stations 1. Patscerkofel Innsbruck 2. Feuerkogel Weyregg 3. Villacher Alpe Bad Bleiberg †

3

Longitude (◦ E)

Latitude (◦ N)

Elevation (m.a.s.l.)

Temperature (◦ C)

Mean annual precipitation (mm)

δ 18 O (‰)†

d-exc (‰)†

11.45 11.35 13.72 13.57 13.67 13.67

47.20 47.27 47.82 47.90 46.60 46.62

2245 580 1598 469 2140 904

1.7 11.9 4.6 10.0 1.3 7.9

875 930 1786 1156 1273 1274

−13.5 −10.3 −11.9 −10.4 −10.8 −10.6

12.1 6.5 12.2 9.2 10.0 9.6

weighted annual averages.

mountain and valley stations is significant, as reflected by the corresponding differences in the average monthly air temperature, δ 18 O, and D-excess values (table 1). Two of the station pairs (Patscherkofel/Innsbruck and Feuerkogel/Weyregg) are located north of the main ridge of the AustrianAlps and the pair Villacher Alpe/Bad Bleiberg south of it. In the Alps, there are three air masses that control weather and climate in this region, including precipitation and its isotopic composition. Atlantic maritime air masses dominate in the northern slopes of the Alps and its foreland; continental air masses are of influence in the eastern part of the Alps. The southern slopes of the Alps and its foreland are to a certain degree controlled by Mediterranean air. This convergence of the three air masses in the Alpine region is demonstrated by the precipitation pattern in the European region including the Alps for July, the month with the highest precipitation in this region (figure 1). The different origin of the air masses north and south of the main ridge appears to be indicated by the average annual δ 18 O values of the respective mountain stations (table 1). The lower δ 18 O values at the northern stations correspond to a stronger rainout (longer travel distance) of the Atlantic air masses in comparison to the southern stations where Mediterranean air masses with shorter travel distance are more frequent. This conclusion is in accordance with the tritium values and back-trajectory analysis of precipitation collected in July and August 1998 [11].

Figure 1. Precipitation pattern in July in Europe (x-axis in ◦ E, y-axis in ◦ N). The data represent average values over the last 45 years. They have been retrieved from the ECMWF (European Centre for Medium-Range Weather Forecasts) database. The given numbers represent the precipitation (in mm) that falls during 6 hours a day. They have to be multiplied by 4 to obtain the daily values for the given month. The location of the station pairs is given by the black dots: PK – Patscherkofel, FK – Feuerkogel, VA – Villacher Alpe. The SB sign has not been removed from the map.

4

151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200


Figure 2. Average monthly precipitation, D-excess and tritium in precipitation of the Station Portorz, Slovenia, for the period 2001–2003. The plotted data have been calculated from the IAEA database, attached to IAEA-TECDOC (2005) and from Vreˆca et al. (2003).

Figure 3. Average seasonal variation of D-excess in precipitation of the stations north (a) and south (b) of main ridge of the Alps. Patscherkofel (PK), Innsbruck (IB), Feuerkogel (FK), Weyregg (WR), Villacher Alpe (VA), Bleiberg (BB).


201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250

5

However, the comparison of the D-excess values (table 1) would disagree with this conclusion, if the common perception is adopted that Mediterranean air masses (precipitation) are characterized by higher D-excess than Atlantic air masses. This apparent discrepancy can be resolved considering the seasonal change of the D-excess in precipitation of nearby Mediterranean stations. For Portoroz at the coast of the Adriatic Sea [12], only about 120 km apart from the stations at the southern slope of the Alps, the seasonal change of the D-excess shows higher values at late autumn and early winter when the tritium values are at minimum (figure 2), which indicates that this precipitation has been formed by moisture evaporated from the Adriatic Sea. However, the precipitation at the southern slope of Alps reaches its maximum in the summer months July and August, when the D-excess of the air moisture from the Mediterranean is at its minimum. The striking feature in the seasonal change of the D-excess of all selected stations is the continuous increase from early spring to late summer (figures 3a and b). The highest D-excess values have been found for the stations north of the Alpine ridge, which also show a distinct difference between mountain and valley stations (figure 3a). The low or negligible difference of the D-excess between the mountain station Villacher Alp and nearby valley-station Bad Bleiberg suggests a high proportion of convective precipitation and thus a strong vertical mixing of the air in this area [11]. The increase of the D-excess in precipitation found in this mountain region is often missing at ‘lowland’ stations because of higher partial evaporation of precipitation between cloud base and ground, which is associated with a decrease of the D-excess and thus masks the increase observed at mountain stations. A more detailed evaluation of the sub-cloud evaporation effect is therefore necessary.

