Norström_Soil and stream water chemistry in a boreal catchment - interactions, influences of dissolved organic matter and effects of wood ash application_2010 $(function(){PrimeFaces.cw("Tooltip","widget_formSmash_items_resultList_11_j_idt799_0_j_idt801",{id:"formSmash:items:resultList:11:j_idt799:0:j_idt801",widgetVar:"widget_formSmash_items_resultList_11_j_idt799_0_j_idt801",showEffect:"fade",hideEffect:"fade",target:"formSmash:items:resultList:11:j_idt799:0:fullText"});});

Norström_Soil and stream water chemistry in a boreal catchment - interactions, influences of dissolved organic matter and effects of wood ash application_2010 $(function(){PrimeFaces.cw("Tooltip","widget_formSmash_items_resultList_11_j_idt799_0_j_idt801",{id:"formSmash:items:resultList:11:j_idt799:0:j_idt801",widgetVar:"widget_formSmash_items_resultList_11_j_idt799_0_j_idt801",showEffect:"fade",hideEffect:"fade",target:"formSmash:items:resultList:11:j_idt799:0:fullText"});});

Thesis for the degree of Doctor of Philosophy, Sundsvall 2010

Soil and stream water chemistry in a boreal catchment – interactions, influences of dissolved organic matter and effects of wood ash application

Sara H. Norström

Supervisors:

Professor Ulla Lundström

Assoc. Prof. Dan Bylund

Department of Natural Sciences, Engineering and Mathematics

Mid Sweden University, SE-851 70 Sundsvall, Sweden

ISSN 1652-893X,

Mid Sweden University Doctoral Thesis 99

ISBN 978-91-86694-04-3

Akademisk avhandling som med tillstånd av Mittuniversitetet i Sundsvall framläggs till offentlig granskning för avläggande av filosofie doktorsexamen fredag den 26:e november 2010, klockan 13.00 i sal O111, Mittuniversitetet

Sundsvall. Disputationen kommer att hållas på engelska.

Soil and stream water chemistry in a boreal catchment – interactions, influences of dissolved organic matter and effects of wood ash application

Sara H. Norström

© Sara H. Norström, 2010

Department of Natural Sciences, Engineering and Mathematics

Mid Sweden University, SE-851 70 Sundsvall

Sweden

Telephone: +46 (0)771-975 000

Printed by Kopieringen Mid Sweden University, Sundsvall, Sweden, 2010 ii

Soil and stream water chemistry in a boreal catchment – interactions, influences of dissolved organic matter and effects of wood ash application

Sara H. Norström

Department of Natural Sciences, Engineering and Mathematics, Mid Sweden University, SE-

851 70 Sundsvall, Sweden

ISSN 1652-893X, Mid Sweden University Doctoral Thesis 99; ISBN 978-91-86694-04-3

ABSTRACT

Two small bordering catchments in Bispgården, Central Sweden, were investigated in regard to soil solution and stream water chemistry during the frost free seasons of 2003-2007. Both catchments were drained by first order streams,

Fanbergsbäcken and Gråbergsbäcken, and in Fanbergsbäckens catchment an extensive investigation of the soil and soil solution chemistry was conducted by lysimeter and centrifugation sampling. The area of intensive soil solution investigation was situated in a slope towards a stream incorporating a recharge area, with podzolic soil, and a discharge area close to the stream with an arenosol soil. Samples were continuously taken in both the recharge- and the discharge area of the slope, and stream water was sampled in the streams of both catchments. The main variables of interest of the study were the interactions, the influence of dissolved organic carbon and the effects of wood ash application to soil solution and stream water.

The natural variations and the interactions between soil solution and stream water were monitored during 2003-2004. In soil solution, most of the investigated substances tended to increase during the growing season, due to weathering and microbial degradation of biota. Ca, Mg, Al and Fe were highly associated to dissolved organic carbon (DOC) throughout the catchment. The low molecular fraction of DOC seemed to have a higher impact on the soil processes in the recharge area, while high molecular DOC was more important for transport of cations in the discharge area and the stream water.

The concentration of different substances in the two streams differed significantly, even though the catchments were similar in size, shape and forestation. The seasonal patterns of most of the substances measured were iii

significantly correlated between the streams, however. Cations and pH correlated well with DOC and flow. The flow pattern driven by precipitation seems to be the driver of the stream water chemistry.

Wood ash was applied at a dosage of 3 ton/ha to one of the catchments in the autumn of 2004, to investigate the initial effects on the soil solution- and stream water chemistry. WAA is recommended by the Swedish Forest Agency to counteract acidification in soil and runoff that may be caused by an intensive biomass harvesting. The impact of the WAA was studied during 2005-2006.

Compared to the control temporarily higher concentrations of K, Ca and SO

4

were observed in the soil solution of the ashed area. In the stream water the effects of the

WAA were easier to distinguish due to higher sampling frequency. The strongest effect was seen for K, but increases in the stream water were also noted for DOC,

Ca, Mg, Si, Cl and malonate. No increase in pH could be statistically verified however, and overall the initial effects of the WAA seem mild.

Keywords: Boreal forest, catchment studies, dissolved organic carbon, soil solution, stream water, wood ash application iv

SAMMANFATTNING

Två angränsande avrinningsområden i Bispgården i centrala Sverige undersöktes under den snöfria säsongen 2003-2006. Båda avrinningsområdena dränerades av första ordningens bäckar och i det ena, Fanbergsbäckens avrinningsområde, gjordes en omfattande undersökning av mark och markvattenkemin. Markvatten provtogs genom centrifugering och med lysimetrar.

Provtagningen gjordes i inströmningsområdet där jordmånen var en typisk podzol samt närmare bäcken, i utströmningsområdet, där jordmånen var en arenosol. Bäckvatten provtogs i båda avrinningsområdena. Interaktioner mellan mark, markvatten och bäckvatten undersöktes med särskilt fokus på inverkan av organiskt kol samt påverkan av askåterföring.

De naturliga variationerna i markkemin samt interaktionerna mellan mark- och bäck vatten undersöktes 2003-2004. I markvatten ökade merparten av de studerade

ämnena under provtagningssäsongen beroende av ökad vittring och recirkulation av biota och fallförna. Ca, Al och Fe var i stor utsträckning associerade till löst organiskt kol (DOC). Den lågmolekylära fraktionen av DOC hade en större inverkan på markprocesserna i inströmningsområdet, medan den högmolekylära delen var viktigare för transport av katjoner i utströmningsområdet samt i bäckvattnet.

Koncentrationsnivåerna av olika substanser i bäckarna uppvisade skillnader trots avrinningsområdenas yttre likheter med avseende på storlek, form och beskogning. Trots de kvantitativa skillnaderna erhölls emellertid liknande säsongsvariationer i koncentrationerna, vilket indikerade att kemin i huvudsak styrdes av mängden DOC som i sin tur berodde av avrinningen som drevs av nederbörden. Höga halter av Ca, Mg, Al och Fe återfanns associerade till högmolekylärt DOC i bäckvattnet i en utsträckning som inte rapporterats tidigare.

På hösten 2004 spreds 3 ton aska/ha till Fanbergsbäckens avrinningsområde för att undersöka de initiala effekterna på mark- och bäckvatten kemin. Askåterföring bör ske minst en gång per omloppstid vid helträdsavverkning, i enlighet med

Skogsstyrelsens rekommendationer, främst för att motverka försurning i mark och avrinnande vatten som antas uppstå vid intensivt uttag av biomassa. De initiala effekterna av askåterföringen på mark- och bäckvattnets kemi studerades under

2005-2006 och Gråbergsbäcken kunde användas som obehandlad kontroll vid undersökning av vattenkemin i Fanbergsbäcken. I undersökningen av markvatten v

återfanns stora säsongsvariationer, vilket gjorde det svårt att urskilja eventuella effekter av askåterföringen. Tillfälligt högre värden av K, Ca och SO

4

återfanns dock i det askade området i jämförelse med kontrollområdet. På grund av högre provtagningsfrekvens i bäckvattnet var det lättare att påvisa förändringar i bäckvatten kemin. Framförallt märktes en signifikant ökning av K jämfört med kontrollbäcken. Sådana ökningar, om än inte lika markanta återfanns också för

DOC, Ca, Mg, Si, Cl och malonat. En tendens till ökat pH kunde observeras, men denna kunde inte verifieras statistiskt. De initiala effekterna av askåterföringen var således till synes milda, och den eftertraktade pH-effekten erhölls ej i denna undersökning. vi

TABLE OF CONTENTS

ABSTRACT ...................................................................................................................... III

SAMMANFATTNING .................................................................................................... V

LIST OF PAPERS ............................................................................................................ IX

LIST OF ABBREVATIONS ............................................................................................. X

1.

INTRODUCTION ...................................................................................................... 1

1.1 G

ENERAL INTRODUCTION

.............................................................................................. 1

1.2 C

ATCHMENT STUDIES

.................................................................................................... 2

1.3 S

OIL AND SOIL SOLUTION

.............................................................................................. 3

1.3.1 Recharge areas ...................................................................................................... 3

1.3.2 Discharge areas .................................................................................................... 4

1.4 S

TREAMS

....................................................................................................................... 4

1.4.1 Headwater ............................................................................................................. 4

1.4.2 Riparian zones.......................................................................................................5

1.5 D

ISSOLVED ORGANIC CARBON

......................................................................................5

1.6 L

OW MOLECULAR MASS ORGANIC ACIDS

(LMMOA

S

) .................................................. 7

1.7 W

OOD ASH APPLICATION

.............................................................................................. 8

1.7.1 Wood ash ............................................................................................................... 9

1.7.2 Effects ..................................................................................................................11

2. MATERIALS AND METHODS ................................................................................ 14

2.1 S

ITE DESCRIPTION

.......................................................................................................14

2.1.1 Study site .............................................................................................................14

2.1.2 Streams ................................................................................................................15

2.2 F

IELD METHODS

..........................................................................................................15

2.2.1 Stream water .......................................................................................................16

2.2.2 Soil sampling .......................................................................................................16

2.2.3 Lysimeter sampling .............................................................................................16

2.2.4 Centrifugation .....................................................................................................17

2.2.5 Soil respiration ....................................................................................................17

2.2.6 Application of wood ash to the Fanbergs catchment ...........................................17

2.3 L

ABORATORY METHODS

.............................................................................................18

2.3.1 Soil and ash .........................................................................................................18

2.3.2 Soil solution and stream water ............................................................................18

2.4 S

TATISTICAL EVALUATION

..........................................................................................19

vii

3. DEVELOPMENT OF AN ANALYTICAL METHOD FOR ANALYSIS OF

LMMOAS IN NATURAL WATERS ............................................................................. 21

3.1 S

EPARATION

................................................................................................................21

3. 2 D

ETECTION

.................................................................................................................22

3. 3 F

URTHER DEVELOPMENT

............................................................................................23

4. SOIL, SOIL SOLUTION AND STREAM WATER ................................................ 25

4.1 S

OIL SOLID

..................................................................................................................25

4.2 S

OIL RESPIRATION

....................................................................................................... 26

4.3 S

OIL SOLUTION AND STREAM WATER

..........................................................................26

4.3.1 pH and DOC .......................................................................................................27

4.3.2 DOC fractionation ...............................................................................................28

4.3.3 Carboxylic content ..............................................................................................29

4.3.4 Low molecular mass organic acids (LMMOAs) ..................................................31

4.3.5 Base cations and their association to DOC.........................................................33

4.3.6 Al and Fe and their association to DOC .............................................................35

5. IN-STREAM VARIATIONS ...................................................................................... 37

6. WOOD ASH APPLICATION (WAA) ...................................................................... 38

6.1 The effect of WAA on stream water- and soil solution chemistry in the Fanbergs catchment .....................................................................................................................38

6.1.1 Respiration ..........................................................................................................38

6.1.2 pH and DOC .......................................................................................................39

6.1.3 Ca and Mg ...........................................................................................................41

6.1.4 K and Na .............................................................................................................43

6.1.5 Transition elements and other metals..................................................................44

6.1.6 Anions..................................................................................................................44

6.1.7 Si..........................................................................................................................45

6.2 I

N

-

STREAM VARIATIONS IN

F

ANBERGSBÄCKEN

...........................................................46

7. CONCLUSIONS AND FURTHER PERSPECTIVES ............................................. 46

8. ACKNOWLEDGEMENTS ......................................................................................... 50

9. REFERENCES ............................................................................................................... 52

viii

LIST OF PAPERS

This thesis is mainly based on the following six papers, herein referred to by their

Roman numerals:

Paper I Dan Bylund, Sara H. Norström, Sofia A. Essén, Ulla S. Lundström

(2007) Analysis of low molecular mass organic acids in natural

Paper II waters by ion exclusion chromatography tandem mass spectrometry. Journal of Chromatography A 1176:89-93.

Jenny L.K. Vestin, Sara H. Norström, Dan Bylund, Per-Erik

Mellander, Ulla S. Lundström (2008) Soil solution and stream water chemistry in a forested catchment I: Dynamics. Geoderma 144:256-

270.

Paper III Jenny L.K. Vestin, Sara H. Norström, Dan Bylund, Ulla S.

Lundström (2008) Soil solution and stream water chemistry in a forested catchment II: Influence of organic matter. Geoderma 144:271-

278.

Paper IV

Sara H. Norström

, Jenny L.K. Vestin, Dan Bylund, Ulla S.

Lundström. Influences of dissolved organic carbon on stream water chemistry in two forested catchments in central Sweden. Accepted for

publication in Biogeochemistry.

Paper V

Sara H. Norström

, Dan Bylund, Jenny L.K. Vestin, Ulla S.

Lundström. Initial effects of wood ash application on the water chemistry of a first order stream in a small, boreal catchment in central Sweden. Submitted to Water, Air, & Soil Pollution.

Paper VI

Sara H. Norström

, Dan Bylund, Jenny L.K. Vestin, Ulla S.

Lundström. Initial effects of wood ash application to soil- and soil solution chemistry in a small, boreal watershed. Manuscript.

Not included Sara J.M. Holmström, Sara H. Norström, Anna Rosling. Functional differences in the genus Piloderma, linking production of LMMOAs and siderophores to vertical distribution in soil. Submitted to Fungal

Biology.

ix

LIST OF ABBREVIATIONS

AD – Anion deficit

DOC – Dissolved organic carbon

DOM – Dissolved organic matter

ESI – Electrospray ionization

HMM – High molecular mass (M w

>1 kDa)

LC – Liquid chromatography

LMM – Low molecular mass (M w

<1 kDa)

LMMOAs – Low molecular mass organic acids

MS – Mass spectrometry

SIM – Selected ion monitoring

WAA – Wood ash application

WTH – Whole tree harvesting x

1.

INTRODUCTION

1.1 General introduction

There is no such thing as an isolated piece of nature. Everything in a natural environment is connected to something else in that environment. Human activities in one part of such an environment will inevitably cause changes somewhere else, and our understanding of these relationships is often incomplete.

To achieve a better understanding it is crucial to expand our knowledge of natural systems and their interfaces in the environments, and how effects travel through different mediums, especially when proposing large-scale interventions for these systems.

A variety of influential environmental factors need to be considered when studying natural systems. Climate, precipitation, elevation, hydrology, topography and parent material are a few of the more important physical factors, while fertility, flora and fauna are examples of some biological factors. Each of these has the potential to influence the environment through interactions with other factors, creating unique circumstances for a particular area. This makes comparisons of results from different studies challenging. Solid background data can help illustrate treatment effects, preferably from the very same spot that the treatment was administered, despite the fact that chemical properties of different parameters in soil may show large variations over very small distances. These observations, when possible, can be combined with vast numbers of other observations from similar environments, and perhaps general conclusions can be drawn. It is this methodological collection of field observations that lays the basis for later analysis and greater understanding.

This thesis addresses the chemistry of soil solution and stream water, two different mediums that are in constant interaction. Our knowledge of these interactions is limited; the experiments described in this thesis aim to remedy at least a small part of that uncertainty in small catchment areas of a boreal forest environment. In addition to this, the initial effects of wood ash application (WAA) to the chemistry of these mediums were investigated.

The aims of this thesis were to examine the dynamics of elements and compounds, the temporal variations in concentrations, and processes determining the soil properties in a slope within a small, boreal catchment area as well as the relation between soil and stream water there (papers II, III and IV).

