Species-specific temporal variation in photosynthesis as a ...

Biogeosciences 14 257ndash269 2017wwwbiogeosciencesnet142572017doi105194bg-14-257-2017copy Author(s) 2017 CC Attribution 30 License

Species-specific temporal variation in photosynthesis as a moderatorof peatland carbon sequestrationAino Korrensalo1 Pavel Alekseychik2 Tomaacuteš Haacutejek3 Janne Rinne4 Timo Vesala25 Lauri Mehtaumltalo6Ivan Mammarella2 and Eeva-Stiina Tuittila1

1School of Forest Sciences University of Eastern Finland Joensuu Finland2Dept of Physics University of Helsinki Helsinki Finland3Faculty of Science University of South Bohemia Ceskeacute Budejovice Czech Republic4Dept of Physical Geography and Ecosystem Science Lund University Lund Sweden5Dept of Forest Sciences University of Helsinki Helsinki Finland6School of Computing University of Eastern Finland Joensuu Finland

Correspondence to Aino Korrensalo (ainokorrensaloueffi)

Received 20 June 2016 ndash Published in Biogeosciences Discuss 13 July 2016Revised 9 December 2016 ndash Accepted 19 December 2016 ndash Published 18 January 2017

Abstract In boreal bogs plant species are low in numberbut they differ greatly in their growth forms and photosyn-thetic properties We assessed how ecosystem carbon (C)sink dynamics were affected by seasonal variations in thephotosynthetic rate and leaf area of different species Pho-tosynthetic properties (light response parameters) leaf areadevelopment and areal cover (abundance) of the species wereused to quantify species-specific net and gross photosynthe-sis rates (PN and PG respectively) which were summed toexpress ecosystem-level PN and PG The ecosystem-level PGwas compared with a gross primary production (GPP) esti-mate derived from eddy covariance (EC) measurements

Species areal cover rather than differences in photosyn-thetic properties determined the species with the highestPG of both vascular plants and Sphagna Species-specificcontributions to the ecosystem PG varied over the growingseason which in turn determined the seasonal variation inecosystem PG The upscaled growing season PG estimate230 g C mminus2 agreed well with the GPP estimated by the EC(243 g C mminus2)

Sphagna were superior to vascular plants in ecosystem-level PG throughout the growing season but had a lower PNPN results indicated that areal cover of the species togetherwith their differences in photosynthetic parameters shape theecosystem-level C balance Species with low areal cover buthigh photosynthetic efficiency appear to be potentially im-portant for the ecosystem C sink Results imply that func-

tional diversity ie the presence of plant groups with dif-ferent seasonal timing and efficiency of photosynthesis mayincrease the stability of C sinks of boreal bogs

1 Introduction

Northern peatlands are a globally important carbon (C) sinkand storage of approximately 500 Gt C (Gorham 1991 Yu2012) as a result of an imbalance between photosynthesis anddecomposition Boreal bogs are peatland ecosystems wherephotosynthetic productivity is limited by midsummer dry pe-riods light induced stress and in particular low nutrientavailability (Frolking et al 1998 Moore et al 2002) Dueto the low rate of photosynthesis the annual C sink of borealbogs is weak and sensitive to changes even a small changein the environmental conditions that regulate the C cycle canturn the ecosystem into a C source (Waddington and Roulet2000 Lund et al 2012) The rate by which CO2 enters theecosystem through photosynthesis of all of the individualplants together is the definition of gross primary production(GPP) When ecosystem respiration (Reco) is subtracted fromGPP the result is net ecosystem exchange (NEE) of CO2 be-tween the ecosystem and the atmosphere On the scale of in-dividual plants the same processes are called gross photo-synthesis (PG) plant respiration (R) and net photosynthesis(PN) (Chapin et al 2011)

Published by Copernicus Publications on behalf of the European Geosciences Union

258 A Korrensalo et al Species-specific temporal variation in photosynthesis

Boreal bogs are ecosystems with low species diversity buta high diversity of growth forms due to the large microtopo-graphical variation and associated diversity of habitats alongthe water table gradient (Turetsky et al 2012 Rydin andJeglum 2013) Several studies (eg Weltzin et al 2000Moore et al 2002 Leppaumllauml et al 2008) have reported thatpatterned bogs produce more biomass and have less varia-tion in gross photosynthesis over the growing season thanfens which receive additional nutrients from the surroundingmineral soil and generally have more homogenous sedge-dominated vegetation (Weltzin et al 2000) Experimentalstudies have shown that bog plant growth forms have dif-ferential responses to warming and water table level manip-ulation which can help to maintain the level of total ecosys-tem productivity under changing environmental conditions(Weltzin et al 2000 Breeuwer et al 2009) Short-term plantremoval experiments have shown the differential roles ofplant functional types for the peatland NEE and GPP (Wardet al 2009 Kuiper et al 2014) Photosynthetic properties ofbog plants are known to differ widely between species of thesame functional type (Small 1972) and between phases ofgrowing season (Korrensalo et al 2016a) So far the role ofspecies-level differences in temporal variation of bog ecosys-tem photosynthesis has not been studied

Here we aim to solve the linkage between the tempo-ral pattern of bog carbon balance and the development ofspecies-specific potential photosynthesis and leaf area Forthis purpose we quantified the contribution of different plantspecies to ecosystem-level photosynthesis over a growingseason As species differ in their photosynthetic propertiesand the properties vary over the growing season we expecttheir importance for the ecosystem carbon sequestration toalso vary over the season To reach our aim we estimatePN and PG for the whole study site based on the monthlyspecies-level light response of photosynthesis and species-specific leaf area development over a growing season Tovalidate the upscaling approach the sum of species-level PGis compared to the GPP derived from eddy covariance (EC)measurements at the study site

2 Methods

21 Study site

The study site (6150179prime N 2410145prime E) is situated in anombrotrophic bog which is a part of the Siikaneva peatlandcomplex in southern Finland located in the southern bo-real vegetation zone The annual temperature sum in the area(base temperature 5 C) is 1318 degree days annual rainfallis 707 mm and the average annual January and July tem-peratures are 42 minus72 and 171 C respectively (30-yearaverages (years 1982ndash2011) from the Juupajoki-Hyytiaumllaumlweather station) The study site has a surface topography typ-ical of raised bogs that varies from open water pools and mud

surfaces to hollows lawns and hummocks An EC flux toweris mounted on a raft in the center of the site

The vegetation is mainly composed of 11 vascular plantand 8 Sphagnum species (Table 1) the abundance of whichvaries markedly along the microtopographical gradient Acontinuous Sphagnum carpet covers the surfaces from hum-mocks to hollows although no Sphagna are present on themud and water surfaces Sphagnum cuspidatum and S ma-jus are dominant in hollows S papillosum S rubellum Sbalticum and S magellanicum in lawns and S fuscum Srubellum and S angustifolium cover the hummocks Vascu-lar plant species composition includes Rhynchospora albaScheuchzeria palustris and Carex limosa vegetation on mudand hollow surfaces dwarf shrubs (Andromeda polifoliaVaccinium oxycoccos) and Eriophorum vaginatum on lawnsurfaces and a shrub layer of Calluna vulgaris Betula nanaEmpetrum nigrum and sedges Eriophorum vaginatum andTrichophorum cespitosum on hummocks

22 Plant level photosynthesis measurements

To quantify the role of plant species in ecosystem-level pho-tosynthesis over the growing season we conducted net pho-tosynthesis (PN) measurements of the 19 most commonspecies at the study site Over the growing season 2013 wemeasured CO2 exchange of three to five samples of eachspecies per month at three light levels with two open flow-through gas exchange measurement devices (GFS-3000Walz Germany and LI-6400 LI-COR USA) Samples werecollected from several locations inside the study area a max-imum of 30 h prior to measurement Sphagna were collectedin small plastic bags and vascular plants in plastic boxeswith an ample amount of roots and peat These were keptmoist until measured Vascular plants were kept in shadedconditions and Sphagna were stored in the dark at 5 C Themoss cuvettes were filled with Sphagnum capitula imitatingthe natural shoot density of each species in the field result-ing in a total number of 6ndash16 capitula inside a cuvette de-pending on the species Before placing the capitula into thecuvettes they were first wetted and then lightly dried of ex-cess water with pulp paper The cuvette was then placed un-der a photosynthetic photon flux density (PPFD) of approx-imately 1000 micromol mminus2 sminus1 to acclimate to light for 20 minThe measured light levels were 2000 25 and 0 micromol mminus2 sminus1

for vascular plants and 2000 35 and 0 micromol mminus2 sminus1 forSphagna Light levels were designed to catch both the lin-ear light-limited beginning and the light-saturated maximumof the light response curve without causing photoinhibitoryreduction of photosynthetic rate (Laine et al 2015) Vas-cular plants were measured with a standard cuvette but forSphagna we used a moss cuvette of our own constructionwith internal dimensions of 3times 2times 1 cm (Haacutejek et al 2009)with a net bottom to allow airflow above and below the sam-ple reducing the aerodynamic resistance After changing thelight level we waited for the PN to reach steady state be-

