Translator Disclaimer
1 February 2013 Modeling Canopy CO2 Exchange in the European Russian Arctic
Author Affiliations +
Abstract

In this study, we use the coupled photosynthesis-stomatal conductance model of Collatz et al. (1991) to simulate the current canopy carbon dioxide exchange of a heterogeneous tundra ecosystem in European Russia. For the parameterization, we used data obtained from in situ leaf level measurements in combination with meteorological data from 2008. The modeled CO2 fluxes were compared with net ecosystem exchange (NEE), measured by the eddy covariance technique during the snow-free period in 2008.

The findings from this study indicated that the main state parameters of the exchange processes were leaf area index (LAI) and Rubisco capacity (Vcmax). Furthermore, this ecosystem was found to be functioning close to its optimum temperature regarding carbon accumulation rates. During the modeling period from May to October, the net assimilation was greater than the respiration, leading to a net accumulation of 58 g C m-2. The model results suggest that the tundra ecosystem could change from a carbon sink to a carbon source with a temperature rise of only 2–3 °C. This is due to the fact that, in the continental Arctic, a global warming of a few degrees might restrict the net assimilation, due to high temperatures, whereas the respiration is predicted to be enhanced. However, future changes in vegetation composition and growth, along with acclimation to the new thermal regime, might facilitate the assimilation to counterbalance the carbon losses.

Introduction

Terrestrial ecosystems in the Arctic play an important role in the global carbon cycle, because they contain at least one-third of the global terrestrial carbon, whilst covering 25% of the global land area (McGuire et al., 2009; Tarnocai et al., 2009). This region is predicted to undergo considerable future climate change, which might alter the net carbon exchange between the land and the atmosphere (IPCC, 2007; Qian et al., 2010).

To make predictions about the future net carbon exchange, it is necessary, however, to assess the present-day fluxes in terms of their controlling mechanisms. Successful modeling of net ecosystem carbon dioxide exchange in the Arctic relies on good estimates of the two major fluxes: carbon assimilation and ecosystem respiration. One challenge for modeling the carbon exchange on a regional scale is that the arctic landscape is a patchwork of differing plant species compositions, microtopography, water-table depth, and soil properties (Bliss and Matveyeva, 1992). Within the global network of micrometeorological flux measurements (FLUXNET), numerous ecosystem studies have been conducted in the boreal and temperate regions (Baldocchi et al., 2001), whereas detailed information about the dominant processes in the arctic region is comparably scarce.

Still, a number of models have been applied on a plot, landscape, or even regional scale to simulate the present-day carbon exchange in the Arctic. They vary in complexity, ranging from empirical models describing responses to environmental factors (e.g. Williams et al., 2006) to complex mechanistic or process-based models (e.g. Ecosys; Grant et al., 2003) (Fig. 1).

Empirical models use the combined representation of photosynthetic irradiance-response and temperature-sensitive respiration to scale carbon exchange from plot to landscape scale (e.g., Fox et al., 2008; Vourlitis et al., 2000), or for the interannual or regional comparison of the exchange (e.g., Lafleur and Humphreys, 2008; Laurila et al., 2001).

These regression-based models can be expanded to a regional scale through the use of multiple observations. Using such an empirical regression-based model, Zamolodchikov and Karelin (2001) simulated the carbon fluxes for the whole Russian tundra by evaluating the relation of the gross primary production to light, temperature, and seasonal phytomass dynamics. Requiring seasonal recalibration, empirical models provide useful information for regional or interannual comparison and the productivity of specific species. However, they are not intended for evaluating future responses of carbon budgets to changing climates, because the temperature response of photosynthesis is not taken into account.

The mechanistic and process-based models, on the other hand, include the temperature dependence of the processes involved, and are therefore suited to predicting future carbon exchange. These models attempt to represent specific interacting ecological process dynamics using a more detailed set of environmental and physiological parameters.

A mechanistic model developed by Farquhar et al. (1980) and further elaborated by Collatz et al. (1991) simulates the gas exchange of a canopy in free air using the big-leaf approach. This model, hereafter called the Collatz model, couples the modeling of net photosynthesis from environmental and leaf physiological parameters with a stomatal conductance model.

When extrapolating canopy models to landscape levels, a detailed knowledge of soil properties and hydrological conditions is required. This was demonstrated by models such as the soil-plant-atmosphere (SPA) continuum (Williams et al., 1996; Jarvis and McNaughton, 1986) and the Met Office Surface Exchange Scheme (MOSES) (Cox et al., 1999), which include the canopy and soil water availability for modeling the carbon assimilation rate. These models use the multilayer approach, separating the canopy and soil into several layers exhibiting different properties. Rennermalm et al. (2005) used MOSES to model the net carbon exchange in a high arctic site in Greenland using a four soil layer approach, while Williams et al. (2000) successfully tested the SPA model in northern Alaska using a two-layer canopy approach. Other process-based models (right-hand column in Fig. 1) have been applied for regional Pan-Arctic estimates of present and future carbon exchange processes, and these demand even more input parameters.

FIGURE 1.

Schematic overview of different models and the main input parameters used in the Arctic.

f01_50.jpg

As only limited knowledge about soil and hydrological conditions in Northeast European Russia is available at present, in this study we evaluate the canopy carbon dioxide exchange in combination with estimates of soil respiration derived from the generalized temperature function described by Lloyd and Taylor (1994). This study focuses on the snow-free period from mid-May to the beginning of October. We recognize the importance of wintertime fluxes, because they can be substantial (Fahnestock et al., 1999; Bjorkman et al., 2010), but wintertime fluxes in remote areas can be hard to quantify. Therefore, the study period is restricted to the interval from early spring until late autumn.

We parameterized and tested the Collatz model for simulating the present-day net assimilation rate in the European Russian tundra, which covers an area of 232,000 km2 (Zamolodchikov and Karelin, 2001). For the parameterization, site-specific input parameters are required. One important input parameter is Rubisco capacity (V cmax). This study benefits from the fact that the V cmax value was measured in situ on the representative plant species. The modeled CO2 fluxes were compared to net ecosystem exchange (NEE) measurements obtained using the eddy covariance technique. We aim to identify the controlling parameters of net assimilation and to develop a tool for upscaling the results to a regional level. Furthermore, we attempt to analyze the temperature dependence in order to provide prospects for future carbon exchange mechanisms.

Material and Methods

SITE DESCRIPTION

This study was conducted in the discontinuous permafrost zone in the subarctic tundra of Komi Republic, Northeast European Russia. The measurement site was located near the village Seida (67°03′N, 62°56′E, 100m a.s.l.) (Fig. 2).

FIGURE 2.

Location of Seida study site (67°03′N, 62°56′E) in the Komi Republic (black area), Russia.

f02_50.jpg

Long-term average air temperatures for 1980–1999 from neighboring Vorkuta station (67°48′N, 64°01′E, 172 m a.s.l.) show a mean annual temperature in the region of -5.7 °C, with the coldest month being January ( -20.1 °C) and the warmest July (12.8 °C). Long-term average precipitation records for 1980–1999 for Salekhard station (66°32′N, 66°36′E, 66 m a.s.l.) show an annual precipitation of 454 mm (data from Komi Republican Centre for Hydrometeorological and Environmental Monitoring).

The study site is a sedge-dwarf shrub tundra type, according to Bliss and Matveyeva (1992). Large parts of the landscape are shrub tundra dominated by Betula nana, Salix spp., and dwarf shrubs of heath species, Empetrum. nigrum., Ledum palustre, and Vaccinium spp. (Table 1). The willow stands grow to a height of up to 1.5 m on low-lying parts of the landscape, which act as water drains in the partially frozen landscape. The remaining parts of the landscape are dominated by peat plateaus, thermokarst lakes, and narrow fens with high water tables, which surround the peat plateaus and thermokarst lake edges. The most characteristic vegetation in the fen is made up of Carex spp., Eriophorum spp., and mosses.

