Elevated CO2 maintains grassland net carbon uptake

under a future heat and drought extreme

Jacques Roya,1,2, Catherine Picon-Cochardb,1, Angela Augustib,c, Marie-Lise Benotb,d, Lionel Thieryb, Olivier Darsonvilleb,
Damien Landaisa, Clment Piela, Marc Defosseza, Sbastien Devidala, Christophe Escapea, Olivier Ravela,
Nathalie Fromine, Florence Volairee,f, Alexandru Milcua,e, Michael Bahng, and Jean-Franois Soussanab
a
Ecotron Europen de Montpellier, Unit Propre de Service 3248, Centre National de la Recherche Scientifique (CNRS), Campus Baillarguet, F-34980
Montferrier-sur-Lez, France; bGrassland Ecosystem Research, Unit de Recherche 874, Institut National de la Recherche Agronomique (INRA), F-63039
Clermont-Ferrand, France; cInstitute of Agroenvironmental and Forest Biology, Consiglio Nazionale delle Ricerche, 2-05010 Porano (TR), Italy; dBiodiversit
Gnes et Communauts, INRA, Universit de Bordeaux, F-33615 Pessac, France; eCentre dEcologie Fonctionnelle et Evolutive, CNRS, Unit Mixte de
Recherche 5175, Universit de Montpellier, Universit Paul Valry, cole Pratique des Hautes tudes, F-34293 Montpellier Cedex 5, France; fUnit Sous
Contrat 1338, INRA, Centre dEcologie Fonctionnelle et Evolutive F-34293 Montpellier Cedex 5, France; and gInstitute of Ecology, University of Innsbruck,
A-6020 Innsbruck, Austria

Extreme climatic events (ECEs) such as droughts and heat waves

are predicted to increase in intensity and frequency and impact the
terrestrial carbon balance. However, we lack direct experimental
evidence of how the net carbon uptake of ecosystems is affected by
ECEs under future elevated atmospheric CO2 concentrations (eCO2).
Taking advantage of an advanced controlled environment facility
for ecosystem research (Ecotron), we simulated eCO2 and extreme
cooccurring heat and drought events as projected for the 2050s
and analyzed their effects on the ecosystem-level carbon and water
fluxes in a C3 grassland. Our results indicate that eCO2 not only
slows down the decline of ecosystem carbon uptake during the
ECE but also enhances its recovery after the ECE, as mediated by
increases of root growth and plant nitrogen uptake induced by the
ECE. These findings indicate that, in the predicted near future climate,
eCO2 could mitigate the effects of extreme droughts and heat waves
on ecosystem net carbon uptake.

climate change extreme events

grassland ecosystem

| elevated CO | carbon fluxes |

ncreased aridity and heat waves are projected to increase in the

21st century for most of Africa, southern and central Europe,
the Middle East, and parts of the Americas, Australia, and southeast Asia (13). These regions have a large fraction of their land
covered by grasslands and rangelands, a biome covering approximately one-quarter of the Earths land area and contributing to the
livelihoods of more than 800 million people (4). There is mounting
evidence that extreme climatic events (ECEs) may significantly affect the regional and global carbon (C) fluxes (3, 59) and potentially feed back on atmospheric CO2 concentrations and the climate
system (7). However, our knowledge concerning the outcome of the
interaction between future ECEs and elevated atmospheric CO2
concentrations (eCO2) for ecosystem C stocks is equivocal (1012).
Studies focusing on plant physiological responses have shown that
eCO2 has the potential to mitigate future drought-related stress on
plant growth by reducing stomatal conductance, thereby increasing
water use efficiency (WUE) (1315) and preserving soil moisture
(1618). However, to date, little is known on whether and how
eCO2 alters the consequences of ECEs for ecosystem net C uptake.
Because the capacity of ecosystems to act as a C sink depends on
the relative effects of eCO2, ECE, and their potential interaction on
both plant and soil processes, an integrated assessment of all C
fluxes during and after the ECEs is important if we are to estimate
the overall C balance.
Using the Montpellier CNRS Ecotron facility (www.ecotron.
cnrs.fr), we tested with 12 large controlled environment units
(macrocosms, SI Appendix, Fig. S1) whether (i) an ECE (severe
drought and heat wave) predicted for the 2050s reduces ecosystem
net C uptake by reducing ecosystem photosynthesis relative to
ecosystem respiration (Reco), (ii) eCO2 buffers the negative effects
www.pnas.org/cgi/doi/10.1073/pnas.1524527113

of the ECE on ecosystem CO2 fluxes and increases ecosystem

water-use efficiency during the ECE, and (iii) eCO2 speeds up
the recovery of ecosystem C uptake after the ECE. We exposed a
seminatural upland grassland (botanical composition see SI
Appendix, Table S1) from the French Massif Central (4543N,
0301E, 800 m above sea level) to the average climatic conditions for the 2050s as projected by the downscaled ARPEGEv4
climate model (19). We exposed large (4 m2 and 0.6 m depth)
and intact ecosystem monoliths in lysimeters to (i) ambient CO2
(aCO2 at 390 molmol1) and predicted elevated levels of atmospheric CO2 concentrations (eCO2 at 520 molmol1) and
(ii) the presence/absence of the most severe drought and heat
wave (ECE) projected by the above-mentioned climate model
over the 2050s. The experiment was based on 12 macrocosm
experimental units, using a two factorial crossed design (elevated
CO2 and ECE, each with two levels) with three replicates per
treatment combination. The ECE included a reduction in precipitation by 50% during 4 wk in midsummer, followed by 15 d with
no precipitation and a concomitant increased temperature by +3.4 C
(Fig. 1). Thereafter, the precipitation was gradually increased
during 26 d (see Methods for details). In addition to vegetation
characteristics, we monitored the net ecosystem CO2 exchange
Significance
Ecosystems are responding to climate change and increasing
atmospheric CO2 concentrations. Interactions between these
factors have rarely been assessed experimentally during and
after extreme climate events despite their predicted increase in
intensity and frequency and their negative impact on primary
productivity and soil carbon stocks. Here, we document how a
grassland exposed to a forecasted 2050s climate shows a remarkable recovery of ecosystem carbon uptake after a severe
drought and heat wave, this recovery being amplified under
elevated CO2. Over the growing season, elevated CO2 entirely
compensated for the negative impact of extreme heat and
drought on net carbon uptake. This study highlights the importance of incorporating all interacting factors in the predictions of climate change impacts.
Author contributions: J.-F.S. designed research; J.R., C.P.-C., A.A., M.-L.B., L.T., O.D., D.L., C.P.,
S.D., C.E., O.R., N.F., and F.V. performed research; J.R., C.P.-C., A.A., D.L., C.P., M.D., N.F., F.V.,
and A.M. analyzed data; and J.R., C.P.-C., A.M., M.B., and J.-F.S. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
Freely available online through the PNAS open access option.
1

J.R. and C.P.-C. contributed equally to this work.

To whom correspondence should be addressed. Email: jacques.roy@ecotron.cnrs.fr.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.

1073/pnas.1524527113/-/DCSupplemental.

PNAS Early Edition | 1 of 6

ECOLOGY

Edited by William H. Schlesinger, Cary Institute of Ecosystem Studies, Millbrook, NY, and approved April 11, 2016 (received for review December 12, 2015)

(NEE) and ecosystem evapotranspiration (ET). Furthermore,

we derived values of gross primary productivity (GPP), Reco, and
WUE for the prestress, stress, and poststress periods.