3. The effect of sub-cloud evaporation on the deuterium excess The estimation of the sub-cloud evaporation effect is based on a model developed by Stewart [1] to study the change of the isotopic composition of a falling water drop under well-defined laboratory conditions. Stewart’s model has been adapted to the conditions in the sub-cloud atmosphere at the selected stations. The values of the parameter to be used in this approach (long-term monthly averages of air temperature, dew point temperature and relative humidity) have been calculated using datasets provided by the Austrian Meteorological Service. Assuming that the initial isotopic composition of precipitation at the cloud base is in isotopic equilibrium with the surrounding water vapour, the following expression has been derived for the change of the D-excess (difference between the D-excess at the level of the sampling station (D) and the cloud base (Dc ): D − Dc = 2F − 818 F.

(1)

The parameter 2 F and 18 F are defined as follows:

i

γ F = 1− i α

i

(f iβ − 1)

(2)

with ‘i’ standing for ‘2’ (deuterium) and ‘18’ (oxygen-18), respectively. The parameter iγ and β are defined by Stewart [1]; they depend on the kinetic and equilibrium fractionation factor for deuterium and oxygen-18 and the relative humidity. The equilibrium fractionation factor has been calculated by the relationship of Majoube [13], and for the kinetic fractionation

i

6

251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300


factors the following values have been used for deuterium αk = 1 + n · 0.024

(2a)

αk = 1 + n · 0.0289

(2b)

2

and oxygen-18 18

For n the value 0.58 was adopted [1]. In equation (2), f represents the remaining fraction of the water-drop mass, which depends on the evaporation rate, initial radius and fall time of the drop. The evaporation rate of the falling drop has been determined on the basis of the laboratory experiments with falling water drops by Stewart [1]. For this purpose, the laboratory data had to be converted to the field conditions characterized by air temperature, dew point temperature and relative humidity at the various stations and months. The time of the drop fall from the cloud base to the ground was derived from the difference between dew point and surface air temperature, the adiabatic lapse rate, the fall velocity and radius of the water drop. The latter two parameters have been used as free (fitting) parameters, and the values selected for the following evaluations are: 250 m/min. and 1.3 mm, respectively. By this approach, the values of the parameter iF (equation 2) have been calculated, and with equation (1), the difference of the D-excess between ground (D) and cloud base (Dc ) is determined. It has been found (figure 4) that this sub-cloud evaporation effect (D − Dc ) is proportional to the evaporated fraction (1 − f ). Although the slope of this ‘evaporation line’ does not explicitly depend on relative humidity and ambient temperature, the evaporated fraction itself is a function of the evaporation rate and, through it, of relative humidity and temperature. The slope increases with the value of n, i.e., the kinetic fractionation factor. Figure 4 clearly indicates the extent to which sub-cloud evaporation modifies the D-excess of precipitation. At mountain stations the sub-cloud evaporation is less than about 1 % resulting in a modification of the D-excess of less than 2 ‰. The D-excess corrected for the sub-cloud effect is defined as difference between the measured D-excess and the value derived from equation (1) and, thus, represents the D-excess at cloud base. Therefore, the corrected D-excess of valley stations is expected to be equal to the

Figure 4. Sub-cloud evaporation effect; change of the D-excess with increasing evaporated fraction (decreasing remaining fraction) estimated for the various mountain and valley stations. EK has not been removed from the plot legend.


