Betula Pubescens) in Rehabilitated Areas in Southern Iceland

Home , Revegetation

CarbBirch (Kolbjörk): Carbon sequestration and soil development under mountain birch (Betula pubescens) in rehabilitated areas in southern Iceland

Master’s Thesis

Presented in Partial Fulfillment of the Requirements for the Degree Master of Science in the Graduate School of The Ohio State University

Páll Valdimar Kolka-Jónsson, B.S.

Graduate Program in Environmental and Natural Resources

The Ohio State University

2011

Thesis Committee:

Dr. Brian K. Slater, Advisor

Dr. Nicholas Basta

Dr. David M. Hix

Copyrighted by

Páll Valdimar Kolka-Jónsson

2011

Abstract

Understanding soil change when restoring severely degraded land is important to be able to determine when and if the ecosystem services that ‘healthy’ soil provides are restored. Carbon sequestration is an important benefit from soil restoration. This thesis studies the trends in soil carbon sequestration and mineral assemblies over a time period of just 65 years when using Betula pubescens for restoration of severely degraded land.

The soils did not reach the undisturbed soil carbon concentrations in the area in 65 years, indicating that the sequestering period is significantly longer. The total sequestration rate was calculated as 0.466 t C ha-1. Effects of significant inputs of aeolian material were seen in the increased variance in 5-15 cm compared to 0-5 and 15-30. This shows the need to account for rapid aeolian deposition in future carbon sequestration models and emphasises the challenges of modelling processes in such a dynamic environment.

Significant changes in the depth distribution of carbon and clays were seen over this short time period, showing the rapid soil forming processes in Andisols compared to other soil orders. Rapid aeolian deposition in the area masks the change in clay contents but the trends for increased concentrations to a greater depth with age were stronger in the more dynamic carbon contents. The soils are multi-genetic, dominated by aeolian transferred material in the surface horizons and fluvial/direct tephra deposition in the subsurface horizons. ii

Dedication

I dedicate this thesis to my sister Katrín Kolka, who showed me how to stay strong

no matter what life throws at you.

iii

Acknowledgments

Firstly, I’d like to thank my fiancée, Sandra Sif Einarsdóttir, who saved my skin by helping me set up the thesis and deal with the joys of Microsoft Word. She also has had to bear my sudden coming and goings as I rocketed between continents. Thank you for being so patient with me these last few months.

I would like to thank my advisor, Brian Slater for support during my studies at The Ohio State and giving me an understanding and love of soil pedology as well as helping me adjust to life in the USA

Ólafur Arnalds at the Agricultural university of Iceland is the reason this project ever happened and has been instrumental in hammering it all together.

At The Ohio State, I’d like to thank Sandy Jones for being there and help me trawl through the labyrinth of soil classification methods and always respond in a productive manner to all my questions and fantasies. Shane Whitacre has also been extremely helpful in the lab, allowing me a freedom I probably shouldn’t have enjoyed. Jerry Bigham has then helped me take the fantasies and make them into a something tangible. Lastly, the lovely ladies in 210 that make everything go tick.

At the Agricultural University of Iceland, Rannveig Guichernaud helped me understand soils (and people) and Brita Berglund showed me how to work the lab.

Vita

2003...... Menntaskólinn á Akureyri (The Akureyri

Junior College)

2007 ...... B.S. in Geology, University of Iceland

2008 – present...... Graduate Teaching and Research

Associate, The Ohio State University

Fields of Study

Major Field: Environmental and Natural Resources

Table of Contents

!"#$%&'$((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((())* +,-)'&$)./ (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( )))* !'0/.12,-34,/$# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( )5* 6)$& ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((5* 7&"2,*.8*9./$,/$# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( 5)* :)#$*.8*7&"2,# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((();* :)#$*.8*<)3=%,# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((;*

!* "#$%&'()$*&#+,#'+-).#.+-.$$*#/000000000000000000000000000000000000000000000000000000000000000000000000000!* !0!* 12.%2*.30000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000 !* !04* 5.&6&/7+,#'+/.&8&%9:&6&/7+&;+").6,#'000000000000000000000000000000000000000000000000000000000000000 !* !0<* =6*8,$. 0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000 4* !0>* ?#'*-&6-000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000 <* >(?(>* @)/,%&2.3)'&2*9A&%&'$,%)#$)'#*.8*!/-)#.2#((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( B* >(?(C* D.)2*8.%4&$)./*)/*5.2'&/)'*,E,'$& ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( F* !0@* ").6,#'*)+A&*6- 000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000 @* !0B* C,#'+'./%,',$*&#+,#'+6,#'+%.-$&%,$*&#+*#+").6,#' 000000000000000000000000000000000000000000 D* !0D* =,%E&#+-$&)F-+*#+;&%.-$.'+.)&-7-$.8-+*#+").6,#' 000000000000000000000000000000000000000000000 G* !0H* I:.+A&*6+=,%E&#+)7)6. 00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000!J* !0G* K.$(6,+-99000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000!J* !0!J* I:.+L&6EMN%F+O=,%EK*%):P+9%&M.)$ 000000000000000000000000000000000000000000000000000000000000000000000!!* >(>G(>* H5,%5),1(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( >>* >(>G(C* D$=-I*!%,& ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( >C* !0!!* I:*-+9%&M.)$ 000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000!<*

4* =:,#/.-+*#+-&*6+),%E&#+,--&)*,$.'+3*$:+%.-$&%,$*&#+&;+-.2.%.67+ './%,'.'+6,#'+*#+A&($:.%#+").6,#'+3*$:+'&3#7+E*%):+O!"#$%&' ($)"*+",*P00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000 !>* 40!* "#$%&'()$*&# 0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000!>* C(>(>* 9A&J$,%*.5,%5),1(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( >?* C(>(C* D.)2*'&%"./*J..2 ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( >?* C(>(B* K',2&/-)'*#.)2# ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( >F* vi

C(>(?* 6,3,$&$)./*&/-*2&/-*-,3%&-&$)./((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( >F* C(>(F* H"E,'$)5,#((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( >L* 404* Q.$:&'-0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000!D* C(C(>* D$=-I*!%,&(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( >M* C(C(C* D&4J2)/3*#'A,4,((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( >N* C(C(B* !/&2I#)# ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( C>* 40<* R.-(6$-00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000044* C(B(>* D.)2#*)/*$A,*#&4J2,*&%,&#(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( CC* C(B(C* 9&%"./(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( CB* C(B(B* +)88,%,/',*",$1,,/*#$=-I*&%,&# ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( CB* C(B(?* 9OP*%&$).# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( C?* 40>* S*-)(--*&# 000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000004>* C(?(>* 9&%"./*#,Q=,#$%&$)./(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( C?* C(?(C* +)88,%,/',*",$1,,/*#$=-I*&%,&# ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( CL* 40@* =&#)6(-*&#-0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000004D*

<* A&*6+;&%8,$*&#+,#'+9%&;*6.+'.2.6&98.#$+(#'.%+!"#$%&'($)"*+",*+*#+ -&($:.%#+").6,#' 00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000 <>* <0!* "#$%&'()$*&# 0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000<>* B(>(>* D.)2*8.%4&$)./ ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( B?* B(>(C* !22.JA&/,(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( BF* <04* 1EM.)$*2.-000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000<@* <0<* Q.$:&'-0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000<@* B(B(>* D$=-I*#)$,# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( BF* B(B(C* D&4J2)/3 ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( BM* B(B(B* 9A,4)'&2*&/&2I#)#(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( BM* B(B(?* D$&$)#$)'&2*&/&2I#)# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( BR* <0>* R.-(6$-+,#'+'*-)(--*�* S%.8)2,*-,#'%)J$)./#(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( BN* B(?(C* D.)2*%,&'$)./((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( ?R* B(?(B* T=20*-,/#)$I((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( ?R* B(?(?* 9&%"./*&/-*/)$%.3,/(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( ?N* B(?(F* H;&2&$,*,;$%&'$&"2,*'2&I# ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( FC* B(?(L* DJ,')8)'*#=%8&',*&/-*J..%2I*'%I#$&22)/,*4)/,%&2#((((((((((((((((((((((((((((((((((((((((((((((((( FF* vii

B(?(M* 9A&/3,*1)$A*&3,((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( FM* <0@* =&#)6(-*&#-000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000B4*

>* R.;.%.#).- 00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000 B>*

viii

List of Tables

7&"2,*>O*D=44&%I*.8*#$=-I*&%,&#(*D$U%)*V2.8)*WDVXY*T.2A.2$*WTX(((((((((((((((((((((((((((((((((((((((((((((((>N* 7&"2,* CO* !5,%&3,* 5&2=,#* 8.%* 39* 4ZC* 1)$A* NF[* './8)-,/',* )/$,%5&2#* 8.%* $A,* J2.$#* 1)$A.=$*8.%,#$*&$*D$U%)*V2.8)*WDVX*&/-*T.2A.2$*WTX((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((CB* 7&"2,*BO*!3,*-)#$%)"=$)./*.8*J%.8)2,# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((BL* 7&"2,*?O*S%.8)2,*-,#'%)J$)./#*&/-*A.%)\./*-,#)3/&$.%#(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((?>* 7&"2,*FO*D,2,'$)5,*-)##.2=$)./*&/-*#J,')8)'*#=%8&',(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((F>* 7&"2,*LO*D,2,'$)5,*-)##.2=$)./*&/-*#J,')8)'*#=%8&',(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((F?*

List of Figures

<)3=%,*>O*:.'&$)./*.8*#$=-I*&%,& (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((>R* <)3=%,*CO*+)#$%)"=$)./*.8*#$=-I*#)$,#(*>*)#*]=//2&=3##0U3=%Y*C*T.2A.2$Y*B*D$U%)V2.8)Y*?* ^%&=/$,)3=%*&/-*F*T_%8,22*W^&22-U%##./*,$*&2(Y*CGGNX ((((((((((((((((((((((((((((((((((((((((((((((((((((((>R* <)3=%,*BO*`,#,&%'A*J2.$#*&$*T.2A.2$((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((CG* <)3=%,*?O*`,#,&%'A*J2.$#*&$*D$U%)*V2.8)((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((CG* <)3=%,*FO*`,#,&%'A*J2.$#*&$*^%&=/$,)3=%(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((C>* <)3=%,*LO*D.)2*'.%,*8%.4*=/5,3,$&$,-*J2.$*&$*D$U%)*V2.8)(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((CC* <)3=%,*MO*D.)2*'.%,*8%.4*=/8.%,#$,-*J2.$*&$*D$U%)*V2.8)((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((CC* <)3=%,*RO*!*#.)2*J%.8)2,*8%.4*=/5,3,$&$,-*J2.$*&$*D$U%)V2.8)(*aA)$,*3%&)/#*&%,*%.=/-,-* $,JA%&Y*"2&'0*"&#&2$)'*32&##*&/-*%.'0(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((CM* <)3=%,*NO*!*#.)2*J%.8)2,*8%.4*=/5,3,$&$,-*J2.$*&$*T.2A.2$(*+&%0,%*2&I,%#*&%,*)/-)5)-=&2* $,JA%&*2&I,%#*1)$A*&,.2)&/*4&$$,%*)/*",$1,,/((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((CM* <)3=%,* >GO* [9* )/* GZF* '4* %,3%,##)./* 8)$O* [9* b* >(BBFMN??* c* G(G?>FBBNd&3,Y* `C* b* G(??NFY*Sb*eG(GG>((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((CN* <)3=%,* >>O*[9*)/*FZ>F* '4Y* %,3%,##)./* 8)$O[9* b* G(LNLNCFL* c* G(G>NCNNRd&3,* * `Cb* G(CLFFNY*S*b*eG(GGG> (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((CN* <)3=%,* >CO* [9* )/* >FZBG* '4Y* %,3%,##)./* 8)$O* [9* b* G(LBC??RM* c* G(GGRBGGMd&3,Y* `Cb* G(>?RLY*Sb*G(GG??((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((BG* <)3=%,*>BO**39*4ZC*)/*GZF*'4Y*%,3%,##)./*8)$O**39*4ZC*b*?>R(?BMM?*c*>>(LMGCFFd&3,Y*`Cb* G(?C?LY*Sb*eG(GG>((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((BG* <)3=%,*>?O*39*4ZC*%,3%,##)./*8)$O*39*4ZCb*L>?(BCN>L*c*C>(B?BMMBd&3,Y*`Cb*G(CLM>Y*Sb* eG(GG>(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((B>* <)3=%,*>FO*39*4ZC*%,3%,##)./*8)$*39*4ZCb*FCL(RNR>N*c*>B(MFBGCRd&3,Y*`Cb*G(FGCNY*Sb* eG(GGG> ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((B>* <)3=%,*>LO*9OP*%&$).#*)/*GZF*'4((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((BC*