Fig. 1. Ecotron-simulated environmental conditions of the year 2050 under

two atmospheric CO2 concentrations (390 and 520 molmol1) with and
without an ECE. (A) Soil moisture relative to field capacity in the top 60 cm.
Bars depict rainfall. (B) Experimentally simulated vs. field recorded temperature (daily means) in the heat wave year 2003 and average of the years 1990
2009 (the simulated temperatures at each CO2 concentrations are too similar
for the blue and red lines to be distinct). (C) Simulated air vapor pressure
deficit. (D) Experimentally simulated atmospheric CO2 concentrations. Data
represent means over 14 (1) d SEM of three replicates. Horizontal bars
represent the experimental periods (solid green, prestress; solid orange,
halved water supply; solid red, no water supply and +3.4 C temperature increase; hatched orange bar, gradual rewetting; hatched green bar, poststress
(see Methods for details).

2 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1524527113

Results
During the prestress period in spring, ecosystem CO2 and water
vapor fluxes increased with the development of the canopy until
the June harvest (Fig. 2). Elevated CO2 increased GPP (+20%)
and WUE (+25%) while slightly decreasing ET (3%) (Fig. 2 and
SI Appendix, Table S2). At the June (prestress) harvest, there was
no effect of eCO2 on above-ground biomass. Root growth, however, increased by 77% under elevated CO2 (marginally significant; SI Appendix, Table S3). Canopy greenness, i.e., the fraction
of green vs. total above-ground shoot biomass, increased by 41%
(SI Appendix, Table S3 and Fig. S3B), whereas the shoot nitrogen
content (%) and pool (g/m2 ground area) decreased under eCO2
relative to aCO2 by 9 and 10%, respectively (Fig. 3B and SI
Appendix, Fig. S4 and Table S4).
When the ECE was imposed (stress period), the mean daily air
temperature and vapor pressure deficit peaked at 25 C and 1.8 kPa,
respectively (Fig. 1 B and C). These climatic conditions led to a
gradual reduction in soil moisture (Fig. 1A) and canopy greenness
(10%; SI Appendix, Table S3 and Fig. S3B). NEE and GPP were
significantly affected by the interaction between eCO2 and ECE, but
in a variable manner over time (SI Appendix, Table S2). Compared
with aCO2 conditions, the eCO2 treatment increased GPP and Reco
(by +55 and 23%, respectively) during the first month of drought,
but this effect disappeared for GPP during the 2 wk with combined drought and heat wave. However, Reco was 35% higher in the
eCO2 + ECE treatment relative to aCO2 + ECE (Fig. 2C) resulting
in a net release of CO2 from the ecosystem under the eCO2 + ECE
treatment during and immediately after the heat wave (Fig. 2A).
During the first month of reduced precipitation, eCO2 buffered the
negative effect of drought on ET and WUE, leading to a higher ET
(+6%) and WUE (+51%) under the eCO2 treatment relative to
aCO2. However, eCO2 had no lasting impact on ET during and
immediately after the heat wave (Fig. 2 D and E and SI Appendix,
Table S2). When precipitation was gradually resumed, in August
(Fig. 1A), a progressive recovery of CO2 and water vapor fluxes was
observed in the macrocosms previously exposed to the ECE. The
recovery of relative soil water moisture was lower in the eCO2 + ECE
treatment relative to aCO2 + ECE (Fig. 1A) due to a stronger
recovery of ET and other fluxes at elevated CO2 (Fig. 2D).
During the poststress period (September to November), fluxes
of CO2 and water decreased progressively with decreasing temperature and light levels in both CO2 treatments without ECE.
The eCO2 treatment significantly increased NEE (+146%), GPP
(+46%), Reco (+25%), and WUE (+32%) (Fig. 2 and SI Appendix,
Table S2). GPP, NEE, and WUE strongly increased in September
and reached values significantly higher than in the treatments
without ECE (Fig. 2 and SI Appendix, Table S2). An eCO2 ECE
interaction affected the NEE and GPP fluxes, suggesting that the
recovery of the CO2 fluxes led to significantly higher net and gross
ecosystem C uptake in the eCO2 + ECE treatment combination
(up to 60% and 240% higher GPP and NEE, respectively in the
eCO2 + ECE treatment compared with the mean of the remaining
treatment combinations). A time-series analysis of daily CO2
fluxes focused on the last two weeks of September shows that
NEE and GPP were significantly affected by the eCO2 ECE
Time interaction during this period with P values <0.001 (SI
Appendix, Table S5 and Fig. S5), and with the highest C uptake in
the eCO2 + ECE treatment.
Above-ground biomass at the November harvest did not differ
between the treatments (Fig. 3A) and confirmed the low productivity of this grassland type during autumn (20). During the
poststress period, compared with aCO2, eCO2 increased the Leaf
Area Index (LAI) (+47.9%; SI Appendix, Fig. S3A) and root
growth rate (+244%; Fig. 3C) and decreased the shoot nitrogen
Roy et al.

Fig. 2. Seasonal dynamics (Left, lines) and cumulative fluxes (Right, bars) of
carbon and water as affected by the CO2 and ECE treatments. (A) NEE. (B) GPP.
(C) Reco. (D) ET. (E) WUE. Data represent means over 14 (1) d SEM of three
replicates. Stars indicate significant terms according to repeated-measures
ANOVAs [***P < 0.001, ** P < 0.01, * P < 0.05, and (*)P < 0.08]. Different letters
above bars denote significant differences between individual means. Horizontal
bottom bars represent the experimental periods as in Fig. 1.

Roy et al.

