Skip Navigation


Tree Physiology Advance Access originally published online on December 6, 2008
Tree Physiology 2009 29(1):1-17; doi:10.1093/treephys/tpn004
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow Supplementary Data
Right arrow All Versions of this Article:
29/1/1    most recent
tpn004v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Grant, R.F.
Right arrow Articles by Bergeron, O.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Grant, R.F.
Right arrow Articles by Bergeron, O.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2008. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Changes in net ecosystem productivity of boreal black spruce stands in response to changes in temperature at diurnal and seasonal time scales

R.F. Grant1,2, H.A. Margolis3, A.G. Barr4, T.A. Black5, A.L. Dunn6, P.Y. Bernier7 and O. Bergeron3

1 Department of Renewable Resources, University of Alberta, Edmonton, AB T6G 2E3, Canada
2 Corresponding author (robert.grant{at}afhe.ualberta.ca)
3 Faculté de Foresterie et de Géomatique, Pavillon Abitibi-Price, Université Laval, Québec, QC G1K 7P4, Canada
4 Climate Research Branch, Meteorological Service of Canada Saskatoon, SK S7N 3H5, Canada
5 Department of Soil Science, University of British Columbia, Vancouver BC V6T 1Z4, Canada
6 Department of Geography, Worcester State College, Worcester, MA 01602, USA
7 Natural Resources Canada Canadian Forest Service, Laurentian Forestry Center, Québec, QC G1V 4C7, Canada


   Abstract
Top
 Notes
 Abstract
Introduction
 Model description
 Field measurements
 Model testing
 Results
 Discussion
 References
 
Net ecosystem productivity (NEP) of boreal coniferous forests is believed to rise with climate warming, thereby offsetting some of the rise in atmospheric CO2 concentration (Ca) by which warming is caused. However, the response of conifer NEP to warming may vary seasonally, with rises in spring and declines in summer. To gain more insight into this response, we compared changes in CO2 exchange measured by eddy covariance and simulated by the ecosystem process model ecosys under rising mean annual air temperatures (Ta) during 2004–2006 at black spruce stands in Saskatchewan, Manitoba and Quebec. Hourly net CO2 uptake was found to rise with warming at Ta < 15 °C and to decline with warming at Ta > 20 °C. As mean annual Ta rose from 2004 to 2006, increases in net CO2 uptake with warming at lower Ta were greater than declines with warming at higher Ta so that annual gross primary productivity and hence NEP increased. Increases in net CO2 uptake measured at lower Ta were explained in the model by earlier recovery of photosynthetic capacity in spring, and by increases in carboxylation activity, using parameters for the Arrhenius temperature functions of key carboxylation processes derived from independent experiments. Declines in net CO2 uptake measured at higher Ta were explained in the model by sharp declines in mid-afternoon canopy stomatal conductance (gc) under higher vapor pressure deficits (D). These declines were modeled from a hydraulic constraint to water uptake imposed by low axial conductivity of conifer roots and boles that forced declines in canopy water potential ({psi}c), and hence in gc under higher D when equilibrating water uptake with transpiration. In a model sensitivity study, the contrasting responses of net CO2 uptake to specified rises in Ta caused annual NEP of black spruce in the model to rise with increases in Ta of up to 6 °C, but to decline with further increases at mid-continental sites with lower precipitation. However, these contrasting responses to warming also indicate that rises in NEP with climate warming would depend on the seasonality (spring versus summer) as well as the magnitude of rises in Ta.

Keywords: autotrophic respiration, climate change, ecosys, gross primary productivity, heterotrophic respiration, modeling, net primary productivity


   Introduction
Top
 Notes
 Abstract
 Introduction
Model description
 Field measurements
 Model testing
 Results
 Discussion
 References
 
The productivity of boreal coniferous forests is generally observed to rise with temperature (Myneni et al. 1997). In a global meta-analysis of CO2 fluxes measured over forests, Luyssaert et al. (2007) found that annual gross primary productivity (GPP) rose continuously with mean annual temperature, whereas net primary productivity (NPP = GPP – Ra, where Ra is autotrophic respiration) reached maximum values at a mean annual temperature of 10 °C. In a meta-analysis of CO2 fluxes measured over temperate and boreal coniferous forests, Arain and Restrepo-Coupe (2005) also found that GPP as well as ecosystem respiration (Re) rose with mean annual air temperature. In boreal ecosystems, warming lengthens the period between the spring thaw and the autumn freeze that determines the annual tree growth, mainly through temperature effects on CO2 uptake in spring (Mellander et al. 2008) and on nutrient uptake in summer (Jarvis and Linder 2000). Increases in tree growth and ecosystem carbon storage with warming were attributed to accelerated nitrogen mineralization and uptake by Ladanai and Ågren (2004).

However, Goetz et al. (2005) found from remote sensing studies that the response of some boreal forest zones to climate warming has been inconsistent with the general correlations between temperature and productivity. They attributed the apparent decoupling of warming and forest growth to possible stress due to drought, nutrient limitation, damages caused because of insect and disease, and changes in resource allocation. Tree ring analyses have also indicated contrasting responses of tree growth to warming. The tree growth inferred from ring analyses has been found to increase with warmer temperatures in Alaska (Garfinkel and Brubaker 1980). More recently, however, growth of Alaskan conifers has been found to respond positively to warmer springs and negatively to warmer summers (Barber et al. 2000, Wilmking et al. 2004). Further south in Alberta and Saskatchewan, growth of black spruce (Picea mariana (Mill.) BSP) was negatively correlated with summer temperatures (Dang and Lieffers 1989, Brooks et al. 1998), although growth of jack pine (Pinus banksiana (L.)) was positively correlated (Brooks et al. 1998).

Further insight into the response of forest to warming has been gained from eddy covariance (EC) studies. Morgenstern et al. (2004) found that high air temperatures (Ta) during an El Niño event reduced annual net ecosystem productivity (NEP = GPP – Re) of a temperate coastal Douglas fir stand by raising Re more than GPP. Griffis et al. (2003) found that boreal conifers experienced a pronounced mid-season decline in NEP because warming lowered GPP and raised Re. Grant (2001) showed that changes in GPP versus Re caused boreal black spruce to change from a sink to a source of CO2 at a daily time scale when maximum/minimum temperature rose above 25/15 °C. Dunn et al. (2007) concluded that increases in Re offset those in GPP with warming in a boreal black spruce stand near the treeline and that NEP was controlled by soil water effects on Re, as determined by precipitation and by thawing.

The complex response of forest growth to climate warming is likely the net result of two contrasting responses of net CO2 uptake to rises in Ta: a positive response at lower Ta and a negative response at higher Ta. The CO2 fixation in boreal pine at seasonal time scales has been found to rise sigmoidally with temperatures from –10 to +20 °C (Kolari et al. 2007). However, instantaneous CO2 fixation by coniferous needles declines at Ta > 25 °C (Man and Lieffers 1997). This decline is unlikely to be biochemical in origin because Medlyn et al. (2002) found the optimum Ta for carboxylation reactions in conifers to be > 35 °C, and Méthy et al. (1997) found degradation of PSII photochemical efficiency in Pinus halepensis Mill. only at Ta > 35 °C. The decline at Ta > 25 °C has been attributed to larger vapor pressure deficits (D) (Kloeppel et al. 2000, Law et al. 2000) that lower canopy stomatal conductance (gc), particularly in boreal conifers (Saugier et al. 1997, Grelle et al. 1999, Ohta et al. 2001). These contrasting responses of CO2 fixation to changes in Ta under cooler versus warmer conditions may explain differing responses of growth to warming in different boreal conifer stands (e.g., Barber et al. 2000).

The sensitivity of gc to D in conifers may be caused by their hydraulic structure. Gao et al. (2002) found that transpiration in pines was several times more sensitive to D than that in broadleaf trees, even though gc of pines declined less with canopy water potential ({psi}c) than that of broadleaf trees, due to the structure and composition of their guard cells. They attributed the sensitivity of transpiration in pines to lower xylem hydraulic conductance caused by the tracheid cells from which their xylem is constructed. Lower xylem conductance was hypothesized to force larger soil–canopy water potential gradients and hence lower {psi}c that reduced gc under high transpiration demand. Predawn {psi}c of conifers was found to be lower than that of deciduous shrubs, due to incomplete overnight recharge of plant water caused by low xylem conductance (Royce and Barbour 2001). Consequently, low xylem conductance may limit primary productivity of mature conifers in climates with larger D (Pothier et al. 1989a, 1989b, Menuccini and Grace 1996, Hubbard et al. 1999).

We propose that the concurrent functioning of both positive and negative responses of CO2 fixation to Ta, combined with the positive response of Re to Ta, may explain much of the interannual variation in the productivity of boreal black spruce ecosystems. These responses are represented mathematically in the terrestrial ecosystem model ecosys (Grant 2001) that we tested against EC fluxes recorded over black spruce stands located in different ecoregions of the North American boreal forest during a period of warming from 2004 to 2006.


   Model description
Top
 Notes
 Abstract
 Introduction
 Model description
Field measurements
 Model testing
 Results
 Discussion
 References
 
A comprehensive description of ecosys with a detailed listing of inputs, outputs, governing equations, parameters, results and references can be found in the research articles of Grant (2001, 2004) and Grant et al. (2007a, 2007c). A more detailed description of model algorithms and parameters that are most relevant to simulating temperature effects on forest NEP is given in the following sections.