Biogeosciences 14 257ndash269 2017 wwwbiogeosciencesnet142572017

A Korrensalo et al Species-specific temporal variation in photosynthesis 259

Table 1 Average projection cover of the most common plant species at the study site maximum leaf area index (LAI) values and cumulativeseasonal gross and net photosynthesis (PG PN) of the species measured in this study The cover values are based on a vegetation inventoryconducted in 2013 at the study site within the 30 m radius footprint of the eddy covariance tower (N = 121) LAI values of vascular plantsare seasonal maxima of each species calculated by converting the cover values into LAI using species-wise linear relationships (SupplementTable S2) Sphagnum LAI is assumed to stay similar over the growing season and is simply the coverage expressed as LAI Sphagnum andvascular species nomenclature according to Laine et al (2009) and Haumlmet-Ahti (1998) respectively

Species Cover mean LAI Seasonal PG Seasonal PNplusmnSE () (m2 mminus2) (g C mminus2) (g C mminus2)

Vascular total 242plusmn 19 029 922 573Rhynchospora alba 69plusmn 08 010 233 151Andromeda polifolia 37plusmn 04 003 144 84Calluna vulgaris 28plusmn 09 004 138 91Rubus chamaemorus 25plusmn 06 003 69 45Eriophorum vaginatum 15plusmn 03 002 60 38Vaccinium oxycoccos 12plusmn 02 001 53 32Drosera rotundifolia 11plusmn 01Empetrum nigrum 10plusmn 05 001 23 12Trichophorum cespitosum 10plusmn 05 002 115 67Drosera longifolia 08plusmn 04Scheuchzeria palustris 08plusmn 01 002 51 30Betula nana 04plusmn 02 0004 11 05Carex limosa 04plusmn 01 0005 24 18

Sphagnum total 638plusmn 37 065 1378 199Sphagnum rubellum 183plusmn 26 018 307 08S papillosum 129plusmn 23 013 339 68S fuscum 110plusmn 23 011 271 74S balticum 83plusmn 15 008 157 minus05S cuspidatum 48plusmn 13 005 134 17S majus 47plusmn 12 005 127 28S angustifolium 13plusmn 05 001 36 06S lindbergii 08plusmn 08S magellanicum 03plusmn 01 0003 07 01

Other mosses and lichens

Pleurozium schreberi 08plusmn 05Mylia anomala 02plusmn 01Cladina rangiferina 04plusmn 02

fore recording the CO2 exchange Only PPFD was variedduring the measurements while we kept air temperature atconstant 20 C the flow rate at 600 micromol sminus1 and the CO2concentration in the incoming air at 400 ppm to be able tocompare the seasonal changes in photosynthetic potentialamong species The relative humidity inside the cuvette waskept at 50 for the vascular plants and below 90 for theSphagna The measured PN value of each sample at the threelight levels was expressed per photosynthesizing leaf area(mg CO2 mminus2 (LA) hminus1) which was the leaf area inside thecuvette measured with a scanner for vascular plants and as-sumed to be the cuvette area for Sphagna Two of the speciesnamely Rhynchospora alba and Rubus chamaemorus werenot yet of measurable size in May R alba had already mostlysenesced in September and therefore the two species were

not measured in those months Altogether the data consistedof 720 measurements

23 Net photosynthesis model

To obtain a species-wise flux reconstruction of PN and PGwe fitted a nonlinear mixed-effects model separately for eachcombination of species and month The mixed-effects mod-eling approach allowed us to take into account the variationbetween samples of which each was measured at three lightlevels We used the hyperbolic light saturation curve of PN(Larcher 2003)

wwwbiogeosciencesnet142572017 Biogeosciences 14 257ndash269 2017


PNsi = Rs+Pmaxs PPFDsi

ks+PPFDsi

+ esi (1)

where PNsi is the observed net CO2 exchange(mg CO2 mminus2 (LA) hminus1) and PPFDsi is the photosyn-thetic photon flux density for measurement i of sample sThe three parameters to be estimated are the maximum rateof light-saturated net photosynthesis (Pmaxs ) the PPFD levelwhere half of Pmax was reached (ks) and respiration (Rs)and they were assumed to be constant for each combinationof species and month esi is the normally distributed residualvariance of the model with a mean of zero The normallydistributed random effect of the sample was included in oneto three of the parameters depending on the model Therandom effect structure was selected based on the Akaikeinformation criterion (AIC) values of the alternative modelswith random effects included in a different combination ofparameters The random effects for the sub-models of eachnonlinear model form a vector of random effect with non-zero correlation Parameter values for the 87 PN models arepresented in the Supplement (Table S1) The separate fittingfor each combination of species and months leads to modelswith similar asymptotic unbiasedness as a single model forcomplete data would do However separate models do notquantify the temporal and between-species correlation ofrandom effects and residuals and therefore do not providea sufficiently detailed model for evaluating the predictionerrors of upscaled estimates of net photosynthesis A propermodel for such a purpose would model both temporal andbetween-species covariance of both random effects andresiduals but model fitting procedures for such modelsare not available in the standard statistical software Allmodels were fitted using the function nlme of the R programpackage nlme (Pinheiro and Bates 2000)

Sphagna were not measured in June nor were vascularplants measured in July due to technical failures of the de-vices Therefore light response curves for these 2 monthswere fitted by combining data from the previous and follow-ing month for each species In 5 of the 95 speciesndashmonthcombinations only one sample of the species had an accept-able measurement The parameters for these months wereestimated separately without the mixed model structure (Ta-ble S1)

24 Upscaling

To upscale species-level photosynthesis to the ecosystemlevel the cover of each species was estimated within thestudy site with a systematic vegetation inventory conductedin July 2013 We estimated the relative cover of each plantspecies in 121 plots of 0071 m2 (Table 1) which were ar-ranged in a regular grid in a 30 m radius circle around the ECtower To link net photosynthesis measured per leaf area tospecies cover we converted species cover in the study area

to leaf area index (LAI) using linear relationships betweenthe two (Table S2) Relationships were based on an inventorymade in July 2012 over a 200 m radius circle where speciescover was estimated and then all living aboveground vege-tation was harvested from 65 inventory plots of 0071 m2 forLAI measurements The vascular plant LAI of these sampleswas measured in the laboratory

We monitored LAI development of each vascular plantspecies over the growing season in 18 permanent samplingplots (036 m2) that represented all the vegetation communi-ties (n= 3 in each vegetation community) along the microto-pographical gradient at the study site LAI was estimated ev-ery third week according to the method described by Wilsonet al (2007) Continuous LAI development of each specieswas then estimated by fitting a log-linear response to the ob-servations The shape of the log-linear LAI development wastaken from this fitting and the growing season LAI maxi-mum for each species was taken from the converted aver-age cover (Table 1) Sphagnum leaf area was assumed to stayconstant over the whole growing season and was obtainedusing the average cover from the 2013 inventory

Using the light response curves estimated daily LAIand half-hourly above-canopy PPFD data from SMEARII measurement station (6150845prime N 2417686prime E)we calculated PN and PG for each half hour period(mg CO2 mminus2 30 minminus1) over the growing season (Juliandays 121ndash273) with the species-wise and monthly lightresponse curves PG was calculated with the same modelwithout the R parameter ie assuming that respiration iszero Model predictions were not meaningfully changed byusing marginal prediction ie averaging the predictionsover the distribution random effects (eg de Miguel et al2012) and were therefore computed using the fixed part ofthe model only Growing season PN and PG of the wholestudy site were calculated as a sum of their daily values

25 Ecosystem-level CO2 exchange measurements andestimation of gross primary production

To validate the measured levels of photosynthesis the calcu-lated values were compared with the GPP estimates obtainedby EC measurements which offer an independent estimate ofthe ecosystem-level CO2 exchange measured directly as tur-bulent vertical fluxes (eg Baldocchi 2003 Aubinet et al2012) The EC system comprised a 3-D ultrasonic anemome-ter (USA-1 METEK Meteorologische Messtechnik GmbHGermany) and an enclosed H2OCO2 gas analyzer (LI-7200LI-COR Biosciences USA) The EC sensors were mountedon the mast 25 m above the peat surface EddyUH softwarewas used to process the raw data and produce the 30 minaverage fluxes of latent heat sensible heat and CO2 (Mam-marella et al 2015) Standard EC data checks based on thewidely accepted quality criteria (eg Aubinet et al 2012)were applied partly automatically by the software and partlymanually the EC data at friction velocity (ulowast) less than