PNAS Early Edition | 3 of 6

ECOLOGY

content (21%; SI Appendix, Fig. S4 and Tables S3 and S4). No

CO2 effect was found on canopy greenness or on the shoot
nitrogen pool. ECE strongly increased the shoot nitrogen pool
(+87%, Fig. 3B) relative to the treatment combinations without
ECE (both under aCO2 and eCO2), primarily through an increase in the shoot nitrogen content (+53%; SI Appendix, Fig.
S4 and Table S4). Belowground, the root growth rate and the
root nitrogen pool were affected by the CO2 ECE Time
interaction (Fig. 3 C and D and SI Appendix, Fig. S6A and
Tables S3 and S4). We also found an ECE Time interaction
on the root nitrogen content, with a significantly higher content
in the ECE treatment in September (+19%, P1/10 = 0.029; SI
Appendix, Fig. S6B).
These results suggest that the recovery of CO2 fluxes following
the ECE treatments was stimulated by increased nitrogen availability. This suggestion was confirmed by showing that shoot and
root nitrogen pools were predictors of late October CO2 fluxes
(GPP and NEE; SI Appendix, Table S6). Furthermore, in agreement with this conjecture, we also found that the soil nitrification
potential measured after the November biomass harvest was
twice as large in the macrocosms that had been exposed to
ECE compared with macrocosms not exposed to ECE (Fig. 4A
and SI Appendix, Table S7).
Integrated over the full growing season (April 26November 3),
eCO2 significantly increased the cumulative seasonal NEE (+79%),
GPP (+30%), and WUE (+32%), whereas the ECE significantly
reduced cumulative NEE (29%), GPP (16%), and ET (18%)
(bar plot of Fig. 2 and SI Appendix, Table S7). Interestingly,
whereas effects of eCO2 and ECE were not additive, the cumulative NEE was 24% higher under the eCO2 + ECE treatment
compared with aCO2 without ECE (P = 0.027). This result shows
that under the projected climatic conditions of the 2050s, eCO2
more than compensated the negative impact of extreme drought
and heat wave. Under aCO2, the ECE reduced the ecosystem
C balance (CNEE Charvested biomass) (Fig. 4B and SI Appendix,
Table S7), and turned the functioning of the grassland from a small
C sink (19.9 gCm2) to a C source for the atmosphere (94 gCm2),
a result consistent with the analysis of the consequence of the summer 2003 heat and drought (5, 21). In contrast, under eCO2, the
ecosystem C balance increased and reached 115.0 gCm2 without
ECE and remaining a C sink, 46.3 gCm2, with ECE (Fig. 4B).
Most previous studies on the impact of eCO2 associated with
drought and/or elevated temperature did not report whole ecosystem responses, but primarily plant responses at various organizational levels. Generally, they show that eCO2 alleviates the stress
associated with drought and/or elevated temperature (4, 16, 18, 22
28), but some studies found inconsistent responses (2936). This
variability could be linked with the seasonality of precipitation (35,
37) or with the intensity and length of the imposed stress, because
the effect of eCO2 has been shown to be positive under moderate
drought and negligible under severe stress when stomata are fully
closed (14). Owing to high-frequency measurements of ecosystemlevel CO2 fluxes, we could detect both mitigation effects of eCO2 on
CO2 fluxes during moderate stress and negative effects of eCO2 on
NEE via an increased Reco during and immediately after the heat
wave. Our results show that these apparently conflicting results,
which have been previously documented separately, occur at different periods during the growing season and depend indeed on the
level of water stress. An increase in Reco under eCO2 has been
shown to be prevalent and to result from improved soil water
content and/or enhanced inputs of labile C that prime the decomposition of soil organic matter (38). We show that the negative
effect of eCO2 on NEE via increased Reco is transient and had no
substantial impact on the cumulative net C uptake during the
growing season. In our experiment, the positive effect of eCO2 on
CO2 fluxes under moderate stress cannot be explained through a
soil water sparing effect as found in some earlier studies (16, 24),

To our knowledge, only a single experiment has so far addressed the consequence of combined eCO2 and elevated temperature on the ecosystem C budget (39) and found that these
combined treatments increased C loss (in one of the four years of
study), a result in contradiction to model simulations (40, 41).
The most notable effect in that experiment was the increased loss
of C under elevated CO2 (through increased heterotrophic soil
respiration) in two of the four years. In our study, we found a
positive C budget at the end of the growing season under the
combined eCO2 and ECE treatment, mainly due to a strong
recovery of GPP fluxes in the autumn (post ECE), where the
interaction between eCO2 and ECE led to the highest GPP. Such
a remarkable recovery of GPP under a realistic future scenario
of eCO2 and combined drought and heat wave has not been
reported before, although at species level, it was previously observed that eCO2 can enhance nitrogen metabolism and photosynthesis after drought (42). In our experiment, the enhanced
recovery of GPP under the eCO2 + ECE treatment combination
was related to increased nitrogen uptake by the newly grown
roots and the shoots, which statistically explain the ECE effect
on GPP and NEE (SI Appendix, Table S6). Increased plant nitrogen content has been previously found to be higher under
eCO2, elevated temperature, and drought, especially when all
three factors were combined (43). We argue that the unexpected
increase in canopy and root nitrogen pools (Fig. 3 B and D), and
a higher LAI (SI Appendix, Fig. S3A) at the beginning of the
poststress period in the ECE treatments, reflect a higher soil
nitrogen availability after the extreme event, as indicated by a
significantly higher microbial nitrification potential in the treatments with ECE (Fig. 4A). Recent analyses of 12 decadal experiments (44, 45) show that such a positive impact of eCO2 on
plant nitrogen acquisition and aboveground net primary production is not decreasing with time and that the associated
progressive nitrogen limitation hypothesis (46) is not ubiquitous.
In conclusion, our study shows that under realistic extreme events
of heat and drought as predicted for the 2050s, eCO2 does not
necessarily exacerbate the drought and heat stress as sometimes
suggested (11, 12), but can contribute to maintain the C sink
function of grasslands via an increased root nitrogen uptake when
the stress is released. Although climatic extremes are likely to become more frequent and more severe toward the end of the century, the current Earth system models do not simulate adequately
the impacts of climate extremes on biogeochemical cycles (47) and
still have large uncertainties in parameterizing regional responses to
such events and often provide contrasting results (4750). Our experiment highlights the importance of accounting for interactions
between drivers of global change, their seasonality effects, and the
importance of poststress recovery processes for more accurate
model predictions of the future C budgets of grasslands.
Methods

Fig. 3. Vegetation response variables as affected by the CO2 and ECE treatments. Shoot biomass (A) and nitrogen (N) pool in the shoots (B) at the two
successive June 9 and November 3 cuts. Total carbon (C) in the new root growth
(C) and total pool of N in the new root growth (D) during the prestress and
rewetting/poststress periods. Data are means SEM of three replicates. Different
letters above bars denote significant differences between individual means.

because soil water content was similar with or without the ECE,
but rather through the CO2 fertilization effect.
4 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1524527113

Extraction and Acclimation of Soil Monoliths. In June 2009, 48 soil monoliths

(1 m2 each) including their intact soil and plant communities [dominated by
perennial C3 grasses and by the pasture legume Trifolium repens (full species
composition; see SI Appendix, Table S1)], were excavated down to the rock
level (0.6 m depth) from an extensively managed upland seminatural
grassland site (Redon), near Saint-Gens Champanelle in the French Massif
Central (SI Appendix, Fig. S2). The monoliths were left to recover after the
extraction disturbance at the Clermont-Ferrand INRA research station (mean
annual temperature 12.4 C) until September 2009 when they were transported to Montpellier. In April 2010, four randomly chosen 1 m2 monoliths
were combined to form one large (4 m2) experimental unit and inserted in
each of the 12 Ecotron macrocosms (SI Appendix, SI Methods). Thereafter,
they were left to acclimate to the climatic conditions representative of
years 2050s and ambient atmospheric CO2 concentrations (390 molmol1)
for 11 mo. The eCO 2 treatment (520 molmol1) started in early March
2011 in six randomly chosen macrocosms (the small departure from the
set points for 10 d in August was due to a malfunction in the automatic
control system). The 30 m 3 transparent dome of each macrocosm had a

Roy et al.