Energy exchange
Higher Ta can slow down CO2 fixation by raising D and hence transpiration, thereby lowering {psi}c and gc. In ecosys, the effect of D on transpiration is solved through a first-order closure of the energy balance, resolved into that between the atmosphere and the leaf and stem surfaces of each plant population, and that between the atmosphere and each of the ground surfaces (soil, plant residue and snow) beneath (Grant et al. 1999). Sensible and latent heat fluxes from these energy balances are coupled with soil heat and water transfers, including surface runoff (Manning) (Eqs. (A88)–(A92) in Grant 2004), and subsurface infiltration (Green–Ampt) and flows through macropores (Poiseuille), and micropores (Richards) (Eqs. (A93)–(A96) in Grant 2004) that determine soil temperatures (Ts) and water contents ({theta}).

The effect of transpiration on {psi}c and gc is calculated from an hourly two-stage convergence solution for the transfer of water and heat through a multilayered, multi-population soil–root–canopy system. The first stage of this solution requires convergence to a value of canopy temperature, Tc, for each plant population at which the first-order closure of the canopy energy balance is achieved (Eqs. (1)–(15) in Grant et al. 1999). These fluxes are controlled by aerodynamic (ra) and canopy stomatal (rc) resistances (r = g–1). Two controlling mechanisms are postulated for rc which are solved in two successive steps:

  1. At the leaf level, leaf resistance rl controls gaseous CO2 diffusion through each leaf surface when calculating CO2 fixation by C3 or C4 plants (Eqs. (A27) or (A1) in Grant et al. 2007a). The CO2 fixation is initially calculated from concurrent solutions for diffusion (Eqs. (A28) or (A2) in Grant et al. 2007a) and carboxylation (Eqs. (A29) or (A3) in Grant et al. 2007a) under ambient Ca, irradiance, Tc, leaf nutrient content and zero {psi}c (Eqs. (A31) or (A5) in Grant et al. 2007a); Tc directly affects carboxylation through the Arrhenius functions for light and dark reactions, using parameters developed by Bernacchi et al. (2001, 2003) for temperatures from 10 to 40 °C. These functions are extended to temperatures outside this range by adding low and high temperatures inactivation parameters that reproduce temperature sensitivity in Medlyn et al. (2002) and Kolari et al. (2007). Values of rl are aggregated by leaf surface area to a canopy value rc for use in the energy balance convergence scheme (Grant et al. 1999).
  2. At the canopy level, rc rises from zero {psi}c from step (1) through an exponential function of canopy turgor potential, {psi}t, calculated from ambient {psi}c and osmotic water potential, {psi}{pi} (Eq. (25) in Grant et al. 1999), during convergence for transpiration versus water uptake.

Water relations
Water uptake by conifers is thought to be constrained by low xylem hydraulic conductivity. This constraint is modeled in a convergence solution for {psi}c at which the difference between canopy transpiration Ec from the energy balance and total water uptake U from all rooted layers in the soil is tested against the difference between canopy water content from the previous hour and that from the current hour (Eq. (A38) in Grant et al. 2007a, Eq. (B13) in Grant et al. 2007c). The difference between {psi}c and soil water potential, {psi}s, determines U by establishing potential differences across soil–root and root–canopy hydraulic resistances, namely {Omega}s and {Omega}r, in each rooted soil layer (Eqs. (B5)–(B12) in Grant et al. 2007c, Eqs. (32)–(37) in Grant et al. 1999). Values of {Omega}s and {Omega}r are calculated from root radial and axial resistivities using root lengths and surface areas and bole lengths from a root system submodel (Grant 1998). Modeled values of conifer hydraulic conductivity calculated from root and bole axial resistivities in Grant et al. (2007c) were consistent with the values experimentally derived from conifers by Ewers and Zimmermann (1984) and Tyree and Ewers (1991).

GPP
The CO2 fixation may be slowed down by lower {psi}c and gc forced by higher Ta and D. After successful convergence for Tc and {psi}c (described in the section Water relations), the leaf carboxylation rates are adjusted from those calculated under full {psi}t to those under ambient {psi}t. This adjustment is required by the decrease in gc from its value at zero {psi}c (Eqs. A5 (C4) or A31 (C3) in Grant et al. 2007a) to that at ambient {psi}c (Eqs. A4 (C4) or A30 (C3) in Grant et al. 2007a) as calculated in steps (1) and (2), respectively (in the section Energy exchange), and by non-stomatal effects of ambient {psi}t on carboxylation (Eq. (A12) in Grant et al. 2007a, Eq. (4) in Grant and Flanagan 2007). The adjustment is achieved through a convergence solution for Ci at each leaf surface in the canopy at which the diffusion rate of gaseous CO2 between Cb and Ci through gl (Eqs. (48)–(53) in Grant et al. 1999) equals the carboxylation rate of aqueous CO2 at mesophyll CO2 concentration Cm, the temperature-dependent aqueous counterpart of Ci (Eqs. (38)–(47) in Grant et al. 1999). The CO2 fixation rate of each leaf surface at convergence is added to calculate GPP by each branch of each plant population (i.e., species or cohort) in the model. The CO2 fixation product is stored in nonstructural C pools ({sigma}C) in each branch that drives Ra.

Provision is made in the model to activate carboxylation reactions after a set number of hours accumulated above a threshold temperature during lengthening photoperiods in spring, based on the observations of seasonal changes in photosynthetic capacity of boreal conifers (e.g., Bergh and Linder 1999, Kolari et al. 2007).

Autotrophic respiration
The Ra responds to Ta differently than does CO2 fixation; Ra in each plant branch or root and mycorrhizal layer is first driven by the oxidation of {sigma}C (Rc) (Eq. (C13) in Grant et al. 2007c) multiplied by functions of Tc or Ts (Arrhenius) and nutrient status (from nonstructural C:N:P ratios). The Rc is first used to meet maintenance respiration requirements (Rm from plant N using Tc or Ts with a Q10 of 2.25), then any excess is expended as growth respiration Rg, constrained by a linear function of {psi}t (Eq. (C16) in Grant et al. 2007c). The Rg is therefore sensitive to Tc or Ts through Rc. The Rg drives the conversion of {sigma}C into structural material in branches and roots according to organ-specific growth yields and phenology-dependent partitioning coefficients. When Rm exceeds Rc, the shortfall is met by the respiration of remobilizable C (Rr) in shoots or roots and mycorrhizae. The Rr drives the withdrawal of remobilizable C, N and P (nonstructural) from shoots or roots and mycorrhizae into {sigma}N and {sigma}P, and the loss of associated non-remobilizable C, N and P (structural) as litterfall (Eq. (C17) in Grant et al. 2007c).

The Ra of each branch or root and mycorrhizal layer is the total of Rc and Rr, and NPP is the difference between GPP and total Ra of all branches and root and mycorrhizal layers. Phytomass net growth is the difference between NPP and litterfall (Eq. (C19) in Grant et al. 2007c).

Nutrient uptake
Uptake of N and P is calculated by solving for solution [NH4+], [NO3–] and [H2PO4–] at root and mycorrhizal surfaces at which radial transport by mass flow and diffusion from the soil solution equals active uptake at the surfaces (Eq. (A36) in Grant et al. 2007c). Path lengths and surface areas of roots and mycorrhizal uptake are calculated from a root and mycorrhizal growth submodel (Eqs. (1)–(17) in Grant 1998). Products of N and P uptake are added to nonstructural pools where they are coupled with nonstructural C pools to drive growth of branches, roots and mycorrhizae as described in the section Autotrophic respiration. The translocation of nonstructural C, N and P among branches, roots and mycorrhizae is driven by concentration gradients generated by production through uptake versus consumption through growth, thus maintaining a functional equilibrium among branches, roots and mycorrhizae which adapts to changing resource availability. Leaf nonstructural N and P concentrations determine leaf structural N and P contents used to calculate leaf carboxylation rates as described in the section Energy exchange.

Heterotrophic respiration
Higher Ta causes higher Ts that hastens organic C, N and P transformations and hence Rh. These transformations occur in five organic matter–microbe complexes (coarse woody litter, fine nonwoody litter, manure, particulate organic matter and humus) in surface detritus and in each soil layer. Decomposition rates of different substrates in each complex are combined functions of substrate concentration (Eqs. (A3) and (A4) in Grant et al. 2007c), of active biomass in heterotrophic microbial populations (Mh) (Eqs. (A1) and (A2) in Grant et al. 2007c), of Ts (the Arrhenius function in Eq. (A5) of Grant et al. 2007c), and of {theta} (Eqs. (A3) and (A4) of Grant et al. 2007c).

Decomposition products are dissolved organic carbon (DOC), N and P and are the substrates for Rh by all Mh in each substrate–microbe complex (Eq. (A11) in Grant et al. 2007c). The Rh is a combined function of DOC concentration (Monod), Ts (Arrhenius) and microbial N or P contents. Aerobic Rh is constrained by O2 uptake (Eqs. (A12)–(A14) in Grant et al. 2007c) calculated from O2 diffusivity to microbial surfaces (Eqs. (A15) and (A16) in Grant et al. 2007c); Rh that is not coupled with O2 uptake is coupled with the sequential reduction of NO3–, NO2– and N2O by heterotrophic denitrifiers, and with the reduction of organic C by fermenters and acetotrophic methanogens. Some soil CO2 exchange is driven by autotrophic microbial populations Ma, including nitrifiers that conduct NH4+ and NO2– oxidation, and NO2– reduction, and autotrophic methanogens and methanotrophs that conduct CH4 production and oxidation.