1

The main parameter of study was organic matter and its association of cations in the catchment. The role of dissolved organic carbon (DOC) as a driver of stream water chemistry induced by stream water flow was of particular interest, especially the quantity and quality of DOC, its seasonal fluctuations and its influence on cation concentrations in the soil and stream water of the investigated area (papers

II, III and IV).

To further investigate the low molecular fraction of DOC, which contains potent complex-forming substances involved in transport of metals and base cations in soil and stream water, an improved method for analysis of low molecular mass organic acids (LMMOAs) was required. Thus a new method employing liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) was developed (paper I) with the aims to provide high selectivity in order to distinguish between structural isomers, and high sensitivity in order to reach the low detection limits needed for the analysis of LMMOAs in stream water.

After two years of monitoring the natural variations of the studied environment, wood ash was applied to one of the investigated catchments. The aim of this catchment scale intervention was to study the initial effects of this application on the soil, soil solution and stream water chemistry (papers V and VI).

All results are presented in the form of published articles or manuscripts, and in the thesis a large summary of the most interesting results is presented together with some additional unpublished material.

1.2 Catchment studies

Restricted catchments are important for studying and interpreting the processes in boreal forest ecosystems. This allows short-term fluctuations to be distinguish from long-term trends and, if possible, to distinguish natural changes from maninduced effects. The nutrient dynamics differ between and within catchments due to the biogeochemical processes and transport routes depending on the hydrological conditions (Stutter et al. 2006). This results in different retention and transformation processes for different substances present in the soil (Johnson et al.

2000).

Stream water reflects the catchments’ hydrological and biogeochemical conditions since it represents the discharge zones. Chemical analysis of a headwater stream draining a catchment is a valuable tool for both quantifying the output and generalising the ongoing processes of a small catchment.

2

1.3 Soil and soil solution

Heterogeneous and complex soil and soil solution systems are the media through which elements and nutrients are transported to and from plant roots. The concentrations of elements in these matrices are highly influenced by biological processes and show large seasonal variations following the activity of biota. To fairly describe the processes taking place in soils, many measurement replicates need to be analysed, as some analytes may vary substantially even over short distances. The temporal variations in concentrations are normally largest in the upper soil horizons and become smaller with increasing depth in the soil profile

(Johnson et al. 2000; Lundström 1993), although the nutrient concentrations also vary with soil depth. The soil solution is also the medium for soil-forming processes, and nutrients such as Na and Si increase in concentration with soil depth as they mostly originate from weathering reactions. Conversely, K, which is intensively recirculated in the biota, is found in highest concentrations in the upper horizons (Johnson et al. 2000). The nutrient concentrations in the soil solution may therefore reflect both the heterogeneity of the soil, the soil forming processes and the activity of biota.

1.3.1 Recharge areas

Recharge is the process by which groundwater is replenished. Recharge areas occupy most of a catchment, and this is where water from precipitation is transmitted downward in the soil profile through a vertical flow. Soil-forming processes are continuous in forest soils. In temperate coniferous forests podzols are the dominating soil type and are generally found in recharge areas. Podzolic soils are characterized by a thick organic layer (O-horizon or mor layer) composed of needles, branches, leaves, cones etc. in different stages of degradation. Most of the organic compounds recovered in soil solution originate from this horizon. Below the mor layer is a strongly weathered eluvial horizon that is grey to white in colour

(the E-horizon). Here, organic acids (such as LMMOAs, fulvic acids and humic acids) from the mor layer and exudates of roots and biota weather minerals by forming complexes with Al and Fe (van Hees & Lundström 2000). The horizon is, slowly drained of these elements, thereby causing the greyish-white colour. This soil layer is usually enriched in residual Si. The illuvial horizon (B) is situated below the E-horizon, and here Al and Fe are immobilized as a result of precipitation of the organic complexes due to enhanced metal to carbon ratio and/or microbial degradation of the organic compounds followed by precipitation of the cations (Lundström et al. 2000a). Al and Fe are accumulated as solid Al-Si-

OH and Fe-OH phases, giving the horizon its characteristic reddish-brown to dark

3

brown colour (Lundström et al. 2000b). Below the B-horizon is the relatively undisturbed parent material, also called the C-horizon.

Podzolisation is a slow process and it takes between 300-3000 years for a podzol profile to fully develop (Lundström et al. 2000a). The thickness of the individual horizons varies depending on soil age, parent material, vegetation and the climate at the site (Ilvesniemi et al. 2000).

1.3.2 Discharge areas

Discharge areas are the opposite of recharge areas. They are the locations where ground water leaves the aquifer and flows to the surface. Discharge areas can consist of stream channels, dams, riparian zones, irrigation ditches etc., and soils in discharge areas are usually saturated. The soil profile is subjected to multidirectional flow patterns. Discharge areas might even be flooded for parts of the year, leading to anaerobic conditions and slow degradations of organic substances, and hence, soils in discharge areas are usually enriched in carbon. The soil-forming processes are thus largely different from those in the recharge area.

No clear vertical division of the soil material is usually visible, in contrast to the soils of recharge areas.

Water may be stored in the unsaturated soils of catchments during dry periods, and during high flows the increased ground water level mobilizes this water; the stream water chemistry varies with flow (Bishop et al. 2004; Kirchner 2003). The discharge area is important for nutrient dynamics and stream water composition, and may act either as a sink for the nutrients, as they can be retained by plant uptake or immobilization, or as a source of nutrients to the stream, especially during high flows like snow melting (Bishop et al. 1993; Fölster 2000; Laudon et al.

2004). Since stream water chemistry is determined by soil chemistry in the riparian zone, as this soil is the last contact before the soil solution reaches the stream, the chemistry in the runoff is essentially a product of flow and riparian soil chemistry

(Bishop et al. 2004).

1.4 Streams

1.4.1 Headwater

A headwater is the part of a stream proximate to its source and a first order stream is a stream that does not have any other stream feeding into it. The headwater is therefore the beginning of every first order stream. The chemistry of stream water in boreal, forested catchments shows large variations during the

4

season with the main driver being precipitation, which in turn affects dissolved organic matter (DOM) and pH.

In Sweden, a catchment size of 15 km

2

is the lower limit for what has been surveyed on a national scale, but more than 90 % of stream length in Sweden is actually present in catchments with areas <15 km

2

(Bishop et al. 2008). It has been shown that variability in stream water chemistry for these small catchments may be in the same order of magnitude as the variability in boreal forest streams in all of northern Sweden (Temnerud & Bishop 2005), and Ågren et al. (2007) even proposed that small headwater streams might be the largest contributor to the terrestrial DOC export per unit area. The importance of, and paucity of knowledge regarding, headwater has recently been discussed by Bishop et al. (2008) who stressed the importance of further research on these largely unknown environments.

1.4.2 Riparian zones

Stream water wetlands, i.e. riparian zones, are strips of wetland areas lining the stream channel. They link the stream environment to the terrestrial catchment and act as a buffer between the environments. Substances can be modified, diluted or concentrated in these zones before they enter the stream (Burt 2005; Luke et al.

2007; Osborne & Kovacic 1993) and riparian areas have in several studies been shown to be a large source of DOC to stream water (Bishop et al. 2004; Hinton et al.

1998). During high flow events the water table rises in the soil, flow paths change and old water stored in the unsaturated part of the soil becomes mobilized. Even superficial flows over the soil may occur. This causes rapid variations in stream water chemistry as a consequence of increased DOC concentration and changes the composition of organic matter, and in turn also the amount of cations and nutrients associated to DOC. Intensive studies of storm events are therefore important as some of the hydrological processes that influence DOC dynamics occurs only during these events, and may be missed with a low sampling frequency. In areas where riparian sources of DOC are important the DOC dynamics will vary between catchments and may even vary within a catchment during different storm events (Hinton et al. 1998).

1.5 Dissolved organic carbon

DOC is a ubiquitous substance in soil and natural waters, and plays a crucial role in establishing the nutrient status of soil as well as in the soil-forming processes. The organic compounds are derived from degradation of biota and from roots and mycorrhiza exudates (van Hees et al. 2005). DOC in soil plays an

5

important!role!in!the!biogeochemistry!of!carbon,!nitrogen,!phosphorus!and!metals,!

with!a!corresponding!impact!on!pedogenesis!(Kalbitz!et!al.!2000;!Lundström!et!al.!

2000b).!Due!to!their!complex"forming!ability,!the!organics!affect!the!transportation!

of!different!elements!within!the!soil!(van!Hees!&!Lundström!2000)!and!also!affect!

the!weathering!of!minerals!(Lundström!&!Öhman!1990;!Lundström!1993;!Raulund"

Rasmussen!et!al.!1998).!A!large!proportion!of!cations!may!be!associated!with!DOC,!

which!also!improves!the!moisture!and!structure!of!soil!(Prescott!et!al.!2000).!!

!

Most!DOC!in!streams!originates!from!terrestrial!sources!(allochthonous!carbon),!

where!litter!and!humus!are!the!most!important!contributors!(Kalbitz!et!al.!2000),!

Definite! compositional! differences! have! been! demonstrated! among! DOC! derived!

from! soil"! and! stream! environments! however,! indicating! that! in"stream!

transformations!take!place!(Malcolm!1990).!DOC!in!stream!ecosystems!is!a!mix!of!

carbon! compounds! produced! by! biota!in! the! stream! (autochthonous! carbon)!and!

allochthonous!carbon!and!holds!great!importance!due!to!its!multi"functional!role!

as!a!primary!food!source!in!the!aquatic!food!web,!in!mobility!and!transport!of!trace!

metals!and!contaminants,!and!in!regulating!acid"base!chemistry!(Hope!et!al.!1994;!

Oliver! et! al.! 1983).! In"stream! processes! are! important! to! DOC! in! stream! water!

(Meyer! et! al.! 1998;! Mulholland! &! Hill! 1997).! For! instance,! leaf! litter! stored! in! a!

stream!channel!may!contribute!to!as!much!as!30!%!of!the!daily!DOC!export!from!a!

stream!in!a!forested!catchment!(Meyer!et!al.!1998).!To!obtain!a!complete!picture!of!

catchment! processes! it! is! important! to! try! to! separate! catchment! hydrological!

processes! from! in"stream! processes.! DOC! in! freshwater! is! also! a! significant!

component!in!the!global!carbon!cycle!(Cole!et!al.!2007).

!!

!

DOC!is!commonly!defined!as!the!organic!carbon!derived!from!a!solution!that!

has!been!passed!through!a!0.45!#m"filter,!and!therefore!includes!a!wide!variety!of!

compounds!with!different!sizes!and!properties.!Due!to!its!heterogeneous!nature,!it!

is!hard!to!characterise!without!dividing!it!into!smaller!groups!according!to!certain!

properties,!one!of!which!is!size.!The!low!molecular!mass!(LMM)!fraction!of!DOC!

(<1! kDa)! contains! compounds! such! as! LMMOAs,! peptides,! amino! acids,!

carbohydrates,!siderophores!and!fulvic!acids!(Lundström!et!al.!2000a;!van!Hees!et!

al.!2005)!most!of!which!possess!a!high!ability!to!complex!metals!such!as!Al!and!Fe!!

(van! Hees! &! Lundström! 2000).! The! LMM! compounds! are! derived! mainly! from!

metabolic! pathways! and! are! mineralised! in! soil! within! a! few! days.! The! high!

molecular!mass!(HMM)!fraction!of!DOC!(>1!kDa)!contains!predominantly!humic!

acid!substances!smaller!than!0.45!#m,!which!also!add!to!the!metal"binding!capacity!

of!DOC!(Tipping!1998).!HMM!substances!are!more!recalcitrant!and!may!persist!in!

6

!

the environment for years (van Hees et al. 2005). The degradation of DOC is highly variable due to microbial community, soil structure, climate and aeration.

Elevated DOC concentrations during snowmelt or storms events are a common feature of nearly all watersheds, indicating that DOC is mobilized by infiltrating water (Hornberger et al. 1994). Stream water wetlands i.e. riparian zones, are a large source of DOC to stream water (Bishop et al. 2004; Hinton et al. 1998), and large parts of Fe and Al in stream water appear to be associated to organic matter

(Pettersson & Bishop 1996).

1.6 Low molecular mass organic acids (LMMOAs)

LMMOAs are versatile molecules defined as aliphatic or aromatic carboxylic acids with a maximum molecular weight of approximately 300 u. They are characterized by the possession of at least one carboxylic acid group. Depending on the dissociation properties and the number of carboxylic groups, these substances can carry various negative charges and can thereby complex cations in solutions. LMMOAs contribute to the DOC pool to a small extent comprising a maximum of 10 % of DOC in soil solution (Strobel 2001; van Hees et al. 2000). They have a similar vertical distribution as the total DOC content in forest soil, i.e. the highest concentrations are found in the organic layers and decrease with depth.

LMMOAs are found in all organisms and are involved in, among other things, energy production in the Krebs cycle and balancing cations in the cells for or maintaining osmotic potential (Marschner 1995). The major input of LMMOAs to soil is from root exudations (Sandnes et al. 2005; van Scholl et al. 2006a), forest litter (dead plant materials and microbial decomposition products; Pohlman &

McColl 1988), and ectomycorrhiza in symbiotic relationship with tree roots

(Ahonen-Jonnarth et al. 2000; van Hees et al. 2006; van Scholl et al. 2006a; van

Scholl et al. 2006b). The LMMOAs are implicated in many soil processes including mobilization and uptake of nutrients (Fransson et al. 2004; van Breemen et al.

2000), dissolution of soil minerals leading to pedogenesis (Lundström et al. 2000b), and microbial proliferation in the rhizosphere (Marschner 1995). They are generally found in higher concentration in rhizosphere soil than in bulk soil (Jones

1998). LMMOAs are also produced by photochemical transformation/degradation by DOC in freshwaters (Bertilsson & Tranvik 2000).

The ability of LMMOAs to complex metals and affect soil pH has been well investigated (Lundström & Öhman 1990; van Hees & Lundström 2000; van Hees et al. 2001). Organic acids with only one carboxyl group (e.g. lactate, formate, acetate)

7

have low metal-complexing abilities. The degree of complex-forming depends on the organic acid in question (number and proximity of carboxyl groups), the type of metal and its concentration, and the pH of the solution. Di- and tri-carboxylic acids can be potent metal complex-formers acting as detoxifiers of Al in soil solution (Hue et al. 1986) and bringing about the dissolution of soil minerals and modulating the size of humic substances (Jones 1998). LMMOAs are also known to increase in different parts of plants i.e. roots, leaves and xylem sap, as a response to

Fe deficiency as reviewed by (Abadia et al. 2002), and species induced weathering has been related to the production of individual LMMOAs in the soil (Holmström et al. submitted).

Organic acids have high turnover rates in soil, with half-lives ranging from 1-5 h in top-soil and approximately 12 h in subsoil (Jones 1998; van Hees et al. 2002).

Basic calculations using typical soil solution concentrations of LMMOAs, amino acids and simple sugars indicates that their contribution to respiration could be as high as 30 % (van Hees et al. 2005). However, concerns have been raised that current methods of quantifying organic acids may vastly underestimate the concentrations in soil solution, due to the high degree of buffering by the soil solid phase, the continual removal by the microbial community and the fact that available sampling techniques do not take into account the spatial heterogeneity of these substances (Jones et al. 2003). LMMOAs have rarely been identified or quantified in stream water, a medium that due to its close proximity to soil may offer useful insight into the behavior of these substances. A wide variety of techniques have previously been used for analyzing LMMOAs in soil solution, e.g. reversed phase liquid chromatography, ion chromatography, gas chromatography and capillary zone electrophoresis (Hagberg 2003; van Hees et al. 1999).

1.7 Wood ash application

Whole tree harvesting (WTH) in Swedish forestry has increased in recent years due to the novel interest in renewable energy sources. Previously only stem wood was recovered at harvest, while branches, tips and needles/foliage - parts that contains the most amounts of mineral nutrients were left to decompose at the harvest site, contributing to the recycling of mineral nutrients to the forest soils.

WTH has been shown to significantly decrease cation exchange capacity and total amounts of exchangeable base cations compared with conventional stem harvest in the uppermost soil layers 15 years after harvest (Olsson et al. 1996) and to increase soil acidity and reduce base saturation 28 years after harvest (Vanguelova et al.