Fig. 4. Soil nitrification potential at the end of the experiment (A) and full
growing season (April 26November 3) ecosystem carbon (C) balance (>0
sink; <0 source) (B) as affected by the CO2 and ECE treatments. Data are
means SEM of three replicates. Different letters above bars denote significant differences between individual means.

photosynthetically active radiation transmission of 0.86, a figure matching

closely the annual global radiation ratio between Saint-Gens Champanelle
and Montpellier during April to September (0.84).
Simulation of ECE. Climate projections for the original location of this
grassland for the 2050s (20402060) were obtained from the ARPEGEv4
atmosphere-ocean general circulation model with the A2-CO2 emission
scenario (51) and using a multivariate statistical downscaling methodology
(19) to generate projections over 8 8 km grid. The climatic conditions of
an average year of that period were applied from April 2010 to November
2, 2011. From June 25 to August 31, 2011, a summer ECE including combined drought and heat wave (stress period), simulating the severest
events projected by the downscaled ARPEGEv4 model, was applied to
three randomly selected macrocosms of the six at each CO2 concentration.
For the first four weeks of the stress period (June 25July 21), the irrigation amount was reduced by half compared with the control. Then, during
two weeks (July 22August 4), irrigation was stopped and the air temperature was increased by 3.4 C compared with the 2050s average. This
increase in air temperature corresponded to 7.1 C above the 20002009
average for the same period, a value above the average of the 14 consecutive hottest days of the exceptional heat wave in summer 2003. From
August 5 to 31, irrigation was progressively increased in the treatment
with ECE to obtain the same cumulative precipitation in the two treatments
during that period.
The Gas Exchange System. Each macrocosm unit of the Montpellier Ecotron is
an independent open gas exchange measurement system (52), with a
constant flow of air circulating through. The CO2 flux measurement (NEE) is
based on mass balance calculation: The flux of CO2 exchanged between the
ecosystem and atmosphere is calculated from the difference between the
fluxes of CO2 entering and leaving the dome. For this purpose, the CO2 and
water vapor mole fractions are measured at the inlet and outlet (infrared
gas analyzers, Licor 7000; Licor), and the mass flow of air at the inlet
thermal mass flow meter (Sensiflow iG; ABB). Water vapor mole fraction

Roy et al.

Other Measurements. Ecosystem evapotranspiration was measured continuously by the lysimeter weight changes over time. Four shear beam load cells
with an accuracy of 200 g (CMI-C3 5,000 kg, Precia-Molen) are used for each
lysimeter. WUE was calculated as mg of CO2 fixed per g of H2O lost. Soil
water content was continuously measured at three soil depths (7, 20, and
50 cm) with TDR probes (TRIME Pico 32, IMKO Micromodul-technik, Germany Ettlingen). Reported data relate to the 060 cm depth and are
expressed relative to soil water content at field capacity. LAI was estimated with a sunfleck ceptometer (Decagon Devices) from April to October, and canopy greenness was visually estimated every three weeks from
June to October. Because extensive management was found to be the best
option for sustaining the production of this type of grassland in the context of greater climate variability (20), no fertilization was applied and the
number of harvests was low, especially in the summer after the start of the
ECE. The vegetation was cut at 5 cm height on March 14, April 26, June 9,
and November 3.
Root growth (015 cm depth) was measured monthly using four (8 cm
diameter) ingrowth cores per macrocosm from February to the end of the
experiment. After harvest, shoots and roots were oven-dried (60 C, 48 h),
weighed, and their C and nitrogen content analyzed. Microbial nitrification
potential was determined on soil samples collected at the end of the experiment, according to the method of Lensi et al. (55) described in more
detail by Pinay et al. (56) (SI Appendix, SI Methods).
Statistical Analysis. Linear mixed effects models as available in the R nlme
package (57) were used to perform repeated-measures ANOVAs on the effects
of CO2, ECE, Time, and their interactions on fortnightly averaged values of
carbon and water fluxes, with the Ecotrons macrocosms as a random factor
accounting for temporal pseudoreplication. The statistical models were then
simplified to reach the most parsimonious models by using the automatic
model simplification step procedure based on Akaikes Information Criterion. Planned pairwise comparisons between the means of the treatments at
different time periods were performed by using ANOVAs. As we found a
marginally significant interaction between CO2 and ECE on NEE and GPP
during the recovery period, additional time-series analyses on daily values
were performed by using mixed effects models with the Ecotrons macrocosms
as a random factor. To account for the temporal autocorrelation of the daily
values, these models included autoregressive covariance structure (correlation=corAR1(form = 1 j macrocosm) at the macrocosm level. Following the
guidelines of Zuur et al. (58), we found that the models with the daydependent variance coefficients (varIdent(form = 1 j Day) fitted best
the data and were therefore retained in the models (see SI Appendix,
Table S4 for the complete R syntax).
ACKNOWLEDGMENTS. Ch. Collin and the experimental field team at Centre
dEcologie Fonctionnelle et Evolutive (CEFE)CNRS as well as the technical
staff of INRA Grassland Ecosystem UREP and UERT groups are thanked for
extracting the intact soil monoliths. We also thank B. Buatois and the Plateforme dAnalyses Chimiques en Ecologie (CEFE and LabEx CEMEB) for the
nitrification analyses. This study was supported by the ANR VALIDATE project grant and the AnimalChange project funded by the European Community (FP7/20072013, Grant 266018). This study benefited from the CNRS
human and technical resources allocated to the ECOTRONS Research Infrastructure as well as from the state allocation Investissement dAvenir
AnaEE-France ANR-11-INBS-0001. The Languedoc Roussillon Region and
the Conseil Gnral de lHrault also participated in the funding of
the Ecotron. A.A. and M.-L.B. received a postdoctoral position through
an INRA scientific package (20102014). A.A. was also supported by the
European FP7 ExpeER Transnational Access program, and M.B. was additionally supported by the AD WTZ-programme AustriaFrance and FWF
Project P28572-B22.

PNAS Early Edition | 5 of 6

ECOLOGY

measurements are used to correct for the dilution of CO2 and the variation
of airflow rate at the outlet due to the vapor added or removed in the dome by
ecosystem transpiration and humidity control. A custom-made manifold system
(interconnected solenoid valves, 750 series; Matrix) allows for switching the air
fluxes to the infrared gas analyzers from one dome to the other. It takes 12 min
to analyze the atmospheric CO2 concentration of the 12 domes.
Missing values or known inaccurate NEE values during unavoidable experimental work (manual watering, mowing, sampling, checks) caused by the
respiration of the persons entering the domes were gap-filled by using the
equation NEE = f(photosynthetically active radiation), where f is a rectangular hyperbola fitted with the data of the day before or after the disturbance (53). To provide conservative results, all statistics were performed
using only days with less than 33% of gap-filled NEE data. The C flux partitioning algorithm Reichstein et al. (54) (www.bgc-jena.mpg.de/MDIwork/
eddyproc/index.php) was used to estimate the daytime Reco.