TABLE 1

Composition of land surface types with the dominant species in a 500 m radius around the measurement mast.

t01_50.gif

COUPLED PHOTOSYNTHESIS-STOMATAL CONDUCTANCE MODEL

The Collatz model estimates the leaf gross photosynthesis rate (Al ) as the minimum process of three potential capacities, which are limited by different factors:

e01_50.gif

JE is the potential photosynthetic rate dependent on light:

e02_50.gif

where a is the absorptivity of the photosynthetic photon flux density (PPFD) for leaf material, a is the initial slope of the light response curve, Q is the PPFD in µmol m-2 s-1 C i , is the internal CO2 pressure in Pa and Γ* is the compensation point in Pa.

Jc is the potential photosynthetic rate dependent on Rubisco capacity:

e03_50.gif

where V cmax is the maximum catalytic capacity of Rubisco per leaf area unit in µmol m-2 s-1 , Kc and K0 are the Michaelis constant for CO2 and the competitive inhibition constant for O2 with respect to CO2 in the Rubisco reaction.

Js is the potential rate of export or utilization of the assimilated products and is proportional to the Rubisco capacity:

e04_50.gif

where CJs is a coefficient originally set to 0.5 (Collatz et al., 1991).

The temperature dependent parameters, i.e. fi01_50.gif V cmax, Kc , K0 , and Rld , are scaled in accordance with Collatz et al. (1991) using the Q10 approach. The Q10 values were chosen according to Collatz et al. (1991), except the Q10 value for V cmax, which was taken from Soegaard and Nordstroem (1999) and Cox et al. (1999) (Table 2).

In order to simulate realistic rates at high and low temperatures, a gradual temperature inhibition of V cmax at high temperatures (s1) and of Js at low temperatures (s2) is included (Alton and Bodin, 2010; Dangetal., 1998).

TABLE 2

Parameters and constants used in the model.

t02_50.gif

The dark respiration was calculated by:

e05_50.gif

where r is a scaling factor derived from leaf level estimates of Rld and V cmax (see below).

The bulk stomatal conductance to water vapor per ground area unit (gsw) is estimated using the Ball, Woodrow, and Berry model (BWB):

e06_50.gif

where An is the net photosynthesis in mol m-2 s-1, hs is the relative humidity at the surface of the leaf, Cs is mole fraction of CO2 at the surface of the leaf, and m and b are the slope and intercept of the BWB function, which can be obtained by linear regression analysis of data from gas exchange studies (Ball et al., 1987; Collatz et al., 1991).

A complete parameter list, including their temperature dependence, is shown in Table 2.

The photosynthesis rate of the upper sunlit leaves (Al) is scaled to the canopy (A11 ) using a canopy factor:

e07_50.gif

where Al is the gross leaf photosynthesis rate, An is the net photosynthesis rate of the canopy, and Rld is the dark respiration per unit leaf area, including the respiration that occurs during both day and night. Π is the canopy factor describing the area of sunlit leaves per ground area unit using Beer's law of light extinction (Sellers et al., 1992):

e08_50.gif

where LAI is the leaf area index, and k is the extinction coefficient (Kull and Jarvis, 1995). The extinction coefficient relates to the solar elevation angle, β, by k = 0.5/sin β (Baldocchi, 1994). In this model, we only consider the sunlit leaves, as the number of shaded leaves is small due to the low LAI.

EDDY COVARIANCE MEASUREMENTS

For the validation of the model, the modeled canopy net assimilation rates are compared to the measured fluxes from the eddy covariance technique. The net ecosystem exchange (NEE) between the surface and the atmosphere was continuously measured using the micrometeorological eddy covariance method (Aubinet et al., 2000; Baldocchi, 2003). The system was set up with an R3 ultrasonic anemometer (Gill Instruments, U.K.) and an LI-7500 open-path (OP) CO2/H2O infrared gas analyzer (IRGA) (LI-COR Inc., U.S.A.) mounted at 3.95 m height on an extendable mast. The OP analyzer was tilted in a northern direction to minimize water coverage of the lower lens during rain events. Measurements were taken at a frequency of 10 Hz from 18 May until 6 October 2008. The raw data were processed using Alteddy software (version 3.5, Alterra, University of Wageningen, The Netherlands;  http://www.climatexchange.nl/projects/alteddy/), which is based on EUROFLUX methodology discussed in Aubinet et al. (2000). The correction for artificial heating of the open-path system, as proposed by Burba et al. (2008), was included in the calculation of the half-hourly NEE. A detailed description of the data processing and footprint modeling is given in Marushchak et al. (2012).

The net assimilation rate (AnEC ) was estimated as:

e09_50.gif
e10_50.gif

where NEP is the net ecosystem production, which equals the negative NEE, as measured by the eddy covariance, Rsoil is the soil respiration, REco is the ecosystem respiration, determined during nighttime, and Rd is the modeled canopy dark respiration. The respiration rates (Ri ) were calculated using an Arrhenius type of function elaborated by Lloyd and Taylor (1994):

e11_50.gif

where R10,i is the respiration rate at 10 °C, Ts is the soil temperature in °C at 5 cm, and i represents either ecosystem, soil, or dark respiration.

The R 10 of the ecosystem respiration (REco) was estimated from the nighttime fluxes measured by the eddy covariance as 2.7 µmol m-2 s-1, as described in Marushchak et al. (2012). The area integrated R 10 of the dark respiration (Rd ) was estimated from the leaf level measurements as 0.5 µmol m-2 s-1 (see below). According to Equation (10), the R 10 of the soil respiration (Rsoil ) was calculated as 2.7 µmol m-2 s-1 -0.5 µmol m-2 s-1 = 2.2 µmol m-2 s-1.

SUPPLEMENTARY MEASUREMENTS

The development of the green leaf area index (LAl) was monitored using an LAI-2000 (LI-COR, Lincoln, Nebraska, U.S.A.) during the peak growing season (July—August). This was supplemented by spectral reflectance measurements in the red (FR ) and near infrared (FNIR ) part of the spectrum using a 2-channel spectrometer (Skye Instruments Ltd, Powys, U.K.). Using regression analysis, it was possible to determine the slope (c = 0.28) and the offset (r0= 1.5) in Equation (12) and thereby also estimate the LAI outside the peak growing season, in accordance with Soegaard et al. (2000):

e12_50.gif

To be able to upscale the LAI to landscape level, the level-4 MODIS global Leaf Area Index product was used as 8-day composites during the period from 8 May to 29 September 2008, at 1-km resolution (MODIS land product, Oak Ridge National Laboratory Distributed Active Archive Center (ORNL DAAC, 2009).

The basic meteorological parameters were recorded at the study site for the duration of the whole measurement campaign in 2008 using a CR23X logger (Campbell Scientific, U.K.). The following parameters were measured at half-hour intervals: air temperature and relative humidity (Hygrometer MP100A, Rotronic, U.S.A.), soil heat flux (HFP01, Hukseflux Thermal Sensors, The Netherlands), net radiation (NR LITE, Kipp and Zonen, The Netherlands), photosynthetic photon flux density (PPFD) (LI-190, LICOR, U.S.A.), precipitation (7852 Rain Collector II, Davis, U.S.A.) and wind speed (A100R, Vector Instruments, U.K.).

The leaf temperature was estimated as:

e13_50.gif

where H is the sensible heat flux in W m-2 measured by the eddy covariance technique, ρ is the air density in kg m-3,c p is the specific heat for air at constant pressure (1005 J kg-1 K-1), Ta is the air temperature in K, and rabw is the total resistance to water vapor calculated according to Monteith and Unsworth (1990).

The photosynthetic response curves of net assimilation vs. light and internal CO2 concentration (A/Ci ) were measured by a portable photosynthesis system (LI-6400, LI-COR Inc., Lincoln, Nebraska, U.S.A.) (Table 2). V cmax was estimated from the initial slope of the A/Ci , curves (Harley et al., 1992) and scaled to common temperature (25 °C; Leuning, 1997).

Results

PARAMETERIZATION OF THE MODEL

The model requires a set of input parameters, which describe the leaf physiological parameters, such as Rubisco capacity (V cmax), quantum use efficiency (α), and scaling factor for Rld (r), leaf area index (LAI), and the meteorological conditions.