<)3=%,*>MO*9OP*%&$).#*)/*FZ>F*'4(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((BC* <)3=%,*>RO*9OP*%&$).*)/*>FZBG*'4 ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((BB* <)3=%,*>NO*S%.8)2,*SVG>*1)$A*A.%)\./*-,#)3/&$.%# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((??* <)3=%,*CGO*S%.8)2,*SVGC*1)$A*A.%)\./*-,#)3/&$.%# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((??* <)3=%,*C>O*S%.8)2,*SVGB*1)$A*A.%)\./*-,#)3/&$.%# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((?F* <)3=%,*CCO*S%.8)2,*SVG?*1)$A*A.%)\./*-,#)3/&$.%# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((?F* <)3=%,*CBO*S%.8)2,*SVGF*1)$A*A.%)\./*-,#)3/&$.%# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((?L* <)3=%,*C?O*S%.8)2,*SVGL*1)$A*A.%)\./*-,#)3/&$.%# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((?L* <)3=%,*CFO*S%.8)2,*SVGM*1)$A*A.%)\./*-,#)3/&$.%# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((?M* <)3=%,*CLO*S%.8)2,*SVGR*1)$A*A.%)\./*-,#)3/&$.%# (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((?M* <)3=%,*CMO*+,J$A*-)#$%)"=$)./*.8*'&%"./*W[X(*P.$,*$A,*)%%,3=2&%*J&$$,%/*)/*.2-3%.1$A* J%.8)2,#*WSVGM*&/-*SVGRX ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((FG* <)3=%,*CRO*9&%"./*-,J$A*-)#$%)"=$)./( (((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((FG* <)3=%,*CNO*9.4")/,-*-,J$A*-)#$%)"=$)./*.8*J..%2I*'%I#$&22)/,*4)/,%&2#((((((((((((((((((((((((((((FB* <)3=%,* BGO* 9.4")/,-* -,J$A* -)#$%)"=$)./* .8* J..%2I* '%I#$&22)/,* 4)/,%&2#* &$* D$U%)* V2.8)* ,;'2=-)/3*$A,*"=%),-*J&2&,.#.2((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((FB* <)3=%,* B>O* D)4J2,* %,3%,##)./* &/&2I#)#* ./* $A,* %,2&$)./#A)J* ",$1,,/* $A,* &22.JA&/,* './$,/$#*&/-*#J,')8)'*#=%8&',*&%,&(**D=%8&',*&%,&*b*Z?(F?F*c*WG(NFL*d*43*3*#.)2Z>* &22.JA&/,XY*%C*b*G(LC?Y*JZ5&2=,*eG(GGG>(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((FL* <)3=%,* BCO* D)4J2,* %,3%,##)./* &/&2I#)#* ./* $A,* %,2&$)./#A)J* ",$1,,/* $A,* $.$&2* '2&I* './$,/$*&/-*#J,')8)'*#=%8&',*&%,&(**D=%8&',*&%,&*b*ZF(NLC*c*WG(LLR*d*43*3*#.)2Z>* '2&IXY*%C*b*G(LFGY*JZ5&2=,*eG(GGG>(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((FL* <)3=%,* BBO* +,J$A* -)#$%)"=$)./* .8* .;&2&$,* ,;$%&'$&"2,* J..%2I* '%I#$&22)/,* 4)/,%&2#* &/-* #=%8&',*&%,&*)/*J%.8)2,#*SVG>*&/-*SVGC ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((FR* <)3=%,* B?O* +,J$A* -)#$%)"=$)./* .8* .;&2&$,* ,;$%&'$&"2,* J..%2I* '%I#$&22)/,* 4)/,%&2#* &/-* #=%8&',*&%,&*)/*J%.8)2,#*SVGB*&/-*SVG? ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((FN* <)3=%,* BFO* +,J$A* -)#$%)"=$)./* .8* .;&2&$,* ,;$%&'$&"2,* J..%2I* '%I#$&22)/,* 4)/,%&2#* &/-* #=%8&',*&%,&*)/*J%.8)2,#*SVGF*&/-*SVGL ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((LG* <)3=%,* BLO* +,J$A* -)#$%)"=$)./* .8* .;&2&$,* ,;$%&'$&"2,* J..%2I* '%I#$&22)/,* 4)/,%&2#* &/-* #=%8&',*&%,&*)/*J%.8)2,#*SVGM*&/-*SVGR ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((L>*

1 Introduction and scene setting

1.1 Overview

While there is a broad understanding of the physical and chemical properties of Iceland soil (Arnalds, 2004; Arnalds, 2009) few studies have focused on soil development during rehabilitation of severely degraded land below the upper 30 cm. The study of Icelandic soil to date has focused on agronomical

(e.g. Gudmundsson et al., 2004) and environmental (e.g. Arnalds, 2000) issues and as such there has been little regard for the field variability of soils or for the most effective approach to effectively sample these soils. Hence a key approach of this study is to consider field variability and multi-genetic origins during pedogenesis.

1.2 Geology and geomorphology of Iceland

Iceland is an active volcanic island located on the North Atlantic ridge, which passes through the country. Total land area is 103,000 km2. Iceland is classically divided into 3 main stratigraphical formations. The Tertiary basalt

(16-3.3 million years), the Plio-Pleistocene (3.3-0.7 million years) and the

Upper Pleistocene palagonite formation (>0.7 million years), which includes

Holocene volcanics. The current volcanic activity is confined to volcanic belts

and the most active are situated on the spreading ridge extending from the southwest to the northeast of the country (Sæmundsson, 1979). Volcanic eruptions are frequent and occur on average every 4-5 years. The chemical composition of the volcanic ejecta varies from basaltic to rhyolitic and the morphology ranges from lava to fine grained tephra (Thordarson and Larsen,

2007). Unlike most volcanic areas, hydromagmatic highly explosive basaltic eruptions are common resulting in widespread distribution of basaltic tephra

(Thordarson and Larsen, 2007).

1.3 Climate

Iceland is situated just below the Arctic Circle between latitudes ~63°-66° N and 13°-24°W. The topography is mountainous with an average altitude of

500 m above sea level. Less than a quarter of the country lies below 200 m.

The climate is humid and cool-temperate, characterized by cool summers and mild winters. Shifts between frost and thaw are common but permafrost is near absent. Annual mean temperature ranges from 2 to 6°C with mean July temperature about 6-10°C (Einarsson, 1976). Mean annual precipitation ranges from 600-2000 mm in lowland areas and <400mm in the interior

(Einarsson, 1976). The growing season is short, 89-144 days per year

(Fri!riksson and Sigur!sson, 1983).

Iceland is situated near the border of warm and cold ocean currents, the warm Irminger current, a part of the Gulfstream and the cold East

Greenland current. During prolonged northerly winds, drift ice is often carried

by the East Greenland current to the coast of Iceland in late winter. As the polar front is often over the North Atlantic, cyclones connected to this front often pass close to Iceland causing strong winds, high precipitation and rapid changes in weather. In winter, the change can be from strong southerly winds and plus 10°C to a northern gale with sub-zero temperatures in a matter of hours (Einarsson, 1976).

1.4 Andisols

Andisols are soils formed in volcanic ejecta and occur in a wide range of climates (Ugolini and Dahlgren, 2002). They are dominated by short-range minerals and Al-humus complexes. Close associations between the organic matter and soil minerals in the colloidal fraction makes the organic matter recalcitrant (e.g. Dahglren et al., 2004; Shoji et al., 1993 and references therein). This results in accumulation of carbon in Andosols and they store the most C per unit area after Histosols (31 and 218 kg m-1 respectively) (Batjes,

1996). Volcanic soils are in general characterized by low bulk density, short- range order minerals (allophane, imogolite and ferrohydrites), high phosphorus adsorption and high levels of organic matter (e.g. Dahglren et al.,

2004; Shoji et al., 1993 and references therein).

1.4.1 Mineralogical Characteristics of Andisols

Rapid weathering of volcanic ejecta commonly leads to formation of poorly crystalline or short-range order minerals that without being crystalline, have a

short range (the same elemental group occurs repeatedly) ordering in their structure. The rapid kinetics of nucleation for these minerals favours their formation over more crystalline minerals (Stumm, 1992). The dominant colloidal assemblage of volcanic soils varies widely and depends on several factors such as composition of the parent material, degree of weathering, pH, soil temperature and moisture regime as well as the accumulation of organic matter (Shoji et al., 1993). These colloids contribute to the unique physical and chemical attributes of volcanic soils due to their high surface area, variable charge surface, high water holding capacity, high phosphate retention, low bulk density and formation of stable soil aggregates (Dahlgren et al., 2004). The dominant colloids in Icelandic soils are allophane, ferryhydrites and Al-humus complexes as well as imogolite (Wada et al.,

1992).

Allophane as defined by Parfitt (1990) as “a group of clay-size minerals with short-range order which contain silica, alumina and water in chemical combination”. It is formed preferentially over Al-humus complexes at pH >5

(Parfitt and Kimble, 1989). They have a pH dependent variable charge and high surface area (Parfitt, 1990). Upon drying the allophanic aggregates interact strongly to form silt and sand sized aggregates that are stable in sodium hexametaphosphate commonly used for texture analysis (Churchman and Tate, 1987). As little as two percent allophane in the soil can be detected by their thixotropic properties during hand texturing (Parfitt, 1990).

1.4.2 Soil formation in volcanic ejecta

Formation of poory-crystalline materials and accumulation of organic matter are the dominant pedogenic processes in most soils formed in volcanic material (Shoji et al., 1993). Translocation of Al, Fe and dissolved organic matter are minimal (Ugolini et al., 1988) and thus the non-crystalline materials in subsurface horizons have formed in situ. Due to often-frequent inundation of tephra it is common to find polygenetic profiles (e.g. Shoji et al., 1994;

Dahlgren et al., 2004). Andisols are often broadly characterized into two groups based on the mineralogical composition of the A-horizon: Allophanic and non allophanic (Dahlgren et al., 2004 and references therein). The availability of Al3+ appears to be the regulating factor with pH 5 being the break point between the two groups with allophane dominating at higher pH.