1. Jacob D, et al. (2013) EURO-CORDEX: New high-resolution climate change projections

for European impact research. Reg Environ Change 14(2):563578.
2. Christidis N, Jones GS, Stott PA (2015) Dramatically increasing chance of extremely hot
summers since the 2003 European heatwave. Nat Clim Chang 5:4650.
3. Seneviratne SI, et al. (2012) IPCC 2012: Managing the Risks of Extreme Events and
Disasters to Advance Climate Change Adaptation (Cambridge Univ Press, Cambridge,
UK), pp 109230.
4. Reynolds SG, Batello C, Baas S, Mack S (2005) Grassland: A Global Ressource, ed
Gilloway DA (Wageningen Academic Publishers, Wageningen, The Netherlands), pp
323338.
5. Ciais P, et al. (2005) Europe-wide reduction in primary productivity caused by the heat
and drought in 2003. Nature 437(7058):529533.
6. Reichstein M, et al. (2007) Reduction of ecosystem productivity and respiration during
the European summer 2003 climate anomaly: A joint flux tower, remote sensing and
modelling analysis. Glob Change Biol 13(3):634651.
7. Reichstein M, et al. (2013) Climate extremes and the carbon cycle. Nature 500(7462):
287295.
8. Frank D, et al. (2015) Effects of climate extremes on the terrestrial carbon cycle:
Concepts, processes and potential future impacts. Glob Change Biol 21(8):28612880.
9. Zscheischler J, et al. (2014) Extreme events in gross primary production: A characterization across continents. Biogeosciences 11:29092924.
10. Leakey ADB, et al. (2009) Elevated CO2 effects on plant carbon, nitrogen, and water
relations: Six important lessons from FACE. J Exp Bot 60(10):28592876.
11. Williams AC (2014) Heat and drought extremes likely to stress ecosystem productivity
equally or more in a warmer, CO2 rich future. Environ Res Lett 9:101002.
12. Williams IN, Torn MS, Riley WJ, Wehner MF (2014) Impacts of climate extremes on
gross primary production under global warming. Environ Res Lett 9:094011.
13. Morgan J, Lecain D, Mosier A, Milchunas D (2001) Elevated CO2 enhances water relations and productivity and affects gas exchange in C3 and C4 grasses of the Colorado shortgrass steppe. Glob Change Biol 7(4):451466.
14. Franks PJ, et al. (2013) Sensitivity of plants to changing atmospheric CO2 concentration: From the geological past to the next century. New Phytol 197(4):10771094.
15. Holtum JAM, Winter K (2010) Elevated CO2 and forest vegetation: More a water issue
than a carbon issue? Funct Plant Biol 37:694702.
16. Robredo A, et al. (2007) Elevated CO2 alleviates the impact of drought on barley
improving water status by lowering stomatal conductance and delaying its effects on
photosynthesis. Environ Exp Bot 59(3):252263.
17. Albert KR, et al. (2011) Effects of elevated CO, warming and drought episodes on
plant carbon uptake in a temperate heath ecosystem are controlled by soil water
status. Plant Cell Environ 34(7):12071222.
18. Morgan JA, et al. (2011) C4 grasses prosper as carbon dioxide eliminates desiccation in
warmed semi-arid grassland. Nature 476(7359):202205.
19. Bo J, Terray L, Habets F, Martin E (2006) A simple statistical-dynamical downscaling
scheme based on weather types and conditional resampling. J Geophys Res 111:
D23106.
20. Zwicke M, et al. (2013) Lasting effects of climate disturbance on perennial grassland
above-ground biomass production under two cutting frequencies. Glob Change Biol
19(11):34353448.
21. Soussana JF, et al. (2007) Full accounting of the greenhouse gas (CO2, N2O, CH4)
budget of nine European grassland sites. Agric Ecosyst Environ 121:121134.
22. Zinta G, et al. (2014) Physiological, biochemical, and genome-wide transcriptional
analysis reveals that elevated CO2 mitigates the impact of combined heat wave and
drought stress in Arabidopsis thaliana at multiple organizational levels. Glob Change
Biol 20(12):36703685.
23. Naudts K, Van den Berge J, Janssens IA, Nijs I, Ceulemans R (2013) Combined effects of
warming and elevated CO2 on the impact of drought in grassland species. Plant Soil
369:497507.
24. Blumenthal DM, et al. (2013) Invasive forb benefits from water savings by native
plants and carbon fertilization under elevated CO2 and warming. New Phytol 200(4):
11561165.
25. Miranda-Apodaca J, Prez-Lpez U, Lacuesta M, Mena-Petite A, Muoz-Rueda A
(2015) The type of competition modulates the ecophysiological response of grassland
species to elevated CO2 and drought. Plant Biol (Stuttg) 17(2):298310.
26. Koutavas A (2013) CO2 fertilization and enhanced drought resistance in Greek firs
from Cephalonia Island, Greece. Glob Change Biol 19(2):529539.
27. Dias De Oliveira E, et al. (2013) Can elevated CO2 combined with high temperature
ameliorate the effect of terminal drought in wheat? Funct Plant Biol 40(2):160171.
28. Yu J, Chen L, Xu M, Huang B (2012) Effects of elevated CO2 on physiological responses
of tall fescue to elevated temperature, drought stress, and the combined stresses.
Crop Sci 52:18481858.
29. Derner JD, et al. (2003) Above- and below-ground responses of C3-C4 species mixtures
to elevated CO2 and soil water availability. Glob Change Biol 9(3):452460.
30. Dukes JS, et al. (2005) Responses of grassland production to single and multiple global
environmental changes. PLoS Biol 3(10):e319.

