(This is a sample cover image for this issue. The actual cover is not yet available at this time.)

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Soil surface CO2 efflux measurements in Norway spruce forests: Comparison betweenfour different sites across Europe — from boreal to alpine forest

Manuel Acosta a,⁎,1, Marian Pavelka a, Leonardo Montagnani b, Werner Kutsch c, Anders Lindroth d,Radosław Juszczak e, Dalibor Janouš a

a Global Change Research Centre — CzechGlobe, ASCR, Brno, Czech Republicb Autonomous Province of Bolzano, Forest Service, Bolzano, Italyc Johann Heinrich von Thünen-Institut, Braunschweig, Germanyd University of Lund, GeoBiosphere Sciences Centre, Physical Geography and Ecosystems Analysis, Lund, Swedene Poznan University of Life Sciences, Department of Meteorology, Poznan, Poland

a b s t r a c ta r t i c l e i n f o

Article history:Received 13 October 2011Received in revised form 9 August 2012Accepted 19 August 2012Available online xxxx

Keywords:Soil CO2 effluxForestChamber methodQ10

Soil temperatureSpatial variability

Extensive measurements of soil surface (including vegetation cover) CO2 effluxwere carried out on 80 positionsat four different forest sites (Sweden, Germany, Czech Republic and Italy) using a closed dynamic chambertechnique. The period of measurement was 4–5 consecutive days per site. Two approaches were used to analyzethe measured data, the Q10 parameter and the Arrhenius relationship. Basic environmental factors such as soiltemperature and moisture were measured. All the four investigated sites showed a positive dependence of thesoil surface CO2 efflux on soil temperature. The four datasets generally resulted in good agreement (up to 93%)between the approaches and residual analysis showed no significant difference between approaches (less than8%). The Q10 ranged between 2.0 and 2.3 among the sites. The highest spatial variation of the soil surface CO2

efflux at 10 °C (expressed by the coefficient of variation CV) ranged from 30 to 65% between sites.© 2012 Elsevier B.V. All rights reserved.

1. Introduction

The efflux of CO2 from soil surface is one of the largest and mostimportant fluxes of carbon in terrestrial ecosystems (Davidson et al.,2002; Raich and Schlesinger, 1992). Soil CO2 efflux results from thecombination of biological processes (i.e. production of CO2 by roots,soil micro-organisms and soil macro fauna) and physical processes(i.e. CO2 diffusion from sources to soil surface) (Le Dante et al.,1999), and is thought to account for 60–80% of ecosystem respiration(Davidson et al., 1998; Epron et al., 1999). Soil CO2 efflux, also called“soil respiration”, is the main pathway for carbon moving from theecosystem to the atmosphere and can strongly influence net carbonuptake from the atmosphere, or net ecosystem production(Valentini et al., 2000; Ryan and Law, 2005). However, as the effluxof CO2 from the soil is typically characterized by a large temporaland spatial variability, a high heterogeneity introduces an uncertaintyin the estimations.

Several studies regarding factors influencing temporal and spatialvariabilities of soil CO2 efflux at local or multi-site scale have been

published, the most common are related to soil temperature andmoisture (Buchmann, 2000; Janssens et al., 1999; Kosugi et al.,2007; Vodkin et al., 2006), vegetation characteristics (Chojnicki etal., 2010; Law et al., 2001; Widén, 2002), root density or biomass(Fang et al., 1998; Rodeghiero and Cescatti, 2006; Stoyan et al.,2000), quantity–quality of organic matter and microbial biomass(Rayment and Jarvis, 2000; Ryan et al. 1996; Taylor et al. 1989), nitro-gen status and soil carbon (Bowden et al., 2004; Robertson et al.,1999), tree photosynthesis (Tang et al., 2005) and net primary pro-duction (Janssens et al., 2001; Korari et al., 2009; Moyano et al.,2008). Nevertheless, among all above mentioned environmental fac-tors, temperature is still a key factor influencing soil CO2 efflux andother processes involved in the production of soil CO2.

The dependence of soil CO2 efflux on soil temperature has beenfrequently described (Lloyd and Taylor, 1994). Soil temperature isthe best predictor of soil CO2 efflux in most of the ecosystems whenno water stress occurs (Janssens et al., 1999). Usually, soil moistureis the other dominant factor influencing the soil CO2 efflux (Epronet al., 1999). Numerous models have been developed to express thetemperature sensitivity of soil CO2 efflux (Reichstein et al., 2003;Tuomi et al., 2008). On the other hand, exponential relationships, es-pecially the Q10 relationship, are the most frequently used to predictrespiration rates using temperature (Davidson et al., 1998; Raichand Potter, 1995). The use of the Q10 relationship has often been

Geoderma 192 (2013) 295–303

⁎ Corresponding author at: University of Helsinki, P.O. Box 48 (Erik Palmenin aukio 1),00014 Helsinki, Finland. Tel.: +358 504154763; fax: +358 9 19148802.

E-mail address: manuelacosta35@gmail.com (M. Acosta).1 Present address: University of Helsinki, Department of Physics, Helsinki, Finland.

0016-7061/$ – see front matter © 2012 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.geoderma.2012.08.027

Contents lists available at SciVerse ScienceDirect

Geoderma

j ourna l homepage: www.e lsev ie r .com/ locate /geoderma

Author's personal copy

criticized because Q10 decreases with increasing temperature and de-pends on soil moisture conditions (Davidson et al., 2006). Anotherwidely used approach to determine the relationship between soilCO2 efflux and temperature is the Arrhenius relationship in whichthe activation energy (Ea) depends on soil temperature as proposedby Lloyd and Taylor (1994).

Generally, the chamber technique has been used for the direct mea-surement of soil surface CO2 efflux. Different chamber methods andtypes of chamber have been employed mainly depending on the typeof investigated ecosystem (Davidson et al., 2002; Fang and Moncrieff,1996; Janssens et al., 2000; Juszczak et al., 2012; Pumpanen et al.,2004; Acosta et al., 2004; Saiz et al., 2006). Although there are advan-tages and disadvantages associated with chamber measurements, thechamber technique is generally considered as a useful tool for the deter-mination of soil CO2 efflux spatial heterogeneity and a complementarytool for eddy covariance measurements. Chamber technique is not af-fected by CO2 advection (Welles et al., 2001); on the other hand, cham-bers can modify the soil environment (e.g., temperature, humidity, CO2

concentration) if they cover the measured area for long periods, andtherefore may introduce biases (Livingston and Hutchinson, 1995). An-other issue related to the chamber technique is a common problem inacross-ecosystem comparison in that soil surface CO2 efflux rates areoftenmeasuredwith different systems introducing an important sourceof uncertainty (Pumpanen et al., 2004).