The composition of the tephra is one of the factors controlling the pH with base poor tephras (e.g. rhyolitic) often resulting in lower pH of the soil

(Dahlgren et al., 2004 and references therein). Andisols generally form rapidly in humid climates and alter to other soil orders with age, but in areas with intermittent inundation of volcanic ash where each additions rejuvenates the soil forming processes Andisols are maintained as a relatively stable soil condition (Shoji et al., 1993).

1.5 Icelandic Soils

Icelandic soils are predominantly volcanic in origin with an estimated 86% belonging to the Andisol soil order (Arnalds, 2004). Soil formation is heavily 5

influenced by a steady flux of eolian materials, frequent tephra events and climate (Arnalds, 2004; Arnalds, 2008). Icelandic soils are unique among volcanic soils as they are young, are of basaltic origin, receive large amounts of eolian input and occur at low temperatures with a wide range of precipitation (Arnalds, 2004; Arnalds, 2008). Cryoturbitation is more pronounced in Iceland than any other Sub-Arctic region (Orradóttir et al.,

2008) with corresponding pedological impacts.

Arnalds (2004) introduced a classification scheme for Icelandic soils based mainly on the world reference base (WRB) (FAO, 1998). Six main soil types were identified, Histosols, Histic Andosols, Gleyic Andosols, Brown

Andosols, Vitrisols and Leptisols. The scheme reflects mineralogical characteristics and organic matter content of the soils, which in turn are related to drainage conditions and intensity of eolian additions (Arnalds, 2004;

Arnalds, 2008). These parameters are more connected with geology than climate with, in general terms, the carbon rich Histosols mostly on impermeable Tertiary formations and the Vitrisols follow the active volcanic zones (Arnalds, 2008).

Aeolian deposition is a controlling factor in the classification scheme described above and its importance is emphasised in the recent revised soil map and classification scheme by Arnalds and Óskarsson (2009). The aeolian dust mainly comes from two sources, extensive deserts and confined plume areas. Erosion fluxes can reach more than 2000 kg m-2 day-1 during storms and the deposition rates vary from <25 g m-2 yr-1 farthest away from

dust sources to > 500 g m-2 yr-1 close to or in sandy deserts (Arnalds, 2010).

The chemistry of the dust is different than normal globally, where quartz and phyllosilicates dominate (Lawrence and Neff, 2009) whereas in Iceland it is dominated by volcanic materials that have a high surface area and high dissolution rates (Gíslason, 2008).

The desert soils are typically sandy Andisols with low water holding capacity, limited sources of macronutrients, rich in volcanic glass and have low amounts of allophane clay and organic C compared with vegetated areas

(0.08-0.5 kg C/m2in desert soils compared to 40-90 kg C/m2 in brown

Andisols) (Óskarsson et al., 2004; Arnalds and Kimble, 2001). Many deserts are erosional surfaces once covered with vegetation (Arnalds and Kimble,

2001).

1.6 Land degradation and land restoration in Iceland

The vegetation cover at the time of settlement (AD 874) is estimated to have been about 50-60% with birch woodlands (Betula pubescens) covering about

25-30% of the country (Anonymous, 2001). Today, only about 40% of Iceland is covered with continuous vegetation of which only 1% is original birch woodland (LMI, 1993). The drastic decrease in birch-dominated ecosystems has been recorded using pollen analysis (Hallsdóttir, 1995; Olafsdottir et al.,

2001) and forest remnants, in tandem with place names (e.g. Bjarnason,

1974). The reason behind the degradation is believed to be a mixture of management practises, e.g. charcoal making, overgrazing and winter grazing

in forests and natural disasters, e.g., large scale volcanic eruptions as well as less favourable climate during the ‘little Ice Age’ (Arnalds et al., 1997;

Dugmore et al., 2005). Perhaps the starkest evidence of the massive soil erosion over the last 11 centuries are the barren landscapes of the black

‘deserts’ covering about 37,000 km2 plus an additional 10-15,000 km2 with limited vegetation production (LMI, 1993). The estimated extent of barren areas at the time of settlement is 5-15,000 km2 (Arnalds, 2000). In comparison, glaciers cover an estimated 10,000 km2.

The ecosystem degradation in Iceland has been conceptualized as a stepwise problem, where ecosystems with a higher erosion tolerance are degraded through overexploitation to ecosystems with a lower tolerance. An ecosystem can be ‘pushed’ over the stability boundary through natural disasters, e.g. volcanic eruptions or prolonged cold periods (Aradottir et al.,

1992; Dugmore et al., 2005; Dugmore et al., 2009). The high susceptibility of

Icelandic Andisols to cryoturbation, landslide, wind and water transport make them highly erodible (Arnalds, 1997; Wada et al., 1992).

Total loss of soil organic carbon (C) is estimated to be 120-500 x 106

Mg since settlement; representing a serious environmental problem still ongoing with estimated yearly losses 50-100 x 103 Mg C year-1 (Óskarsson et al., 2004).

In response to severe land degradation Iceland has a long history of concern for the land, and rehabilitation efforts have continued for more than

100 years. Carbon sequestration is a major benefit of land rehabilitation and

revegetation programs, particularly in the severely degraded desert areas

(Arnalds et al. 2000). The carbon storage in young non-forested reclaimed areas is 0.01 to 0.5 t C/ha for vegetation, both above and below ground

(Aradottir et al., 2000) and the sequestering rate 0.6 t C/ha/yr in soils for reclaimed areas a rate maintained for >50 years (Arnalds et al., 2000).

1.7 Carbon stocks in forested ecosystems in Iceland

Relatively few studies have focused on carbon stocks and carbon sequestration potential in afforestation in Iceland. Snorrason et al. (2002) studied areas both planted with native Betula pubescens and non-native species (Larix sibrica and Picea sitchensis). The stands were compared with unplanted adjacent pastures to determine the carbon sequestration rate (C- sequestration). They found that B. pubescens had the lowest C-sequestration potential, or 1 Mg ha-1yr-1 of the three. The major change in total carbon was due to the larger biomass of those species compared to B. pubescens but the largest carbon pool was in the soil. No significant increase in soil carbon was detected, in part due to the large spatial variability of the soils.

Ritter (2007) studied a chronosequence of planted stands of B. pubescens and L. sibirica established on native heath land. Increase in soil carbon at >10 cm was found to be insignificant in forested areas compared to the native heath but C-sequestration rates were found to be 0.23 MgC ha-1yr-1 in the upper 10 cm.

1.8 The Soil Carbon cycle

The soil carbon pool is the largest actively cycling C in terrestrial ecosystems

(Amundson, 2001). Plants take up carbon dioxide (CO2) from the atmosphere during photosynthesis forming organic molecules by trapping sun light into carbon-to-carbon bonds. The majority of the molecules are used by the plant as sources of energy and are released back to the atmosphere as CO2 via respiration. Part is stored as part of the standing vegetation, most of which will in time be returned to the soil as plant litter or root deposition. The larger portion of the litter is consumed by other organisms and returned to the atmosphere as CO2 (Brady and Weil, 2004). The stability of carbon in the soil is dependent on several factors and carbon oxidation is enhanced by sufficient O2 supply within the soil, soil temperature, soil moisture and soil aggregate stability (Gobat et al., 2003). Decomposition of soil organic matter in Andisol is mainly limited by close association of OM with the mineral fraction and high phosphate adsorption limiting P availability to decomposers

(Dahlgren et al., 2004).

1.9 Betula spp.

Birch species are light-demanding pioneer species capable of rapid colonization and occupation of open areas and are an important ecological component in northern temperate and boreal forests (Hynynen et al., 2009).

Two birch species have been identified in Iceland, downy or mountain birch

(Betula pubescens) and dwarf birch (B. nana). They co-exist in Iceland and 10

hybridization between the two species is common (Thorsson et al., 2007).

Birch colonized Iceland shortly after the glaciers started to retreat and the distribution reached a maximum 9000-7000 years ago (Hallsdóttir, 1995). The morphology and characteristics of Icelandic forests are very variable, reflecting site conditions, forest management and genetics (Aradottir et al.,

2001). Birch trees are highly variable with often with multiple, crooked and leaning stems of various sizes (Jonsson, 2004). The ground vegetation of birch habitats is generally thick and commonly comprised of shrubs,

Vaccinum uliginosum L., Salix lanata L. and Empetrum nigrum N., the forb

Rubus saxatilis L. and the grasses Deschampsia flexuosa L. Tron. and

Agrostis capillaris L. among others (Aradottir et al., 2001). It must be noted that minimum values of crown cover, tree height and stand size of most

Icelandic birch forests may not fulfil the general forest definition of UNFCCC.

1.10 The Kolbjörk (CarbBirch) project

1.10.1 Overview

Kolbjörk is a research project focusing on ecosystem development and potential for carbon sequestration using mountain birch (Betula pubescens) for land restoration on severely degraded areas. In view of government plans to reforest large parts of the country with native mountain birch, further research on the ecosystem development in afforested areas is essential. Only one study is available for carbon sequestration in areas afforested with native

birch (Snorrason et al., 2002). The Kolbjörk project is a comprehensive study of ecosystem changes, carbon sequestration and carbon flux in reclaimed mountain birch areas. The project covers changes in understory vegetation community composition, carbon fluxes and productivity of plant biomass, colonization of mycorrhizal fungi and soil development in a chronosequence of established tree plots dating back 60 years with comparison to natural old growth forests in the area. Typically restoration efforts are directed at areas that are distant from remnant natural forests.

1.10.2 Study Area

The research sites are in a severely degraded area close to an active volcano, Hekla in southern Iceland (fig 1). The landscape has been highly eroded since settlement times. The land is inundated on a regular basis with volcanic ejecta from nearby volcanic systems, covering vegetation and causing destabilisation of surfaces by both directly killing vegetation and subsequent erosion due to aeolian movement of tephra cutting and covering above ground biomass. Mature forested systems appear to be tolerant to tephra fall as several old-growth forests still exist in the area and probably have survived since before settlement times (Aradottir et al., 2007). In order to stabilize the area and prevent further erosion due to volcanic events a large reforestation project is underway. The aim is to establish forests over approximately 900 km2 using native species, both mountain birch and willows

(Aradottir et al., 2007).

1.11 This project

The role of this study is twofold. First I will establish the rate of carbon sequestration in the soils when using Betula pubescens for land restoration. It has been speculated that sequestration of soil carbon is enhanced when using forest vegetation for restoration. In addition, add a significant contribution to understanding soil carbon dynamics during reclamation and will help establishing rates for carbon budget calculations.

Secondly, I aim is to put some constraints on soil development and evolution with age in the restored areas and how quickly they reach the

‘steady state’ expected in the region. The time frame is short, only 65 years, but due the rapid kinetics of soil formation in volcanic ejecta, significant trends in concentration and depth distribution of short-range minerals and carbon are expected with age. The distribution of these minerals in turn determines the nutrient and water holding capacities of the soil as well as contaminant transport. The information published here will help create overall ecosystem change model for these areas.

2 Changes in soil carbon associated with rehabilitation of severely degraded land in Southern Iceland with downy birch (Betula pubescens)

2.1 Introduction

2.1.1 Chapter overview

The aim of this chapter is to determine the carbon sequestration rate from cores taken from all the study areas.

