Seasonality and variability ofstratospheric water vapour Tongmei Wang

Tongmei W

ang    Seasonality an

d variability of stratospheric w

ater vapour

Doctoral Thesis in Atmospheric Sciences and Oceanography at Stockholm University, Sweden 2019

Department of Meteorology

ISBN 978-91-7797-853-4

Seasonality and variability of stratospheric watervapourTongmei Wang

Academic dissertation for the Degree of Doctor of Philosophy in Atmospheric Sciences andOceanography at Stockholm University to be publicly defended on Friday 29 November 2019 at10.00 in Högbomsalen, Geovetenskapens hus, Svante Arrhenius väg 12.

AbstractStratospheric water vapour (SWV) plays a critical role in the climate system by modulating the radiation budget andinfluencing the stratospheric chemistry. Studying changes of SWV on global scale is helpful for our understanding ofclimate change. This thesis aims to gain an improved understanding of the stratospheric processes and dynamic mechanismsthat determine the seasonality and variability of SWV.

Water vapour is characterized by its compound, which leaves an isotopic fingerprint in relevant atmospheric andhydrologic processes. The thesis starts with analyzing the global features of three stable water isotopes (SWIs) in thestratosphere by using satellite retrievals from Odin/SMR. The spatial pattern of SWI indicates clear effects of methaneoxidation in the upper stratosphere, dehydration at the tropopause and stratospheric transport via the Brewer-Dobsoncirculation (BDC). In addition to the tropical tape recorder in the lower stratosphere, a pronounced downward propagationof the seasonal signal from the upper to the lower stratosphere is observed in high-latitudes. These observed features arefurther compared to model outputs to identify possible causes of model deficiencies in reproducing the distribution of SWV.

The downward propagation signal of zonal wind has been demonstrated in the high-latitude stratosphere in springseasonal transition in the Southern Hemisphere, but not in the Northern Hemisphere. This inter-hemispheric difference isdue to the stronger stratospheric planetary wave activity in austral spring than in boreal spring. With strong wave activityin spring, the transition is inclined to occur first at the stratopause followed by a downward propagation to the lowerstratosphere. In particular, the stronger the upward propagation of planetary waves in high-latitudes in spring the earlierthe stratospheric seasonal transition.

The new generation reanalysis ERA5 represents climatological distribution and seasonal cycle of SWV better than itspredecessor ERA-Interim by assimilating more satellite observations. The variability of SWV in ERA5 is highly consistentwith SDI MIM observation. The interannual variability of water vapour in the lower stratosphere is found to be closelylinked to the tropical Quasi-Biennial Oscillation (QBO) and QBO-induced residual circulation. On decadal scale, the deficitof SWV in boreal winter is associated with a warm sea surface temperature (SST) anomaly in the North Atlantic, whichleads to stronger upward propagation of planetary waves, resulting in a warmer pole in the lower stratosphere, coldertropical tropopause and stronger BDC, hence less water vapour enters the stratosphere through the tropopause and theanomaly extends to the entire stratosphere.

Sensitivity experiments for a CO2 doubling scenario are performed with the model WACCM to investigate the SWVresponse to climate change. The response of SWV is dominated by the warm SST, which is induced by CO2 doubling ina coupled ocean-atmosphere system. The enhanced SST leads to a moist troposphere and warmer tropical and subtropicaltropopause, resulting in more water vapour entering the stratosphere from below. A large increase of SWV in the lowerstratosphere, in turn, affects stratospheric temperature. It results in a warming in the tropical and subtropical lowerstratosphere, offsetting the cooling caused by CO2 doubling in general.

Keywords: stratospheric water vapour, seasonality, variability.

Stockholm 2019http://urn.kb.se/resolve?urn=urn:nbn:se:su:diva-173284

ISBN 978-91-7797-853-4ISBN 978-91-7797-854-1

Department of Meteorology

Stockholm University, 106 91 Stockholm

SEASONALITY AND VARIABILITY OF STRATOSPHERIC WATERVAPOUR 

Tongmei Wang

Seasonality and variability ofstratospheric water vapour 

Tongmei Wang

©Tongmei Wang, Stockholm University 2019 ISBN print 978-91-7797-853-4ISBN PDF 978-91-7797-854-1 Cover image: Planet Earth seen from Space.(From NASA Image and Video Library) Printed in Sweden by Universitetsservice US-AB, Stockholm 2019

To my beloved family

iii

Abstract

Stratospheric water vapour (SWV) plays a critical role in the climate system by modulating the radiation budget and influencing the stratospheric chemi-stry. Studying changes of SWV on global scale is helpful for our understan-ding of climate change. This thesis aims to gain an improved understanding of the stratospheric processes and dynamic mechanisms that determine the seasonality and variability of SWV.

Water vapour is characterized by its compound, which leaves an isotopic fingerprint in relevant atmospheric and hydrologic processes. The thesis starts with analyzing the global features of three stable water isotopes (SWIs) in the stratosphere by using satellite retrievals from Odin/SMR. The spatial pattern of SWI indicates clear effects of methane oxidation in the upper stra-tosphere, dehydration at the tropopause and stratospheric transport via the Brewer-Dobson circulation (BDC). In addition to the tropical tape recorder in the lower stratosphere, a pronounced downward propagation of the sea-sonal signal from the upper to the lower stratosphere is observed in high-latitudes. These observed features are further compared to model outputs to identify possible causes of model deficiencies in reproducing the distribution of SWV.

The downward propagation signal of zonal wind has been demonstrated in the high-latitude stratosphere in the spring seasonal transition in the Sout-hern Hemisphere, but not in the Northern Hemisphere. This inter-hemispheric difference is due to the stronger stratospheric planetary wave activity in austral spring than in boreal spring. With strong wave activity in spring, the transition is inclined to occur first at the stratopause followed by a downward propagation to the lower stratosphere. In particular, the stronger the upward propagation of planetary waves in high-latitudes in spring the earlier the stratospheric seasonal transition.

The new generation reanalysis ERA5 represents climatological distribut-ion and seasonal cycle of SWV better than its predecessor ERA-Interim by assimilating more satellite observations. The variability of SWV in ERA5 is highly consistent with SDI MIM observation. The interannual variability of water vapour in the lower stratosphere is found to be closely linked to the tropical Quasi-Biennial Oscillation (QBO) and QBO-induced residual cir-culation. On decadal scale, the deficit of SWV in boreal winter is associated with a warm sea surface temperature (SST) anomaly in the North Atlantic. It

iv

leads to stronger upward propagation of planetary waves, resulting in a warmer pole in the lower stratosphere, colder tropical tropopause and stronger BDC. Hence, less water vapour enters the stratosphere through the tropopause and the anomaly extends to the entire stratosphere.

To investigate the SWV response to climate change, sensitivity experi-ments for a CO2 doubling scenario are performed with the model WACCM. The response of SWV is dominated by the warm SST, which is induced by CO2 doubling in a coupled ocean-atmosphere system. The enhanced SST leads to a moist troposphere and warmer tropical and subtropical tropopause, resulting in more water vapour entering the stratosphere from below. A large increase of SWV in the lower stratosphere, in turn, affects stratospheric tem-perature. It results in a warming in the tropical and subtropical lower stra-tosphere, offsetting the cooling caused by CO2 doubling in general.

v

Sammanfattning

Stratosfärisk vattenånga (SWV) spelar en viktig roll i klimatsystemet genom att modulera strålningsbudgeten och påverka kemiska processer i stratosfä-ren. Genom att studera förändringar av stratosfärisk vattenånga på en global skala har vi möjlighet att öka vår förståelse för den pågående klimatföränd-ringen. Målet med denna avhandling är att öka förståelsen för de stratosfä-riska processer samt dynamiska mekanismer som bestämmer årstidsvariat-ioner samt variabilitet av SWV.

Vattenånga kännetecknas av dess sammansättning, som lämnar ett isotopiskt fingeravtryck i olika atmosfäriska och hydrologiska processer. Avhandlingen börjar med att analysera de globala attributen hos tre stabila vatten isotoper (SWIs) i stratosfären genom satellitobservationer från Sub-Millimeter Radiometer instrumentet (SMR) på Odin satelliten. Mönstret från SWI indikerar tydliga effekter av metanoxidation i den övre stratosfären, uttorkning vid tropopausen och stratosfärisk transport via Brewer-Dobson-cirkulationen. Utöver processen kallad tropical taperecorder i den nedre stra-tosfären, ses även en nedåtgående propagerande säsongsbaserad signal från övre till nedre stratosfären på höga latituder. Dessa observerade strukturer jämförs vidare med atmosfäriska modeller för att identifiera möjliga orsaker till modellbrister vid reproduktion av fördelningen av SWV.

Den nedåtpropagarande signalen i det zonala vind mönstret har påvisats över höga latituder i stratosfären under övergången från vinter till sommar på södra hemisfären, men inte på norra hemisfären. Denna interhemisfäriska skillnad orsakas av att den stratosfäriska planetära vågaktiviteten är starkare under södra hemisfärens vår än norra hemisfärens vår. Med stark vågaktivi-tet på våren tenderar övergången att inträffa först vid stratopausen och följs sedan av en nedåtgående propagering till den nedre stratosfären. Tydligt ses att med desto starkare uppåtgående propagering av planetära vågor under på höga latituder under våren, desto tidigare inträffar den stratosfäriska års-tidsövergången.