In our study, we measured and compared the soil CO2 efflux at fourdifferent sites across Europe. Three identical soil CO2 efflux systemswere transported to each investigated site and the same measuringmethodology (including man power) was applied in all sites in orderto eliminate any uncertainty regarding to equipment and methodologyacross-ecosystem comparison, thus, doing this study was unique andchallenging. This study deals with short-term temporal variability aswell as within-site spatial variability of the soil surface CO2 efflux. Ourobjectives were: (1) to provide a reliable estimate of the short-termsoil surface CO2 efflux at the four forest sites with different locationsin Europe applying the same chamber measurement system, (2) todetermine spatial variation in the soil surface CO2 efflux, (3) to con-struct a map of the soil surface CO2 efflux on the base of measurementsdone for each investigated site and (4) to identify possible factorscontrolling spatial variation of the soil surface CO2 efflux.

2. Material and methods

2.1. Site description

The four sites were investigated during this study. Three siteswere from the CarboEurope-Integrated Project (CE-IP) advection

activities ADVEX performed from 2005 to 2006: Renon (Italy),Wetzstein (Germany), and Norunda (Sweden). The fourth investigat-ed site was Bílý Kříž (Czech Republic), (for more details on the exper-imental set-up and first results of ADVEX, see Feigenwinter et al.,2008).

Renon (RE) site is influenced by an alpine, windy and humid cli-mate. The unevenly aged forest is heterogeneous with gaps betweengroups of older and younger trees, and sparse understorey vegetationis dominated by blueberries (Vaccinium myrtillus (L.)) and grasses,(for more details about the site, see Marcolla et al., 2005). Wetzstein(WS) site is situated nearly on the crest of ridge with steep slopes tothe ESE and WNW directions. The climate is temperate humid. Theforest is a managed homogeneous stand and ground vegetation issparse with occasional areas covered by blueberries and mosses, formore details about the site see Anthoni et al. (2004). Norunda (NO)site is a flat area located in a cold-temperate to boreal climate. It is amanaged forest and the forest floor is mainly covered by boulders,rocks (with a diameter of up to 1.5 m), dwarf shrubs and mosses.For more details about the NO site see Morén and Lindroth (2000).Bíly Kríž (BK) site is located on a gentle slope with SSW exposure.The climate is cold-temperate. The forest is a managed homogeneousstand. The ground vegetation is dominated by blueberries, for moredetails about the site see Urban et al. (2007). Main characteristics ofthe sites are presented in Table 1.

2.2. Soil surface CO2 efflux measurements

Extensive measurements of the soil surface CO2 efflux (includingvegetation cover — low growing plants covering the ground) werecarried out for 80 positions at all the investigated sites. At each posi-tion a PVC collar (19 cm in diameter, height 8.5 cm) was placed andinserted into the soil at about 3 cm depth one day before measure-ments began. During the installation of the ring a cutting effectcould be generated a paulatim increase in the soil CO2 efflux thatcould last for weeks or months depending of the impact due to thecut of root biomass. However, installation of rings one day beforemeasurements is the minimum time necessary to obtain reliable soilCO2 effluxmeasurements (see Heinemeyer et al., 2011 for more infor-mation about collar insertion). At WS, NO and BK sites a grid designwas used in order to investigate the soil CO2 efflux spatial variation.In this design a collar was installed at each intersecting point of thegrid with 5 m distance between positions at sites WS and BK sinceboth sites present a homogeneous soil surface, and with 10 m dis-tance at NO due to the presence of boulders and rocks on the forestfloor. Because the forest stand in RE was heterogeneous with gapsbetween groups of older and younger trees, collars were installed in

Table 1Main characteristics of the investigated sites.

Site name Renon Wetzstein Norunda Bílý Kříž

Country Italy Germany Sweden Czech RepublicRegion South Tyrol Thuringia Uppland BeskydyLatitude 46° 35′ N 50° 27′ N 60° 05′ N 49° 30′ NLongitude 11° 26′ E 11° 27′ E 17° 28′ E 18° 32′ EElevation (m) 1730 785 45 908Topography Alpine slope (11°) Top of a hill Flat Slope (13 °C)Mean annual temperature (°C) 4.1 5.9 5.5 5.5Mean annual precipitation (mm) 1010 840 527 1200Forest type Evergreen needleleaf forest

(Picea abies 85%)Evergreen needleleaf forest(Picea abies)

Evergreen needleleaf forest(Picea abies, Pinus sylvestris)

Evergreen needleleaf forest(Picea abies)

Stand age Uneven 50 50–100 27Stand density (trees/ha) 745 670 600 2000Mean canopy height (m) 25 22 28 12Mean DBH (cm) 17 27 29 13LAI (m2 m−2) 4–5.5 4 3–6 10.5Soil description — FAO Haplic podzols Dystric cambisol and podzols Dystric regosol Haplic podzolsSoil depth (m) 0.3 0.4 2.0 0.6Depth of the main rooting zone (m) 0.2 0.3 0.4 0.2

296 M. Acosta et al. / Geoderma 192 (2013) 295–303

Author's personal copy

groups (27 groups of three collars) covering the most representativesoil surface. The 80 selected sampling points, chosen on a systematicgrid (WS, NO, BK) or according to the site morphology and vegetationcharacteristics (RE) were based on a similar model by Rodeghiero andCescatti (2008) and covered the overall CO2 spatial variation of theinvestigated sites.

We measured the soil surface CO2 efflux at each site simultaneouslywith three identical systems working in a closed dynamic mode. Eachsystem consisted of an infrared gas analyzer (Li-6250, Li-Cor Inc., Lin-coln, NB, USA) and control console (Li-6200), linked to a self-madesoil chamber based on the model proposed by Norman et al. (1992).The chamber was made of white PVC (cylinder-shape, 17.0 cm heightand 15.1 in diameter) and the base part of the chamber has a borderin order to provide better sealing and stability when placed on thecollar. The chamber is equipped with a small fan for gentle air-mixingand tubing (4 mm diameter) to ensure pressure equilibrium betweenthe inside and the outside. The infrared gas analyzers were calibratedat each site (using a calibration gas — 420 ppm CO2), this being donedaily at the same time before the measurement. Moreover, the systemswere previously compared to a standard system for soil CO2 effluxmea-surement (Li6400, Li-Cor, US), analysis of the data showed availabilityand confidence of the systems for the soil surface CO2 efflux measure-ments. Soil CO2 efflux (μmol CO2 m−2 s−1) was measured over4–6 min during the closing of the chamber (one measurement).A total of 15 measurements at different times and temperature werecarried out on each position. The same procedure of the soil surfaceCO2 efflux measurement was used at all the sites. The period ofmeasurement was 4–5 consecutive days, at RE in July 2005, at WS inJune 2006, at BK in August 2006 and at NO in June 2007.