2010). It has also been shown that leaving slash after forest harvest has a positive effect on nutrient status in foliage for successive forest stands (Olsson et al. 2000)

8

and that WTH can have a negative effect on tree growth already in the second rotation (Proe et al. 1996; Walmsley et al. 2009).

Increased use of forestry products as bio-energy also gives rise to by-products in the form of wood ash. Wood ash is a highly alkaline substance that up until recently has been dumped in landfills. To address the concerns that depletion of nutrients in forest soil will arise with the more extended use of WTH, the Swedish

Forest Agency now recommends WAA to recycle the nutrients in these forests. The main aim of WAA is to counteract acidification in soil and runoff from forests subjected to WTH. A positive side effect is an alternative disposal route for byproducts from the forest industry. It is also conceivable that increased growth rates in the treated forests may result. In Sweden a maximum amount of 6 ton/ha per rotation is to be applied to areas subjected to WTH, and not more than 3 ton/ha per

10-year period. WAA is recommended at least once in every rotation (Anonymous

2008).

When wading through the literature discussing the effect of wood ash application to forest ecosystems the most striking feature is the abundance of variables that can be changed in an investigation. Ash type is a typical example variable; different wood products and incineration ovens can be used resulting in markedly varying ash quality. Similarly, treatments of the ash can encompass any of the following: selfhardening, granulation or pelletzing. Additives may include any combination of dolomite, lime, N- and P-fertilizers. The quantity and frequency of application is also highly variable, as is the season of distribution.

Geographical features such as clearcuts and mires, as well as fertility status in the forest (Saarsalmi et al. 2001a) are additional variables. Investigation parameters such as the time after application until investigation, timing and presence of control plot establishment, number of replicates used, choice of medium sampled and method used to do so, choice of analytes for analysis and method used to do so etc. all make it difficult to draw general conclusions from the literature.

1.7.1 Wood ash

The main types of crystalline compounds formed during combustion of forest products, such as sawmill residues, are oxides, carbonates, sulphates and chlorides. Nitrogen compounds are lost in fuel gases when wood is combusted.

Reactive oxides and soluble salts are compounds that can cause undesirable effects in forest biotopes such as pH shock, burning damages to plant tissues and salt effects (Steenari & Lindqvist 1997). To avoid this, the ash must be stabilized.

Stabilisation can be achieved by pelletizing, granulation or self-hardening. The self-

9

!

hardening!process!utilizes!the!ability!of!most!ash!materials!to!solidify!on!addition!

of! water.! Water! is! also! added! in! the! pelletizing! and! granulation! technique! but!

instead!a!mechanical!process!is!employed,!while!self"hardening!ash!is!usually!left!

out!in!the!open!to!solidify!after!which!it!is!crushed.!!

!

During!the!self"hardening!process!the!dominating!reactions!in!the!presence!of!

CO

2

!are:!

!"# $ %

&

# ' !"(#%)

&

!

!

!"(#%)

&

$ !#

&

' !"!#

*

$ %

&

#

!

These! reactions! occur! in! consecutive! steps.! The! hydration! of! CaO! is! rapid! and!

exothermic,! then! the! CO

2

! dissolved! in! the! pore! fluid! of! the! hydrated! ash! is!

transported!by!diffusion!into!the!material!where!crystalline!CaCO

3

!is!formed.!The!

carbonation! of! Ca(OH)

2

! (portlandite)! is! hence! dependent! on! the! presence! of! a!

water!phase!in!which!the!reactants!can!be!dissolved!and!transported.!The!calcite,!

CaCO

3

,!that!forms!precipitates!from!the!solution!and!creates!a!product!layer!on!the!

ash!surface!and!in!the!pores.!Both!the!hydroxide!and!carbonate!formation!lead!to!a!

solidification! of! the! wetted! ash.! During! the! hardening! process,! the! ash! surface!

layer! gradually! becomes! hard!and! brittle! as! the! crystallization! of!CaCO

3

! proceed!

(Steenari!&!Lindqvist!1997;!Steenari!et!al.!1998).!

!

Ashes!with!high!content!of!unburnt!matter!do!not!harden!on!addition!of!water,!

and! the! degree! of! hardening! decides! the! reactivity! of! the! ash.! Calcite! is!

approximately! a! hundred! times! less! soluble! than! calcium! oxide! or! calcium!

hydroxide,! which! is! why! carbonatisation! of! the! ash! is! an! important! part! of! the!

stabilization.!The!formation!of!a!hard,!dense!structure!generally!leads!to!reduced!

leaching!rates!for!most!nutrients!(Steenari!&!Lindqvist!1997).!

!

Since! carbonatisaton! primarily! occurs! in! the! surface! layer! of! the! ash! pile,! the!

reaction!in!the!interior!part!is!slow.!This!can!be!due!to!the!formation!of!a!product!

layer!of!calcite!hindering!the!transport!of!carbon!dioxide!to!the!interior!of!the!pile,!

or! a! drying! of! the! surface! layer! though! the! reaction! takes! place! in! solution.! To!

speed!up!the!carbonation!process!the!ash!may!be!turned!over!and!wetted.!As!the!

ash!has!been!spread!in!the!forest,!the!ash!particles!will!become!more!available!for!

carbonatisation! and! this! process! may! continue! (Steenari! et! al.! 1999).! In! a!

comparison!between!untreated!ash!and!ash!that!had!been!stored!outdoors!or!in!the!

laboratory! it! was! found! that! the! leaching! of! Ca! significantly! decreased! by! the!

carbonation!reaction.!!

10

!

The amount of Ca released from the ash is highly dependent on the form it is present in. If the ash is well stabilized, Ca is present as carbonate while a less stabilised ash will have Ca in the form of hydroxides and oxides (Steenari et al.

1998). The dissolution of Ca species at the ash surface, i.e. in the pore fluid, is pH dependent. One example given is Ca(OH)

2

, whose solubility is very low at a pH above 12.5 but rapidly increases with decreasing pH owing to CO

2

equilibration and loss of alkali metal hydroxides by leaching.

The self-hardening processes increased the leaching of K and Na, probably due to a decrease in particle size. A larger surface area is thus exposed to the leachate and this facilitating the dissolution and transport of these substances (Steenari et al.

1999). In a column leaching experiment, Ca, K, Na Cl and SO

4 leached rapidly, while Mg and P seem to be released more slowly from the ash (Eriksson 1998b).

Similar observations were made by Steenari et al. (1999) who also found low leaching rates for Fe, Mn, Zn, Cu, Si and Pb, Cd and very low Hg levels in leachates.

1.7.2 Effects

A number of studies have investigated WAA in forests and its subsequent effects on soil processes (Lundström et al. 2003a), forests and aquatic systems

(Aronsson & Ekelund 2004), temperate forest ecosystems (Augusto et al. 2008) and environmental impacts in forestry (Pitman 2006). In most of the investigations, soil samples and the exchangeable pool of nutrients have been studied, while fewer studies addressed the effects on soil solution chemistry and stream water chemistry.

Application of wood ash has different effects depending on soil type and nutrient status in the soil. For instance, addition of stabilized wood ash to coniferous stands on mineral soil did not significantly affect stem growth (Jacobson

2003). Increased growth occurred when N was added however, alone or in combination with wood ash, and there were indications that WAA would increase stem growth at nutrient rich sites while conversely decreasing growth in less fertile sites (Jacobson 2003). Högbom et al. (2001) found that WAA may increase leaching of NO

3-

from sites with relatively high N-input.

Wood ash is sometimes referred to as fertilization, and indeed increased growth rate of soil microorganisms and increased respiration at the soil surface has been observed (Zimmermann & Frey 2002). Perkiömäki & Fritze (2002) found that WAA onto the forest floor of coniferous forests caused a dose-dependent increase in

11

microbial activity and a change in the community structure in the mor layer independent of the fertility of the site. Other groups have had more difficulty in confirming this. Maljanen et al. (2006a) found no initial change in the respiration rate due to WAA following application of 7 ton/ha of loose wood ash to a forested site in southern Finland. However, when investigating WAA (3-8 ton/ha) in northern Finland on both mineral- and peat soils increased respiration rate on old ash plots (14-50 years) was observed, but no increase in newly established experiments (1 year; Maljanen et al. 2006b). Rosenberg et al. (2010) found increased respiration 12 years after application of 6 ton/ha to a Norway spruce site in southern Sweden and Taylor & Finlay (2003) found changes in the mycorrhizal community in a Norway spruce forest 4 years after application of 4.28 ton/ha of wood ash. Jokinen et al. (2006) tried to distinguish what change, increased pH or

DOC, caused by the WAA was responsible for the increase in microbial activity and demonstrated that both pH and DOC quality influenced subsets of the bacterial community.

When applying ash to forests the aim is generally to raise the pH of acidic soils and surface waters, and to improve the nutrient status of the soil. Increases in cation exchange capacity (CEC), base saturation (BS), pH and decreases in total acidity (TA) in the mor layer have been found in several studies after application of

3 ton DS/ha of wood ash to forest stands of different ages in Sweden and Finland

(Arvidsson & Lundkvist 2003; Eriksson 1998a; Saarsalmi et al. 2001b).

In investigations of the effects of WAA on soil solution chemistry, Geibe et al.

(2003) found increased concentration of Ca, Mg and DOC in soil solution 4 years after application of 4.28 ton TS/ha and in the same investigation Holmström et al.

(2003) found that the increase in DOC was attributed to an increase in the HMM fraction, in turn leading to a more hydrophobic constitution of DOC. Ring et al.

(2006) observed increases in K, Na, SO

4

and TOC in soil solution sampled at 50 cm depth after application of 3 ton/ha crushed wood ash.

In investigations of the effects of WAA (2.4 ton/ha) to the soil solution chemistry in a pine stand in northern Germany, the DOC concentration did not increase. An initial decrease of pH due to desorption of exchangeable cations when soluble salts were leached from the ash or desorption of Al followed by precipitation of Al(OH)

3

was noted, however. Additionally, elevated concentrations of Ca, K, SO

4

and NO

3

were found furthermost in the topsoil, but some changes were visible down to 100 cm (Ludwig et al. 2002; Rumpf et al. 2001).

12

Information regarding the effects of wood ash application on the chemistry of streaming water is scarce, with a few exceptions. A study by Tulonen et al. (2002), for instance, reported that WAA (6.4 ton/ha) to 12 % of a catchment area drained by two small streams to a humic lake led to a slight increase of conductivity, alkalinity, K, Cl, and SO

4 concentrations for the stream running through peat soil and pH, K and Cl in stream water running through mineral soil. The increases were only significant for K in the peat soil stream however, and for pH and K in the mineral soil the first year after ash application. Changes in the same parameters were also observed in the recipient lake. In another study, three adjacent catchments of 20 ha were treated with 2.2 ton/ha granulated wood fly ash, 5 ton/ha dolomitic limestone or remained untreated as a reference respectively (Fransman &

Nihlgård 1995). Compared to the reference, the ash treatment resulted in increased

K concentrations and a tendency towards increased pH five years after treatment.

In the limed catchment stream, pH, Ca and Mg increased and Al, Fe and Mn decreased over the same period.

Studies of stream water draining lime-treated catchments are more plentiful.

Löfgren et al. (2009) studied nine stream waters draining catchments treated with 3 ton/ha of lime or dolomite and ten untreated stream waters in southern Sweden.

No significant differences were observed between treated and untreated stream waters 12-16 years after treatment. Regardless of treatment, however, all stream waters exhibited a general pattern of declining concentrations of SO

4

, Ca, BS and increasing acid neutralizing capacity (ANC), likely due to the reduced sulphur deposition. In contrast, Hindar et al. (2003) investigated a stream water draining a

84 ha catchment treated with 3 ton/ha of dolomite and reported significant increases in pH, Ca, Mg, ANC and TOC and decrease of inorganic monomeric Al over a 6-year period after liming compared to a reference stream.

Aronsson & Ekelund (2008) investigated Fanbergsbäcken, the same stream studied in this thesis, 2 years after WAA and found no changes in bentic diatom community, but indices of elevated concentrations of K in aquatic moss and leaves from alder. No heavy metal accumulation was detected by either Moilanen et al.

(2006) or Levula et al. (2000) in berries or mushrooms in different time intervals after subjection to different doses of wood ash. In contrast, the concentrations were often lower than those in the controls (except for As) in older experiments (13-52 years; Moilanen et al. 2006).

13

2. MATERIALS AND METHODS

Figure 1. The Gråbergs catchment (control) and Fanbergs catchment to the left. The numbers 1-3 denotes the sampling points in each stream. Area for intensive soil sampling in the Fanbergs catchment enlarged to the right. White circles denotes lysimeter sampling pits in the discharge and recharge area.

2.1 Site description

2.1.1 Study site

The field studies (papers II, III, IV, V and VI) took place in two small, bordering forested catchments, 50 and 40 ha respectively, in Bispgården (63 07N, 16 70E), central Sweden (Figure 1). Each catchment is drained by a first order forest stream.

The forest consists mainly of 50 to 80 year-old Norway spruce (Picea abies) and

Scots pine (Pinus sylvestris). The catchments are located at an altitude of 258 m above sea level, which is above the highest coastline in this area, so postglacial surficial sediments that are common to postglacial deltas are lacking. The area has only low levels of anthropogenic deposition. The bedrock is mainly composed of granite and gneiss. The soil in the recharge area is classified as podzol and in the discharge area the soil is classified as arenosol (FAO, 1990).

The investigation in this thesis involved six plots at 10 m from the stream

(arenosol) and six plots at 80 m from the stream (podzol). The plots were separated by 10-15 m each (Figure 1). The organic horizon was on average 28 cm thick in both podzol and arenosol. The podzol had an average depth of 9 cm for E-horizons and

7 cm for the B-horizons above the C-horizons. The arenosol had an average depth

14

of 50 cm for the A-horizons above the C-horizons. When sampling the mor layer, the living vegetation was removed and the remaining mor was divided into two parts. The upper part (O1) consisted of partly degraded litter and the lower part

(O2) consisted of well-degraded litter.

2.1.2 Streams

The location and distribution of wetlands and the morphology of the streams differed between the two catchments. Both streams, Fanbergbäcken and

Gråbergsbäcken, originate from small mires, with estimated areas of about 0.2 ha and 1 ha, respectively. Gråbergsbäcken, however, also receives water from a spring, making it less sensitive to drought. Like the majority of the forest streams in central and northern Sweden, these streams have been subjected to ditching, which is common in many peatlands and wet mineral soils to improve forest productivity (Ivarsson & Jansson 1994). The scars from this treatment are more visible in Gråbergsbäcken, where the streambed cuts deeper into the soil and the stream water lacks contact with the shallow organic soil layers for approximately

80 % of its length. An increase in flow would therefore raise the surface of

Gråbergsbäcken, but essentially no flooding of the banks would occur. The deeper streambed might also lead to a greater influence of groundwater from deeper flow paths in the soil. In Fanbergsbäcken, the traces of the ditching are less obvious, and the stream is shallower with a more pronounced riparian zone that lines at least 60

% of the stream and is evenly distributed throughout its length. This would cause the riverbed to expand up over the banks in several areas along the stream at high flows. Even though the areas of wetlands are similar between the catchment their location differs, with wetlands lining Fanbergsbäcken as a pronounced riparian zone, while located apart from the stream in the Gråbergs catchment.

2.2 Field methods

The sampling seasons lasted from April to November in 2003-2006 as the soil and streams were frozen in winter. The schedule of sampling methods is listed in

Table 1.

15

Table 1. Schematic description of the water sampling conducted. C - control, A - wood ash applied.

Soil solution centrifugation

Soil solution lysimeter

Stream water Fanbergsbäcken

Stream water Gråbergsbäcken

2003

C

C

C

2004

C

C

C a

C a

2005

A/C

A

C b

2006

A/C

A

C b

2007

A c

C

a

Ash was applied to the catchment in late September 2004 b

A control area without ash saved within the catchment (see Figure 1) c

Forest was harvested in the Fanberg catchment possibly causing an interfering effect

2.2.1 Stream water

Stream water samples were collected manually every second week and with higher frequency during high flows. The sampling points were located at the outlet of each catchment and at two additional sites upstream (Fan1, Fan2, Fan3, Grå1

Grå2, Grå3; Figure 1). The water was collected in polyethylene bottles and kept cool until pre-treatment within 24 h after sampling.