Den nya generationens ERA5 reanalys representerar den klimatologiska fördelningen och säsongsbetonade cykeln av SWV bättre än dess föregång-are ERA-Interim genom att assimilera en större mängd satellit observationer. Variabiliteten av SWV i ERA5 överensstämmer mycket väl med observat-ioner från SPARC Data Initiative Multi-Instrument Mean. Variabiliteten av vattenånga i nedre stratosfären för olika år har visat sig vara nära samman-

vi

kopplad med det stratosfäriska tropiska vindfenomenet Quasi-Biennial Osci-lation (QBO) och dess associerade vind cirkulation. På en tidsskala över flera decennium föreslår våra resultat att underskottet av SWV på norra he-misfärens vinter är förknippad med en varm anomali i Nordatlantens havsy-tetemperatur. Den varmare Nordatlanten leder till starkare uppåtgående pro-pagering av planetära vågor, vilket resulterar i högre temperaturer över den lägre stratosfären, lägre temperaturer över den tropiska tropopausen och starkare Brewer Dobson Cirkulation. Följden blir att mindre vattenånga tränger in i stratosfären genom tropopausen och denna avvikelsen sträcker sig över hela stratosfären.

För att undersöka hur SWV reagerar på klimatförändringen utförs käns-lighetsexperiment med den globala atmosfärsmodellen The Whole At-mosphere Community Climate Model (WACCM) där effekten av en för-dubbling av mängden koldioxid i atmosfären jämfört med förindustriell tid undersöks. Våra resultat visar att SWV är starkt påverkat av den förhöjda havsytetemperatur som orsakas av den fördubblade halten av koldioxid i ett kopplat havs-atmosfärssystem. De förhöjda havsytetemperaturerna leder till en fuktig troposfär och varmare tropisk och subtropisk tropopaus, vilket resulterar i att mer vattenånga tränger in i stratosfären underifrån. En stor ökning av SWV i den lägre stratosfären påverkar i sin tur temperaturen i stratosfären. Det resulterar i en uppvärmning i den tropiska och subtropiska nedre stratosfären, vilket kompenserar för den allmänna avkylningen orsakad av koldioxidfördubblingen.

vii

List of papers

The following papers, referred to in the text by their Roman numerals, are included in this thesis:

Paper I: Wang, T., Zhang, Q., Lossow, S., Chafik, L., Risi, C., Mur-tagh, D., and Hannachi, A. (2018). Stable Water Isoto-pologues in the Stratosphere Retrieved from Odin/SMR Measurements. Remote Sensing, 10(2), 166, 1-21. https://doi.org/10.3390/rs10020166.

Paper II: Wang, T., Zhang, Q., Hannachi, A., Lin. Y., and Hirooka, T. (2019). On the dynamics of the spring seasonal transition in the two hemispheric high-latitude stratosphere. Tellus A: Dy-namic Meteorology and Oceanography, 71(1), 1-18. https://doi.org/10.1080/16000870.2019.1634949.

Paper III: Wang, T., Zhang, Q., Hannachi, A., Hirooka, T., and Hegglin, M.I. (2019). Tropical water vapour in the lower stratosphere in ERA5 and its relationship to tropical/extra-tropical dynamic processes. Under review in Quarterly Journal of the Royal Meteorological Society.

Paper IV: Wang, T., Zhang, Q., Kuilman, M., and Hannachi, A. (2019). Response of stratospheric water vapour to CO2 doubling in WACCM. Manuscript.

Reprints were made with permission from the publishers

ix

Author’s contribution

The project for paper I was proposed by my supervisor Qiong Zhang. The idea was developed during discussions between me, Qiong Zhang and Stefan Lossow. Camille Risi provided model outputs from LMDZ4. I did all data analysis and produced the plots. I wrote the paper with support from my two supervisors Qiong Zhang and Abdel Hannachi. I conducted all revisions from the reviewers with input from all co-authors.

The idea for paper II was initiated in a discussion between me, Abdel Han-nachi, Yihua Lin and Qiong Zhang, developed in discussions between me, Toshihiko Hirooka and Abdel Hannachi. I performed all analysis, produced the plots and wrote the paper. I conducted all revisions from the reviewers with input from all co-authors.

The idea for paper III was initially developed during discussions between me, Qiong Zhang, Abdel Hannachi and Toshihiko Hirooka. Michaela I. Hegglin provided the SDI MIM data and Figure 2. I analysed the data, plot other figures and wrote the paper. All co-authors contributed to revision of the manuscript.

The idea for paper IV resulted from discussions between me, Qiong Zhang and Abdel Hannachi. Maartje Kuilman provided model outputs from WACCM. I did all data analysis and produced the plots. I wrote the paper with input from all co-authors.

xi

Contents

Abstract iii Sammanfattning v List of Papers vii Author’s contribution ix Abbreviations xiii

1. Introduction .................................................................................... 1

2. Observations and sources of stratospheric water vapour ........... 4 2.1 Observations .......................................................................................... 4 2.2 Methane oxidation in the stratosphere .................................................. 5 2.3 Dehydration at the tropical tropopause layer ........................................ 6

3. Stratospheric dynamics and water vapour transport ................. 8 3.1 The Brewer-Dobson circulation (BDC) ................................................ 8 3.2 Planetary waves and stratospheric circulation .................................... 10 3.3 Water vapour transport and seasonality .............................................. 14

4. Variability of stratospheric water vapour .................................. 16 4.1 Stratospheric oscillations .................................................................... 16 4.2 Variability of SWV in observations and reanalyses ........................... 17 4.3 Response of SWV to a changing climate ............................................ 20

5. Summary and outlook .................................................................. 23 5.1 Summary of papers included in this thesis .......................................... 23 5.2 Outlook ................................................................................................ 25

Acknowledgements ........................................................................... 27

References .......................................................................................... 28

xiii

Abbreviations

BDC Brewer-Dobson Circulation

DJF December-January-February

ERA European Reanalysis

GCM General Circulation Model

JJA June-July-August

MIM Multi-Instrument Mean

MLS Microwave Limb Sounder

NAO North Atlantic Oscillation

NH Northern Hemisphere

PI Pre-Industry

PSC Polar Stratospheric Cloud

QBO Quasi-Biennial Oscillation

SDI SPARC Data Initiative

SH Southern Hemisphere

SMR Sub-Millimetre Radiometer

SPARC Stratosphere-troposphere Processes And their Role in Climate

SST Sea Surface Temperature

SSW Sudden Stratospheric Warming

SWI Stable Water Isotope

SWV Stratospheric Water Vapour

TEM Transformed Eulerian Mean

UTLS Upper Troposphere and Lower Stratosphere

WAF Wave Activity Flux

WCRP World Climate Research Programme

WACCM Whole Atmosphere Community Climate Model

1

1. Introduction

Invisible water vapour is the largest and most important greenhouse gas in the atmosphere. It provides the largest positive feedback in model project-ions of climate change (Solomon et al., 2007). Stratospheric water vapour (SWV) is chemically, physically, and radiatively active (e.g. Forster and Shine, 1999, 2002; Solomon et al., 2010; Riese et al., 2012). It plays mul-tiple roles and makes a significant contribution to climate change. SWV influences the radiation budget and modulates stratospheric ozone chemistry (Vogel et al., 2011). Any change in its abundance may have non-negligible impacts on various processes. For example, an increase of SWV can affect atmospheric temperatures from the stratosphere to the Earth’s surface (Riese et al., 2012). Solomon et al. (2010) attribute 30% of the surface temperature change to the radiative forcing of increased SWV since 1980. Wang et al. (2017) suggest that a persistent change in SWV could have contributed to the observed hiatus in global warming after 2000. According to Stenke and Grewe (2005), 10% of the global total ozone decline from 1960 to 1999 can be explained by the increase in SWV.

As a valuable trace gas, water vapour provides a comprehensive picture of the physical and dynamical processes in the entire atmosphere, especially in middle atmosphere where observations of the circulation are quite defici-ent. For example, the vertical distribution of SWV provides evidence for a large-scale circulation, the Brewer-Dobson Circulation (BDC) (Brewer, 1949; Dobson, 1950). In BDC the tropospheric air enters the stratosphere in tropics and then moves upward and poleward before descending in the middle and high latitudes. The knowledge of the spatial and temporal distri-bution of the SWV is of great interest for a better understanding of the pro-cesses controlling the water vapour content in the atmosphere. Studying the current and future variability of SWV on a global scale is invaluable for our understanding of climate and climate change.

SWV is determined by in situ methane oxidation and the intrusion of water vapour through the tropical tropopause layer, which is considered as the “gate to the stratosphere” (Fuglistaler et al., 2009). The water vapour mixing ratio in the lower stratosphere is strongly governed by microphysical

2

processes that act on tropical air masses as they penetrate through the tro-pical tropopause layer into the stratosphere (Hoskins, 1991; Holton et al., 1995). This leads to condensation (or dehydration), freezing out most of the remaining water content due to the extremely cold tropical tropopause. Above the middle stratosphere, water vapour gradually increases due to the oxidation of methane. Stratospheric transport is dominated by mean diabatic advection, such as upwelling in the tropics and downwelling in the surf zone at higher latitudes. Together with mixing, it determines the global-scale dis-tribution of tracers including water vapour (Plumb, 2002). Therefore, the variation of SWV embeds information on changes of these relevant stra-tospheric processes. However, the contribution of individual processes to the overall water budget of stratosphere is complex and not well understood yet.