6 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1524527113

31. Duan H, et al. (2014) Elevated [CO2] does not ameliorate the negative effects of elevated temperature on drought-induced mortality in Eucalyptus radiata seedlings.
Plant Cell Environ 37(7):15981613.
32. Godfree RC, et al. (2013) Nonindigenous plant advantage in native and exotic australian grasses under experimental drought, warming, and atmospheric CO2 enrichment. Biology (Basel) 2(2):481513.
33. Bloor JMG, Pichon P, Falcimagne R, Leadley P, Soussana JF (2010) Effects of warming,
summer drought, and CO2 enrichment on aboveground biomass production, flowering phenology, and community structure in an upland grassland ecosystem.
Ecosystems (N Y) 13(6):888900.
34. Kongstad J, et al. (2012) High resilience in heathland plants to changes in temperature, drought, and high resilience in heathland plants to changes in temperature,
drought, and CO2 in combination: results from the CLIMAITE experiment. Ecosystems
(N Y) 15(2):269283.
35. Brookshire ENJ, Weaver T (2015) Long-term decline in grassland productivity driven
by increasing dryness. Nat Commun 6:7148.
36. Way DA (2011) The bigger they are, the harder they fall: CO2 concentration and tree
size affect drought tolerance. Tree Physiol 31(2):115116.
37. Hovenden MJ, Newton PCD, Wills KE (2014) Seasonal not annual rainfall determines
grassland biomass response to carbon dioxide. Nature 511(7511):583586.
38. van Groenigen KJ, Qi X, Osenberg CW, Luo Y, Hungate BA (2014) Faster decomposition under increased atmospheric CO2 limits soil carbon storage. Science
344(6183):508509.
39. Pendall E, et al. (2013) Warming reduces carbon losses from grassland exposed to
elevated atmospheric carbon dioxide. PLoS One 8(8):e71921.
40. Luo Y, et al. (2008) Modeled interactive effects of precipitation, temperature, and
CO2 on ecosystem carbon and water dynamics in different climatic zones. Glob
Change Biol 14(9):19861999.
41. Zscheischler J, et al. (2014) Carbon cycle extremes during the 21st century in CMIP5
models: Future evolution and attribution to climatic drivers. Geophys Res Lett 41(24):
88538861.
42. Robredo A, et al. (2011) Elevated CO2 reduces the drought effect on nitrogen metabolism in barley plants during drought and subsequent recovery. Environ Exp Bot
71(3):399408.
43. Arndal MF, et al. (2013) Net root growth and nutrient acquisition in response to
predicted climate change in two contrasting heathland species. Plant Soil 369(1):
615629.
44. Feng Z, et al. (2015) Constraints to nitrogen acquisition of terrestrial plants under
elevated CO2. Glob Change Biol 21(8):31523168.
45. Reich PB, Hobbie SE (2012) Decade-long soil nitrogen constraint on the CO2 fertilization of plant biomass. Nat Clim Chang 3:278282.
46. Luo Y, et al. (2004) Progressive nitrogen limitation of ecosystem responses to rising
atmospheric carbon dioxide. Bioscience 54(8):731739.
47. Bahn M, Reichstein M, Dukes JS, Smith MD, McDowell NG (2014) Climate-biosphere
interactions in a more extreme world. New Phytol 202(2):356359.
48. Smith J, et al. (2007) Projected changes in the organic carbon stocks of cropland
mineral soils of European Russia and the Ukraine, 1990-2070. Glob Change Biol 13(2):
342356.
49. Van Oijen M, et al. (2014) Impact of droughts on the carbon cycle in European vegetation: A probabilistic risk analysis using six vegetation models. Biogeosciences
11(22):63576375.
50. Soussana JF, Graux AI, Tubiello FN (2010) Improving the use of modelling for projections of climate change impacts on crops and pastures. J Exp Bot 61(8):22172228.
51. Nakicenovic N, et al. (2000) IPCC Special Report on Emission Scenarios (Cambridge
Univ Press, Cambridge, UK).
52. Milcu A, et al. (2014) Functional diversity of leaf nitrogen concentrations drives
grassland carbon fluxes. Ecol Lett 17(4):435444.
53. Casella E, Soussana JF (1997) Long-term effects of CO2 enrichment and temperature
increase on the carbon balance of a temperate grass sward. J Exp Bot 48(311):
13091321.
54. Reichstein M, et al. (2005) On the separation of net ecosystem exchange into assimilation and ecosystem respiration: Review and improved algorithm. Glob Change Biol
11(9):14241439.
55. Lensi R, Mazurier S, Gourbiere F, Josserand A (1986) Rapid determination of the nitrification potential of an acid forest soil and assessment of its variability. Soil Biol
Biochem 18(2):239240.
56. Pinay G, et al. (2007) Impact of atmospheric CO2 and plant life forms on soil microbial
activities. Soil Biol Biochem 39(1):3342.
57. Pinheiro J, Bates D, DebRoy S, Sarkar D, R Development Core Team (2016) nlme:
Linear and Nonlinear Mixed Effects Models. R Package version 3.197. Available at
CRAN.R-project.org/package=nlme. Accessed April 2016.
58. Zuur AF, Ieno EN, Walker N, Saveliev AA, Smith GM (2009) Mixed Effects Models and
Extensions in Ecology with R (Springer, New York), 10.1007/978-0-387-87458-6.

Roy et al.

SUPPORTING INFORMATION
========================================================
SI Appendix for:
Elevated CO2 maintains grassland net carbon uptake under a future heat
and drought extreme
Jacques Roy, Catherine Picon-Cochard, Angela Augusti, Marie-Lise Benot, Lionel Thiery,
Olivier Darsonville, Damien Landais, Clment Piel, Marc Defossez, Sebastien Devidal,
Christophe Escape, Olivier Ravel, Nathalie Fromin, Florence Volaire, Alexandru Milcu,
Michael Bahn, and Jean-Franois Soussana

Table of content:
Supplementary Methods

page 2

Supplementary Figures

page 3

Fig. S1 .. page 3
Fig. S2... page 4
Fig. S3 .. page 5
Fig. S4 .. page 6
Fig. S5... page 7
Fig. S6 .. page 8
Supplementary Tables

page 9

Table S1 page 9
Table S2 page 10
Table S3 page 11
Table S4 page 12
Table S5 page 13
Table S6 page 14
Table S7 page 15

Supplementary Methods
The Ecotron Macrocosms platform.
The Ecotron Macrocosms platform houses twelve identical and independent experimental
units (macrocosms) oriented east-west. Two additional fake units at the east and west sides
eliminate the edge effect associated with self-shading of the domes at the early and late hours
of the day in fall and winter. Each unit is covered by a 30 m3 transparent dome (Teflon-FEP
film, 250m thick, DuPont, USA) that allows for the confinement and control of the
atmosphere. Below each dome a dedicated technical room hosts the belowground
compartment of the ecosystem contained in a lysimeter, the lysimeter weighing strain gauges,
various soil-related sensors and the air temperature conditioning units (SI Appendix Fig. S1).
The access into the domes is through an airlock situated on the north side. Each dome has a
circular base area of 25 m2, of which 20 m2 is the concrete roof of the technical room and a
central 5m2 area allows receiving 2, 4 or 5 m2 lysimeters (4 m in this experiment). The
airflow from the dome area is prevented to leak into the technical room by the means of
fitting metal plates and rubber seals between the concrete roof and the upper ring of the
lysimeters. Concrete surfaces inside the domes are covered with epoxy-resin to prevent CO2
absorption. The lysimeter rests on a raising platform allowing adjusting for various lysimeter
depths up to 2 m. An internal air circulation of 2 volume minute-1 in each dome ensures a
good air temperature and relative humidity regulation. It results in a wind speed of 1 m s -1 at
the canopy level.
Microbial nitrification.
Microbial nitrification potential was determined according to the method of Lensi et al.
(1986) which is described in more detail by Pinay et al. (2007) and consists of quantifying
nitrate produced by nitrification as follows. Soil was sampled after the November canopy
harvest on each of the four monoliths per macrocosm (0-15 cm depth). A first 20 g DW soil
sample was allowed to nitrify for 48 h under oxic condition and optimal (80%) water holding
capacity after enrichment with 0.2 mg (NH4)2SO4- N g-1 soil. The resulting nitrate was fully
converted into N2O during a 24 h anaerobic incubation by active (overnight cultured)
bacterial Pseudomonas fluorescens strain AK15 in the presence of 1 mg C (provided half as
glucose and half as glutamic acid) and a 90:10 He-C2H2 atmosphere (vol:vol) to ensure
anaerobic condition and N2O reductase inhibition. The resulting N2O was analyzed by gas
chromatography with 63Ni electron capture detector (Varian Star 3800, Varian). The initial
residual nitrate content was determined the same was on a second soil subsample. The
nitrification activity, as amount of N nitrified g-1 soil h-1, was deduced by subtracting the
nitrate-N initially present from the nitrate-N accumulated after nitrification step and
expressed as g N nitrified g-1 soil h-1.
Lensi, R., Mazurier, S., Gourbiere, F. & Josserand, A. (1986) Rapid determination of the nitrification
potential of an acid forest soil and assessment of its variability. Soil Biol. Biochem. 18, 239240.
Pinay, G. et al. (2007) Impact of atmospheric CO2 and plant life forms on soil microbial activities.
Soil Biol. Biochem. 39, 3342.