01 m sminus1 were rejected The resulting EC fluxes representthe exchange over a quasi-elliptical source area (footprint)located within about 30 m upwind of the EC mast as sug-gested by footprint calculations using the model by Kormannand Meixner (2001)

NEE measured by the EC method was then partitioned intoecosystem GPP and Reco The daytime Reco estimates wereobtained from the Q10-type temperature response curve fit-ted to the nighttime EC data when respiration is the onlycomponent of NEE Nighttime was defined as all the periodswhen the sun elevation angle was lower than 5 Peat tem-perature at 5 cm depth was used as the driver of Reco in thefollowing relationship

Reco = RrefQ

(TpminusTref

10

)10 (2)

where Tp is the peat temperature at a 5 cm depth (C) andTref is the peat reference temperature of 12 C Parameters tobe estimated from the fit of the model (Eq 2) to all availablenighttime NEE data were Rref the reference respiration at thetemperature of 12 C and Q10 the temperature sensitivitycoefficient

The GPP estimates were calculated by subtracting themodeled Reco from the EC-derived NEE values Finally inorder to gap-fill the GPP time series a model using PPFD(from the SMEAR II measurement station) and the footprint-scale LAI was fitted to the data following

GPP=Pmax PPFDk+PPFD

(LAI+ b) (3)

where GPP is expressed in milligrams (CO2) per square me-ter per hour LAI is the modeled daily vascular leaf area indexdescribed above while b represents the temporally constantcontribution of the Sphagnum to total LAI

One of the major sources of random uncertainty in cumu-lative GPP originates from the uncertainty in the Reco andGPP model parameters Random uncertainty was calculatedas the 95 confidence interval of a set of 1000 cumulativeGPP estimates obtained using Rref Q10 Pmax and k variedwithin their respective 95 confidence intervals Since lab-oratory measurements of PG were conducted at a constanttemperature of 20 C and EC measurements at the air tem-peratures present at the field site the temperature limitationof GPP was studied by fitting to the GPP data a model similarto Eq (3) but complemented with a Gaussian-type tempera-ture response (Maanavilja et al 2011)

GPP=Pmax PPFDk+PPFD

( LAI+ b)exp

(minus05 (Taminus Topt)

2

T 2tol

) (4)

where Ta is the air temperature Topt is the temperature op-timum of GPP and Ttol is the temperature tolerance of GPP

(deviation from the optimum at which GPP is 60 of themaximum) Using Eq (4) GPP at 20 C and at ambient PARand LAI was simulated for the study site over the growingseason

3 Results

31 Cumulative growing season gross photosynthesis

Fitting the temperature response curve of Reco Eq (2) intothe nighttime eddy covariance data yielded a reference respi-ration (Rref) of 123 mg (CO2) mminus2 hminus1 and Q10 of 35 In theGPP model (Eq 3) fit Pmax was 17218 mg CO2 mminus2 hminus1k was 1283 micromol mminus2 sminus1 and b was 008 After gap-fillingthe GPP data (Eq 3) the resulting cumulative growing sea-son GPP estimate was 243 g C mminus2 (95 confidence inter-val 220ndash265 g C mminus2) In the GPP model fit complementedwith the temperature response (Eq 4) Pmax was 1852 mg(CO2) mminus2 hminus1 k was 1703 micromol mminus2 sminus1 b was 01 Toptwas 226 C and Ttol was 209 C

Cumulative growing season PG upscaled to the ecosys-tem level using the separate light response curves for speciesand months (Eq 1) was 230 g C mminus2 (Julian days 121ndash273)Daily PG estimates were higher than GPP values from theEC tower in spring and lower in the middle of the summer(Fig 1a) The GPP simulated at 20 C the same tempera-ture as during the laboratory measurements was similar tothe upscaled PG in spring but closer to the measured GPP inthe middle of the summer (Fig 1a) In the autumn all of thethree estimates showed fairly similar levels (Fig 1a)

Sphagna at the study site had a higher cumulative grow-ing season PG value (138 g C mminus2) than vascular plants(92 g C mminus2) Sphagna had higher daily PG than vascularplants in spring and autumn but were almost at the same levelin the middle of the summer (Fig 2) A small increase inSphagnum photosynthesis was observed during May (Figs 2and 3b) due to increments of daily PPFD towards midsum-mer Otherwise Sphagnum PG decreased steadily over thegrowing season (Fig 2) Seasonal changes in vascular PGshowed similar patterns to vascular LAI development al-though the maximum PG was reached slightly earlier in theseason than maximum LAI (Figs 1a c and 2)

The three vascular plant species with the highest PG inthe ecosystem were C vulgaris R alba and A polifoliaA polifolia was the most productive species in May andSeptember R alba in June and July and C vulgaris in Au-gust (Figs 3a and 4a) With 13 cover altogether (Table 1)they formed 22 of the seasonal ecosystem PG and 56 ofthe vascular plant PG (Fig 4) The three Sphagnum specieswith the highest PG at the ecosystem level were S papillo-sum S fuscum and S rubellum (Figs 3b and 4b) As withall of the Sphagnum species their PG per ground area de-creased steadily over the growing season (Figs 2 and 3b)With 42 cover altogether (Table 1) they formed 40 of



120 150 180 210 240 270

05

10

15

20

25

30

35

PG

GP

P (

g C

mminus2

dminus1

)

PG

GPP in ambient T

GPP in 20 degC

(a)

May June July August September

JDMonth

120 150 180 210 240 270

05

1015

2025

30

Air

T (

degC)

(b)

May Jun Jul Aug Sep

JDMonth

120 150 180 210 240 270

000

010

020

030

LAI (

m2 m

minus2 )

(c)

May Jun Jul Aug Sep

120 150 180 210 240 270

minus20

minus15

minus10

minus5

WT

(cm

)

(d)

May Jun Jul Aug Sep

Figure 1 (a) Comparison of total daily ecosystem-level gross photosynthesis estimate of all plants (PG) derived from laboratory measure-ments with GPP estimates derived directly from the eddy covariance measurements and with GPP simulated at constant temperature of20 C The temperature in laboratory photosynthesis measurements was kept constant at 20 C during the whole growing season (b) Dailymean air temperature (in black) and daily temperature variation (in grey fill Hyytiaumllauml Forest Research Station 10 km from the study siteFinnish Meteorological Institute 2016) (c) sum of modeled vascular leaf area during the growing season 2013 and (d) average daily lawn(ie intermediate) surface water table (WT) at the study site Lines represent locally weighted scatterplot smoothing (Loess smoothingparameter= 025) curves

the seasonal ecosystem PG and 67 of the PG of Sphagnummosses (Fig 4)

32 Cumulative growing season net photosynthesis

The aboveground vegetation of the study site was a car-bon sink of 77 g C mminus2 over the growing season as esti-mated by PN values upscaled to the ecosystem level using thespecies- and month-wise light response curves PN results forSphagna and vascular plants were reversed in comparison toPG estimates PN of Sphagna was 20 g C mminus2 and vascularPN was 57 g C mminus2

The same vascular plant species had the highest growingseason PN and PG R alba C vulgaris and A polifolia hadthe highest PN estimates of 151 91 and 84 g C mminus2 re-spectively (Table 1) These three species made up 57 ofthe total vascular PN and 42 of the whole ecosystem-levelPN

S fuscum S papillosum and S majus had the highest sea-sonal PN of Sphagnum species 74 68 and 28 g C mminus2 re-spectively (Table 1) The PN of these three species was 85 of the total Sphagnum PN and 22 of the seasonal ecosys-tem PN Although having one of the highest coverage and PGvalues S rubellum was not among the three most productivespecies in terms of PN



120 150 180 210 240 270

00

05

10

15

PG

(g

C m

minus2 d

minus1 )

Vasculars Sphagna


JD

Month

Figure 2 Daily gross photosynthesis estimates (g C mminus2 dminus1) ofvascular plants and Sphagna upscaled to the ecosystem level us-ing the species-specific monthly light response curves derivedfrom laboratory measurements Lines represent Loess averaging(smoothing parameter= 025)