The seasonal cycle of the lower SWV in the tropics is characterised by the tape recorder (Mote et al., 1996). It exhibits a hydrated lower stra-tosphere in boreal summer and a dehydrated lower stratosphere in boreal winter. Therefore, most of the water vapour enters the stratosphere during boreal summer, when Monsoon systems (e.g. Gettelman and Kinnison, 2004) and enhanced deep convection over the tropics (e.g. Khaykin et al., 2009) transport water vapour into the tropical tropopause level. In order to quantify the contributions of each individual processes to the SWV budget, a detailed understanding of these processes is needed. This can lead to an im-proved representation of the hydrological cycle in GCMs, and thus a better representation of the variations of SWV can be achieved.

This thesis investigates observations and model outputs of SWV from different perspectives. In particular, it highlights the importance of the BDC, the planetary wave activity and water vapour changes and feedback proces-ses within the stratosphere. The first paper presents SWV and its isotopes retrieved from Odin/SMR. The second paper focuses on the stratospheric dynamics of seasonal transition and provides an improved understanding of the influence of planetary wave activity on the stratospheric circulation. Using the recently released new reanalysis ERA5, the third paper studies variability of SWV on interannual and decadal time scales, discusses the relevant mechanisms, including dynamical processes affecting the SWV. The last paper investigates the response of SWV to a CO2 doubling by using the atmospheric model, WACCM, to assess changes in SWV, and its feed-back to a warming climate. The following important questions constitute the skeleton of this thesis:

3

• The main features of the SWV and its isotopes in Odin/SMR observa-tions. How do SWV observations help to improve the isotope-enabled models?

• What is the inter-hemispheric difference in stratospheric seasonal transition and what are the causes for these differences?

• How is SWV represented in the new reanalysis ERA5? What process-es control the interannual and decadal variabilities of SWV?

• How does SWV respond to greenhouse warming and feedback to the stratospheric temperature?

Following the introduction section, chapter 2 introduces the observations, and some important stratospheric processes related to sources of SWV, which have been evidenced by the observations. Chapter 3 describes basic concepts of stratospheric dynamics and water vapour transport. Chapter 4 discusses stratospheric oscillation, SWV variability in observations and rea-nalyses, and SWV changes in a warming climate. Chapter 5 gives a sum-mary of papers included in this thesis and an outlook of this work.

4

2. Observations and sources of stratospheric water vapour

Observation is a core aspect in understanding atmospheric water vapour. The temporal and spatial distributions of SWV described by observations provide clear evidence of stratospheric processes, e.g., the methane oxidation in the stratosphere and the dehydration at the tropical tropopause.

2.1 Observations Water vapour is measured in situ by balloon and aircraft-based instruments, and remotely by satellite- and ground-based sensors. Unfortunately, it is difficult to obtain accurate, height-resolved global-scale measurements of water vapour in the middle atmosphere. The change in water vapour concen-tration of five orders of magnitude from the ground to the mesopause ex-plains the lack of a standard instrument that could measure it at every alti-tude. Measurements use a wide range of techniques and observational plat-forms, all with strengths and weaknesses. The inter-comparison among measurements is needed to make a progress in understanding the water va-pour distribution.

Water vapour measurements with global coverage are provided by satel-lites. Various techniques and different spectral ranges have been used to retrieve water vapour in the middle atmosphere. The first satellite limb ob-servations started with the launch of the Nimbus-6 satellite on 12 June 1975 (Gille et al., 1980). Since then, various water vapour products have appeared from different instruments aboard satellites. The most advanced instruments are MLS on the Aura Satellite (Waters et al., 2006), and the SMR on the Odin satellite (Urban et al., 2007), which are examples of satellite observa-tions in the frequency range above 200 GHz. The operational Odin/SMR processing covers stratospheric and mesospheric altitudes, and measures several atmospheric emission lines of water vapour and its isotopes (Urban et al., 2007). These data obtained from Odin/SMR are analyzed here to in-vestigate the climatological features and seasonal cycles of SWIs.

5

The Stratosphere-troposphere Processes And their Role in Climate (SPARC) core project of the World Climate Research Programme (WCRP) initiated the SPARC Data Initiative (SDI) in 2009. The aim of SDI is to co-ordinate an assessment of available, vertically resolved chemical tracers and aerosol observations obtained from a multi-national suite of space-based satellite instruments. The first comprehensive comparison of water vapour measurements has been performed. The quality of 13 water vapour products from 11 limb-viewing satellite instruments from the international space agencies obtained since 1978 have been assessed. A Multi-Instrument Mean (MIM) of stratospheric limb sounders has been presented (Hegglin et al., 2013; Hegglin et al., 2014). A combination of the SDI water vapour datasets, SDI-MIM, shows great potential in improving previous model-measurement comparisons based on the HALOE dataset only.

2.2 Methane oxidation in the stratosphere Methane (CH4) oxidation is an important water vapour production in the middle and upper stratosphere, where substantial amounts of SWV are pro-duced in situ through this mechanism (Bates and Nicolet, 1950; LeTexier et al., 1988). The primary corresponding chemical reactions are between met-hane and hydroxyl radical (OH), chlorine (Cl), singlet oxygen (O1D) and photons (hv). These reactions are described through the following equations (Ravishankara, 1988):

CH4 + OH à CH3 + H2O (2.1)

CH4 + Cl à HCl + CH3 (2.2)

CH4 + O(1D) à CH2 + H2O (2.3)

CH4 + hv à CH3 + H (2.4)

The product of HCl of Eq. 2.2 produces water in a second reaction:

HCl + OH à Cl + H2O (2.5)

The reaction (Eq. 2.1) with OH is the most important mechanism in the upper troposphere and lower stratosphere (UTLS). The reaction with O(1D) (Eq. 2.3) is significant in the middle stratosphere. It becomes more important than the reaction with OH in the upper stratosphere. The photolysis of met-hane (Eq. 2.4) produces CH3 and H in the upper stratosphere, and leads with further reactions to a production of water (Röckmann et al., 2003) as:

CH3 + H à CH2 + H2 (2.6)

6

CH3 à CH2 + H (2.7)

H2 + OH à H+ H2O (2.8)

H + HO2 à O + H2O (2.9)

The production of SWV through methane oxidation is therefore stronger in the upper stratosphere (Hurst et al., 2011). As shown in Figure 2.1, the maximum of H2O mixing ratio shows the water is produced at the summer pole in the upper stratosphere by methane oxidation. Here we show the re-actions that involve CH4. If methane occurs in the form of CH3D but not CH4, the reactions produce HDO. The oxygen could be any of the oxygen isotopes, e.g. 17OH, 18OH, 17O(1D) or 18O(1D), which is not highlighted here. A detailed description of the initial methane destruction reactions can be found in Bechtel and Zahn (2003).

Figure 2.1 The climatological distribution of water vapour (shading, da-ta from Odin/SMR) and temperature (contour, data from MERRA) in bo-real winter (DJF mean). The brown arrow at the tropical tropopause and the blue arrows represent the tropical upwelling and the upper branch of the BDC, respectively. The climatology is based on the period 2002–2009.

2.3 Dehydration at the tropical tropopause layer Another source of SWV is moist air mass entering the stratosphere from the tropical troposphere. Dehydration of air masses transiting the cold-point tropopause is one of the key processes driving changes in SWV (Brewer, 1949; Holton and Gettelman, 2001). It appears as the minimum of the H2O mixing ratio and the coldest temperature in the tropical lower stratosphere in Figure 2.1. The tropical tropopause temperature is highly correlated with the

Methane oxidation freshwaterproduction�

Upwelling Dehydration

7

water vapour in the lower stratosphere due to the dehydration resulting from the water vapour condensation (Fueglister et al., 2009; Dessler et al., 2014).

In addition, the effect of dehydration is demonstrated by the tropical tape recorder, which has been discovered in the tropical stratosphere by Mote et al. (1996). In the lower stratosphere (Figure 2.2), relatively dry air in boreal winter and spring alternates with moist air in boreal summer and autumn. This corresponds to cold tropopause temperatures in winter and warm tem-peratures in summer. As the air enters the rising branch of the BDC in the tropical stratosphere, it is advected upwards, conserving its signature in wa-ter vapour mixing ratio. The high water vapour mixing ratios in the tropical tropopause layer during boreal summer are associated with the enhanced convective events linked with the summer monsoon circulations (Gettelman et al., 2004). Seasonal and QBO-related variabilities affect the amplitude of the tape recorder signal, as well as its vertical velocity (Niwano et al., 2003).

Figure 2.2 Stratospheric tropical (15°S–15°N mean) water vapour tape recorder. Data are from SDI MIM observations.