Supplementary Figures

Figure S1 | Simplified schematic of one macrocosm experimental unit of the Montpellier

CNRS Ecotron facility. Ecotron

Figure S2| Snapshot taken during the extraction of the soil monoliths. The soil around the
monoliths was excavated before the vertical and horizontal (bottom) cuts were made with the
help of a hydraulic press. The soil is characterized as a Cambisol (59.5% sand, 19.2% silt,
21.4% clay, pHH2O = 5.9). Ecotron

Leaf Area Index (m2 m-2)

Figure S3 | Seasonal dynamics of canopy response variables as affected by the CO 2 and

extreme climatic event (ECE) treatments. (a) Leaf area index. (b) Greenness. Data
represent individual measurement points SEM of three replicates. Stars indicate significant
terms according to repeated measures ANOVAs (* for P<0.05, ** for P<0.01 and *** for
P<0.001). Horizontal bottom bars represent the distinct experimental periods (solid green:
pre-stress; solid orange: halved water supply; solid red: no water supply +3.4C temperature
increase; hatched orange bar: gradual rewetting; hatched green bar: post-stress (see Methods
for details).

Pre-stress

3.0

Post-stress

390CO2
390CO2 ECE
520CO2
520CO2 ECE

2.5
2.0
1.5
1.0
0.5
0.0

a)
May
CO2 *

80

Greenness (%)

Stress

Jun

Jul

Aug

Sep

Oct

Sep

Oct

CO2*** ECExTime ***

60

40

20

b)
0

May

Jun

Jul

Aug

Shoot N content (mg g -1)

Figure S4 | Effect of CO2 and extreme climatic event (ECE) treatments on shoot
nitrogen (N) content. Data are means SEM of 3 replicates.

Cut Nov 3

Cut Jun 9

390 CO2
CO2 ** ECE ***
390 CO2 ECE
520 CO2
520 CO2 ECE

CO2 ***

May

J un

J ul

Aug

Sep

Oc t

Nov

Figure S5 | Effect of CO2 and extreme climatic event (ECE) treatments on a) gross
primary productivity (GPP) and b) net ecosystem CO2 exchange (NEE) during the
second two weeks of September when the CO2 x ECE interaction is significant. Data are
means SEM of 3 replicates. Error bars are shown only upwards to avoid lines overlapping.

390CO2
390CO2 ECE
520CO2
520CO2 ECE

CO2 x ECE x Day ***

-2

-1

NEE (gCO2 m day )

4
3
2
1
0

a)
-1
Sep20

Sep22

Sep24

Sep26

Sep28

Sep30

-2

-1

GPP (gCO2 m day )

Sep18

2
CO2 x ECE x Day ***

b)

0
Sep18

Sep20

Sep22

Sep24

Sep26

Sep28

Sep30

Figure S6 | Effect of CO2 and extreme climatic event (ECE) treatments on a) root
nitrogen (N) pool and b) root N content. Data are means SEM of 3 replicates and
correspond to the post-stress soil core samplings.

390CO2
390CO2 ECE
520CO2
520CO2 ECE

-2
Root N pool (gN m )

2.0

1.5

CO2 x ECE x Time: P2/6 = 0.056

1.0

Sep 20
Aug 9

0.5

0.0
1
Root N content (mg g - )

Nov 3

a)
Aug 9

2.5

Sep 20

Nov 3

2.0
1.5
1.0
0.5
0.0

b)

Supplementary Tables

Table S1| Average botanical composition of the sampled monoliths assessed on 20 contact
points per monolith.
Species

Functional group

Percentage
cover (%)

Trifolium repens

Legume

28.5

Lolium perenne

Grass

15.5

Holcus lanatus

Grass

13.6

Alopecurus pratensis

Grass

6.7

Lathyrus pratense

Legume

6.0

Agrostis tenuis

Grass

5.9

Trisetum flavescens

Grass

5.0

Poa trivialis

Grass

3.8

Dactylis glomerata

Grass

3.8

Rumex acetosa

Forb

3.4

Ranonculus acris

Forb

3.3

Trifolium pratense

Legume

2.3

Poa pratensis

Grass

2.2

Table S2 | Effects of CO2 and the extreme climatic event (ECE) including a combined drought and heat
wave on ecosystem-level response variables related to C and water fluxes (NEE = net ecosystem
exchange, GPP = gross primary productivity, Reco = ecosystem respiration, ET = evapotranspiration,
WUE = water-use efficiency) as analyzed by repeated measures factorial ANOVAs on fortnightly
averages. We present minimal adequate models after model simplification based on the Akaikes
Information Criterion (AIC) as performed by the stepAIC function in R. Treatments or interactions that
have been found to be non-significant have been excluded (see Excl.) from the minimal adequate models.
P-values and numerator and denominator degrees of freedom are shown. The three experimental periods
have been defined as: pre-stress (April 26th to June 24th), stress (June 25th to August 31th) and recovery
(September 1st to October 30th). Significant differences (P < 0.05) are marked in bold.
Pre-stress
Treatment / Variable

NEE

GPP

Reco

ET

WUE

CO2

Excl.

P1/10 = 0.021

Excl.

P1/10 = 0.039

P1/10 = 0.001

Time

P3/21 < 0.001

P3/19 < 0.001

P3/22 < 0.001

P3/30 < 0.001

P3/19 < 0.001

CO2 Time

Excl.

P3/19 = 0.073

Excl.

P3/30 < 0.001

P3/19 = 0.063

Treatment / Variable

NEE

GPP

Reco

ET

WUE

CO2

P1/8 = 0.022

P1/8 = 0.014

P1/8 = 0.075

P1/8 = 0.030

P1/8 = 0.006

ECE

P1/8 < 0.001

P1/8 < 0.001

P1/8 = 0.005

P1/8 < 0.001

P1/8 < 0.001

Time

P4/32 < 0.001

P4/32 < 0.001

P4/32 < 0.001

P4/32 < 0.001

P4/30 < 0.001

CO2 ECE

P1/8 = 0.555

P1/8 = 0.808

Excl.

P1/8 = 0.191

P1/8 = 0.979

CO2 Time

P4/32 < 0.001

P4/32 < 0.001

Excl.

P4/32 = 0.488

P4/30 < 0.001

ECE Time

P4/32 < 0.001

P4/32 < 0.001

P4/40 < 0.001

P4/32 < 0.001

P4/30 < 0.001

CO2 ECE Time

P4/32 = 0.003

P4/32< 0.001

Excl.

P4/32 = 0.151

P4/30 = 0.121

Treatment / Variable

NEE

GPP

Reco

ET

WUE

CO2

P1/8 = 0.004

P1/8 < 0.001

P1/8 = 0.009

P1/8 = 0.202

P1/8 = 0.016

ECE

P1/8 = 0.032

P1/8 = 0.006

P1/8 = 0.341

P1/8 = 0.134

P1/8 = 0.018

Time

P3/27 < 0.001

P3/27 < 0.001

P3/27 < 0.001

P3/27 < 0.001

P3/30 < 0.001

CO2 ECE

P1/8 = 0.067

P1/8 = 0.073

Excl.

P1/8 = 0.171

Excl.

CO2 Time

P3/27 = 0.137

P3/27 = 0.008

P3/27 = 0.003

Excl.

Excl.

ECE Time

P3/27 < 0.001

P3/27 < 0.001

P3/27 = 0.015

P3/27 < 0.001

P3/30 = 0.004

CO2 ECE Time

Excl.

Excl.

Excl.

Excl.

Excl.

