UNIVERSITY OF OULU P .O. Box 8000 F I -90014 UNIVERSITY OF OULU FINLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

University Lecturer Tuomo Glumoff

University Lecturer Santeri Palviainen

Postdoctoral research fellow Sanna Taskila

Professor Olli Vuolteenaho

University Lecturer Veli-Matti Ulvinen

Planning Director Pertti Tikkanen

Professor Jari Juga

University Lecturer Anu Soikkeli

Professor Olli Vuolteenaho

Publications Editor Kirsti Nurkkala

ISBN 978-952-62-1852-6 (Paperback)ISBN 978-952-62-1853-3 (PDF)ISSN 0355-3191 (Print)ISSN 1796-220X (Online)

U N I V E R S I TAT I S O U L U E N S I SACTAA

SCIENTIAE RERUM NATURALIUM

U N I V E R S I TAT I S O U L U E N S I SACTAA

SCIENTIAE RERUM NATURALIUM

OULU 2018

A 711

Kaisa Lehosmaa

ANTHROPOGENIC IMPACTS AND RESTORATION OF BOREAL SPRING ECOSYSTEMS

UNIVERSITY OF OULU GRADUATE SCHOOL;UNIVERSITY OF OULU, FACULTY OF SCIENCE

A 711

ACTA

Kaisa Lehosm

aa

ACTA UNIVERS ITAT I S OULUENS I SA S c i e n t i a e R e r u m N a t u r a l i u m 7 1 1

KAISA LEHOSMAA

ANTHROPOGENIC IMPACTS AND RESTORATION OF BOREAL SPRING ECOSYSTEMS

Academic dissertation to be presented with the assent ofthe Doctoral Training Committee of Technology andNatural Sciences of the University of Oulu for publicdefence in Kaljusensali (KTK112), Linnanmaa, on 4 May2018, at 12 noon

UNIVERSITY OF OULU, OULU 2018

Copyright © 2018Acta Univ. Oul. A 711, 2018

Supervised byDoctor Jussi JyväsjärviProfessor Timo Muotka

Reviewed byDoctor Lenka KuglerováDoctor Marco Cantonati

ISBN 978-952-62-1852-6 (Paperback)ISBN 978-952-62-1853-3 (PDF)

ISSN 0355-3191 (Printed)ISSN 1796-220X (Online)

Cover DesignRaimo Ahonen

JUVENES PRINTTAMPERE 2018

OpponentAssociate Professor Tenna Riis

Lehosmaa, Kaisa, Anthropogenic impacts and restoration of boreal springecosystems. University of Oulu Graduate School; University of Oulu, Faculty of ScienceActa Univ. Oul. A 711, 2018University of Oulu, P.O. Box 8000, FI-90014 University of Oulu, Finland

AbstractHuman activities have increasingly altered freshwater ecosystems. Land use is a major driver ofhabitat loss and land use-related input of nutrients and other pollutants from agriculture, forestryand urbanization have deteriorated water quality. Freshwater research has mainly focused on lakesand streams while the effects of anthropogenic stressors on groundwater-dependent ecosystems(GDEs) are poorly known. Likewise, the effectiveness of ecological restoration in mitigatinghuman disturbance in GDEs remains understudied. In this thesis, I studied the effects of two mainanthropogenic stressors – land drainage and groundwater contamination – on boreal springecosystems and evaluated the recovery of spring biodiversity and ecosystem functioning afterhabitat restoration. I applied several structural (macroinvertebrates, bryophytes, leaf-decomposingfungi and groundwater bacteria) and functional (organic matter decomposition and primaryproductivity) measures to provide a comprehensive insight into these issues.

Both stressors modified spring ecosystems. Land drainage reduced the key ecosystemprocesses. Long-term monitoring of drainage-impacted springs showed a marked biodiversity lossand change of spring-dwelling bryophytes, and no signs of recovery were observed after about 20years since the intial land drainage. Groundwater contamination, indicated by elevated nitrate andchloride concentrations, altered the structure of spring biota, reduced their taxonomic diversityand suppressed primary productivity in the most severely contaminated springs. Spring restorationimproved habitat quality by reducing drainage-induced inflow of surface water, thus re-establishing groundwater-dominated hydrological conditions. Restoration increased abundance ofhabitat-specialist bryophytes and shifted macroinvertebrate composition towards naturalconditions, despite the restoration actions being fairly recent. Anthropogenic activities can thuscause severe structural and functional degradation of spring ecosystems, and their self-recoverypotential from these stressors seems low. Habitat restoration bears great promise as a cost-effective approach to mitigate drainage-induced impacts on spring ecosystems, but protection andco-management of groundwater resources are urgently needed to secure the role of springs asbiodiversity hotspots in the boreal forest landscape.

Keywords: aquatic hyphomycetes, bacteria, benthic invertebrates, biodiversity,bryophytes, conservation, crenobiontic, crenophiles, ecosystem functions, forestdrainage, groundwater contamination, groundwater-dependent ecosystems, habitatspecialists

Lehosmaa, Kaisa, Ihmistoiminnan vaikutukset boreaalisissa lähde-ekosysteemeissä. Oulun yliopiston tutkijakoulu; Oulun yliopisto, Luonnontieteellinen tiedekuntaActa Univ. Oul. A 711, 2018Oulun yliopisto, PL 8000, 90014 Oulun yliopisto

TiivistelmäIhmistoiminta muuttaa yhä enemmän vesiekosysteemejä. Maankäyttö on johtanut elinympäristö-jen häviämiseen, ja siihen liittyvä ravinne- ja haitta-ainekuormitus maa- ja metsätaloudesta sekä kaupunkiympäristöistä on merkittävästi huonontanut veden laatua johtaen maailmanlaajuiseen vesiluonnon monimuotoisuuden heikentymiseen. Vesiekosysteemien tutkimus on keskittynyt pääasiassa järvi- ja jokiympäristöihin, kun ihmistoiminnan vaikutukset pohjavesiriippuvaisiin ekosysteemeihin tunnetaan edelleen huonosti. Samoin kunnostusten merkitys pohjavesiriippu-vaisten ekosysteemien tilan parantamiseksi on selvittämättä. Väitöskirjassani tarkastelin kahden keskeisen ihmistoiminnan – metsäojituksen ja pohjaveden laadun heikkenemisen – vaikutuksia lähde-ekosysteemeihin sekä arvioin elinympäristökunnostusten vaikutuksia niiden rakenteeseen ja toimintaan. Sovelsin työssäni rakenteellisia (pohjaeläimet, sammalet, lehtikariketta hajottavat sienet ja pohjavesibakteerit) ja toiminnallisia (eloperäisen aineksen hajoaminen ja perustuotan-to) mittareita tuottamaan kattavan käsityksen tutkimuskysymyksiini.

Sekä metsäojitukset että pohjaveden laadun heikkeneminen aiheuttavat muutoksia lähteiden rakenteessa ja toiminnassa. Metsäojitukset hidastavat keskeisiä ekosysteemitoimintoja ja johta-vat lähdesammallajiston muutokseen ja monimuotoisuuden taantumiseen. Pohjaveden pilaantu-minen, jota työssä ilmennettiin kohonneilla nitraatti- ja kloridipitoisuuksilla, heikentää lähdela-jiston monimuotoisuutta, muuttaa lajikoostumusta ja johtaa perustuotannon laskuun voimak-kaimmin kuormitetuissa lähteissä. Kunnostus parantaa lähde-elinympäristön laatua vähentämäl-lä metsäojien aiheuttamaa pintavesivaikutusta palauttaen pohjavesivaltaisen hydrologisen tilan. Lähdekunnostusten myötä lähdesammaleet runsastuvat ja pohjaeläinyhteisön rakenne palautuu luonnontilaisten lähteiden kaltaiseksi, vaikka kunnostuksista on kulunut vasta muutamia vuosia. Väitöskirjan tulokset osoittavat, että ihmisen toiminta voi aiheuttaa muutoksia lähde-ekosystee-mien rakenteessa ja toiminnassa ja lähteiden luontainen palautuminen häiriöstä on hidasta. Läh-de-elinympäristöjen kunnostus vaikuttaa lupaavalta suojelutoimenpiteeltä metsäojitusten vaiku-tusten vähentämisessä, mutta lähteiden säilyttäminen monimuotoisena ja suojelullisesti arvok-kaana luontotyyppinä edellyttää pohjavesivarojen hallinnan ja tilan suojelun tehostamista.

Asiasanat: akvaattiset sienet, bakteerit, ekosysteemitoiminnot, habitaattispesialisti, lähdelaji, metsäojitus, monimuotoisuus, pohjaeläimet, pohjaveden pilaantuminen, pohjavesiriippuvainen ekosysteemi, sammalet, suojelu

We’ll never know the worth of water till well go dry. – Scottish Proverb

8

9

Acknowledgements

I owe my deepest gratitude to my supervisors Jussi Jyväsjärvi and Timo Muotka for their endless support, encouragement and patience. Jussi, I want to thank you for inspiring discussions, exciting field and conference trips and all the help and effort you gave to this project. Timo your advice and contribution has been indispensable - your enthusiasm for science is just inspiring. Thank you both for guiding me through the process, but most of all, you let me go deep into the world of aquatic microbes for which I am really grateful.

This work would not be the same without co-authors, thank you Risto Virtanen, Jari Ilmonen, Pekka Rossi, Annamari Markkola, Dimitrios Rados, Tatiana Chuzhekova, Lauri Paasivirta and Jouko Saastamoinen for collaboration, valuable help in the field and laboratory and manuscript writing. Special thanks go to Risto for teaching and helping me with bryophyte identifications and Annamari for all the help, among other things, with fungi and molecular analyses. Thanks also to hard-working master students Dimitrios Rados, Maria Rajakallio, Reetta Peuraniemi and Iikka Jaakola. I also want to acknowledge the reviewers Marco Cantonati and Lenka Kuglerová for their valuable comments on the thesis manuscript. And I am grateful for Tenna Riis for agreeing to be my opponent. This work was supported by Maj and Tor Nessling foundation, Maa- ja vesitekniikan tuki ry, The Finnish Concordia Fund and University of Oulu Graduate School. Their support is highly acknowledged.