Leaf Physiological Input Parameters

For the description of the leaf physiological parameters, representative values for the study site were chosen from four measured species (Salix lanata, Salix phylicifolia, Carex aquatilis, and Rubus chamaemorus). The quantum use efficiency (α) was found to vary between 0.06 and 0.09 µmol CO2 µmol photon-1. This can be compared to the values of 0.07 µmol CO2 µmol photon-1 derived from the eddy covariance data. The scaling factor for Rld (r) was estimated as 0.035 for the four species, which lies in a similar magnitude to that reported in other studies (Miao et al., 2009; Rodeghiero et al., 2007). The Rubisco capacity at 25 °C (V cmax25) was found to range from 27 to 51 µmol m-2 s-1 among the four species. Taking the composition of the landscape into account, the (dwarf) shrubs/willow (Salix spp.) were the representative vegetation type, because they are the most abundant species. Even if land-cover classes (willow and Betula nana heath) only cover 12% of the study site, the dwarf shrubs were found in most of the other land-cover classes. For this reason V cmax25, representing the entire study area, was derived from two Salix species to 45 µmol m-2 s-1.

FIGURE 3.

The slope (m) and intercept (b) of the response of stomatal conductance (gsw) to the index of the rate of net CO2 uptake (An ), the relative humidity (hs ) and the CO2 mole fraction (Cs ) of the air at the leaf surface are obtained by linear regression analysis of ambient measurements from gas exchange studies.

f03_50.jpg

The slope (m) and intercept (b) of the response of the stomatal conductance (gsw ) to the rate of net CO2 uptake (An ), relative humidity (hs ), and CO2 mole fraction (C3 ) of the air at the leaf surface were obtained from the leaf-level measurements [Equation (6)]. A linear regression analysis of the ambient measurement of the A/Ci curves is shown in Figure 3.

Leaf Area Index for Scaling from Leaf to Canopy

The MODIS LAI data were compared to the handheld LAI measurements, showing that values nearly doubled during the peak growing season. Additionally, we found that the satellite-derived LAI values also increased during early spring. The bud break of the deciduous shrubs was noted on 19 June. The re-greening of the evergreen dwarf shrubs was observed on 17 June. A similar increase of satellite-derived LAI prior to the bud break was made by Verbyla (2005) in Alaska, who attributed this artificial increase to the snowmelt. In accordance with this, the LAI increase was delayed until 17 June through extrapolation of the measured LAI by the spectrometer. The last snow patches disappeared on 25 June, so the recalibrated MODIS data were used after that date. By combining the three methods, the development of the green LAI during the growing season was derived (Fig. 4).

FIGURE 4.

Development of measured and interpolated LAI, the sinus of the sun elevation (β), the extinction coefficient (k) and the canopy factor (Π) during the modeling period. The dotted line indicates the start of the bud break.

f04_50.jpg

The daily midday extinction coefficient is around 0.7 until the end of July. After that, the extinction coefficient starts increasing, because the solar elevation angle decreases, leading to a longer path length through the canopy. The extinction coefficient continues to increase as the solar elevation angle decreases until it reaches a maximum value of 1.7 in the beginning of October. At that time, the sine of the solar elevation angle reaches half of its initial value in May, which equals a decrease from the maximum solar elevation angle of 46° in June to 17° in October. Comparable extinction coefficients were measured in arctic tundra ecosystems ranging from 0.9 during April and May in subarctic Canada (Bewley et al., 2007) and during July and August in a high arctic site in Greenland (Soegaard and Nordstroem, 1999) to 1.9 during July and August in a modeling study for tundra ecosystems (Shaver et al., 2007).

Meteorological and Radiation Input Parameters

The seasonal variation of the meteorological input parameters of air, soil, and leaf temperature; photosynthetic photon flux density (PPFD); and relative humidity is shown in Figure 5.

The measurements started on 18 May 2008 (DOY 139), when the half-hourly average air temperatures were mainly below 0 °C, and the landscape was partly covered in snow. The average soil temperatures at 2 cm depth remained negative until 27 May (DOY 148). After that, only nighttime soil temperatures reached 0 °C, while daytime soil temperatures approached 2–3 °C. From 10 June onwards, no negative soil temperatures were measured until 26 September (DOY 270). The daily temperature rose until the end of July, with a daily average maximum of 23 °C in the air and 24 °C at the leaf surface on 23 July (DOY 205). The soil temperatures in 5 cm depth reached a maximum value of 16 °C.

A maximum half-hourly PPFD value of 1300 µmol m-2 s-1 was measured at the middle of July. Low PPFD levels (<50 µmol m-2 s-1) only occurred in May and June during the period from 9:30 p.m. to 3:00 a.m. local summer time, when the midnight sun occurred.

SEASONAL AND DIURNAL COURSE OF MODELED NET ASSIMILATION

The net assimilation (An ) was modeled using the parameters listed in Table 2. The seasonal course of the modeled daily average An (black), the measured An as the sum of eddy covariance measurements (NEP) and the soil respiration (gray) are shown in Figure 6. The daily averages were calculated only from those half-hour intervals when an original flux from the eddy covariance data was available to avoid artificial error being possibly introduced by the gap-filling procedure.

FIGURE 5.

Seasonal course of selected meteorological parameters during the measurement campaign 2008: (a) air and soil temperature, (b) leaf temperature according to Equation (13), (c) photosynthetic photon flux density (PPFD), and (d) relative humidity.

f05_50.jpg

The comparison of the modeled and measured fluxes showed an overall good agreement of the daily average fluxes during the growing season, as the values are scattered around the 1:1 line (gray line in Fig. 6 [right]). However, during early autumn, a difference between the modeled and measured fluxes was observed. By analyzing the diurnal course of the modeled An, we found that, especially during that period, the fluxes were often excessively limited by the export (Js ), leading to an underestimation of the modeled An compared to the measured An. For this reason, the scaling factor for export limitation (CJs ) was increased from 0.5 to 1. This modification weakened the influence of the limitation of the assimilation due to the removal of assimilates and thus increased the modeled An. A similar approach was taken in other northern latitude studies (Dang et al., 1998; Soegaard and Nordstroem, 1999). This change led to a decrease in root square mean error (RSME) from 1.68 µmol m-2 s-1 to 1.49 µmol m-2 s-1 because the underestimation in autumn decreased (Fig. 7). A statistical t-test demonstrated that the slope of the regression line with 0.9 differs significantly from 0, but not from the 1:1 line (t = 4.3).

FIGURE 6.

Daily average net assimilation rates derived from the Collatz model (black) and the original, not gap-filled fluxes measured by the eddy covariance technique, defined as An EC = -NEE + Rsoil (gray), (RSME = 1.68 µmol m-2 s-1).

f06_50.jpg

Taking these aspects into account, the net carbon assimilation for a heterogeneous tundra ecosystem can be adequately simulated using the Collatz model. As shown in Figure 8, the modeled cumulative An amounts to 306 g C m-2 over the measurement period from mid-May to the beginning of October. The measured cumulative An is 31 g C m-2 lower, with the largest deviation from the model during the start of the growing season, but both curves feature the same slope.

FIGURE 7.

Daily average net assimilation rates derived from the Collatz model (black) and the original, not gap-filled fluxes measured by the eddy covariance technique, defined as An EC = -NEE + Rsoil (gray), including a change of CJs from 0.5 to 1 (RSME = 1.49 µmol m-2 s-1).

f07_50.jpg

For further examination of the model, four evenly distributed days were chosen to demonstrate the diurnal course of the measured and modeled An. Figure 9 shows the fluxes together with the three limiting rates on 22 June, 23 July, 23 August, and 10 September 2008. It is evident that during nighttime, light was the limiting factor of the fluxes, whilst during daytime, Rubisco capacity gained in importance as the limiting factor. On 22 June, 23 July, and 10 September, the CO2 fluxes were limited only by the Rubisco limited rate (Jc ), whereas the fluxes on 23 August were limited by both Rubisco (Jc ) and light (JE ). The weakening of the influence of the export limitation (Js ) restricted the flux only as a co-limitation, e.g., on 22 June in the morning.