2.3. Soil vegetation

A detailed photo-documentation of the aboveground vegetationcover on the measured positions and vicinity areas (about 1 m aroundthe measured position) was done at each site using a digital compactcamera (Canon PowerShoot A2200). Thus, vegetation cover on eachmeasured position and their surroundings were determined at allthe sites along with further detailed observations for each position(e.g. waterlogged forest).

2.4. Ancillary data

Soil temperature wasmeasured at each position about 3 cm outsideof each ring during the efflux measurement using a penetration ther-mometer (Roth, Germany) at 1.5 cm depth from the soil surface, sincethemost suitable soil temperaturemeasurement depth for Q10 determi-nation is close to the soil surface (Pavelka et al., 2007). Data on soilmoisture was obtained from the measurements at the site: at NOTDR-probes were installed at three depths (10, 20 and 30 cm) at six lo-cations, atWS Delta-T Devices Theta probe (ML-2x, Delta-T, Cambridge,UK) installed at 8 cm depth at the three locations, at RE Delta T DevicesTheta probe (ML-2x, Delta-T, Cambridge, UK) installed at 15 cm depthat the three locations and at BK soil moisture was measured at threedifferent soil depths (1–17 cm, 16–32 cm, 27–34 cm) using the TDRtechnique (IMKO, Germany) at the four locations. An extra study onthe soil rock fragments (depth of stones in the measured position)was carried out at NO and BK by penetrating the soil with a thin metalrod in order to determine the dependence of R10 on soil rock fragments.

2.5. Data processing

Two approaches were used to analyze the measured data. The firstwas using the Q10 parameter (the proportional change in respirationresulting from 10 °C increase in temperature) from van't Hoff (1898).Q10 expresses the dependence of soil respiration on soil temperature.In our analysis we determined a Q10 value on the basis of our

measurements for each site in order to compare soil respirationsensitivity among the sites. The Q10 values were calculated accordingto Eq. (1) (Buchmann, 2000):

Q10 ¼ e10⋅β1 ð1Þ

where β1 is the regression coefficient obtained from the natural loga-rithm of the CO2 efflux and soil temperature. Then, a reference valueof R10 (CO2 efflux normalized to a temperature of 10 °C) for eachmeasured position was calculated as:

R10 ¼ SR � Q1010−T10ð Þ ð2Þ

where SR is the measured soil CO2 efflux (μmol CO2m−2 s−1) at Tsoil temperature. All our measured data were related to temperature.During our campaign measurements and before it no significant pre-cipitation events occurred at any of the four sites.

Both Eqs. (1) and (2) were calculated using Microsoft Office Excel2007. The second approach was related to the Arrhenius relationshipin which the activation energy Ea depended on soil temperature asproposed by Lloyd and Taylor (1994):

SR ¼ R10 � eEa1

283:15−To− 1

T−To

� �ð3Þ

where SR is the measured soil CO2 efflux (μmol CO2m−2 s−1), R10 isthe soil CO2 efflux at 10 °C (μmol CO2m−2 s−1), T is the soil temper-ature (K), To is the reference temperature and Ea is the activationenergy (J mol−1). The parameter To was fixed to 227.13 K (Lloydand Taylor, 1994) and the parameters R10 and Ea were obtainedfrom a regression using the Lloyd and Taylor equation by the curvefitting program (curve Expert 1.38, Daniel Hyams and Microsoft).

All the data for each position were taken into account and analyzedindividually regarding temperature dependence and using Q10 parame-ter and Arrhenius relationship. On the basis of the obtained R10 values, amap of the soil surface CO2 efflux was created using the grid design(WS, NO, BK) or group position (RE). The R10 (soil surface CO2 effluxat 10 °C) was interpolated using a conditional stochastic simulationmethod (Herbst et al., 2010) — geostatistic software (R version 2.14.2,the R Foundation for Statistical Computing). Even though the use ofthe Kriging method in geostatistical techniques is still recognized asthe most favorable algorithm in terms of univariate spatial estimationand reproduction of mean values (Hengl et al., 2004), the conditionalstochastic simulation technique is more appropriate and suitable forour datasets. Semivariograms were calculated from the R10 data persite and a spherical function was fitted. A typical semivariogram has anugget variance (C0). With increasing lag distance, the variance willincrease up to the sill variance (C0+C). The nugget represents eitherrandom error or spatial dependence at scales smaller than theminimum distance examined. The sill represents spatially independentvariance. The distance at which the sill is reached is called the range.The measurement points estimated within this range are spatiallyautocorrelated, whereas points outside this range are considered inde-pendent (Kosugi et al., 2007; Robertson, 1987). The spatial variation ofthe soil surface CO2 efflux was expressed by the coefficient of variationCV—which is defined as the ratio of the standard deviation to themean.

3. Results

3.1. Weather conditions

The mean daily air temperature and mean wind velocity (at 1.5 mheight) during the campaign period ranged between 7 and 24 °C,0.5–1.3 m/s at RE; 10–19 °C, 0.6–0.9 m/s at WS; 6–22 °C, 0.2–0.5 m/sat NO and 10–26 °C, 0.5–1.0 m/s at BK, respectively. During the mea-surement campaigns at RE, NO and BK no precipitation events occurred

297M. Acosta et al. / Geoderma 192 (2013) 295–303

Author's personal copy

while at WS measurements were stopped for a couple of hours due tovery light precipitation events (less than 1 cm per hour).

3.2. Soil surface CO2 efflux and temperature dependence

The highest measured soil surface CO2 efflux was found at WS andthe lowest at NO (Table 2). R10 (soil surface CO2 efflux normalized to10 °C temperature) was determined at each measured position persite using both approaches, the Q10 and Lloyd and Taylor equation.The four datasets generally resulted in good agreement (similarity upto 93%) when comparing calculated R10 per position at each site usingboth approaches. While residual analysis of R10 showed no significantdifference between approaches (dissimilarity less than 8%). However, atfew positions at NO and WS (6 and 5 positions, respectively) residualswere higher than 20%. In our study we used the exponential equationwithQ10 and soil temperature in order to determineR10 per eachpositionat the individual site. We choose the Q10 to determine the soil CO2 effluxtemperature sensitivity because it is the most widely used. The highestR10 was found at RE (4.09±1.23 μmol μmol CO2m−2 s−1) (mean±SD) followed by WS (3.49±1.14 μmol CO2 m−2 s−1), BK (2.66±1.04 μmol CO2 m−2 s−1) and NO (2.09±1.36 μmol CO2 m−2 s−1),(Fig. 1).