MIM H2O tape recorder

Jan 06 Jan 08 Jan 10 Jan 12 Jan 14 Jan 16 Jan 18

100

10

press

ure [h

Pa]

2

3

4

5

6 10500

ppmv

8

3. Stratospheric dynamics and water vapour transport

The stratosphere is heated radiatively from within and is, for the most part, dynamically stable, as its very name suggests. It is primarily distinguished from the troposphere by its stability. However, the stratosphere is not a bor-ing zonal flow without meridional circulation. It is forced away from radia-tive equilibrium by waves propagating up from the troposphere, which trans-fer momentum and energy. Such wave-driven circulations alter the tempera-ture and, hence, the zonal wind in the stratosphere. The spatial distribution of tracers in the stratosphere is strongly affected by circulation, and hence transport. In the meantime, the distribution of tracers is also an indicator of the transport.

3.1 The Brewer-Dobson circulation (BDC) To obtain a general view of the global-scale circulation in the stratosphere, it is useful to observe transport processes in a zonally averaged sense. This meridional circulation is known as the BDC, which describes a diabatic mean mass transport, with upwelling in the tropics, poleward transport and descending in the extratropics. Brewer (1949) first proposed this circulation pattern to explain the lack of water in the stratosphere. He assumed that water vapour is “freeze dried” as it moves vertically through the cold equatorial tropopause, where dehydration occurs by condensation as a result of cold temperature. Dobson (1956) later suggested that this type of circulat-ion could also explain the observed high ozone concentrations in the tropical middle stratosphere. BDC exerts a crucial control on stratospheric tempera-ture and thereby on winds (Shepherd, 2000). It also plays a key role in the transport of water vapour, ozone and other chemical species. A changing BDC makes impact on many aspects of the stratosphere, e.g., the recovery of stratospheric ozone (Weber et al., 2011), the changes of some greenhouse gases including water vapour (Butchart and Scaife, 2001), and the exchange of mass between the stratosphere and the troposphere (Hegglin and Shepherd, 2009).

9

To estimate this zonal-mean meridional transport circulation, Andrews and McIntyre (1978) introduced the transformed Eulerian mean (TEM). The TEM residual circulation (𝑣∗, 𝑤∗) is defined as follows (Andrews et al., 1987):

(3.1)

(3.2)

where 𝑣 𝑎𝑛𝑑 𝑤 are the zonal means of the meridional and vertical veloci-ties, respectively. 𝜌!, 𝜃, 𝑎, 𝑧 𝑎𝑛𝑑 𝜙 represent air density, potential tempera-ture, the Earth’s radius, height and latitude, respectively. A bar and prime denote the zonal-mean and perturbation of it, respectively.

Figure 3.1 Schematic of the residual mean meridional circulation in the atmosphere according to Plumb (2002). The ellipse in bold indicates the Hadley circulation of the troposphere. The shaded regions (labelled “S”, “P”, and “G”) denote regions of breaking waves (synoptic- and plane-tary-scale waves, and gravity waves, respectively), responsible for driv-ing branches of the stratospheric and mesospheric circulation. The surf zone is a region of mixing of potential vorticity by planetary waves, which acts as a drag force on the mean zonal flow.

The BDC is driven by Rossby and gravity waves breaking, and saturation in the middle atmosphere (Butchart, 2014). The downward control principle

10

(Haynes et al., 1991) expresses the BDC as a response to breaking and saturating waves aloft. It means that the mean meridional circulation at any level is determined by the vertically integrated momentum forcing above that level. The gravity wave breaking in the mesosphere contributes signifi-cantly to the stratospheric circulation at high latitudes in winter (Garcia and Boville, 1994). The schematic Figure 3.1 illustrates that the residual circulat-ion of the stratosphere comprises a two-cell structure in the lower stra-tosphere, with upwelling in the tropics and subsidence in middle and high latitudes, a single cell from the tropics into the winter hemisphere at higher altitudes, and a branch from the summer pole to the winter pole in the upper stratosphere and mesosphere.

3.2 Planetary waves and stratospheric circulation As shown in Figure 3.1, the planetary-scale Rossby waves, also known as planetary waves, dominate the dynamics of the winter stratosphere. Plane-tary waves have horizontal scale of thousands of kilometers and vertical scale of several kilometers. They originate from a combination of meridional temperature gradients and the Coriolis effect. Besides, large-scale topo-graphy, like the Rocky Mountains and the Himalaya-Tibet complex, together with the meridional temperature gradients and Coriolis deflection, generate standing long Rossby waves with wavelengths up to 10,000 kilometers. These waves propagate vertically into the stratosphere, where they grow in amplitude and eventually break leading to turbulent mixing and dissipation.

The wave activity flux (WAF) is a conventional measure of the propaga-ting planetary waves in three-dimensional space from the troposphere to the stratosphere. Following Plumb (1985), the WAF vector 𝐹! defined on the sphere, is calculated in log-pressure coordinates and takes the form given by equation (3.3), which quantifies a propagating packets of planetary waves in three-dimensional space,

𝐹! = 𝑝𝑐𝑜𝑠𝜙

!!!!!"#!!

!!!

!"

!− 𝜓! !

!!!

!!!

!!!!!"#$

!!!

!"!!!

!"− 𝜓! !

!!!

!"!#!!!!"#!!!!!"#$!!

!!!

!"!!!

!"− 𝜓! !

!!!

!"!#

(3.3)

where 𝑝 is pressure, and ϕ, λ and z are latitude, longitude and height, respectively. A prime denotes perturbations of the zonal-mean field. The

11

stream function, Earth’s rotation rate, radius of the Earth, and buoyancy fre-quency are represented respectively by ψ, Ω, a, and N.

The vertical propagation is permitted when the mean winds are westerly and not very large (Charney and Drazin, 1961). The consequent feedback between waves and the mean flow determines the strong variability of the stratospheric circulation (Polvani et al., 2010). It has been recognised that the dominant planetary waves are responsible for the dramatic sudden stra-tospheric warming (SSW) events, which is the essential feature of the extra-tropical northern winter stratosphere (Schoeberl, 1978; Labitzke, 1981). The planetary waves, together with contributions from smaller-scale internal gravity waves, control almost all aspects of the extra-tropical stratospheric circulation.

Figure 3.2 Composite anomalies of the vertical component of the plane-tary wave activity flux (shading) and zonal-mean zonal wind (contour) (a, b), and temperature (shading) and the vertical velocity of residual circu-lation w* (contour) (c, d) during the period 2001–2006 (a, c) and 1991–1996 (b, d).

Figure 3.2 shows that the vertical propagation of the planetary waves is responsible for the decadal difference of circulation between the 1990s and the 2000s. During the period 2001-2006, a strong easterly anomaly in zonal wind is obtained across the depth of the stratosphere (Figure 3.2a); the ano-

12

maly in temperature is warm in the lower and middle stratosphere, and it is cold in the upper stratosphere (Figure 3.2c). This is due to a strong upward propagation of planetary waves in high latitudes of the Northern Hemisphere (NH) (Figure 3.2a). As typically observed during SSW events, warming anomalies in the lower and middle stratosphere are generally accompanied by cooling anomalies above them (Andrews et al., 1987; Holton and Hakim, 2012), which are caused by the upward anomaly motion (or significantly weakened downward motion) in the polar upper stratosphere (Figure 3.2c). In comparison, the period 1991–1996, associated with a weak planetary wave activity, shows the opposite picture in the winter polar region (Figure 3.2b, d).

Andrews et al. (1987) provide the theoretical background for a planetary wave influence on the thermodynamic structure of the stratosphere. The planetary waves break and their easterly angular momentum are released in the stratosphere. This process induces a deceleration of the westerlies and/or a poleward meridional circulation. Thus, the mean westerly wind (the polar night jet) is weakened, leading to more planetary waves propagating upward. If this process continues the wind eventually reverses, forming a critical layer where the mean zonal wind matches the wave phase speed. This would inhibit further upward wave propagation. The wave breaking would be in-tensified, inducing an easterly wind up to the level reached by the propaga-ting wave. A rapid change to easterlies then takes place and the critical layer descends. Eventually, the polar night jet is replaced by easterlies from the upper stratosphere down to the lower stratosphere and upper troposphere. This process is schematised in Figure 3.3.

Figure 3.3 Schematic illustration of the interaction between upward propagating planetary waves and the mean flow in the stratosphere (adapted from Andrews et al., 1987).

F•∇

A

WaveActivity

13

As a consequence of this process involving strong planetary wave acti-vity, the signal of seasonal change from winter to summer occurs first in the upper stratosphere, then propagates downward to the lower stratosphere and eventually reaches the upper troposphere. This transition is apparent in the Southern Hemisphere (SH) in spring, as shown in Figure 3.4b: the strong upward propagation of planetary waves penetrates deep into the stratosphere in austral spring, causing the zonal wind reversal initiated from the stratopa-use, which then propagates downward to the lower stratosphere. This is re-flected by the zero contour of zonal-mean zonal wind, which is tilted from the upper to the lower stratosphere. During austral winter, the upward propa-gation of planetary waves is weak, because the zonal wind is too strong to satisfy the Charney-Drazin condition (Charney and Drazin, 1961; Holton and Hakim, 2012).

Figure 3.4 Seasonal cycle of the zonal-mean WAFz (shading) and zonal wind (contour, zonal winds equal zero and greater than 30 m/s are shown) in the atmosphere over (a) 60°N–85°N and (b) 60°S–85°S. Data are from ERA-Interim (1979–2017).