I also want to thank every present and former member of Stream Ecology research group. In particular, Mari Tolkkinen, you have been a great friend and a tutor from my first day at the University of Oulu until today, our scientific and sometimes not so scientific discussions have been encouraging and sometimes so needed! I really think ‘nomen est omen’; thank you Kaisa Huttunen, Kaisa Mustonen and Maare Marttila for all your support and friendship. Warm thanks to Romain Sarremejane for coffee breaks and jokes, always not so good ones, but those moments and support have meant a lot. And also thank you Jarno, Pauliina, Mira, Heli, Mikko, Leif and Joel for all your help and support. Along my journey I have met amazing scientists from University of Oulu, Finnish Environment Institute and Natural Resources. Thank you Piippa, Annu, Saija, Henni, Kati, Karoliina, Tuija, Lauralotta, Anna, Vlad, Marko, Riina, Jukka, Heikki, Satu-Maaria, Tiina, Annika, Kirsti, Katri, Hannu, Anna-Kaisa among all others for your help, advises and collaboration. Special thanks to Piippa Wäli for taking me into the field in Kittilä and inspiring me to think further. Also, Saija Ahonen and Tiina Laamanen,

10

it has been a privilege to work with you in the laboratory and in front of computer, our lab discussions, bioinformatics courses and your help have meant so much.

I thank my dearest friends Jaana, Anna, Essi and Juulia for our trips, parties and discussions –many thanks for being always there for me. I am surrounded by many great friends and relatives who I owe many thanks. Also, parents-in-law and Maarit, Jussi, Henriikka and Eemil, it is always memorable go to summer cottage, ice-fish or even drive to Nordkapp with you. Mom and dad and my sister Elina, your support and engourament through the years has been invaluable. Special thanks to my mom, who in the first place engouraged me to move and study in England; that was in a way my first step towards this day. I have to say that a dog really is a man’s best friend: many thanks Heimo, you are the best listener and test audience for conference presentations. Last but definitely not the least, I want to thank my beloved husband Marko. Words are powerless to express my gratitude for your support and patience, it is an honour to share joys and sorrows with you. Our hiking, fishing and skiing trips have meant the world to me, John Muir once said ‘In every walk with nature one receives far more than he seeks’.

February 2018 Kaisa Lehosmaa

11

Abbreviations

ANOVA Analysis of variance BLAST Basic Local Alignment Search Tool Cl- Chloride DOC Dissolved organic carbon EU European Union GDE Groundwater-dependent ecosystem GWD The Groundwater Directive IndVal Indicator species analysis ITS Internal transcribed spacer NGS Next-generation sequencing OTU Operational taxonomic unit PCA Principal component analysis QIIME Quantitative Insights into Microbial Ecology 16S 16S ribosomal RNA NO3--N Nitrate-nitrogen WFD The Water Framework Directive

12

13

List of original articles

This thesis is based on the following publications, which are referred throughout the text by their Roman numerals:

I Lehosmaa K, Jyväsjärvi J, Virtanen R, Ilmonen J, Saastamoinen J & Muotka T (2017) Anthropogenic habitat disturbance induces a major biodiversity change in habitat specialist bryophytes of boreal springs. Biological Conservation. 215: 169–178.

II Lehosmaa K, Jyväsjärvi J, Ilmonen J, Rossi P.M, Paasivirta L & Muotka T (2018) Groundwater contamination and land drainage induce divergent responses in boreal spring ecosystems. Manuscript.

III Lehosmaa K, Jyväsjärvi J, Virtanen R, Rossi P.M, Rados D, Chuzhekova T, Markkola AM, Ilmonen J & Muotka T (2017) Does habitat restoration enhance spring biodiversity and ecosystem functions? Hydrobiologia. 793: 161–173.

Author’s contribution

In all three studies, I had the principal responsibility for data analyses and writing of manuscript drafts. Also, I was responsible for molecular analyses, bioinformatics and data collection in studies II and III, while R. Virtanen, T. Muotka, J. Ilmonen and J. Saastamoinen collected data for study I, except 2015 when I also contributed. The original study designs for studies I and III were planned by J. Jyväsjärvi, R. Virtanen, J. Ilmonen and T. Muotka, and I strongly contributed to the design of study II. All co-authors contributed to writing of the manuscripts.

14

15

Contents

Abstract Tiivistelmä Acknowledgements 9 Abbreviations 11 List of original articles 13 Contents 15 1  Introduction 17 

1.1  Springs – a key component of boreal forest landscapes .......................... 17 1.1.1  Spring flora and fauna .................................................................. 18 

1.2  Human-induced stressors on spring ecosystems ..................................... 19 1.2.1  Spring habitat restoration ............................................................. 20 

1.3  Structural and functional responses to human-induced stressors ............ 21 2  Aims of the thesis 23 3  Materials and methods 25 

3.1  Study areas and site selection .................................................................. 25 3.2  Environmental data ................................................................................. 28 3.3  Taxonomic measures ............................................................................... 28 3.4  Functional measures ................................................................................ 30 3.5  General data analyses .............................................................................. 30 

3.5.1  Specific data analyses: study I ...................................................... 31 3.5.2  Specific data analyses: study II .................................................... 31 3.5.3  Specific data analyses: study III ................................................... 32 

4  Results and discussion 33 4.1  Land drainage causes long-term diversity change and loss of

spring specialist bryophytes .................................................................... 33 4.2  Groundwater contamination and habitat degradation alter the

biodiversity and ecosystem functioning of boreal springs ...................... 35 4.3  Spring restoration improves hydrology and potentially also

biodiversity ............................................................................................. 37 5  Concluding remarks 39 References 43 Original publications 51 

16

17

1 Introduction

Never before has one species affected the environment with such a global magnitude as today. Humans have globally altered terrestrial and aquatic ecosystems by intensifying land use, pollution and climate change. These changes have led to an unprecedented biodiversity loss and have also altered crucial ecosystem processes (Pimm et al. 2014). Freshwater ecosystems have suffered the greatest biodiversity loss (Dudgeon et al. 2006, Strayer & Dudgeon 2010) and habitat modification is often the key driver of this unwanted development (Pimm et al. 2014). The alarming degradation of freshwater ecosystems has resulted in the establishment of numerous legislative mandates for the protection and management of surface waters and groundwater; e.g. the Water Framework Directive (WFD; EC 2000) and the Groundwater Directive (EC 2006) in Europe. The protection of freshwater ecosystems in WFD, however, focuses on larger rivers and lakes, and monitoring and ecological status assessement of small water bodies – particularly groundwater-depedent ecosystems (GDEs), such as cold-water springs – remain largely non-existent (Barquín & Scarsbrook 2008, Ilmonen et al. 2012). In Finland, springs are classified as woodland key habitats in the National Forest act, yet the lack of specific conservation protocols for springs has led to poor current status of spring ecosystems in many parts of the country.

1.1 Springs – a key component of boreal forest landscapes

Non-thermal springs are a special case of freshwater ecosystems; they are the places where groundwater arises onto the Earth’s surface forming a spring ecosystem (Fig. 1). In contrast to other freshwater types, they are entirely sustained by a continuous flux of groundwater that provides a unique thermally and chemically stable environment. The majority of springs have a stable flow rate, which together with thermal stability are major factors determining the biotic composition of spring ecosystems (Glazier 2009). Spring hydrological regime in boreal areas is highly dependent on seasonality, i.e. springs are dependent on groundwater and groundwater in turn is reliant on precipitation (Fig. 1). Spring habitat type can thus vary from lentic to lotic water which modifies the geomorphological setting of springs. Three main spring habitat types are rheocrenes, limnocrenes and helocrenes. Rheocrenes include spring brooks and streams, limnocrenes retain spring pools with an outflow channel and helocrenes incorporates marshes and bogs. Springs generally form heterogenous habitats, with

18

a variety of lentic and lotic habitats, helocrenes and groundwater seepage areas (Cantonati et al. 2012). Springs are environmentally diverse but patchy and isolated aquatic ‘islands’ within a terrestrial matrix (von Fumetti et al. 2006). The role of springs as key ecotones between surface water and groundwater, and also between terrestrial and aquatic habitats (Ward & Tockner 2001, Cantonati et al. 2012), supports diverse biotic communities which host several specialist plant and animal species rare in other freshwater habitats (Cantonati et al. 2012, Ilmonen et al. 2012, Kuglerová et al. 2016).

Fig. 1. Illustrative presentation on how springs are formed. A spring is formed when surface of groundwater meets Earth’s surface. Groundwater occurs between permeable (porous) and impermeable rock material.

1.1.1 Spring flora and fauna

Bryophytes are the dominant plant group in boreal springs (Heino et al. 2005), providing microhabitat and shelter for many macroinvertebrate species (Turunen et al. 2017), being the most important retentive structure in boreal fluvial systems (Muotka & Laasonen 2002). Benthic macroinvertebtebrates have an important role in maintaining healthy ecosystems as they are consumers of algae and organic matter; they are therefore often used in bioassessment and classification of freshwaters (Wallace &Webster 1996). Both bryophytes and macroinvertebrates

19

include several endemic spring specialists that are relatively well studied (Ilmonen et al. 2005, von Fumetti et al. 2006, Virtanen et al. 2009), while the microbiome of GDEs remains understudied. Microbes are the most numerous and diverse organisms on the planet and the dominant organism also in groundwater (Mauffret et al. 2017). Today, detection of DNA has made quantative microbial diversity analysis possible for both prokaryotic and eukaryotic cells. In aquatic ecosystems, microbes have an important role in ecosystem functioning as they are key components driving biogeochemical cycles (Fenchel et al. 1998 and references therein) of which their occurrence and activity is dependent on (Falkowski et al. 2008, Mauffret et al. 2017).