All Mh and Ma undergo maintenance respiration (Rm from microbial N using Ts with a Q10 of 2.25) (Eqs. (A17) and (A18) in Grant et al. 2007c). All Mh and Ma also undergo decomposition (Eq. (A22) in Grant et al. 2007c). The Rh in excess of Rm is used in growth respiration, Rg (Eq. (A19) in Grant et al. 2007c), the energy yield of which drives growth in Mh and Ma from uptake of DOC or CO2 according to the energy requirements of biosynthesis (Eqs. (A20) and (A21) in Grant et al. 2007c). The Rm in excess of Rh accelerates microbial decomposition, the products of which are humified according to soil clay content (Eqs. (A34) and (A35) in Grant et al. 2007c). Net growth in Mh and Ma drives mineralization – immobilization (Eq. (A25) in Grant et al. 2007c), thereby controlling solution [NH4+], [NO3–] and [H2PO4–] that determine root and mycorrhizal uptake and hence plant nutrient status as described in the section Nutrient uptake.


   Field measurements
Top
 Notes
 Abstract
 Introduction
 Model description
 Field measurements
Model testing
 Results
 Discussion
 References
 
The ability of ecosys to simulate environmental controls on productivity of boreal black spruce was tested with the data recorded at boreal old black spruce sites in Saskatchewan (SOBS 53.99° N and 105.12° W), Manitoba (NOBS 55.88° N and 98.48° W) and Quebec (EOBS 49.69° N and 74.34° W) of the Fluxnet-Canada Research Network (FCRN) during successively warmer years from 2004 to 2006 (Table 1). These sites are considered typical of mature black spruce stands in these different regions. Descriptions of sites and EC measurement protocol at SOBS, NOBS and EOBS are given in Bergeron et al. (2007), Dunn et al. (2007) and Krishnan et al. (2008). Gaps in EC measurements caused by inadequate turbulence or equipment malfunction were filled using FCRN protocol described by Barr et al. (2004). Gap-filled fluxes were used to derive GPP, Re and NEP as described in Bergeron et al. (2007), Dunn et al. (2007) and Krishnan et al. (2008).


View this table:
[in this window]
[in a new window]

  		Table 1. Average annual temperature and total precipitation recorded during 2004–2006 at old black spruce sites in Saskatchewan (SOBS), Manitoba (NOBS) and Quebec (EOBS).

 

   Model testing
Top
 Notes
 Abstract
 Introduction
 Model description
 Field measurements
 Model testing
Results
 Discussion
 References
 
Modeled versus measured CO2 fluxes and NEP
Before testing ecosys with the CO2 fluxes recorded at the three black spruce sites, the model had to reproduce site conditions by simulating site history. This was accomplished by initializing ecosys at each site with the biological properties of black spruce and moss (Grant et al. 2001a, 2001b, Grant 2004), and with the physical and chemical properties of the Cumic Humic Regosols at SOBS (Table 1 in Grant et al. 2001b) and NOBS (Table 1 in Grant et al. 2001a), and the ferro-humic podzol at EOBS (Table 2). Ecosys was also initialized with stocks of coarse and fine residue estimated to remain after a stand-replacing fire. For the ecosys runs at SOBS, black spruce and moss were seeded in the model year 1796, and grown from 1797 to 1900 under eight cycles of continuous hourly weather data (radiation, Ta, dewpoint, wind speed and precipitation) recorded at SOBS from 1994 to 2006. Black spruce and moss were burnt in 1901, reseeded in 1902, and regrown from 1903 to 2006 under another eight cycles of weather data. For the run at NOBS, black spruce and moss were seeded in the model year 1692, grown from 1693 to 1848 under 12 cycles of hourly weather data recorded at NOBS from 1994 to 2006, burnt in 1849, reseeded in 1850, and regrown from 1851 to 2006 under another 12 cycles of weather data. For the run at EOBS, black spruce and moss were seeded in the model year 1808 and then grown from 1809 to 1896 under hourly weather data recorded at EOBS during 2005. Black spruce and moss were burnt in 1897, reseeded in 1898, and regrown from 1899 to 2006 under 2005 weather until 1919, under daily weather data (radiation, maximum and minimum Ta, relative humidity, wind speed and precipitation) constructed from meteorologic records during 1920–2003 by Régnière and St-Amant (2007), and then from 2004 to 2006 under hourly weather data recorded at EOBS during 2004–2006. These chronologies enabled the ages of the modeled stands in 2004–2006 to be similar to those at the field sites as derived from fire scars. A background mortality rate of 0.8% (SOBS and NOBS) or 1.2% (EOBS) per year was applied to spruce during the model runs, simulating natural self-thinning of the stands (Aakala et al. 2007). During the model runs, Ca rose exponentially from 280 to 385 µmol mol–1, and precipitation NH4+ and NO3– concentrations used to simulate wet N deposition rose exponentially from historic values based on Holland et al. (1999) at the start of each run to current values based on Meteorological Service of Canada (2004) at the end of each run. Atmospheric concentration of NH3 used to simulate dry N deposition was maintained at 0.002 µmol mol–1.


View this table:
[in this window]
[in a new window]

  		Table 2. Representative physical and chemical properties of the ferro-humic podzol measured at the Quebec old black spruce site (EOBS) and used in ecosys.

 
The ability of the model to simulate NEP at hourly and annual time scales was evaluated by comparing CO2 fluxes simulated during 2004–2006 of the model runs with EC and surface chamber measurements recorded at SOBS, NOBS and EOBS during 2004–2006.

Model predictions of CO2 fluxes and NEP under higher temperatures
The sensitivity of modeled CO2 exchange to sustained rises in Ta, such as those predicted under future boreal climates (IPCC 2007), was examined by initializing ecosys with output from the end of the model year 2000 in the run at each site, and then running the model under weather recorded during 2001–2006 with hourly Ta raised by 2, 4, 6, or 8 °C. During these runs, hourly relative humidity was assumed to remain unchanged so that D rose with Ta. Changes in CO2 exchange under elevated Ta reached stable values by the model year 2003 (the 3rd year under elevated Ta). These changes were then examined from 2004 to 2006 at diurnal, seasonal and annual time scales to determine how modeled NEP might change under higher Ta.


   Results
Top
 Notes
 Abstract
 Introduction
 Model description
 Field measurements
 Model testing
 Results
Discussion
 References
 
Daily NEP
Seasonal temperature effects on daily NEP
Daily NEP, calculated as daily totals of hourly (modeled) or 1/2-hourly (EC) CO2 fluxes, remained negative (net C source) at all three sites until April, rose to maximum positive values (net C sink) during May and early June, declined rapidly in July, remained low or rose slightly in August, declined again in September and became negative by early October during 2004–2006 at SOBS (Figure 1), NOBS (Figure 2) and EOBS (Figure 3). Rises in NEP during April and May coincided with rises in max./min. Ta above 10/0 °C, declines in NEP during July and August coincided with rises in max./min. Ta above 25/15 °C, and declines in NEP during September and October coincided with declines in max./min. Ta below 10/0 °C.


Figure 1
View larger version (59K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Left: daily net ecosystem productivity (3-day moving mean values) calculated as daily totals of 1/2-hourly gap-filled EC (symbols) or hourly model (lines) CO2 fluxes and right: hourly-averaged air temperatures during 2004–2006 at SOBS. Open symbols represent EC values with < 24 accepted 1/2-hour EC flux measurements. A color version of this figure is available as Supplementary Data at Tree Physiology Online.

 

Figure 2
View larger version (59K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Left: daily net ecosystem productivity (3-day moving mean values) calculated as daily totals of 1/2-hourly gap-filled EC (symbols) or hourly model (lines) CO2 fluxes and right: hourly-averaged air temperatures during 2004–2006 at NOBS. Open symbols represent EC values with < 24 accepted 1/2-hour EC flux measurements. A color version of this figure is available as Supplementary Data at Tree Physiology Online.

 

Figure 3
View larger version (57K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3. Left: daily net ecosystem productivity (3-day moving mean values) calculated as daily totals of 1/2-hourly gap-filled EC (symbols) or hourly model (lines) CO2 fluxes and right: hourly-averaged air temperatures during 2004–2006 at EOBS. Open symbols represent EC values with < 24 accepted 1/2-hour EC flux measurements. A color version of this figure is available as Supplementary Data at Tree Physiology Online.

 
Tests of modeled versus EC-derived daily NEP
Regressions of modeled versus EC-derived daily NEP during 2004–2006 gave intercepts within 0.13 g C m–2 day–1 of zero, slopes between 0.95 and 1.1 except at EOBS in 2005 and 2006 which were higher, correlation coefficients (R2) of 0.5–0.7 (P < 0.0001) and root mean squares for differences (RMSD) of 0.5–0.6 g C m–2 day–1 (Table 3a). These regressions indicated a tendency for a slightly greater variation in modeled versus EC-derived NEP caused by larger effluxes modeled during autumn and winter, and larger peak influxes modeled during late spring and early summer. However, regression parameters suggested that bias in annual aggregations of NEP arising from this variation should be small.


View this table:
[in this window]
[in a new window]

  		Table 3. Intercepts (a), slopes (b), coefficients of determination (R2) and root mean square of differences (RMSD) from regressions of modeled CO2 fluxes versus (a) daily-aggregated and (b and c) hourly-averaged eddy covariance CO2 fluxes (b) measured and (c) gap-filled during 2004–2006 at old black spruce sites in Saskatchewan (SOBS), Manitoba (NOBS) and Quebec (EOBS) and (d) versus hourly-averaged soil CO2 fluxes (averages of nine values less standard errors) measured by automated surface flux chambers.