4 Discussion

41 Comparison of upscaled gross photosynthesisvalues with eddy covariance gross primaryproduction estimates

Accounting for the differences in photosynthetic parametersbetween species and between phases of the growing sea-son appeared to accurately estimate ecosystem PG when up-scaling species-level measurements Sphagnum mosses es-pecially showed a large seasonal variation in their photosyn-thetic light response which could be accounted for in thisupscaling approach The similarity of the PG estimates cal-culated with species-wise and monthly light response curvesand GPP estimates derived from EC measurements (Fig 1a)adds credibility to the methods used and indicates that thephotosynthetic parameters measured under laboratory condi-tions are comparable with field measurements Both methodshave their error sources Annual CO2 flux balances from ECmeasurements are prone to significant systematic bias some-times in excess of 30 but usually between 10 and 30 of the cumulative flux (eg Baldocchi 2003 Rannik et al2006) The underestimation of the EC fluxes implicit in theunclosed energy balance (70 for Siikaneva-1 unpublisheddata) might be partly compensated for by the Kok effectwhich might be more significant than previously thought asindicated by Wehr et al (2016) Our PG estimates include

errors related to the LAI development measurements visualspecies cover estimation the conversion from cover to LAIand the laboratory measurements of photosynthetic param-eters Although the shading of the moss layer by vascularplants may figure as a potential error source of PG upscaledwith PPFD measured above the vegetation it is not likelyto be caused by the sparse cover of vascular plants at thesite (Supplement Fig S3) with low seasonal maximum LAI(Fig 1c) By taking into account the variation between sam-ples in the 87 PN models (Eq 1) we aimed at a more accurateestimation of the light response parameters Neverthelessour ecosystem-level PG estimate may contain bias causedby not accounting for the random effects of the 87 modelsin the upscaling procedure The cumulative growing seasonPG of 230 g C mminus2 is very similar to the 205 g C mminus2 ob-tained by Alm et al (1999) at an ombrotrophic bog site undersimilar climatic conditions and comparable water levels butwhere the exceptionally dry conditions during the measuredseason reduced the photosynthetic capacity of many Sphag-num species Our growing season PG was considerably lowerthan the 500 g C mminus2 obtained by Moore et al (2002) andRoulet et al (2007) at a temperate ombrotrophic bog withmuch lower water table levels While our value only coversthe period from May until September it falls just below thelarge range of annual GPP values (250 to 900 g C mminus2) mea-sured with the EC method from seven northern peatland sites(Lund et al 2010)

The shapes of PG and GPP development differed over thegrowing season especially at the beginning of the summerwhich is largely due to the constant temperature of 20 C inour laboratory measurements (Fig 1a) The constant temper-ature allowed us to investigate how the changes in species-specific photosynthetic parameters were affected by the sea-sonal changes in moisture conditions in the field Since themeasuring of species-specific temperature responses of PGwas unachievable due to the large number of species weinstead chose to model the temperature dependence of EC-derived GPP (Eq 4 Fig 1a) Our upscaled PG values werehigher than GPP in May when vascular plants had a high ca-pacity to use low light levels (low k value) and Sphagna hadhigh Pmax (Fig 1a Table S1) but this was the case whentemperatures in the field remained mostly below 20 C andlimited the measured GPP (Fig 1b) The temperature limita-tion of measured GPP is demonstrated by the lower spring-time measured GPP in comparison with GPP simulated at20 C (Eq 4 Fig 1a) Both measured GPP and GPP simu-lated at 20 C show higher levels than PG in July and Au-gust (Fig 1a) the reason for which remains partly unclearBecause the difference between GPP and PG lasted for 2months the lack of vascular plant PG measurements in Julycan only partly explain this midsummer deviation betweenthe two methods In September when Sphagnum Pmax val-ues and k values of both vascular plants and Sphagna wereat their lowest PG measured GPP and GPP at 20 C wereall similar despite the difference between air and laboratory



120 150 180 210 240 270

00

01

02

03

04

05

PG

(g

C m

minus2 d

minus1 )

AndrPol CallVul RhynAlb

(a)


JD

Month

120 150 180 210 240 270

00

01

02

03

04

05

PG

(g

C m

minus2 d

minus1 )

SphaFus SphaPap SphaRub

(b)


JD

Month

Figure 3 Daily gross photosynthesis estimates (g C mminus2 dminus1) ofthe three most productive (a) vascular and (b) Sphagnum speciesupscaled to the ecosystem level using the species-specific monthlylight response curves derived from laboratory measurementsLines represent Loess averaging (smoothing parameter= 025) Thespecies cover within the study site (EC footprint) is given in Table 1Abbreviations of the speciesrsquo names are AndrPol Andromeda po-lifolia CallVul Calluna vulgaris RhynAlb Rhynchospora albaSphaFusc Sphagnum fuscum SphaPapi Sphagnum papillosumSphaRube Sphagnum rubellum

temperatures (Fig 1a) According to our results peatlandphotosynthesis is temperature limited especially in springSphagna had a high photosynthetic potential due to favorable

moisture conditions at that time (Fig 2) but the low fieldtemperatures limited ecosystem-level GPP (Fig 1a and b)Temperature limitation of springtime photosynthesis is wellknown for boreal forests (Tanja et al 2003 Ensminger et al2004) as well as for bog Sphagna (Moore et al 2006) Meanannual temperature together with PPFD during the growingseason are the most important factors explaining Sphagnumproductivity on the global scale (Gunnarsson 2005 Loiselet al 2012) and the temperature optimum of Sphagnumphotosynthesis is known to change over the growing season(Gaberšcik and Martincic 1987) However the temperaturedependence and acclimatization of species-level photosyn-thesis in peatlands has been studied only with a few Sphag-num species (Gaberšcik and Martincic 1987 Robroek et al2007)

42 The contribution of plant species to ecosystem-levelgross photosynthesis

Among both Sphagna and vascular plants the species withthe highest seasonal upscaled PG (g C per mminus2 of groundarea) ndash and hence the most productive species on the ecosys-tem scale ndash were also the ones with the highest areal coverNo interspecies differences in photosynthetic properties ei-ther within vascular plants or Sphagna could change thisorder On the ecosystem scale Sphagna covering on aver-age 63 of the ground area had higher upscaled daily PGvalues for the whole summer than vascular plants coveringonly 24 despite the lower Pmax values of Sphagna (Ta-ble S1) In ombrotrophic bogs Sphagna are known to be thefirst group to start photosynthesizing in early spring (Mooreet al 2006) which was also evident at our site (Fig 2) Thecombination of low Pmax values in July and September andhigh respiration rates in August and September (Table S2) re-sulted in an almost linear decrease in ecosystem-scale Sphag-num PG over the growing season (Fig 2) The seasonally de-creasing Sphagnum PG is likely to reflect the change in themoisture conditions Water table depth which together withprecipitation is known to be the most important moderatorof Sphagnum photosynthesis (Hayward and Clymo 1983Backeacuteus 1988 Lindholm 1990 Nijp et al 2014) decreasedat the study site over the growing season (Fig 1d)

Despite low Pmax values R alba was among the three vas-cular plant species with highest PG on the ecosystem scaledue to its high cover at the site (Table 1) It also had a verysharp but short-lived LAI and PG peak at the end of June(Fig 3a) which was largely the reason for the peak in vas-cular plant PG (Fig 2) occurring slightly earlier in the sea-son than maximum vascular LAI (Fig 1c) Evergreen shrubshave been observed to be the second group to start photo-synthesizing after Sphagna in spring (Moore et al 2006)Similarly the vascular plants with the highest upscaled PGin May were the evergreen shrubs A polifolia and C vul-garis (Fig 3a) The contributions of different species to to-tal Sphagnum PG did not differ over the growing season


A Korrensalo et al Species-specific temporal variation in photosynthesis 265P

G (

g C

mminus2

mon

thminus1

)

05

1015

2025

3035 AndrPol

CallVulRhynAlbOthers

(a)

PG

(g

C m

minus2 m

onth

minus1 )

05

1015

2025

3035 SphaRub

SphaFusSphaPapOthers

(b)

May June July August SeptemberMonth

PG

(g

C m

minus2 m

onth

minus1 )

010

2030

4050

60

SphagnaVasculars

(c)

Figure 4 Speciesrsquo proportions of monthly ecosystem-scale grossphotosynthesis (PG) of (a) vascular plants (b) Sphagnum mossesand (c) the contribution of those two groups to total monthlyecosystem-scale PG Ecosystem-level PG was calculated using thespecies-specific monthly light response curves derived from labo-ratory measurements For abbreviations of the speciesrsquo names seeFig 3

(Fig 4) Based on these observations phenology and arealcover rather than differences in photosynthetic parametersamong species seem to be the key factor in determining thespecies with the highest PG of a bog ecosystem

43 Ecosystem-level net photosynthesis

The variation in photosynthetic properties changed the rolesof the plant species into a seasonal ecosystem-level carbonsink Although Sphagna had more than twice the cover ofvascular plants seasonal PN was much lower than vascu-lar plants (Table 1) The seasonal PN of Sphagnum specieswas not proportional to their areal cover for example thespecies with the highest cover S rubellum had a small sea-sonal PN (Table 1) S rubellum has earlier found to havelower light-saturated photosynthesis and higher respirationthan most of the other Sphagnum species (Table S1 Kor-rensalo et al 2016a) The differences in photosynthetic pa-rameters of Sphagnum species seem to become much morevisible in ecosystem-level PN than in that of vascular plantspecies since the leaf area of Sphagna stays similar overthe growing season The vascular plants most important forthe ecosystem-level PN were the same as the species withthe greatest cover However T cespitosum with only 1 ofareal cover made up 12 of the seasonal vascular PN