All the four investigated sites showed a positive dependence of thesoil surface CO2 efflux on soil temperature. The range of soil tempera-ture at 1.5 cm depth for the four sites during the measurement cam-paigns, Q10 and its coefficient of determination (r2) are shown inTable 2. The Q10 ranged between 2.0 and 2.3 among the sites thathave a relative narrow range. RE site was divided in the three areasdue to its uneven aged forest heterogeneity and complex terrain:dense young self-seedling forest, mature forest with gaps and waterlogged mature forest with gaps. The averaged Q10 value at RE was 2.2and was determined using all the positions in the three areas. WS, NOand BK sites presented a homogeneous soil surface and canopy.

At all the four sites a significant spatial variation of R10 was foundduring the measurement campaigns. The highest spatial variation ofthe soil surface CO2 efflux at 10 °C (R10) (expressed by the coefficientof variation) during our measurement campaigns was determined atNO (65%) and the lowest at RE (30%), WS and BK presented slightlydifferent CV 33% and 39%, respectively. On the basis of the obtainedR10 for each position, maps of spatial variation were created foreach site according to the grid design used (Fig. 2). The spatial auto-correlation for the soil CO2 efflux rate per site ranged between19.3 m at NO and 51.3 m at BK. The model parameters of thesemivariograms are presented in Table 3. The semivariograms fromthe R10 data show similar trend and spatial patterns in semivariancerelated to sampling distance (about 40 m) for RE, WS and BK(Fig. 4). On the other hand, NO showed an early decrease ofsemivariance with increasing distance (about 20 m) in comparisonwith the other three investigated sites, however, the increase ofsemivariance increased in all the sites with increasing distance. Thenugget in NO was 0, while for the other investigated sites it washigher. RE and WS presented similar sill (about 0.5) while NO thehighest and BK the lowest (Table 3). These differences could beexplained by site vegetation, understory, soil properties, topographyand climatic conditions of individual site.

RE site showed high soil CO2 efflux rates in the areas with matureforest with gaps (covered by blueberries and grasses) and denseyoung self-seedling forest, while low soil CO2 efflux rates wereobtained in the area of mature water logged forest with gaps(Fig. 3). At NO a relationship between stone depth and normalizedsoil respiration (R10) was found (r2=0.4) (Fig. 5). The same studywas carried out at WS and BK but no relationship between depth ofstones at individual measuring points and soil CO2 efflux was found.Thus, we confirm the dependence of R10 on depth of the stonedepth at NO and the water logged limitation of the soil CO2 efflux atRE. Furthermore, at RE no significant difference of the soil CO2 effluxbetween young forest and mature forest was found (Fig. 5).

3.3. Soil vegetation

At NO the soil vegetation layer was sparse and consisted mainly ofdwarf shrubs (V. myrtillus (L.), Vaccinium vitis-idaea (L.), Linnaeaborealis(L.), Luzula pilosa (L.)willd,), and mosses (Hylocomium spledens,Pleurózium schreberi (Brid.) Mitt., Ptilium crista-castrensis, Dicranumscoparium, Polytrichum commune). Hylocomium sp. and Pleuróziumsp. were two of the most frequent mosses. About 65% of the forestfloor was covered with varying proportions of moss and V. myrtillus(L.), approximately 20% was covered by moss, and the remaining 15%was a mosaic of V. myrtillus (L.) with sphagnum moss (Sphagnum)and bracken (Pteridium aquilinum L.). At WS ground vegetation wassparsewith occasional areas covered bymosses (Dicranum sp., Atrichumundulatom, P. commune,Hypnum sp.) and dwarf shrubs (V.myrtillus (L.).Small patches of lichens (Dryopteris carthusiana agg., Cladonia sp.) andbunchgrass (Deschampsia flexuosa (L.)) also covered the soil surface.Up to 25% of the forestfloorwas covered bymoss and 55% byV.myrtillus(L.). The remaining 30% was without vegetation or covered withsporadic patches of lichens and bunchgrass. RE site was characterizedby gaps between groups of older and younger trees. The soil vegetationwas mainly composed of grasses (D. flexuosa (L.)), Phleum sp.), horse-tails (Equisetum sp.) and shrubs (V. myrtillus (L.), Rhododendron sp.).RE presented a large area without understory in the older forest areas(about 45%). The remaining areas were covered by grasses (32%),shrubs (18%) and horsetails (5%). The soil vegetation in BK was mainlycharacterized by shrubs (V. myrtillus (L.)) distributed over all the forestarea (70%) and the remaining 30% is characterized by areas withoutunderstory, small patches with grasses (Calamagrotis villosa, D. flexuosa(L.)) and seldom mosses (Polytrichum sp., less than 2%).

Summarizing, all the four sites presented higher soil surface CO2

effluxes in positions covered by shrubs compared to positions withgrasses or mosses (Fig. 6). The highest CO2 effluxes rates were deter-mined at locations with blueberries. On the other hand, mosses mainlyrepresented by Dicranum sp., Polytrichum sp. and Ptilium sp. influencedthe spatial distribution of the soil surface CO2 efflux, more in the boreal

Table 2Range of soil temperature, soil surface CO2 efflux, Q10 and coefficient of determinationduring the measurement campaigns per investigated site.

Site Renon Wetzstein Norunda Bílý Kříž

Range of temperature at1.5 cm depth (°C)

7.6–26.1 6.1–15.9 9.7–19.8 10.5–24.5

Range of measured soil CO2

efflux (μmol CO2 m−2 s−1)0.67–13.83 1.05–15.02 0.12–12.62 1.38–14.03

Averaged Q10 and coefficientof determination (r2)

2.0 2.0 2.2 2.3(0.62) (0.55) (0.47) (0.58)

0

1

2

3

4

5

6

RE WS NO BK

R10

[µm

ol C

O2m

-2 s

-1]

Fig. 1. Mean normalized soil surface CO2 efflux (R10 — respiration at 10 °C) and SD(standard deviation) at the investigated sites. Renon (RE), Wetzstein (WS), Norunda(NO) and Bílý Kříž (BK).

298 M. Acosta et al. / Geoderma 192 (2013) 295–303

Author's personal copy

forest (NO) than in the other sites (WS, RE, BK) due to their commonoccurrence. Concerning tree vicinity (less than 1 m) to the measuredposition, at BK and WS the tree vicinity was more frequent (about26%) than in RE and NO due to management of planting and standdensity. Moreover, measured positions with tree vicinity in all thefour sites presented higher CO2 effluxes (up to 14%) compared to thepositions without tree vicinity, which may be explained by the influ-ence of the tree root system.

4. Discussion

Applied functions, Q10 andmodified Arrhenius function, presenteda good fit to the data at all the four sites. Residuals between the twofunctions showed a similarity at all the four sites with the exceptionof a few positions (up to 6 of the 80 positions ) at only two sites NOand WS. Nevertheless, in our calculation Q10 function was opted toexpress the temperature sensitivity of respiration because it is themost commonly applied to compare soil respiration activity betweendifferent sites.

