Accepted Article

Received Date : 10-Jul-2014 Accepted Date : 14-Dec-2014 Article type

: Primary Research Articles

Tree growth variation in the tropical forest: understanding effects of temperature, rainfall and CO2 Running head: Tree growth variation in the tropical forest Peter Schippers1,2, Frank Sterck1, Mart Vlam1, Pieter A. Zuidema1 1

Forest Ecology and Forest Management Group, Wageningen University, PO Box 47, 6700 AA Wageningen, The Netherlands 2 Team Biodiversity and Policy, ALTERRA, Wageningen University and Research Centre, PO Box 47, 6700 AA Wageningen, Netherlands Corresponding Author Peter Schippers Email: [email protected] Phone: +31317485018 Keywords: carbon dioxide, climate change, CO2, NPP, pathway, photosynthesis, precipitation, temperature, transpiration, tree-ring

Abstract Tropical forest responses to climatic variability have important consequences for global carbon cycling, but are poorly understood. Since empirical, correlative studies cannot disentangle the interactive effects of climatic variables on tree growth, we used a tree growth model (IBTREE) to unravel the climate effects on different physiological pathways and in turn on stem growth variation. We parameterized the model for canopy trees of Toona ciliata (Meliaceae) from a Thai monsoon forest and compared predicted and measured variation from a tree-ring study over a 30-year period. We used historical climatic variation of This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/gcb.12877 This article is protected by copyright. All rights reserved.

Accepted Article

minimum and maximum day temperature, precipitation and carbon dioxide (CO2) in different combinations to estimate the contribution of each climate factor in explaining the inter-annual variation in stem growth. Running the model with only variation in maximum temperature and rainfall yielded stem growth patterns that explained almost 70% of the observed interannual variation in stem growth. Our results show that maximum temperature had a strong negative effect on the stem growth by increasing respiration, reducing stomatal conductance and thus mitigating a higher transpiration demand, and - to a lesser extent - by directly reducing photosynthesis. Although stem growth was rather weakly sensitive to rain, stem growth variation responded strongly and positively to rainfall variation owing to the strong inter-annual fluctuations in rainfall. Minimum temperature and atmospheric CO2 concentration did not significantly contribute to explaining the inter-annual variation in stem growth. Our innovative approach – combining a simulation model with historical data on tree-ring growth and climate – allowed disentangling the effects of strongly correlated climate variables on growth through different physiological pathways. Similar studies on different species and in different forest types are needed to further improve our understanding of the sensitivity of tropical tree growth to climatic variability and change.

Introduction Tropical forests cover 12 % of the earth’s land surface, but are responsible for 33.5% of the terrestrial gross primary production and contain 25% of the carbon of the terrestrial biosphere (Beer et al., 2010; Bonan, 2008). Because of the large photosynthetic capacity and massive carbon storage, tropical forests are a crucial factor in the global carbon cycle (Luyssaert et al., 2007; Pan et al., 2011). Carbon fluxes and stocks in tropical forests are largely determined by

This article is protected by copyright. All rights reserved.

Accepted Article

where the respiration coefficients of the plant organs are leaf (Rl), sapwood (Rs), fine roots (Rr) and reserves (Rrs) (kg CH2O kg DM-1 d-1). The reserves are used to bridge unfavourable conditions when maintenance respiration exceeds photosynthesis, for instance at night and during the dry season. We assumed that reserves also respire because storage organs in plants often have a considerable respiration (Goudriaan & Van Laar, 1994; Penningdevries, 1975). We assume that the maintenance respiration doubles with a 10oC (Q10=2) temperature increase (Atkinson et al., 2007; Ryan et al., 1994; Schippers & Kropff, 2001) so:

(4)

where T is the actual temperature (oC) and Tr is a reference temperature at which respiration coefficients are measured (oC). This relation, however, is a simplification because plants can thermally adapt by reducing the Q10 at higher temperature (Atkin & Tjoelker, 2003; Slot et al., 2014). Because the temperature roughly varies between of 20oC in winter and 27oC in summer at our study site, we expected values of Q10 of 2.4 in winter and 1.8 in summer (Atkin & Tjoelker, 2003) and considered Q10 = 2 a reasonable approximation. Since long term acclimation may nevertheless mitigate actual Q10 values, we ran additional simulations with Q10=1.5, but this did not change mean growth by more than 1% and did not alter the main drivers of variation in inter-annual growth. Furthermore, Q10 was not a very sensitive parameter because an increase in respiration below 25oC (reference temperature) is compensated for temperature increases above 25oC, and vice versa.