2.1.2 Soil carbon pool

Soils are the largest terrestrial reservoir for carbon with global stocks believed to be 1500-2300 Gt C compared to 760 Gt C in the atmosphere and 500-560

Gt C in the biomass (Jobbágy and Jackson, 2000; Amundsson, 2001). The high level of organic carbon in soils has made management practices that increase the carbon sequestration more tempting to mitigate the increase in atmospheric CO2. The vast spatial variability in soils makes quantification of carbon sequestration difficult. Variations in soil carbon levels depend on many factors, e.g., soil type, environmental conditions, land use and land use history (Brady and Weil, 2004). The two soil types that store the most carbon

per unit area are Histosols and Andisols with global average values of 218 and 31 kgC m-2 respectively (Batjes, 1996).

2.1.3 Icelandic soils

The soils in Iceland are predominantly volcanic in origin belonging to the

Andisol order (Arnalds, 2004; Arnalds. 2008). The soils are young, having mostly developed since the younger Dryas period around 10 000 years ago.

They are heavily influenced by aeolian deposition, in part due to volcanic activity but also due to severe wind erosion in large parts of the country. Other influential factors are the cold maritime climate and intense cryoturbation

(Arnalds et al., 2000; Arnalds, 2008; Arnalds and Kimble, 2001). The main broad categories of Icelandic soils are freely drained brown Andisols, both mineral and organic wetlands soils and soils of the barren deserts (Arnalds,

2008). The desert soils are typically sandy Andisols with low water holding capacity, limited sources of macronutrients, rich in volcanic glass and have low amounts of allophane clay and organic C compared to vegetated areas

(0.08-0.5 kg C m-2 in desert soils compared to 40-90 kg C m-2 in brown

Andisols (Óskarsson et al., 2004; Arnalds and Kimble, 2001). Many present- day deserts in Iceland were once covered with vegetation (Arnalds and

Kimble, 2001).

2.1.4 Vegetation and land degradation

The vegetation cover at the time of settlement (AD 874) is estimated to have been about 50-60% with birch woodlands (Betula pubescens) covering about 15

25-30% of the country (Anonymous, 2001). Today, only about 40% of Iceland is covered with continuous vegetation of which only 1% is original birch woodland. Total loss of soil organic carbon (C) is estimated to be 120-500 x

106 Mg since settlement; representing a serious environmental problem still on-going with estimated yearly losses 50-100 x 103 Mg C year-1 (Óskarsson et al., 2004). In response to severe land degradation Iceland has a long history of concern for the land, and restoration efforts have continued for more than 100 years. Carbon sequestration is a major benefit of land restoration and revegetation programs, particularly in the severely degraded desert areas. The carbon storage is 0.01 to 0.5 t C/ha for vegetation, both above and below ground (Aradottir et al., 2000) and the sequestering rate 0.6 t C/ha/yr in soils for reclaimed areas a rate maintained for >50 years (Arnalds et al.,

2000). Both numbers are for unforested reclaimed areas.

2.1.5 Objectives

The objectives of this project are to show that there is a significant trend of increasing carbon content with age over just 65 years when using Betula pubescens for restoration of severely degraded land and to create a simple model for the carbon increase.

2.2 Methods

2.2.1 Study Area

The research sites are in a severely degraded area close to an active volcano, Hekla in southern Iceland (fig 1 and 2). The landscape has been severely eroded since settlement times. The land is inundated on a regular basis with volcanic ejecta from nearby volcanic systems, covering vegetation and causing destabilisation of surfaces by both directly killing vegetation and subsequent erosion due to aeolian movement of tephra cutting and covering above ground biomass. Mature forested systems appear to be tolerant to tephra fall as several old-growth forests still exist in the area and probably have survived since before settlement times (Aradottir et al., 2007). In order to stabilize the area and prevent further erosion due to volcanic events a large reforestation project is underway. The aim is to establish forests over approximately 900 km2 using native species, both mountain birch and willows

(Aradottir et al., 2007).

Figure 1: Location of study area

Figure 2: Distribution of study sites. 1 is Gunnlaugsskógur, 2 Bolholt, 3 StóriKlofi, 4 Hraunteigur and 5 Búrfell (Halldórsson et al., 2009)

2.2.2 Sampling scheme

The research sites were split into age groups (table 1). Within each age group in each area, three 20x10 m plots were randomly chosen. Each plot was split into two 10x10 m sub-plots, one for sampling and one for monitoring. Soil samples were taken on transects, with 5 sub-samples taken at 2 m intervals combined into one sample. Samples were taken in two replicates. Each sample was split into three depth increments 0-5, 5-15 and 15-30 cm. Bulk density was determined by the core method in small profiles taken adjacent to the areas. Figures 3, 4 and 5 show location of research sites.

Table 1: Summary of study areas. Stóri Klofi (SK), Bolholt (B) and Hraunteigur (H)

Land type Research sites Unvegetated, eroded SK, B Unforested, revegetated SK, B Pioneering stage, 0-15 year birch SK Young forest, 10-20 year birch B Young forest, 20-30 year birch B Forest, 35-45 year birch SK Forest, 45-65 year birch SK Oldgrowth forest H

Figure 3: Research plots at Bolholt

Figure 4: Research plots at Stóri Klofi

Figure 5: Research plots at Hraunteigur

2.2.3 Analysis

All samples were sieved through a 2 mm sieve and the fine fraction dried at

60°C. Total carbon and nitrogen was measured at the Agricultural University of Iceland using the dry combustion method using dry combustion (Vario MAX

C/N – Macro Elementar Analyzer). Samples were checked for moisture content at the time of analysis and results adjusted accordingly (minor adjustments) (Blakemore et al., 1987). Bulk density was determined by using the inner diameter of the sample probe as well as with the core method in profiles dug beside the plots. The average of the two bulk densities were used for calculations. Soil pH was measured in water, KCl (Blakemore et al., 1987).

The data was analysed by age grouping (table 1) and study areas. All depth 21

increments were analysed separately. Difference between sites were analysed using the unvegetated and unforested plots in each site. Graphing was done in SigmaPlot 11 using 95% confidence intervals. Other statistical analyses were done using JMP 9.

2.3 Results

2.3.1 Soils in the sample areas

Due to intensive soil erosion and subsequent movement of material with wind and water, the spatial variability of the soils is very great inside each area. In many areas buried soils can be found at less than 30 cm depth from current surface with aeolian and fluvial sand in between (fig 6 and 7). Soils in the old- growth areas show effects of aeolian deposition, but limited erosion.

Figure 6: Soil core from Figure 7: Soil core from unvegetated plot at Stóri Klofi unforested plot at Stóri Klofi

2.3.2 Carbon

Figures 10 to 12 show total soil carbon as weigh percent plotted against treatments for all plots and depths. Figures 13 to 15 show the same for soil carbon adjusted to weight per area using measured bulk density.

The regression analysis shows that there is a trend towards increased carbon content with age of reclaimed plots in all depth increments.

2.3.3 Difference between study areas

Comparisons on carbon levels in unvegetated, plots with less than 5% vegetation cover, and unforested, fully vegetated but no forest vegetation, plots between the two study areas, Bolholt and StóriKlofi show that there is a significant difference (p-value 0.0025) in the means of both C% and g C m-2

(table 2).

Table 2: Average values for gC m-2 with 95% confidence intervals for the plots without forest at Stóri Klofi (SK) and Bolholt (B).

Site Treatment Average values (g C m-2) SK Unvegetated 134.88 ± 35.35 B Unvegetated 808.90 ± 378.63 SK Unforested 434.04 ± 132.06 B Unforested 709.44 ± 90.99

2.3.4 C:N ratios

C:N ratios are an indicator of the availability of N to plants and the type of organic matter with low values indicating more readily available N and less

‘evolved’ soil carbon. There is no significant difference in C:N values but the trend is towards stabilizing at a C:N ratio of 15-20 (figures 16 - 18) with age.

2.4 Discussion

2.4.1 Carbon sequestration

It is difficult to make ‘perfect’ models of soil carbon sequestration due to the heterogeneity of soils in general. The data presented here is an addition to understanding carbon sequestration in reclaimed areas in Iceland. The sequestration rate calculated here, 46.76 g C m-2 yr-1 (0.466 t ha-1 yr-1) in 0-30 cm, is similar to what Arnalds et al. (2000) report for reclamation sites spread around Iceland but lower than reported for agricultural soil under long-term fertilizer application in Iceland (Gudmundsson et al., 2004). The lower sequestration rate can be attributed to lack of nutrient input inhibiting biomass production and the effects of acidification due to fertilizer application inhibiting microbial decomposition.

The trend for increased carbon contents in restored soils with time from initial restoration is clear. This trend can be seen down to 30 cm, which is in contrast to other studies on reforested sites in Iceland (see Snorrason et al.,

2002; Ritter, 2007). The reason could be that those studies were carried out

on relatively ‘healthy’ soil while this study focuses on severely degraded soil in an area where erosion is still significant and ongoing with considerable aeolian deposition each year. Arnalds et al. (2000) does not offer a comparison between depth increments, but it can be speculated that the presence of trees with extensive root systems compared with the grasses normally used for restoration help explain this soil carbon increase at depth.

Aeolian deposition in the area is estimated to be >250 g m-2 yr-1 and as much as 500 g m-2 yr-1 (Arnalds, 2010) so it can be concluded that over 65 years soil thickening is substantial due to this. Additions like these result in lower percentages of carbon in the soil, but overall greater sequestration

(Óskarsson et al., 2004). To further establish the effects of aeolian deposition, rates soil thickening would have to be determined.

The stronger correlation of carbon increase with age at depth, 15-30 cm, indicates that the aeolian additions are muddling the numbers closer to the surface, possibly because the windblown material does include some remnants of eroding soil that include ‘older’ organic matter.

The IPCC (2000) cites a short period of carbon sequestration in soils, or 4-25 years but the trend for these sites indicate that even after 65 years of forest vegetation the carbon levels in the soils are not as high as in old-growth forests in the area, which indicates that the sequestration period is longer than 65 years. The data is not yet sufficient to make elaborate models on changes in soil carbon sequestration with age but a strong trend can be seen.

As the unforested plots cannot be accurately dated, no effects of forest vegetation can be usefully determined from this data. A comparison with revegetated, but unforested, sites of different ages in the area would give a better understanding and help determine if there is a significant difference in carbon sequestration in soils under forest vegetation. Previous studies on afforestation in Iceland have been unable to show a significant increase in soil carbon below 5 cm (Snorrason et al., 2002; Ritter, 2007) but as those studies were mostly carried out on non-eroded soil, comparisons are difficult but it is likely that the high level of carbon in the soils prior to afforestation complicate matters somewhat.

2.4.2 Difference between study areas

The difference between the unvegetated and unforested plots in Stóri Klofi and Bolholt is explained by the difference between the morphology of the soils in the two areas. The plots at Bolholt are situated on a lava field in areas were aeolian deposition from the eroding neighbourhood gets caught in crevasses while at Stóri Klofi the soil, whilst also on a lava field, is formed on fluvial and aeolian sand (see figures 8 and 9).

Figure 8: A soil profile from Figure 9: A soil profile from unvegetated plot at unvegetated plot at Bolholt. StóriKlofi. White grains are Darker layers are individual rounded tephra, black tephra layers with aeolian basaltic glass and rock. matter in between.

2.5 Conclusions

There is a strong trend for increase in soil carbon with time from restoration.