2.2.2 Soil sampling

In 2002, soil samples for determining soil pH and soil, exchangeable cations, and total content of nitrogen and carbon were collected from all horizons in all pits. On the same occasion soil samples for determination of total composition were taken from two plots in podzol and two plots in arenosol including all horizons except for the mor layer (E, B1, C in podzol; 0-5 cm, 10-15 cm, 45-50 cm in arenosol). Additional soil samples for determination of exchangeable cations, and pH in centrifugated soil were collected from all horizons in all pits in 2004 and

2007.

2.2.3 Lysimeter sampling

Suction lysimeters (Rhizon MOM, Rhizosphere, Wageningen, Netherlands) were installed at four horizons: O2 (lower mor layer), E, B1 (upper B-horizon) and

C in podzol soil and O2, 0-5 cm, 10-15 cm and 45-50 cm depth in arenosol. Three lysimeters were installed at each horizon in each pit. Lysimeters were used to sample the percolating soil solution that might influence stream water composition. After sampling, the soil solutions from the triplicate lysimeters were mixed to obtain a composite sample from each horizon and pit. Soil water from the lysimeters was collected every month during the sampling season.

16

2.2.4 Centrifugation

In June 2004 soil was collected to obtain soil solution samples by centrifugation according to Giesler & Lundström (1993). The soil samples were taken from O1,

O2, E, B1 and C-horizons in podzol and from O1, O2, 0-5 cm, 10-15 cm and 45-50 cm depths in arenosol. After centrifugation, the soil solution was filtrated through

0.45 µm membrane filters (Millex-HV, Millipore) before analyses.

2.2.5 Soil respiration

The soil respiration was measured every month during the growing seasons at two pits in the recharge area and at two pits in the discharge area. The measurements were made at eight points around each pit and an average value was calculated. The respiration was measured using an EGM-4 (PP-Systems, UK) equipped with a soil respiration chamber (SRC-1) and a soil temperature probe

(STP-1).

2.2.6 Application of wood ash to the Fanbergs catchment

In late September 2004 self-hardened crushed wood ash was spread in the

Fanbergs catchment, at a dosage of 3 ton DS/ha as recommended by the Swedish

Forest Agency (Anonymous 2008; Figure 1). No ash was spread in the control area and a 10 m buffer zone surrounding it (paper VI). The results from previous studies conducted in the same area (papers II, III and IV) were used as background data. The stream draining the ashed catchment, Fanbergsbäcken, was intensively monitored, and the stream draining the bordering catchment, Gråbergsbäcken, was used as an untreated reference.

The elemental composition of the wood ash, the recommended minimum amounts for some of the elements, and the composition of the top mineral soil close to Fanbergsbäcken are described in Table 2. The wood ash met the Swedish

Forest Agency standards for the elements investigated in this study (Anonymous

2008).

17

Table 2. Composition (g/kg DS) of the wood ash, recommended minimum amounts from the

Swedish Forest Agency (Min; Anonymous 2008), and the total composition found in the superficial mineral soil close to the stream (Arenosol 0-5).

Element Ash Min

Arenosol

0-5 cm

Si

Al

Ca

Fe

97.2

29.7

205.1

13.0

>125

238.9

42.8

6.6

10.9

K

Mg

Mn

Na

30.6

15.7

12.4

6.2

>30

>15

18.7

1.8

0.2

12.3

2.3 Laboratory methods

2.3.1 Soil and ash

Soil samples for determination of total composition were dried at 105 °C, melted with LiBO

4

and dissolved in HNO

3

before analysis by inductively coupled plasma mass spectrometry (ICP-MS). The analyses were performed at Analytica in

Luleå, Sweden. Samples for N and C content were dried at 70°C before analysed with a Perkin Elmer CHN 2400 elemental analyzer. These analyses were made at the Soil Science Laboratory at the Swedish University of Agricultural Sciences

(SLU) in Umeå, Sweden. Determination of exchangeable cations, soil content of N and C and pH was conducted twice, first in 2002 at HSMiljölab AB in Kalmar,

Sweden (paper II), and for centrifugated soil from 2004 and 2007 at Agrilab AB in

Uppsala, Sweden (paper VI). The total composition of the ash was determined by melting with LiBO

2

followed by dissolution in diluted HNO

3

. For determination of

As, Cd, Co, Cu, Hg, Ni, Pb, S and Zn the sample was dried at 50°C and dissolved in HNO

3

/HCl/HF in a closed Teflon container in a microwave oven. The samples were then analyzed with a combination of ICP-AES and ICP-SFMS. The analyses were performed at Analytica.

2.3.2 Soil solution and stream water

All samples were stored refrigerated until analysis unless otherwise mentioned.

The pH was determined within 24 h using a Beckman Φ32 pH-meter. For all other analyses the stream water was filtrated through a 0.45 µm membrane filter (Millex-

HV, Millipore). Samples for cation analysis (Ca, Mg, Fe, Al, K, Na, Si, Cr, Mn, Co,

18

Ni, Cu, Zn, Cd, Pb) were acidified with ultrapure HNO

3

before analysis with ICP-

MS (VG PQ ExCell, Thermoelemental, Winsford, England). DOC was determined within 48 h of sampling and quantification was achieved by catalytic oxidation and

IR detection with a Shimadzu TOC analyzer, TOC-5050A (Shimadzu, Kyoto,

Japan). Anion determination of F, Cl and SO

4

was performed within 48 h of sampling by ion chromatography with Dionex DX-120 (Dionex, Sunnyvale, CA,

USA). Flow injection analysis (FIA) with a FIAstar 5000 (Foss, Höganäs, Sweden) was used for the determination of quickly reactive aluminium (Al qr

; Clarke et al.

1992) and NO

3

within 48 h of sampling. Samples for analysis of NH

4

and PO

4

were stored frozen until analysis with FIA.

Some stream water samples were separated into a smaller size fraction

(nominal molecular weight cutoff 1 kDa), hereinafter referred to as the LMM fraction of the substance, by means of a stirred cell method (Amicon model 8050), in which the water samples are passed through a YM-1 membrane of regenerated cellulose (Millipore Corporation, Billerica, MA 01821) under a N

2

pressure of 3.5 bar. For further details on this method see van Hees et al. (2001). DOC and cations were determined for these samples and the amount of elements associated with the high molecular mass (HMM) fraction (M w

>1 kDa) was calculated by subtracting the concentration recovered in the LMM-fraction (M w

<1 kDa) from the total concentration.

Samples for determination of total acidity were stored frozen until analysis by titration. After thawing, the sample was first passed through a column with a strong cation exchange resin (Dowex 50 8), and then titrated with 0.01 M NaOH to pH 7. A constant flow of N

2

was supplied to the sample to remove any atmospheric CO

2

during the titration, which was performed with an autoburette system (702 SM Titrino, Metrohm Ltd., Herisau, Switzerland) and a glass electrode

(Mettler-Toledo). For further details on this method, see Hruska et al. (1996).

Samples for analysis of LMMOAs were stored frozen and then determined by

LC-MS/MS (paper I). The LC-MS/MS system was comprised of a Shimadzu LC-

10AD pump, an Agilent 1100 autoinjector, a Supelcogel C610-H column (300 7.8 mm) and an API3000 mass spectrometer (MDS Sciex, Concord, Canada). The mobile phase consisted of 10 % (v/v) methanol and 0.01 % (v/v) formic acid in

MilliQ-water.

2.4 Statistical evaluation

Microsoft Office Excel 2007 (Microsoft, Redmond, WA), MINITAB version 12

(Minitab Inc., State College, PA), MATLAB 6.1 (The MathWorks Inc., Natick, MA)

19

and Sigmaplot 11 (Systat Software Inc., San Jose, CA) were used for mathematical treatment and statistical evaluation of data. Unless otherwise stated, the significance level was set to 0.05, i.e. an observed effect was considered as significant and the null hypothesis of no effect rejected if it was indicated that its probability to be valid was below 5%.

Correlations between time series were evaluated by linear regression followed by t-tests to evaluate the significance of the Pearson correlation coefficients obtained (paper IV). Student’s t-test was in general used to determine the significance of differences in the means between two data series (papers IV, V, VI), and whenever appropriate, paired t-tests were used in order to minimise the influence of temporal variations in the data. When more than two means or underlying causes of variations were to be evaluated, analysis of variance

(ANOVA) was applied (paper II).

In some cases, it could be doubted whether the requirement of naturally distributed data for the above mentioned tests was fulfilled. In paper IV, a robust sign test (Miller & Miller 2000) was thus used to evaluate differences in the size fractionation of different cations expressed in percentage, and in paper V, a random intervention analysis (RIA; Carpenter et al. 1989) was used as a complement to t-tests when evaluating the impact of WAA to stream water chemistry. In RIA, random permutations of the test variable are performed in order to establish a distribution of the treatment effect, which enables determination of the likelihood of obtaining the experimental value. In this particular case, Monte

Carlo simulations with 10 000 permutations were performed for each analyte to establish the p-values for the measured effects.

Finally, cumulative sum (cusum) charts (Figure 2) was used to visualise small changes within time series and to evaluate whether observed effects are persistent or due to more abrupt, short-lived changes in the water chemistry (paper V).

20

Figure 2. Seasonal variation of stream water concentration of Ca in Fanbergsbäcken (Fan) and Gråbergsbäcken (Grå) in 2003-2006 (left) and the concentration of Ca in

Gråbergsbäcken subtracted from Fanbergsbäcken and the cumulative difference in Ca

(right).

3. DEVELOPMENT OF AN ANALYTICAL METHOD FOR

ANALYSIS OF LMMOAS IN NATURAL WATERS

LMMOAs have frequently been analyzed in soil solution, where they are present in relatively high concentrations and may constitute of up to 10 % of the total DOC in soil solution (Strobel 2001; van Hees et al. 2000). Comparatively few investigations of stream water have been conducted, as this water is usually more diluted, contains less organic matter and requires a more sensitive analytical method. Of the acids investigated in the current study, only the levels of lactic acid and citric acid had previously been reported in surface waters (Hlavacova et al.

2005; Jonsson et al. 2007). A robust and selective method with low detection limits for LMMOAs present in natural waters was therefore developed as part of the research work described in this thesis. The new method coupled HPLC on-line, via an electrospray ionization interface, to mass spectrometry detection.

3.1 Separation

A number of other methods have previously been applied in the separation of

LMMOAs in natural waters. Most common was the use of ion exclusion chromatography (van Hees et al. 1999) and capillary electrophoresis (Hagberg

2003; Hagberg et al. 2002). Since ion exclusion chromatography is well compatible with MS-detection, the original method by van Hees et al. (1999) was here selected for further development.

21

Ion exclusion chromatography of acidic compounds uses a negatively charged stationary phase, typically partly sulfonated divinyl benzene, which repels anions due to so-called Donnan exclusion. Compounds, exhibiting a negative charge under the chromatographic conditions applied, will elute in the void volume of the column. Nonionic or weakly ionic substances, on the other hand, will be retained as they can penetrate into the pores of the packing material. Thus, in pure ion exclusion chromatography, the retention time of a solute is solely dependent on its net charge. Size exclusion effects and secondary interactions with the stationary phase such as van der Waal and hydrophobic interactions will, however, also influence the separation to various degrees and add to the selectivity of the system.

Thus the LMMOAs will separate on the ion exclusion column primarily in order of their pK a

values, and secondarily according to size and other properties of the analyte.

3. 2 Detection

Low UV-range spectrometry (~210 nm) has been used to detect aliphatic acids, which essentially lack strong chromophores (Hees et al. 1999). In this UV-range a large number of organic substances absorb light, making peak separation difficult and peak identification uncertain due to interference and coelution. An alternative is to apply indirect UV detection as described by Hagberg (2003), but this mode of operation also suffers from selectivity problems. In the work described in this thesis, electrospray ionization (ESI) in combination with a tandem mass spectrometer (MS/MS) improved the overall selectivity and attained the low detection limit needed for the applications intended. In electrospray, the liquid mobile phase gets converted into a fine arenosol of charged droplets that dries until a critical volume where a coulombic explosion occurs, a process finally resulting in the production of gas phase ions from the analyte. The ions then enter the MS where their mass-to-charge ratios (m/z) and abundances are measured

(SIM mode). Alternatively, the ions can be accelerated into a collision cell, where collisions with inert gas molecules results in fragmentation producing a unique mass pattern for detection.

A number of adjustments were made to modify the original method based on

HPLC-UV developed by van Hees et al. (1999) to allow proper function with MS detection. The eluent was adapted to the ESI-interface by exchanging the phosphoric acid with formic acid (0.01 % v/v) for its higher volatility, and by adding 10 % methanol the hydrophobic interaction with the stationary phase was lowered and a more stable electrospray was obtained. A multiple reaction

22

monitoring mode was used for detection and the 17 acids included in the method were separated within 30 minutes (Figure 3).

Figure 3. A 5 µM standard of 17 different LMMOAs analyzed with LC-ESI-MS/MS

3. 3 Further development

The chromatographic separation for the structural isomers, fumarate and maleat and trans- and cis-aconitate, was excellent (Figure 3). Citrate and isocitrate may, however, due to their similar retention time constitute a problem while integrating, especially if large differences in concentrations are present. A high signal from citrate may shield or in the worst case completely swallow a low signal from isocitrate. To circumvent this, an additional mass channel, unique to isocitrate, has been added to the method. Figure 4 shows the product ion spectrum for citrate with the mother ion m/z=191 and the main daughter ion at m/z=111 representing neutral losses of two H

2

O and one CO

2

.

23

Figure 4. ESI-MS/MS spectrum for citrate after collision induced fragmentation at E lab

=18 eV.

The same mass channel was used for isocitrate in the original method (paper I).

Isocitrate, however, has several more usable fragments that are not formed in the fragmentation of citrate, and as a complement the ion with m/z=117 was chosen

(Figure 5). This mass channel does not possess the same sensitivity as the original, but it contributes to a more definite identification of the isocitrate peak, and may also be used for quantification in difficult samples.

24

Figure 5. ESI-MS/MS spectrum for isocitrate after collision induced fragmentation at

E lab

=23eV.

4. SOIL, SOIL SOLUTION AND STREAM WATER

The result from the pre-ash application study, based on the data sets marked with a C in table 1 are discussed in this section.

4.1 Soil solid

The parent material had a similar composition throughout the investigated slope in the catchment. A trend towards higher carbon content was found throughout the soil profile in the arenosol compared to podzol with the exception of the O horizon (paper II).

The higher carbon content in the discharge areas was due to less oxidizing conditions and a slower decomposition rate induced by a wetter environment. The content of exchangeable Ca was generally higher in arenosol throughout the soil profile, and the exchangeable Mg, K and Na contents were higher in superficial layers of the arenosol compared to podzol. This resulted in a higher base saturation and a trend towards a higher content of exchangeable

25

cations in all mineral horizons in arenosol, probably an effect of the higher carbon content (paper II).

4.2 Soil respiration

Soil respiration measurements did not show any significant difference between the recharge and discharge area and varied in the range of ~1-10 µmol CO

2

/m

2 s over the 2003-2004 sampling season (Figure 12). The activity of biota in the superficial soil was therefore assumed to be similar for podzol and arenosol. Any difference in soil solution most likely reflected soil properties and soil water content rather than biotic activity.

4.3 Soil solution and stream water

The concentration in the E- and O-horizon of most of the measured substances peaked in late summer due to build up of products of weathering and degradation of biota that were subsequently washed out during autumn rains. The concentration of many of the measured substances differed between the recharge and the discharge area (Table 3). As seen in Table 3, most of the analyzed substances are higher in the discharge area, which is in agreement with Johnson et al. (2000) who found that both soil and drainage waters showed a consistent pattern of nutrient enrichment downslope when investigating biogeochemical processes in small catchments.

Table 3. Significant differences in seasonal averages for soil solution from lysimeter sampling 2003-2004. P= significantly higher in podzol (recharge area), A= significantly higher in arenosol (discharge area) (p≤0.05).