The light use efficiency (equation 2) depends on both temperature (Fp) (Pepper et al., 2008; Schippers & Joenje, 2002; Schippers et al., 2004) and internal CO2 concentration Ci (FCi, ppm) (Haxeltine & Prentice, 1996):

This article is protected by copyright. All rights reserved.

Accepted Article

order to maintain seasonality in our simulations. We ran these simulations for single climatic variables, for several climatic variables combined and for all climatic variables. For each of the simulations, we correlated the simulated temporal variation in stem growth with the observed pattern of growth variation to assess the degree of similarity and the contribution of variation in driving stem growth patterns. A comparison of the correlation coefficients reveal the relative importance of inter-annual variation of minimum temperature, maximum temperature, CO2, and rainfall in driving the observed variation in Toona stem growth.

Results Sensitivity of simulated stem growth to climatic variation We used a hierarchical regression model to quantify the impacts of three pathways by which climatic variables influence tree physiology and finally stem growth. This model explained 99.9% of the inter-annual variation in simulated sapwood production. The photosynthesis, respiration and water availability pathways resulting from this analyses are presented in Fig. 2, along with the results of the sensitivity analysis. Our model assumptions for temperature impacts imply that a 1%-increase in maximum temperature and minimum temperature both increased tree respiration, but also that the impact of maximum temperature was twice as large as the impact of minimum temperature. Moreover, maximum temperature had a strong negative effect on relative stomatal conductance Fw, and this was only partially compensated by smaller, positive, effects of minimum temperature, rainfall, and CO2 on this variable. The maximum temperature also directly decreased photosynthesis whereas the direct effect of the minimum temperature was marginal. Both stomatal conductance and CO2 increased photosynthesis directly, with the effect of CO2 being stronger (Fig. 2). Overall, stem biomass growth was most sensitive to maximum day temperature because this had a strong negative effect on stem growth through negative impacts on photosynthesis and

This article is protected by copyright. All rights reserved.