Meanwhile, for the NH (Figure 3.4a), the maximum of the vertical wave activity flux above 200 hPa is in winter, which is related to the SSW events. Therefore, the SSW events occur more often in boreal winter than in austral winter. The upward propagation of planetary waves is not so strong in boreal

14

spring, resulting in a non-titled zero contour of zonal wind, which means that the seasonal transition occurs simultaneously across the stratosphere.

In summary, the inter-hemispheric difference in stratospheric circulation (in winter and spring) is basically determined by the seasonal difference of the planetary wave activity between the two hemispheres.

3.3 Water vapour transport and seasonality The residual circulation can be employed to illustrate water vapour transport. As shown in Figure 3.5, there is an upwelling from the tropical tropopause, followed by the poleward transport to the extratropics in the lower stra-tosphere. In the upper stratosphere and mesosphere, the inter-hemispheric transport from the summer to the winter hemisphere forms the upper branch of the BDC (see also Figure 3.1). Water vapour is transported to the winter pole and descends there, resulting in a large water vapour content in the win-ter pole of the middle stratosphere. The subsidence in DJF (Figure 3.5a) is closer to the winter pole compared to that in JJA (Figure 3.5b). This is due to the fact that the polar vortex in austral winter is much stronger than that in boreal winter. It is also seen from Figure 3.4, where the zonal wind of 60°S–85°S mean in JJA is much stronger than that of 60°N–85°N mean in DJF. The poleward transport is inhibited from penetrating the strong polar vortex.

Figure 3.5 Seasonal mean of SWV (shading) and residual circulation (stream) for, (a) the boreal winter (DJF) mean and (b) the austral winter (JJA) mean. Data are from ERA5 (1979–2017).

The seasonal cycle of tropical SWV shows a well-known phenomenon of the tropical tape recorder (Figure 3.6a). It depicts the wet and dry signals propagating upward from the tropopause to about 30 km. This is observed in Figure 2.2 and in other satellite observations as well as in model simulations

15

(Mote et al., 1996; Steinwagner et al., 2010; Hegglin et al., 2013; Eichinger et al., 2015; Wang et al., 2018). Over the south polar region (Figure 3.6b), the anomaly shows a distinct wet summer and dry winter contrast in the up-per stratosphere, resulting from enhanced methane oxidation in the persi-stently sunlit summer stratosphere. This feature, along with a similar ano-maly pattern in the lower stratosphere (below 20 km) and the opposite ano-maly pattern in the middle stratosphere, constitutes a triple vertical structure over the south polar region. Over the north polar region (Figure 3.6c), the seasonal evolution of SWV resembles a mix of the former two. The down-ward propagation in the upper stratosphere is similar to that of south polar region, save for the upward propagation in the lower stratosphere more follows the tape recorder in the lower tropical stratosphere.

Figure 3.6 Seasonal cycle of SWV anomalies from the climatological an-nual mean for, (a) the tropics (15°S–15°N), (b) the south polar region (60°S–85°S) and (c) the north polar region (60°S–85°S). Data are from ERA5.

16

4. Variability of stratospheric water vapour

The interannual and long-term variabilities in SWV are also modulated by changes in processes related to the sources of SWV and stratospheric trans-port. The multi-timescale variations of SWV are investigated using recent observational and reanalyses data. The changes of SWV in a warming cli-mate are simulated in climate modles.

4.1 Stratospheric oscillations The interannual variability of the equatorial stratosphere (16–50km) is dom-inated by the quasi-biennial oscillation (QBO) (Baldwin et al., 2001). QBO is observed as downward propagation of easterly and westerly wind regimes (Figure 4.1), with a variable period averaging approximately 28 months. It has been known that QBO modulates the interannual variations in tropical tropopause temperature (Randel et al., 1998, 2004). The contribution of QBO to the tropopause temperature variation can go up to 0.5 K. These changes could result in a significant modulation of the water vapour content that enter the stratosphere. The downward propagation of the QBO is also observed in the tropical temperature and vertical velocity of the residual circulation (see Fig. 7 in paper III), indicating that QBO would further mo-dulate SWV through QBO-induced changes in temperature and circulation.

Figure 4.1 Interannual variability of the tropical zonal-mean zonal wind. Data are from ERA5.

17

On decadal timescale, the North Atlantic Oscillation (NAO) is known to be the dominant mode of variability of the NH atmosphere (Hurrell, 1995). The NAO is particularly important in boreal winter, when it exerts a strong control on the climate of the NH (Osborn, 2011). This is also the season that exhibits strong inter-decadal variability (Osborn, 2004). It has been shown that the stratospheric component of the NAO is responsible for a large part of the stratospheric variability (Ambaum and Hoskins, 2002). An increase in the NAO index is normally associated with a stronger stratospheric vortex, as a result of increased equatorward refraction of upward-propagating plane-tary waves (Ambaum and Hoskins, 2002). As shown in Figure 4.2, the NAO index shows multiple timescales ranging from interannual to multi-decadal. On decadal scale, it is positive in most years of the 1990s, and negative in the 2000s. This decadal change is also seen in the variability of SWV.

Figure 4.2 The winter NAO index based on the difference of normalized sea level pressure between Lisbon, Portugal and Stykkisholmur / Reykja-vik, Iceland since 1864. The SLP values at each station were normalized by removing the long-term mean and by dividing by the long-term stand-ard deviation. Both the long-term means and standard deviations are based on the period 1864–1983. Normalization is used to avoid the series being dominated by the greater variability of the northern station. (Cli-mate Data Guide; A. Phillips)

4.2 Variability of SWV in observations and reanalyses Variability of water vapour entering the stratosphere through the tropical tropopause is largely dominated by the upward transport and the tropopause temperature. The variability of tropical SWV (deseasonalized) shows that the anomalies propagate upward in the lower stratosphere, which is similar

18

to the tape recorder of the seasonal cycle (Figure 4.3). In the upper stra-tosphere above 30 km, there is no vertical propagation and the phase is nearly constant with height. In addition to the interannual variability that bears a signature of the QBO, a decadal signal with positive anomalies during the 1980s and 1990s and negative in the 2000s is observed (Figure 4.3a, b). The time series of the mean tropical SWV in the lower stratosphere 15-20km from ERA5 and SDI MIM are consistent with a high correlation coefficient of 0.85 (over the common period), indicating that the ERA5 data represents the SWV variability very well. The power spectrum of the time series shows a pronounced 2-year peak which is related to the tropical QBO, and a 22-year peak which is probably related to low-frequency oscillation originating from high latitudes such as the NAO.

Figure 4.3 Tropical SWV anomalies (15°S–15°N mean) relative to the monthly climatology using (a, b) ERA5, and (c) the SDI MIM. The panels (b, c) are based on the 15–20km mean. Unit: 0.1 ppm.

The residual circulation anomaly corresponding to QBO westerly phase is shown in Figure 4.4a. It exhibits a rising motion from the equatorial tropo-pause and a poleward divergence from the tropics to the subtropics. Unlike the upwelling in the tropics associated with the general stratospheric BDC, the QBO-induced motion in the vertical direction is different at different levels. Around 30 km, there is divergence with upwelling above, and downwelling below the westerlies. At around 20 km, there is convergence with downwelling above, and upwelling below the easterlies (also see Fig.14

19

in paper III). This residual circulation anomaly results in a higher and colder tropical tropopause, and less water vapour entering the tropical stratosphere. Along the poleward motion in the lower stratosphere the global water vapour anomaly is negative. The opposite takes place during the QBO easterly phase (Figure 4.4b).

Figure 4.4 Composite anomalies of SWV (shading) and the residual cir-culation (vectors and stream), associated with QBO phases (a, b) and de-cadal NAO phases (c, d), respectively. The decadal periods are over 2001–2006 and 1991–1996. Data are from ERA5.

On decadal timescale, during the low SWV period 2001–2006 (negative NAO), the water vapour anomaly is negative almost in the entire stra-tosphere (Figure 4.4c). Meanwhile, as shown in Figure 3.2(a, c), with strong upward propagation of planetary waves in NH high latitudes, there is an easterly anomaly across the depth of the stratosphere, a warm temperature anomaly in the lower and middle stratosphere and a cold anomaly in the upper stratosphere. This is concurrent with a weakening of the polar vortex, which is favourable for downwelling in the lower stratosphere. In this case the BDC is enhanced, with a stronger winter-poleward flow in the upper stratosphere (Figure 4.4c). The enhanced poleward flow brings a stronger tropical upwelling and poleward divergence into the lower stratosphere. The stronger tropical upwelling results in a colder tropical tropopause, which leads to cold temperature anomalies in the lower tropical and subtropical stratosphere, and further results in dehydration and less water vapour in the

20

stratosphere. In comparison, the high SWV period 1991–1996 shows preci-sely the opposite picture, especially in the winter polar region (Figure 4.4d).