4.1. Soil temperature sensitivity

Regarding Q10, there have been several published studies with dif-ferent ranges of Q10 values of soil temperature at different depthsranging from 1 to 4 (Davidson et al., 2006; Hanson et al., 2000;Raich and Schlesinger, 1992). Our determined Q10 per site was withinthis range, about 2 (Table 2). The relatively low coefficients of deter-mination of temperature dependence determined for all the four siteswere caused by the narrow temperature range during measurements(Table 2), excluding RE where the range of soil temperature was large(about 20 °C) due to the heterogeneity of the forest (uneven aged for-est with gaps between groups of older and younger trees and sparse

N

Fig. 2.Maps of the soil surface CO2 efflux spatial variation at 10 °C (R10) per investigated site. Where A is Bílý Kříž, B is Norunda, C is Renon and D is Wetzstein. The scale value for allthe four sites is in μmol CO2m−2 s−1, X and Y are the distance in meters. Soil surface CO2 efflux includes ground vegetation. R10 was determined as the average of 15 measurementscarried out at each measured position per site.

Table 3Semivariogram model parameters per site for the soil surface CO2 efflux at 10 °C (R10).

Site Model Nugget (C0) Sill (C0 +C) Range (m)

Renon Spherical 0.018 0.561 44.3Wetzstein Spherical 0.081 0.594 47.9Norunda Spherical 0.0 0.798 19.3Bílý Kříž Spherical 0.114 0.378 51.3

0

1

2

3

4

5

6

young mature waterlogged mature

R10

[µm

ol C

O2

m-2 s

-1]

Fig. 3. Spatial variation of the soil surface CO2 efflux at 10 °C (R10) at the Renon sitedivided by areas: young forest, mature forest with gaps and waterlogged mature forestwith gaps.

299M. Acosta et al. / Geoderma 192 (2013) 295–303

Author's personal copy

understory vegetation). Despite our results one essential questionconcerning the Q10 parameter and soil temperature depth for the pre-cise and correct determination of the soil surface CO2 efflux at differ-ent sites is still unclear. Pavelka et al. (2007) pointed out that soiltemperature near the soil surface better represented the dynamic ofthe soil CO2 efflux. It could be explained by the fact that the highestamount of soil organic carbon, and thus respiration potential, is veryoften located at the upper soil layer. Previous experiments at theNO site (Morén and Lindroth, 2000) found a high correlation betweensoil respiration and soil temperature measured at 5 cm depth. How-ever, the authors measured only at two depths of 5 and 10 cm andit was not specified whether the authors took into account the thick-ness of the moss layer. Measurements of the soil respiration at RE(Rodeghiero and Cescatti, 2005) showed a good relationship betweensoil respiration and the only soil temperature measured at 10 cm

depth. Recent studies at WS (Moyano et al., 2008) showed a strongrelationship between soil respiration and soil temperature at 2 cmdepth, and thus confirming that the most suitable depth of soil tem-perature measuring point corresponding to the measured soil CO2

efflux is close to the soil surface.

4.2. Factors influencing spatial and short-term temporal soil CO2 effluxes

The variability of the normalized respiration (R10) at the investi-gated sites was probably driven by different factors. In several studies,soil temperature explains most of the temporal and spatial variationsof soil respiration (Buchmann, 2000; Janssens et al., 2001). However,the main factors influencing the spatial variation of soil respirationare vegetation composition of under canopy and root biomass distri-bution (Law et al., 2001; Moyano et al., 2008; Rodeghiero andCescatti, 2006; Widén, 2002). Among these factors, litter fall, amountof litter on soil, soil microbial biomass and soil pH are soil descriptorswhich are supposed to be related to spatial variation of soil respira-tion (Hanson et al., 2000). At RE the highest spot of soil respirationwas related to the patch covered by blueberries and grasses. On theother hand, particularly high spots were found in the area of youngself-seedling trees. Thus, it could be explained by the participationof root biomass and microbial biomass to total soil respiration, in ayoung forest a great portion of living fine root biomass (up to 50%)is located in the upper 10 cm of the soil (Todd and Shepard, 1995).Patches with lower soil respiration values at RE were related to theinfluence of soil moisture (water logged) limiting aerobic respiration.Kosugi et al. (2007) reported similar influence of soil moisture on thesoil CO2 efflux spatial variation on a tropical rain forest as in ourstudy, but in their conditions temporally the soil CO2 efflux waslower when soil was dry. They also suggested that a physical factorsuch as the restriction of gas diffusivity in a water-saturated soil isnot the primary cause of the low soil respiration rates in wetterplaces, and that several physical, biological and chemical properties

Fig. 4. Variogram of the semivariance computed from the data of the soil CO2 efflux at 10 °C (R10) for the individual investigated sites.

0

1

2

3

4

5

6

7

8

0 5 10 15 20 25 30

Depth [cm]

R10

[µm

ol C

O2

m-2 s

-1]

y = 3.861 * exp(-exp(1.639 - 0.208 * x))

R2 = 0.396

Fig. 5. Relationship between the stone depth and soil surface CO2 efflux at 10 °C (R10)at measuring points at the Norunda site.

300 M. Acosta et al. / Geoderma 192 (2013) 295–303

Author's personal copy

co-varying with soil water drive the bipolar spatial and temporal var-iations of soil respiration rate. Furthermore, the soil respirationresponse to moisture is the result of several processes – includingosmotic stress, diffusion and oxygen limitations – that combined toproduce a net effect on the rate of carbon decomposition (Moyanoet al., 2012).

WS and BK sites are characterized by a homogeneous soil surfacecover. In both sites, ground vegetation (mainly blueberries) is sparsewith a few gaps with no vegetation and inWS occasional areas coveredby moss. WS showed low spatial heterogeneity of soil R10 (CV 33%), itcould be explained by the flat terrain on the top of the hill and homog-enous mature forest stand. Higher CVs of soil R10 spatial heterogeneityat BK (39%) could be explained with high portions of gravel in soil(30–40%). Another explanation for the difference in CV between WSand BK was related to stand age and density. The forest stand at BKwas younger and denser than the WS forest stand. Thus, in BK weexpected a higher ratio of fine root biomass influencing spatial varia-tion. Claus and George (2005) reported a clear effect of stand age onstanding fine root biomass, with the highest fine root biomass in adultbut not mature stands. Another possible reason for differences in CVof soil R10 spatial heterogeneity between BK and WS could be theratio of rhizosphere at each site. Tang et al. (2005) suggested that thesoil CO2 efflux is largely controlled by rhizosphere respiration in areascovered by trees, while Fang et al. (1998) found an increase in the soilCO2 efflux due to an increase in fine root biomass, litter and humusamount on the forest floor.