Accepted Article

Clark DA (2004) Sources or sinks? The responses of tropical forests to current and future climate and atmospheric composition. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences, 359, 477-491. Clark DA, Clark DB, Oberbauer SF (2013) Field-quantified responses of tropical rainforest aboveground productivity to increasing CO2 and climatic stress, 1997–2009. Journal of Geophysical Research: Biogeoscience, 118, 1-12. Clark DA, Piper SC, Keeling CD, Clark DB (2003) Tropical rain forest tree growth and atmospheric carbon dynamics linked to interannual temperature variation during 1984-2000. Proceedings of the National Academy of Sciences of the United States of America, 100, 5852-5857. Clark DB, Clark DA, Oberbauer SF (2010) Annual wood production in a tropical rain forest in NE Costa Rica linked to climatic variation but not to increasing CO2. Global Change Biology, 16, 747-759. Clark DB, Olivas PC, Oberbauer SF, Clark DA, Ryan MG (2008) First direct landscape-scale measurement of tropical rain forest Leaf Area Index, a key driver of global primary productivity. Ecology Letters, 11, 163-172. Comins HN, Mcmurtrie RE (1993) Long-term response of nutrient-limited forest to CO2 enrichment - Equilibrium behavior of plant-soil models. Ecological Applications, 3, 666-681. Couralet C, Sterck FJ, Sass-Klaassen U, Van Acker J, Beeckman H (2010) Species-Specific Growth Responses to Climate Variations in Understory Trees of a Central African Rain Forest. Biotropica, 42, 503-511. Cramer W, Bondeau A, Woodward FI et al. (2001) Global response of terrestrial ecosystem structure and function to CO2 and climate change: results from six dynamic global vegetation models. Global Change Biology, 7, 357-373. Dong SX, Davies SJ, Ashton PS et al. (2012) Variability in solar radiation and temperature explains observed patterns and trends in tree growth rates across four tropical forests. Proceedings of the Royal Society B-Biological Sciences, 279, 3923-3931. Dunisch O, Montoia VR, Bauch J (2003) Dendroecological investigations on Swietenia macrophylla King and Cedrela odorata L. (Meliaceae) in the central Amazon. Trees-Structure and Function, 17, 244-250. Edwards NT, Hanson PJ (1996) Stem respiration in a closed-canopy upland oak forest. Tree Physiology, 16, 433-439. Farquhar GD, Buckley TN, Miller JM (2002) Optimal stomatal control in relation to leaf area and nitrogen content. Silva Fennica, 36, 625-637. Feeley KJ, Wright SJ, Supardi MNN, Kassim AR, Davies SJ (2007) Decelerating growth in tropical forest trees. Ecology Letters, 10, 461-469. Gharun M, Turnbull TL, Adams MA (2013) Validation of canopy transpiration in a mixed-species foothill eucalypt forest using a soil-plant-atmosphere model. Journal of Hydrology, 492, 219-227. Goudriaan J, Van Laar HH (1994) Modelling potential crop growth processes, Dordrecht, Kluwer Academic Publishers. Grant RF, Goulden ML, Wofsy SC, Berry JA (2001) Carbon and energy exchange by a black spruce-moss ecosystem under changing climate: Testing the mathematical model ecosys with data from the BOREAS experiment. Journal of Geophysical Research-Atmospheres, 106, 33605-33621. Haxeltine A, Prentice IC (1996) A general model for the light-use efficiency of primary production. Functional Ecology, 10, 551-561. Hickler T, Smith B, Sykes MT, Davis MB, Sugita S, Walker K (2004) Using a generalized vegetation model to simulate vegetation dynamics in northeastern USA. Ecology, 85, 519-530. Holmes RL (1983) Computer-assisted quality control in tree-ring dating and measurement. Tree-ring Bulletin, 43, 69-78. Holtum JaM, Winter K (2010) Elevated CO2 and forest vegetation: more a water issue than a carbon issue? Functional Plant Biology, 37, 694-702. Keenan TF, Baker I, Barr A et al. (2012) Terrestrial biosphere model performance for inter-annual variability of land-atmosphere CO2 exchange. Global Change Biology, 18, 1971-1987. Kirk JTO (1994) Light an photosynthesis in aquatic systems, Cambridge, Cambridge University Press. Kirschbaum MUF (1999) CenW, a forest growth model with linked carbon, energy, nutrient and water cycles. Ecological Modelling, 118, 17-59. Koerner C (2009) Responses of Humid Tropical Trees to Rising CO(2). In: Annual Review of Ecology Evolution and Systematics. pp Page. Leakey ADB, Press MC, Scholes JD (2003) High-temperature inhibition of photosynthesis is greater under sunflecks than uniform irradiance in a tropical rain forest tree seedling. Plant Cell and Environment, 26, 1681-1690. Lewis SL, Brando PM, Phillips OL, Van Der Heijden GMF, Nepstad D (2011) The 2010 Amazon Drought. Science, 331, 554-554. Lewis SL, Lopez-Gonzalez G, Sonke B et al. (2009) Increasing carbon storage in intact African tropical forests. Nature, 457, 1003-U1003. Liu N, Dang QL, Parker WH (2006) Genetic variation of Populus tremuloides in ecophysiological responses to CO2 elevation. Canadian Journal of Botany-Revue Canadienne De Botanique, 84, 294-302. Lloyd J, Farquhar GD (2008) Effects of rising temperatures and CO2 on the physiology of tropical forest trees. Philosophical Transactions of the Royal Society B-Biological Sciences, 363, 1811-1817. Luo YQ, Gerten D, Le Maire G et al. (2008) Modeled interactive effects of precipitation, temperature, and CO2 on ecosystem carbon and water dynamics in different climatic zones. Global Change Biology, 14, 1986-1999. Luyssaert S, Inglima I, Jung M et al. (2007) CO2 balance of boreal, temperate, and tropical forests derived from a global database. Global Change Biology, 13, 2509-2537.

This article is protected by copyright. All rights reserved.