1.2 Human-induced stressors on spring ecosystems

Due to their small size and relative isolation, springs are highly vulnerable to land use activities. In Finland, peatland drainage has strongly altered forests since the 19th century (Holden et al. 2004, Peltomaa 2007). The primary goal of this vast drainage activity was to enhance forest growth by channeling surplus water from moist peat-dominated soils to streams and lakes (Holden et al. 2004). In Finland, land drainage peaked during the 1970s and 1980s, but it was practiced until 1997 when the Finnish Forest Act declined further land drainage; currently only maintenance of old drainage networks is allowed. By 1997, approximately 50% of the total pristine peatland area of Finland was impacted by drainage activities (Peltomaa 2007), which has led to a substantial loss of mires. Land drainage has deteriorated aquatic ecosystems in many ways by, for example, intensifying siltation, enhancing nutrient enrichment and altering catchment-scale hydrology (Holden et al. 2004, Jyväsjärvi et al. 2014, Nieminen et al. 2017). Also spring ecosystems have been disturbed by land drainage. In many cases, ditches have been drained directly into or across the spring pool and/or spring outflow. Previous hydrological and ecological studies have shown that land drainage changes wetland hydrology by disrupting natural groundwater inflow (Rossi et al. 2014), thus resulting in changes in the hydrology of groundwater-dependent ecosystems (Ilmonen et al. 2013).

A myriad of harmful agents originating from agriculture, industry and urbanization pose an increasing threat to groundwater and ecosystems depedent on it. Nutrient enrichment is a major threat to aquatic ecosystems worldwide (Woodward et al. 2012), and nitrate (NO3-) is the most ubiquitous contaminant of groundwater (Spalding & Exner 1993), because it is easily leached into

20

groundwater. Nitrate is derived both from diffuse sources (Mahvi et al. 2005) and a mixture of urban and agricultural point sources (Wakida & Lerner 2005). In Finland, nitrate concentrations exceeding 15 mg L-1 are considered to originate from anthropogenic pollution and nitrate content of drinking water should not exceed 50 mg L-1 (Finnish Water and Waste Water Works Association 2000). Similarly, the Nitrates Directive of EU allows groundwater nitrate concentrations

21

1.3 Structural and functional responses to human-induced stressors

Despite the vast evidence of human impact on aquatic and terrestrial biodiversity, the magnitude and even the direction of biodiversity loss has been recently debated (see Vellend et al. 2013, Cardinale 2014, Dornelas et al. 2014, Gonzalez et al. 2016). However, the general consensus is that anthropogenic impact leads to a non-random sequence of species loss, with habitat specialists being eradicated first (Pimm et al. 2014). Replacement of specialists by generalist species may result in taxonomic and functional homogenization thereby modifying ecosystem functioning (Clavel et al. 2011). Land use reduces connectivity between habitat patches and specialist species with limited dispersal abilities suffer disproportional losses and therefore are under the greatest risk of local extinction (Kuussaari et al. 2009, Pimm et al. 2014). Recovery of habitat specialist species may remain slow even after the initial stressor is no longer present, potentially causing delayed extinctions several decades or generations after the initial disturbance (extinction debt; Kuussaari et al. 2009). Furthermore, amalgamation of species with different habitat affinities may hinder the detection of patterns of conservation interest (Matthews et al. 2014). Division of species to habitat specialists and generalists is particularly meaningful for spring biota, as it consists of habitat generalists but also habitat specialist species which are either exclusively (crenobionts) or preferentially (crenophiles) inhabiting springs (Glazier 2009, Cantonati et al. 2012).

Taxonomic and functional measures can respond differently to the same anthropogenic stressor, which complicates impact assessment. At low levels, some stressors can be beneficial for the biota and ecosystem functions. For instance, aquatic plants can use surplus nitrate-nitrogen as a source of nitrogen if it is otherwise limited (Bornette & Puijalon 2011), and the same applies to groundwater bacteria and inorganic nutrients (Smith & Prairie 2004). Responses to stressors can be linearly negative or show a unimodal trend, in which case a response variable first benefits from a stressor until it reaches a level where a subsidy turns into a stress (‘subsidy-stress response’; Odum et al. 1979, Wagenhoff et al. 2011).

22

23

2 Aims of the thesis

Regardless of the recent increase in scientific interest towards spring ecology, no consensus exists on how i) spring biota and ii) the related key ecosystem processes are affected by different human activities or iii) if ecological restoration is effective in mitigating the negative effects of human impact on spring biodiversity. This information is urgently needed, as springs harbour valuable components of regional biodiversity, but are increasingly threatened by human activities, such as land drainage, agriculture, urbanization, water abstraction and global warming. The objective of my thesis was to assess the structural and functional properties of spring ecosystems in relation to different antropogenic stressors. The thesis also evaluates the effectiveness of spring habitat restoration in the mitigation of spring habitat or water quality impairment.

This thesis consists of three separate studies, where I focus on the following research questions:

1. Does long-term land drainage-induced habitat modification affect the biodiversity of spring bryophytes and, specifically, do habitat specialist bryophytes recover from the past land use?

2. How do different stressors (land drainage vs. groundwater contamination) modify spring ecosystem structure and functioning?

3. Does habitat restoration enhance spring biodiversity and ecosystem functioning?

More specifically, we first used long-term monitoring data (study I) to assess if habitat specialist bryophytes have recovered from historical land drainage or, alternatively, if long-term habitat disturbance has caused delayed extinctions of specialist species, potentially leading to a gain of generalist bryophytes. We hypothesized that the old, unmanaged drainage ditches might have been gradually revegetated, facilitating the recovery of spring specialist bryophytes (via reduced surface water input). We also tested if habitat generalist peatland bryophytes have attained greater dominance through time, posing a further threat to native specialist species.

In study II, we evaluated the roles of two human-induced stressors in determining the taxonomic structure and ecosystem functioning of spring ecosystems. We expected that habitat degradation should affect particularly macroinvertebrates, while autotrophs and microorganims are more sensitive to changes in groundwater quality (see Virtanen et al. 2009). We expected nitrogen

24

enrichment first to accelerate ecosystem functions, while at high concentrations it should reduce ecosystem process rates, thus showing a subsidy-stress response to groundwater contamination (Odum et al. 1979, Wagenhoff et al. 2011). We also expected that habitat degradation via land drainage might suppress ecosystem functions already at low stressor levels, due mainly to altered thermal regime and water brownification (increased DOC).

In study III, we expected spring habitat restoration to reduce surface water inflow and elevate the water table resulting in thermally stabilized conditions and increased spring habitat size. We also expected that, as a result, biodiversity and ecosystem functions should improve, approaching those in near-pristine springs.

25

3 Materials and methods

3.1 Study areas and site selection

This thesis was based on three spring datasets from Finland (Fig. 2). The first (63° - 64° N; 28° - 31° E; Study I) and the third (64° - 65° N; 27° - 29° E; Study III) study were conducted in eastern Finland and the second study in southern Finland (60° - 62° N; 22° - 27° E; Study II). The study region in eastern Finland is characterized by mixed and coniferous forests and peatlands, and the main land use is silviculture, particularly land drainage with limited agricultural activities. In contrast, the southernmost study region is the most populated part of Finland where >20% of land area is used for agriculture.

Fig. 2. Location of the study springs in Finland.

26

The 34 springs for study I were first sampled by Saastamoinen (1989) in 1987, when all these springs were still in near-pristine condition. Altogether 29 springs were disturbed by land drainage between 1988 and 1997 (Fig. 3c) whereas five of the springs retained their natural status throughout the study period (Fig. 3a). To assess long-term changes of spring bryophytes in relation to land drainage, the sites were revisited in 2000, 2010 and 2015.

For study II, seven near-natural, six drainage-impacted and eight water quality contaminated springs were surveyed to assess the impacts of land drainage and groundwater contamination on the taxonomic structure and functioning of spring ecosystems. ‘Natural’ springs were undisturbed in terms of both habitat and groundwater quality, ‘Drainage’ sites were disturbed by land drainage but had no changes in groundwater quality, while groundwater contaminated (‘WQ’) springs had unmodified habitat but were impaired by nitrate enrichment (Fig. 3d).

Study III was based on a space-for-time substitution design (Pickett 1989) and included nine near-natural springs (“Natural”), seven springs impaired by land drainage (“Impacted”) and seven previously drainage-impacted springs which had been restored between 2009 and 2011 (“Restored”; Fig. 3b). Restoration actions were similar in all springs, i.e. inflowing ditches were filled and the spring outflow was dammed with stones, wooden structures and mosses to increase water level in spring pools. The aim of these actions was to restore pre-disturbance hydrological conditions.

27

Fig. 3. Visual presentation of (a) near-pristine, (b) restored, (c) drainage-impacted and (d) contaminated springs. The outlet of the restored spring (b) has been dammed to increase the coverage of a spring pool and moss-dominated terrestrial-aquatic interface. In the drainage-impacted spring (c), a forest ditch is draining directly into a spring (on top of the photo). Also note a dense algal bloom in the contaminated spring (d).

28

3.2 Environmental data

Habitat integrity of each study spring and its immediate surroundings was assessed using a four-scaled (0–3) classification procedure (Heino et al. 2005). Sites assigned the status score 0 were considered completely destroyed, i.e. their groundwater discharge was disrupted and they lacked a distinct spring habitat. Severely or moderately deteriorated sites belonged to classes 1 (drainage ditches in the immediate proximity of the spring) and 2 (minor in-spring structures and/or drainage marks within more than 30 m from the spring), while class 3 refers to near-pristine springs with no visible human impact within 50 m of the spring. Areas (m2) of different mesohabitat types (i.e. spring pools, helocrenes, and spring brooks with either minerogenic or organogenic substrate) were estimated in the field. Electrical conductivity, pH and water temperature were measured in situ during sampling visits. In studies II and III, water samples of one litre were collected from the spring pool and they were analyzed for dissolved organic carbon (DOC), total phosphorus (TP), nitrate-nitrogen (NO3--N) and chloride (Cl-; only in study II) using Finnish national standards (National Board of Waters 1981). Furthermore, various environmental variables were measured at each study site. These included percentage forest shading, depth of spring pool, depth and width of outflow channel, spring discharge and current velocity. Water temperature variation was measured in studies II and III with temperature loggers (iButton, Thermochron; Maxim Integrated, San Jose, CA, USA) which were set to measure water temperature at 60 or 90-min intervals. In study II, percentages of different land uses (agriculture, forestry and urban areas) within a 1-km buffer around each study site were estimated with ArcMap 10.1 Desktop (ESRI 2011) using Corine 2012 Land Cover data. In study III, stable isotope techniques were applied to determine the balance of groundwater and surface water in each study spring. Stable isotopes are increasingly used as tracers in groundwater hydrology, as the differing fractionation of the 2O and 18H isotopes by precipitation and evaporation processes enables numerical quantification of groundwater in a water sample.