The general sequestration rate is calculated as 11.67 gC m-2 (116 kg C ha-1 yr-1) in 0-5 cm, 21.34 g C m-2 (213.4 kg C ha-1 yr-1) in 5-15 cm and 13.75 g C m-2 yr-1 (137.5 kg C ha-1 yr-1) which gives a total sequestration rate of 46.76 g

C m-2 yr-1 (0.466 t ha-1 yr-1) in 0-30 cm. These rates are slightly lower than the average rates, 0.6 t C m-2, Arnalds et al. (2000) reported for soil restoration 27

areas spanning most of the country. The difference is possibly due to differences in modelling and the fact that most of the areas in Arnalds et al.

(2000) have received significant inputs of fertilizer over the time period of 60 years.

The C:N ratios indicate that plant available nitrogen is a limiting factor in plant production. Adding fertilizer would help increase carbon sequestration.

Aeolian deposition, significant in the area, masks the increase in carbon with age to some extent, increasing the variance in 5-15 cm depth increment especially. This shows the need for accounting for rapid aeolian deposition in all carbon sequestration models in high deposition areas in

Iceland e.g. by measuring soil thickening rates using tephra chronology, by direct measurements, or estimation from generalized models of deposition in

Iceland.

As the oldest restored forest areas still have significantly less carbon in

0-30 cm compared to the old growth forest, it can be concluded that the carbon sequestration period is longer than 65 years. In fact, as a result of rapid soil thickening due to windblown material, it can be assumed that the old growth forest is still sequestering soil carbon.

Figure 10: %C in 0-5 cm regression fit: %C = 1.3357944 + 0.0415339*age, R2 = 0.4495, P= <0.001

Figure 11: %C in 5-15 cm, regression fit:%C = 0.6969256 + 0.0192998*age R2= 0.26559, P = <0.0001

Figure 12: %C in 15-30 cm, regression fit: %C = 0.6324487 + 0.0083007*age, R2= 0.1486, P= 0.0044

Figure 13: gC m-2 in 0-5 cm, regression fit: gC m-2 = 418.43774 + 11.670255*age, R2= 0.4246, P= <0.001

Figure 14: gC m-2 in 5-15 cm, regression fit: gC m-2= 614.32916 + 21.343773*age, R2= 0.2671, P= <0.001

Figure 15: gC m-2 in 15-30 cm, regression fit: gC m-2= 526.89819 + 13.753028*age, R2= 0.5029, P= <0.0001

Figure 16: C:N ratios in 0-5 cm

Figure 17: C:N ratios in 5-15 cm

Figure 18: C:N ratio in 15-30 cm

3 Soil formation and profile development under Betula pubescens in southern Iceland

3.1 Introduction

This chapter aims to establish soil change under forest vegetation over a period of 65 years. The objective is to show changes in morphology and the depth distribution and concentrations of carbon and poorly crystalline minerals and the subsequent effects on the specific surface of the soils and how this relates back to the field description.

3.1.1 Soil formation

Formation of poorly crystalline materials and accumulation of organic matter are the dominant pedogenic processes in most soils formed in volcanic material (Shoji et al., 1993). Translocation of Al, Fe and dissolved organic matter is minimal (Ugolini et al., 1988) and thus the non-crystalline materials in subsurface horizons have formed in situ. Rapid weathering of volcanic ejecta commonly leads to formation of poorly crystalline or short-range order minerals. The rapid kinetics of nucleation for these minerals favors their formation over more crystalline minerals (Stumm, 1992).

3.1.2 Allophane

Allophane is a group of short-range order minerals that contain Al and Si in a specific combination. The composition depends on leaching and weathering conditions as well as the chemical composition of the parent materials during allophane formation (Parfitt, 1990). The specific surface area of allophane is large: 435 – 534 m2 g-1 (Wada, 1989). Therefore the surface area of the soils may change significantly during soil formation, which again affects the nutrient and pollutant holding and recycling capacity.

3.2 Objectives

The objectives of this study is to determine whether there is a significant change in soil properties over the time period of just 65 years when using Betula pubescens for restoration of severely degraded land. This includes significant differences in concentrations and depth distribution of poorly crystalline minerals, carbon and the subsequent changes in specific surface.

3.3 Methods

3.3.1 Study sites

The study took place in a reforested area close to an active volcano, Hekla, in southern Iceland. The area has suffered from severe soil erosion and land degredation in the last 1200 years (Hjartarson, 1995). The main study site, 35

Stóri Klofi, has been revegetated in the last 70-80 years. 60-70 years ago downy birch (Betula pubescens) seeds were sown in an area close to a small stream and the area protected from grazing. In the years since, the forest has been spreading naturally from that original core. The parent material is both water and wind reworked tephra covering paleosol formed before the area lost most of its topsoil due to soil erosion. The area receives continuously a significant amount of aeolian deposition originating from desert areas further inland. The deposited material is mostly silt-size basaltic glass and primary minerals, with a significant amount of eroded soil materials (Arnalds, 2010).

The soil profiles were dug in three areas where a certain age group of birch trees dominated, two profiles in each age group. The groups are as follows: the oldest part of the forest (50-65 yrs), younger stand (5-15 yrs) and fully vegetated (vegetation cover >90%) but non-forested site. All the profiles are within an area of 1.5 km2. Profiles in an old undisturbed forest 12 km away

(Hraunteigur) were described and sampled for comparison (table 3).

Table 3: Age distribution of profiles

Group Profiles

50-60 year old forest PK01, PK02

5-15 years, pioneering stage PK03, PK04

Unforested PK05, PK06

Oldgrowth PK07, PK08

3.3.2 Sampling

Representative locations for the pedons were chosen at Stóri Klofi after all sites were checked for variability using a soil corer to measure the soil thickness between prominent tephra layers. Stóri Klofi soils showed less variation than those at the Hraunteigur site. Profiles were described in the field following USDA guidelines (Schoeneberger, 2002), modified for Icelandic conditions (Arnalds et al., 2005) and sampled by horizons from the wall of the pit. Bulk density of the soil was determined by the core method. A metal cylinder with a known volume was forced into the profile and the soil and cylinder weighed before and after drying for 24 hrs at 105°C. The bulk density is the weight of the soil divided by the volume of the core.

3.3.3 Chemical analysis

Soil samples for chemical analysis were dried at 40°C and passed through a

2 mm sieve for homogenization. Soil pH was measured by shaking 5 g of soil

(<2 mm) with 25 ml of de-ionized water for two hours and reserve acidity in a weak (1 M) KCl solution. The pH was then determined with a glass calomel electrode (Blakemore et al., 1987). Total C and N were measured in 1.5 g of soil (<2 mm) by dry combustion (Blakemore et al., 1987) using Vario MAX CN

– Macro Elementar Analyzer (Elementar Analysing systems GmbH). Do to the fact that upon drying the allophanic aggregates interact strongly to form silt and sand sized aggregates that are stable in sodium hexametaphosphate commonly used for texture analysis (Churchman and Tate, 1987).Ammonium

oxalate extractable Al, Fe, Mn and Si (Alox, Feox, Mnox and Siox) were determined by the method described by Blakemore et al. (1987) and the metal concentrations in the extract measured by inductively coupled plasma

(ICP).

Concentrations of non-crystalline minerals were calculated as summarized by Ulery and Drees (2008). Non-crystalline iron minerals are calculated as ferrihydrite using the formula

%ferrihydrite = %Feox * f

Where f = 1.7.

Allophane concentrations are calculated using the equation

%allophane = %Si * f

The factor f depends on the Al/Si molar ratio and distinguishes Si-rich allophane (Al/Si near 1:1) and Al-rich allophane (Al/Si 2:1). For an Al/Si ratio of 1:1 the factor is 5, for 2:1 the factor is 7.

Specific surface was determined by nitrogen adsorption. The method assumes that the non-polar gas molecules are adsorbed in a monolayer on the surface, and the amount of gas adsorbed can be calculated by constructing an adsorption isotherm and using the BET equation (Ulery and

Drees, 2008).

3.3.4 Statistical analysis

Changes in properties with soil depth were graphed using SigmaPlot 11. All other statistical analysis were completed using JMP 9.

3.4 Results and discussion

3.4.1 Profile descriptions

Table 4 shows the morphological descriptions of all the profiles and figures

19-26 show what they look like in the field. Noticeable is a buried soil occurred at the bottom of all profiles at Stóri Klofi (profiles PK01 to PK06)

(probed to with a soil corer, if not described). This soil has several easily describable characteristics that do not change significantly between profiles.

The age of the palaeosol is not known but is at least several hundred years old, as the area has undergone severe erosion. It is likely that the top of the palaeosol not the surface from when erosion started, as both wind and water would have cut into the soil. The C-horizons at Stóri Klofi show artifacts from the intense reworking expected in an actively eroding area, such as rounding of grains, cross stratification and a mixture of lithologies (basaltic, rhyolitic and andesitic tephra). The agents of transport, water or wind, are not easily distinguished but the proximity of the stream would indicate that water reworking would most likely play a significant role. In all likelihood, both erosional agents are at play. The influence of aeolian depositions can be seen in the relative increase of the silt fraction in the surface horizons. As the areas become vegetated, plants act as a sediment trap, capturing suspended particles.

The fibric surface horizon found in many of the profiles did not meet the criterion for an O horizon (>12% C) except in PK02 and PK07 (see table 5).

The fibric nature of these layers in the field made textural and structural estimation difficult. The profiles do show a visual and descriptive difference in soil development with age of tree stands. The older the trees, the deeper and clearer is soil development as indicated by structure grade and color (fig 19-

26).

All profiles at Stóri Klofi except PK06 had water tables present within the depth of each soil pit indicated by evidence of redoximorphic features, limiting the description depth. This potentially limits the value of comparison with the old-growth forest, where the profiles did not exhibit saturation at the time of sampling. All profiles at Stóri Klofi, again except PK06, had a dense layer below the free water surface (sampled and checked for by probing further down at the bottom of the soil pit). The origin of this dense layer has not been explained as yet, but it is plausible that water is perched on top of this poorly permeable layer. The layer was observed as base level in the adjacent stream.