Accepted Article

The changes of leaf biomass (Wl), the sapwood biomass (Ws), the root biomass (Wr), and the reserves (Wrs) can be described as the change in the individual plant organ weight, Wo, (kg DM tree-1, DM = dry matter):

(1)

where A is the assimilation rate (kg CH2O tree-1 d-1; CH2O = carbohydrates), Rm is the maintenance respiration rate of the tree (kg CH2O tree-1 d-1 ), Fo is allocation factor to organs that comes from the allocation module, CVFo is conversion factor between dry matter and carbohydrates and biomass (kg DM kg CH2O-1) and Mo is the turnover of the plant part (d-1). The assimilation rate A depends on the amount of absorbed radiation and the light use efficiency (Monteith, 1977; Haxeltine & Prentice, 1996; Hickler et al., 2004; Makela et al., 2008; Pepper et al., 2008):

(2)

where I is the amount radiation on top of the canopy (MJ PAR m-2 d-1), LUE(T,Ci) is the light use efficiency dependent on temperature and internal CO2 concentration (Ci) (kg CO2 Mj PAR-1), k is the light extinction coefficient of the canopy (m2 ground area/m2 leaf area), SLA is the specific leaf area (m2 kg-1), Area is the total crown area (m2), and 30/44 is the molar ratio between carbohydrates (CH2O) and CO2. The maintenance respiration of the tree Rm (kg CH2O tree-1 d-1) is dependent on the dry matter weight of living tree organs and a temperature dependent factor Fr:

(3) This article is protected by copyright. All rights reserved.

Accepted Article

where the respiration coefficients of the plant organs are leaf (Rl), sapwood (Rs), fine roots (Rr) and reserves (Rrs) (kg CH2O kg DM-1 d-1). The reserves are used to bridge unfavourable conditions when maintenance respiration exceeds photosynthesis, for instance at night and during the dry season. We assumed that reserves also respire because storage organs in plants often have a considerable respiration (Goudriaan & Van Laar, 1994; Penningdevries, 1975). We assume that the maintenance respiration doubles with a 10oC (Q10=2) temperature increase (Atkinson et al., 2007; Ryan et al., 1994; Schippers & Kropff, 2001) so:

(4)

where T is the actual temperature (oC) and Tr is a reference temperature at which respiration coefficients are measured (oC). This relation, however, is a simplification because plants can thermally adapt by reducing the Q10 at higher temperature (Atkin & Tjoelker, 2003; Slot et al., 2014). Because the temperature roughly varies between of 20oC in winter and 27oC in summer at our study site, we expected values of Q10 of 2.4 in winter and 1.8 in summer (Atkin & Tjoelker, 2003) and considered Q10 = 2 a reasonable approximation. Since long term acclimation may nevertheless mitigate actual Q10 values, we ran additional simulations with Q10=1.5, but this did not change mean growth by more than 1% and did not alter the main drivers of variation in inter-annual growth. Furthermore, Q10 was not a very sensitive parameter because an increase in respiration below 25oC (reference temperature) is compensated for temperature increases above 25oC, and vice versa.

The light use efficiency (equation 2) depends on both temperature (Fp) (Pepper et al., 2008; Schippers & Joenje, 2002; Schippers et al., 2004) and internal CO2 concentration Ci (FCi, ppm) (Haxeltine & Prentice, 1996):

This article is protected by copyright. All rights reserved.

Accepted Article

(5)

where LUEm is the maximum light use efficiency under optimal conditions (kg CO2 MJ PAR1

), i.e. optimal temperature and at saturating water supply and CO2 levels. Fp reflects the

direct effect of temperature on photosynthesis and FCi the carbon limitation ruled by the internal CO2 concentration of the leaves. Both water availability (Fw) and atmospheric CO2 affect this internal CO2 concentration. Since water availability determines the stomatal conductance, it is also crucial in determining the internal CO2 concentration (Ci). For CO2 limitation we use a Michaelis Menten reduction based on the internal CO2 concentration of the leaves Ci (ppm) (Farquhar et al., 2002; Sterck & Schieving, 2011; Sterck et al., 2011):

.

(6)

Here, Ci is the CO2 concentration in the leaves (ppm), Cc is the CO2 compensation concentration (ppm), and KmC is the Michaelis Menten constant for the carboxylation process (ppm). We use the fraction Cai to relate Ci to the atmospheric CO2 concentration Ca (ppm), in the absence of water stress. We use the value 0.7 for Cai (Grant et al., 2001, Liu et al., 2006, Tricker et al., 2005). Note that the parameterised equation 6 according to Table 1 resembles the CO2-LUE response as published by (Haxeltine & Prentice, 1996). The next step is to relate water availability to relative stomatal conductivity that in turn determines Ci and FCi. We use a simple bucket soil water model (Fig. 1). The dimension of the soil volume from which the tree is able to extract water is determined by the product of

This article is protected by copyright. All rights reserved.