The SST composite analyses of these two periods show a significant anomaly over the North Atlantic, with a positive anomaly during 2001–2006 and a negative anomaly during 1991–1996 (see Fig.13 in paper III). The correlation between the tropical SWV and the North Atlantic SST, on deca-dal timescales is -0.7, and it is similarly correlated with the NAO index. One possible mechanism for the influence of the North Atlantic SST on the SWV can be explained by the fact that the warmer North Atlantic (associated with a negative NAO-like anomaly) in winter enhances the thermal contrast between land and sea. This may force stronger planetary wave activity, which in turn affects the thermal and dynamical conditions in the stra-tosphere (warmer high-latitude stratosphere in the NH and stronger BDC). This mechanism is consistent with the study of Omrani et al. (2014).

4.3 Response of SWV to a changing climate The climate change response in the stratosphere has been a hot field in recent years. In order to increase confidence in climate change projections, we need to better understand the physical mechanisms causing changes in the SWV as well as its possible feedback on the changing climate.

With the global high-top atmosphere model WACCM (Marsh et al., 2013), the responses of SWV to a CO2 doubling scenario is investigated (Figure 4.5). The radiative photochemical and dynamical responses have been separated by considering individual response to changes in CO2 con-centration and changes in SSTs. The warm SSTs is induced by CO2 doubling in a coupled model. The doubled CO2 only forcing leads to SWV decrease by ~2–6% in most of the stratosphere, and by more than 10% in the two polar regions of the lower stratosphere. The warm SSTs forcing results in SWV increase by ~6–10% in most of the stratosphere, and by more than 10% in the lower stratosphere, especially in the tropics. The SWV response to the combined CO2 doubling and warm SSTs is approximately equal to the sum of the two former responses, increase by ~6–10% in most of the stra-tosphere, and decrease by ~2–6% in the two polar regions in the lower stra-tosphere. This means that the CO2 doubling influences the SWV in two ways, namely, (i) radiatively through in situ changes associated with changes in CO2 or (ii) dynamically through changes in stratospheric wave forcing. In reality, these two effects may not be of equal magnitude and it is likely that changes in the stratospheric response are primarily a result of changing the

21

SSTs (Shepherd, 2008). This is indeed shown in WACCM simulations. The effect of warm SSTs dominates the change of SWV in most of the stra-tosphere, except for the two polar regions in the lower stratosphere, where the CO2 doubling induces a strong decrease in SWV. Previous studies have shown that the lower stratospheric cooling associated with CO2 doubling may increase the probability of polar stratospheric cloud (PSC) formation (e.g., Pitari et al., 1992). An increase in PSC formation in this area yields a decrease in water vapour.

Figure 4.5 Comparison of change (from the control run) in seasonal mean SWV in climate change experiments of combined doubled CO2 and warm SSTs (a, d), doubled CO2 only (b, e) and warm SSTs only (c, f), for DJF mean (a–c) and JJA mean (d–f). The shading indicates significance at the 5% level.

The SWV increase is mainly associated with the warm SSTs, which al-lows for a warmer, and hence a moister troposphere (by the Clausius-Clapeyron relation). This provides a good source for the water vapour re-aching the stratosphere. Besides, a warmer tropopause appears over the reg-ion of 30°S–30°N in warm SSTs experiments (Figure 4.6), which in turn weakens the condensation at the tropopause and allows for more water vapour to enter the stratosphere. Consequently, more water vapour is trans-ported upward to the middle and upper stratosphere and poleward to the high-latitudes of both hemispheres through the BDC, resulting in more water vapour in the entire stratosphere. The tropical and subtropical tropopause is colder in doubled CO2 only experiment than in control experiment, resulting in less water vapour entering the stratosphere. Eventually, the SST effect on

22

SWV offsets the CO2 effect and leads to an increase of water vapour in the stratosphere.

Figure 4.6 Comparison of change (from the control run) in tropopause temperature in the climate change experiments of combined doubled CO2 and warm SSTs (black), doubled CO2 only (blue) and warm SSTs only (red), for DJF (a) and JJA (b) means.

Figure 4.7 shows that the stratospheric temperature changes is dominated by the doubled CO2 forcing, except for the tropical lower stratosphere, where a warming anomaly emerges due to warm SSTs. The SWV increases greatly in the lower stratosphere in the warm SSTs experiment (Figure 4.5c, f), which is related to the high tropical and subtropical tropopause temperature. And in turn, as a greenhouse gas, the increased water vapour results in war-ming the lower stratosphere, namely, there is a positive feedback between SWV and the temperature in the tropical and subtropical lower stratosphere.

Figure 4.7 Same as Figure 4.5, but for temperature changes.

23

5. Summary and outlook

5.1 Summary of papers included in this thesis

Paper I

Using eight years (2002–2009) of retrievals from Odin/SMR, this paper pre-sents the global climatological features of three SWIs in the stratosphere. Vertically the SWIs increase with altitude, which is attributed to stratospher-ic methane oxidation. In the tropics, highly depleted water isotopes in the lower stratosphere indicate the effect of dehydration when the air enters through the cold tropopause. At higher latitudes more enriched water iso-topes in the upper stratosphere during summer are produced and transported to the other hemisphere via the Brewer–Dobson circulation. The tape record-er in H2O is observed in the lower tropical stratosphere, and a pronounced downward propagating seasonal signal in SWIs is observed from the upper to the lower stratosphere over the polar regions. These observed features in water isotopes are used to evaluate the SWI-enabled models. A few possible causes of model deficiencies in reproducing the main seasonal features of SWI in the stratosphere have been identified. For instance, choosing a better advection scheme and including the process of methane oxidation in the model, can immediately improve the model’s representation of SWV fea-tures.

Paper II

This paper investigates observational features of the annual cycle and sea-sonal transition in the high-latitude stratosphere by using the ERA-interim reanalysis. Climatologically, the tropospheric planetary waves propagate to the stratosphere and affect significantly the winter-to-summer stratospheric seasonal transition over both hemispheres, with a much stronger planetary wave activity in austral spring than its boreal counterpart. The austral spring seasonal transition occurs first at the stratopause then propagates downward to the lower stratosphere due to enhanced planetary wave breaking, which weaken the westerlies. In boreal spring, the seasonal transition occurs simul-

24

taneously across the depth of the stratosphere, mainly due to the solar radi-ation and weaker planetary wave activity. The characteristics of the seasonal transition exhibit hemispheric differences. Before the transition, stronger planetary wave activity leads to asynchronous rather than synchronous tran-sition in the NH, which propagates downward from the stratopause. In the SH, a moderate rather than steep transition is obtained, which occurs earlier and takes longer to propagate from the upper to the lower stratosphere. The timing of stratospheric seasonal transition is closely linked to the intensity of upward propagation of planetary wave activity, i.e. the stronger the upwel-ling planetary wave activity in spring the earlier the stratospheric seasonal transition.

Paper III

This study investigates the dynamic processes that determine SWV on inte-rannual to decadal timescales. First we evaluate two SWV reanalysis pro-ducts and show that SWV is better represented in the new generation reana-lysis, ERA5, than in the former ERA-Interim product. In particular, it is shown that ERA5 SWV is highly consistent with observational data obtained from the SDI MIM. The interannual variability in the lower tropical SWV is closely linked to the tropical QBO. When the QBO westerly phase occupies the middle stratosphere, a decrease in water vapour is observed in the lower stratosphere due to a colder tropical tropopause and a QBO-induced en-hanced residual circulation. The opposite is found for the QBO easterly phase. On decadal time scale, the SWV variability is related to North Atlan-tic SST anomaly. The warm North Atlantic SST anomaly leads to stronger upward propagation of planetary waves in the northern high latitudes, resul-ting in a warmer pole in the lower stratosphere, weaker polar vortex, en-hanced residual circulation and colder tropical tropopause, and leads to furt-her less SWV. The opposite occurs during the cold North Atlantic SST ano-maly period.

Paper IV

This study performs sensitivity experiments with WACCM to investigate the SWV response to a CO2 doubling scenario, its possible feedback on the stra-tospheric temperature and relevant mechanisms. The CO2 concentration and SSTs are changed together, as well as separately, to separate the radiative-photochemical and dynamical response. The response of SWV is dominated by the warm SSTs, with an increase by about ~6–10% in most of the stra-tosphere and more than 10% in the lower stratosphere, except for the winter

25

pole in the lower stratosphere where the doubled CO2 increases the proba-bility of polar stratospheric cloud formation and hence decreases water vapour. The increase of SWV is obtained mainly through a dynamical re-sponse to the warm SSTs, which leads to a moist troposphere and a warmer tropical and subtropical tropopause, and results in more water vapour ente-ring the stratosphere from below. A large increase of SWV in the lower stra-tosphere in turn affects the stratospheric temperature, and results in a war-ming in the tropical and subtropical lower stratosphere, which offsets the cooling caused by CO2 doubling in general.

5.2 Outlook The observations, reanalyses data and climate model simulations provide valuable information on studying the seasonality and variability of SWV and the involved mechanisms. This thesis has focused on seasonal, interannual and decadal variabilities of SWV, and how the SWV changes in a warming climate. Nevertheless, many interesting questions arose from this thesis, which need further investigations.