2003

O

E 0-5

B1 10-15

C 45-50

2004

O

E 0-5

B1 10-15

C 45-50 pH DOC NO

3-

F Cl SO

42-

Na Mg Al Si K Ca Mn Fe

A

A

A

A

A

A

P

P

P

P

P

P

A

A

A

A A

A

A

A

A

A

A

P

A

A

P

A

A

A

A

A

P

A

A A

A

A

P

A

A

A

P

A

A

A

A

A

A

A

A

A

A

A

A

A

A

A

A

A

A

A

A

26

There were large seasonal variations in soil solution during the 2003-2004 sampling seasons. The 2003 growing season was characterized as normal to dry with a major storm event in the middle of August, resulting in a tenfold increase in stream flow for a couple of hours. This resulted in a large response in the stream water chemistry, which dominated the seasonal pattern and could be traced by rapid changes in concentration for most of the different parameters measured

(paper IV). The 2004 season was dry, lacking heavy rain events and thus any large fluctuations of the stream water chemistry (papers II and IV).

4.3.1 pH and DOC

The pH increased with depth in both arenosol and podzol in soil solution recovered with both centrifugation- and lysimeter sampling. The pH was higher and DOC lower in the three lowest soil horizons of the podzol compared to corresponding horizons in the arenosol for both 2003 and 2004, demonstrating the impact of DOC on pH (papers II and III).

DOC generally decreased with soil depth in both recharge- and discharge areas, but the decrease was more pronounced in the recharge area, likely due to the constant vertical transport in this soil, inducing this pattern through filtering, adsorption or degradation across the vertical soil profile.

In the lysimeter samplings pH and DOC were usually low during the first sampling occasion in early spring, likely due to acidic water from the snow melt penetrating the soil in combination with frozen conditions and thus lack of contact with the soil organic matter. During the rest of the seasons there was a tendency towards increasing DOC, due to increased degradation of organic matter and microbial activity, and also decreasing pH, as a consequence of increased DOC and nutrient uptake by the flora (paper II).

During the driest periods in the middle of summer 2003 and in the end of summer 2004, pH in the lower horizons peaked as there was no water percolating downwards from the upper and more acidic horizons (paper II).

The pH and DOC in Fanbergsbäcken was usually in the same range as the deeper soil horizons in the discharge area, indicating these as a main contributors to stream water chemistry, except for in early spring and the storm event in 2003 where increased water content in the soil and presumably overland flow resulted in the stream water mirroring the more superficial soil layers (papers II and IV).

27

Fanbergsbäcken had a significantly lower pH and higher DOC content than

Gråbergsbäcken. The pH and DOC were negatively correlated in both streams with a major decline in pH and a corresponding increase in DOC occurring during the storm event in 2003. In 2004 as well there was a significant correlation between these parameters in Fanbergsbäcken but not in Gråbergsbäcken (paper IV).

Earlier studies of stream water chemistry during spring flood and autumn storms in northern Sweden have shown that DOC is the most important driving mechanism of the pH decline in forested catchments (Laudon & Bishop 2002). The organic acidity is thus of major importance for the pH of the stream water. At high flow, superficial flow paths that activate new sources and an increased washout of

DOC have been proposed as the reason for the large increase in DOC (Bishop &

Pettersson 1996). The same patterns were observed in both streams in this investigation, and DOC was significantly correlated between the streams in both

2003 and 2004 and pH in 2003, which indicates that the same response was triggered by the precipitation, and that the process was independent of the different concentrations of DOC (paper IV). The flow induced by the precipitation governs the concentration of DOC as also found by Haaland et al. (2010) and Riise et al. (1994).

4.3.2 DOC fractionation

The 2003 stream water and the soil solution and stream water from the centrifugation sampling in 2004 were subjected to fractionation of DOC (papers III and IV). There were large differences in size distribution between the recharge- and the discharge area in the soil. The recharge area had a higher relative proportion of LMM DOC (paper III). This is in accordance with the podzolisation theory, where LMM compounds play a crucial role. The relative amount of HMM

DOC also decreased with depth in accordance with the findings by Riise et al.

(2000), who suggested that HMM DOC were filtered or adsorbed when passing through the podzolic soil. In the discharge area the fraction of HMM DOC was higher, likely due to different flow conditions (less vertical filtering) and waterlogged conditions for parts of the year, giving rise to anaerobic environments slowing down the degradation of the HMM substances into smaller sizes (Figure

6).

The DOC concentration was much lower in stream water than in the soil solution. The relative amount of HMM DOC in stream water, however, sampled at the same time as the centrifugation sampling of soil solution was made, was in the same range as in the soil solution of the shallower soil layers of the arenosol

28

indicating a close relationship between the discharge area and the stream water.

During 2003 the fractionation of DOC from both streams showed high amounts in the HMM fraction, approximately 70 % in Fanbergsbäcken, while it was somewhat lower, around 60 %, in Gråbergsbäcken. The concentration of both DOC and HMM

DOC was significantly correlated between the streams (paper IV).

Figure 6. Size distribution for DOC in soil solution from centrifugation sampling 2004 in

the recharge (left) and discharge area (right).

4.3.3 Carboxylic content

The carboxylic concentration was estimated by titration of soil solution from the centrifuged samples and of stream water. There was a high content of carboxylic groups in the organic horizons that were not associated with cations recovered in the HMM fraction, which indicates the presence of free carboxyl groups or carboxyl groups belonging to the LMM fraction (Figure 7). In the mineral horizons of the arenosol, HMM organics increased with depth. Here, there were no unassociated carboxylic groups and thus no LMM complexes. Instead an anion deficit (AD) was found indicating the presence of HMM inorganic complexes

(Figure 7).

29

Figure 7. The total carboxylic content (striped bar) and the cations recovered in the high molecular fraction of DOC (stacked bars) in soil solution from centrifugation sampling

2004 in the recharge (left) and discharge area (right).

The carboxylic concentration was correlated between the streams for both sampling seasons and also to DOC in both streams, but showed a higher concentration (µeq/l) in Fanbergsbäcken in accordance with the higher content of

DOC in this stream. The difference in site density for the streams was large (Table

4). Fanbergsbäcken was well within the range of the universal model proposed by

(Oliver et al. 1983) and of that reported for Swedish sites by Bergelin et al. (2000) and Hruska et al. (2003).

In Gråbergsbäcken the temporal variation of site density was greater with higher averages. The discrepancy present between site densities of the streams might be attributable to the large differences in DOC concentration and fractionation. In Gråbergsbäcken the relative fraction of LMM DOC was larger, indicating a greater presence of fulvic compounds with higher charge density

(COOH/C-ratio; Stevenson 1993). It should be noted however that, compared to

Fanbergsbäcken, the low DOC concentration and higher concentrations of the strong acid anions F and NO

3 in this stream would yield a higher uncertainty for the calculated site density values due to error propagation (Hruska et al. 2001). The highest site density notation was for Gråbergsbäcken in 2004, which was a very dry year with the lowest DOC values detected.

30

Table 4. Site density for stream water (µeq/mg C) in the outlet of the Fan- and Grå catchment, the third sampling point in Fanbergsbäcken (Fan3) and Gråbergsbäcken (Grå3).

2003

2004

2005

2006

Fan3

average stddev

10.02 1.47

9.40

9.49

8.72

3.03

1.81

1.99 n

36

28

23

27

Grå3

average stddev

14.13 6.84

20.83

15.73

14.07

13.95

9.39

6.93

4.3.4 Low molecular mass organic acids (LMMOAs)

n

36

20

20

30

LMMOAs are small acids that are ubiquitous in soil solution and have high complex binding abilities to cations and metals. They play an important role in soil processes such as weathering of minerals in the podzolation process (van Hees &

Lundström 2000) and are recovered in the LMM fraction of DOC (<1 kDa). Using the method developed in paper I, LMMOAs were analyzed in both soil solution from centrifugation and stream water.

There was no significant difference in the amount of LMMOAs recovered in the podzol and the arenosol, which was expected due to the difference in DOC concentration. The LMMOAs comprised <3.5 % of the DOC in the podzol, and <1.5

% of the DOC in the arenosol.

The most common LMMOAs recovered in the soil solution were, in declining order, citric-, malic- and shikimic acid (Figure 8). All acids, except lactic acid, were found in the highest concentration in the O-horizons and diminished with depth in the soil profile, similar to the DOC concentration (Figure 6).

31

Figure 8. The vertical distribution, average values with standard error (n=6), of the three most common LMMOAs in soil solution from the centrifugation sampling in 2004.

There have been few investigations of LMMOAs in stream water, possibly because of the low concentrations present in this medium. The method developed in paper I proved to be sensitive enough for this purpose, however. LMMOAs were measured in Fanbergsbäcken during 2003 and 2004 and Gråbergsbäcken during 2003-2007. The total amount of LMMOAs did not differ significantly between the streams in 2003, but in 2004 it was higher in Fanbergsbäcken.

The most common acids in the stream water were oxalic- and lactic acid, of which peak concentrations of oxalic acid coincided with those of DOC. Oxalate in the stream water in 2004 was in the same concentration range as found in the Olayers of the discharge area/arenosol in the centrifugation sampling the same year.

In Gråbergsbäcken the seasonal variation for DOC and oxalate were correlated

(p<0.001) for the entire investigated period (Figure 9), while no such correlation was seen in Fanbergsbäcken during 2003-2004 (although a positive correlation could be seen for oxalic acid in the first two sampling points in Fanbergsbäcken

2003). There was no relation between the concentrations of DOC and lactate in either of the streams, which seems consistent with its behavior in soil. Some of the most common acids in the soil solution, shikimate, citrate and malate, were hardly ever found, and only then in very low concentrations in stream water, and therefore did not seem to make the transition from soil to stream water in the same way as oxalic acid did.

One of the few investigations of LMMOAs in streaming water by Jonsson et al.

(2007) reported much higher concentrations of citrate than found in this investigation, and also a negative correlation of LMMOAs to DOC in the inlet and

32

outlet of a subarctic lake in northern Sweden. This indicates a very different

LMMOA constitution compared to the present investigation, and highlights the influences of the environment on these substances.

Figure 9. Seasonal variations for DOC (n=147) and oxalate (n=124) in Gråbergsbäcken during 2003-2007.

4.3.5 Base cations and their association to DOC

The association of cations and metals to DOC was investigated in the stream water in 2003 and in the soil solution sampled by centrifugation in 2004.

The monovalent base cations, Na and K, were not significantly associated to

HMM DOC in either soil solution or stream water. Na and K were most likely present as free ions and experienced only weak or no association to DOC. K showed large variations in the O- and E-horizon in the soil solution from lysimeter sampling due to high recirculation of this substance in biota (paper II) while Na that was more abundant, being less susceptible for plant uptake, was found in higher concentration in the mineral soil. Both substances showed only minor variations in the stream water over the season (papers II and IV).

33

Ca and Mg were found in the highest amounts in the topsoil solution and then diminished with depth in the soil profile in both the recharge- and the discharge area. During the summer months both Ca and Mg tended to increase in the soil solution, probably due to weathering and degradation of biota, and then a decrease was seen in the autumn due to wash out with the increase in precipitation and decreasing microbial activity, leading to a lower weathering rate (papers II and III).

Fifty percent of the Ca was found to be associated to HMM DOC in solution throughout the soil profile both in podzol and arenosol, except for the organic horizon in arenosol where the fraction associated to HMM DOC was 60-80 %

(paper III). The distribution of Mg in soil solution was similar to Ca, except for a slightly higher HMM fraction of Mg in the mineral soil in the discharge area.

There was a higher concentration of Ca in Gråbergsbäcken than in

Fanbergsbäcken during both seasons, while the opposite was observed for Mg. The higher concentration of Ca in Gråbergsbäcken might be attributable to the greater influence of groundwater due to the deeper stream channel in this stream. Both Ca and Mg were positively correlated between the two streams during 2003 and 2004, and both were positively correlated with DOC and its HMM fraction in both streams during 2003, which indicates that these elements have a considerable association with organic matter in stream water. Surprisingly high amounts of both

Ca and Mg were associated to HMM DOC in both streams during 2003 (Table 5), and peak values of around 80 % of Ca and Mg associated to HMM DOC were reached in Fanbergsbäcken during the high flow event in August, with slightly lower values for Gråbergsbäcken (paper IV).

Such high association of Ca to HMM DOC has previously been reported by

Benedetti et al. (2003) from river water in Rio Negro, Brazil. A similar range was also found in the soil solution sampled by centrifugation in the Fanbergsbäcken discharge area in 2004, as mentioned above (papers III and IV). This confirms a superficial flow in the soil during high flow/high precipitation events where the stream water adopts the chemistry of the soil it was last in contact with (similar to findings by Bishop et al. (2004)). Despite this, Dahlqvist et al. (2004) reported a considerably smaller association of Ca to DOC in a study of river water from

Kalixälven in northern Sweden, and noted a risk of overestimation of the HMM Ca with use of conventional ultra filtration methods, as used in this study. On the other hand, DOC quantity and quality differs substantially between headwaters and large rivers and also between catchments. This subject therefore requires further investigation.

34

Table 5. The median value (1 st - 3 rd quartile) of the fraction of the different substances recovered in the HMM-fraction (M w

>1 kDa) in the third sampling point (the outlet of the catchments) of Fanbergsbäcken and Gråbergsbäcken during 2003. Sign tests (Miller &

Miller 2000) were performed on seasonal series to evaluate differences between the streams.

DOC (n=29,31)

Ca (n=17, 17)

Mg (n=17, 17)

Al (n=17, 17)

Fe (n=17, 17)

Fan3

68% (55-81%)

54% (47-67%)

48% (36-60%)

73% (61-86%)

79% (73-87%)

Grå3

58% (50-69%)

35% (23-45%)

24% (19-37%)

67% (50-82%)

40% (0-71%)

P (sign test)

0.043

0.013

0.057

1

0.023

4.3.6 Al and Fe and their association to DOC

In the recharge area with a clear podzol profile, the concentrations of Al and Fe in solution were decreasing with soil depth in the mineral soil due to immobilization/precipitation. Fe was found almost entirely in the LMM fraction in the B-horizon while in the other horizons the association to HMM DOC was around 70 % (Figure 10). Fifty percent of the Al was recovered as HMM in the Ohorizons, while the association was slightly higher (~75 %) in the mineral horizons

(Figure 10;paper III).

Figure 10. Size distribution for Fe and Al in soil solution in the recharge area from the centrifugation sampling 2004.

An increasing trend with depth in the arenosol soil profile was visible for Al and Fe, and both substances were found in highest concentrations in the deep mineral soil solution. Compared to podzol, a higher proportion of Al and Fe were bound to HMM organics, highest in the shallow mineral soil (0-5 cm) where the association was 95 % for Fe and 90 % for Al.

35

The seasonal variations of Al and Fe concentration in soil solution were somewhat less clear than those for Ca and Mg, but generally followed the same pattern with increases during the summer months and a decrease in the autumn.

The highest concentrations for both substances, however, were usually recovered in the E- and 0-5 horizons in the lysimeter investigation, clearly differing from the centrifugation investigation of the arenosol (Figure 11). Previously, differences in the recovery of different elements with different sampling techniques i.e. centrifugation and tension lysimeters among others, has been described by Geibe et al. (2006).

Figure 11. Size distribution for Fe and Al in soil solution in the discharge area from centrifugation sampling 2004.

Both total Fe and total Al were found in significantly higher concentrations in

Fanbergsbäcken than in Gråbergsbäcken, and the concentration of Al was substantially higher than Fe in both streams (paper IV). Total Fe and Al were significantly correlated with DOC in both streams during 2003 and 2004, except for

Fe, which was not correlated to DOC in Gråbergsbäcken during 2004 likely due to the low concentrations and lack of high flow events. The correlations emphasize the importance of DOC for the transport of cations and metals in stream water.

Fanbergsbäcken had a significantly higher amount of Fe associated with the

HMM fraction (Table 5). The lower association of Fe to HMM DOC in

Gråbergsbäcken was possibly due to higher fraction LMM DOC and the low total concentration of Fe. Björkvald et al. (2008) and Kortelainen et al. (2006) have shown that the hydrogeochemistry of Fe is highly dependent on the catchments coverage of wetlands, and in this investigation the greater abundance of wetlands lining the

36

stream channel in Fanbergsbäcken might explain the significantly higher concentrations of Fe in this stream.

During the 2003 season, a majority of the Al in the streams was found associated to the HMM fraction of DOC (Table 5), in accordance with previous studies of the behaviour of Al in soil solution (Riise et al. 2000; van Hees et al. 2001; paper III). There was no significant difference in the association of Al to the HMM

DOC between the streams (Table 5). The HMM fraction of Al was strongly correlated to both DOC and total Al in both streams. During 2003, the concentrations of Fe and Al were positively correlated between the streams, but in

2004 no such correlation was noted. This is presumably a direct effect of the low concentrations of these substances in combination with the absence of high flow events during the 2004 season.