3.3 Taxonomic measures

Aquatic and semiaquatic bryophytes were sampled using one to seven 0.5 × 0.5 m quadrats (the number of quadrats varied depending on the size of the spring) along the main course of the spring outflow. The first quadrat was placed at the outflow point and, in study I, the remaining quadrats were sampled at 1, 2, 3, 5, and 10 m

29

intervals, while in studies II and III, quadrats were sampled at 1 m intervals from the spring outflow. For each quadrat, all bryophyte species were identified to genus or species and their percent cover (%) was estimated.

Macroinvertebrates (studies II and III) were sampled using a 20-cm wide D-frame hand net (mesh size 500 m). Five 20 s subsamples representing all available mesohabitat types (see 3.2. Environmental data) were taken within the first 10 m from the spring source by sweeping submerged substrates or by pressing muddy and mossy substrates and collecting loose material into the net (see Ilmonen & Paasivirta 2005). Animals were preserved in 70% ethanol in the field and later identified to the lowest feasible level (in most cases, species). In study II, all chironomids were sorted and identified, while in study III, due to extensive number of chironomids, only the first 300 randomly selected individuals per sample were sorted and identified.

Microorganisms in studies II and III were analyzed using DNA sequencing methods. Fungal DNA was extracted from subsamples of frozen alder leaves (see 3.4. Functional measures) using PowerSoil DNA Isolation Kit (MO BIO laboratories, Carlsbad, California, USA). Groundwater bacteria in study II was extracted from 100 ml of spring water using PowerWater DNA isolation Kit (MO BIO laboratories). The rRNA coding for fungi was amplified using ITS primers ITS1F (Gardes & Bruns 1993) and 58A2R (Martin & Rygiewicz 2005) and for groundwater bacteria 16S rRNA using 519F and R926 primers (Lane 1991). The amplicons were sequenced using Ion Torrent PGM sequencer. All sequences were analyzed using Quantitative Insights into Microbial Ecology (QIIME) pipeline (Caporaso et al. 2010). Sequences were clustered into operational taxonomic units (OTUs) using the Usearch61 algorithm, clustering OTUs at 97% identity (Edgar 2010). OTU composition was determined using BLAST (basic local alignment search tool) findings of NCBI (National Center for Biotechnology Information, USA) GenBank’s non-redundant nucleotide database and against reference databases Unite (Kõljalg et al. 2013) and Greengenes (DeSantis et al. 2006). As sequence numbers varied among samples, OTU data for each dataset were rarefied to the lowest shared sample size and singletons were discarded from further analyses.

In study III, a proxy of metabolically active fungal biomass, i.e. ergosterol content (Gessner 2005), was determined from alder leaf litter using a modified ergosterol assay (Nylund & Wallander 1992). Ergosterol extracts were quantified with high-pressure liquid chromatography (HPLC).

30

3.4 Functional measures

We measured ecosystem processes using organic matter breakdown rate (studies II and III) and primary productivity (study II). Periphyton accrual, a surrogate of primary productivity, was measured by incubating three replicate tiles in spring pools for 49 days. After the incubation period, periphyton biomass (µg cm-2) was measured from moist tiles with a phytobenthos fluorescence measurement instrument (bbe Benthotorch®). Organic matter decomposition was measured with leaf litter breakdown assays using 15 × 15 cm mesh bags. Grey alder (Alnus incana) leaves were collected prior to abscission and air-dried for one week. Four grams of leaves were weighed into two different mesh bags: fine bags (0.2 mm) measured microbial decomposition while coarse (8 mm) mesh allowed also leaf-shredding invertebrates to enter the bags. Three coarse and fine bags were each incubated in spring pools for 49 days in spring 2015 (Study II) and 62 days in autumn 2014 (Study III). In laboratory, the remaining leaf material was cleaned of invertebrates and other material, and subsamples of frozen leaf material were taken for DNA and fungal biomass (Study III) analyses. The remaining leaf material was dried at 60 °C for 48 hours and ashed for four hours at 550 °C to determine ash-free dry mass (AFDM). Leaf breakdown rates (k) were calculated using the negative exponential decay model (Benfield 1996). As mean water temperatures in studies II and III did not differ among the study groups, k values were not corrected for temperature before statistical analyses.

3.5 General data analyses

All statistical analyses were carried out in the R program (R Core Team 2014). Differences in bryophyte, macroinvertebrate, fungal and groundwater bacteria community composition among the spring groups (studies II and III) or study years (study I) were tested using one-way nonparametric permutational multivariate analysis of variance (PERMANOVA; Anderson 2001) with adonis function of vegan package (Oksanen et al. 2018). Bray-Curtis similarity coefficients were calculated for the abundance data and similarity matrices were permuted 9999 times to reveal the statistical significances of PERMANOVA tests. After significant global tests, pairwise comparisons were conducted with separate PERMANOVAs. Species compositional patterns were demonstrated visually with non-metric multidimensional scaling (NMDS) using metaMDS function of vegan package.

31

In studies II and III, the environmental differences among the spring groups were visually assessed using principal component analysis. Among-group differences in environmental conditions were tested with PERMANOVA using Euclidean distance matrices. Among-group differences of individual environmental variables of particular interest were tested with one-way ANOVA and, in case of significant difference, Tukey’s post-hoc tests were used to reveal differences between spring groups.

3.5.1 Specific data analyses: study I

Temporal changes in mean bryophyte species richness and abundance were tested with generalized linear mixed models (GLMMs). GLMMs were carried out using lme4 (Bates et al. 2017) and lmerTest (Kuznetsova 2016) packages in R program. Poisson error structure was used for species richness, while Gaussian error distribution was assumed for relative abundance data (square-root transformed prior to analyses). Due to the repeated sampling design, year was used as fixed effect and individual springs as random effect to avoid pseudoreplication; plot number among years remained the same. Pre-disturbance condition (the first sampling year 1987) was used as a baseline in the models. Statistical analyses were conducted separately for habitat specialist and generalist bryophytes. The classification of specialists followed Eurola et al. (1984), Ulvinen et al. (2002) and Juutinen (2011); the remaining taxa were assigned as generalists.

Changes in abundance-occupancy relationships (AOR) between sampling periods (1987, 2000, 2010 and 2015) and bryophyte groups (habitat specialists vs. generalists) were visually demonstrated with scatterplots, and the among-year and among-group differences in abundance-occupancy slopes were tested using analysis of covariance (ANCOVA). Species-specific abundance changes between sampling periods were tested with Wilcoxon signed rank tests.

3.5.2 Specific data analyses: study II

Differences in species richness (except bryophyte richness) among natural, drainage-impacted and contaminated springs were tested with Generalized Linear models (GLMs) using either Poisson or Quasi-Poisson link functions (in case of overdispersion). Taxonomic richness of macroinvertebrates and bryophytes were calculated separately for habitat specialist and generalist species using the classification by Ilmonen et al. (2009) for macroinvertebrates.

32

Among-group differences in functional measures and bryophyte species richness were tested with linear mixed-effects models (LMEs). Spring group (natural, drainage, WQ) was used as fixed effect and individual leaf bags or algal tiles nested within site as random effect, while for bryophytes, individual plot was used as random effect. Relationships between functional measures (leaf decomposition rate and periphyton accrual rate) and NO3--N or DOC concentration were tested with linear regression models. The explanatory variables and their quadratic term were included in the initial models, if they improved the model fit i.e. inspection of residual plots, adjusted r2-values and associated P-values. Also, if needed explanatory and response variables were log10-transformed.

In case of a significant global PERMANOVA (macroinvertebrates; groundwater bacteria), beta diversity components (nestedness and turnover) were specified with bray.part function in betapart R package (Baselga et al. 2018) using species abundance data (see Barwell et al. 2015). Beta diversity components were assessed across the entire data set to determine pair-wise dissimilarities and multiple-site dissimilarities of macroinvertebrates and groundwater bacteria (Baselga 2010). Beta diversity component matrices (turnover, nestedness) were correlated with the environmental matrix (nitrate concentration; strongly charged to PC1) using partial Mantel tests with geographical distances between sites being partialled out.

Indicator species for each spring group were identified using indicspecies package (De Cáceres & Jansen 2016) in R with IndVal settings (Dufrêne & Legendre 1997). Indval combines information from species abundance and occurrence in each group, the strongest indicators being those present in each site within a group but absent in other groups (Dufrêne & Legendre 1997).

3.5.3 Specific data analyses: study III

Among-group differences in species richness, abundance, leaf breakdown rates and environmental variables were tested with Generalized Linear Models (GLMs). Bryophyte species richness was summed across all plots. Quasipoisson or Poisson error distribution was applied for count data (to meet the assumptions of the model), while Gaussian error structure was used for environmental variables and functional measures.

Macroinvertebrate, bryophyte and fungal species responsible for among-group assemblage-level differences were tested using similarity percentage analysis (SIMPER; Clarke 1993) with simper function in the vegan R package.

33

4 Results and discussion

4.1 Land drainage causes long-term diversity change and loss of spring specialist bryophytes

Habitat quality in drainage-disturbed springs did not improve despite the cessation of land drainage about 20 years prior to our latest survey. Abundance of habitat specialist bryophytes decreased strongly between 1987 and 2000, and remained low through the rest of the study (Study I, Fig. 1a), while richness of habitat specialists showed a lagged species loss in 2015 (Study I, Fig. 1b). Contradictory to our hypothesis, habitat disturbance did not increase the abundance of generalist bryophytes (Study I, Fig. 1a). Community composition of specialist bryophytes in disturbed springs shifted strongly from the predisturbance condition, whereas no such change was recorded for generalist bryophytes (Study I, Fig. 2). These biodiversity changes were caused by drainage-induced modification of the spring habitat and/or groundwater-surface water dynamics, as similar changes were not observed in the five springs that retained their near-pristine status throughout the study (Study I, Appendix B).