During stress

Post-stress

10

Table S3 | Effects of atmospheric CO2 concentration and extreme climatic events (ECE) including a
combined drought and heat wave on measured vegetation-related response variables as analyzed by
ANOVA or repeated measures ANOVA when appropriate. We present minimal adequate models after
model simplification based on the Akaikes Information Criterion (AIC) as performed by the stepAIC
function in R. Treatments or interactions that have been found to be non-significant and have been
excluded (Excl.) from the minimal adequate models. NA denotes non-applicable analyses of Time when
only one sampling data points is available. P-values and numerator and denominator degrees of freedom
are shown. The three experimental periods have been defined as: pre-stress (April 26th to June 24th),
stress (June 25th to August 31th) and recovery (September 1st to October 30th).
Pre-stress
Shoot
biomass
P1/10 = 0.662

LAI

Greenness

Soil moisture

CO2

Root
growth rate
P1/10 = 0.078

Excl.

P1/10 = 0.025

Excl.

Time

P1/10 = 0.976

NA

P3/33 < 0.001

NA

P3/33 < 0.001

CO2 Time

P1/10 = 0.090

NA

Excl.

NA

Excl.

Shoot
biomass
NA

LAI

Greenness

Soil moisture

CO2

Root
growth rate
P1/9= 0.064

P1/8 = 0.118

P1/9 = 0.006

P1/8 = 0.708

ECE

P1/9 =0.017

NA

P1/8 = 0.181

P1/9 < 0.001

P1/8 < 0.001

Time

NA

NA

P4/32 < 0.001

P1/18 < 0.001

P4/32 < 0.001

CO2 ECE

Excl.

NA

P1/8 = 0.686

Excl.

P1/8 = 0.378

CO2 Time

NA

NA

P4/32 = 0.682

P4/32 = 0.094

P4/32 = 0.081

ECE Time

NA

NA

P4/32= 0.004

P2/18 < 0.001

P4/32< 0.001

CO2 ECE Time

NA

NA

P4/32 = 0.092

Excl.

P4/32 = 0.135

Shoot
biomass
Excl.

LAI

Greenness

CO2

Root
growth rate
P1/8 = 0.031

ECE

P1/8 = 0.100

Time

Treatment / Variable

During stress
Treatment / Variable

Recovery

P1/9 = 0.029

Excl.

Soil
moisture
P1/8 = 0.073

Excl.

P1/9 = 0.144

Excl.

P1/8 = 0.182

P1/8 = 0.055

NA

P3/30 < 0.001

Excl.

P3/30 < 0.001

CO2 ECE

P1/8 = 0.067

Excl.

Excl.

Excl.

P1/8= 0.008

CO2 Time

P1/8 = 0.247

NA

Excl

Excl.

Excl.

ECE Time

P1/8= 0.029

NA

P3/30 = 0.064

Excl.

P1/30 = 0.002

CO2 ECE Time

P1/8= 0.042

NA

Excl.

Excl.

Excl.

Treatment / Variable

11

Table S4 | Effects of atmospheric CO2 concentration and extreme climatic events (ECE)
including a combined drought and heat wave on shoot and root nitrogen (N) pool and %
measured by ANOVA or repeated measures ANOVA when appropriate. We present
minimal adequate models after model simplification based on the Akaikes Information
Criterion (AIC) as performed by the stepAIC function in R. Treatments or interactions that
have been found to be non-significant and have been excluded (Excl.) from the minimal
adequate models. NA denotes non-applicable analyses of Time when only one sampling data
points is available. P-values and numerator and denominator degrees of freedom are shown.
For root and shoot data two experimental periods have been defined as: pre-stress and
recovery periods.

Pre-stress
Treatment / Variable Shoot N
pool
CO2
Excl.

Shoot N
Root N
Root N
content
pool
content
P
=
0.068
Excl.
P1/10 < 0.001 1/10

Time

NA

NA

3/30 <

CO2 Time

Excl.

Excl.

Excl.

0.001

P3/30 < 0.001

Excl.

Recovery
Treatment / Variable Shoot N
pool
CO2
Excl.

Shoot N
Root N
Root N
content
pool
content
Excl.
P1/9 = 0.006 P1/8 = 0.024

ECE

P1/10 = 0.011 P1/9 < 0.001 P1/8 = 0.218

Excl.

Time

NA

NA

P2/16 < 0.001

Excl.

CO2 ECE

Excl.

Excl.

P1/8 = 0.442

Excl.

CO2 Time

NA

NA

P2/16 = 0.047

Excl.

ECE Time

NA

NA

P2/16= 0.003

Excl.

CO2 ECE Time

NA

NA

P2/16= 0.056

Excl.

12

Table S5 | Table output presenting the results from a repeated measures analysis performed
with daily values for last two weeks of September using a mixed effect models with the
Ecotrons macrocosms as a random factor and including an autoregressive covariance
structure between the individual macrocosms and day-dependent variance coefficients.
R syntax: model = lme(NEE~CO2*ECE*Day, random= ~ 1|macrocosm,
correlation=corAR1(form = ~ 1 | macrocosm), weights=varIdent(form = 1 | Day),
method="REML).
Source

num/den NEE

NEE

GPP

GPP

DF

F-value P-value

F-value P-value

CO2

1/8

17.56

0.003

24.81

0.001

ECE

1/8

16.06

0.004

15.14

0.001

Day

13/99

153.46

<0.001

115.68

<0.001

CO2 ECE

1/8

4.19

0.075

3.41

0.102

CO2 Day

13/99

8.16

<0.001

8.14

<0.001

ECE Day

13/99

10.78

<0.001

7.98

<0.001

4.68

<0.001

3.01

<0.001

CO2 ECE Day 13/99

13

Table S6 | Comparison of outputs of analysis of variance (ANOVA) and covariance

(ANCOVA) introducing community root and shoot nitrogen (N) pools as a co-variables in
the models for ecosystem gross primary production (GPP) and net ecosystem exchange
(NEE) averaged from the last two weeks of October.

Source

NEE

NEE

GPP

GPP

P-values

P-values

P-values

P-values

ANOVA

ANCOVA ANOVA

ANCOVA

Root N

without

0.002

without

<0.001

Shoot N

without

0.072

without

0.018

CO2

0.114

0.731

0.011

0.113

ECE

0.021

0.573

0.015

0.389

CO2 ECE

0.076

0.270

0.128

0.313

1/8

1/6

1/8

1/6

DF

14

Table S7 | ANOVA table output for the effect of atmospheric CO2 concentration and extreme
climatic events (ECE) including a combined drought and heat wave on cumulative (April 26th
November 3rd) ecosystem fluxes of net ecosystem exchange (NEE), gross primary
productivity (GPP), respiration (Reco), evapotranspiration (ET), water-use efficiency (WUE)
soil nitrification potential (SNP) and ecosystem carbon incorporation (as the difference
between the carbon in biomass harvested on June 9th and November 3rd and NEE flux). Table
statistics apply to the bar plot charts of Fig. 2 (right side) and Fig.4a,b for SNP and carbon
balance.

Source

num/

NEE

GPP

Reco

ET

WUE

SNP

den DF

Carbon
balance

CO2

1/8

0.001

0.002

Excl.

Excl.

0.009

Excl.

0.018

ECE

1/8

0.030

0.019

Excl.

<0.001

Excl.

0.023

0.051

CO2 ECE

1/8

Excl.

Excl.

Excl.

Excl.

Excl.

Excl.

Excl.

15