Accepted Article

the crown area (Area in m2) and the rooting depth (D in m). The relative stomatal conductance (Fw) is reduced when the water in the soil gets below a certain critical level Hc. (Goudriaan & Van Laar, 1994; Pepper et al., 2008)

(7)

where Ha is the actual relative soil moisture content (m3 water/m3 soil), Hw is the relative soil moisture content at wilting point (m3 water/m3 soil), and Hc is the critical relative soil moisture content (m3 water m-3soil) below which the stomata are starting to close and Fw becomes smaller than one. From Fw and the assimilation without water stress, we calculate the absolute stomatal conductance Gs and Ci. Ci determines the assimilation (equation 5,6) and Gs determines the plants transpiration assuming a stomatal conductance of water on a molar basis that is 1.6 times larger than that of CO2 (Sterck & Schieving, 2007). The (air) temperature T (oC) effect on C3 photosynthesis is ruled by the equation.

(8)

This temperature function fitted temperature response curves of (Leakey et al., 2003, Yamori et al., 2006) very well. It has a maximum value of one at 25oC and photosynthetic rate is halved at 10 oC and at 40 oC. It also resembles C3 photosynthesis- temperature relations as published by Goudriaan & Van Laar (1994) and Kirschbaum (1999). According to the pipe model assumption optimal organ sizes were derived from the optimal LAI, the LAI where the canopy production is maximal (see ESM-Model description, eq 7), This article is protected by copyright. All rights reserved.

Accepted Article

and the optimal reserve pool was 20% of the sapwood biomass (VanNieuwstadt, 2002; Veneklaas & den Ouden, 2005). In the model leaves, fine roots and reserves have a higher priority than sapwood (Waring & Pitman, 1985; Weinstein et al., 1991). So, assimilates are allocated to sapwood after leaves, fine roots and reserves have reached their optimal size. We used species-specific values for the parameterization of the model but when those values were unknown, we used published general C3 plant values (Table 2). For more information we provide a complete model description in the ESM-Model description.

Study area and climatic variables Toona ciliata is a ring porous, long-lived pioneer species. Adult trees of this species are typically >30 m tall and their crowns are fully exposed. The species is widely distributed over in South Asia, Southeast Asia and Australia. Our study site was the Huai Kha Khaeng (HKK) Wildlife Sanctuary located in Uthai Thani province in west-central Thailand, approximately 250 km north west of Bangkok (15o60’ N 99o20’ E) at an altitude between 490 and 650 m. The soil texture is sandy clay-loam, mean annual precipitation is 1473 mm (1993-2003), with a demarcated dry season from November until March (ESM Fig. S1; Bunyavejchewin et al., 2009) in which Toona loses its leaves for a short period. From this texture we esitmated the soil parameters Hw and Hc according to (Saxton et al., 1986). Mean annual temperature of the reserve is 23.5oC (1992-1998). In our simulations, we used long-term monthly maximum and minimum temperatures and rainfall of the Nakhon Sawan (NS) meteorological station (15o80’ N 100o20’ E; Table 2). The amount of photosynthetic active radiation (PAR) on top of the crown was calculated from the solar elevation as calculated from earth axis deviation, latitude, time and the solar constant (Goudriaan & Van Laar, 1994; Schippers et al., 2004). We used local CRU cloud cover values to estimate inter-annual variation in light, however, this did not contribute to explaining inter-annual variation in growth probably due to the low year to year

This article is protected by copyright. All rights reserved.

Accepted Article

light variation, so we used a constant atmospheric transmission instead of 0.47 according to (Kirk, 1994). This means that radiation did not contribute to the year to year variation but the daily and seasonal variation of light is able to interfere with other climatic variables in the model that do vary between years. Tree-ring samples of 26 canopy trees of Toona ciliata were obtained by manually extracting 2-3 5.15 mm diameter cores per tree, using Suunto (Vantaa, Finland) or Haglöf (Långsele, Sweden) increment borers. Based on allometric relationships for estimating tree dimensions (Poorter et al., 2006), we selected only trees that were expected to be canopy trees since the start of the studied time series of 30 years. Average diameter at breast height (1.3 m, dbh) of sampled trees was 50.4 cm and estimated average tree height was 29.5 m. The cores were glued to a wooden holder and cut perpendicular to the tree-ring boundaries with a large sliding microtome (WSL, Switzerland) in order to obtain an even surface. Subsequently every core was scanned with a high resolution flatbed scanner (Epson Expression 10000 XL) and analysed in the WinDENDRO program for tree-ring analysis (version 2009b; Regent Instruments Canada Inc.). Ring-width series were visually cross-dated within trees and then among trees (WinDENDRO). Afterwards they were checked for dating errors with the computer program COFECHA (Holmes, 1983).