301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350

7

one of the nearby mountain station, if the altitude of the cloud base does not change. In fact, for the two station pairs north of the main ridge of the Alps, the corrected D-excess values of the valley stations appear in reasonably good agreement with the one of the mountain stations. However, during the winter season the corrected D-excess values of the two mountain stations are remarkably higher than the corresponding values of the valley stations. This difference indicates that the sub-cloud evaporation model is not applicable for the winter season where precipitation predominantly is formed by snow rather than rain. In case of Villacher Alpe/Bad Bleiberg, there is no remarkable difference in D-excess between mountain and valley station (figure 5c), which suggests the dominance of convective precipitation that is associated with vertical mixing of the atmospheric water. The inappropriately high corrected D-excess values at the valley station Bad Bleiberg prove that in this case the application of the sub-cloud evaporation model is invalid. In conclusion, figures 5a–c demonstrate that mountain stations above about 1600 m.a.s.l. are nearly free of the sub-cloud evaporation effect. The continuous increase of the D-excess from spring to late summer is suggested to be associated with moisture recycling, especially water vapour evaporated from surface water and soil of the continent during air mass movement. In the following, the recycled fraction is estimated on the basis of the seasonal variation of the corrected D-excess at the mountain stations on the north ridge of the Austrian Alps corresponding to the recycling domain from north-west/west.

Figure 5. Long-term average seasonal change of d-excess derived from measurements and corrected for sub-cloud evaporation at the stations Patscherkofel/Innsbruck (a), Feuerkogel/Weyregg (b), Villacher Alpe/Bad Bleiberg (c).

8

Q1

351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400

4.


Recycling of continental moisture

A first attempt in estimating moisture recycling by changes of the D-excess in precipitation has been made by Salati et al. [14] for the Amazon Basin and by Gat et al. [5] for the Great Lakes in North America. Evaporation of surface water from swamps and lakes in Siberia and subsequent moisture transport towards Western Mongolia has been suspected to be the cause of higher D-excess values in firn of a Mongolian glacier [15]. Recently, Peng et al. [9] have suggested a simplified method to estimate the vapour contribution from regional evaporation in local precipitation. The method assumes that the D-excess of local precipitation represents a mixture of two components: one component with a fraction fc derived from the continental evaporation flux with D-excess Devap , and the other component with a fraction (1 − fc ) from advected vapour flux with D-excess Dadv . From the mixing equation it follows that: D − Dadv fc = . (3) Devap − Dadv This simplified method is consistent with a ‘modified Rayleigh fractionation model’ (Peng et al. [9]) and provides a good approximation of the recycled moisture evaporated in the continental region along the air mass trajectory. The model requires estimating the D-excess of the advection and evaporated flux components. For the advected flux component, the average D-excess value of the months with negligible evaporation (January to March) has been used, and the D-excess of the evaporated moisture is estimated from the Craig–Gordon equation [16]: Revap =

RW /α − hRA , (1 − h)αK

(4)

where Ri (i = evap, W, A) represents the isotopic ratio (Ri = 1 + δi ). The average of the annual weighted mean isotopic composition of precipitation of relevant German stations has been derived from the GNIP database (GNIP, IAEA [17]). The values obtained are δW 18 O = −8.6 ‰, δW 2 H = −60.9 ‰ and have been used to calculate the isotopic ratio of the evaporating water RW for 18 O and 2 H. The mean annual temperature used to calculate the average equilibrium fractionation factor α has been determined to be 8.8 ◦ C. Assuming isotopic equilibrium between atmospheric vapour and precipitation, it follows RA = RW /α. In this way, Devp and Dadv can be determined, and equation 3 is used to derive the recycled fraction fc from the corrected D-excess values. With the estimated values of Devp and Dadv , equation 3 suggests that the value of the corrected D-excess changes by 2 ‰ if the evaporated fraction changes about 1 %.

5.