None of the initially (1987) abundant habitat specialist species, such as Warnstorfia exannulata, became extinct during the monitoring period (Study I, Fig. 3 and Fig. 4b). In contrast, abundance and occupance of several initially rare species (e.g. Scorpidium revolvens and Scapania irriqua; Study I, Fig. 4 and Appendix C), decreased (Wilcoxon signed rank test P< 0.05) during the study, becoming extinct or at the margin of extinction. It is highly likely that more habitat specialist species will become extinct in the future, although this can take as long as a century from the initial disturbance (Kuussaari et al. 2009, Jackson & Sax 2010, Essl et al. 2015), particularly for species with a perennial lifestyle (Lindborg 2007). A strong post-disturbance decline of habitat specialist bryophytes might be occurring because the distribution of spring specialist bryophytes is very dispersal limited. In addition, springs are the upmost reaches within a river network: they do not receive any downstream dispersal of fragments, and many species are missing from otherwise suitable habitats even without human impact (Astorga et al. 2012). Overall, the dispersal of aquatic bryohytes is a highly stochastic process, where isolation, spore production and patch size influence dispersal success (Stream Bryophyte Group 1999).

34

Extinction debt of organisms that live in such small and isolated habitat patches as springs is usually paid off fastest if metapopulation processes act on the organism scale (Kuussaari et al. 2009). Therefore, we could assume that the majority of extinctions have already taken place. However, for perennial species in isolated habitats, metapopulation processes are usually slow and exctinction debt should therefore be a relatively fast process. It is thus likely that many more spring specialist species will become regionally or locally extinct, even without further habitat loss.

Species-specific responses revealed that not all spring specialists suffered equally from land-use disturbance and some species were found to be rather tolerant to anthropogenic pressure (e.g. Brachythecium rivulare; Study I, Fig. 3 and Fig. 4f). Generalist bryophytes did not gain greater dominance through time. However, some species, such as the common peatmoss Sphagnum angustifolium (Study I, Fig. 4g), clearly benefited from land use activities.

Lacking historical data, we are unable to estimate the number of boreal springs that have already been lost through land-use activities. Nevertheless, it can be safely assumed that the number of springs impaired by historical land drainage must be very high. Thus, land drainage activities in the late 1980s and early 1990s have left a strong imprint on spring bryophyte communities. Our results pointed out that the self-recovery potential of boreal springs is very low, as none of the 29 drained springs showed any signs of recovery towards the predisturbance conditions. Reduction and eventual loss of bryophytes may have far-reaching ecosystem-level impacts, as they often play a key functional role in headwater systems (Suren 1991, Muotka & Laasonen 2002). Therefore, to halt the continual spring biodiversity loss, restoration measures are needed to facilitate the recovery of disturbed springs (Hylander & Ehrlén 2013). There is evidence that restoration aiming to rehabilitate the hydrology (Menberu et al. 2016) and biodiversity (Haapalehto et al. 2011) of drainage-impacted mires can have positive side-effects on springs associated with the restored mires (Ilmonen et al. 2013), although with a time lag. The safest management action to conserve springs as an integral component of boreal forest landcapes is to refrain from any kind of land use activities within broad buffers around springs. More scientific evidence of ecologically sufficient and cost-effective buffers is needed to help forest managers find the best practices to protect springs from land drainage.

35

4.2 Groundwater contamination and habitat degradation alter the biodiversity and ecosystem functioning of boreal springs

Both habitat degradation and groundwater contamination had clear but divergent impacts on the structure and functioning of boreal spring ecosystems. A cross-taxon comparison revealed that changes in groundwater quality modified macroinvertebrate and groundwater bacteria community structure and reduced their species diversity (Study II, Fig. 3 and 4). Also, species richness of habitat specialist bryophytes was lower in the contaminated springs. In contrast, none of the organism groups showed a strong reponse to land drainage (Study II, Fig. 3 and 4).

Land drainage reduced both periphyton accrual and total leaf decomposition rates (Study II, Fig. 5), whereas in groundwater contaminated sites, neither of the ecosystem processes deviated from natural conditions (Study II, Fig 5). Microbial decomposition did not differ from reference conditions in either drained or WQ springs, implying that the among-group differences in leaf decomposition were primarily caused by leaf-shredding macroinvertebrates. The reduction of leaf decomposition rates was strongly associated with water brownification (increasing DOC concentration) but not with groundwater contamination (Study II, Fig. 6). Increased DOC reduced also algal productivity, and both ecosystem functions responded strongly to even small increases in DOC, with the relationship reaching a plateau at relatively low DOC concentrations (Study II, Fig. 6). Groundwater contamination resulted in a unimodal subsidy-stress type of response in periphyton accrual rate (Study II, Fig. 6d).

Nitrate concentration in contaminated springs ranged between 0.015 and 8 mg L-1 being, on average, 30 times higher compared to natural springs (or drained) springs (Study II, Fig. 2a), and there were substantial differences in community composition between contaminated and non-contaminated sites. The results of this work imply that the nitrate concentration limit for groundwater quality set by EU legislation (Nitrates Directive:

36

Pesticides or heavy metals were not measured in our study, but we did find a strong correlation between nitrate and chloride concentrations, thus supporting the assumption that sites high in nitrate are probably stressed by other contaminants as well (Hosen et al. 2017).

Primary productivity was unimodally related to groundwater quality (Study II, Fig. 6d) – first increasing, plateauing at approximately 2.5 mg L-1 NO3--N and then decreasing at even higher concentrations. The decrease in most contaminated sites is likely associated to ”cocktail of contaminants” typical of agricultural streams (Relyea 2009). Woodward et al. (2012) reported a similar unimodal response to nutrient enrichment for leaf decomposition rate. We found no relationship between groundwater contamination and leaf decomposition rate (Study II, Fig. 6b). This is surprising, as several studies have documented a strong inhibition of leaf decomposition by agricultural pesticides (e.g. Rasmussen et al. 2012, Fernández et al. 2015). However, impact by nutrients and/or pesticides in these studies was likely much higher than in our impacted springs.

Groundwater contamination altered groundwater bacteria community composition and, in contrast to the current paradigm of groundwater contamination increasing bacterial diversity (Röling et al. 2001, Johnson et al. 2004, Stein et al. 2010), we found the opposite: groundwater bacteria species diversity was lowest in the nutrient-enriched springs. Our results were, however, similar to those reported for forest soils after addition of nitrogen fertilizers (Liu & Greaver 2010). Partitioning of beta diversity into species turnover and nestedness components suggested that groundwater contamination introduced new bacterial OTU’s than replaced native species. A similar finding was recently reported for urban stream bacterial assemblages (Hosen et al. 2017). The majority of the non-native bacterial OTU’s are involved in purifying contaminated groundwater (denitrification process) and breaking down organic contaminants (Griedler & Leuders 2009). For example, OTUs of the family Acidovorax are used as denitrifying bacteria in wastewater treatment plants (Ehsani et al. 2015), and Paucibacter is an effective indicator of increased nitrogen content in urban streams (Hosen et al. 2017). The majority of the 722 groundwater bacteria OTU’s were cosmopolitan heterotrophic bacteria, Proteobacteria, Bacteroides, Actinobacteria and Firmicutes (Griedler & Lueders 2009, Salis et al. 2017). Our results were thus consistent with the view that groundwater or groundwater-dependent ecosystems do not support specialized bacterial taxa (Stein et al. 2010, Sirisena et al. 2013).

Land drainage increased DOC concentration of the spring water (Study II, Fig. 2c), and drainage-impacted sites also showed a substantially different seasonal

37

trajectory of water temperature compared to other springs (Study II, Appendix 4). The influx of snowmelt waters lowered spring water temperature in the beginning of the study period and the opposite pattern was observed by the end of it, as the temperature of the inflowing surface water exceeded the ambient groundwater temperature. Both ecosystem processes were disturbed by land drainage and showed a strong negative relationship with DOC concentration (Study II, Fig. 6 a and c). The role of DOC in reducing primary productivity likely relates to reduced light levels (Brett et al. 2017) and reduced substrate availability for algal attachment (Wojtal & Sobczyk 2012). Also the reduced decomposition rates in drainage-impacted sites were likely caused by increased input of DOC and fine organic sediments from drainage ditches, as the leaf bags were frequently covered by organic fines (K. Lehosmaa, pers. obs), thus potentially preventing the utilization of the buried leaves by shredding macroinvertebrates (Lecerf & Richardson 2010, Sanpera-Calbet et al. 2012).

4.3 Spring restoration improves hydrology and potentially also biodiversity

Restoration actions reduced surface water inflow into the springs. Both temperature and stable isotope data indicated that the restored springs regained near-natural groundwater-dominated hydrological conditions (Study III, Fig. 3 and 4), which is a prerequisite for the restoration of biodiversity and functioning of spring ecosystems. Macroinvertebrates responded positively to improved habitat quality, as their community composition shifted towards natural communities. Abundance of habitat specialist bryophytes recovered to natural levels (Study III, Fig. 6e), which increased the coverage of moss-dominated helocrenic mesohabitats (Study III, Fig. 6).

Despite apparent improvements in spring habitat quality and hydrology, some ecosystem components did not show any response to restoration. For instance, leaf litter decomposition (Study III, Fig. 5), diversity and community composition of leaf-decomposing fungi (Study III, Fig. 6c) and bryophyte community composition were similar among the spring groups. Restoration actions were carried out relatively recently (3–5 years prior to our sampling), and complete recovery to near-natural levels may take as long as several decades, as recorded for riparian (Hasselquist et al. 2015) and peatland vegetation (Haapalehto et al. 2011). Dispersal limitation may be a key factor constraining the post-restoration recovery of spring communities, as recolonization of restored springs by bryophytes is

38

known to be a slow and stochastic process (see Juutinen 2011, Ilmonen et al. 2013). This is generally the case for organisms that rely on spores (i.e. bryophytes and fungi) on long-distance dispersal (Stream Bryophyte Group 1999, Astorga et al. 2012). In contrast, many macroinvertebrate species have winged adult stages which enable more efficient and deterministic dispersal between springs (Bilton et al. 2001).