Our model simulates the amount of sapwood connecting leaves with fine roots, including coarse roots and branches. To allow comparison of model results in terms of sapwood NPP with the empirical information on tree growth we transformed tree-ring width data (mm year1

) to sapwood biomass production (kg wood per m2 of crown area, per year) assuming a

closed canopy. To this end, we first estimated tree height (H) calibrating the height equation of (Poorter et al., 2006) to our trees yielding H = 80*(1-exp(-0.05743*dbh0.533)) (r2 = 0.74). Then we used a cylindrical approximation formula (Philip, 1994; Shinozaki et al., 1964),

This article is protected by copyright. All rights reserved.

Accepted Article

using H and dbh data, to estimate total wood biomass of (B kg): B = *(dbh/200)2*H*Wd . Here Wd is the wood density of Toona (480 kg m-3; Nock et al., 2009). Finally, we calculated the crown area (A in m2) by coupling the dbh-height relation and height crown area relation of (Poorter et al., 2006) yielding A = 1.3*dbh. So, we estimated the sapwood biomass of each tree and each year which allowed us to calculate the mean biomass increase of these tree during the monitoring period. This approach yielded mean productivity values (16.5 Mg ha-1 y-1) that closely match the simulations (15 Mg ha-1 y-1).

Parameterization, simulation and model fit We used species-specific values of plant traits for the parameterization of the model but when those values were unknown, we used published general C3 plant values (Table 1). We initialized the state variables of our model at equilibrium values for leaves (1.94 Mg ha-1), living sapwood (127 Mg ha-1), and fine root biomass (2.74 Mg ha-1). We obtained those values after running the model for 10 years, using average monthly temperature and average monthly precipitation between 1950 and 1960. Subsequently, we ran the model over 60 years starting in 1950 using measured climatic variables. We compared the simulated inter-annual production of sapwood over the last 30 years (1981-2010) with that based on tree-ring data. To adjust the model to our study site, we optimized the rooting depth D for water supply with respect to the fit between the tree ring data and model results. We therefore maximized the coefficient of determination (r2) by running a series of simulations varying D between 0-10 meters, keeping D constant in each simulation. The best simulation explained 68.5% of the variation of the tree ring chronology at a calibrated rooting depth of 4.28 m (see Fig.4f). This soil depth is close to published values for tropical monsoon forests (Schenk & Jackson, 2002a; Schenk & Jackson, 2002b) and was subsequently used in all our simulations.

This article is protected by copyright. All rights reserved.

Accepted Article

Sensitivity and regression analyses To analyse the sensitivity of the model under local conditions we ran 150 simulations in which we vary each climatic variable randomly between -5% and +5%. We used this model output to quantify the change in the simulated rate of relative stomatal conductance (Fw), photosynthesis, respiration and stem growth due to the change in each of the climate variables using model equations as a guideline to create pathways. To do this we performed a hierarchical multiple regression (see ESM hierarchical regression for details) using the simulated values to obtain standardized regression coefficients. From these regression coefficients, we characterized three pathways by which climatic variables influence stem growth, and the relative importance of these pathways. (1) The water availability pathway comprises all effects of climatic variables on the soil water availability and transpiration demand that in turn determines stomatal conductance and photosynthesis (Fw). (2) The respiration pathway comprises the temperature effects via the plant respiration per unit of biomass (Fr). (3) The photosynthesis pathway comprises the direct effect of all climatic variables on photosynthesis. These are temperature effects on the trees photosynthesis, and the direct effects of CO2. In addition, we performed a direct multiple linear regression that related the climatic variables and their interactions to simulated sapwood growth. This additional analysis was intended to evaluate the importance of potential interactive effects of climate variables on simulated tree growth. Finally, we quantified the contribution of each of the climatic variables in determining the observed temporal variation in stem growth of Toona trees. To this end, we ran simulations in which inter-annual variation in these climatic variables was switched on or off. For the climatic variable of interest, we fed IBTREE with 30-year meteorological data of this variable and set inter-annual variation of the other variables to zero by using their 30-year average values each year. We used averages of monthly data for these constant climatic variables in

This article is protected by copyright. All rights reserved.