Our results indicate that in addition to the areal coverof the species differences in photosynthetic parameters be-tween species shape the ecosystem-level carbon sink of abog Species with low areal cover may be important for theecosystem carbon sink because of their high photosyntheticefficiency However the PN results have to be interpretedwith care since they contain the R parameter estimated basedon respiration measurements done at 20 C which is higherthan the field temperature for most of the growing seasonThis general overestimation of respiration may be the reasonbehind a slightly negative seasonal PN of S balticum (Ta-ble 1)

44 The role of functional diversity for peatland carbonsink

According to the insurance hypothesis species diversity bothenhances productivity and decreases the temporal varianceof productivity of a plant community (Yachi and Loreau1999) This hypothesis has gained support from testing inseveral ecosystem types especially in grasslands (Hector etal 2010 Cardinale et al 2011 Morin et al 2014) In addi-tion to species and genotype diversity (Hughes et al 2008)the functional diversity ie the presence of species and plantfunctional types with different physiology morphology re-source requirements seasonal growth patterns and life his-tory may increase the productivity of an ecosystem (Tilmanet al 1997 Cadotte et al 2008) Although this study didnot directly test the insurance hypothesis our results also in-dicate that functional diversity especially in regard of differ-ences in phenology and seasonal changes in photosyntheticparameters decreased the temporal variation of ecosystem-level PG and could therefore decrease the variation of theecosystem C sink Vascular plant species of different phenol-ogy had the highest ecosystem-level photosynthesis at dis-



tinct phases of the growing season (Fig 3a) In additionSphagna and evergreen shrubs formed two stable baselines ofecosystem PG which was further increased by the midsum-mer PG peak of the sedge R alba (Fig 3a and b) Especiallyecosystem-level Sphagnum PG was modified by the seasonaldecrease in photosynthetic potential (Table S1) This sug-gests that the growing season PG of our study site is notonly more stable but it is also larger than it would be with amore functionally homogenous assemblage of species Sev-eral studies have suggested that the C sink function of bogsis more stable over the growing season than that of fenswhich have more homogenous and sedge-dominated vegeta-tion (Bubier et al 1998 Leppaumllauml et al 2008) Hence our re-sults should be compared with the patterns of photosyntheticproductivity of a peatland site with a more homogenous plantassemblage

Based on small-scale experimental studies bog speciesand growth forms are known to vary in terms of their con-tribution to ecosystem productivity and to differ in theirresponses to manipulations of environmental conditions(Weltzin et al 2000 Ward et al 2009 Kuiper et al 2014)In this study the laboratory measurements of species photo-synthetic parameters were for the first time upscaled to theecosystem level over a whole growing season and verified bythe comparison with EC measurements Boreal bog specieswere found to differ in the timing of their maximum PG overa growing season (Fig 3a and b) Diversity in species re-sponses to environmental factors is hypothesized to makea plant community more resilient towards changing condi-tions (Yachi and Loreau 1999 Gunderson 2000) In addi-tion to species diversity plant community diversity withinan ecosystem has been shown to increase ecosystem stabil-ity during a severe drought in grasslands (Frank and Mc-Naughton 1991) In boreal bogs Sphagnum mosses createmicrotopographic variations that ndash according to model sim-ulations ndash increase resilience towards environmental pertur-bations both through the diversity of growth forms they sup-port and by variation in physical properties between micro-forms (Turetsky et al 2012) To find out about the effect ofbog spatial heterogeneity on ecosystem resilience studies ex-tending over several growing seasons are needed As demon-strated in an arctic sedge fen the impact of extreme weatherconditions on an ecosystem C sink may occur with a lag ofone growing season (Zona et al 2014) Our study providestools to empirically study the role of species and communitydiversity on the ecosystem scale The combination of labora-tory measurements of photosynthetic parameters phenolog-ical monitoring and EC measurements opens up the possi-bility of long-term and experimental ecosystem-level studieson the effect of functional diversity on the peatland ecosys-tem carbon sink The long-term measurements would permitthe inclusion of a wider range of environmental conditionsIn particular the EC method would allow for a comparisonof the effect of diversity at sites with different plant assem-blages

5 Conclusions

The areal cover of the species determined the species with thehighest gross photosynthesis while phenology in leaf areaand photosynthetic activity drove the variation in ecosystem-level gross photosynthesis In spring potential ecosystem-level gross photosynthesis was much higher than measuredgross primary production which appeared to be due to tem-perature limitation of photosynthesis Ecosystem-level netphotosynthesis was more of a combination of the differencesin (i) photosynthetic parameters which were important inSphagna (ii) phenology which largely defined vascular pro-ductivity and (iii) areal coverage which acted in both vascu-lar plants and Sphagna

The different growth strategies of the plant species presentat our study site appeared to increase the ecosystem-levelphotosynthesis and decrease its variation within a growingseason We are looking forward to the future studies findingout if the diversity of growth forms has the same stabilizingeffect on the interannual variation of ecosystem-level photo-synthesis

6 Data availability

The data associated with the paper are published in the PAN-GAEA repository under doi101594PANGAEA862986(Korrensalo et al 2016b)

The Supplement related to this article is available onlineat doi105194bg-14-257-2017-supplement

Competing interests The authors declare that they have no conflictof interest

Author contributions Eeva-Stiina Tuittila formulated the idea forthis paper Aino Korrensalo Tomaacuteš Haacutejek and Eeva-Stiina Tuit-tila designed the measurements which were done by Aino Korren-salo and Tomaacuteš Haacutejek Aino Korrensalo Tomaacuteš Haacutejek and Eeva-Stiina Tuittila were responsible for the primary photosynthesis dataprocessing Eddy covariance data collection and analysis was doneby Pavel Alekseychik Janne Rinne Timo Vesala and Ivan Mam-marella The mixed-effects models were developed by Lauri Mehtauml-talo and Aino Korrensalo Aino Korrensalo Pavel Alekseychik andEeva-Stiina Tuittila wrote the paper on which all the other authorscommented

Acknowledgement The work presented here is supported bythe Faculty of Science and Forestry at the University of East-ern Finland the Finnish Cultural Foundation the Academy of



Finland (287039 118780 1284701 1282842) ICOS (271878)ICOS-Finland (281255) and the Nordic Centre of Excellence ndashDEFROST We would also like to thank the staff at Hyytiaumllauml ForestResearch Station and Salli Uljas Janne Sormunen Mariacutea Gutier-rez Laura Kettunen and Eva-Stina Kerner for their help withthe measurements and Nicola Kokkonen for revising the Englishlanguage of the paper

Edited by P StoyReviewed by three anonymous referees

References

Alm J Schulman L Walden J Nykaumlnen H Martikainen PJ and Silvola J Carbon balance of a boreal bog during ayear with an exceptionally dry summer Ecology 80 161ndash174doi1018900012-9658(1999)080[0161CBOABB]20CO21999

Aubinet M Vesala T and Papale D Eddy Covariance A Practi-cal Guide to Measurement and Data Analysis Springer Nether-lands 2012

Backeacuteus I Weather variables as predictors of Sphagnum growthon a bog Holarctic Ecol 11 146ndash150 1988

Baldocchi D D Assessing the eddy covariance technique forevaluating carbon dioxide exchange rates of ecosystems pastpresent and future Glob Change Biol 9 479ndash492 2003

Breeuwer A Robroek B J M Limpens J Heijmans M M PD Schouten M G C and Berendse F Decreased summerwater table depth affects peatland vegetation Basic Appl Ecol10 330ndash339 doi101016jbaae200805005 2009

Bubier J Crill P Moore T Savage K and Varner R Seasonalpatterns and controls on net ecosystem CO2 exchange in a bo-real peatland complex Global Biogeochem Cy 12 703ndash714doi10102998GB02426 1998

Cadotte M W Cardinale B J and Oakley T H Evolutionaryhistory and the effect of biodiversity on plant productivity PNatl Acad Sci USA 105 17012ndash17017 2008

Cardinale B J Matulich K L Hooper D U Byrnes J EDuffy E Gamfeldt L Balvanera P OrsquoConnor M I and Gon-zalez A The functional role of producer diversity in ecosys-tems Am J Bot 98 572ndash592 doi103732ajb1000364 2011

Chapin F S Matson P A Vitousek P M and Chapin M CPrinciples of terrestrial ecosystem ecology 2nd Edn SpringerNew York NY 1991

de Miguel S Mehtaumltalo L Shater Z Kraid B and Pukkala TEvaluating marginal and conditional predictions of taper modelsin the absence of calibration data Can J Forest Res 42 1383ndash1394 doi101139x2012-090 2012

Ensminger I Sveshnikov D Campbell D A Funk C Jans-son S Lloyd J Shibistova O and Atildequist G Intermittentlow temperatures constrain spring recovery of photosynthesis inboreal Scots pine forests Glob Change Biol 10 995ndash1008doi101111j1365-2486200400781x 2004