5. IN-STREAM VARIATIONS

The in-stream variation was more pronounced in Fanbergsbäcken where many of the measured parameters showed significant differences between all three sampling points, while in Gråbergsbäcken differences were mostly found between the last two sampling points. In-stream differences were most noticeable for pH, where Fanbergsbäcken showed a significant decrease between all three sampling points downstream in both 2003 and 2004, in accordance with a significant increase of DOC. In Gråbergsbäcken a significant increase in pH occurred between all three sampling points in both years, without the corresponding decreases in DOC.

Fanbergsbäcken showed higher concentration values downstream for Al and Fe as these substances were highly associated to DOC in this stream. In

Gråbergsbäcken, Fe increased between all three sampling points downstream, while DOC, Ca and Al increased between the second and the third sampling point

(Table 6). The increase of Fe and Al and the decrease of pH downstream in

Fanbergsbäcken is clearly a consequence of the increasing DOC concentration; this is also visible in Gråbergsbäcken between the last two sampling locations for Fe, Al and possibly Ca (paper IV).

37

Table 6. Differences (paired t-test, P<0.05) for in-stream variations between sample points in Fanbergsbäcken and in Gråbergsbäcken during the 2003 sampling season. should be interpreted as an increase and as a decrease between the sampling points.

Fan1 Fan2 Fan2 Fan3 Grå1 Grå2 Grå2 Grå3 pH

DOC tot-LMMOAs

Ca

Al

Fe

Cl

F ≡ ≡

6. WOOD ASH APPLICATION (WAA)

6.1 The effect of WAA on stream water- and soil solution chemistry in the

Fanbergs catchment

Spatially uniform application of wood ash in forests is nearly impossible, due to the variable cover and difficulty of access. Irregularities in the ash distribution were avoided to a certain degree at the sampling site, as the ash was spread by hand. However, a perfect evenness of the ash layer was not certain and the lysimeter sampling, representing a small volume of soil solution close to the installation, may therefore reflect this. To compensate, the samples from three lysimeters were pooled at each pit and soil solution concentrations were averaged for three pits representing, at the most, nine lysimeters for each horizon and each

“treatment” (ash podzol, ash arenosol, control podzol and control arensosol).

Monitoring the water chemistry in the last sampling point in Fanbergsbäcken is representative of the discharge processes from the whole catchment, 50 ha, where

150 ton of wood ash had been applied.

6.1.1 Respiration

Soil respiration measurements were not different between the ashed area and the control area, and varied in the range of ~1-12 µmol CO

2

/m

2 s over the sampling season (Figure 12). In 2005, the first year after WAA, the respiration seemed to increase, but this increase was present in both the ashed area and the control area.

No change in respiration could therefore be related to the WAA in this investigation.

38

Figure 12. Respiration measurements (µmol/m 2 s) 2003-2006 in the recharge area (a) and the discharge area (b) in the Fanbergs catchment.

6.1.2 pH and DOC

No difference in pH resulting from WAA was found in soil solution at any depth in either the podzol or the arenosol in this investigation. No difference in

DOC was noted either, even though a tendency towards an increase could be seen in 2005 and 2006 for the O- and 0-5 horizon in the arenosol. High seasonal variations, foremost in the superficial soil layers, made it difficult to distinguish any effects caused by the WAA from the natural variations (paper VI). The pH of the treated area’s stream water after WAA remained unchanged.

A significant increase in DOC concentration in Fanbergsbäcken compared to

Gråbergsbäcken was observed, however. The lack of corresponding decrease in pH may indicate a pH-related change since this shows a deviation from previously established strong negative relationships between pH and DOC in these streams during the 2003-2004 pre-treatment period (paper IV).

The WAA did not affect the total LMMOAs concentration in Fanbergsbäcken and there was no difference during the investigated period in Gråbergsbäcken either. During the four sampling seasons, oxalic acid was the most dominant acid in Fanbergsbäcken whereas it alternated between oxalic- and lactic acid for

Gråbergsbäcken. Malonate was the only of the 17 analysed LMMOAs that seems to be affected by WAA. There was an increase in Fanbergsbäcken when looking at the average differences (paper V). Data on individual LMMOAs in stream water are scarce in the literature, but malonate has previously been found to increase in soil solution in superficial soil layers at a forested site in southern Sweden due to treatment with 8.75 tonnes/ha dolomite (Holmström et al. 2003), indicating that

39

changes in the production of this organic acid can be induced by catchment scale treatments with ash or lime.

No increase could be seen in the total carboxylic concentration of the stream water in the present study, despite an increase in DOC, which would imply a shift towards more hydrophobic DOC as previously observed by Holmström et al.

(2003). No decrease in site density was found to confirm this, however. This means that the small increase of carboxylic groups compensated the increase in DOC and

WAA did not induce any significant changes in the quality of the DOC.

Increased DOC concentration in soil caused by application of lime and wood ash has been observed in several studies and attributed to a fertilization effect occurring from the treatment that increases the biological activity and therefore the turnover time for different carbon compounds in the soil (Nilsson et al. 2001;

Perkiömäki & Fritze 2002).

In a study of both soil solution and stream water chemistry over a five-year period following the application of 2.2 ton/ha of granulated fly ash to a small catchment area, Fransman & Nihlgård (1995) found an increasing pH tendency in runoff, while no significant change was seen in soil solution at 30 cm depth. A slight increase of the pH in a brook running through mineral soil was also noted by

Tulonen et al. (2002) one year after application of 6.4 ton/ha of wood ash to a small subcatchment.

Geibe et al. (2003) observed higher pH in the O2 horizon, but lower pH deeper in the soil horizon in comparison to a control area four years after application of

4.28 ton/ha of wood ash, possibly due to ion exchange of protons with applied cations. Higher DOC concentration was reported in the topsoil of the sites subjected to WAA, compared to the control (Geibe et al. 2003; Holmström et al.

2003).

When investigating WAA to a pine stand in northern Germany, Rumpf et al.

(2001) noted an initial decrease in the pH of superficial soil layers in a sandy podzol after application of 2.4 ton/ha wood ash. Twenty-four months after WAA, pH in the treated plots was still lower than in the control. This indicates the same ion exchange effect as noted by Geibe et al. (2006) through WAA.

Ring et al. (2006) observed an increased pH in soil solution at 50 cm depth four years after application of 9 ton/ha of crushed wood ash, and elevated TOC

40

concentration in soil solution one year after application of 3, 6 and 9 ton/ha. Other studies have also noted increased pH-values in the soil of superficial soil layers after application of different doses of wood ash to forest soil (Eriksson 1998a;

Jacobson et al. 2004; Saarsalmi et al. 2001a).

Increased pH of stream water is a desirable effect since it may reduce mobilization of inorganic Al. In this investigation we found no increase of pH and no reduction of Al qr

or total concentration of Al after WAA in either soil solution or stream water. The species in the ash foremost responsible for pH increases are OH and HCO

3

, however large amounts of base cations may induce an increased ion exchange, releasing exchangeable protons and possibly inorganic Al. This would lead to a more acidic soil solution as suggested by Lundström et al. (2003b), and the effect may therefore be a potential suppression of the increase in pH (paper VI).

6.1.3 Ca and Mg

Lysimeter soil solution data for cations and metals were fewer in quantity than for pH and DOC, especially in the superficial soil layers, due to periodically dry conditions in combination with the requirement of a larger sample volume. This, in combination with large seasonal variations, made it hard to achieve statistical significance and results are reported as “trends” or “indications” in some cases.

Large amount of base cations were added to the soil with the wood ash and foremost was Ca, which compared to the area closest to the stream, was added in amounts more than 30 times (g/kg DS) higher than the amount present in the most superficial mineral soil layers (Table 2). Effects of WAA on Ca or Mg concentrations were hard to distinguish by visually examining the data pattern, but t-test analysis revealed differences not visible before the WAA appeared for Ca in the mineral soil of the arenosol in 2005 (0-5; p=0.03, 10-15; p=0.003, 45-50; p=0.02; paper VI). This indicates an increase in the ashed plot compared to the control plot after WAA. This enhancement in Ca-content was still visible in the 0-5 horizon in

2006 (p=0.04). There were no detectable changes in Mg following WAA.

Geibe et al. (2003) reported significant increases of both Ca and Mg in soil solution achieved by centrifugation for the whole soil profile four years after

WAA, and a similar result was observed for areas treated with dolomitic lime. The same investigation found that Ca was retained to a greater extent in the shallower soil layers, while Mg was more mobile and recovered to a greater extent in deeper soil layers. Higher mobility for Mg than for Ca have also been reported in limed plots by van Hees et al. (2003). Ring et al. (2006) found only elevated Ca in soil

41

solution from lysimeters however, from the highest wood ash dose (9 ton/ha) four years after application.

The stream water investigation yielded much higher resolution than soil sampling, which made it easier to distinguish small changes in concentration.

There were significant increases in both Ca and Mg after WAA (paper V; Figure

13), which is perhaps not that surprising considering that large amounts of these substances had been added to the catchment (Table 2). Both substances in the preinvestigation were associated to DOC in a similar fashion (papers III and IV), and the increased concentration of DOC in Fanbergsbäcken seen after WAA might partly explain the increase of Ca and Mg in the stream. The addition of Ca to the catchment was far greater than that of Mg however - 30 ton Ca in total compared to 2 ton Mg - while the percent increase in stream water after WAA was similar for both substances (paper V). This indicates a somewhat faster release of Mg from the ash.

Fransman & Nihlgård (1995) observed an increase in Ca, although not statistically different from the reference, in a stream draining a treated catchment; no change was found in soil water from 30 cm depth though. Mg in stream water remained unchanged as a result of WAA, but an increase was noted in stream water in a limed catchment, the latter confirmed by Hindar et al. (2003). Tulonen et al. (2002) observed no change in Ca in two brooks draining a small wood-ashed area. Steenari et al. (1998) noted total Ca losses of up to 40 % in both a simulated weathering study, and when extracting weathered ash particles from the forest floor two years after WAA, while a much slower leaching rate was reported for

Mg.

Figure 13. Mg in Fanbergsbäcken and Gråbergsbäcken 2003-2006 (a) and cusum chart of differences (b). The vertical line denotes WAA to the Fanbergs catchment.

42

6.1.4 K and Na

Large variations in K concentration were visible in the soil solution of all sampled plots in 2005, although these variations were slightly larger in the ashed area (paper VI). The concentration of K increased significantly in the O-horizon of the podzol (p=0.009), and there were indications of elevated concentrations compared to control even in 2006. K is highly mobile and is known to increase in soil initially after WAA (Kahl et al. 1996). Ring et al. (2006) found significant increases in K in soil solution from 3, 6 and 9 ton/ha of crushed wood ash, and the effect persisted for 4, 7 and 9 years respectively after treatment. Geibe et al. (2003) found an increase in K at a depth of 25-30 cm when investigating soil sampled by centrifugation. Several studies have also reported increases in exchangeable K after

WWA (Jacobson et al. 2004; Saarsalmi et al. 2001b) but the increase was usually located in a deeper horizon or disappeared faster than the increase in Ca due to the higher mobility of K (Kahl et al. 1996).

In stream water K concentration showed the most pronounced increase after

WAA. This is clearly distinguishable in the cusum chart in the spring 2005 and

2006 and at a high flow event in the middle of the 2006-season (paper V). Eriksson

(1998b) demonstrated in a laboratory study that K leached very rapidly from the mor layer, implying weak interactions with organic soils. This may explain the large increases that can be seen at high flow, as K is washed out from superficial soil layers. Jacobson et al. (2004) verified this behaviour for K in soil 5 years after application of wood ash, where increased concentrations of exchangeable K were found much deeper in the soil profile than Ca and Mg for instance, where the latter substances increased in the mor layer. Nieminen et al. (2005) observed significant decreases in K in investigations of ash particles 3 and 5 years after ash application to the topsoil layer. K is thus released very quickly from the ash particles and, as this study demonstrates, also reaches the runoff shortly after ash application to forest soil. Tulonen et al. (2002) and Fransman & Nihlgård (1995) observed similar increases in K in stream water, which seems to be the most mobile of the cations and therefore shows the strongest leakage pattern.

No significant changes in Na concentration were detected in either soil solution or stream water, which was unexpected as Na is known to increase soil and soil solution after WAA (Kahl et al. 1996; Ring et al. 2006). In this investigation loose wood ash was spread in late September, possibly leading to a low penetration of the soluble substances of the soil before the frost, which would contribute to a large washout in early spring from the soil surface or superficial soil layers, similar to that seen for K. Na is more abundant however, and is less limiting nutrient in

43

forest soils and therefore would probably leach to deeper soil layers at a faster rate than K. This in combination with a lower amount off Na added with the ash would make any possible effect too low to distinguish.

6.1.5 Transition elements and other metals

The only effect of WAA on the other metals analyzed in the soil solution was an increase of Fe compared to the control in the E horizon of the podzol in 2006 (paper

VI). Fransman and Nihlgård (1995) observed a decrease in Fe in soil solution from

30 cm depth after application of 2.2 ton/ha of wood ash. They attributed this decrease to the increase in pH, making Fe less soluble. Reductions of Fe in water from brooks draining a wood ash treated (6.4 ton/ha) area have also been reported by Tulonen et al. (2002). There were no increases of any metals in stream water in that investigation.

6.1.6 Anions

Anions are present in high amounts as counter ions in the ash, and Cl and SO

4 have most prominently shown rapid initial leaching from wood ash in experiments

(Steenari et al. 1998). No ash-related effect was seen for Cl in any of the investigated soils although the seasonal variations were very high and may have been masking potential effects (paper VI). Cl was however the only anion observed to substantially increase after WAA in the stream (Figure 14; paper V)), similar to findings reported for stream water by Tulonen et al. (2002).

Figure 14. Cl concentration in Fanbergsbäcken and Gråbergsbäcken 2003-2006 (a) and cusum chart of differences (b). The vertical line denotes the WAA to the Fanbergs catchment.

A higher concentration of SO

4

in the three mineral horizons of the ashed arenosol was detected during 2005 (paper VI). Of these horizons, the 10-15 horizon showed higher concentrations already in 2003 and 2004, but the increase in 0-5 and

44

45-50 horizons was newer (p=0.009, p=0.012). In 2006, the only notable increase was in the O-horizon. Somewhat unexpectedly, SO

4 remained unchanged in stream water. Previous observations have shown increases of SO

4

in soil solution after

WAA (Ring et al. 2006) and it was also detected in high concentration in initial leaches in a laboratory leaching experiment of self-hardened crush ash by Steenari et al. (1999). Ludwig et al. (2002) noted large increases in SO

4

concentration in soil solution at in a pine stand treated with 4.8 ton/ha of stabilized wood ash. Two years after the application the concentration of SO

4

was still higher at 0, 10 and 100 cm compared to the control plot.

Higher F concentrations were found in the O, 0-5 and 10-15 in the wood ashtreated arenosol compared to control during 2005, probably due to a slight decrease of this substance in the control area. A tendency towards an increase in the ashed arenosol was seen for 2006 however, but due to lack of samples and high standard deviations this increase was not statistically confirmed. F was the only anion that decreased in stream water after WAA (paper V).

There was an increase in AD in stream water as an effect of WAA (paper V).

The AD consists of bicarbonate, hydroxyl groups and carboxyl groups, both free and associated to cations. These substances are all potential proton acceptors, depending on the pK a

of the substance and the complex binding characteristic, and may thus contribute to the buffer capacity of the stream water. There was no change in pH after WAA, while the carboxylic concentration only tended towards increase. This would imply that the conditions for cation association to DOC had not changed substantially, although Ca and Mg had increased slightly. A significant amount of the Ca, Mg, Al and Fe in the stream water had previously been shown associated to organic matter (paper IV). That leaves mainly bicarbonate to account for the possible increase in buffer capacity, and the effect on the buffer capacity is likely smaller than the observed change in AD concentration.

6.1.7 Si

The concentrations of Si varied in soil solution, with occasional peak values, most likely due to dry conditions, making the data hard to interpret. The lowest concentrations were found in the O-horizon while higher concentrations of Si were found in the mineral horizons. Si concentration was significantly higher in the ashed podzol plot compared to the control plot in 2006.