 
Hourly CO2 exchange
Diurnal temperature effects on CO2 exchange
Seasonal variability in NEP was associated with seasonal variability in Ta through effects of Ta on net CO2 exchange. Earlier rises in spring Ta from 2004 to 2006 caused earlier rises in daily NEP (Figures 1–3GoGo), driven by earlier rises in hourly CO2 influxes modeled and measured at all three sites (e.g., for SOBS in Figure 4). These rises were modeled from parameters for temperature sensitivity functions of radiation- and CO2-limited carboxylation reactions developed by Bernacchi et al. (2001, 2003) and used in ecosys.


Figure 4
View larger version (82K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4. (A) Solar radiation, air temperature; (B) energy fluxes; and (C) CO2 fluxes measured (closed symbols), gap-filled (open symbols) or modeled (lines) at SOBS during DOY 130–139 in 2004–2006. Downward fluxes are represented by positive values and upward fluxes by negative values. A color version of this figure is available as Supplementary Data at Tree Physiology Online.

 
Higher Ta during July and August caused sharp declines in daily NEP, followed by partial recovery with subsequent cooling (Figures 1–3GoGo). These declines were driven by declines in hourly CO2 influxes and rises in hourly CO2 effluxes modeled and measured under higher Ta. These changes in CO2 fluxes were apparent when a warming front passed from west to east through SOBS, NOBS and EOBS during day of the year (DOY) 185–193 in 2005, raising Ta above 25 °C for several days at each site (Figure 5A). Warming caused earlier mid-afternoon declines in modeled CO2 influxes (Figure 5C) when large conifer {Omega}r forced lower {psi}c and hence {psi}t that drove lower gc that in turn constrained further declines in {psi}c in the convergence solutions described in the sections Energy exchange and Water relations. The stomatal constraint to CO2 uptake that was modeled under higher Ta was corroborated by the absence of any rise in latent heat of evaporation (LE) effluxes modeled or measured under rises in midday D from 1 to 3 kPa that accompanied the rises in Ta, except briefly following a rainfall event on DOY 188 at NOBS (Figure 5B).


Figure 5
View larger version (82K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5. (A) Solar radiation, air temperature; (B) energy fluxes; and (C) CO2 fluxes measured (closed symbols), gap-filled (open symbols) or modeled (lines) at SOBS, NOBS and EOBS during DOY 184–193 in 2005. Downward fluxes are represented by positive values and upward fluxes by negative values. A color version of this figure is available as Supplementary Data at Tree Physiology Online.

 
Warming also caused CO2 effluxes in the model to rise according to the Arrhenius functions for Ra and Rh described in the sections Autotrophic respiration and Heterotrophic respiration. Much of these effluxes originated in the soil, as indicated by rising CO2 emissions modeled and measured by automated surface flux chambers during periods of soil warming (e.g., EOBS in Figure 6). These emissions were largely offset by CO2 uptake into moss during daytimes, causing net CO2 fluxes to approach zero, but attained rates close to those measured or gap-filled from the EC flux tower during nights (e.g., EOBS 2005 in Figure 5 versus Figure 6). Smaller CO2 influxes and larger CO2 effluxes modeled and measured at higher Ta (Figure 5) caused sharp declines in daily NEP when max./min. Ta exceeded 25/15 °C at all three sites in all 3 years (Figures 1–3GoGo).


Figure 6
View larger version (42K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 6. Soil temperatures and surface CO2 fluxes measured (symbols: mean values of nine chambers; dashes: mean values less standard deviations of nine chambers) and modeled (lines) during periods of soil warming at EOBS in 2004–2006. Downward fluxes are represented by positive values and upward fluxes by negative values. A color version of this figure is available as Supplementary Data at Tree Physiology Online.

 

Figure 7
View larger version (54K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 7. Relationships between air temperature and CO2 influxes (closed symbols) when solar irradiance > 500 W m–2, and between air temperature and CO2 effluxes (open symbols) when solar irradiance = 0 W m–2 modeled and measured during 2004–2006 at SOBS. A color version of this figure is available as Supplementary Data at Tree Physiology Online.

 
Tests of modeled versus EC CO2 fluxes
Regressions of modeled CO2 fluxes versus hourly-averaged EC CO2 fluxes measured during 2004–2006 gave intercepts within 0.4 µmol m–2 s–1 of zero, and slopes between 0.95 and 1.12 (Table 3b). These regression parameters indicated minimal bias, but a tendency for variation in modeled fluxes slightly to exceed that in measured values, possibly because EC fluxes were not corrected for incomplete energy balance closure that was 85–90% for most site-years. Values of R2 and RMSD between modeled and EC fluxes were 0.7–0.8 (P < 0.0001) and 1.2–1.8 µmol m–2 s–1, respectively. Both daily and hourly RMSDs at SOBS and NOBS were considerably smaller than values from earlier modeling studies at these sites (Amthor et al. 2001, Arain et al. 2002, Yuan et al. 2008), indicating the ongoing progress in modeling CO2 exchange over boreal forests. Hourly RMSDs were also smaller than random errors in EC CO2 flux measurements, estimated by Richardson et al. (2006) to rise from ~ 1.4 to ~ 3.5 µmol m–2 s–1 for net CO2 influxes from 0 to 10 µmol m–2 s–1 and for net CO2 effluxes from 0 to 5 µmol m–2 s–1, the range of fluxes measured at the experimental sites. Much of the unexplained variance in EC fluxes could be attributed to a random error of ca. 20% in EC methodology (Wesely and Hart 1985).

Tests of modeled versus gap-filled and soil CO2 fluxes
Modeled CO2 fluxes were well correlated with gap-filled values used to replace EC measurements during unfavorable conditions (b = 0.97 – 1.14, R2 ~ 0.85, P < 0.0001 in Table 3c), except at EOBS in 2006 where b was larger. Ratios between modeled and gap-filled fluxes tended to be slightly larger than those between modeled and measured fluxes (b in Table 3c versus 3b), except at NOBS. These regression parameters indicated that no substantial bias was introduced into modeled versus EC-derived comparisons of annual NEP (see below) by including gap-filled values in the latter.

Soil CO2 effluxes in the model were usually smaller than the average of those measured from nine automated surface chambers at EOBS (Figure 6), although concurrent ecosystem CO2 effluxes in the model were close to EC measurements (e.g., EOBS 2005 in Figure 5 versus Figure 6). The difference between ecosystem and soil effluxes in the model represented the contribution of aboveground Ra that can be a substantial fraction of Re (Lavigne et al. 1997). Bergeron (2007) estimated that measurements with the automated chambers may have overestimated the CO2 effluxes by 28%. Modeled soil effluxes were well correlated with the mean measured values less standard deviations representing their spatial variability (b ~ 0.8, R2 ~ 0.6, P < 0.0001).

Differences between modeled and EC CO2 fluxes
Variation in EC and gap-filled CO2 fluxes not explained by the model (Table 3) were mostly attributed to three sources:

  1. Variable EC CO2 influxes frequently recorded during daylight hours with apparently stable weather conditions, during which scattering in the EC CO2 influxes caused modeled values to exceed most of the measured values (e.g., DOY 136 versus 135 at SOBS 2006 in Figure 4). This scattering in EC fluxes requires further examination.
  2. Anomalously large EC CO2 fluxes occasionally recorded at an hour or two after sunrise (e.g., DOY 188, 190–193 at EOBS 2005 in Figure 5) were not modeled. These fluxes were usually associated with large changes in canopy CO2 storage, likely caused by rapidly changing turbulence with early morning warming.
  3. Larger rises in modeled versus gap-filled CO2 effluxes during soil warming (e.g., DOY 185–187 at SOBS 2005 and 188–191 at EOBS 2005 in Figure 5), even with comparable rises in modeled versus measured soil CO2 effluxes (e.g., DOY 188–191 at EOBS 2005 in Figure 6). The larger rises in modeled ecosystem effluxes were partially driven by the response of aboveground Ra to rising Ta not accounted for in gap-filling protocol at SOBS and EOBS that is based on Ts. Agreement between modeled and gap-filled effluxes was better at NOBS (b closer to 1 in Table 3c) where gap-filling was based on Ta.

General temperature effects on modeled and EC CO2 exchange
The CO2 influxes measured and modeled under non-limiting irradiance during 2004–2006 rose with Ta to 15 °C due to more rapid carboxylation (Figure 4), and fell at Ta above 20 °C due to declining gc (Figure 5). These declines were driven by higher D so that CO2 influxes fell sharply when D > 1.5 kPa (Figure 8). The CO2 effluxes measured and modeled under zero irradiance rose nonlinearly with Ta (Figure 7), driven in the model by the Arrhenius functions for Ra and Rh (e.g., Figure 6). Rises in mean annual Ta from 2004 to 2005 and 2006 (Table 1) caused a number of influxes measured and modeled at suboptimum Ta (< 10 °C) in 2004 to be replaced by higher influxes measured and modeled at more optimum Ta (10–20 °C) in 2005 and 2006, respectively (e.g., Figure 4). These gains in CO2 influxes with warming were only partially offset by smaller increases in the number of lower influxes measured and modeled at superoptimum Ta (> 20 °C) in 2005 and 2006 (e.g., Figure 5).


Figure 8
View larger version (12K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 8. Relationships between vapor pressure deficit and CO2 influxes when solar irradiance > 500 W m–2 modeled and measured during 2005 at SOBS.