Frank D A and McNaughton S J Stability Increaseswith Diversity in Plant Communities Empirical Evidencefrom the 1988 Yellowstone Drought Oikos 62 360ndash362doi1023073545501 1991

Frolking S E Bubier J L Moore T R Ball T BellisarioL M Bhardwaj A Carroll P Crill P M Lafleur P MMcCaughey J H Roulet N T Suyker A E Verma S BWaddington J M and Whiting G J Relationship betweenecosystem productivity and photosynthetically active radiationfor northern peatlands Global Biogeochem Cy 12 115ndash126doi10102997GB03367 1998

Gaberšcik A and Martincic A Seasonal dynamics of net photo-synthesis and productivity of Sphagnum papillosum Lindbergia105ndash110 1987

Gorham E Northern Peatlands Role in the Carbon Cycle andProbable Responses to Climatic Warming Ecol Appl 1 182doi1023071941811 1991

Gunderson L H Ecological resiliencendashin theory and applicationAnnu Rev Ecol Syst 31 425ndash439 2000

Gunnarsson U Global patterns of Sphagnum productivity JBryol 27 269ndash279 doi101179174328205X70029 2005

Haacutejek T Tuittila E-S Ilomets M and Laiho R Light re-sponses of mire mosses ndash a key to survival after water-level drawdown Oikos 118 240ndash250 doi101111j1600-0706200816528x 2009

Hayward P M and Clymo R S The growth of Sphagnum ex-periments on and simulation of some effects of light flux andwater-table depth J Ecol 71 845ndash863 1983

Haumlmet-Ahti L Suominen J Ulvinen T and Uotila P (Eds)Retkeilykasvio (Field Flora of Finland) Edn 4 Finnish Museumof Natural History Helsinki 1998

Hector A Hautier Y Saner P Wacker L Bagchi R Joshi JScherer-Lorenzen M Spehn E M Bazeley-White E Weilen-mann M Caldeira M C Dimitrakopoulos J A Finn J AHuss-Danell K Jumpponen A Mulder C P H Palmborg CPereira J S Siamantziouras A S D Terry A C TroumbisA Y Schmid B and Loreau M General stabilizing effectsof plant diversity on grassland productivity through populationasynchrony and overyielding Ecology 91 2213ndash2220 2010

Hughes A R Inouye B D Johnson M T J Underwood Nand Vellend M Ecological consequences of genetic diversityEcological effects of genetic diversity Ecol Lett 11 609ndash623doi101111j1461-0248200801179x 2008

Kormann R and Meixner F X An analytical footprint model fornon-neutral stratification Bound-Lay Meteorol 99 207ndash2242001

Korrensalo A Haacutejek T Vesala T Mehtaumltalo L and TuittilaE-S Variation in photosynthetic properties among bog plantsBotany 94 1127ndash1139 doi101139cjb-2016-0117 2016a

Korrensalo A Haacutejek T Alekseychik P Rinne J Vesala TMehtaumltalo L Mammarella I and Tuittila E-S Eddy co-variance and environmental data of boreal bogs plant speciesdoi101594PANGAEA862986 2016b

Kuiper J J Mooij W M Bragazza L and Robroek B J Plantfunctional types define magnitude of drought response in peat-land CO2 exchange Ecology 95 123ndash131 2014

Laine A M Ehonen S Juurola E Mehtaumltalo L and TuittilaE-S Performance of late succession species along a chronose-quence Environment does not exclude Sphagnum fuscum fromthe early stages of mire development edited by Zobel M JVeg Sci 26 291ndash301 doi101111jvs12231 2015



Laine J The intricate beauty of ldquoSphagnumrdquo mosses a Finnishguide to identification University of Helsinki Department of For-est Ecology 2009

Larcher W Physiological Plant Ecology Ecophysiology andStress Physiology of Functional Groups Springer 2003

Leppaumllauml M Kukko-Oja K Laine J and Tuittila E-S Seasonaldynamics of CO 2 exchange during primary succession of borealmires as controlled by phenology of plants Ecoscience 15 460ndash471 doi10298015-4-3142 2008

Lindholm T Growth dynamics of the peat moss Sphagnum fuscumon hummocks on a raised bog in southern Finland Ann BotFenn 27 67ndash78 1990

Loisel J Gallego-Sala A V and Yu Z Global-scale pattern ofpeatland Sphagnum growth driven by photosynthetically activeradiation and growing season length Biogeosciences 9 2737ndash2746 doi105194bg-9-2737-2012 2012

Lund M Lafleur P M Roulet N T Lindroth A ChristensenT R Aurela M Chojnicki B H Flanagan L B HumphreysE R Laurila T Oechel W C Olejnik J Rinne J SchubertP and Nilsson M B Variability in exchange of CO2 across12 northern peatland and tundra sites exchange of CO2 in wet-lands Glob Change Biol 16 2436ndash2448 doi101111j1365-2486200902104x 2009

Lund M Christensen T R Lindroth A and Schubert P Effectsof drought conditions on the carbon dioxide dynamics in a tem-perate peatland Environ Res Lett 7 45704 doi1010881748-932674045704 2012

Maanavilja L Riutta T Aurela M Pulkkinen M LaurilaT and Tuittila E-S Spatial variation in CO2 exchangeat a northern aapa mire Biogeochemistry 104 325ndash345doi101007s10533-010-9505-7 2011

Mammarella I Peltola O Nordbo A Jaumlrvi L and Rannik UumlQuantifying the uncertainty of eddy covariance fluxes due to theuse of different software packages and combinations of process-ing steps in two contrasting ecosystems Atmos Meas Tech 94915ndash4933 doi105194amt-9-4915-2016 2016

Moore T R Bubier J L Frolking S E Lafleur P M andRoulet N T Plant biomass and production and CO2 exchangein an ombrotrophic bog J Ecol 90 25ndash36 doi101046j0022-0477200100633x 2002

Moore T R Lafleur P M Poon D M I Heumann B WSeaquist J W and Roulet N T Spring photosynthesis ina cool temperate bog Glob Change Biol 12 2323ndash2335doi101111j1365-2486200601247x 2006

Morin X Fahse L de Mazancourt C Scherer-Lorenzen Mand Bugmann H Temporal stability in forest productivity in-creases with tree diversity due to asynchrony in species dy-namics edited by Rejmanek M Ecol Lett 17 1526ndash1535doi101111ele12357 2014

Nijp J J Limpens J Metselaar K van der Zee S E A TM Berendse F and Robroek B J M Can frequent pre-cipitation moderate the impact of drought on peatmoss car-bon uptake in northern peatlands New Phytol 203 70ndash80doi101111nph12792 2014

Pinheiro J and Bates D Mixed-Effects Models in S and S-PLUSSpringer New York 2000

Rannik Uuml Kolari P Vesala T and Hari P Uncertain-ties in measurement and modelling of net ecosystem ex-

change of a forest Agr Forest Meteorol 138 244ndash257doi101016jagrformet200605007 2006

Robroek B J M Limpens J Breeuwer A and SchoutenM G C Effects of water level and temperature on perfor-mance of four Sphagnum mosses Plant Ecol 190 97ndash107doi101007s11258-006-9193-5 2007

Roulet N T Lafleur P M Richard P J H Moore T RHumphreys E R and Bubier J Contemporary carbon bal-ance and late Holocene carbon accumulation in a northernpeatland Glob Change Biol 13 397ndash411 doi101111j1365-2486200601292x 2007

Rydin H and Jeglum J K The Biology of Peatlands OUP Ox-ford 2013

Small E Photosynthetic rates in relation to nitrogen recycling asan adaptation to nutrient deficiency in peat bog plants Can JBot 50 2227ndash2233 1972

Tanja S Berninger F Vesala T Markkanen T Hari P MaumlkelaumlA Ilvesniemi H Haumlnninen H Nikinmaa E Huttula T Lau-rila T Aurela M Grelle A Lindroth A Arneth A Shibis-tova O and Lloyd J Air temperature triggers the recovery ofevergreeen boreal forest photosynthesis in spring Glob ChangeBiol 9 1410ndash1426 2003

Tilman D Knops J Wedin D Reich P Ritchie Mand Siemann E The Influence of Functional Diversity andComposition on Ecosystem Processes Science 277 1300doi101126science27753301300 1997

Turetsky M R Bond-Lamberty B Euskirchen E Talbot JFrolking S McGuire A D and Tuittila E-S The resilienceand functional role of moss in boreal and arctic ecosystemsTansley review New Phytol 196 49ndash67 doi101111j1469-8137201204254x 2012

Waddington J M and Roulet N T Carbon balance of aboreal patterned peatland Glob Change Biol 6 87ndash97doi101046j1365-2486200000283x 2000