The lowest rates of soil respiration but the highest spatial varia-tion were observed at NO. It is a complex site where the forest flooris mainly covered by boulders, rocks (of a diameter up to 1.5 m)and understory vegetation is dominated by mosses and blueberries.NO is also characterized by occasionally water logged patches. Atthese places deeper soil profiles were detected. Deep soil profile(more than 25 cm) together with a higher soil moisture in thesepositions resulted in increased rates of soil respiration. On theother hand, low respiration rates were observed at places with boul-ders. These places were characterized by a very shallow soil depth(about 2 cm). Contrary to RE, at NO higher soil moistures increasedthe soil CO2 efflux. The relationship between soil moisture and soilrespiration is known to be variable and depends on site specificconditions. Reichstein et al. (2003) pointed out that three phases ofthe response of soil respiration to soil moisture can be identified:(1) when soils are relatively dry, metabolic activity increases stronglywith water availability, (2) there is a broad range of near optimumsoil water content where changes in soil moisture only have a littleeffect, if any, on soil respiration and (3) above field capacity and towardsaturation, oxygen deficiencies inhibit aerobic respiration. Thus, at NO ahigher soil moisture was considered as optimum in the mentionedplaces and site specific characteristics.

The ground vegetation, besides temperature, was an importantfactor influencing soil CO2 spatial variation in all the investigatedsites (Figs. 3 and 6). At the same time the magnitude of this influencedepends on the physiological state of the vegetation, distribution andquantity of the soil vegetation. In our study, shrub vegetationinfluenced the soil CO2 efflux spatial variation more than grassesand mosses in all the investigated sites due to its occurrence. Eventhough in our study we determined some factors influencing spatialsoil surface CO2 efflux, studies concerning root distribution and den-sity, litter fall, soil texture, physiological state of the ground vegeta-tion, etc. at the individual sites are needed to properly assess factorsinfluencing the spatial distribution of the soil surface CO2 efflux.

Rayment and Jarvis (2000) pointed out that spatial heterogeneityin soil respiration increased with a distance greater than 1 m, but didnot increase substantially with further increased distance. Kosugi etal. (2007) in a study in a tropical rain forest reported soil respirationvariability up to 100% at locations of 0.5 m. In our study, soil CO2

efflux spatial variation also increased with an increased distance inall the investigated sites. However, it is necessary to point out thatthe systematic grid at NO was of 10 m, BK and WS of 5 m while RErandomly according to vegetation and topography. Thus these condi-tions influenced our semivariance results. However, our determinedCV of the soil CO2 efflux represents the specific conditions per site.

5. Conclusions

The total soil surface CO2 efflux at the four different forest siteswas estimated in order to determine spatial variation and identifypossible factors influencing this variation. At RE no significant differ-ence of soil R10 between the young forest and mature forest wasfound due to characteristics of understory vegetation. Moreover, alimitation of soil CO2 efflux was found at water logged parts of themature forest. At NO spatial variability of R10 was mostly explainedby the contrast in the soil depth, ground vegetation and water avail-ability. The lowest rates of soil respiration were found in placeswith a high ratio of soil stones and shallow soil depth. The highestrates were found in the terrain depressions with a deeper soil profile,close to the waterlogged areas. The soil surface CO2 effluxes at WSand BK were characterized as homogeneous and mostly influencedby stand age and structure. Relatively low R10 and its high coefficientof variation at BK could be caused by a high ratio of soil rock frag-ments, which indicates more heterogeneous distribution of CO2

sources in the soil. Tree vicinity at the sample position influencesthe total soil surface CO2 efflux. In our study, soil temperature, soilmoisture and floor vegetation were the main factors influencing spa-tial and short-term temporal variabilities of soil CO2 efflux dependingon site specific conditions. Although our study had a limited durationof observations to completely determine spatial and short-term

Fig. 6. Mean CO2 efflux by main vegetation composition at the investigated sites. Renon (RE), Wetzstein (WS), Norunda (NO) and Bílý Kříž (BK). Bar indicates standard deviation.Number of asterisks above the columns denotes significantly different mean values of individual vegetation components between sites on alpha level=0.05.

301M. Acosta et al. / Geoderma 192 (2013) 295–303

Author's personal copy

temporal variabilities of the soil surface CO2 efflux, the presented datagives a general overview of the soil surface CO2 efflux using the samemethodology and equipment and factors influencing it in differentgeographically located forest ecosystems.

Acknowledgment

Special thanks are expressed to Jan Novotný and Victor Rolo for theassistance with Geostatistics work. We also thank two anonymousreviewers for their valuable remarks and constructive comments in theprevious version of the manuscript. This work was supported by grantsMSMT OC08021, MZP SP/2d1/70/08 and SP/2d1/93/07 and the researchintention AV0Z60870520 from the Academy of Sciences of the CzechRepublic. This article is an output of the CzechGlobe Centre that is beingdeveloped within the OP RDI and co-financed by EU funds and the StateBudget of the Czech Republic (Project: CzechGlobe — Centre for GlobalClimate Change Impacts Studies, Reg. No. CZ.1.05/1.1.00/02.0073).Manuel Acosta was supported by the Greenflux project: MTDK-CT-2006-042445 from EC during the preparation of this manuscript.

References

Acosta, M., Janouš, D., Marek, M.V., 2004. Soil surface CO2 fluxes in a Norway sprucestand. Journal of Forest Science 50 (12), 573–578.

Anthoni, P.M., Knohl, A., Rebmann, C., Freibauer, A., Mund, M., Ziegler, W., Kolle, O.,Schulze, E.D., 2004. Forest and agricultural land-use-dependent CO2 exchange inThuringia. Global Change Biology 10, 2005–2019.

Bowden, R., Davidson, E., Savage, K., Arabia, C., Steudle, P., 2004. Chronic nitrogenadditions reduce total soil respiration and microbial respiration in temperate forestsoils at the Harvard forest. Forest Ecology and Management 196, 43–56.

Buchmann, N., 2000. Biotic and abiotic factors controlling soil respiration rates in Picea abiesstands. Soil Biology and Biochemistry 32, 1625–1635.

Chojnicki, B.H., Michalak, M., Acosta, M., Juszczak, R., Augustin, J., Drösler, M., Olejnik, J.,2010. Measurements of carbon dioxide fluxes by chamber method at the Rzecinwetland ecosystem, Poland. Polish Journal of Environmental Studies 19 (2),283–291.

Claus, A., George, E., 2005. Effect of stand age on fine root biomass and biomass distributionin three European forest chronosequences. Canadian Journal of Forest Research 35 (7),1617–1625.