 
Annual ecosystem productivity
GPP
The increased frequency of CO2 influxes occurring at more optimum Ta from 2004 to 2005 and 2006 caused comparatively similar rises in annual GPP from 2004 to 2006 in the model and in values derived from gap-filled EC measurements at SOBS, NOBS and EOBS (modeled versus EC-derived annual GPP: R2 = 0.70, P < 0.005, n = 9) (Table 4). Annual GPP in the model varied with mean annual Ta at the three sites (SOBS > EOBS > NOBS, 2006 > 2005 > 2004 in Table 1), as did gap-filled GPP except at EOBS where it remained consistently lower than that at the other two sites. Bergeron et al. (2007) found that GPP at NOBS and EOBS did not differ significantly in 2004. Moss contributed 20–25% of ecosystem GPP in the model, as was estimated by Goulden and Crill (1997) from earlier measurements of surface CO2 exchange at NOBS. The important contribution of moss to GPP at EOBS was apparent in the large diurnal amplitude of soil CO2 fluxes caused by moss CO2 uptake (Figure 6).


View this table:
[in this window]
[in a new window]

  		Table 4. Gross primary productivity (GPP), ecosystem respiration (Re) and net ecosystem productivity (NEP) modeled (M) and derived from gap-filled eddy covariance fluxes (EC), and root + mycorrhizal N uptake (NH4+ + NO3–) modeled during 2004–2006 at old black spruce sites in Saskatchewan (SOBS), Manitoba (NOBS) and Quebec (EOBS).

 
The modeled rise in GPP from 2004 to 2005 and 2006 was sustained by a rise in N uptake by spruce and moss driven by greater root length density and more rapid active uptake. This rise occurred because functional equilibria in the model increased allocation of plant nonstructural C, N and P to root growth and nutrient uptake under increased CO2 fixation with Ta, as described in the section Nutrient uptake, thereby rebalancing N uptake with CO2 fixation. Rises in N uptake allowed mid-season foliar C:N ratios to be maintained at 58 (SOBS), 63 (NOBS) and 61 (EOBS) during 2004–2006, similar to the values of 57, 63 and 63 for these sites reported in 2004 by Bergeron et al. (2007).

Ra and NPP
Modeled Ra and hence NPP varied with GPP among years, so that the ratio of spruce annual NPP:GPP in the model remained within a narrow range of 0.40–0.44 (Table 4). These ratios were smaller than the ones of 0.47 found for a temperate conifer by Jassal et al. (2007) and 0.45 for a wide range of forest types by Waring and Running (1998), but larger than the ones of 0.28 and 0.23 found by Ryan et al. (1997) from scaled chamber measurements at SOBS and NOBS.

Of this Ra, 0.39–0.44 was allocated belowground (= soil respiration Rs – Rh in Table 4) depending on year, compared to ratios of 0.24 and 0.46 found by Ryan et al. (1997) at SOBS and NOBS. Annual belowground Ra modeled at SOBS in 2004, excluding moss, was (245 – 17) = 228 g C m–2, but root-driven respiration Rr in the model rose to (228 + 110) = 338 g C m–2 by adding microbial respiration of root exudates. This value was comparable to annual rhizospheric respiration of 285 g C m–2 measured and gap-filled at SOBS during 2004 by Gaumont-Guay et al. (2008) with automated soil flux chambers in control versus root exclusion plots. The ratio of belowground Ra to Rs was 0.36–0.41 (Table 4), but the ratio of Rr to Rs rose to 0.48–0.54, similar to the one of 0.52 for rhizosphere to total soil respiration (excluding moss) at SOBS in 2004 derived by Gaumont-Guay et al. (2008).

Rh and NEP
Annual Rs modeled at SOBS in 2004, including aboveground moss Ra, was (655 + 51) = 706 g C m–2, versus 611 g C m–2 measured by Gaumont-Guay et al. (2008). Annual Rh modeled at SOBS in 2004, excluding microbial respiration of root exudates, was (410 – 110) = 300 g C m–2, versus 269 g C m–2 measured and gap-filled in root exclusion plots by Gaumont-Guay et al. (2008); Rh and hence Rs changed little from 2004 to 2006 because limitations to microbial activity from soil drying in the summers of 2005 and 2006 offset gains in microbial activity from soil warming. Declines in respiration with soil drying at SOBS in the summer of 2006 were found experimentally by Krishnan et al. (2008). Annual Rs in the model contributed 0.61–0.68 of Re (Table 4), within the range of 0.48–0.71 reported by Lavigne et al. (1997) from scaled chamber measurements in diverse boreal forest stands, and close to a ratio of 0.62 derived by Jassal et al. (2007) from automated chamber and EC measurements in a temperate coniferous stand. The remaining 0.32–0.39 of Re was contributed by aboveground phytomass. However, EC-derived Re was close to chamber-derived Rs (e.g., 665 versus 611 g C m–2 at SOBS in 2004), as were EC and chamber CO2 effluxes (e.g., EOBS 2005 in Figures 5 and 6), so that EC-derived Re remained less than the modeled values.

Both modeled and EC-derived annual Re rose similarly with mean annual Ta from 2004 to 2006 (modeled versus EC-derived annual Re: R2 = 0.72, P < 0.005, n = 9) (Table 4). Modeled Re rose slightly less than did GPP with warming at each site so that NEP rose with GPP (Table 4). However, EC-derived Re sometimes rose more with warming than did GPP, so that EC-derived NEP did not always rise with mean annual Ta (e.g., EOBS 2005 in Table 4). However, EC-derived annual NEP is the sum of 17,520 measured or gap-filled 1/2-hourly fluxes (or 17,568 in a leap year), the accumulated uncertainty of which has been estimated to be ±21 g C m–2 year–1 at SOBS (Krishnan et al. 2008). As these fluxes are the net result of much larger opposing fluxes, their uncertainty is large with respect to their sum. Nonetheless, there was a tendency to derive larger NEP from EC measurements in 2006 versus 2004 and 2005, although this tendency was smaller than that modeled.

Annual GPP and Re in the model exceeded those derived from EC measurements by 30–40% (Table 4), although hourly net CO2 fluxes (= GPP – Re) in the model were within 10% of EC measurements (Table 3). These differences could be partly accounted for by correcting EC-derived GPP and Re for incomplete energy balance closure, which would raise annual values at SOBS by 10% (Krishnan et al. 2008) and at EOBS by 20% (data not shown), and by resolving differences in modeled versus EC and gap-filled CO2 fluxes discussed in the section Hourly CO2 exchange. Long-term NEP in the model allowed accumulation of wood C during forest regeneration since last stand-replacing fires that corresponded with wood growth derived from recent inventory measurements (Figure 9).


Figure 9
View larger version (15K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 9. Wood C growth modeled (lines) and estimated from inventory measurements (symbols) after stand-replacing fires corresponding approximately in time to those from which the current stands at SOBS, NOBS and EOBS are regenerated. Abbreviations: CCP, inventory at SOBS and EOBS in 2004 by the Canada Carbon Project; Gower et al. (1997), inventory in 1994 at SOBS and NOBS; AFS, Alberta Forest Service (1985) inventory growth curve for black spruce at medium site index; Pothier and Savard (1998), merchantable wood (dbh > 9 cm) in the Chibougamau region. A color version of this figure is available as Supplementary Data at Tree Physiology Online.

 
Changes in CO2 exchange and NEP under rising temperatures
The contrasting effects of warming on net CO2 exchange at different Ta (Figures 4, 5 and 7) caused rises in Ta of 2, 4, 6 and 8 °C to have contrasting effects on net CO2 exchange modeled at different times of the year (e.g., at EOBS during 2005 in Figure 10). Fluxes modeled under current Ta (+0 °C) were compared with EC values in Figures 3 and 5. CO2 influxes rose with higher Ta in spring but declined in summer, whereas CO2 effluxes rose with higher Ta in both spring and summer (Figure 10A and B). Influxes rose more than did effluxes with warming at lower Ta but less at higher Ta, so that daily NEP modeled during the year rose with warming at lower Ta and fell at higher Ta (Figure 10C).


Figure 10
View larger version (42K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 10. Changes in (A, B) CO2 fluxes and (C) net ecosystem productivity (NEP) modeled at EOBS during 2005 under rises in air temperature of 0, 2, 4 and 6 °C. Numbers in brackets in (C) are annual totals of NEP in g C m–2 year–1.

 
Larger influxes and lengthening growing seasons caused annual NEP in the model to rise with increases in Ta of up to 6 °C, corresponding to a mean annual Ta of 6–8 °C (Figure 11). However, further increases in Ta increased the frequency and intensity of C losses during high Ta events, so that annual NEP did not rise with further warming. These increases in Ta also exacerbated water limitations at the mid-continental sites with lower precipitation in Saskatchewan (SOBS) and Manitoba (NOBS) (Table 1), causing substantial declines in NEP.


Figure 11
View larger version (6K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 11. Annual net ecosystem productivity (NEP) versus mean annual temperature (MAT) modeled at SOBS, NOBS and EOBS with weather data from 2004–2006 raised by 0, 2, 4, 6 and 8 °C.

 
Rises in NEP with Ta were sustained by more rapid N uptake driven by more rapid Rh and N mineralization with soil warming (described in the section Heterotrophic respiration). For example, more rapid N uptake caused foliar C:N ratios modeled at EOBS to decline from 61 under ambient Ta to 58, 53, 49 and 46 with rises in Ta of 2, 4, 6 and 8 °C, respectively, during August 2005, the 5th year under elevated Ta. Rises in modeled foliar N densities contributed to more rapid CO2 fixation that raised leaf area index (LAI) modeled at EOBS in August 2005 from 2.81 (projected) under ambient Ta to 2.93 and 3.09 with rises in Ta of 2 and 4 °C. However, more rapid respiration and hence litterfall caused declines in modeled LAI to 3.08 and 2.97 with rises in Ta of 6 and 8 °C.