In paper III, we have investigated the effects of the QBO on the inte-rannual variability of SWV, and the influence of NAO-associated planetary wave activity (and stratospheric warming) on the decadal variability of SWV during boreal winter. Some studies show that the QBO modulates the deca-dal oscillation associated with the 11-year solar cycle in the stratosphere (Labitzke and van Loon, 1995), and that the effect of the QBO on the nort-hern stratospheric polar vortex yields decadal changes (Lu et al., 2008). Here, we reveal that the QBO is different during the two decadal periods, and that the tropical SWV anomaly pattern in the 2000s is more connected to the QBO pattern. More questions remain to be answered and need further investigations, namely (i) what is the connection between QBO and the decadal oscillation? And, (ii) what is the precise dynamical mechanism of the QBO impact on the SWV decadal variability? For the decadal variability in SWV associated with the SST anomaly in the North Atlantic, we ex-plained that SST anomalies influence the planetary wave activity, thus affecting the residual circulation and the temperature distribution in the stra-tosphere, which further affect the SWV transport and distribution. This mechanism needs to be verified by model experiments with prescribed SST anomalies.

The tropical SWV shows an abrupt drop at the end of the 1990s (Figure 4.3). This finding is consistent with observations from previous works

26

(Wang et al., 2017; Solomon et al., 2010; Ding and Fu, 2018). We show that the amount of tropical SWV is related to the tropical upwelling, and suggest that the tropical SST may contribute to the abundance of water vapour brought to the stratosphere via convection in the troposphere, as well as the tropopause temperature. However, whether the tropical SST anomaly deter-mines tropical SWV is not entirely clear. Solomon et al. (2010) show that the tropical SWV is negatively correlated with SST of the warm pool in the Pa-cific. Ding and Fu (2018) suggest that the tropical central Pacific SST could contribute a step-wise drop of the lower SWV from 1992–2000 to 2001–2005. More detailed investigation is required.

In paper IV, the model experiments have shown that SWV response to CO2 doubling is dominated by warm SSTs, which is induced by CO2 dou-bling in the coupled model simulation. It would be interesting to investigate the response of SWV to low atmospheric CO2 concentration, such as those that happened during the Last Glacier Maximum. This would improve our understanding of the response and feedback of SWV to different climate forcings.

27

Acknowledgements

First of all, I would like to thank my supervisors Abdel Hannachi and Qiong Zhang for giving me the opportunity to work with them. Thanks for all your support during the last four years. Thanks to Abdel for sharing his know-ledge and pushing me to present my work to other colleagues, which helped me overpowering my procrastination. Thanks to Qiong for inviting me to visit her in 2014, and providing me another choice of my career. Thanks for her patience, encouragement and understanding during these years.

Further, I would like to thank my committee members Jörg Gumbel and Bodil Karlsson for making sure I was on track during this period. Thanks to Léon Chafik for jumping in as a committee member and for his concern and support.

I would also like to thank all co-authors of the papers included in this thesis, for all their discussions, constructive comments and revisions. These papers wound not have been completed without their help. Thanks to Jenson Zhang and Qiang Li for their technical support. Many thanks to the climate model-ling group: Charlotta, Maartje, Ellen, Zixuan and Josefine, for sharing every step of our research. Thanks to Lena and Xiangyu for proofreading my kappa. Thanks to Lina for checking the sammanfattning of my kappa.

I really enjoyed my time at MISU. We have such an awesome research at-mosphere, super nice people and wonderful traditional activities here. I wish to extend a huge thank you to everyone at MISU. Some special words of gratitude to Waheed, Lena and Etienne for being my amazing officemates, and thanks to Aitor, Maartje and Xiangyu for all the dinners and cozy mo-ments we spent together. I also like to thank Malin, Anna, Lina, Evelyne, Koen, Marin, Wing, Erik, Dipanjan, Qin and Cheng for running, cycling, playing badminton and all the interesting conversations. Thanks to Friderike for sharing the office with me in the last couple of months and helping me with the final steps of the thesis. Thanks also to my friend Qian for sharing my sadness and happiness, and for all her useful advice.

Finally, I would like to thank my family. Thanks to my parents, my husband and my daughter, for always being there and supporting me. You are the softest part of my heart which makes me strong.

28

References

Ambaum MH, Hoskins BJ (2002) The NAO troposphere–stratosphere connection. Journal of Climate, 15(14), 1969-1978.

Andrews AE, Boering KA, Daube BC, Wofsy SC, Hintsa EJ, Weinstock EM, and Bui TP (1999) Empirical age spectra for the lower tropical stratosphere from in situ observations of CO2: Implications for stratospheric transport. Journal of Geophysical Research: Atmospheres, 104(D21), 26581-26595.

Andrews DG, McIntyre ME (1978) Generalized Eliassen-Palm and Charney-Drazin theorems for waves on axismmetric mean flows in compressible atmospheres. Journal of the Atmospheric Sciences, 35(2), 175-185.

Andrews DG, Holton JR, Leovy CB (1987) Middle Atmosphere Dynamics (Vol. 40). Academic Press, Orland, 489 pp

Baldwin MP, Gray LJ, Dunkerton TJ, Hamilton K, Haynes PH et al (2001) The quasi-biennial oscillation. Reviews of Geophysics 39(2): 179-229

Bates DR, Nicolet M (1950) The photochemistry of atmospheric water vapor. Journal of Geophysical Research, 55(3), 301-327.

Bechtel C, Zahn A (2003) The isotope composition of water vapour: A powerful tool to study transport and chemistry of middle atmospheric water vapour.

Brewer AW (1949) Evidence for a world circulation provided by the measurements of helium and water vapour distribution in the stratosphere. Quart. J. Roy. Meteor. Soc. 75(326): 351-363

Butchart N, Scaife AA (2001) Removal of chlorofluorocarbons by increased mass exchange between the stratosphere and troposphere in a changing climate. Nature, 410(6830), 799.

Butchart N (2014) The Brewer‐Dobson circulation. Reviews of geophysics, 52(2), 157-184.

Charney JG, Drazin PG (1961) Propagation of planetary-scale disturbances from the lower into the upper atmosphere. J. Geophys. Res. 66, 83–109

Dessler AE, Schoeberl MR, Wang T, Davis SM, Rosenlof KH (2013) Stratospheric water vapor feedback. Proceedings of the National Academy of Sciences 110(45): 18087-18091

Dessler, AE, Schoeberl MR, Wang T, Davis SM, Rosenlof KH, Vernier JP (2014) Variations of stratospheric water vapor over the past three decades. J. Geophys. Atmos. 119(22): 12-588

Ding Q, Fu Q (2018) A warming tropical central Pacific dries the lower stratosphere. Climate dynamics, 50(7-8), 2813-2827.

29

Dobson GMB (1956) Origin and distribution of the polyatomic molecules in the atmosphere. Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences, 236(1205), 187-193.

Dvortsov VL, Solomon S (2001) Response of the stratospheric temperatures and ozone to past and future increases in stratospheric humidity. J. Geophys. Res. Atmos. 106(D7): 7505-7514

Eichinger R, Jöckel P, Lossow S (2015) Simulation of the isotopic composition of stratospheric water vapour–Part 2: Investigation of HDO/H2O variations. Atmos. Chem. and Phys. 15(12): 7003-7015

Fomichev, V. I., Jonsson, A. I., De Grandpre, J., Beagley, S. R., McLandress, C., Semeniuk, K., and Shepherd, T. G. (2007) Response of the middle atmosphere to CO2 doubling: Results from the Canadian Middle Atmosphere Model. Journal of Climate, 20(7), 1121-1144.

Forster PMdF, Shine KP (1999) Stratospheric water vapour changes as a possible contributor to observed stratospheric cooling. Geophys. Res. Lett. 1999, 26.21: 3309-3312

Forster PMdF, Shine KP (2002) Assessing the climate impact of trends in stratospheric water vapor. Geophys. Res. Lett. 29(6): 10-1

Fueglistaler S, Dessler AE, Dunkerton TJ, Folkins I, Fu Q, Mote PW (2009) Tropical tropopause layer. Reviews of Geophysics 47(1)

Garcia RR, Boville BA (1994) “Downward control” of the mean meridional circulation and temperature distribution of the polar winter stratosphere. Journal of the atmospheric sciences, 51(15), 2238-2245.

Gettelman A, Kinnison DE, Dunkerton TJ, Brasseur GP (2004) Impact of monsoon circulations on the upper troposphere and lower stratosphere. Journal of Geophysical Research: Atmospheres, 109(D22).

Gille JC, Bailey PL, Russell JM, Russell J M (1980) Temperature and composition measurements from the LRIR and LIMS experiments on Nimbus 6 and 7. Philosophical Transactions of the Royal Society of London. Series A, Mathematical and Physical Sciences, 296(1418), 205-218.

Haynes PH, McIntyre ME, Shepherd TG, Marks CJ, Shine KP (1991) On the “downward control” of extratropical diabatic circulations by eddy-induced mean zonal forces. Journal of the Atmospheric Sciences, 48(4), 651-678.

Hegglin MI, Shepherd TG (2009) Large climate-induced changes in ultraviolet index and stratosphere-to-troposphere ozone flux. Nature Geo, 2(10), 687.

Hegglin MI, Tegtmeier S, Anderson J, Froidevaux L, Fuller R et al (2013) SPARC Data Initiative: Comparison of water vapor climatologies from international satellite limb sounders. J. Geophys. Res. Atmos. 118(20): 11-824

Hegglin MI, Plummer DA, Shepherd TG, Scinocca JF, Anderson J et al (2014) Vertical structure of stratospheric water vapour trends derived from merged satellite data. Nat. Geosci. 7(10): 768

Holton JR, Gettelman A (2001) Horizontal transport and the dehydration of the stratosphere. Geophysical Research Letters, 28(14), 2799-2802.