Table 4: Profile descriptions and horizon designators Horizon Depth (cm) Description PK01 7.5 YR 3/2, loam, medium fine granular structure. Many very fine to medium and moderately few A 0-7 medium to very coarse roots. Boundary abrupt and smooth. 5YR 2.5/2; fine sandy loam; fine medium sub- A2 7-10 angular blocky structure; few fine to medium roots; abrupt and smooth boundary. 5YR 3/2; sandy loam; fine weak sub-angular Bw 10-17 blocky structure; few fine and few medium roots; abrupt and smooth boundary. 7.5YR 3/0; loamy sand; weak, fine sub-angular 2C 17-33 blocky; few fine roots; abrupt boundary; common coarse 5YR 5/6 mottles. 5YR 3/1; pebbly loamy sand; weak, fine sub- C 33-45 angular blocky; very few fine roots; wavy, clear boundary; few prominent 5YR 5/6 mottles. 7.5YR 3/2; pebbly loam; moderate, fine sub- angular blocky; very few fine roots; common, clear 3Bw 45-57 5YR 5/6 concentrations; common, distinct coarse 7.5YR 3/2 depletions; clear wavy boundary. Black; loamy sand; variable thickness; clear 4C 57-63 boundary. 7.5YR 3/2; silt loam; moderate, fine sub-angular 5Bw 63+ blocky; common, distinct 5YR 5/6 mottles; common, distinct 7.5YR 3/2 depletions. PK02 Oi 0-5 Fibric; abrupt boundary 7.5YR 3/2; loamy sand; moderate, medium A 5-11 granular; common fine roots; clear boundary 7.5YR 3/2; loamy fine sand; moderate, fine sub- Bw 11-23 angular blocky; few fine, few medium roots; clear wavy boundary. 7.5YR 2.5/0; loamy sand; weak, fine sub-angular 2C 23-46 blocky; few, medium, distinct 7.5YR 4/4 mottles; abrupt boundary; distinct cross stratification. 5YR 3/2; loam; medium, fine sub-angular blocky; 3Bwg 46-71+ few fine roots; prominent 5YR 4/6 root-channel concentrations. PK03 A 0-4 Fibric; abrupt boundary 5YR 2.5/2; sandy loam; moderate, fine, sub- A2 4-10 angular blocky; many fine roots; abrupt boundary. 5YR 2.5/2; gravelly sandy loam; moderate, fine Bw 10-17 sub-angular blocky; common fine and medium roots; clear wavy boundary. 5YR 2.5/1; gravelly loamy sand; weak, fine sub- 2C 17-36.5 angular blocky; common fine, few medium and coarse roots; clear wavy boundary. Light/dark tephra; pebbly, loamy coarse sand; 3C 36.5-47.5 massive; common fine roots; clear wavy boundary. Reworked tephra; gravelly very coarse sand; 4C 54.5-61.5+ massive; very few fine roots; lighter tephra coated with 5YR 5/6 concentrations Continued 41

Table 4: Continued PK04 Oi 0-3 Fibric, gradual boundary 5YR 3/2; fine sandy loam; moderate fine granular; A 3-7 many fine, few coarse roots; abrupt boundary. 5YR 3.5/2; fine, sandy loam, 5% fine sub-rounded tephra; moderate, fine, sub-angular blocky; Bw 7-17 common fine and few coarse roots; abrupt wavy boundary. 5YR 2.5/1; gravelly coarse sand; single grain; few 2C 17-33 fine roots; clear wavy boundary. 7.5YR 2/0; medium sand, lenses with coarser 2C2 33-68+ white tephra; single grain; very few fine roots. PK05 A 0-3 7.5YR 2/3; fibric; abrupt boundary 5YR 2.5/2; loamy sand; weak, fine granular; many AC 3-23 fine, few medium and coarse roots; abrupt wavy boundary. 5YR 3/1; fine sandy loam; single grain; common 2C 23-33 fine, few medium and coarse roots; clear, wavy boundary; stratified. 7.5YR 7.5/0; loamy fine sand; single grain; few fine 2C2 33-50 and medium roots; clear wavy boundary. 7.5YR 3/4; loam; moderate, fine sub-angular blocky; few fine and medium roots; common faint 3Bw 50-74+ to distinct coarse depletions 7.5YR 4/2, few fine root channel concentrations 7.5YR 5/6; clear wavy boundary. PK06 A 0-3 Fibric; abrupt boundary 5YR 2.5/1; loamy sand; weak, fine granular; many A2 3-9 fine, common medium roots; abrupt boundary. 5YR 3/1; pebbly coarse sand; weak fine sub- Bw 9-23 angular blocky; common fine and medium roots; clear wavy boundary. 5YR 2.5/1; pebbly loamy fine sand; weak, fine, sub-angular blocky; common fine, few medium 2C 23-41 roots; multiple deposit lenses; clear, irregular boundary. 5YR 3/2; pebbly, loamy sand; weak, fine, sub- 2C3 41-56 angular blocky; few, fine roots; abrupt boundary. 5YR 3/3; loam; moderate, fine, sub-angular blocky; 3Bw 56-96 few fine roots; common faint 5YR 4/2 coarse depletions.

Continued

Table 4: Continued PK07 Oi 0-5 Fibric; clear boundary 5YR 2.5/1; fine sandy loam; strong, fine granular; A 5-12 many fine, common medium, few coarse roots; abrupt boundary. 7.5YR 3/2; fine sandy loam; moderate, fine sub- Bw 12-32 angular blocky; many fine, common medium, few coarse roots; clear boundary. 5YR 3/2; fine sandy loam; moderate, fine sub- Bw2 32-55 angular blocky; common fine, few medium roots; clear boundary. 5YR 3/2; fine sandy loam; moderate, medium sub- Bw3 55-78 angular blocky; few fine, few medium roots; clear wavy boundary. 7.5YR 3/2; loamy sand; weak, fine sub-angular 2C 78-85 blocky; few fine, few medium roots; irregular, abrupt boundary. 7.5YR 3/4; fine sandy loam; moderate, fine sub- 3Bw 85-112 angular blocky; Strongly pedoturbated and irregular; layer of rocks on top. PK08 7.5YR 3/2; silt loam; strong, fine granular; many A 0-9 fine, many medium, common coarse roots; clear boundary. 5YR 4/2; silt loam; weak fine granular; fine AB 9-21 common, many medium, few coarse; clear boundary. 7.5YR 3/4; sandy loam; medium, fine, sub-angular Bw 21-36 blocky; common fine, few medium roots; clear boundary. 5YR 3/2; sandy loam; medium, fine sub-angular Bw2 36-56 blocky; common fine roots; clear boundary. 7.5YR 3/2; sandy loam; moderate, fine sub- Bw3 56-64 angular blocky; few fine roots; abrupt, wavy boundary. 5YR 3/2; pebbly loamy sand; weak, fine, sub- 2BC 64-80 angular blocky; very few fine roots; abrupt, irregular, broken boundary. 7.5YR 3/4; loam; medium, fine, sub-angular 3Bw 80-97 blocky; few fine roots; clear boundary. Dark tephra, coarse to very coarse ash; abrupt 4C 97-111 boundary 7.5YR 3/4; silt loam; moderate, fine sub-angular 5Bw 111-131+ blocky; no roots; common large stones at top of layer.

Figure 19: Profile PK01 with horizon designators

Figure 20: Profile PK02 with horizon designators

Figure 21: Profile PK03 with horizon designators

Figure 22: Profile PK04 with horizon designators

Figure 23: Profile PK05 with horizon designators

Figure 24: Profile PK06 with horizon designators

Figure 25: Profile PK07 with horizon designators

Figure 26: Profile PK08 with horizon designators

3.4.2 Soil reaction

The soils are all slightly acid to neutral with lower pH in the surface horizons than at depth. A trend towards lower pH at a greater depth at Stóri Klofi is most likely due to the higher carbon content with higher age and the secretion of organic acids by the vegetation. This rapid decrease in pH in the surface horizon when Vitrisols (desert Andisols) are vegetated is well documented in

Iceland (e.g. Arnalds and Kimble, 2004). The pH values in the surface horizons are higher than commonly reported for Andisols (Shoji et al., 1994;

Dahlgren et al., 2004) but within the range for Icelandic soils (Arnalds, 2004).

The higher pH values in surface horizons are attributed to the steady influx of aeolian materials, mostly of basaltic origin (Jóhannesson, 1960; Arnalds,

1995).

Potassium chloride pH (table 5) is generally more than 1 unit lower than pH in H2O, which indicates that the soil colloids have a net negative charge that is pH dependent. Notable exceptions are the surface horizons in

PK02 with delta pH of 0.3-0.5.is attributed to the steady influx of aeolian materials, mostly of basaltic origin (Jóhannesson, 1960; Arnalds, 1995).

3.4.3 Bulk density

Bulk density in the more developed surface horizons (excluding C horizons) is significantly lower in the 50-60 year old birch profiles (P – value = 0.0023) than in the overlying younger soil materials. As bulk density and mineralogy are closely correlated in Andisols (Dahlgren et al., 2004) this is explained by

higher carbon content (table 5) and more abundance of short-range minerals in the older soil horizons. The C-horizons generally have a bulk density around 1 g cm-3, which is similar to what is reported for Icelandic Vitrisols

(Arnalds and Kimble, 2001; Arnalds, 2004). The lower bulk density in the 2C horizon in PK04 can be explained by the high concentrations of porous and lightweight, rhyolitic pumice grains.

3.4.4 Carbon and nitrogen

The carbon and nitrogen content in the surface horizons is roughly similar in all profiles with differences probably reflecting the difference in vegetation with higher carbon content were moss is dominant. There is a trend towards a greater carbon concentration to a greater depth with age (fig 27 and fig 28).

The carbon in the profiles below ~15 cm is most likely carbon sequestered from roots, as it sits below the aeolian influenced surface horizons and vertical movement of carbon is limited in volcanic soils (Ugolini and Dahlgren,

2002). The levels of carbon in the developing soil profiles are lower in the subsurface horizons than in the old-growth. This does not just reflect the greater time for sequestration, but also the fact that due to constant aeolian deposition, soil surfaces become gradually buried changing to B-horizons.

The carbon, as stated earlier, does not migrate significantly, and due to the recalcitrant properties, is not readily decomposed by microbes. The increase in carbon at about 20 cm (fig 26 and 27) with age is consistent with the results presented in chapter 2 in this thesis.

Figure 27: Depth distribution of carbon. Note the irregular pattern in oldgrowth profiles (PK07 and PK08)

Figure 28: Carbon concentrations.