   Discussion
Top
 Notes
 Abstract
 Introduction
 Model description
 Field measurements
 Model testing
 Results
 Discussion
References
 
Variation in GPP and Re with temperature
Both modeled and EC-derived annual GPP rose with mean annual Ta (Table 4), as found in meta-analyses of forest CO2 exchange (Arain and Restrepo-Coupe 2005, Luyssaert et al. 2007). In the model, Re and hence NEP rose with GPP because Ra was driven by the CO2 fixation product {sigma}C at hourly to daily time scales (described in the section Autotrophic respiration), and Rh was driven mostly by plant litterfall derived from growth products of {sigma}C at seasonal to interannual time scales (described in the section Heterotrophic respiration). Model coefficients used to calculate Ra and Rh from C substrates were well constrained from basic research (e.g., Waring and Running 1998). Therefore, CO2 effluxes in the model were not independent of CO2 influxes, but rather were dependent on these influxes over a range of time scales, as observed experimentally by Ekblad and Högberg (2001) and Ekblad et al. (2005). Therefore, modeled Re and hence NEP tended to vary with GPP (Table 4), consistent with a linear relationship between GPP and NEP reported across terrestrial ecosystems with GPP from 600 to 2200 g C m–2 year–1 (Law et al. 2002). However, this relationship in the model could be altered by drought (Grant et al. 2006), forest age (Grant et al. 2007b, 2007c), and other factors. EC-derived Re also rose with mean annual Ta (Table 4), but varied more independently of GPP so that NEP did not rise with Ta as consistently as did modeled values.

Variation in NEP with temperature
The response of NEP to warming in the model was the net result of two opposing responses: a rise with warming at Ta < 15 °C (Figure 4), and a decline with warming at Ta > 20 °C (Figure 5) as seen in Figures 7 and 9. Rises in CO2 uptake with warming in spring (e.g., Figures 4 and 9A) and declines with warming in summer (e.g., Figures 5 and 9B) have been observed in boreal black spruce elsewhere (Myneni et al. 1997, Arain et al. 2002, Welp et al. 2007, Krishnan et al. 2008). A rise in CO2 uptake with spring warming would explain the positive correlation between tree ring index and spring temperatures in boreal conifers found by Wilmking et al. (2004). A decline in CO2 uptake with summer warming would explain the negative correlation between tree ring index and June–August maximum temperatures found in boreal black spruce by Dang and Lieffers (1989) and in other boreal conifers by Barber et al. (2000) and Wilmking et al. (2004).

The rise in NEP with warming at lower Ta in spring or autumn was attributed to three sources in the model:

  1. A lengthening of the C net uptake period by 15–30 days in spring (modeled versus measured lengthening from 2004 to 2006 in days: 16 versus 16 at SOBS in Figure 1, 31 versus 42 at NOBS in Figure 2 and 25 versus 25 at EOBS in Figure 3) with warming of ca. 2 °C in 2005 and 2006 versus 2004 (Table 1). Lengthening arose from seasonal temperature effects on photosynthetic capacity (e.g., Bergh and Linder 1999, Jarvis and Linder 2000, Medhurst et al. 2006), as reported in earlier and current studies at SOBS (Arain et al. 2002, Krishnan et al. 2008). Similar lengthening of the C net uptake period was modeled under 2 °C rises in Ta (Figure 10C). There may be some evidence from satellite images and CO2 measurements that warming is already causing boreal growing seasons to lengthen. Myneni et al. (1997) estimated from satellite normalized difference vegetation index (NDVI) that the boreal growing season had lengthened by 8 days in spring and 4 days in autumn in response to a rise in Ta of ca. 1 °C from 1982 to 1990. Growing season lengths estimated from NDVI in boreal coniferous forests have been positively correlated with C net uptake duration, and thereby with NEP (Churkina et al. 2005). However, rises in NDVI and NPP with autumn warming may be offset by greater rises in Rh, causing earlier declines in late-season C net uptake (Piao et al. 2008). Such earlier declines could not be corroborated from seasonal NEP measured or modeled at the sites in this study.
  2. A rise in carboxylation activity (Figure 4), based on parameters for the Arrhenius temperature functions of key carboxylation processes derived by Bernacchi et al. (2001, 2003), extended to lower and higher temperatures by parameters for low and high temperature inactivation. These parameters reproduced the temperature sensitivity of CO2 fixation in boreal and temperate conifers found by Medlyn et al. (2002) and Kolari et al. (2007).
  3. More rapid N uptake that maintained or lowered foliar C:N ratios during warming so that N constraints on CO2 fixation remained constant or declined with rising Ta. Foliar nutrient concentrations of boreal conifers have been observed to rise in soil warming experiments, indicating accelerated mineralization and uptake (Jarvis and Linder 2000). This might explain the tendency towards higher foliar N concentrations in conifers growing in warmer climates found in a meta-analysis of spruce and pine productivity by Ladanai and Ågren (2004).

The decline in NEP with warming at higher Ta in summer was attributed to three sources in the model:

  1. A sharp decline in gc under higher D, modeled from a hydraulic constraint to water uptake imposed by low axial conductivity of conifer xylem in roots and boles (Tyree and Ewers 1991) as in Gao et al. (2002). Low axial conductivity forced declines in {psi}c and hence in gc to avoid further declines in {psi}c to adverse levels. Declines in gc were apparent in larger Bowen ratios measured by EC over boreal conifers under higher Ta (Jarvis et al. 1997) and in the absence of a marked rise in LE measured or modeled under exponential rises in D with Ta (Figure 5B). The consequent declines in mid-afternoon CO2 influxes (Figure 5C) have consistently been measured over conifers under higher D at other sites (e.g., Chen et al. 2002, Grant et al. 2005).
  2. Declining ratios of aqueous CO2:O2 in warmer leaves that lowered rubisco substrate specificity and hence CO2 fixation rates. The high temperature inactivation parameter in the Arrhenius functions for carboxylation processes did not affect carboxylation rates over the range of Ta in this study.
  3. A rise in Ra and Rh driven by the Arrhenius functions of Ta and Ts (Figures 6, 7 and 9). Declines in mid-summer NEP at SOBS have been attributed to sharp rises in Re with seasonal warming (Arain et al. 2002).

The net effect of these contrasting responses to warming was to increase NEP modeled with warming of 2–4 °C from 2004 to 2006 (Table 4), and with further warming up to mean annual temperatures (MATs) of 6–8 °C (Figure 11), above which water limited NEP at mid-continental sites with lower precipitation. These increases in NEP were driven by the ones in GPP and Re which were well correlated with increases in EC-derived GPP and Re with warming from 2004 to 2006.

Rises in NEP with Ta modeled here were consistent with the experimental findings of Medhurst et al. (2006) that an increase in Ta of 3 °C raised net CO2 exchange by about one-third in Norway spruce after 3 years of growth in whole-tree chambers in northern Sweden (MAT 2.3 °C). However, raising Ta by more than 3 °C above a mean annual value of 7.5 °C caused modeled NEP of a temperate conifer to decline from water stress (Thornley and Cannell 1996), as found for mid-continental sites at SOBS and NOBS in this study (Figure 11). In sensitivity studies with other models, a rise in Ta of 2 °C caused rises in NEP at SOBS (Kang et al. 2006) and rises or declines in NEP at NOBS (Potter et al. 2001).

Variation in NEP with climate
The impact of higher Ta on NEP is likely affected by the seasonality of Ta. The annual NEP of warmed boreal conifers modeled here (Figure 11) remained smaller than that of a coastal temperate conifer measured and modeled at similar mean annual Ta (e.g., 300–400 g C m–2 year–1 at 8 °C MAT in Grant et al. 2007c) because the temperate climate had earlier, warmer springs and cooler summers than did the warmed boreal climate modeled here. However, NEP of this coastal temperate conifer was adversely affected by rises in Ta caused by El Niño events (Morgenstern et al. 2004).

The contrasting responses of CO2 fixation to higher Ta modeled here indicate that rises in NEP with long-term climate warming would depend on the seasonality of rises in Ta. These rises may be greater in spring and fall than in summer (IPCC 2007), increasing the beneficial effects of warming at lower Ta and reducing the adverse effects of warming at higher Ta. Therefore, rises in boreal NEP with warming modeled during the 4–6 years after a step change in Ta (Figure 11) may not be comparable to the ones that would be modeled under a longer-term climate warming scenario that would include gradual and seasonally variable rises in Ta, Ca, and likely precipitation over several decades (e.g., Grant et al. 2007b).


   Acknowledgements
 
The Fluxnet-Canada Research Network is funded by the Natural Sciences and Engineering Research Council of Canada (NSERC), the Canadian Foundation for Climate and Atmospheric Sciences (CFCAS), and the Biological Implications of CO2 Policy in Canada (BIOCAP). Computational facilities for ecosys were provided by the Westgrid high performance computing infrastructure. The contributions of Marc-André Giasson, Carole Coursolle, Praveena Krishnan, and other members of the Quebec and BC Flux teams for obtaining and processing the flux and meteorologic data at EOBS and SOBS are gratefully acknowledged. The contributions of staff from the Department of Earth and Planetary Sciences, Harvard, and from the Department of Soil Science, University of Manitoba, for operating the NOBS site in 2004–2005 and 2005–2006 respectively, are also gratefully acknowledged.


   Notes
Top
 Notes
Abstract
 Introduction
 Model description
 Field measurements
 Model testing
 Results
 Discussion
 References
 
Supplementary Data

Supplementary data for this article are available at Tree Physiology Online. Back


   References
Top
 Notes
 Abstract
 Introduction
 Model description
 Field measurements
 Model testing
 Results
 Discussion
 References
 

    Aakala T., Kuuluvainen T., De Grandpré L., Gauthier S. Trees dying standing in the northeastern boreal old-growth forests of Quebec: spatial patterns, rates, and temporal variation. Can. J. For. Res. (2007) 37:50–61.