Davidson, E.A., Belk, E., Boone, R.D., 1998. Soil water content and temperature asindependent or confounded factors controlling soil respiration in a temperaturemixed hardwood forest. Global Change Biology 4, 217–227.

Davidson, E., Savage, K., Verchot, L.V., Navarro, R., 2002. Minimizing artifacts and biases inchamber-basedmeasurements of soil respiration. Agricultural and ForestMeteorology113, 21–37.

Davidson, E., Jansens, I., Luo, Y., 2006. On the variability of respiration in terrestrialecosystems: moving beyond Q10. Global Change Biology 12 (2), 154–164.

Epron, D., Farque, L., Lucot, E., Badot, P., 1999. Soil CO2 efflux in a beech forest: dependenceon soil temperature and soil water content. Annals of Forest Science 56, 221–226.

Fang, C., Moncrieff, J.B., 1996. An improved dynamic chamber technique for measuringCO2 efflux from the surface of soil. Functional Ecology 10, 297–305.

Fang, C., Moncrieff, J.B., Gholz, H.L., Clark, K.L., 1998. Soil CO2 efflux and its spatialvariation in a Florida slash pine plantation. Plant and Soil 205, 135–146.

Feigenwinter, C., Bernhofer, C., Eichelmann, U., Heinesch, B., Hertel, M., Janouš, D.,Kolle, O., Lagergren, F., Lindroth, A., Minerbi, S., Moderow, U., Model, M.,Montagnani, L., Queck, R., Rebmann, C., Vestin, P., Yernaux, M., Zeri, M., Ziegle, W.,Aubinet, M., 2008. Comparison of horizontal and vertical advection CO2 fluxes atthree forest sites. Agricultural and Forest Meteorology 148, 12–24.

Hanson, P.J., Edwards, N.T., Garten, C.T., Andrews, J.A., 2000. Separating root and soilmicrobial contributions to soil respiration: a review of methods and observations.Biogeochemistry 48, 115–146.

Heinemeyer, A., Di Bene, C., Lloyd, A.R., Tortorella, D., Baxter, R., Huntley, B., Gelsomino, A.,Ineson, P., 2011. Soil respiration: implications of the plant-soil continuum and respi-ration chamber collar-insertion depth on measurement and modelling of soil CO2

efflux rates in three ecosystems. European Journal of Soil Science 62, 82–94.Hengl, T., Heuvelnik, G.B.M., Stein, A., 2004. A generic framework for spatial prediction

of soil variables based on regression-kriging. Geoderma 120, 75–93.Herbst, M., Prolingheuer, N., Graf, A., Huisman, J.A., Weihermüller, J., Vanderborght, H.,

vereecken, H., 2010. Mutivariate conditional stochastic simulation of soil heterotropicrespiration at plot scale. Geoderma 160, 78–82.

Janssens, I.A., Meiresonne, L., Ceulemans, R., 1999. Mean soil CO2 efflux from a mixedforest. temporal and spatial integration. In: Ceulemans, R., et al. (Ed.), ForestEcosystem Modelling, Upscaling and Remote Sensing. SPB Academic Publishing,The Hague, pp. 19–33.

Janssens, I.A., Kowalski, A.S., Longdoz, B., Ceulemans, R., 2000. Assessing forest soil CO2

efflux: an in situ comparison of four techniques. Tree Physiology 20, 23–32.

Janssens, I.A., Lankreijer, H., Matteucci, G., et al., 2001. Productivity overshadowstemperature in determining soil and ecosystem respiration across European forest.Global Change Biology 7, 269–278.

Juszczak, R., Acosta, M., Olejnik, J., 2012. Comparison of daytime and nighttime ecosystemrespiration measured by the close chamber technique on temperate mire in Poland.Polish Journal of Environmental Studies 21 (3), 643–658.

Korari, P., Kulmala, L., Pumpane, J., Launiainen, S., Ilvesniemi, H., Hari, P., Nikimma, E.,2009. CO2 exchange and component CO2 fluxes of a boreal Scots pine forest. BorealEnvironment Research 14, 761–783.

Kosugi, Y., Mitani, T., Itoh, M., Noguchi, S., Tani, M., Naoko, S., Takanashi, S., Ohkubo, A.R.,Rahim Nik, A., 2007. Spatial and temporal variation in soil respiration in a SoutheastAsia tropical rainforest. Agricultural and Forest Meteorology 147, 35–47.

Law, B.E., Kelliher, F.M., Baldocchi, D.D., Anthoni, P.M., Irvine, J., Moore, D., Van Tuyl, S.,2001. Spatial and temporal variation in respiration in a young ponderosa pine forestduring a summer drought. Agricultural and Forest Meteorology 110, 27–43.

Le Dante, V., Epron, D., Dufrene, E., 1999. Soil CO2 efflux in a beech forest: comparisonof two closed dynamic systems. Plant and Soil 214, 125–132.

Livingston, G.P., Hutchinson, G.L., 1995. Enclosure-based measurement of trace gasexchange: applications and sources of error. In: Matson, P.A., Harriss, R.C. (Eds.),Biogenic Trace Gases: Measuring Emissions from Soil and Water. Blackwell Science,Cambridge, pp. 14–50.

Lloyd, J., Taylor, J.A., 1994. On the temperature dependence of soil respiration. FunctionalEcology 8, 15–323.

Marcolla, B., Cescatti, A., Montagnani, L., Manca, G., Kerschbaumer, G., Minerbi, S., 2005.Importance of advection in the atmospheric CO2 exchange of an alpine forest.Agricultural and Forest Meteorology 130, 193–206.

Morén, A.S., Lindroth, A., 2000. CO2 exchange at the floor of a boreal forest. Agriculturaland Forest Meteorology 101, 1–14.

Moyano, F.E., Werner, L.K., Rebmann, C., 2008. Soil respiration fluxes in relation tophotosynthetic activity in broad-leaf and needle-leaf forest stands. Agriculturaland Forest Meteorology 148, 135–143.

Moyano, F.E., Vasilyeva, N., Bouckaert, L., Cook, F., Craine, J., Yuste, J.C., Don, A., Epron, D.,Formanek, P., Franzluebbers, A., Ilstedt, U., Kätterer, T., Orchard, V., Reichstein, M.,Rey, A., Ruamps, L., Subke, J.A., Thomsen, I.K., Chenu, C., 2012. The moisture responseof soil heterotrophic respiration: interaction with soil properties. Biogeosciences 9,1173–1182.

Norman, J.M., Garcia, R., Verma, S.B., 1992. Soil surface CO2 fluxes and the carbon budgetof a grassland. Journal of Geophysical Research 97, 18845–18853.