Table 5: Selective dissolution and specific surface

pH Organic C Organic N -3 -3 Horizon Bd H2O KCl !pH %C kg m %N kg m C:N g cm-3 PK01 A 0.23 5.92 5.07 0.85 8.84 20.33 0.45 1.04 19.54 A2 0.36 6.21 5.08 1.13 3.49 12.41 0.18 0.63 19.58 Bw 0.95 6.26 4.8 1.46 1.74 16.55 0.11 1.08 15.36 2C 1.21 6.49 4.73 1.76 0.35 4.29 0.02 0.21 20.19 2C 0.87 6.58 4.74 1.84 0.71 6.21 0.05 0.43 14.55 3Bw 0.69 6.77 4.72 2.05 0.42 2.92 0.05 0.36 8.10 4C 0.78 6.75 4.74 2.01 0.14 1.11 0.03 0.20 5.50 5Bw 0.64 6.86 4.91 1.95 0.42 2.67 0.03 0.20 13.15 PK02 Oi 0.30 5.79 5.08 0.30 11.05 32.98 0.58 0.033 19.04 A 0.52 6.49 5.11 0.52 4.66 23.99 0.28 0.024 16.35 Bw 0.96 6.72 4.72 0.96 0.91 8.72 0.06 0.009 14.94 2C 1.14 6.89 4.84 1.14 0.19 2.12 0.02 0.002 10.36 3Bwg 0.42 6.77 5.11 0.42 1.79 7.59 0.13 0.008 13.56 PK03 A 0.55 6.02 4.84 1.18 8.04 44.14 0.28 1.51 29.14 A2 0.55 6.35 5.08 1.27 4.81 26.44 0.22 1.20 22.10 Bw 1.04 6.63 5.11 1.52 0.53 5.58 0.02 0.24 22.92 2C 1.02 6.92 5.11 1.81 0.13 1.37 0.01 0.11 12.42 PK04 Oi 0.95 5.81 4.65 1.16 5.57 53.02 0.21 2.03 26.08 A 1.00 6.12 4.89 1.23 4.08 40.77 0.15 1.46 27.99 Bw 0.59 6.78 5.17 1.61 0.56 3.31 0.03 0.17 19.84 2C 0.59 6.67 5.34 1.33 0.11 0.64 0.01 0.07 9.38 2C2 1.11 6.84 5.29 1.55 0.07 0.73 0.01 0.09 8.15 PK05 A 0.49 6.36 5.14 1.22 4.46 21.75 0.22 1.08 20.10 AC 0.65 6.76 5.08 1.68 0.36 2.37 0.02 0.15 15.63 2C 0.99 6.9 4.99 1.91 0.35 3.42 0.02 0.24 14.27 2C2 1.22 6.94 4.91 2.03 0.13 1.58 0.01 0.18 8.93 3Bw 0.57 6.81 5.07 1.74 1.32 7.49 0.11 0.60 12.55 PK06 A 0.47 6.08 4.92 1.16 6.17 29.00 0.33 1.57 18.48 A2 0.83 6.37 5.03 1.34 1.90 15.70 0.11 0.89 17.55 Bw 0.91 6.77 5.17 1.6 0.60 5.44 0.04 0.36 15.09 2C 1.15 7.01 5.26 1.75 0.34 3.88 0.02 0.25 15.24 2C3 0.79 6.99 5.29 1.7 0.72 5.68 0.05 0.42 13.47 3Bw 0.51 6.96 5.39 1.57 2.57 13.05 0.20 1.00 13.10 PK07 Oi 0.37 6.99 4.9 2.09 12.39 46.22 0.54 0.046 23.05 A 0.66 5.85 5.04 0.81 4.89 32.16 0.28 0.032 17.76 Bw 0.66 6.27 5.35 0.92 1.51 9.96 0.08 0.010 18.30 Bw2 0.79 6.82 5.44 1.38 1.74 13.69 0.11 0.014 16.28 Bw3 0.87 7.01 5.51 1.5 1.15 10.03 0.07 0.010 15.96 2C 0.72 7.18 5.55 1.63 0.63 4.55 0.03 0.005 20.13 3Bw 0.60 7.26 5.65 1.61 1.29 7.75 0.07 0.008 19.22 PK08 A 0.64 6.45 5.16 1.29 4.27 27.14 0.23 1.48 18.36 AB 0.64 6.68 5.2 1.48 2.18 13.93 0.13 0.81 17.23 Bw 0.73 6.85 5.19 1.66 1.77 12.98 0.11 0.81 15.97 Bw2 0.85 6.92 5.24 1.68 1.65 14.05 0.11 0.93 15.03 Bw3 0.83 6.96 5.24 1.72 1.29 10.79 0.07 0.57 18.91 2BC 0.54 7.34 5.31 2.03 0.45 2.38 0.02 0.11 22.60 3Bw 0.47 7.17 5.49 1.68 1.73 8.15 0.10 0.46 17.85 51

3.4.5 Oxalate extractable clays

Quantities of oxalate extractable aluminum, iron and silica (Alo, Feo and Sio) are significantly lower (p-value <0.001) in the Stóri Klofi profiles compared to the oldgrowth (table 6). The quantities of Alo, Feo and Sio in the Hraunteigur profiles is in turn low compared to ‘stable’ Andisols in Iceland, (Arnalds 2004;

Arnalds et al., 1995) probably due to the significant aeolian influx in the area.

The Al:Si values are around 1.

There are no significant differences between the allophane contents between profiles when comparing ages, but if allophane and ferrihydrite are looked at together, a trend can be observed. The surface horizons all show similar amounts of short-range minerals, which indicates that the aeolian deposition equals out the age differences. In fact, it can be said that the surface is in many ways ‘younger’ than the subsurface horizons and of similar

‘age’ in all the profiles. However, a difference can be observed in the subsurface horizons (fig 29 and 30) with higher concentrations of poorly crystalline minerals to a greater depth with age. A distribution difference can even be seen between the 5-15 year old forest and the unforested profiles.

This is possibly an effect of more extensive root system of the birch increasing hydrolysis by chelation/organic acid secretation thus accelerating mineral formation. The low number in the A horizon in PK02 can be attributed to a measurement error in the preparation of the sample for oxalate extraction.

Figure 29: Combined depth distribution of poorly crystalline minerals

Figure 30: Combined depth distribution of poorly crystalline minerals at Stóri Klofi excluding the buried palaeosol

Table 6: Selective dissolution and specific surface Oxalate extractable Horizon Al Fe Si Mn Al:Si Allophane Ferrihydrite Specific surface mg g-1 PK01 mg g-1 m2 g-1 A 2.86 5.09 2.62 0.14 1.09 13.11 8.66 4.79 A2 3.27 6.07 3.23 0.14 1.01 16.17 10.33 6.07 Bw 3.48 6.16 3.55 0.11 0.98 17.73 10.46 8.22 2C 2.82 4.39 3.38 0.04 0.84 16.88 7.46 9.27 C 4.37 6.15 4.34 0.05 1.01 21.72 10.46 22.42 3Bw 9.05 16.32 8.72 0.09 1.04 43.62 27.75 79.67 4C 5.40 10.13 4.77 0.14 1.13 23.85 17.22 29.50 5Bw 4.45 5.60 4.12 0.17 1.08 20.58 9.52 48.24 PK02 Oi 2.42 4.52 2.33 0.15 1.04 11.65 7.68 3.72 A 2.38 4.37 2.24 0.11 1.06 11.20 7.43 7.46 Bw 2.95 5.35 3.51 0.11 0.84 17.55 9.09 10.47 2C 1.80 3.36 2.31 0.06 0.78 11.57 5.71 8.63 3Bwg 11.14 21.42 15.02 0.47 0.74 75.09 36.42 86.98 PK03 A 2.87 5.15 2.75 0.14 1.04 13.75 8.76 13.88 A2 2.82 5.20 2.85 0.16 0.99 14.23 8.83 6.70 Bw 2.75 4.90 3.25 0.09 0.85 16.24 8.33 11.70 2C 1.23 2.50 1.62 0.05 0.76 8.09 4.26 6.70 PK04 Oi 2.26 4.25 2.31 0.10 0.98 11.56 7.22 5.83 A 2.99 5.49 3.09 0.13 0.97 15.47 9.33 6.91 Bw 2.82 5.16 3.50 0.09 0.80 17.51 8.77 11.78 2C 1.45 2.96 2.00 0.06 0.72 9.99 5.03 4.70 2C2 1.09 2.54 1.65 0.04 0.67 8.23 4.31 4.79 PK05 A 2.36 4.15 2.51 0.10 0.94 12.53 7.06 8.13 AC 2.19 4.22 2.70 0.09 0.81 13.51 7.18 9.97 2C 2.46 4.44 3.05 0.10 0.81 15.24 7.54 11.40 2C2 1.63 3.41 2.25 0.06 0.72 11.26 5.79 5.37 3Bw 4.97 8.67 5.87 0.22 0.85 29.37 14.74 51.87 PK06 A 2.58 4.75 2.65 0.10 0.97 13.25 8.07 5.13 A2 2.66 4.98 2.96 0.10 0.90 14.82 8.47 3.88 Bw 2.62 4.67 3.12 0.09 0.84 15.58 7.94 5.89 2C 2.21 3.91 2.82 0.07 0.78 14.10 6.64 8.58 2C3 4.20 7.08 5.07 0.15 0.83 25.34 12.03 14.77 3Bw 11.17 18.95 13.84 0.44 0.81 69.20 32.21 60.63 PK07 Oi 1.55 2.33 1.10 0.08 1.41 5.51 3.97 3.88 A 3.70 6.73 3.75 0.13 0.99 18.73 11.45 5.56 Bw 6.60 11.24 8.03 0.20 0.82 40.14 19.11 21.78 Bw2 6.43 11.35 7.91 0.23 0.81 39.57 19.29 21.33 Bw3 7.17 12.56 9.06 0.27 0.79 45.29 21.35 29.35 2C 6.55 11.51 8.22 0.23 0.80 41.12 19.56 23.57 3Bw 7.45 12.10 9.71 0.24 0.77 48.54 20.57 45.89 PK08 A 3.99 7.14 3.88 0.16 1.03 19.42 12.13 5.60 AB 7.07 12.25 7.96 0.22 0.89 39.80 20.83 21.89 Bw 7.98 13.41 9.63 0.27 0.83 48.15 22.79 28.24 Bw2 5.97 10.53 7.23 0.20 0.82 36.16 17.89 20.57 Bw3 7.34 12.80 9.13 0.26 0.80 45.66 21.77 29.24 2BC 5.41 10.01 7.04 0.21 0.77 35.20 17.02 NE 3Bw 11.39 15.84 13.49 0.38 0.84 67.43 26.93 NE 54

3.4.6 Specific surface and poorly crystalline minerals

The large specific surface area of allophanes, 435 – 534 m2 g-1 (Wada, 1989), shows that relatively low amounts can change the nutrient and pollutant holding capacities of the soil. There is a moderate correlation between surface area and allophanes (fig 31) (r2 = 0.624). The other big contributors are ferrihydrite, specific surface 3-5 m2 g-1 (Schwertman and Taylor, 1989), and organic matter. Not all the specific surface can be attributed to these components, as the morphology of the sand and silt sized fractions can also play a significant role. Pumice is highly porous with low specific weight and counting all the vesicles, can have a significant specific surface but limited sorption capacities (Esposito and Guadagno, 1998).

The depth distribution of ferrihydrite and allophane calculated from oxalate extractable Al, Si and Fe follows the same pattern as specific surface area. This shows the close connection between these factors that in turn control nutrient holding capacities, phosphorous and fluoride retention and how recalcitrant the organic matter is. The similarities in the pattern also indicate that the assumptions behind the mineral calculations using selective dissolution hold, and interference is not significant.

Figure 31: Simple regression analysis on the relationship between the allophane contents and specific surface area. Surface area = -4.545 + (0.956 * mg g soil-1 allophane), r2 = 0.624, p-value <0.0001

Figure 32: Simple regression analysis on the relationship between the total clay content and specific surface area. Surface area = -5.962 + (0.668 * mg g soil-1 clay), r2 = 0.650, p- value <0.0001

3.4.7 Change with age

When comparing the profiles visually using figures 33-36, the bulge at ~20 cm is prominent in PK01 and PK02 (65 year old forest profiles), less prominent in

PK03 and PK04 (5-15 year old trees) and at a shallower depth and almost non-existent in PK05 and PK06 (unforested). This, as discussed above, indicates the change in concentrations and depth distribution of poorly crystalline minerals and, as a result, in specific surface area. The high values at the bottom of PK01, PK02, PK05 and PK06 are from the buried palaeosol and correlate well with the hand texturing in the field.

58 ! !

! ! !

Figure 33: Depth distribution of oxalate extractable poorly crystalline minerals and surface area in profiles PK01 and PK02

59 ! ! !

! ! !

Figure 34: Depth distribution of oxalate extractable poorly crystalline minerals and surface area in profiles PK03 and PK04

60 ! ! !

! ! !

Figure 35: Depth distribution of oxalate extractable poorly crystalline minerals and surface area in profiles PK05 and PK06

61 ! ! !

! ! !

Figure 36: Depth distribution of oxalate extractable poorly crystalline minerals and surface area in profiles PK07 and PK08

3.5 Conclusions

1. Soil development is a process that is usually (i.e. in most soils)

measured in centuries and millennia. Andisols, however, exhibit more

rapid development than most other soil orders due to unique

chemical, physical and biological properties that promote relatively

rapid pedogenesis (Shoji et al., 1994; Dahlgren et al., 2004). Yet,

most studies of pedogenesis in Andisols, including chronosequence

studies are based on time frames of hundreds of years at least. By

contrast, this study concentrates on the initial stages of pedogenesis

to measure significant soil change within a century. Soil properties

that vary so quickly (i.e. within the “human” time scale) have recently

been termed “dynamic”, indicating that they are properties

susceptible to human management. In the case of this study,

significant changes that fundamentally affect the morphology,

mineralogy and physical and chemical properties of the soils have

been measured over a time scale of decades.