A sensitivity test was conducted on the modeled half-hourly fluxes for the input parameters LAI, leaf temperature, and Rubisco capacity (V cmax), while other parameters were unchanged and set to a PPFD of 1150 µmol m-2 s-1, a relative humidity of 70%, and the extinction coefficient of 0.7 (Fig. 10). For all three parameters, An shows a distinct response. The V cmax has the greatest potential for an increase in An , because An almost doubles when doubling the V cmax to 90 µmol m-2 s-1. An also almost doubles when LAI is increased from 0.5 to 1, but subsequently increases less with higher LAI. The temperature exerts a positive influence on the assimilation at first, but at high temperatures, above 30 °C, the modeled rates decrease due to the high-temperature constraints.

FIGURE 8.

Modeled and measured cumulative net assimilation rates over the measurement period 2008. The estimated error of the model was found to be ±5.1 g C m-2 (Soegaard et al., 2000), while the estimated error of the measured flux data is ±49.8 g C m-2, due to the uncertainty in the gapfilling procedure (Marushchak et al., 2012).

f08_50.jpg

NET ECOSYSTEM CARBON DIOXIDE EXCHANGE IN THE CONTEXT OF CLIMATE CHANGE

For the evaluation of future carbon balance, the Collatz model was used for the simulation of An at variable air and soil temperatures and a higher atmospheric CO2 concentration. The air and soil temperatures were changed in 2 °C intervals from 4 °C below to 6°C above 2008 levels, while the atmospheric CO2 concentration was increased to a value of 470 ppm (according to the recent growth rates and concentrations from  http://www.esrl.noaa.gov/gmd/ccgg/trends/). It was assumed that a temperature increase would affect the soil and air temperature equally. All other meteorological conditions, such as precipitation, radiation, and relative humidity, remained the same as in the year 2008. Only the length of the growing season was adjusted, as a temperature change results in a longer growing season. In order to estimate the prolongation, we calculated the growing season according to the sum of days exceeding 5 °C on five consecutive days. This temperature threshold was previously applied in the Arctic by Groendahl et al. (2007) and Aurela et al. (2001). By increasing the air temperature, the length of the growing season extended on average by 3 days per 1 °C temperature increase.

The model showed that this ecosystem accumulates 58.1 g C m-2 from mid-May to the beginning of October. The contributing fluxes include 305.6 g C m-2 net assimilation and 247.5 g C m-2 carbon loss through soil respiration. When soil and air temperatures rise, the Collatz model predicts that both net assimilation and soil respiration rates will increase (Fig. 11).

But as the soil respiration rates increase faster than the net assimilation rates, the carbon budget turns from sink to source, when temperatures rise more than 2 °C at current atmospheric CO2 concentrations. The enhanced carbon loss is further facilitated as the simulated net assimilation rates cease to increase due to the high temperature inhibition of the photosynthesis. This restriction of the photosynthetic gains is important due to the relatively high summer temperatures experienced in continental arctic Russia.

Considering the present rate of increase in atmospheric CO2, higher CO2 concentrations are likely to accompany future temperature increases. When the ecosystem is not limited by nutrients (nitrogen or phosphorus), the CO2 fertilization due to the higher CO2 concentrations increases net assimilation rates, counteracting the enhanced losses due to the initial temperature increase. At higher CO2 concentrations, the change from sink to source occurs beyond a temperature increase of 4 °C.

FIGURE 9.

Four selected days illustrate the temporal variability of the net assimilation, including their three limiting rates during the growing season.

f09_50.jpg

Discussion

MODELING THE NET ASSIMILATION RATE OF A HETEROGENEOUS TUNDRA

This study shows that the net carbon assimilation An amounted to 306 g C m-2 in Northeast European Russia over the measurement period from mid-May to the beginning of October 2008. When adding the modeled dark respiration (47g C m-2) to An, the modeled gross primary production (GPP) amounted to 353g C m-2. This carbon assimilation agrees with a GPP of 334.3 ± 41.3 g C m-2 estimated by Zamolodchikov and Karelin (2001) for the entire European Russian tundra. Also, other arctic studies measured GPP of a similar magnitude: approximately 310 g C m-2were measured in a Canadian tundra by Nobrega and Grogan (2008) and 374 g C m-2 in a tussock tundra exhibiting extensive thaw in Alaska by Vogel et al. (2009).

FIGURE 10.

Results from a sensitivity test of the modeled half-hourly fluxes for the input parameters (a) LAI, (b) leaf temperature, and (c) Rubisco capacity, while other parameters were unchanged and set to measured values: 1150 µmol m-2 s-1 for PPFD, 70% for relative humidity, and 0.7 for the extinction coefficient.

f10_50.jpg

The Collatz model was used in its standard setup with site-specific parameters (Table 2). The only necessary change was to weaken the export limitation, especially at cold temperatures, because we detected a systematic underestimation of the measured fluxes by this limitation. This might indicate that the translocation of the assimilates exerts a minor limitation, as demonstrated in models presented by Haxeltine and Prentice (1996) and Sitch et al. (2003). These models are able to simulate the assimilation rates considering only the light and Rubisco limitation processes.

FIGURE 11.

Carbon budget for the period from mid-May to beginning of October, with different temperatures and atmospheric CO2 concentrations, separated into (a) net assimilation (An ) and soil respiration (Rsoil ), and (b) combined to net ecosystem exchange (NEE).

f11_50.jpg

The evaluation of the modeled fluxes also shows that the Rubisco capacity (V cmax) is an important limiting factor for determining An. During the peak growing season, comprising July and August, 68% of the daytime fluxes were limited by the Rubisco capacity, whereas the remaining 32% were light limited. The influence of the export limitation was weakened, as discussed above, and thus exerted only marginal constraints. Consequently, V cmax might have a significant influence on the magnitude of the net assimilation. Only sparse V cmax data are available for arctic species. The V cmax25 of 45 µmol m-2 s-1, determined in this study by leaf gas exchange measurements, is found to be of a similar magnitude as found in other arctic studies, as summarized in Table 3. Due to the wide range of values for Rubisco capacity, the representative value for the study area might be altered, affecting the net assimilation, when the composition of the plant species changes substantially.

TABLE 3

Overview of Rubisco capacity values in the Arctic.

t03_50.gif

Despite the good agreement between the modeled and measured daily values, there are discrepancies that reflect potential bias in the measurements and the model. The comparison shows that, during spring, the model overestimates the measured fluxes, accumulating to a difference of ca. 31 g C m-2. The opposite was found in August, when the model underestimates the measured fluxes. These discrepancies can either be attributed to the measurements or the model, or a combination of both.

The model might introduce bias due to its parameterization, leading to the noted overestimation in spring and the underestimation in August. There are three possible reasons for the overestimation in spring. The biochemical parameters, such as V cmax, might exhibit a seasonal course, starting with lower values during spring, increasing to larger values during peak growing season, and subsequently decreasing again. This seasonal pattern of V cmax was found through inverse modeling in boreal forests (Thum et al., 2008) and might lead to an overestimation of the measured fluxes when using a constant V cmax value. Dang et al. (1998) demonstrated that a single mid-growing-season parameterization gave reasonable results, but the correlation coefficient was improved by including the seasonal variation. In this study, an introduction of a seasonal course would improve the model in the beginning of the growing season, but would increase the discrepancy in autumn by further reducing the modeled flux.

Another reason might be related to the estimation of the LAI. Using a combination of satellite data, spectrometer, and optical measurements, the LAI value reached 1.1 – 1.3 during the peak growing season. Other studies in the Arctic have found LAI values similar to this study (Table 4). Compared to the LAI-2000 measurements, the MODIS-derived LAI was twice as high, which made a recalibration necessary.

Besides the peak value of the LAI, the start of the increase in LAI, in particular, governed An. Even if the meteorological conditions would allow photosynthesis in the model, the LAI has to be greater than 0 to simulate the carbon uptake at canopy level. However, during the start of the growing season, in particular, LAI estimates are challenging, because direct measurements using the LAI-2000 were not feasible, and the satellite-derived LAI might lead to an artificial increase due to snowmelt as shown by Verbyla (2005). This leads to an uncertainty of the timing of the vegetation growth, which might contribute to the earlier start of the modeled carbon assimilation compared to the measured values.

A third reason might be related to overestimation of soil respiration rates, resulting in larger measured net assimilation rates. Water saturated soil conditions in spring might limit the oxygen transport in the soil column and thus the aerobic decomposition (Hobbie et al., 2000). This might result in a lower actual soil respiration rate than the rate calculated from Equation (9), and thus lower measured An by the EC technique.