Fungal communities in restored sites were very different from both impacted and natural springs. Similarly, microbially-mediated leaf decomposition in restored springs was lower than in natural conditions (Study III, Fig. 5), indicating that changes in fungal communities possibly translated into changes in the decomposition process. Restored springs had deeper spring pools than did the non-restored and natural springs (Study III, Fig. 4d). These results thus indicate that deepening of the spring pools as part of the restoration protocol may have disrupted the terrestrial-aquatic linkage crucial for microbial communities (Chauvet et al. 2015, Ruiz-González et al. 2015). Despite little is known of fungal communities in springs, and of their roles in the functioning of spring ecosystems, this work indicated that boreal springs support fungal communities dominated by hyphomycetes typically found in aquatic-terrestrial ecotones (see Mäkelä, 1972). In addition, the higher occurrence of Chytridiomycota compared to streams (see Tolkkinen et al. 2015), and presence of Glomeromycota, suggest a strong linkage between boreal spring ecosystems and their terrestrial surroundings.

39

5 Concluding remarks

Despite recent increase in scientific interest towards GDEs in general and springs in particular, little is known about how human-induced stressors affect spring biota and ecosystem processes. This thesis provides evidence on how human activities have altered the biodiversity and ecosystem functioning of boreal springs. Long-term monitoring data in study I showed that land drainage has reduced species diversity and altered community composition of habitat specialist bryophytes, inducing a major biodiversity change. Furthermore, no signs of community recovery were detected after about 20 years since the initial disturbance, suggesting that the self-recovery potential of spring ecosystems is almost non-existent. The safest management action is therefore to refrain from any land use within broad buffers around springs, because past land use can compromise a species’ viability to such an extent that any level of future protection cannot prevent extinctions (Pimm et al. 2014). Our findings also underline the importance of separate analyses for species with different habitat affinities (see Matthews et al. 2014) for a more effective ecological assessment of spring ecosystems.

Study II showed that springs are differently disturbed by different anthropogenic stressors. Groundwater contamination mainly caused biodiversity loss, while land drainage reduced ecosystem process rates due mainly to increased brownification; this inhibitory effect could be seen already at very low levels of increased DOC. Furthermore, primary productivity was sensitive to both stressors, as groundwater contamination also altered periphyton accrual rates with a strong unimodal response. Single-stressor responses revealed that both stressors caused changes in biodiversity and ecosystem functioning of springs, and neither taxonomic nor functional measures alone allowed a wholesale detection of human impact in springs. Our results indicated that functional measures responded particularly well to the first signs of spring ecosystem impairement; therefore, functional measures should be fully incorporated into GDE bioassessment protocols. Groundwater bacteria was highly sensitive to groundwater contamination and should therefore be considered as a potential indicator group in biological assessment of GDEs. To protect water quality in GDEs and aquifers, future research should focus on finding cost-effective methods to purify contaminated aquifers. However, this remains challenging due mainly to the inaccessibility of subsurface waters. The ability of microbes to degrade contaminants (Jacobsen & Hlemsø 2014) could be potentially applied in

40

groundwater rehabilitation, but this requires deeper understanding of groundwater microbes and their functional roles in groundwater purification processes.

Studies I and II suggest the importance of spring habitat restoration to mitigate the effects of land drainage, and study III revealed that spring habitat restoration could indeed effectively reduce the drainage-induced impairment of springs by blocking surface water inflow and increasing the water level in spring pools, thus re-establishing the near-natural groundwater-surface water interface. Despite apparent improvements in hydrology and habitat quality in restored springs, their biodiversity and ecosystem functioning responded inconsistently to habitat restoration. Springs vary greatly in their physical settings and there is therefore no straigthforward recipe to successful spring restoration. However, damming the spring outflow with single or multiple wooden thresholds increased helocrenic mat abundance and diversified outflow paths, yielding a near-natural habitat structure, which was visually pleasant but also seemed to favour habitat specialist bryophytes and macroinvertebrates. Some springs were dug deeper and cleared from mosses and wood, creating unnaturally deep spring pools which harmed the spring biota, particularly aquatic hyphomycetes. Clearly more research is needed on the long-term post-restoration trajectory of spring communities. Before-control-impact designs, BACI, (see Underwood 1994) would be particularly useful because space-for-time substitution designs (see Study III) can be biased by various unmeasured environmental variables that may vary among study springs.

The lack of response by habitat specialist bryophytes to land drainage in study II was unexpected, as studies I and III showed that land drainage reduces diversity and abundance and alters community composition of habitat specialist bryophytes. Overall, our results suggest that as long as groundwater inflow remains at a sufficient level it can support crenophilous bryophyte species (see also Kuglerová et al. 2016). The partly contrasting results for leaf decomposition rates in studies II and III do not seem to reflect drainage intensity as DOC concentrations did not vary appreciably between the two studies/regions. Decomposition rates were overall higher in study III, most probably as a result from much higher nutrient concentrations in these springs, with an order of a magnitude difference between the studies. Higher nutrient concentrations in study III likely indicate longer timing of initial drainage (see Nieminen et al. 2017).

Anthropogenic impact homogenizes the taxonomic and functional properties of boreal spring ecosystems, potentially leading to impaired ecosystem services (Clavel et al. 2011, Johnson & Angeler 2014). Studies included in this thesis focused on single-stressor responses, but it is a rule rather than an exception that

41

freshwater ecosystems are threatened by multiple, simultaneously operating environmental stressors. More research on multiple stressor responses is needed, as any additional anthropogenic pressure may cause ‘ecological surprises’ (sensu Paine et al. 1998). For example, additional threats such as groundwater warming (Jyväsjärvi et al. 2015) can have effects on, for example, bacterial-mediated carbon cycling (Hall & Cotner 2007), potentially leading to unexpected changes in ecosystem services related to groundwater resources.

Currently there are no specific protocols for the effective protection of GDEs, as they are not specified in the EU level legislation and national laws are very diverse. The National Water Act and Forest Act of Finland protect near-pristine springs and their terrestrial surroundings from further degradation, but the concept of ‘near-pristine’ in this context is open to interpretation. Water Framework Directive (EC 2000) does provide some protection to GDEs, as it states that groundwater quality and quantity should be sustained at a sufficient level without causing any damage to terrestrial ecosystems directly dependent on the groundwater body.

The anthropogenic pressures addressed in this thesis – land drainage and groundwater contamination – deteriorated the taxonomic structure and ecosystem functioning of boreal springs. Impaired habitat quality can be restored by re-establishing the predisturbance groundwater hydrology, which seems a prerequisite for a successful recovery of spring biodiversity. Little is known about the recovery of contaminated groundwater, and clearly stricter protection of groundwater resources, particularly in areas with intensive urban and agricultural land use, is urgently needed. More intense cooperation among researchers, environmental managers and decision makers is needed to attract public awareness to the loss of these valuable ecosystems and maintain spring ecosystems and other GDEs as biodiversity hotspots of boreal landscapes also in the future.

42

43

References Anderson MJ (2001) A new method for non-parametric multivariate analysis of variance.

Austral. Ecol. 26: 32–46. Astorga A, Oksanen J, Luoto M, Soininen J, Virtanen R & Muotka T (2012) Distance decay

of similarity in freshwater communities: do macro - and microorganisms follow the same rules? Global. Ecol. Biogeogr. 21: 365–375.

Barquín J & Scarsbrook M (2008) Management and conservation strategies for coldwater springs. Aquat. Cons. 18: 580–591.

Barwell LJ, Isaac NJB & Kunin WE (2015) Measuring β-diversity with species abundance data. J. Anim. Ecol. 84: 1112–1122.

Baselga A (2010) Partitioning the turnover and nestedness components of beta diversity. Global. Ecol. Biogeogr. 19: 134–143.

Baselga A, Orme D, Villerger S, De Portoli J & Leprieur F (2018) Package ‘betapart’, Partitioning beta diversity into turnover and nestedness components version 1.5.0. R Foundation for Statistical Computing, Vienna, Austria. URI: https://cran.r-project.org/web/packages/betapart/index.html Cited 2018/03/14.

Bates D, Maechler M, Bolker B, Walker S, Christensen RHB, Singmann H, Dai B, Scheipl F, Grothendieck G & Green P (2017) Package “lme4” version 1.1-15. R Foundation for Statistical Computing, Vienna, Austria. URI: https://cran.r-project.org/web/packages/lme4/index.html Cited 2018/03/14.

Benfield EF (1996) Leaf breakdown in stream ecosystems. In: Hauer FR & Lamberti GS (eds.) Methods in Stream Ecology. San Diego, CA, Academic Press: 579–589.

Bilton DT, Freeland JR & Okamura B (2001) Dispersal in freshwater invertebrates. Annu. Rev. Ecol. Evol. Syst. 32: 159–181.

Bornette G & Puijalon S (2011) Response of aquatic plants to abiotic factors: a review. Aquat. sci. 73: 1–14.

Brett MT, Bunn SE, Chandra S, Galloway AWE, Guo F, Kainz MJ, Kankaala P, Lau DCP, Moulton TP, Power ME, Rasmussen JB, Taipale SJ, Thorp JH & Wehr JD (2017) How important are terrestrial organic carbon inputs for secondary production in freshwater ecosystems? Fresh. Biol. 62: 833–853.

Cañedo-Argüellus M, Hawkins CP, Kefford BJ, Schäfer RB, Dyack BJ, Brucet S, Buchwalter D, Dunlop J, Frör O, Lazorchak J, Coring E, Fernandez HR, Goodfellow W, González Achem AL, Hatfield-Dodds S, Karimov BK, Mensah P, Olson JR, Piscart C, Prat N, Ponsá S, Schulz CJ & Timpano AJ (2016) Saving freshwater from salts. Science. 351: 914–916.

Cantonati M, Füreder L, Gerecke R, Jüttner I & Cox E (2012) Crenic habitats, hotpots for freshwater biodiversity conservation: toward an understanding of their ecology. Freshw. Sci. 31: 463–480.

Cardinale B (2014) Overlooked local biodiversity loss. Science. 344: 1098.

44

Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N, Gonzalez Peña A, Goodrich JK, Gordon JI, Huttley GA, Kelley ST, Knights D, Koenig JE, Ley RE, Lozupone CA, McDonald D, Muegge BD, Pirrung M, Reeder J, Sevinsky JR, Turnbaugh PJ, Walters WA, Widmann J, Yatsunenko T, Zaneveld J & Knight R (2010) QIIME allows analysis of high-throughput community sequencing data. Nature Methods. 7: 335–336.