    Alberta Forest Service. Alberta phase 3 forest inventory: yield tables for unmanaged stands (1985) Edmonton, Alberta: Alberta Energy and Natural Resources.

    Amthor J.S., Chen J.M., Clein J., et al. Boreal forest CO2 exchange and evapotranspiration predicted by nine ecosystem process models: inter-model comparisons and relationships to field measurements. J. Geophys. Res. (2001) 106. 33623–33648.

    Arain A., Restrepo-Coupe N. Net ecosystem production in a temperate pine plantation in southeastern Canada. Agric. For. Meteorol. (2005) 128:223–241.

    Arain M.A., Black T.A., Barr A.G., Jarvis P.G., Massheder J.M., Verseghy D.L., Nesic Z. Effects of seasonal and interannual climate variability on net ecosystem productivity of boreal deciduous and conifer forests. Can. J. For. Res. (2002) 32(5):878–891.[CrossRef]

    Barber V.A., Juday G.P., Finney B.P. Reduced growth of Alaska white spruce in the twentieth century from temperature-induced drought stress. Nature (2000) 405:668–673.[CrossRef]

    Barr A.G., Black T.A., Hogg E.H., et al. Inter-annual variability in the leaf area index of a boreal aspen-hazelnut forest in relation to net ecosystem production. Agric. For. Meteorol. (2004) 126:237–255.[CrossRef]

    Bergeron O. Dynamique des échanges de dioxide de carbone de la pessière noire boréale de l’est du Canada (2007) Quebec City, Canada: Univeristé Laval. Ph.D. Thesis.

    Bergeron O., Margolis H.A., Black T.A., Coursolle C., Dunn A.L., Barr A.G., Wofsy S.C. Comparison of carbon dioxide fluxes over three boreal black spruce forests in Canada. Global Change Biol. (2007) 13:89–107.

    Bergh J., Linder S. Effects of soil warming during spring on photosynthetic recovery in boreal Norway spruce stands. Global Change Biol. (1999) 5:245–253.

    Bernacchi C.J., Singsaas E.L., Pimentel C., Portis A.L., Long S.P. Improved temperature response functions for models of rubisco-limited photosynthesis. Plant Cell Environ. (2001) 24:253–259.[CrossRef]

    Bernacchi C.J., Pimentel C., Long S.P. In vivo temperature response functions of parameters required to model RuBP-limited photosynthesis. Plant Cell Environ. (2003) 26:1419–1430.[CrossRef]

    Brooks J.R., Flanagan L.B., Ehleringer J.R. Responses of boreal conifers to climate fluctuations: indications from tree-ring widths and carbon isotope analyses. Can. J. For. Res. (1998) 28:524–533.

    Chen J., Falk M., Euskirchen E., Paw U K.T., Suchanek T.H., Ustin S.L., Bond B.J., Brosofske K.D., Phillips N., Bi R. Biophysical controls of carbon flows in three successional Douglas-fir stands based on eddy covariance measurements. Tree Physiol. (2002) 22:169–177.[Abstract/Free Full Text]

    Churkina G., Schimel D., Braswell B.H., Xiao X. Spatial analysis of growing season length control over net ecosystem exchange. Global Change Biol. (2005) 11:1777–1787.[CrossRef]

    Dang Q.L., Lieffers V.J. Climate and annual ring growth of black spruce in some Alberta peatlands. Can. J. Bot. (1989) 67:1885–1889.

    Dunn A.L., Barford C.C., Wofsy S.C., Goulden M.L., Daube B.C. A long term record of carbon exchange in a boreal black spruce forest: means, responses to interannual variability, and decadal trends. Global Change Biol. (2007) 13:577–590.[CrossRef]

    Ekblad A., Högberg P. Natural abundance of 13C in CO2 respired from forest soils reveals speed of link between tree photosynthesis and root respiration. Oecologia (2001) 127:305–308.[CrossRef][Web of Science]

    Ekblad A, Boström B., Holm A., Comstedt D. Forest soil respiration rate and {delta}13C is regulated by recent above ground weather conditions. Oecologia (2005) 143:136–142.[CrossRef][Web of Science][Medline]

    Ewers F.W., Zimmermann M.H. The hydraulic architecture of eastern Hemlock (Tsuga canadiensis). Can. J. Bot. (1984) 62:940–946.[CrossRef]

    Gao Q., Zhao P., Zeng X., Cai X., Shen W. A model of stomatal conductance to quantify the relationship between leaf transpiration, microclimate and soil water stress. Plant Cell Environ. (2002) 25:1373–1382.

    Garfinkel H.L., Brubaker L.B. Modern climate–tree–growth relationships and climatic reconstruction in sub-Arctic Alaska. Nature (Lond.) (1980) 286:872–874.

    Gaumont-Guay D., Black T.A., Barr A.G., Jassal R.S., Nesic Z. Biophysical controls on rhizospheric and heterotrophic components of soil respiration in a boreal black spruce stand. Tree Physiol. (2008) 28:161–171.[Abstract/Free Full Text]

    Goetz S.J., Bunn A.G., Fiske G.J., Houghton R.A. Satellite-observed photosynthetic trends across boreal North America associated with climate and fire disturbance. PNAS (2005) 102:13521–13525.[Abstract/Free Full Text]

    Goulden M.L., Crill P.M. Automated measurements of CO2 exchange at the moss surface of a black spruce forest. Tree Physiol. (1997) 17:537–542.[Abstract/Free Full Text]

    Gower S.T., Vogel J.G., Norman J.M., Kucharik C.J., Steele S.J., Stow T.K. Carbon distribution and aboveground net primary production in aspen, jack pine and black spruce stands in Saskatchewan and Manitoba, Canada. J. Geophys. Res. (1997) 102:29029–29041.[CrossRef][Web of Science]

    Grant R.F. Simulation in ecosys of root growth response to contrasting soil water and nitrogen. Ecol. Model. (1998) 107:237–264.[Web of Science]

    Grant R.F. A review of the Canadian ecosystem model ecosys. In: Modeling Carbon and Nitrogen Dynamics for Soil Management—Shaffer M., ed. (2001) Boca Raton, FL: CRC Press.

    Grant R.F. Modelling topographic effects on net ecosystem productivity of boreal black spruce forests. Tree Physiol. (2004) 24:1–18.[Abstract]

    Grant R.F., Flanagan L.B. Modelling stomatal and non-stomatal effects of water deficits on CO2 fixation in a semi-arid grassland. J. Geophys. Res. (2007) 112:G03011. doi:10.1029/2006JG000302.

    Grant R.F., Wall G.W., Kimball B.A., Frumau K.F.A., Pinter P.J. Jr., Hunsaker D.J., Lamorte R.L. Crop water relations under different CO2 and irrigation: testing of ecosys with the free air CO2 enrichment (FACE) experiment. Agric. For. Meteorol. (1999) 95:27–51.[CrossRef]

    Grant R.F., Goulden M.L., Wofsy S.C., Berry J.A. Carbon and energy exchange by a black spruce–moss ecosystem under changing climate: testing the mathematical model ecosys with data from the BOREAS experiment. J. Geophys. Res. (2001a) 106:33605–33621.[CrossRef]

    Grant R.F., Jarvis P.G., Massheder J.M., Hale S.E., Moncrieff J.B., Rayment M., Scott S.L., Berry J.A. Controls on carbon and energy exchange by a black spruce–moss ecosystem: testing the mathematical model ecosys with data from the BOREAS experiment . Global Biogeochem. Cycles. (2001b) 15:129–147.[Web of Science]

    Grant R.F., Arain A., Arora V., Barr A., Black T.A., Chen J., Wang S., Yuan F., Zhang Y. Intercomparison of techniques to model high temperature effects on CO2 and energy exchange in temperate and boreal coniferous forests. Ecol. Model. (2005) 188:217–252.[Web of Science]

    Grant R.F., Black T.A., Gaumont-Guay D., Kljun N., Barr A.G., Morgenstern K., Nesic Z. Net ecosystem productivity of boreal aspen forests under drought and climate change: mathematical modelling with ecosys. Agric. For. Meteorol. (2006) 140:152–170.

    Grant R.F., Arkebauer T.J., Dobermann A., Hubbard K.G., Schimelfenig T.T., Suyker A.E., Verma S.B., Walters D.T. Net biome productivity of irrigated and rainfed maize–soybean rotations: modelling versus measurements. Agron. J. (2007a) 99:1404–1423.[Abstract/Free Full Text]

    Grant R.F., Barr A.G., Black T.A., et al. Net ecosystem productivity of boreal jack pine stands regenerating from clearcutting under current and future climates. Global Change Biol. (2007b) 13:1423–1440.

    Grant R.F., Black T.A., Humphreys E.R., Morgenstern K. Changes in net ecosystem productivity with forest age following clearcutting of a coastal Douglas fir forest: testing a mathematical model with eddy covariance measurements along a forest chronosequence. Tree Physiol. (2007c) 27:115–131.[Abstract/Free Full Text]

    Grelle A., Lindroth A., Mölder M. Seasonal variation of boreal forest surface conductance and evaporation. Agric. For. Meteorol. (1999) 98–99:563–578.

    Griffis T.J., Black T.A., Morgenstern K., Barr A.G., Nesic Z., Drewitt G.B., Gaumont-Guay D., McCaughey J.H. Ecophysiological controls on the carbon balances of three southern boreal forests. Agric. For. Meteorol. (2003) 117. 53–71.