In contrast to the overestimation in spring, the model underestimates the measured fluxes in August in particular. The monthly average air temperature in August was 9 °C, with daily air temperatures ranging from 6 to 15 °C. This was significantly colder than July, with daily air temperatures ranging from 12 to 23 °C and a monthly average of 16 °C. The underestimation of the measured fluxes could be caused by the fact that the photosynthesis may not be as constrained at low temperatures as previously assumed. The original model setup includes a cold temperature inhibition in the export limitation (Js ). We already adapted this constraint by weakening the influence of Js and lowering the cold inhibition temperature (s2) from 15 to 0 °C. This improved the model by decreasing the RSME from 1.68 µmol m-2 s-1 to 1.49 µmol m-2 s-1, but did not fully solve the discrepancy. A similar observation was made by Williams et al. (2000). They were also able to improve the model by adjusting the low temperature inhibition to the extent that the photosynthetic rates were kept at 95% of their maxima at 5 °C.

TABLE 4

Overview of leaf area index (LAI) values in the Arctic.

t04_50.gif

CARBON BUDGET UNDER A CHANGING CLIMATE

The mean annual air temperature is likely to increase up to 7 °C in arctic Russia by the end of this century (ACIA, 2005). The simulations with increasing air and soil temperatures showed how fragile this ecosystem is and how changes in temperature can weaken the sink strength or even turn the ecosystem into a carbon source. Under present-day meteorological and leaf physiological conditions (Fig. 5), the net assimilation will still be enhanced when the air temperature increases by 2 °C. As the net assimilation operates close to its optimum temperature, the biological activity would become restricted with further temperature increases, which was demonstrated in the sensitivity test. Over a longer period a temperature increase could lead to a shift in optimum temperature, as plants are able to adapt to changes in growing temperature (Sage and Kubien, 2007). The ability of shifts in the optimum temperature on the ecosystem level was demonstrated for subtropical to boreal ecosystems and related to temperature acclimation of gross ecosystem production (GEP) (Niu et al. 2012). A higher acclimation potential due to a wider range of temperature variability was found in northern populations of trees relative to their southern counterparts (Ghannoum and Way, 2011). This thermal acclimation would delimit the high-temperature inhibition, enabling the net assimilation rates to counterbalance the respiratory losses even at higher temperatures.

Generally, experimental warming studies (e.g. Welker et al., 2004; Oberbauer et al., 2007) or model projections (e.g. Sitch et al., 2007; Qian et al., 2010) indicate a greater increase in net primary production or GEP than in RECO in the Arctic. The increase in GEP is mainly caused by an increase in green biomass (Marchand et al., 2004) and more favorable temperature conditions (Callaghan et al., 2004). The response of GEP to warming can also be related to a latitudinal gradient (Oberbauer et al., 2007). They found that the smallest response of GEP to warming was in the southern part with higher temperatures. Considering that our study site is also experiencing relatively high temperatures during the growing season (Fig. 5), it seems plausible that the response to warming is less pronounced than in studies in the High Arctic (e.g. Welker et al., 2004; Marchand et al., 2004).

Additionally, increasing temperatures will not only lead to a longer growing season, which is positively related to carbon uptake (Lund et al., 2010; Groendahl et al., 2007), but might also introduce vegetation changes (Elmendorf et al., 2012). Observations based on satellite data and photographs taken during the last few decades show that this change has already started in the transition zone between taiga and tundra. An increase in normalized difference vegetation index (NDVI), visible as a “greening” of the tundra, has been observed in the circumpolar Arctic (e.g., Sturm et al., 2001; Forbes et al., 2010, Myers-Smith et al., 2011) and attributed to the expansion or increase in the height of shrubs, or both these factors, manifested in a larger LAI (Walker et al., 2006; Raynolds et al., 2008). Even though this model does not include a dynamic response of the carbon exchange process to changes in carbon and nitrogen pools, the sensitivity test indicates the ecosystem's potential to assimilate carbon. When running the existing model with double the LAI values, keeping all other parameters at present-day values, the net carbon uptake would increase from 306 g C m-2 to 477 g C m-2.

The warming experiments and model projections (e.g. Sitch et al., 2007; Qian et al., 2010) agree that the RECO will increase with rising temperatures. This is primarily caused by the exponential temperature dependence (Lloyd and Taylor, 1994), but also through unlocking carbon stored from the permafrost (Schuur et al., 2009). The effect of warming is further influenced by changes in the hydrology. Drier sites reacted more than wet sites (Oberbauer et al., 2007). Thus, a great uncertainty remains regarding the future hydrological regime in the study area, which will determine the increase in RECO- Under present hydrological conditions, an increase of RECO is projected, which is not constrained by oxygen-limited conditions.

Thus, if all other conditions were stable, this ecosystem would turn from a sink into a source with only a small increase in temperature. However, due to the changes, e.g. in plant composition and stature and hydrology, the response will be more complex.

Conclusion

The great advantage of the Collatz model compared to the complex process-based model is the relatively simple parameterization (Table 2). This study showed that, despite being a heterogeneous ecosystem, the model was able to simulate the fluxes using a single parameter set. The leaf area index (LAI) and the Rubisco capacity (V c1max) were identified as two important parameters for modeling carbon dioxide exchange on a canopy scale. It was also shown that a constant value of V cmax was able to capture the variation in An during the growing season. Furthermore, V cmax is found to be a robust measure (45–62 µmol m-2 s-1) for arctic ecosystems (Table 3).

This study provides a tool for scaling carbon fluxes to a larger area, such as the European Russian tundra. The combination of a constant V cmax and regional satellite-based LAI values facilitates the upscaling of carbon exchange to a regional scale, where care should be taken that the satellite data should be validated against ground truth data in order to avoid possible errors.

Furthermore, this study showed that European Russian tundra is close to its optimum temperature with regards to carbon accumulation rates. At present, the net assimilation is greater than the respiration, leading to a net accumulation of 58 g C m-2 during our model period from mid-May to the beginning of October. This net gain becomes even smaller when considering the cold season respiratory losses used to calculate an annual budget. With increasing temperatures, the photosynthesis is predicted to become restricted by the high temperatures, exerting a strong limitation on the carbon uptake if no long-term acclimation occurs. The respiration, on the other hand, is predicted to be enhanced with increasing temperatures. Under these conditions, the European Russian tundra might turn into a net carbon source with rising temperatures. However, future changes in vegetation composition and growth along with acclimation to the new thermal regime might enhance carbon assimilation. The future carbon balance will depend on whether or not the increase in assimilation will be able to keep up with the increase in soil respiration.

Acknowledgments

The authors would like to thank M. Marushchak, V. Elsakov, R. Jensen, D. Pedersen, and I. Samarina for their contribution in field work and the Center for Permafrost (CENPERM), University of Copenhagen, for its support. This research was financially supported by the Danish Council for Independent Research | Natural Sciences (FNU) (Reference number: 645-06-0493) and the EU 6th Framework Programme Global Change and Ecosystems (CARBO-North, project contract number 036993).

References Cited

1.

ACIA, 2005: Arctic Climate Impact Assessment. Cambridge and New York: Cambridge University Press, 1042 pp. Google Scholar

2.

P. Alton , and P. Bodin , 2010: A comparative study of a multilayer and a productivity (light-use) efficiency land-surface model over different temporal scales. Agricultural and Forest Meteorology , 150: 182–195. Google Scholar

3.

A. Aubinet , A. Grelle , A. Ibrom , U. Rannik , J. Moncrieff , T. Foken , A. S. Kowalski , P. H. Martin , P. Berbigier , Ch. Bernhofer , R. Clement , J. Eibers , A. Granier , T. Grunwald , K. Morgenstern , K. Pilegaard , C. Rebmann , W. Snijders , R. Valentini , and T. Vesala , 2000: Estimates of the annual net carbon and water exchange of forests: the EUROFLUX methodology. In A. H. Fitter , and D. G. Raffaelli (eds.), Advances in Ecological Research. San Diego: Academic Press, 113–175. Google Scholar

4.