Ward S E Bardgett R D McNamara N P and Ostle NJ Plant functional group identity influences short-term peat-land ecosystem carbon flux evidence from a plant removalexperiment Funct Ecol 23 454ndash462 doi101111j1365-2435200801521x 2009

Wehr R Munger J W McManus J B Nelson D D ZahniserM S Davidson E A Wofsy S C and Saleska S R Season-ality of temperate forest photosynthesis and daytime respirationNature 534 680ndash683 2016

Weltzin J F Pastor J Harth C Bridgham S D Up-degraff K and Chapin C T Response of bog andfen plant communities to warming and water-table ma-nipulations Ecology 81 3464ndash3478 doi1018900012-9658(2000)081[3464ROBAFP]20CO2 2000

Wilson D Alm J Riutta T Laine J Byrne K A Far-rell E P and Tuittila E-S A high resolution green area in-dex for modelling the seasonal dynamics of CO2 exchange inpeatland vascular plant communities Plant Ecol 190 37ndash51doi101007s11258-006-9189-1 2007

Yachi S and Loreau M Biodiversity and ecosystem productivityin a fluctuating environment the insurance hypothesis P NatlAcad Sci USA 96 1463ndash1468 1999

Yu Z C Northern peatland carbon stocks and dynamics a re-view Biogeosciences 9 4071ndash4085 doi105194bg-9-4071-2012 2012



Zona D Lipson D A Richards J H Phoenix G K LiljedahlA K Ueyama M Sturtevant C S and Oechel W C De-layed responses of an Arctic ecosystem to an extreme summerimpacts on net ecosystem exchange and vegetation functioningBiogeosciences 11 5877ndash5888 doi105194bg-11-5877-20142014


Abstract

Introduction

Methods

Study site

Plant level photosynthesis measurements

Net photosynthesis model

Upscaling

Ecosystem-level CO2 exchange measurements and estimation of gross primary production

Results

Cumulative growing season gross photosynthesis

Cumulative growing season net photosynthesis

Discussion

Comparison of upscaled gross photosynthesis values with eddy covariance gross primary production estimates

The contribution of plant species to ecosystem-level gross photosynthesis

Ecosystem-level net photosynthesis

The role of functional diversity for peatland carbon sink

Conclusions

Data availability

Competing interests

Author contributions

Acknowledgement

References


Boreal bogs are ecosystems with low species diversity buta high diversity of growth forms due to the large microtopo-graphical variation and associated diversity of habitats alongthe water table gradient (Turetsky et al 2012 Rydin andJeglum 2013) Several studies (eg Weltzin et al 2000Moore et al 2002 Leppaumllauml et al 2008) have reported thatpatterned bogs produce more biomass and have less varia-tion in gross photosynthesis over the growing season thanfens which receive additional nutrients from the surroundingmineral soil and generally have more homogenous sedge-dominated vegetation (Weltzin et al 2000) Experimentalstudies have shown that bog plant growth forms have dif-ferential responses to warming and water table level manip-ulation which can help to maintain the level of total ecosys-tem productivity under changing environmental conditions(Weltzin et al 2000 Breeuwer et al 2009) Short-term plantremoval experiments have shown the differential roles ofplant functional types for the peatland NEE and GPP (Wardet al 2009 Kuiper et al 2014) Photosynthetic properties ofbog plants are known to differ widely between species of thesame functional type (Small 1972) and between phases ofgrowing season (Korrensalo et al 2016a) So far the role ofspecies-level differences in temporal variation of bog ecosys-tem photosynthesis has not been studied

Here we aim to solve the linkage between the tempo-ral pattern of bog carbon balance and the development ofspecies-specific potential photosynthesis and leaf area Forthis purpose we quantified the contribution of different plantspecies to ecosystem-level photosynthesis over a growingseason As species differ in their photosynthetic propertiesand the properties vary over the growing season we expecttheir importance for the ecosystem carbon sequestration toalso vary over the season To reach our aim we estimatePN and PG for the whole study site based on the monthlyspecies-level light response of photosynthesis and species-specific leaf area development over a growing season Tovalidate the upscaling approach the sum of species-level PGis compared to the GPP derived from eddy covariance (EC)measurements at the study site

2 Methods

21 Study site

The study site (6150179prime N 2410145prime E) is situated in anombrotrophic bog which is a part of the Siikaneva peatlandcomplex in southern Finland located in the southern bo-real vegetation zone The annual temperature sum in the area(base temperature 5 C) is 1318 degree days annual rainfallis 707 mm and the average annual January and July tem-peratures are 42 minus72 and 171 C respectively (30-yearaverages (years 1982ndash2011) from the Juupajoki-Hyytiaumllaumlweather station) The study site has a surface topography typ-ical of raised bogs that varies from open water pools and mud

surfaces to hollows lawns and hummocks An EC flux toweris mounted on a raft in the center of the site

The vegetation is mainly composed of 11 vascular plantand 8 Sphagnum species (Table 1) the abundance of whichvaries markedly along the microtopographical gradient Acontinuous Sphagnum carpet covers the surfaces from hum-mocks to hollows although no Sphagna are present on themud and water surfaces Sphagnum cuspidatum and S ma-jus are dominant in hollows S papillosum S rubellum Sbalticum and S magellanicum in lawns and S fuscum Srubellum and S angustifolium cover the hummocks Vascu-lar plant species composition includes Rhynchospora albaScheuchzeria palustris and Carex limosa vegetation on mudand hollow surfaces dwarf shrubs (Andromeda polifoliaVaccinium oxycoccos) and Eriophorum vaginatum on lawnsurfaces and a shrub layer of Calluna vulgaris Betula nanaEmpetrum nigrum and sedges Eriophorum vaginatum andTrichophorum cespitosum on hummocks

22 Plant level photosynthesis measurements

To quantify the role of plant species in ecosystem-level pho-tosynthesis over the growing season we conducted net pho-tosynthesis (PN) measurements of the 19 most commonspecies at the study site Over the growing season 2013 wemeasured CO2 exchange of three to five samples of eachspecies per month at three light levels with two open flow-through gas exchange measurement devices (GFS-3000Walz Germany and LI-6400 LI-COR USA) Samples werecollected from several locations inside the study area a max-imum of 30 h prior to measurement Sphagna were collectedin small plastic bags and vascular plants in plastic boxeswith an ample amount of roots and peat These were keptmoist until measured Vascular plants were kept in shadedconditions and Sphagna were stored in the dark at 5 C Themoss cuvettes were filled with Sphagnum capitula imitatingthe natural shoot density of each species in the field result-ing in a total number of 6ndash16 capitula inside a cuvette de-pending on the species Before placing the capitula into thecuvettes they were first wetted and then lightly dried of ex-cess water with pulp paper The cuvette was then placed un-der a photosynthetic photon flux density (PPFD) of approx-imately 1000 micromol mminus2 sminus1 to acclimate to light for 20 minThe measured light levels were 2000 25 and 0 micromol mminus2 sminus1

for vascular plants and 2000 35 and 0 micromol mminus2 sminus1 forSphagna Light levels were designed to catch both the lin-ear light-limited beginning and the light-saturated maximumof the light response curve without causing photoinhibitoryreduction of photosynthetic rate (Laine et al 2015) Vas-cular plants were measured with a standard cuvette but forSphagna we used a moss cuvette of our own constructionwith internal dimensions of 3times 2times 1 cm (Haacutejek et al 2009)with a net bottom to allow airflow above and below the sam-ple reducing the aerodynamic resistance After changing thelight level we waited for the PN to reach steady state be-
















ks+PPFDsi

+ esi (1)



24 Upscaling











Reco = RrefQ

(TpminusTref

10

)10 (2)



GPP=Pmax PPFDk+PPFD

(LAI+ b) (3)



GPP=Pmax PPFDk+PPFD

( LAI+ b)exp


2

T 2tol

) (4)



3 Results








120 150 180 210 240 270

05

10

15

20

25

30

35

PG

GP

P (

g C

mminus2

dminus1

)

PG

GPP in ambient T

GPP in 20 degC

(a)


JDMonth

120 150 180 210 240 270

05

1015

2025

30

Air

T (

degC)

(b)

May Jun Jul Aug Sep

JDMonth

120 150 180 210 240 270

000

010

020

030

LAI (

m2 m

minus2 )

(c)

May Jun Jul Aug Sep

120 150 180 210 240 270

minus20

minus15

minus10

minus5

WT

(cm

)

(d)

May Jun Jul Aug Sep









120 150 180 210 240 270

00

05

10

15

PG

(g

C m

minus2 d

minus1 )

Vasculars Sphagna


JD

Month


4 Discussion







120 150 180 210 240 270

00

01

02

03

04

05

PG

(g

C m

minus2 d

minus1 )


(a)


JD

Month

120 150 180 210 240 270

00

01

02

03

04

05

PG

(g

C m

minus2 d

minus1 )