Results

In the following, the seasonal change of the recycled fraction over the domain north-west and west of the station Patscherkofel and Feuerkogel is estimated by this approach. The mountain stations rather than the corresponding valley stations have been selected to minimize the uncertainties inherent in the estimation of the sub-cloud evaporation effect. Figure 6 demonstrates that the use of uncorrected D-excess values would result in an underestimation of the recycled fraction of about 20–25 %. It should also be noted that the lower seasonal change of the D-excess at the stations south of the main ridge (figure 5c) suggests a lower recycling of evaporated moisture to air masses of predominantly Mediterranean origin with a relatively short over-land travel distance [20].


401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450

9

Figure 6. Evaporated fraction in precipitation of the mountain stations Patscherkofel (PK) and Feuerkogel (FK). The full lines have been estimated using d-excess values corrected for sub-cloud evaporation effect; the broken lines are for the uncorrected d-excess values.

The estimated fc values are in the order of a few percent and, thus, in good agreement with estimates based on other approaches. Taking into consideration the estimate of the total evapotranspiration recycling rate, which for the Mississippi valley has been estimated to be in the range of about 20–50 % [3], the contribution of continental evaporation to the evapotranspiration recycling appears to be in the order of 10 %. This figure compares well with 14 %, the fraction of evaporated soil/surface water in the total evapotranspiration flux estimated by Yepez et al. [18]. For the specific conditions of the Tibetan Plateau, Tsujimura et al. [7] found that 27 % of the precipitation is due to evaporation from the soil surface. Furthermore, the increase of the evaporation fraction from April to July in figure 6 corresponds with the increase of the total evapotranspiration due to increased leaf area index in Middle and Northern Europe [19]. In July, the soil moisture content reaches a critically low value [19], which explains the reversal in the seasonal change of the evaporation fraction (figure 6).

6.

Conclusions

It has been shown that sub-cloud evaporation and continental moisture recycling are local, respectively, regional processes affecting the seasonal variation of the D-excess in local precipitation. At Alpine mountain stations with altitudes of more than 1600 m.a.s.l. the sub-cloud evaporation has been estimated to be lower than about 1 % resulting in a modification of the D-excess of less than 2 ‰. For the selected valley stations with an altitude difference of more than 1000 m, the maximum evaporated fraction is 7 % and the corresponding change of the D-excess about 8 ‰. The difference in D-excess between mountain and valley station south of the main Alpine ridge (Villacher Alpe/Bad Bleiberg) has found to be negligible, which suggests prevailing convective precipitation with strong vertical mixing. A linear relation between the seasonal change of the corrected D-excess and the recycled fraction of surface evaporated water has been found using a two-component mixing model. At the mountain stations in the north of the Alpine ridge, which predominantly receive Atlantic air masses, the maximum seasonal change of the D-excess is 5–6 ‰ suggesting a maximum recycled fraction in July/August of 2.5–3 %. The lower seasonal change of the D-excess at the stations south of the main ridge seems to be consistent with a lower contribution of evaporated