45

Si increased considerably in the stream water in the treated area (Figure 15), which is hard to reconcile with the results of Steenari et al. (1999), who identified Si in stabilised wood ash in the form of highly insoluble silicates and quartz. They also found very low concentrations of Si in leachates. To the author’s knowledge, a major increase of Si in runoff after WAA has not previously been reported.

Figure 15. Si in Fanbergsbäcken and Gråbergsbäcken 2003-2006 (a) and cusum chart of

differences (b). The vertical line denotes the WAA to the Fanbergs catchment.

6.2 In-stream variations in Fanbergsbäcken

There were no visible effects of WAA through comparison of the in-stream variation, but of note is that K, which had a decreasing trend downstream during

2003-2005, showed an increase downstream in 2006. Other noteworthy changes include the decreases in Na and Ca between all sampling points in 2004 and 2005, which later became increases between the second and third sampling point in 2006 for Na, and a tendency towards an increase for Ca. The in-stream effects seen for K,

Ca and Na in 2006 might be secondary effects caused by ash constituents reaching the stream through different flow paths, rather than the effects discussed hitherto.

The ash effect on water subjected to slower hydrological pathways with longer residence times in the soil have begun to contribute to the stream chemistry at the end of this investigation, and the impact of the wood ash may therefore have entered a new phase after this initial time span. The effects are mild and may easily be missed in an investigation with lower resolution.

7. CONCLUSIONS AND FURTHER PERSPECTIVES

Most of the studied elements in soil solution, especially DOC, Ca, Al, SO

4

, and

Si increased in concentrations during the growing season due to increased

46

weathering and increased degradation of biota and litter. These seasonal patterns, however, were not recognized in the concentrations in stream water. The stream water more likely reflected the soil horizons from which the water was received during different flows and was thus more dependent on the soil solution concentrations at different soil depth.

In the recharge area, Al and Fe were precipitated in the ongoing podzolisation process as indicated by the low concentrations recovered in the deep mineral soil, and Al and Fe were to a greater extent associated to LMM DOC than in the discharge area and stream water, where they were mainly associated to HMM

DOC. Hence, there was a shift in the DOC composition throughout the catchment.

In the discharge area the soil forming process was strongly affected by the HMM organic matter and the highest concentrations of Al and Fe were found in the deeper mineral soil.

It has been shown that the quantity and quality of the DOC differed greatly between the streams, with higher DOC, total caboxylic concentration and a higher concentration and percentage of HMM-DOC in Fanbergsbäcken, resulting in a higher association of cations to HMM-DOC in this stream.

High fractions of Ca and Mg were associated to HMM DOC in stream water of two first order streams draining bordering catchments. This, however, has not been reported in other studies and needs further investigations. An investigation of the associations of metals to HMM DOC using SEC-ICP-MS is planned for 2011.

Even though there were large differences between the concentrations of DOC and associated cations in the streams, the seasonal patterns showed significant correlations indicating that precipitation induced changes in flow that triggered similar responses in the stream water chemistry of both catchments. The flow pattern driven by precipitation therefore seems to be the driver of the stream water chemistry. This made it possible to use Gråbergsbäcken as an untreated reference when investigating the effect of a catchment scale wood ash application on the water chemistry of Fanbergsbäcken.

A selective method identifying and quantifying 17 different LMMOAs down to nM-concentrations has been developed and thoroughly tested. The method is fast and efficient and has so far been tested on a wide range of media, including stream water, soil solution sampled by lysimeters and by centrifugation and growth medium for ectomycorrhizal fungi with good performance. To further improve the

47

method, the addition of a new mass track for isocitrate is planned, to make it easier to separate the structural isomers citrate and isocitrate.

There was large divergence in the constitution of LMMOAs in soil solution and stream water, with some of the acids recovered in high concentrations in soil entirely absent in the stream water samples, while others were recovered in the same order of magnitude in both mediums. LMMOAs do not seem to make up a constant fraction of DOC in either soil solution or stream water, and further studies are needed to investigate processes influencing the behavior of LMMOAs in the transfer between and interfaces of these mediums.

The Swedish Forest Agency recommends WAA to counteract acidification in soil and runoff. The effect of wood ash on soil solution was small. Temporary increases of K, Si and Fe concentrations in soil solution compared to the control were detected in the recharge area, as well as for Ca and SO

4

in the discharge area when comparing ashed and control areas.

Investigation of the initial effects of WAA to a first order stream revealed no effect on pH, but a significant increase in base cations balanced by leaching of Cl and an increase of the anion deficit, interpreted partly as an increase in buffer capacity. The DOC concentration increased, probably as a response to raised microbial activity in the soil, even though this was not seen in the respiration measurements, leading to enhanced leaching. As a tool to counteract acidification of surface waters, WAA seems to have limited initial effects, raising the buffer capacity slightly but not affecting the pH. Small but statistically significant effects could be reported, by means of high sampling frequency and enhanced by cumulative sum chart, which, in combination with Students t-test and RIA, proved to be useful in investigating time trends in stream water.

In these investigations, chemical analysis of stream water seems like a more representative method of evaluating the effect of WAA to a catchment than lysimeter data. Stream water represented the whole catchment to which 150 tonnes of wood ash had been applied, while lysimeters represented a very small soil compartment immediately surrounding the lysimeter. Unevenly distributed wood ash would therefore have far greater impact on the soil solution than on the stream water chemistry in this case. A long series of continuous data is also much easier to achieve when sampling stream water than soil solution, since the former is less sensitive to drought.

48

So, what is the most important that I have learned? I think it’s best summarized by Susan Sto Helit while reading a bedtime story

“…and then Jack chopped down what was the world's last beanstalk, adding murder and ecological terrorism to the theft, enticement and trespass charges already mentioned…” (from “Hogfather” by Terry Pratchett).

That is, some tales are told so many times that they almost turn into facts. But that does not mean they’re the truth.

49

8. ACKNOWLEDGEMENTS

Thanks to...

Ulla Lundström for giving me the opportunity to do this work, for her never ending enthusiasm and for introducing me into new projects.

Dan Bylund for sharing his extended knowledge in analytical chemistry and statistics, for his patience with students that needs more than one explanation and for his excellent memory.

Jenny Vestin, a fantastic co-worker, a great friend and a machine in the forest and in the lab. We have made a lot of trips back and forth to Bispgården, shared a lot of laughs and fika breaks, and I still have your orange.

Mattias Fredriksson for sharing the misery of writing a thesis, thing would have been a lot harder and more boring without your company.

Andreas Aronsson for sharing his extended knowledge of limnology, and nature at large (and some football, fly fishing, and roofing).

Kei Nambu for lots of help in the lab, and to Pelle Mellander for great company in the forest and sharing his knowledge about hydrology.

Madelen Olofsson for bringing some more fun into our group, for great company, friendship and crazy ideas.

Former and present members of the analytical- and soil chemistry group Tara Ali,

Sofia Essén, Sara Holmström, Rickard Thurdin, Christine Geibe and Tomas

Östlund for valuable cooperation and interesting discussions.

Katarina Brorson, Agnes Hedenström and Martin Eriksson for great help with field work and in the laboratory.

Torborg Jonsson, Håkan Norberg, Anna Haeggström and Christina Olsson, whose service-mindedness has been invaluable in different mattes.

Matt Richardson, Ann Almsåker and Nils Ekelund for reading my thesis.

50

All great friends at work; Christina, Ida, Hanna, Anna, Lotta, Per, Johan, Cicci,

Birgitta and many more, who made fika breaks a lot of fun and always showed support.

Neglected friends; Mari Jönsson, Åsa Kestrup, Åsa Knies, Susanne Büchner and

Lollo and Ceasar Guevara, whose wedding I missed while writing the final version of this thesis. I hope that I can spend some more time with all of you soon.

Terry Pratchett for making me laugh, when I really needed to.

Love and thanks to Paul and Mary Ann for great support.

Love and thanks to my family, Mor och Jan-Olof, Far och Sylvia, Maria, Tomas,

Jessica och Sofie samt Jakob, Susan, Carmen, Oscar och Ross. Ni är bäst!

And to the pack, for being there Let’s start spending a lot more time together.

51

9. REFERENCES

Abadia J, Lopez-Millan AF, Rombola A, Abadia A (2002) Organic acids and Fe deficiency: a review. Plant Soil 241(1): 75-86

Ahonen-Jonnarth U, Van Hees PAW, Lundström US, Finlay RD (2000) Organic acids produced by mycorrhizal Pinus sylvestris exposed to elevated aluminium and heavy metal concentrations. New Phytol. 146(3): 557-567

Anonymous (2008) Rekommendationer vid uttag av avverkningsrester och askåterföring. In. vol 2/2008. Swedish Forest Agency.

Aronsson KA, Ekelund NGA (2004) Biological effects of wood ash application to forest and aquatic ecosystems. J. Environ. Qual. 33(5): 1595-1605

Aronsson KA, Ekelund NGA (2008) Limnological effects on a first order stream after wood ash application to a boreal forest catchment in Bispgarden,

Sweden. For. Ecol. Manage. 255(1): 245-253

Arvidsson H, Lundkvist H (2003) Effects of crushed wood ash on soil chemistry in young Norway spruce stands. For. Ecol. Manage. 176(1-3): 121-132

Augusto L, Bakker MR, Meredieu C (2008) Wood ash applications to temperate forest ecosystems - potential benefits and drawbacks. Plant Soil 306(1-2):

181-198

Benedetti MF, Mounier S, Filizola N, Benaim J, Seyler P (2003) Carbon and metal concentrations, size distributions and fluxes in major rivers of the Amazon basin. Hydrol. Process. 17(7): 1363-1377

Bergelin A, van Hees PAW, Wahlberg O, Lundström US (2000) The acid-base properties of high and low molecular weight organic acids in soil solutions of podzolic soils. Geoderma 94(2-4): 223-235

Bertilsson S, Tranvik LJ (2000) Photochemical transformation of dissolved organic matter in lakes. Limnol. Oceanogr. 45(4): 753-762

Bishop K, Buffam I, Erlandsson M, Fölster J, Laudon H, Seibert J, Temnerud J

(2008) Aqua Incognita: the unknown headwaters. Hydrol. Process. 22(8):

1239-1242

Bishop K, Pettersson C (1996) Organic carbon in the boreal spring flood from adjacent subcatchments. Environment International 22(5): 535-540

Bishop K, Seibert J, Köhler S, Laudon H (2004) Resolving the Double Paradox of rapidly mobilized old water with highly variable responses in runoff chemistry. Hydrol. Process. 18(1): 185-189

Bishop KH, Lundström US, Giesler R (1993) Transfer of organic C from forest soils to surface waters: example from northern Sweden. Applied Geochemistry

(2): 11-15

Björkvald L, Buffam I, Laudon H, Mörth CM (2008) Hydrogeochemistry of Fe and

Mn in small boreal streams: The role of seasonality, landscape type and scale. Geochimica Et Cosmochimica Acta 72(12): 2789-2804

52

Burt TP (2005) A third paradox in catchment hydrology and biogeochemistry: decoupling in the riparian zone. Hydrol. Process. 19(10): 2087-2089

Carpenter SR, Frost TM, Heisey D, Kratz TK (1989) Randomized intervention analysis and the interpretation of whole-ecosystem experiments. Ecology

70(4): 1142-1152

Clarke N, Danielsson LG, Sparen A (1992) The Determination of quickly reacting aluminum in natural-waters by kinetic discrimination in a flow system.

International Journal of Environmental Analytical Chemistry 48(2): 77-100

Cole JJ, Prairie YT, Caraco NF, McDowell WH, Tranvik LJ, Striegl RG, Duarte CM,

Kortelainen P, Downing JA, Middelburg JJ, Melack J (2007) Plumbing the global carbon cycle: Integrating inland waters into the terrestrial carbon budget. Ecosystems 10(1): 171-184

Dahlqvist R, Benedetti MF, Andersson K, Turner D, Larsson T, Stolpe B, Ingri J

(2004) Association of calcium with colloidal particles and speciation of calcium in the Kalix and Amazon rivers. Geochimica et Cosmochimica Acta

68(20): 4059-4075

Eriksson HM (1998a) Short-term effects of granulated wood ash on forest soil chemistry in SW and NE Sweden. Scand. J. Forest Res.: 43-55

Eriksson J (1998b) Dissolution of hardened wood ashes in forest soils: Studies in a column experiment. Scand. J. Forest Res.: 23-32

FAO (1990) Guidelines for soil profile description. Soil Resources, Management and Conservation Service, Land and Water Developement division, FAO,

Rome.

Fransman B, Nihlgård B (1995) Water chemistry in forested catchments after topsoil treatment with liming agents in South Sweden. Water Air Soil Pollut.

85(2): 895-900

Fransson AM, Valeur I, Wallander H (2004) The wood-decaying fungus

Hygrophoropsis aurantiaca increases P availability in acid forest humus soil, while N addition hampers this effect. Soil Biol. Biochem. 36(11): 1699-1705

Fölster J (2000) The near-stream zone is a source of nitrogen in a Swedish forested catchment. J. Environ. Qual. 29(3): 883-893

Geibe CE, Danielsson R, van Hees PAW, Lundstrom US (2006) Comparison of soil solution chemistry sampled by centrifugation, two types of suction lysimeters and zero-tension lysimeters. Appl. Geochem. 21(12): 2096-2111

Geibe CE, Holmström SJM, Van Hees PAW, Lundström US (2003) Impact of Lime and Ash Applications on Soil Solution Chemistry of an Acidified Podzolic

Soil. Water, Air, & Soil Pollution: Focus 3(4): 77-96

Giesler R, Lundström U (1993) Soil sollution chemistry - effects of bulking soil samples. Soil Sci. Soc. Am. J. 57(5): 1283-1288

53

Haaland S, Hongve D, Laudon H, Riise G, Vogt RD (2010) Quantifying the Drivers of the Increasing Colored Organic Matter in Boreal Surface Waters. Environ.

Sci. Technol. 44(8): 2975-2980

Hagberg J (2003) Analysis of low-molecular-mass organic acids using capillary zone electrophoresis-electrospray ionization mass spectrometry. J.

Chromatogr. A 988(1): 127-133

Hagberg J, Duker A, Karlsson S (2002) Determination of dissociation constants of low molecular weight organic acids by capillary zone electrophoresis and indirect UV detection. Chromatographia 56(9-10): 641-644

Hindar A, Wright RF, Nilsen P, Laessen T, Hogberget R (2003) Effects on stream water chemistry and forest vitality after whole-catchment application of dolomite to a forest ecosystem in southern Norway. For. Ecol. Manage.

180(1-3): 509-525

Hinton MJ, Schiff SL, English MC (1998) Sources and flowpaths of dissolved organic carbon during storms in two forested watersheds of the Precambrian

Shield. Biogeochemistry 41(2): 175-197

Hlavacova E, Rulik M, Cap L (2005) Anaerobic microbial metabolism in hyporheic sediment of a gravel bar in a small lowland stream. River Research and

Applications 21(9): 1003-1011

Holmström SJM, Riise G, Strand LT, Geibe C, Van Hees PAW, Wu Q, Lundström

US (2003) Effects of Lime and Ash Treatments on DOC Fractions and Low

Molecular Weight Organic Acids in Soil Solutions of Acidified Podzolic

Soils. Water, Air, & Soil Pollution: Focus 3(4): 97-120

Holmström SJM, Norström SH, Rosling A. Functional differences in the genus

Piloderma, linking production of LMMOAs and siderophores to vertical distribution in soil profiles. Submitted to Fungal Biology

Hope D, Billett MF, Cresser MS (1994) A Review of the export of carbon in river water - fluxes and processes. Environmental Pollution 84(3): 301-324

Hornberger GM, Bencala KE, McKnight DM (1994) Hydrological Controls on dissolved organic carbon during snowmelt in the Snake River near

Montezuma, Colorado. Biogeochemistry 25(3): 147-165

Hruska J, Johnson CE, Kram P (1996) Role of organic solutes in the chemistry of acid-impacted bog waters of the western Czech Republic. Water Resources

Research 32(9): 2841-2851

Hruska J, Köhler S, Laudon H, Bishop K (2003) Is a universal model of organic acidity possible: Comparison of the acid/base properties of dissolved organic carbon in the boreal and temperate zones. Environmental Science &

Technology 37(9): 1726-1730

Hruska J, Laudon H, Johnson CE, Köhler S, Bishop K (2001) Acid/base character of organic acids in a boreal stream during snowmelt. Water Resources Research

37(4): 1043-1056

54

Hue NV, Craddock GR, Adams F (1986) Effect of organic acids on aluminum toxicity in subsoils. Soil Sci. Soc. Am. J. 50(1): 28-34

Högbom L, Nohrstedt HÖ, Nordlund S (2001) Effects of wood-ash addition on soilsolution chemistry and soil N dynamics at a Picea abies (L.) Karst. site in southwest Sweden. In: Reports from SkogForsk. The Forestry Research iInstitute of Sweden, Uppsala, Sweden.