30

Holton JR, Hakim GJ (2012) An Introduction to Dynamic Meteorology. Vol. 88. Elsevier Academic Press, Waltham, MA, 552 pp

Holton JR, Tan HC (1980) The influence of the equatorial quasi-biennial oscillation on the global circulation at 50 mb. J. Atmos. Sci. 37(10): 2200-2208

Holton JR, Haynes PH, McIntyre ME, Douglass AR, Rood RB, Pfister L (1995) Stratosphere-troposphere exchange. Rev. Geophys. 33(4): 403-439

Hoskins BJ (1991) Towards a PV-θ view of the general circulation. Tellus A 43(4): 27-36

Hurrell JW (1995) Decadal trends in the North Atlantic Oscillation: regional temperatures and precipitation. Science, 269(5224), 676-679.

Jones PD, Jónsson T, Wheeler D (1997) Extension to the North Atlantic Oscillation using early instrumental pressure observations from Gibraltar and South-West Iceland. Int. J. Climatol. 17,1433-1450. doi: 10.1002/(SICI)1097-0088(19971115) 17:13<1433::AID-JOC203>3.0.CO;2-P

Khaykin S, Pommereau J P, Korshunov L, Yushkov V, Nielsen J, Larsen N, Williams E (2009) Hydration of the lower stratosphere by ice crystal geysers over land convective systems. Atmospheric Chemistry and Physics, 9(6), 2275-2287.

Labitzke K (1981) Stratospheric‐mesospheric midwinter disturbances: A summary of observed characteristics. Journal of Geophysical Research: Oceans, 86(C10), 9665-9678.

Labitzke K, Van Loon H (1995) Connection between the troposphere and stratosphere on a decadal scale. Tellus A 47(2): 275-286

Le Texier H, Solomon S, Garcia RR (1988) The role of molecular hydrogen and methane oxidation in the water vapour budget of the stratosphere. Quarterly Journal of the Royal Meteorological Society, 114(480), 281-295.

Lu H, Baldwin MP, Gray LJ, Jarvis MJ (2008) Decadal-scale changes in the effect of the QBO on the northern stratospheric polar vortex. J. Geophys. Res. Atmos. 113(D10)

Marsh DR, Mills MJ, Kinnison DE, Lamarque JF, Calvo N, Polvani LM (2013) Climate change from 1850 to 2005 simulated in CESM1 (WACCM), J. Climate, 26, 7372–7391, doi:10.1175/JCLI-D-12-00558.1.

Mote PW, Rosenlof KH, McIntyre ME, Carr ES, Gille JC et al (1996) An atmospheric tape recorder: The imprint of tropical tropopause temperatures on stratospheric water vapor. J. Geophys. Res. Atmos. 101(D2): 3989-4006

Naito Y, Taguchi M, Yoden S (2003) A parameter sweep experiment on the effects of the equatorial QBO on stratospheric sudden warming events. J. Atmos. Sci. 60(11): 1380-1394

Omrani NE, Keenlyside NS, Bader J, Manzini E (2014) Stratosphere key for wintertime atmospheric response to warm Atlantic decadal conditions. Clim. Dyn. 42(3-4): 649-663

Osborn TJ (2004) Simulating the winter North Atlantic Oscillation: the roles of internal variability and greenhouse gas forcing. Climate Dynamics, 22(6-7), 605-

31

623.

Osborn TJ (2011) Variability and changes in the North Atlantic Oscillation index. In Hydrological, socioeconomic and ecological impacts of the North Atlantic Oscillation in the Mediterranean region (pp. 9-22). Springer, Dordrecht.

Pitari G, Palermi S, Visconti G, Prinn RG (1992) Ozone response to a CO2 doubling: Results from a stratospheric circulation model with heterogeneous chemistry. Journal of Geophysical Research: Atmospheres, 97(D5), 5953-5962.

Plumb RA (1985) On the three-dimensional propagation of stationary waves. J. Atmos. Sci. 42(3): 217-229

Plumb RA (2002) Stratospheric transport. Journal of the Meteorological Society of Japan Ser. II, 80(4B): 793-809

Plumb RA, Bell RC (1982) A model of the quasi-biennial oscillation on an equatorial beta-plane. Quart. J. Roy. Meteor. Soc. 108: 335-352

Pumphrey, HC, Boone C, Walker KA, Bernath P, Livesey NJ (2008) Tropical tape recorder observed in HCN. Geophysical Research Letters, 35(5).

Polvani LM, Sobel AH, Waugh DW (2010) The stratosphere: Dynamics, transport, and chemistry pp.196

Randel WJ, Wu F, Russell JM, Roche A, Waters JW (1998) Seasonal cycles and QBO variations in stratospheric CH4 and H2O observed in UARS HALOE data. Journal of the Atmospheric Sciences, 55(2), 163-185.

Randel WJ, Wu F, Oltmans SJ, Rosenlof K, Nedoluha GE (2004) Interannual changes of stratospheric water vapor and correlations with tropical tropopause temperatures. Journal of the Atmospheric Sciences, 61(17), 2133-2148.

Riese M, Ploeger F, Rap A, Vogel B, Konopka P et al (2012) Impact of uncertainties in atmospheric mixing on simulated UTLS composition and related radiative effects. J. Geophys. Res. Atmos. 117(D16)

Ravishankara AR (1988) Kinetics of radical reactions in the atmospheric oxidation of CH4. Annual Review of Physical Chemistry, 39(1), 367-394.

Röckmann T, Rhee TS, Engel A (2003) Heavy hydrogen in the stratosphere. Atmospheric Chemistry and Physics, 3(6), 2015-2023.

Schoeberl MR (1978) Stratospheric warmings: Observations and theory. Reviews of Geophysics, 16(4), 521-538.

Schoeberl MR, Duncan BN, Douglass AR, Waters J, Livesey N, Read W, Filipiak M (2006) The carbon monoxide tape recorder. Geophysical research letters, 33(12).

Shepherd TG, Koshyk JN, Ngan K (2000) On the nature of large-scale mixing in the stratosphere and mesosphere. Journal of Geophysical Research: Atmospheres, 105(D10), 12433-12446.

Shepherd TG (2007) Transport in the middle atmosphere. Journal of the Meteorological Society of Japan. Ser. II, 85, 165-191.

Shepherd TG (2008) Dynamics, stratospheric ozone and climate change, Atmosphere-Ocean, 46,1, 117-138, doi:10.3137/ao.460106

32

Solomon S, Rosenlof KH, Portmann RW, Daniel JS, Davis SM et al (2010) Contributions of stratospheric water vapor to decadal changes in the rate of global warming. Science 327(5970): 1219-1223

Steinwagner J, Fueglistaler S, Stiller G, Von Clarmann T, Kiefer M et al (2010) Tropical dehydration processes constrained by the seasonality of stratospheric deuterated water. Nat. Geosci. 3(4), 262

Stenke A, Grewe V (2005) Simulation of stratospheric water vapor trends: impact on stratospheric ozone chemistry. Atmospheric Chemistry and Physics, 5(5), 1257-1272.

Vogel B, Feck T, Grooß JU (2011) Impact of stratospheric water vapor enhancements caused by CH4 and H2O increase on polar ozone loss, J. Geophys. Res. Atmos. 116

Wang T, Zhang Q, Lossow S, Chafik L, Risi C, Murtagh D, Hannachi A (2018) Stable Water Isotopologues in the Stratosphere Retrieved from Odin/SMR Measurements. Remote Sensing 10(2): 166

Wang T, Zhang Q, Hannachi A, Lin Y, Hirooka T (2019) On the dynamics of the spring seasonal transition in the two hemispheric high-latitude stratosphere. Tellus A 71(1): 1-18

Wang Y, Su H, Jiang JH, Livesey NJ, Santee ML et al (2017) The linkage between stratospheric water vapor and surface temperature in an observation-constrained coupled general circulation model. Clim. Dyn. 48(7-8): 2671-2683

Waters JW, Froidevaux L, Harwood RS, Jarnot RF, Pickett HM, Read WG, Holden JR (2006) The earth observing system microwave limb sounder (EOS MLS) on the Aura satellite. IEEE Transactions on Geoscience and Remote Sensing, 44(5), 1075-1092.

Weber M, Dikty S, Burrows JP, Garny H, Dameris M, Kubin A, Langematz U (2011) The Brewer-Dobson circulation and total ozone from seasonal to decadal time scales. Atmospheric Chemistry and Physics, 11(21), 11221-11235.

Urban J, Lautié N, Murtagh D, Eriksson P, Kasai Y, Loßow S, Le Flochmoën E (2007) Global observations of middle atmospheric water vapour by the Odin satellite: An overview. Planetary and Space Science, 55(9), 1093-1102.

Zahn A (2001) Constraints on 2-way transport across the Arctic tropopause based on O3, stratospheric tracer (SF6) ages, and water vapor isotope (D, T) tracers. Jounal of atmospheric chemistry. 39(3), 303-325.