(b)


JD

Month









G (

g C

mminus2

mon

thminus1

)

05

1015

2025

3035 AndrPol


(a)

PG

(g

C m

minus2 m

onth

minus1 )

05

1015

2025

3035 SphaRub


(b)


PG

(g

C m

minus2 m

onth

minus1 )

010

2030

4050

60

SphagnaVasculars

(c)












5 Conclusions



6 Data availability










References
































































Abstract

Introduction

Methods

Study site



Upscaling


Results



Discussion





Conclusions

Data availability

Competing interests


Acknowledgement

References















ks+PPFDsi

+ esi (1)



24 Upscaling











Reco = RrefQ

(TpminusTref

10

)10 (2)



GPP=Pmax PPFDk+PPFD

(LAI+ b) (3)



GPP=Pmax PPFDk+PPFD

( LAI+ b)exp


2

T 2tol

) (4)



3 Results








120 150 180 210 240 270

05

10

15

20

25

30

35

PG

GP

P (

g C

mminus2

dminus1

)

PG

GPP in ambient T

GPP in 20 degC

(a)


JDMonth

120 150 180 210 240 270

05

1015

2025

30

Air

T (

degC)

(b)

May Jun Jul Aug Sep

JDMonth

120 150 180 210 240 270

000

010

020

030

LAI (

m2 m

minus2 )

(c)

May Jun Jul Aug Sep

120 150 180 210 240 270

minus20

minus15

minus10

minus5

WT

(cm

)

(d)

May Jun Jul Aug Sep









120 150 180 210 240 270

00

05

10

15

PG

(g

C m

minus2 d

minus1 )

Vasculars Sphagna


JD

Month


4 Discussion







120 150 180 210 240 270

00

01

02

03

04

05

PG

(g

C m

minus2 d

minus1 )


(a)


JD

Month

120 150 180 210 240 270

00

01

02

03

04

05

PG

(g

C m

minus2 d

minus1 )


(b)


JD

Month









G (

g C

mminus2

mon

thminus1

)

05

1015

2025

3035 AndrPol


(a)

PG

(g

C m

minus2 m

onth

minus1 )

05

1015

2025

3035 SphaRub


(b)


PG

(g

C m

minus2 m

onth

minus1 )

010

2030

4050

60

SphagnaVasculars

(c)












5 Conclusions



6 Data availability










References
































































Abstract

Introduction

Methods

Study site



Upscaling


Results



Discussion





Conclusions

Data availability

Competing interests


Acknowledgement

References




Reco = RrefQ

(TpminusTref

10

)10 (2)



GPP=Pmax PPFDk+PPFD

(LAI+ b) (3)



GPP=Pmax PPFDk+PPFD

( LAI+ b)exp


2

T 2tol

) (4)



3 Results








120 150 180 210 240 270

05

10

15

20

25

30

35

PG

GP

P (

g C

mminus2

dminus1

)

PG

GPP in ambient T

GPP in 20 degC

(a)


JDMonth

120 150 180 210 240 270

05

1015

2025

30

Air

T (

degC)

(b)

May Jun Jul Aug Sep

JDMonth

120 150 180 210 240 270

000

010

020

030

LAI (

m2 m

minus2 )

(c)

May Jun Jul Aug Sep

120 150 180 210 240 270

minus20

minus15

minus10

minus5

WT

(cm

)

(d)

May Jun Jul Aug Sep









120 150 180 210 240 270

00

05

10

15

PG

(g

C m

minus2 d

minus1 )

Vasculars Sphagna


JD

Month


4 Discussion







120 150 180 210 240 270

00

01

02

03

04

05

PG

(g

C m

minus2 d

minus1 )


(a)


JD

Month

120 150 180 210 240 270

00

01

02

03

04

05

PG

(g

C m

minus2 d

minus1 )


(b)


JD

Month









G (

g C

mminus2

mon

thminus1

)

05

1015

2025

3035 AndrPol


(a)

PG

(g

C m

minus2 m

onth

minus1 )

05

1015

2025

3035 SphaRub


(b)


PG

(g

C m

minus2 m

onth

minus1 )

010

2030

4050

60

SphagnaVasculars

(c)












5 Conclusions



6 Data availability










References
































































Abstract

Introduction

Methods

Study site



Upscaling


Results



Discussion





Conclusions

Data availability

Competing interests


Acknowledgement

References


120 150 180 210 240 270

05

10

15

20

25

30

35

PG

GP

P (

g C

mminus2

dminus1

)

PG

GPP in ambient T

GPP in 20 degC

(a)


JDMonth

120 150 180 210 240 270

05

1015

2025

30

Air

T (

degC)

(b)

May Jun Jul Aug Sep

JDMonth

120 150 180 210 240 270

000

010

020

030

LAI (

m2 m

minus2 )

(c)

May Jun Jul Aug Sep

120 150 180 210 240 270

minus20

minus15

minus10

minus5

WT

(cm

)

(d)

May Jun Jul Aug Sep









120 150 180 210 240 270

00

05

10

15

PG

(g

C m

minus2 d

minus1 )

Vasculars Sphagna


JD

Month


4 Discussion







120 150 180 210 240 270

00

01

02

03

04

05

PG

(g

C m

minus2 d

minus1 )


(a)


JD

Month

120 150 180 210 240 270

00

01

02

03

04

05

PG

(g

C m

minus2 d

minus1 )


(b)


JD

Month









G (

g C

mminus2

mon

thminus1

)

05

1015

2025

3035 AndrPol


(a)

PG

(g

C m

minus2 m

onth

minus1 )

05

1015

2025

3035 SphaRub


(b)


PG

(g

C m

minus2 m

onth

minus1 )

010

2030

4050

60

SphagnaVasculars

(c)












5 Conclusions



6 Data availability










References
































































Abstract

Introduction

Methods

Study site



Upscaling


Results



Discussion





Conclusions

Data availability

Competing interests


Acknowledgement

References


120 150 180 210 240 270

00

05

10

15

PG

(g

C m

minus2 d

minus1 )

Vasculars Sphagna


JD

Month


4 Discussion







120 150 180 210 240 270

00

01

02

03

04

05

PG

(g

C m

minus2 d

minus1 )


(a)


JD

Month

120 150 180 210 240 270

00

01

02

03

04

05

PG

(g

C m

minus2 d

minus1 )


(b)


JD

Month









G (

g C

mminus2

mon

thminus1

)

05

1015

2025

3035 AndrPol


(a)

PG

(g

C m

minus2 m

onth

minus1 )

05

1015

2025

3035 SphaRub


(b)


PG

(g

C m

minus2 m

onth

minus1 )

010

2030

4050

60

SphagnaVasculars

(c)












5 Conclusions



6 Data availability










References
































































Abstract

Introduction

Methods

Study site



Upscaling


Results



Discussion





Conclusions

Data availability

Competing interests


Acknowledgement

References


120 150 180 210 240 270

00

01

02

03

04

05

PG

(g

C m

minus2 d

minus1 )


(a)


JD

Month

120 150 180 210 240 270

00

01

02

03

04

05

PG

(g

C m

minus2 d

minus1 )


(b)


JD

Month









G (

g C

mminus2

mon

thminus1

)

05

1015

2025

3035 AndrPol


(a)

PG

(g

C m

minus2 m

onth

minus1 )

05

1015

2025

3035 SphaRub


(b)


PG

(g

C m

minus2 m

onth

minus1 )

010

2030

4050

60

SphagnaVasculars

(c)












5 Conclusions



6 Data availability










References
































































Abstract

Introduction

Methods

Study site



Upscaling


Results



Discussion





Conclusions

Data availability

Competing interests


Acknowledgement

References


G (

g C

mminus2

mon

thminus1

)

05

1015

2025

3035 AndrPol


(a)

PG

(g

C m

minus2 m

onth

minus1 )

05

1015

2025

3035 SphaRub


(b)


PG

(g

C m

minus2 m

onth

minus1 )

010

2030

4050

60

SphagnaVasculars

(c)












5 Conclusions



6 Data availability










References
































































Abstract

Introduction

Methods

Study site



Upscaling


Results



Discussion





Conclusions

Data availability

Competing interests


Acknowledgement

References




5 Conclusions



6 Data availability










References
































































Abstract

Introduction

Methods

Study site



Upscaling


Results



Discussion





Conclusions

Data availability

Competing interests


Acknowledgement

References




References
































































Abstract

Introduction

Methods

Study site



Upscaling


Results



Discussion





Conclusions

Data availability

Competing interests


Acknowledgement

References




































Abstract

Introduction

Methods

Study site



Upscaling


Results



Discussion





Conclusions

Data availability

Competing interests


Acknowledgement

References

Species-specific temporal variation in photosynthesis as a ...

Documents

Transcript of Species-specific temporal variation in photosynthesis as a ...