Pavelka, M., Acosta, M., Marek, M.V., Kutsch, W., Janouš, D., 2007. Dependence of Q10

values on the depth of the soil temperature measuring point. Plant and Soil 292,171–179.

Pumpanen, J., Kolari, P., Ilvesniemi, H., Minkkinen, K., Vesala, T., Niinistö, S., Lohila, A.,Larmola, T., Morero, M., Pihlatie, M., Janssens, I., Yuste, J.C., Grünzweig, J.M., Reth, S.,Subke, J.A., Savage, K., Kutsch, W., Østreng, G., Ziegler, W., Anthoni, P., Lindroth, A.,Hari, P., 2004. Comparison of different chamber techniques for measuring soil CO2

efflux. Agricultural and Forest Meteorology 123, 159–176.Raich, J.W., Schlesinger, W.H., 1992. The global carbon dioxide flux in soil respiration

and its relationship to vegetation and climate. Tellus 44, 81–99.Rayment, M.B., Jarvis, P.G., 2000. Temporal and spatial variation of soil CO2 efflux in a

Canadian boreal forest. Soil Biology and Biochemistry 32, 35–45.Reichstein, M., Rey, A., Freibauer, A., Tenhunen, J., Valentini, R., Banza, J., Casals, P.,

Cheng, Y., Grünzweig, J.M., Israel, R., Irvine, J., Joffre, R., Law, B.E., Loustau, D.,Miglietta, F., Oechel, W., Ourcival, J.M., Pereira, J.S., Peressotti, A., Ponti, F., Qi, Y.,Rambal, S., Rayment, M., Romanya, J., Rossi, F., Tedeschi, V., Tirone, G., Xu, M.,Yakir, D., 2003.Modeling temporal and large-scale spatial variability of soil respirationfrom soil water availability, temperature and vegetation productivity indices. GlobalBiogeochemical Cycles 17, 1104 (15 pp.).

Robertson, G.P., 1987. Geostatistics in ecology: interpolating with known variance.Ecology 68, 744–748.

Robertson, G.P., Wedin, D., Groffman, P.M., Blair, J.M., Holland, E.A., Nedelhoffer, K.J.,Harris, D., 1999. Soil Carbon and Nitrogen Availability. Nitrogen Mineralization,Nitrification and Soil Respiration Potentials. In: Robertson, G.P., Coleman, D.C.,Bledsoe, C.S., Sollins, P. (Eds.), Standard soil methods for long-term ecologicalresearch. Oxford University Press, New York, pp. 258–271.

Rodeghiero, M., Cescatti, A., 2005. Main determinants of forest soil respiration along anelevation temperature gradient in the Italian Alps. Global Change Biology 11,1024–1041.

Rodeghiero, M., Cescatti, A., 2006. Indirect partitioning of soil respiration in a series ofevergreen forest ecosystems. Plant and Soil 284, 7–22.

Rodeghiero, M., Cescatti, A., 2008. Spatial variability and optimal sampling strategy ofsoil respiration. Forest Ecology and Management 225, 106–112.

Ryan, M., Law, B.E., 2005. Interpreting, measuring, and modeling soil respiration.Biogeochemistry 73, 3–27.

Ryan, M.G., Hubbard, R.M., Pongracic, S., Raison, R.J., McMurtie, R.E., 1996. Foliage fine-root, woody-tissue and stand respiration in Pinus radiata in relation to nitrogenstatus. Tree Physiology 16, 333–343.

Saiz, G., Green, C., Butterbach-Bahl, K., Kiese, R., Avitabile, V., Farrell, E.P., 2006. Season-al and spatial variability of soil respiration in four Sitka spruce stands. Plant andSoil 287, 161–176.

Stoyan, H., De-Polli, H., Böhm, S., Robertson, G.P., Paul, E.A., 2000. Spatial heterogeneityof soil respiration and related properties at the plant scale. Plant and Soil 222,203–214.

Tang, J., Baldocchi, D.D., Xu, L., 2005. Tree photosynthesis modulates soil respiration ona diurnal scale. Global Change Biology 11, 1298–1304.

Taylor, B.R., Parkinson, D., Parsons, W.F.J., 1989. Nitrogen and lignin content as predictorsof litter decay rates. A microcosm test. Ecology 70, 97–104.

302 M. Acosta et al. / Geoderma 192 (2013) 295–303

Author's personal copy

Todd, S.F., Shepard, M., 1995. Fine root biomass, distribution, and production in youngpine-hardwoods stand. New Forest 10 (1), 99–110.

Tuomi, M., Vanhala, P., Karhu, K., Fritze, H., Liski, J., 2008. Heterotrophic soil respiration—

comparison of different models describing its temperature dependence. EcologicalModelling 211, 182–190.

Urban, O., Janouš, D., Acosta, M., Czerný, R., Marková, I., Navrátil, M., Pavelka, M.,Pokorný, R., Šprtová, M., Zhang, R., Špunda, V., Grace, J., Marek, M.V., 2007. Eco-physiological controls over the net ecosystem exchange of mountain sprucestand. Comparison of the response in direct vs. diffuse solar radiation. GlobalChange Biology 13 (1), 157–168.

Valentini, R., Matteucci, G., Dolman, A.J., Schulze, E.-D., Rebmann, C., Moors, E.J.,Granier, A., Gross, P., Jensen, N.O., Pilegaard, K., Lindroth, A., Grelle, A., Bernhofer,C., Grünwald, T., Aubinet, M., Ceulemans, R., Kowalski, A.S., Vesala, T., Rannik, Ü.,Berbigier, P., Loustau, D., Gudmundsson, J., Thorgeirsson, H., Ibrom, A.,

Morgenstern, K., Clement, R., Moncrieff, J., Montagnani, L., Minerbi, S., Jarvis, P.G.,2000. Respiration as the main determinant of carbon balance in European forests.Nature 404, 961–865.

Van’t Hoff, J.H., 1898. Lectures on theoretical and Physical Chemistry. Part I ChemicalDynamics (translated by R.A. Lehfeldt). Edward Arnold, London, pp. 224–229.

Vodkin, D., Kastelec, D., Pfanz, H., Macek, I., Turk, B., 2006. Small-scale spatial variationin soil CO2 concentration in a natural carbon dioxide spring and some related plantresponses. Geoderma 133, 309–319.

Welles, J.M., Demetriades-Shah, T.H., Mcdermitt, D.K., 2001. Considerations for mea-suring ground CO2 efflux with chambers. Chemical Geology 177, 3–13.

Widén, B., 2002. Seasonal variation in forest-floor CO2 exchange in a Swedish coniferousforest. Agricultural and Forest Meteorology 111, 283–297.

303M. Acosta et al. / Geoderma 192 (2013) 295–303