10

451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500


moisture in the air masses, which mainly are of Mediterranean origin with a relatively short travel distance. The results of these evaluations have been found to be in good agreement with values derived by other approaches and, thus, encourage further use of the D-excess in studying moisture recycling processes. References [1] M.K. Stewart. Stable isotope fractionation due to evaporation and isotopic exchange of falling water drops: applications to atmospheric processes and evaporation of lakes. J. Geophys. Res., 80(9), 1133–1146 (1975). [2] K.E. Trenberth.Atmospheric moisture recycling: role of advection and local evaporation. J. Clim., 12, 1368–1381 (1999). [3] K.L. Brubaker, P.A. Dirmeyer, A. Sudradjat, B.S. Levy, F. Bernal. A 36-year climatological description of the evaporative sources of warm-season precipitation in the Mississippi River Basin. J. Hydrometeorol., 2(6), 537–557 (2001) [4] J.R. Gat, E. Matsui. Atmospheric water balance in the Amazon Basin: an isotopic evapo-transpiration model. J. Geophys Res., 96(D7), 13179–13188 (1991). [5] J.R. Gat, C.J. Bowser, C. Kendall. The contribution of evaporation from the Great Lakes to the continental atmosphere: estimate based on stable isotope data. Geophys. Res. Lett., 21, 557–560 (1994). [6] J. Jouzel, K. Froehlich, U. Schotterer. Deuterium and oxygen-18 in present-day precipitation: data and modelling. Hydrol. Sci. J., 42, 747–763 (1997). [7] M. Tsujimura, A. Numaguti, L. Tian, S. Hashimoto, A. Sugimoto, M. Nakawo. Behaviour of subsurface water revealed by stable isotope and tensiometric observation in the Tibetan Plateau. J. Meteorol. Soc. Japan, 79(1B), 599–605 (2001). [8] A. Sugimoto, A. Numaguti, M. Tsujimura, K. Fujita, S. Hashimoto, A. Yatagai, M. Nakawo. Water vapour transport to the Tibetan Plateau revealed with stable isotopes of precipitation: a new hypothesis for unusual isotope signals, Abstract, paper presented at American Geophysical Union, Fall Meeting 2003. 2003AGUFM.A31C0057S. [9] H. Peng, B. Mayer, A. Norman, H.R. Krouse. Modelling of hydrogen and oxygen isotope compositions for local precipitation. Tellus, 57B, 273–282 (2005). [10] A. Kaiser, H. Scheifinger, M. Kralik, W. Papesch, D. Rank, W. Stichler. Links between meteorological conditions and spatial/temporal variations in long-term isotope records from the Austrian precipitation network. C&S Paper Series 13/P, IAEA (2001). [11] D. Rank, W. Papesch. Isotopic composition of precipitation in Austria in relation to air circulation patterns and climate. In Isotopic Composition in the Mediterranean Basin in Relation to Air Circulation Patterns and Climate. IAEA-TECDOC-1453, IAEA, Vienna, pp. 19–35 (2005). [12] P. Vreˆca, T. Kanduˆc, S. Žigon, Z. Trkov. Isotopic composition of precipitation in Slovenia. In Isotopic Composition of Precipitation in the Mediterranean Basin in Relation to Air Circulation Patterns and Climate. IAEA-TECDOC-1453. IAEA, Vienna, pp. 157–172 (2005). [13] M. Majoube. Fractionnement en oxygène-18 et en deutérium entre l’eau et sa vapeur. J. Chem. Phys., 197, 1423–1436 (1971). [14] E. Salati, A. DallÓlio, E. Matsui, J.R. Gat. Recycling of water in the Amazon Basin: an isotopic study. Water Resour. Res., 15(5), 1250–1258 (1979). [15] U. Schotterer, K. Froehlich, H.W. Gäggeler, S. Sandjordj, W. Stichler. Isotope records from Mongolian and Alpine ice cores as climate indicators. Clim. Change, 36, 519–530 (1997). [16] H. Craig, L.I. Gordon. Deuterium and oxygen-18 variations in the ocean and in the marine atmosphere. In Proceedings of the Conference on Stable isotopes in Oceanographic studies and Paleotemperatures, E. Tongiorgi (Ed.). Laboratory of Geology and Nuclear Science, Pisa, 9–130 (1965). [17] IAEA. Global Network for Isotopes in Precipitation (GNIP), IAEA, Vienna (2001); IAEA, Isotopic Composition of Precipitation in the Mediterranean Basin in Relation to Air Circulation Patterns and Climate. IAEA-TECDOC-1453. IAEA, Vienna (2005). [18] E.A. Yepez, D.G. Williams, R.L. Scott, G. Lin. Partitioning overstory and understory evapotranspiration in a semiarid savannah woodland from the isotopic composition of water vapour. Agric. Forest Meteorol., 119, 53–68 (2003). [19] P. Heck, D. Lüthi, Ch. Schär. The Influence of Vegetation on the Summertime Evolution of European Soil Moisture. Phys. Chem. Earth (B), 24, 609–614 (1999).