M. Aurela , T. Laurila , and J.-P. Tuovinen , 2001: Seasonal CO2 balances of a subartic mire. Journal of Geophysical ResearchAtmospheres , 106: 1623–1637. Google Scholar

5.

D. Baldocchi , 1994: Acomparative study of mass and energy exchange rates over a closed C3 (wheat) and an open C4 (corn) crop: II. CO2 exchange and water use efficiency. Agricultural and Forest Meteorology , 67: 291–321. Google Scholar

6.

D. Baldocchi , E. Falge , L. H. Gu , R. Olson , D. Hollinger , S. Running , P. Anthoni , C. Bernhofer , K. Davis , R. Evans , J. Fuentes , A. Goldstein , G. Katul , B. Law , X. H. Lee , Y. Malhi , T. Meyers , W. Munger , W. U. K. Oechel , K. Pilegaard , H. P. Schmid , R. Valentini , S. Verma , T. Vesala , K. Wilson , and S. Wofsy , 2001 : FLUXNET: a new tool to study the temporal and spatial variability of ecosystem-scale carbon dioxide, water vapor, and energy flux densities. Bulletin of the American Meteorological Society , 82: 2415–2434. Google Scholar

7.

D. D. Baldocchi , 2003: Assessing the eddy covariance technique for evaluating carbon dioxide exchange rates of ecosystems: past, present and future. Global Change Biology , 9: 479–492. Google Scholar

8.

J. T. Ball , I. E. Woodrow , and J. A. Berry , 1987: A model predicting stomatal conductance and its contribution to the control of photosynthesis under different environmental conditions. In J. Biggens (ed.), Progress in Photosynthesis Research. Dordrecht: Martinus Nijhoff Publishers, 221–224. Google Scholar

9.

D. Bewley , J. W. Pomeroy , and R. L. H. Essery , 2007: Solar radiation transfer through a subarctic shrub canopy. Arctic, Antarctic, and Alpine Research , 39: 365–374. Google Scholar

10.

M. P. Bjorkman , E. Morgner , E. J. Cooper , B. Elberling , L. Klemedtsson , and R. G Bjork , 2010: Winter carbon dioxide effluxes from arctic ecosystems: an overview and comparison of methodologies. Global Biogeochemical Cycles , 24: GB3010, 10 pp.,  http://dx.doi.org/10.1029/2009GB003667Google Scholar

11.

L. C. Bliss , and N. V. Matveyeva , 1992: Circumpolar arctic vegetation. In F. S. Chapin , R. L. Jefferies , J. F. Reynolds , G. R. Shaver , J. Svoboda , and E. W. Chu (eds.), Arctic Ecosystems in a Changing Climate: an Ecophysiological Perspective. San Diego: Academic Press, 59–90. Google Scholar

12.

G. G. Burba , D. K. McDermitt , A. Grelle , D. J. Anderson , and L. K. Xu , 2008: Addressing the influence of instrument surface heat exchange on the measurements of CO2 flux from open-path gas analyzers. Global Change Biology , 14: 1854-1876. Google Scholar

13.

T. V. Callaghan , L. O. Björn , Y. Chernov , T. Chapin , T. R. Christensen , B. Huntley , R. A. Ims , M. Johansson , D. Jolly , S. Jonasson , N. Matveyeva , N. Panikov , W. Oechel , and G. Shaver , 2004: Effects on the function of arctic ecosystems in the short- and longterm perspectives. AMBIO: A Journal of the Human Environment , 33: 448–458. Google Scholar

14.

G. J. Collatz , J. T. Ball , C. Grivet , and J. A. Berry , 1991 : Physiological and environmental regulation of stomatal conductance, photosynthesis and transpiration: a model that includes a laminar boundary layer. Agricultural and Forest Meteorology , 54: 107–136. Google Scholar

15.

C. Corradi , O. Kolle , K. Walter , S. A. Zimov , and E. D. Schulze , 2005 : Carbon dioxide and methane exchange of a north-east Siberian tussock tundra. Global Change Biology , 11: 1910–1925. Google Scholar

16.

P. M. Cox , R. A. Betts , C. B. Bunton , R. L. H. Essery , P. R. Rowntree , and J. Smith , 1999: The impact of new land surface physics on the GCM simulation of climate and climate sensitivity. Climate Dynamics , 15: 183–203. Google Scholar

17.

Q. L. Dang , H. A. Margolis , and G. J. Collatz , 1998: Parameterization and testing of a coupled photosynthesis stomatal conductance model for boreal trees. Tree Physiology , 18: 141–153. Google Scholar

18.

S. C. Elmendorf , G. H. R. Henry , R. D. Hollister , R. G. Björk , A. D. Bjorkman , T. V. Callaghan , L. S. Collier , E. J. Cooper , J. H. C. Cornelissen , T. A. Day , A. M. Fosaa , W. A. Gould , J. Grétarsdottir , J. Harte , L. Hermanutz , D. S. Hik , A. Hofgaard , F. Jarrad , I. S. Jónsdóttir , F. Keuper , K. Klanderud , J. A. Klein , S. Koh , G Kudo , S. I. Lang , V. Loewen , J. L. May , J. Mercado , A. Michelsen , U. Molau , I. H. Myers-Smith , S. F. Oberbauer , S. Pieper , E. Post , C. Rixen , C. H. Robinson , N. M. Schmidt , G. R. Shaver , A. Stenström , A. Tolvanen , Ø. Totland , T. Troxler , C. H. Wahren , P. J. Webber , J. M. Welker , and P. A. Wookey , 2012: Global assessment of experimental climate warming on tundra vegetation: heterogeneity over space and time. Ecology Letters , 15: 164–175. Google Scholar

19.

J. T. Fahnestock , M. H. Jones , and J. M. Welker , 1999: Wintertime CO2 efflux from arctic soils: implications for annual carbon budgets. Global Biogeochemical Cycles , 13: 775–779. Google Scholar

20.

G D. Farquhar , S. Caemmerer , and J. A. Berry , 1980: A biochemical model of photosynthetic CO2 assimilation in leaves of C3 species. Planta , 149: 78–90. Google Scholar

21.

B. C. Forbes , M. M. Fauria , and P. Zetterberg , 2010: Russian arctic warming and 'greening' are closely tracked by tundra shrub willows. Global Change Biology , 16: 1542–1554. Google Scholar

22.

A. M. Fox , B. Huntley , C. R. Lloyd , M. Williams , and R. Baxter , 2008: Net ecosystem exchange over heterogeneous arctic tundra: scaling between chamber and eddy covariance measurements. Global Biogeochemical Cycles , 22: GB2027, 15 pp.,  http://dx.doi.org/10.1029/2007GB003027Google Scholar

23.

O. Ghannoum , and D. A. Way , 2011 : On the role of ecological adaptation and geographic distribution in the response of trees to climate change. Tree Physiology , 31: 1273–1276. Google Scholar

24.

R. F. Grant , W. C. Oechel , and C. L. Ping , 2003: Modelling carbon balances of coastal arctic tundra under changing climate. Global Change Biology , 9: 16–36. Google Scholar

25.

L. Groendahl , T. Friborg , and H. Soegaard , 2007: Temperature and snow-melt controls on interannual variability in carbon exchange in the High Arctic. Theoretical and Applied Climatology , 88: 111–125. Google Scholar

26.

P. C. Harley , R. B. Thomas , J. F. Reynolds , and B. R. Strain , 1992: Modeling photosynthesis of cotton grown in elevated CO2. Plant, Cell and Environment , 15: 271–282. Google Scholar

27.

A. Haxeltine , and I. C. Prentice , 1996: A general model for the light-use efficiency of primary production. Functional Ecology , 10: 551–561. Google Scholar

28.

S. E. Hobbie , J. P. Schimel , S. E. Trumbore , and J. R. Randerson , 2000: Controls over carbon storage and turnover in high-latitude soils. Global Change Biology , 6: 196–210. Google Scholar

29.

IPCC, 2007: Climate Change 2007: Working Group I: The Physical Science Basis. Cambridge and New York: Cambridge University Press, 996 pp. Google Scholar

30.