Accepted Article

order to maintain seasonality in our simulations. We ran these simulations for single climatic variables, for several climatic variables combined and for all climatic variables. For each of the simulations, we correlated the simulated temporal variation in stem growth with the observed pattern of growth variation to assess the degree of similarity and the contribution of variation in driving stem growth patterns. A comparison of the correlation coefficients reveal the relative importance of inter-annual variation of minimum temperature, maximum temperature, CO2, and rainfall in driving the observed variation in Toona stem growth.

Results Sensitivity of simulated stem growth to climatic variation We used a hierarchical regression model to quantify the impacts of three pathways by which climatic variables influence tree physiology and finally stem growth. This model explained 99.9% of the inter-annual variation in simulated sapwood production. The photosynthesis, respiration and water availability pathways resulting from this analyses are presented in Fig. 2, along with the results of the sensitivity analysis. Our model assumptions for temperature impacts imply that a 1%-increase in maximum temperature and minimum temperature both increased tree respiration, but also that the impact of maximum temperature was twice as large as the impact of minimum temperature. Moreover, maximum temperature had a strong negative effect on relative stomatal conductance Fw, and this was only partially compensated by smaller, positive, effects of minimum temperature, rainfall, and CO2 on this variable. The maximum temperature also directly decreased photosynthesis whereas the direct effect of the minimum temperature was marginal. Both stomatal conductance and CO2 increased photosynthesis directly, with the effect of CO2 being stronger (Fig. 2). Overall, stem biomass growth was most sensitive to maximum day temperature because this had a strong negative effect on stem growth through negative impacts on photosynthesis and

This article is protected by copyright. All rights reserved.

Accepted Article

stomatal conductance (the water availability pathway) and through increased respiration (respiration pathway; Fig. 3a,b). CO2 was the second important factor and had a positive effect on stem growth, mainly through the photosynthesis pathway but it also increases the water availability. The third important climatic variable was rainfall, because of its positive impacts on stomatal conductance and, in turn, photosynthesis. For minimum temperature, the positive impacts on photosynthesis via the water pathway were almost completely offset by an increase in respiration yielding a near zero overall sensitivity (Fig. 3a,b).

The direct regression model, aimed to evaluate the interactive effects of climate variables on tree growth also explained 99.99 % of the simulated inter-annual variation in tree growth. While several interaction terms in this regression model were significant, their regression coefficients were very small compared to those of the single factors (ESM Fig. S2).

Effect of inter-annual variation in climatic variables on temporal variation in growth To evaluate the contribution of climatic variables in determining the temporal variation in tree growth, we ran a number of simulations in which inter-annual variation of climatic variables was switched on or off. The model simulation in which inter-annual variation in minimum temperature was switched on and 30-year averages were used for all other climatic variables yielded very little temporal variation and did not show a significant correlation with observed tree growth variation (Fig. 4a; Table 3). The temporal variation (i.e., rise) in atmospheric CO2 pressure also did not generate simulated tree growth patterns that resembled those observed (Fig. 4d; Table 3). In contrast, model simulations in which inter-annual variation in rainfall was included (while keeping all other climatic variables constant) produced growth patterns with strong annual variation in tree growth, that correlated with the observed variation in stem growth (r2 = 0.47; Fig. 4c). An even better match was found in simulations that included only

This article is protected by copyright. All rights reserved.

Accepted Article

inter-annual variation in maximum temperature (r2 = 0.53; Fig. 4b). Including inter-annual variation in both maximum temperature and rainfall produced simulated tree growth patterns that closely correlated with observed patterns (r2 = 0.69; Fig. 4e) and had very similar values for maximum and minimum growth values. A simulation in which inter-annual variation of all climatic variables was included, produced a growth pattern that matched equally well (Fig. 4f).