Ilvesniemi H, Giesler R, van Hees P, Magnussson T, Melkerud PA (2000) General description of the sampling techniques and the sites investigated in the

Fennoscandinavian podzolization project. Geoderma 94(2-4): 109-123

Jacobson S (2003) Addition of stabilized wood ashes to Swedish coniferous stands on mineral soils - Effects on stem growth and needle nutrient concentrations.

Silva. Fenn. 37(4): 437-450

Jacobson S, Hogbom L, Ring E, Nohrstedt HO (2004) Effects of wood ash dose and formulation on soil chemistry at two coniferous forest sites. Water Air Soil

Pollut. 158(1): 113-125

Johnson CE, Driscoll CT, Siccama TG, Likens GE (2000) Element fluxes and landscape position in a northern hardwood forest watershed ecosystem.

Ecosystems 3(2): 159-184

Jokinen HK, Kiikkila O, Fritze H (2006) Exploring the mechanisms behind elevated microbial activity after wood ash application. Soil Biol. Biochem. 38(8): 2285-

2291

Jones DL (1998) Organic acids in the rhizosphere – a critical review. Plant Soil

205(1): 25-44

Jones DL, Dennis PG, Owen AG, van Hees PAW (2003) Organic acid behavior in soils - misconceptions and knowledge gaps. Plant Soil 248(1-2): 31-41

Jonsson A, Ström L, Åberg J (2007) Composition and variations in the occurrence of dissolved free simple organic compounds of an unproductive lake ecosystem in northern Sweden. Biogeochemistry 82(2): 153-163

Kahl JS, Fernandez IJ, Rustad LE, Peckenham J (1996) Threshold application rates of wood ash to an acidic forest soil. J. Environ. Qual. 25(2): 220-227

Kalbitz K, Solinger S, Park JH, Michalzik B, Matzner E (2000) Controls on the dynamics of dissolved organic matter in soils: A review. Soil Science 165(4):

277-304

Kirchner JW (2003) A double paradox in catchment hydrology and geochemistry.

Hydrol. Process. 17(4): 871-874

Kortelainen P, Mattsson T, Finer L, Ahtiainen M, Saukkonen S, Sallantaus T (2006)

Controls on the export of C, N, P and Fe from undisturbed boreal catchments, Finland. Aquat. Sci. 68(4): 453-468

Laudon H, Bishop K (2002) Episodic stream water pH decline during autumn storms following a summer drought in northern Sweden. Hydrol. Process.

16(9): 1725-1733

55

Laudon H, Köhler S, Buffam I (2004) Seasonal TOC export from seven boreal catchments in northern Sweden. Aquat. Sci. 66(2): 223-230

Levula T, Saarsalmi A, Rantavaara A (2000) Effects of ash fertilization and prescribed burning on macronutrient, heavy metal, sulphur and Cs-137 concentrations in lingonberries (Vaccinium vitis-idaea). For. Ecol. Manage.

126(3): 269-279

Ludwig B, Rumpf S, Mindrup M, Meiwes KJ, Khanna PK (2002) Effects of lime and wood ash on soil-solution chemistry, soil chemistry and nutritional status of a pine stand in Northern Germany. Scand. J. Forest Res. 17(3): 225-237

Luke SH, Luckai NJ, Burke JM, Prepas EE (2007) Riparian areas in the Canadian boreal forest and linkages with water quality in streams. Environ. Rev. 15:

79-97

Lundström U, Öhman LO (1990) Dissolution of feldspars in the presence of natural, organic solutes. Journal of Soil Science 41(3): 359-369

Lundström US (1993) The Role of Organic-Acids in the Soil Solution Chemistry of a

Podzolized Soil. Journal of Soil Science 44(1): 121-133

Lundström US, Bain DC, Taylor AFS, van Hees PAW (2003a) Effects of

Acidification and its Mitigation with Lime and Wood Ash on Forest Soil

Processes: A Review. Water, Air, & Soil Pollution: Focus 3(4): 5-28

Lundström US, Bain DC, Taylor AFS, Van Hees PAW, Geibe CE, Holmström SJM,

Melkerud PA, Finlay R, Jones DL, Nyberg L, Gustafsson JP, Riise G, Strand

LT (2003b) Effects of Acidification and its Mitigation with Lime and Wood

Ash on Forest Soil Processes in Southern Sweden. A Joint Multidisciplinary

Study. Water, Air, &amp; Soil Pollution: Focus 3(4): 167-188

Lundström US, van Breemen N, Bain D (2000a) The podzolization process. A review. Geoderma 94(2-4): 91-107

Lundström US, van Breemen N, Bain DC, van Hees PAW, Giesler R, Gustafsson JP,

Ilvesniemi H, Karltun E, Melkerud PA, Olsson M, Riise G, Wahlberg O,

Bergelin A, Bishop K, Finlay R, Jongmans AG, Magnusson T, Mannerkoski

H, Nordgren A, Nyberg L, Starr M, Strand LT (2000b) Advances in understanding the podzolization process resulting from a multidisciplinary study of three coniferous forest soils in the Nordic Countries. Geoderma

94(2-4): 335-353

Löfgren S, Cory N, Zetterberg T, Larsson PE, Kronnäs V (2009) The long-term effects of catchment liming and reduced sulphur deposition on forest soils and runoff chemistry in southwest Sweden. For. Ecol. Manage. 258(5): 567-

578

Malcolm RL (1990) The uniqueness of humic substances in each of soil, stream and marine environments. Analytica Chimica Acta 232: 19-30

56

Maljanen M, Jokinen H, Saari A, Strömmer R, Martikainen PJ (2006a) Methane and nitrous oxide fluxes, and carbon dioxide production in boreal forest soil fertilized with wood ash and nitrogen. Soil Use Manage. 22(2): 151-157

Maljanen M, Nykänen H, Moilanen M, Martikainen PJ (2006b) Greenhouse gas fluxes of coniferous forest floors as affected by wood ash addition. For. Ecol.

Manage. 237(1-3): 143-149

Marschner H (1995) Mineral nutrition of higher plants. Academic Press, London

Meyer JL, Wallace JB, Eggert SL (1998) Leaf litter as a source of dissolved organic carbon in streams. Ecosystems 1(3): 240-249

Miller JN, Miller JC (2000) Statistics and chemometrics for analytical chemistry.

Pearson Education limited, Harlow,Essex, England

Moilanen M, Fritze H, Nieminen M, Sirpa P, Issakainen J, Piispanen J (2006) Does wood ash application increase heavy metal accumulation in forest berries and mushrooms? For. Ecol. Manage. 226(1-3): 153-160

Mulholland PJ, Hill WR (1997) Seasonal patterns in streamwater nutrient and dissolved organic carbon concentrations: Separating catchment flow path and in-stream effects. Water Resources Research 33(6): 1297-1306

Nieminen M, Piirainen S, Moilanen M (2005) Release of mineral nutrients and heavy metals from wood and peat ash fertilizers: Field studies in Finnish forest soils. Scandinavian Journal of Forest Research 20(2): 146-153

Nilsson SI, Andersson S, Valeur I, Persson T, Bergholm J, Wiren A (2001) Influence of dolomite lime on leaching and storage of C, N and S in a Spodosol under

Norway spruce (Picea abies (L.) Karst.). For. Ecol. Manage. 146(1-3): 55-73

Oliver BG, Thurman EM, Malcolm RL (1983) The contribution of humic substances to the acidity of colored natural waters. Geochimica et Cosmochimica Acta

47(11): 2031-2035

Olsson BA, Bengtsson J, Lundkvist H (1996) Effects of different forest harvest intensities on the pools of exchangeable cations in coniferous forest soils.

For. Ecol. Manage. 84(1-3): 135-147

Olsson BA, Lundkvist H, Staaf H (2000) Nutrient status in needles of Norway spruce and Scots pine following harvesting of logging residues. Plant Soil

223(1-2): 161-173

Osborne LL, Kovacic DA (1993) Riparian vegetated buffer stripsin water-quality restoration and sstream management. Freshw. Biol. 29(2): 243-258

Perkiömäki J, Fritze H (2002) Short and long-term effects of wood ash on the boreal forest humus microbial community. Soil Biol. Biochem. 34(9): 1343-1353

Pettersson C, Bishop K (1996) Seasonal variations of total organic carbon, iron, and aluminium on the Svartberget catchment in northern Sweden. Environment

International 22(5): 541-549

Pitman RM (2006) Wood ash use in forestry - a review of the environmental impacts. Forestry 79(5): 563-588

57

Pohlman AA, McColl JG (1988) Soluble organics from forest litter and their role in metal dissolution. Soil Sci. Soc. Am. J. 52(1): 265-271

Prescott CE, Maynard DG, Laiho R (2000) Humus in northern forests: friend or foe?

For. Ecol. Manage. 133(1-2): 23-36

Proe MF, Cameron AD, Dutch J, Christodoulou XC (1996) The effect of whole-tree harvesting on the growth of second rotation Sitka spruce. Forestry 69(4): 389-

401

Raulund-Rasmussen K, Borggaard OK, Hansen HCB, Olsson M (1998) Effect of natural organic soil solutes on weathering rates of soil minerals. Eur. J. Soil

Sci. 49(3): 397-406

Riise G, Salbu B, Vogt RD, Ranneklev SB, Mykkelbost TC (1994) Moility of humic substances, major and minor elements in lake Skjervatjern and its catchment area. Environ. Int. 20(3): 287-298

Riise G, Van Hees P, Lundström U, Tau Strand L (2000) Mobility of different size fractions of organic carbon, Al, Fe, Mn and Si in podzols. Geoderma 94(2-4):

237-247

Ring E, Jacobson S, Nohrstedt HÖ (2006) Soil-solution chemistry in a coniferous stand after adding wood ash and nitrogen. Can. J. For. Res.-Rev. Can. Rech.

For. 36(1): 153-163

Rosenberg O, Persson T, Högbom L, Jacobson S (2010) Effects of wood-ash application on potential carbon and nitrogen mineralisation at two forest sites with different tree species, climate and N status. For. Ecol. Manage.

260(4): 511-518

Rumpf S, Ludwig B, Mindrup M (2001) Effect of wood ash on soil chemistry of a pine stand in Northern Germany. J. Plant Nutr. Soil Sci.-Z. Pflanzenernahr.

Bodenkd. 164(5): 569-575

Saarsalmi A, Malkonen E, Piirainen S (2001a) Effects of wood ash fertilization on forest soil chemical properties. Silva. Fenn. 35(3): 355-368

Saarsalmi A, Mälkönen E, Piirainen S (2001b) Effects of wood ash fertilization on forest soil chemical properties. Silva. Fenn. 35(3): 355-368

Sandnes A, Eldhuset TD, Wollebæk G (2005) Organic acids in root exudates and soil solution of Norway spruce and silver birch. Soil Biology and

Biochemistry 37(2): 259-269

Steenari BM, Karlsson LG, Lindqvist O (1999) Evaluation of the leaching characteristics of wood ash and the influence of ash agglomeration. Biomass

Bioenerg. 16(2): 119-136

Steenari BM, Lindqvist O (1997) Stabilisation of biofuel ashes for recycling to forest soil. Biomass Bioenerg. 13(1-2): 39-50

Steenari BM, Marsic N, Karlsson LG, Tomsic A, Lindqvist O (1998) Long-term leaching of stabilized wood ash. Scand. J. Forest Res.: 3-16

58

Stevenson FJ (1993) Humus chemistry: genesis, composition, reactions. John Wiley

& Sons, Inc., New York

Strobel BW (2001) Influence of vegetation on low-molecular-weight carboxylic acids in soil solution--a review. Geoderma 99(3-4): 169-198

Stutter MI, Deeks LK, Low D, Billett MF (2006) Impact of soil and groundwater heterogeneity on surface water chemistry in an upland catchment. J. Hydrol.

318(1-4): 103-120

Taylor AFS, Finlay RD (2003) Effects of Liming and Ash Application on Below

Ground Ectomycorrhizal Community Structure in Two Norway Spruce

Forests. Water, Air, &amp; Soil Pollution: Focus 3(4): 63-76

Temnerud J, Bishop K (2005) Spatial variation of streamwater chemistry in two

Swedish boreal catchments: Implications for environmental assessment.

Environmental Science & Technology 39(6): 1463-1469

Tipping E (1998) Humic ion-binding model VI: An improved description of the interactions of protons and metal ions with humic substances. Aquat.

Geochem. 4(1): 3-48

Tulonen T, Arvola L, Ollila S (2002) Limnological effects of wood ash application to the subcatchments of boreal, humic lakes. J. Environ. Qual. 31(3): 946-953

Walmsley JD, Jones DL, Reynolds B, Price MH, Healey JR (2009) Whole tree harvesting can reduce second rotation forest productivity. For. Ecol.

Manage. 257(3): 1104-1111 van Breemen N, Finlay R, Lundström U, Jongmans AG, Giesler R, Olsson M (2000)

Mycorrhizal weathering: A true case of mineral plant nutrition ?

Biogeochemistry 49(1): 53-67 van Hees PAW, Dahlen J, Lundström US, Boren H, Allard B (1999) Determination of low molecular weight organic acids in soil solution by HPLC. Talanta

48(1): 173-179 van Hees PAW, Jones DL, Finlay R, Godbold DL, Lundström US (2005) The carbon we do not see--the impact of low molecular weight compounds on carbon dynamics and respiration in forest soils: a review. Soil Biology and

Biochemistry 37(1): 1-13 van Hees PAW, Jones DL, Godbold DL (2002) Biodegradation of low molecular weight organic acids in coniferous forest podzolic soils. Soil Biology and

Biochemistry 34(9): 1261-1272 van Hees PAW, Lundström US (2000) Equilibrium models of aluminium and iron complexation with different organic acids in soil solution. Geoderma 94(2-4):

201-221 van Hees PAW, Lundström US, Giesler R (2000) Low molecular weight organic acids and their Al-complexes in soil solution - composition, distribution and seasonal variation in three podzolized soils. Geoderma 94(2-4): 173-200

59

van Hees PAW, Nyberg L, Holmström SJM, Lundström US (2003) Pools and fluxes of cations, anions and DOC in two forest soils treated with lime and ash.

Water, Air, &amp; Soil Pollution: Focus 3(4): 145-165 van Hees PAW, Rosling A, Essen S, Godbold DL, Jones DL, Finlay RD (2006)

Oxalate and ferricrocin exudation by the extramatrical mycelium of an ectomycorrhizal fungus in symbiosis with Pinus sylvestris. New Phytol.

169(2): 367-377 van Hees PAW, van Hees AMT, Lundström US (2001) Determination of aluminium complexes of low molecular organic acids in soil solution from forest soils using ultrafiltration. Soil Biol. Biochem. 33(7-8): 867-874 van Scholl L, Hoffland E, van Breemen N (2006a) Organic anion exudation by ectomycorrhizal fungi and Pinus sylvestris in response to nutrient deficiencies. New Phytol. 170(1): 153-163 van Scholl L, Smits MM, Hoffland E (2006b) Ectomycorrhizal weathering of the soil minerals muscovite and hornblende. New Phytol. 171(4): 805-814

Vanguelova E, Pitman R, Luiro J, Helmisaari H-S (2010) Long term effects of whole tree harvesting on soil carbon and nutrient sustainability in the UK.

Biogeochemistry: 1-17

Zimmermann S, Frey B (2002) Soil respiration and microbial properties in an acid forest soil: effects of wood ash. Soil Biol. Biochem. 34(11): 1727-1737

Ågren A, Buffam I, Jansson M, Laudon H (2007) Importance of seasonality and small streams for the landscape regulation of dissolved organic carbon export. Journal of Geophysical Research-Biogeosciences 112(G3):

60

Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Related manuals