P. G Jarvis , and K. G. McNaughton , 1986: Stomatal control of transpiration—Scaling up from leaf to region. Advances in Ecological Research , 15: 1–49. Google Scholar

31.

O. Kull , and P. G Jarvis , 1995: The role of nitrogen in a simple scheme to scale-up photosynthesis from leaf to canopy. Plant Cell and Environment , 18: 1174–1182. Google Scholar

32.

P. M. Lafleur , and E. R. Humphreys , 2008: Spring warming and carbon dioxide exchange over low arctic tundra in central Canada. Global Change Biology , 14: 740–756. Google Scholar

33.

T. Laurila , H. Soegaard , C. R. Lloyd , M. Aurela , J. P. Tuovinen , and C. Nordstroem , 2001 : Seasonal variations of net CO2 exchange in European arctic ecosystems. Theoretical and Applied Climatology , 70: 183–201. Google Scholar

34.

S. Le Dizes , B. L. Kwiatkowski , E. B. Rastetter , A. Hope , J. E. Hobbie , D. Stow , and S. Daeschner , 2003: Modeling biogeochemical responses of tundra ecosystems to temporal and spatial variations in climate in the Kuparuk River Basin (Alaska). Journal of Geophysical Research—Atmosphere , 108:8165,  http://dx.doi.org/10.1029/2001JD000960Google Scholar

35.

R. Leuning , 1997: Scaling to a common temperature improves the correlation between the photosynthesis parameters Jmax and Vcmax. Journal of Experimental Botany , 48: 345–347. Google Scholar

36.

J. Lloyd , and J. A. Taylor , 1994: On the temperature dependence of soil respiration. Functional Ecology , 8: 315–323. Google Scholar

37.

M. Lund , P. M. Lafleur , N. T. Roulet , A. Lindroth , T. R. Christensen , M. Aurela , B. H. Chojnicki , L. B. Flanagan , E. R. Humphreys , T. Laurila , W. C. Oechel , J. Olejnik , J. Rinne , P. Schubert , and M. B. Nilsson , 2010: Variability in exchange of CO2 across 12 northern peatland and tundra sites. Global Change Biology , 16: 2436–2448. Google Scholar

38.

F. L. Marchand , I. Nijs , H. J. de Boeck , F. Kockelbergh , S. Mertens , and L. Beyens , 2004: Increased turnover but little change in the carbon balance of high-arctic tundra exposed to whole growing season warming. Arctic, Antarctic, and Alpine Research , 36: 298–307. Google Scholar

39.

M. E. Marushchak , I. Kiepe , C. Biasi , V. Elsakov , T. Friborg , T. Johansson , H. Soegaard , T. Virtanen , and P. J. Martikainen , 2012: Carbon dioxide balance of subarctic tundra from plot to regional scales. Biogeosciences, Discussion, 9: 9945–9991,  http://dx.doi.org/10.5194/bgd-9-9945-2012Google Scholar

40.

J. P. McFadden , W. Eugster , and F. S. Chapin , 2003: A reginoal study of the controls on water vapour and CO2 exchange in arctic tundra. Ecology , 84: 2762–2776. Google Scholar

41.

A. D. McGuire , J. S. Clein , J. M Melillo, D. W. Kicklighter , R. A. Meier , C. J. Vorosmarty , and M. C. Serreze , 2000: Modeling carbon responses of tundra ecosystems to historical and projected climate: sensitivity of pan-Arctic carbon storage to temporal and spatial variation in climate. Global Change Biology , 6:141–159. Google Scholar

42.

A. D. McGuire , L. G. Anderson , T. R. Christensen , S. Dallimore , L. D. Guo , D. J. Hayes , M. Heimann , T. D. Lorenson , R. W. Macdonald , and N. Roulet , 2009: Sensitivity of the carbon cycle in the Arctic to climate change. Ecological Monographs , 79: 523–555. Google Scholar

43.

Z. W. Miao , M. Xu , R. G Lathrop , and Y. F. Wang , 2009: Comparison of the A-Cc?? curve fitting methods in determining maximum ribulose 1.5-bisphosphate carboxylase/oxygenase carboxylation rate, potential light saturated electron transport rate and leaf dark respiration. Plant, Cell and Environment , 32: 109–122. Google Scholar

44.

J. L. Monteith, and M. H. Unsworth, 1990: Principles of Environmental Physics. New York: Edward Arnold, 291 pp. Google Scholar

45.

H. Muraoka , M. Uchida , M. Mishio , T. Nakatsubo , H. Kanda , and H. Koizumi , 2002: Leaf photosynthetic characteristics and net primary production of the polar willow (Salix polaris) in a high arctic polar semi-desert, Ny-Alesund, Svalbard. Canadian Journal of Botany—Revue Canadienne de Botanique , 80: 1193–1202. Google Scholar

46.

I. H. Myers-Smith , B. C. Forbes , M. Wilmking , M. Hallinger , T. Lantz , D. Blok , K. D. Tape , M. Macias-Fauria , U. Sass-Klaassen , E. Levesque , S. Boudreau , P. Ropars , L. Hermanutz , A. Trant , L. S. Collier , S. Weijers , J. Rozema , S. A. Rayback , N. M. Schmidt , G. Schaepman-Strub , S. Wipf , C. Rixen , C. B. Menard , S. Venn , S. Goetz , L. Andreu-Hayles , S. Elmendorf , V. Ravolainen , J. Welker , P. Grogan , H. E. Epstein , and D. S. Hik , 2011: Shrub expansion in tundra ecosystems: dynamics, impacts and research priorities. Environmental Research Letters , 6: 045509,  http://dx.doi.org/10.1088/1748-9326/6/4/045509Google Scholar

47.

S. Niu , Y. Luo , S. Fei , W. Yuan , D. Schimel , B. E. Law , C. Ammann , M. Altaf Arain , A. Arneth , M. Aubinet , A. Barr , J. Beringer , C. Bernhofer , T. Andrew Black , N. Buchmann , A. Cescatti , J. Chen , K. J. Davis , E. Dellwik , A. R. Desai , S. Etzold , L. Francois , D. Gianelle , B. Gielen , A. Goldstein , M. Groenendijk , L. Gu , N. Hanan , C. Helfter , T. Hirano , D. Y. Hollinger , M. B. Jones , G. Kiely , T. E. Kolb , W. L. Kutsch , P. Lafleur , D. M. Lawrence , L. Li , A. Lindroth , M. Litvak , D. Loustau , M. Lund , M. Marek , T. A. Martin , G. Matteucci , M. Migliavacca , L. Montagnani , E. Moors , J. W. Munger , A. Noormets , W. Oechel , J. Olejnik , Kyaw, U Tha Paw, K. Pilegaard , S. Rambal , A. Raschi , R. L. Scott , G. Seufert , D. Spano , P. Stoy , M. A. Sutton , A. Varlagin , T. Vesala , E. Weng , G. Wohlfahrt , B. Yang , Z. Zhang , and X. Zhou , 2012: Thermal optimality of net ecosystem exchange of carbon dioxide and underlying mechanisms. New Phytologist , 194: 775–783. Google Scholar

48.

S. Nobrega , and P. Grogan , 2008: Landscape and ecosystem-level controls on net carbon dioxide exchange along a natural moisture gradient in Canadian low arctic tundra. Ecosystems , 11: 377–396. Google Scholar

49.

S. F. Oberbauer , C. E. Tweedie , J. M. Welker , J. T. Fahnestock , G. H. R. Henry , P. J. Webber , R. D. Hollister , M. D. Walker , A. Kuchy , E. Elmore , and G Starr , 2007: Tundra CO2 fluxes in response to experimental warming across latitudinal and moisture gradients. Ecological Monographs , 77: 221–238. Google Scholar

50.

ORNL DAAC [Oak Ridge National Laboratory Distributed Active Archive Center], 2009: MODIS subsetted land products, Collection 5. Available from  http://daac.ornl.gov/MODIS/modis.html.OakRidge, Tennessee: Oak Ridge National Laboratory, accessed May 2011. Google Scholar

51.

H. F. Qian , R. Joseph , and N. Zeng , 2010: Enhanced terrestrial carbon uptake in the northern high latitudes in the 21st century from the coupled carbon cycle climate model intercomparison project model projections. Global Change Biology , 16: 641–656. Google Scholar

52.