Chauvet E, Cornut J, Sridhar KR, Selosse MA & Bärlocher F (2015) Beyond the water column: aquatic hyphomycetes outside their preferred habitat. Fungal. Ecol. 19: 112–127.

Chowdhury SH, Kehew AE, Paasero RN (2003) Correlation between nitrate contamination and groundwater pollution potential. Groundwater. 41: 735–745.

Clarke KR (1993) Non-parametric multivariate analysis of changes in community structure. Aust. J. Ecol. 18: 117–143.

Clavel J, Julliard R & Devictor V (2011) Worldwide decline of specialist species: Toward a global functional homogenization? Front. Ecol. Environ. 9: 222–228.

De Cáceres M & Jansen F (2016) Package ‘indicspecies’: Relationship between Species and Groups of sites version 1.7.6. R Foundation for Statistical Computing, Vienna, Austria. URI: https://cran.r-project.org/web/packages/indicspecies/index.html Cited 2018/03/14.

DeSantis TZ, Hugenholtz P, Larsen N, Rojas M, Brodie EL, Keller K, Huber T, Dalevi D, Hu P & Andersen GL (2006) Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Appl. Environ. Microb. 72: 5069–5072.

Dornelas M, Gotelli NJ, McGill B, Shimadzu H, Moyes F, Sievers C & Magurran AE (2014) Assemblage time series reveal biodiversity change but not systematic loss. Science. 344: 296–299.

Dudgeon D, Arthington AH, Gessner MO, Kawabata Z-I, Knowler DJ, Lévêque C, Naiman RJ, Prieur-Richard A-H, Soto D, Stiassny MLJ & Sullivan CA (2006) Freshwater biodiversity: importance, threats, status and conservation challenges. Biol. Rev. 81: 163–182.

Dufrêne M & Legendre P (1997) Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecol. Monog. 67: 345–366.

EC (2000) Directive 2000/60/EC of the European Parliament and of the Council establishing a framework for Community action in the field of water policy. L 327.

EC (2006) Directive 2006/118/EC of the European Parliament and of the Council on the protection of groundwater against pollution and deterioration. L 372.

Edgar RC (2010) Search and clustering orders of magnitude faster than BLAST. Bioinformatics. 26: 2460–2461.

Eerkes-medrano D, Thompson RC & Aldridge DC (2015) Microplastics in freshwater systems: A review of the emerging threats, identification of knowledge gaps and prioritisation of research needs. Water. res. 75: 63–82.

Ehsani E, Jauregui R, Geffers R, Jarek M, Boon N, Pieper DH & Vilchez-Vargas R (2015) First draft genome sequence of the Acidovorax caeni sp. nov. type strain R-24608 (DSM 19327). Genome. Announc. 3:e01378–15.

45

ESRI (2011) ArcGIS Desktop: Release 10. Environmental Systems Research Institute, Redlands, CA.

Essl F, Dullinger S, Rabitsch W, Hulme PE, Pyšek P, Wilson JRU & Richardson DM (2015) Delayed biodiversity change: No time to waste. Trends. Ecol. Evol. 30: 375–378.

Eurola S, Hicks S & Kaakinen E (1984) Key to Finnish mire types. In: Moore PD (ed.) European mires. London, Academic press: 11–117.

Falkowski PG, Fenchel T, Delong EF (2008) The microbial engines that drive Earth’s biochemical cycles. Science. 320: 1034–1039.

Fenchel T, King G.M, Blackburn T.H (1998) Bacterial Biogeochemistry: The Ecophysiology of Mineral Cycling. Vol. 1. Great Britain, Academic Press.

Fernández D, Voss K, Bundschuh M, Zubrod JP, Schäfer RP (2015) Effects of fungicides on decomposer communities and litter decomposition in vineyard streams. Sci. Tot. Environ. 533: 40–48.

Finnish Water and Waste Water Works Association (2000) Soveltamisopas talousvesiasetukseen 461/2000, Sosiaali- ja terveysministeriön asetus 461/2000 talousveden laatuvaatimuksista ja valvontatutkimuksista. Helsinki, Suomen Kuntaliitto.

Gardes M & Bruns TD (1993) ITS primers with enhanced specificity for basidiomycetes – application to the identification of mycorrhizae and rusts. Mol. Ecol. 2: 113–118.

Gessner MO (2005) Ergosterol as a measure of fungal biomass. In: Methods to Study Litter Decomposition. Dordrecht, Springer: 189–195.

Glazier D S (2009) Springs. Encyclopedia of Inland Waters. (ed.) Gene. E. Linkens. 734–755.

Gonzalez A, Cardinale BJ, Allington GRH, Byrnes J, Endsley KA, Brown DG, Hooper DU, Isbell F, O’Connor MI & Loreau M (2016) Estimating local biodiversity change: A critique of papers claiming no net loss of local diversity. Ecology. 97: 1949–1960.

Griedler C & Lueders T (2009) Microbial biodiversity in groundwater ecosystems. Freshwater. Biol. 54: 649–677.

Haapalehto TO, Vasander H, Jauhiainen S, Tahvanainen T & Kotiaho JS (2011) The effects of peatland restoration on water-table depth, elemental concentrations, and vegetation: 10 years of changes. Restor. Ecol. 19: 587–598.

Hall EK & Cotner JB (2007) Interactive effect of temperature and resources on carbon cycling by freshwater groundwater bacteria communities. Aquat. Microb. Ecol. 49: 35–45.

Hasselquist EM, Nilsson C, Hjältén J, Jørgensen D, Lind L & Polvi LE (2015) Time for recovery of riparian plants in restored northern Swedish streams: a chronosequence study. Ecol. Appl. 25: 1373–1389.

Heino J, Virtanen R, Vuori KM, Saastamoinen J, Ohtonen A & Muotka T (2005) Spring bryophytes in forested landscapes: Land use effects on bryophyte species richness, community structure and persistence. Biol. Conserv. 124: 539–545.

Holden J, Chapman PJ & Labadz JC (2004) Artificial drainage of peatlands: hydrological and hydrochemical process and wetland restoration. Prog. in Phys. Geogr. 28: 95–123.

Hosen JD, Febria CM, Crump BC & Palmer MA (2017) Watershed Urbanization linked to differences in stream bacterial community composition. Front. Microbiol. 8: 1452.

46

Hylander K & Ehrlén J (2013) The mechanisms causing extinction debts. Trends. Ecol. Evol. 28: 341–246.

Ilmonen J & Paasivirta L (2005) Benthic macrocrustacean and insect assemblages in relation to spring habitat characteristics: patterns in abundance and diversity. Hydrobiologia. 533: 99–113.

Ilmonen J, Paasivirta L, Virtanen R, & Muotka T (2009) Regional and local drivers of macroinvertebrate assemblages in boreal springs. J. Biogeogr. 36: 822–834.

Ilmonen J, Mykrä H, Virtanen R, Paasivirta L & Muotka T (2012) Responses of spring macroinvertebrate and bryophyte communities to habitat modification: community composition, species richness and red-listed species. Freshw. Sci. 31: 657–667.

Ilmonen J, Virtanen R, Paasivirta L & Muotka T (2013) Detecting restoration impacts in inter-connected habitats: Spring invertebrate communities in a restored wetland. Ecol. Indic. 30: 165–169.

Jackson ST & Sax DF (2010) Balancing biodiversity in a changing environment: extinction debt, immigration credit and species turnover. Trends. Ecol. Evol. 25: 153–160.

Jacobsen CS & Hjelsø MH (2014) Agricultural soils, pesticides and microbial diversity. Curr. Opin. Biotechnol. 27: 15–20.

Johnson A, Llewellyn N, Smith J, van der Gast J, Lilley A, Singer A & Thompson I (2004) The role of microbial community composition and groundwater chemistry in determining isoproturon degradation potential in UK aquifers. FEMS Microbial. Ecol. 49: 71–82.

Johnson RK & Angeler DG (2014) Effects of agricultural land use on stream assemblages: Taxon-specific responses of alpha and beta diversity. Ecol. Indic. 45: 386–393.

Juutinen R (2011) The decrease of rich fen bryophytes in springs as a consequence of large-scale environmental loss. A 50-year re-sampling study. Lindbergia. 34: 2–8.

Jyväsjärvi J, Suurkuukka H, Virtanen R, Aroviita J & Muotka T (2014) Does the taxonomic completeness of headwater stream assemblages reflect the conservation status of the riparian forest? Forest. Ecol. Manag. 334: 293–300.

Jyväsjärvi J, Marttila H, Rossi PM, Ala-Aho P, Olofsson B, Nisel J, Backman B, Ilmonen J, Virtanen R, Paasivirta L, Britschgi R, Kløve B & Muotka T (2015) Climate-induced warming imposes a threat to North European spring ecosystems. Global. Change. Biol. 21: 4561–4569.

Katz BG, Griffin DW & Davis JH (2009) Groundwater quality impacts from the land application of treated municipal wastewater in a large karstic spring basin: chemical and microbiological indicators. Sci. Total. Environ. 407: 2872–2886.

Kuglerová L, Dynesius M, Laudon H & Jansson R (2016) Relationships between plant assemblages and water flow across a boreal forest landscape: a comparison of liverworts, mosses, and vascular plants. Ecosystems. 19: 170–184.

Kuussaari M, Bommarco R, Heikkinen RK, Helm A, Krauss J, Lindborg R, Öckinger E, Pärtel M, Pino J, Rodà F, Stefanescu C, Teder T, Zobel M & Steffan-Dewenter I (2009) Extinction debt: a challenge for biodiversity conservation. Trends. Ecol. Evol. 24: 564–571.

47

Kuznetsova A (2016) Package lmerTest: Tests in Linear Mixed Effects Models Version 2.0-36. R Foundation for Statistical Computing, Vienna, Austria. URI: https://cran.r-project.org/web/packages/lmerTest/index.html Cited 2018/03/14.