2. The profiles at Stóri Klofi are multi-genetic, with aeolian additions

dominating the uppermost horizons and fluvial/single tephra events

the lower.

3. There are significant changes in the depth distribution of carbon with

age. The older the trees, the higher the carbon concentration at

depth. This is attributed to root carbon being sequestered in situ.

4. The rapid aeolian accumulation in the area masks the difference in

concentration of poorly crystalline minerals between age groups. The

upper horizons are clearly dominated by aeolian inputs. Carbon is

more responsive over the short time period so stronger trends of

depth distribution with age can be seen.

5. Overall, significant change in the morphological and chemical

properties of the soil can be seen over the short time period of 65

years. The change can be attributed to the trees capturing particles

suspended in the air resulting in soil thickening ‘from the bottom up’.

The roots of the trees also accelerate soil development in the non-

aeolian part of the soil by releasing carbon and possibly by increasing

hydrolysis by chelation/organic acid secretion.

4 References

Anonymous. 2001. Biological diversity in Iceland: national report to the

convention on biological diversity. Ministry of the Environment,

Icelandic Institute of Natural History.

Amundson, R. 2001. The carbon budget in soils. Annual Review of Earth

and Planetary Sciences 29: 535-562.

Aradottir, AL. 2007. Restoration of birch and willow woodland on eroded

areas. Effects of afforestation on ecosystems, landscape and rural

development. 408: 67-74

Aradottir, AL, Arnalds, O, Archer, S. 1992. A model for land degradation.

Yearbook of the Soil Conservation Service. 1 13-31. (In Icelandic).

Aradottir ÁL, Svavarsdóttir K, Jónsson !H and Gu"bergsson, G. 2000.

Carbon Accumulation in vegetation and soils by reclamation of

degraded areas. Icel. Agr. Sci. 13, 99-113.

Aradottir ÁL, !orsteinsson I, Sigur"sson, S. 2001. Distribution and

characteristics of birch woodlands in North-Iceland in F. E.

Wielgolaski, editor. Nordic Mountain Birch Ecosystems. CRC Press -

Parthenon Publishing 51-61.

Arnalds O. 2000. The Icelandic ‘rofabar"’ soil erosion features. Earth

Surface Proc. 25: 17-18.

Arnalds O. 2004. Volcanic soils of Iceland. Catena 26:3-20.

Arnalds O. 2008. Soils of Iceland, Jökull 58:409-421.

Arnalds, O. 2010. Dust sources and deposition of aeolian materials in

Iceland. Icel. Agric. Sci. 23: 3-21.

Arnalds O, Hallmark CT, Wilding LP. 1995. Andisols from four different

regions of Iceland. Soil Society of America Journal. 58, 161-169.

Arnalds O, Gísladóttir FO, Sigurjónsson H. 2000. Sandy deserts of Iceland:

an overview. Journal of arid environments. 47 459-371.

Arnalds O, Gudbergsson G, Gudmundsson J. 2000. Carbon sequestration

and reclamation of severely degraded soils in Iceland. Icel. Agr. Sci. 65

13:87–97.

Arnalds O and Kimble J. 2001. Andisols of deserts in Iceland. Soil Society

of America Journal 65, 1778-1786.

Arnalds O and Óskarsson H. 2009. Íslenskt jar"vegskort (Icelandic soil

map). Náttúrufræ"ingurinn 78 107-121. (In Icelandic with English

summary).

Batjes, NH. 1996. Total carbon and nitrogen in the soils of the world.

European Journal of Soil Science 47:151 – 163.

Bjarnason H. 1974. Comments on the history of Icelanders in relation to

woodland destruction. Yearb. Icel. For. Soc. 30-43. (In Icelandic with

English summary).

Blakemore LC, Searle, PL, Daly BK. 1987. Methods for Chemical Analysis

of Soils. New Zealand Soil Bureau Scientific Report 80. 103 p.

Brady NC and Weil RR. 2004. Elements of the nature and properties of

soils. Pearson Education Inc., Upper Saddle River.

Churchman GJ, Tate KR. 1987. Stability of aggregates of different size

grades in allophanic soils from volcanic ash in New Zealand. J. Soil

Sci. 38: 19-27.

Dahlgren, R A, Saigusa M, Ugolini F C. 2004. The nature, properties and

management of volcanic soils. Advances in Agronomy, 82. 182 pp.

Dugmore, AJ, Church MJ, Buckland, PC, Edwards KJ, Lawson I,

McGovern, TH, Panagiotakopulu E, Simpson IA, Skidmore P,

Sveinbjarnardóttir G. 2005. The Norse landnám on the North Atlantic

islands: an environmental impact assessment. Polar record 41:21-37.

Dugmore, AJ, Gisladottir G, Simpson, IA, Newton A. 2009. Conceptual

models of 1200 years of Icelandic soil erosion reconstructed using

tephrochronology. Journal of the North Atlantic. 2:1-18.

Einarsson MÁ. 1976. Climate of Iceland (In Icelandic with English

summary). I"unn, Reykjavík, Iceland. 55 pp.

FAO-Unesco. 1998. Soil map of the World. Revised Legend. Soil

Resources Reports 60, FAO, Rome.

Fri"riksson S and Sigur"sson FH. 1983. The effect of air temperature on 67

grass growth (In Icelandic with English summary). J. Agr. Res. Icel.

15 :1–2: 41–54.

Gíslason SR. 2008. Weathering in Iceland. Jökull 58: 387-408.

Gobat JM, Aragno M, Matthey W. 2004. The Living Soil – Fundamentals of

soil science and soil biology. Science Publishers, Inc., Enfield.

Gudmundsson T. Bjornsson H., Thorvaldsson, G., 2004. Organic

accumulation and pH in an Andic Gleysol under long-term fertilizer

experiment in Iceland. Catena 56, 213-224.

Halldórsson G, Snorrason A, Aradottir A, Sigurdsson B, Oddsdóttir E,

Eggertsson O, Kolka P and Arnalds O. 2009. Kolbjörk endurheimting

birkivistkerfa og kolefnisbinding. Fræ!a"ing Landbúna!arins 6. (In

Icelandic).

Hallsdóttir M. 1995. On the pre-settlement history of Icelandic vegetation.

Búvísindi 9:17-29.

Hjartarson A. 1995. Á Heklusló"um (The Hekla area). Árbók Fer"afélags

Íslands, 1-236. (In Icelandic).

Hynynen J, Niemistö P, Viherä-Aarnio A, Brunner A, Hein S, Velling P.

2009. Silviculture of birch (Betula pendula Roth and Betula

pubescens Ehrh.) in northern Europe. Forestry 2010 83: 103-119.

IPCC, 2000. Land Use, Land-use Change, and Forestry. A special Report

of the IPCC (eds R.T. Watson, IR Noble, B Bolin, NH Ravindranath,

DJ Verardo, DJ Dokken). Cambridge University Press, Cambridge:

377 pp.

Jobbágy E and Jackson RB. 2000. The vertical distribution of soil organic

carbon and its relation to climate and vegetation. Ecological

Applications 10: 423–436.

Jóhannesson B. 1960. Íslenzkur jar"vegur (Icelandic Soil).

Menningarsjó"ur, Reykjavík, Iceland. 132 pp.

Jonsson, TH. 2004. Stature of Sub-Arctic birch in relation to growth rate,

lifespan and tree forms. Annals of Botany, 94(5). 753-762.

Lawrence CR and Neff JC. 2009. The contemporary physical and chemical

flux of aeolian dust: A synthesis of direct measurments of dust

deposition Chemical Geology 267 46-63.

LMI. 1993. Digital vegetation index map of Iceland. National Land Survey of

Iceland, Akranes, Iceland.

Olafsdottir, R, Schlyter P, Haraldsson, HV. 2001. Simulating Icelandic

vegetation cover during the Holocene: Implications for long-term land

degradation. Geografiska Annaler 83 A:203-215.

Orradóttir, B, Archer SR, Arnalds O, Wilding LP, Thurow TI. 2008.

Infiltration in Icelandic Andisols: The role of vegetation and soil frost.

Arctic, Antarctic and Alpine Research. 40(02):412-421.

Óskarsson H, Arnalds O, Gudmundsson J, Gudbergsson G. 2004. Organic

carbon in Icelandic Andosols: geographical variation and impact of

erosion. Catena 56:225–238.

Parfitt RL. 1990. Allophane in New Zealand – A review. Aust. J. Soil 28:343-

360.

Parfitt RL and Kimble JM. 1989. Conditions for formation of allophane in

soils. Soil Sci. Soc. Am. J. 53:971-977.

Ritter R. 2007. Carbon, nitrogen and phosphorus in volcanic soils following

afforestation with native birch (Betula pubescens) and introduced

larch (Larix sibirica) in Iceland. Plant and Soil 295 (1-2): 239-251.

Schoeneberger PJ, Wysocki DA, Benham EC, Broderson WD (editors).

2002. Field book for describing and sampling soils, Version 2.0.

Natural Resources Conservation Service, National Soil Survey

Center, Lincoln, NE.

Schwertman U and Taylor RM. 1989. Iron oxides. In: Minerals in Soil

environment. Soil Sci. Soc. of America, Madison, Wisconsin, 379-438

Shoji S, Nanzyo M, Dahlgren RA. 1993. Volcanic ash soils: Genesis,

properties and utilization. Developments in soil science 21. Elsevier.

Netherlands.

Snorrason A, Sigurdsson BD, Gudbergsson G, Svavarsdóttir K, Jonsson

TH. 2002. Carbon sequestration in forest plantations in Iceland.

Icelandic Agricultural Sciences 15, 81-93.

Stumm W. 1992. “Chemistry of the soil-water interface”, Wiley, New York.

428 pp.

Sæmundsson K. 1979. Outline of the geology of Iceland. Jökull 29:7-28.

Thordarson T and Larsen G. 2007. Volcanism in Iceland in historical time:

Volcano types, eruption styles and eruptive history. Journal of

Geodynamics 43:118-152.

Thorsson AT, Palsson SP, Sigurgeirsson A, Anamthawat-Jonsson K. 2007.

Morphological variation among Betula nana (diploid), B pubescens

(tetraploid) and their triploid hybrids in Iceland. Annals of Botany, 99(6),

1183-1193.

Ugolini FC, Dahlgren RA, Shoji S, Ito T. 1988. An example of andosolization

and podzolization as revealed by soil solution studes, S. Hakkoda,

N.E. Japan. Soil Science 145:111-12.

Ugolini, FC and Dahlgren, RA. 2002. Soil development in volcanic ash.

Glob. Environ. Res. 6:69-81.

Ulery AL, Drees LR (editors). 2008. Methods of Soil Analysis Part 5 –

Mineralogical Methods. Soil Sci. Soc. Am. Madison Wisconsins. 521 pp.

Wada K. 1989. Allophane and Imogolite. Minerals in Soil Environment. Soil

Sci. Soc. of America, Madison, Wisconsis. 1051-1987. 72

Wada K, Arnalds O, Wilding LP, Hallmark, C. T. 1992. Clay minerals in four

soils formed in eolian and tephra materials in Iceland. Geoderma 52

351-165.