M. K. Raynolds , J. C. Comiso , D. A. Walker , and D. Verbyla , 2008: Relationship between satellite-derived land surface temperatures, arctic vegetation types, and NDVI. Remote Sensing of Environment , 112: 1884–1894. Google Scholar

53.

A. K. Rennermalm , H. Soegaard , and C Nordstroem , 2005: Interannual variability in carbon dioxide exchange from a high arctic fen estimated by measurements and modeling. Arctic, Antarctic, and Alpine Research , 37: 545–556. Google Scholar

54.

M. Rodeghiero , U. Niinemets , and A. Cescatti , 2007: Major diffusion leaks of clamp-on leaf cuvettes still unaccounted: how erroneous are the estimates of Farquhar et al. model parameters? Plant, Cell and Environment , 30: 1006–1022. Google Scholar

55.

R. F. Sage , and D. S. Kubien , 2007: The temperature response of C3 and C4 photosynthesis. Plant, Cell and Environment , 30: 1086–1106. Google Scholar

56.

E. A. G. Schuur , J. G. Vogel , K. G. Crummer , H. Lee , J. O. Sickman , and T. E. Osterkamp , 2009: The effect of permafrost thaw on old carbon release and net carbon exchange from tundra. Nature , 459: 556–559. Google Scholar

57.

P. J. Sellers , J. A. Berry , G. J. Collatz , C. B. Field , and F. G. Hall , 1992: Canopy reflectance, photosynthesis, and transpiration. 3. A reanalysis using improved leaf models and a new canopy integration scheme. Remote Sensing of Environment , 42: 187–216. Google Scholar

58.

G. R. Shaver , L. E. Street , E. B. Rastetter , M. T. Van Wijk , and M. Williams , 2007: Functional convergence in regulation of net CO2 flux in heterogeneous tundra landscapes in Alaska and Sweden. Journal of Ecology , 95: 802–817. Google Scholar

59.

S. Sitch , B. Smith , I. C. Prentice , A. Arneth , A. Bondeau , W. Cramer , J. O. Kaplan , S. Levis , W. Lucht , M. T. Sykes , K Thonicke , and S. Venevsky , 2003: Evaluation of ecosystem dynamics, plant geography and terrestrial carbon cycling in the LPJ dynamic global vegetation model. Global Change Biology , 9: 161–185. Google Scholar

60.

S. Sitch , A. D. McGuire , J. Kimball , N. Gedney , J. Gamon , R. Engstrom , A. Wolf , Q. Zhuang , J. Clein , and K. C. McDonald , 2007: Assessing the carbon balance of circumpolar arctic tundra using remote sensing and process modeling. Ecological Applications , 17: 213–234. Google Scholar

61.

H. Soegaard , and C. Nordstroem , 1999: Carbon dioxide exchange in a high-arctic fen estimated by eddy covariance measurements and modelling. Global Change Biology , 5: 547–562. Google Scholar

62.

H. Soegaard , C. Nordstroem , T. Friborg , B. U. Hansen , T. R. Christensen , and C. Bay , 2000: Trace gas exchange in a high-arctic valley 3. Integrating and scaling CO2 fluxes from canopy to landscape using flux data, footprint modeling, and remote sensing. Global Biogeochemical Cycles , 14: 725–744. Google Scholar

63.

M. Sturm , J. P. McFadden , G. E. Liston , F. S. Chapin , C. H. Racine . and J. Holmgren , 2001 : Snow-shrub interactions in arctic tundra: a hypothesis with climatic implications. Journal of Climate , 14: 336–344. Google Scholar

64.

C. Tarnocai , J. G. Canadell , E. A. G. Schuur , P. Kuhry , G. Mazhitova , and S. Zimov , 2009: Soil organic carbon pools in the northern circumpolar permafrost region. Global Biogeochemical Cycles , 23: GB2023, 11 pp.,  http://dx.doi.org/10.1029/2008GB003327Google Scholar

65.

T. Thum , T. Aalto , T. Laurila , M. Aurela , A. Lindroth , and T. Vesala , 2008: Assessing seasonality of biochemical CO2 exchange model parameters from micrometeorological flux observations at boreal coniferous forest. Biogeosciences , 5: 1625–1639. Google Scholar

66.

D. L. Verby la , 2005 : Assessment of the MODIS leaf area index product (MOD15) in Alaska. International Journal of Remote Sensing , 26: 1277–1284. Google Scholar

67.

J. Vogel , E. A. G. Schuur , C. Trucco , and H. Lee , 2009: Response of CO2 exchange in a tussock tundra ecosystem to permafrost thaw and thermokarst development. Journal of Geophysical ResearchBiogeosciences , 114: G04018, 14 pp.,  http://dx.doi.org/10.1029/2008JG000901Google Scholar

68.

G. L. Vourlitis , W. C. Oechel , A. Hope , D. Stow , B. Boynton , J. Verfaillie , R. Zulueta , and S. J. Hastings , 2000: Physiological models for scaling plot measurements of CO2 flux across an arctic tundra landscape. Ecological Applications , 10: 60–72. Google Scholar

69.

M. D. Walker , C. H. Wahren , R. D. Hollister , G. H. R. Henry , L. E. Ahlquist , J. M. Alatalo , M. S. Bret-Harte , M. P. Calef , T. V. Callaghan , A. B. Carroll , H. E. Epstein , I. S. Jonsdottir , J. A. Klein , B. Magnusson , U. Molau , S. F. Oberbauer , S. P. Rewa , C. H. Robinson , G. R. Shaver , K. N. Suding , C. C. Thompson , A. Tolvanen , O. Totland , P. L. Turner , C. E. Tweedie , P. J. Webber , and P. A. Wookey , 2006: Plant community responses to experimental warming across the tundra biome. Proceedings of the National Academy of Sciences of the United States of America , 103: 1342–1346. Google Scholar

70.

J. M. Welker , J. T. Fahnestock , G H. R. Henry , K. W. O'Dea , and R. A. Chimner , 2004: CO2 exchange in three Canadian high arctic ecosystems: response to long-term experimental warming. Global Change Biology , 10: 1981–1995,  http://dx.doi.org/10.1111/j.l3652486.2004.00857.xGoogle Scholar

71.

M. Williams , E. B. Rastetter , D. N. Fernandes , M. L. Goulden , S. C. Wofsy , G. R. Shaver , J. M. Melillo , J. W. Munger , S. M. Fan , and K. J. Nadelhoffer , 1996: Modelling the soil-plant-atmosphere continuum in a Quercus Acer stand at Harvard Forest: the regulation of stomatal conductance by light, nitrogen and soil/plant hydraulic properties. Plant, Cell and Environment , 19: 911–927. Google Scholar

72.

M. Williams , W. Eugster , E. B. Rastetter , J. P. McFadden , and F. S. Chapin , 2000: The controls on net ecosystem productivity along an arctic transect: a model comparison with flux measurements. Global Change Biology , 6: 116–126. Google Scholar

73.

M. Williams , L. E. Street , M. T. Van Wijk , and G. R. Shaver , 2006: Identifying differences in carbon exchange among arctic ecosystem types. Ecosystems , 9: 288–304. Google Scholar

74.

S. D. Wullschleger , 1993: Biochemical limitations to carbon assimilation in C3 Plants—A retrospective analysis of the A/Ci, curves from 109 species. Journal of Experimental Botany , 44: 907–920. Google Scholar

75.

D. G. Zamolodchikov , and D. V. Karelin , 2001: An empirical model of carbon fluxes in Russian tundra. Global Change Biology , 7: 147–161. Google Scholar
© 2013 Regents of the University of Colorado
Isabell Kiepe, Thomas Friborg, Mathias Herbst, Torbjörn Johansson, and Henrik Soegaard "Modeling Canopy CO2 Exchange in the European Russian Arctic," Arctic, Antarctic, and Alpine Research 45(1), 50-63, (1 February 2013). https://doi.org/10.1657/1938-4246-45.1.50
Accepted: 1 August 2012; Published: 1 February 2013
JOURNAL ARTICLE
14 PAGES


SHARE
ARTICLE IMPACT
Back to Top