Kõljalg U, Nilsson RH, Abarenkov K, Tedersoo L, Taylor AFS, Bahram M, Bates ST, Bruns TD, Bengtsson-Palme J, Callaghan TM, Douglas B, Drenkhan T, Eberhardt U, Dueñas M, Grebenc T, Griffit GW, Hartmann M, Kirk PM, Kohout P, Larsson E, Lindahl BD, Lücking R, Martín MP, Matheny PB, Nguyen NH, Niskanen T, Oja J, Peay KG, Peintner U, Peterson M, Põldmaa K, Saag L, Saar I, Schüßler A, Scott JA, Senés C, Smith ME, Suija A, Taylor DL, Telleria MT, Weiss M & Larsson K-H (2013) Towards a unified paradigm for sequence-based identification of Fungi. Mol. Ecol. 22: 5271–5277.

Lane DJ (1991) 16S/23S rRNA sequencing. In: Stackebrandt E & Goodfellow M (ed.) Nucleic acid techniques in bacterial systematics. Chichester, United Kingdom, John Wiley & Sons: 115–175.

Lecerf A & Richardson JS (2010) Litter decomposition can detect effects of high and moderate levels of forest disturbance on stream condition. For. Ecol. Manage. 259: 2433–2443.

Lindborg R (2007) Evaluating the distribution of plant life history traits in relation to current and historical landscape configurations. J. Ecol. 95: 555–564.

Liu L & Greaver TL (2010) A global perspective on belowground carbon dynamics under nitrogen enrichment. Ecol. Lett. 13: 819–828.

Logue JB & Lindström ES (2008) Biogeography of groundwater bacteria in inland waters. Freshw. Rev. 1: 99–114.

Mahvi AH, Nouri J, Babaei AA & Nabizadeh R (2005) Agricultural activities impact on groundwater nitrate pollution. IJEST. 2: 41–47.

Martin KJ & Rygiewich PT (2005) Fungal-specific PCR primers developed for analysis of the ITS region of environmental DNA extracts. BMC. Microbiol. 5: 3–11.

Mauffret A, Baran N & Joulian C (2017) Effects of pesticides and metabolites on groundwater bacterial community. Sci. Total. Environ. 576: 879–887.

Matthews TJ, Borges PAV, Whittaker RJ (2014) Multimodal species abundance distributions: a deconstruction approach reveals the processes behind the pattern. Oikos. 123: 533–544.

Menberu MW, Tahvainen T, Marttila H, Irannezhad M, Ronkainen A-K, Penttinen J & Kløve B (2016) Water-table-dependent hydrological changes following peatland forestry drainage and restoration: Analysis of restoration success. Water. Resour. Res. 52: 3742–3760.

Muotka T & Laasonen P (2002) Ecosystem recovery in restored headwater streams: the role of enhanced leaf retention. J. Appl. Ecol. 39: 145–156.

Mäkelä K (1972) Some aquatic hyphomycetes on grasses in Finland. Karstenia 13: 16–22. National Board of Waters (1981) The analytical methods used by National Board of Waters.

Report 213. Helsinki, National Broad of Waters.

48

Nieminen M, Sallantaus T, Ukonmaanaho L, Nieminen TM, Sarkkola S (2017) Nitrogen and phosphorus concentrations in discharge from drained peatland forests are increasing. Sci. Tot. Environ. 609: 974–981.

Nolan J & Weber KA (2015) Natural uranium contamination in major US aquifers linked to nitrate. Environ. Sci. Technol. Lett. 2: 215–220.

Nylund JE & Wallander H (1992) Ergosterol analysis as a means of quantifying mycorrhizal biomass. Method. Microbiol. 24: 77–88.

Odum EP, Finn JT & Franz EH (1979) Perturbation theory and the subsidy-stress gradient. BioScience. 29: 349–352.

Oksanen J, Blanchet R, Friendly M, Kindt P, Legendre P, McGlinn P, Minchin PR, O'Hara RB, Simpson GL, Solymos P, Stevens MPP, Szoecs E & Wagner H (2018) Vegan: Community Ecology Package Version 2.4-6. R Foundation for Statistical Computing, Vienna, Austria. URI: https://cran.r-project.org/web/packages/vegan/index.html Cited 2018/03/14.

Paine RT, Tegner MJ & Johnson EA (1998) Ecological Surprises. Ecosystems. 1: 535–545. Palmer MA, Bernhardt ES, Allan JD, Lake PS, Alexander G, Brooks S, Carr J, Clayton S,

Dahm CN, Follstad Shah J, Galat DL, Loss SG, Goodwin P, Hart DD, Hassett B, Jenkinson R, Kondolf GM, Lave R, Meyer JL, O’Donnell TK, Pagano L & Sudduth E (2005) Standards for ecologically successful river restoration. J. Appl. Ecol. 42: 208–217.

Palmer MA, Hondula KL & Koch BJ (2014) Ecological restoration of streams and rivers: shifting strategies and shifting goals. Annu. Rev. Ecol. Evol. Syst. 45: 247–269.

Peltomaa R (2007) Drainage of forested in Finland. Irrig. Drain. 56: 151–156. Pickett ST (1989) Space-for-time substitution as an alternative to long-term studies. Long-

term studies in ecology: approaches and alternatives (ed.) Gene E. Likens. 110–135. Pimm SL, Jenkins CN, Abell R, Brooks TM, Gittleman JL, Joppa LN, Raven PH, Roberts

CM & Sexton JO (2014) The biodiversity of species and their rates of extinction, distribution, and protection. Science. 344: 1246752.

Rasmussen JJ, Wiberg-Larsen P, Baatruo-Pedersen A, Monberg RJ, Kronvang B (2012) Impacts of pesticides and natural stressors on leaf litter decomposition in agricultural streams. Sci. Tot. Environ. 416: 148–155.

R Core Team (2014) Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. URI: https://www.r-project.org/ Cited 2018/03/14.

Relyea RA (2009) A cocktail of contaminants: how mixtures of pesticides at low concentrations affect aquatic communities. Oecologia. 159: 363–376.

Rossi PM, Ala-aho P, Doherty J & Kløve B (2014) Impact of peatland drainage and restoration on esker groundwater resources: modeling future scenarios for management. Hydrogeol. J. 22: 1131–1145.

Ruiz-González, Niño-Carcía JB & del Giorgio PA (2015) Terrestrial origin of bacterial communities in complex networks. Ecol. Lett. 18: 1198–1206.

49

Röling WFM, van Breukelen BM, Braster M, Lin B & van Verseveld HW (2001) Relationship between microbial community structure and hydrochemistry in a landfill leachate-polluted aquifer. Appl. Environ. Microbiol. 67: 4619–4629.

Saastamoinen J (1989) Harjujen ja moreenimaiden lähteiden ekologiasta, sammallajistosta ja sammalkasvillisuudesta Pohjois-Karjalassa ja Etelä-Kainuussa. MSc thesis, University of Jyväskylä, Finland.

Salis RK, Bruder A, Piggott JJ, Summerfield TC, Matthaei CD (2017) High-throughput amplicon sequencing and stream benthic bacteria: indentifying the best raxonomic level for multiple-stressor research. Scientific Reports. 7: 44657.

Sanpera-Calbet I, Chauvet E & Richardson JS (2012) Fine sediments on leaves: Shredder removal of sediment does not enhance fungal colonization. Aquat.Sci. 74: 527–538.

Schuler MS, Hintz WD, Jones DK, Lind LA, Mattes BM, Stoler AB, Sudol KA & Relyea RA (2017) How common road salts and organic additives alter freshwater food webs: in search of safer alternatives. J. Appl. Ecol. 54: 1353–1361.

Sirisena KA, Daughne CJ, Moreau-Fournier MJ, Ryan KG & Chambers GK (2013) National survey of molecular bacterial diversity of New Zealand groundwater: relationships between biodiversity, groundwater chemistry and aquifer characteristics. FEMS Microbiol. Ecol. 86: 490–504.

Smith EM & Prairie YT (2004) Bacterial metabolism and growth efficience in lakes: The importance of phosphorus availability. Limnol. Oceanogr. 49: 137–147.

Spalding RF & Exner ME (1993). Occurrence of nitrate in groundwater - a review. J. Environ. Qual. 22: 392–402.

Stanton JS & Qi SL (2006) Ground-water quality of the northern High Plains aquifer. US Geological Survey, 1997, 2002–04.

Stein H, Kellermann C, Schmidt SI, Brielmann H, Steube C, Berkhoff SE, Fuchs A, Hahn HJ, Thulin B, Griebler C (2010) The potential use of fauna and bacteria as ecological indicators for the assessment of groundwater quality. J. Environ. Monit. 12: 242–254.

Strayer DL & Dudgeon D (2010) Freshwater biodiversity conservation: recent progress and future challenges. J. N. Am. Benthol. Soc. 29: 344–358.

Stream Bryophyte Group (1999) Roles of bryophytes in stream ecosystems. J. N. Am. Benthol. Soc. 18: 151–184.

Stuart M, Lapworth D, Crane E & Hart A (2012) Review of risk from potential emerging contaminants in UK groundwater. Sci. Total. Environ. 416: 1–21.

Suren AM (1991) Bryophytes as invertebrate habitat in two New Zealand alpine streams. Freshw. Biol. 26: 399–418.

Tolkkinen M, Mykrä H, Annala M, Markkola AM, Vuori K-M & Muotka T (2015) Multi-stressor impacts on fungal diversity and ecosystem functions in streams: natural vs. anthropogenic stress. Ecology. 96: 672–683.

Turunen J, Aroviita J, Marttila H, Louhi P, Laamanen T, Tolkkinen M, Luhta P-L, Kløve B & Muotka T (2017) Differential responses by stream and riparian biodiversity to restoration of forestry-impacted streams. J. Appl. Ecol. 54: 1505–1514.

Ulvinen T, Syrjänen K & Anttila S (2002) Bryophytes of Finland: distribution, ecology and red list status. Finnish. Environ. 560: 313–354.

50

Underwood AJ (1994) On beyond BACI: sampling designs that might reliably detect environmental disturbances. Ecol. Appl. 4: 3–15.

Vellend M, Baeten L, Myers-Smith IH, Elmendorf SC, Beauséjour R, Brown CD, De Frenn