    Holland E.A., Dentener F.J., Braswell B.H., Sulzman J. M. Contemporary and pre-industrial global reactive nitrogen budgets. Biogeochemistry (1999) 46:7–43.

    Hubbard R.M., Bond B.J., Ryan M.G. Evidence that hydraulic conductance limits photosynthesis in old Pinus ponderosa trees. Tree Physiol. (1999) 19:165–172.[Abstract]

    IPCC. Climate change 2007. Fourth assessment report of the intergovernmental panel on climate change (2007) Cambridge, UK: Cambridge University Press.

    Jarvis P., Linder S. Constraints to growth of boreal forests. Nature (2000) 405:904–905.[CrossRef][Medline]

    Jarvis P.G., Massheder J.M., Hale S.E., Moncrieff J.B., Rayment M., Scott S.L. Seasonal variation in carbon dioxide, water vapor, and energy exchanges of a boreal black spruce forest. J. Geophys. Res. (1997) 102:28953–28966.[CrossRef]

    Jassal R.S., Black T.A., Cai T., Morgenstern K., Li Z., Gaumont-Guay D., Nesic Z. Components of ecosystem respiration and an estimate of net primary productivity of an intermediate-aged Douglas-fir stand. Agric. For. Meteorol. (2007) 144:44–57.

    Kang S., Kimball J.S., Running S.W. Simulating effects of fire disturbance and climate change on boreal forest productivity and evapotranspiration. Sci. Total Environ. (2006) 362:85–102.[CrossRef][Medline]

    Kloeppel B.D., Gower S.T., Vogel J.G., Reich P.B. Leaf-level resource use for evergreen and deciduous conifers along a resource availability gradient. Funct. Ecol. (2000) 14(3):281–292.

    Kolari P., Lappalainen H.K., Hänninen H., Hari P. Relationship between temperature and the seasonal course of photosynthesis in Scots pine at northern timberline and in southern boreal zone. Tellus B (2007) 59(3):542–552.

    Krishnan P., Black T.A., Barr A.G., Grant N.J., Gaumont-Guay D., Nesic Z. Factors controlling the interannual variability in the carbon balance of a southern boreal black spruce forest. J. Geophys. Res. (2008) 113:D09109. doi:10.1029/2007JD008965.

    Ladanai S., Ågren G.I. Temperature sensitivity of nitrogen productivity for Scots pine and Norway spruce. Trees – Struc. Funct. (2004) 18:312–319.

    Lavigne M.B., Ryan M.G., Anderson D.E., et al. Comparing nocturnal eddy covariance measurements to estimates of ecosystem respiration made by scaling chamber measurements at six coniferous boreal sites. J. Geophys. Res. (1997) 102:28977–28985.[CrossRef][Web of Science]

    Law B.E., Waring R.H., Anthoni P.M., Aber J.D. Measurements of gross and net ecosystem productivity and water vapour exchange of a Pinus ponderosa ecosystem, and an evaluation of two generalized models. Global Change Biol. (2000) 6:155–168.

    Law B.E., Falge E., Gu L., et al. Environmental controls over carbon dioxide and water vapor exchange of terrestrial vegetation. Agric. For. Meteorol. (2002) 113:97–120.[CrossRef]

    Luyssaert S., Inglima I., Jung M., et al. CO2 balance of boreal, temperate, and tropical forests derived from a global database. Global Change Biol. (2007) 13:2509–2537.

    Man R., Lieffers V.J. Seasonal photosynthetic responses to light and temperature in white spruce (Picea glauca) seedlings planted under an aspen (Populus tremuloides) canopy and in the open. Tree Physiol. (1997) 17:437–444.[Abstract/Free Full Text]

    Medhurst J., Parsby J., Linder S., Wallin G., Ceschia E., Slaney M. A whole-tree chamber system for examining tree-level physiological responses of field-grown trees to environmental variation and climate change. Plant Cell Environ. (2006) 29:1853–1869.[Medline]

    Medlyn B.E., Dreyer E., Ellsworth D., et al. Temperature response of parameters of a biochemically based model of photosynthesis. II. A review of experimental data. Plant Cell Environ. (2002) 25:1167–1179.[CrossRef]

    Mellander P.-E., Bergh J., Lundmark T., Bishop K. Recovery of photosynthetic capacity in Scots pine: a model analysis of forest plots with contrasting soil temperature. Eur. J. For. Res. (2008) 127:71–79.

    Menuccini M., Grace J. Hydraulic conductance, light interception and needle nutrient concentration in Scots pine stands and their relations with net primary productivity. Tree Physiol. (1996) 16:459–468.[Abstract/Free Full Text]

    Méthy M., Gillon D., Houssard C. Temperature-induced changes of photosystem II activity in Quercus ilex and Pinus halepensis. Can. J. For. Res. (1997) 27:31–38.[CrossRef]

    Meteorological Service of Canada. Canadian acid deposition science assessment (2004) Ottawa ON: Environment Canada.

    Morgenstern K., Black T.A., Humphreys E.R., Griffis T.J., Drewitt G.B., Cai T., Nesic Z., Spittlehouse D.L., Livingstone N.J. Sensitivity and uncertainty in the carbon balance of a Pacific northwest Douglas fir forest during an El Niño/El Niña cycle. Agric. For. Meteorol. (2004) 123:201–219.

    Myneni R.B., Keeling C.D., Tucker C.J., Asrar G., Nemani R.R. Increased plant growth in the northern high latitudes from 1981 to 1991. Nature (1997) 386:698–702.[CrossRef][Web of Science]

    Ohta T., Hiyama T., Tanaka H., Kuwada T., Maximov T.C., Ohata T., Fukushima Y. Seasonal variation in the energy and water exchanges above and below a larch forest in eastern Siberia. Hydrol. Proc. (2001) 15:1459–1476.

    Piao S., Ciais P., Friedlingstein P., et al. Net carbon dioxide losses of northern ecosystems in response to autumn warming. Nature (2008) 451(7174):49–52.[CrossRef][Medline]

    Pothier D., Savard F. Actualisation des tables de production pour les principales espèces forestières du Québec (1998) Ministère des Ressources Naturelles du Québec. 183.

    Pothier D., Margolis H.A., Poliquin J., Waring R.H. Relation between the anatomy and the permeability of jack pine sapwood with stand development. Can. J. For. Res. (1989a) 19:1564–1570.[CrossRef]

    Pothier D., Margolis H.A., Waring R.H. Changes in saturated sapwood permeability and total sapwood conductance with stand development. Can. J. For. Res. (1989b) 19:432–439.

    Potter C.S., Wang S., Nikolov N.T., et al. Comparison of boreal ecosystem model sensitivity to variability in climate and forest site parameters. J. Geophys. Res. (2001) 106:33671–33688.[CrossRef]

    Régnière J., St-Amant R. Stochastic simulation of daily air temperature and precipitation from monthly normals in North America north of Mexico. Int. J. Biometeorol. (2007) 51:415–430.

    Richardson A.D., Hollinger D.Y., Burba G.G., et al. A multi-site analysis of random error in tower-based measurements of carbon and energy fluxes. Agric. For. Meteorol. (2006) 136:1–18.[CrossRef]

    Royce E.B., Barbour M.G. Mediterranean climate effects. I. Conifer water use across a Sierra Nevada ecotone. Am. J. Bot. (2001) 88:911–918.[Abstract/Free Full Text]

    Ryan M.G., Lavigne M.B., Gower S.T. Annual carbon cost of autotrophic respiration in boreal forest ecosystems in relation to species and climate. J. Geophys. Res. (1997) 102:28871–28884.[CrossRef]

    Saugier B., Granier A., Pontailler J.Y., Dufrêne E., Baldocchi D.D. Transpiration of a boreal pine forest measured by branch bag, sap flow and micrometeorological methods. Tree Physiol. (1997) 17:511–519.[Abstract/Free Full Text]

    Thornley J.H.M., Cannell M.G.R. Temperate forest responses to carbon dioxide, temperature and nitrogen: a model analysis. Plant Cell Environ. (1996) 19:1331–1348.[CrossRef]

    Tyree M.T., Ewers F.W. The hydraulic architecture of trees and other woody plants. New Phytol. (1991) 119:345–360.[CrossRef][Web of Science]

    Waring R.H., Running S.W. Forest ecosystems: analysis at multiple scales (1998) 2nd Edn. San Diego, CA: Academic Press.

    Welp L.R., Randerson J.T., Liu H.P. The sensitivity of carbon fluxes to spring warming and summer drought depends on plant functional type in boreal forest ecosystems. Agric. For. Meteorol. (2007) 147:172–185.

    Wesely M.L., Hart R.L. The Forest–Atmosphere Interaction—Hutchinson B.A., Hicks B.B., eds. (1985) Dordrecht: D. Reidel. 591–612.

    Wilmking M., Juday G.P., Barber V.A., Zald H.J. Recent climate warming forces contrasting growth responses of white spruce at treeline in Alaska through temperature thresholds. Global Change Biol. (2004) 10:1724–1736.

    Yuan F., Arain M.A., Barr A.G., Black T.A., Bourque C.P.-A., Coursolle C., Margolis H.A., McCaughey J.H., Wofsy S.C. Modeling analysis of primary controls on net ecosystem productivity of seven boreal and temperate coniferous forests across a continental transect. Global Change Biol. (2008) 14:1765–1784.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow Supplementary Data
Right arrow All Versions of this Article:
29/1/1    most recent
tpn004v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Grant, R.F.
Right arrow Articles by Bergeron, O.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Grant, R.F.
Right arrow Articles by Bergeron, O.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?