Discussion We analysed the effects of climatic variability on temporal variation in stem growth of tropical canopy trees using a combination of tree-ring analysis and a new tree growth model, IBTREE. As far as we know this is a new approach in studies on the impact of climatic variation on tree growth (Zuidema et al., 2012; Zuidema et al., 2013). The tree-ring study allowed us to obtain long-term series of growth variation of adult trees in response to natural climate variability without an expensive and time-consuming monitoring program. The simulation model allowed us to unravel the contrasting effects of temperature, rainfall, and CO2 concentration on tree stem growth and to understand the physiological pathways by which these climatic variables influence tree growth. We found that inter-annual fluctuations in maximum temperature and rainfall strongly drive temporal stem growth variation of Toona ciliata canopy trees at our study site, and that the temporal variation in minimum temperature and CO2 concentration hardly contributes to the observed growth variation in our study species. Our tree growth model was able to realistically simulate the strong inter-annual variation in tree growth, and was more successful in doing so than other models (Keenan et al., 2012). This comparatively strong temporal variation in simulated tree growth can be explained by the fact that we modelled fully exposed canopy trees of a single tree species compared to plot approaches in Keenan et al.

This article is protected by copyright. All rights reserved.

Accepted Article

(2012) and by the allocation rules in our model which give a low priority to stem growth, meaning that the stem growth is very sensitive to inter-annual climate fluctuations. Below, we discuss the effects of climatic variables on tree growth, comparing our results with climategrowth relations obtained in empirical studies.

Temperature effects Simulation results suggest that yearly tree growth was highly sensitive to maximum day temperature (Tmax) and that simulations including inter-annual variation in the maximum temperature yielded temporal growth patterns that closely matched observed stem growth patterns (Fig. 4). The important role of Tmax is caused by a strong positive effect of Tmax on maintenance respiration, a negative effect of Tmax on photosynthesis and a negative effect on tree water availability as maximum temperature importantly determines transpiration rates (Gharun et al., 2013). A strong and negative effect of Tmax was also apparent from direct negative correlations with observed annual tree growth of our study species (ESM Table S1) as well as sympatric species at the study site (Table 3; Vlam et al., 2014). Negative correlation of stem growth and Tmax were also obtained for a large number of species in a moist Malaysian rainforest site (Table 4; Feeley et al., 2007). In contrast, two studies at wetter sites (>2500 mm rain/year) in the Neotropics (Table 4), did not find evidence that diameter growth was influenced by Tmax (Clark et al., 2003; Feeley et al., 2007). While the number of studies assessing effects of Tmax on tree growth is very limited, the observation that effects of maximum temperature were found in somewhat drier tropical forests suggests a strong inhibiting effect of high temperatures through the water availability pathway, which is supported by our modelling results (Fig. 3a). In contrast with the strong effect of maximum temperature, variation in minimum temperature (Tmin) hardly affected simulated stem growth. This difference in effect is partly

This article is protected by copyright. All rights reserved.

Accepted Article

due to the fact that respiration rates roughly double with a temperature increase of 10⁰C, leading to much weaker effects of fluctuations in minimum temperature compared to those in maximum temperature (Atkinson et al., 2007; Ryan et al., 1994; Schippers & Kropff, 2001). Furthermore, the difference between Tmax and Tmin is thought to be an important determinant of the air water pressure during the day (Kirschbaum, 1999). Consequently, an increase in the minimum temperature at dawn (with maximum temperature remaining constant) results in a smaller air humidity differences between leaf and air during the day, and will result in transpiration decrease and a larger water availability. The direct effect of Tmin on photosynthesis was low because it reflects night and early morning temperatures where photosynthesis is absent or low. In contrast, Tmax better reflects the temperature during the day, which drives variation in photosynthesis. The low sensitivity of Tmin in our simulations contrasts with findings of several empirical studies, which showed that minimum day temperature is one of the most important factors driving the year-to-year variation in tropical tree stem growth (Table 4). This apparent discrepancy may result from a difference in approach. In empirical studies, climate sensitivity is assessed through correlations between climate variables and growth, while in our simulation approach, individual climatic variables can be switched on or off. Minimum and maximum temperatures are often correlated (e.g. ESM Table S1), which does not allow to disentangle their separate contributions to growth variation. This methodological constraint possibly explains why Tmin correlated negatively with diameter growth in empirical tests of our study species (ESM Table S1; Vlam et al. 2014) and why it did not affect growth in this simulation study. A second explanation for the discrepancy between our simulation results and the empirical results is related to the relative cool climate at our study site. Annual average temperature at our study site (23.5 ⁰C) is 1.54⁰C lower than that of sites for which a negative effect of minimum temperature on tree growth was found. Overall, these results suggest that subtle differences in temperature among

This article is protected by copyright. All rights reserved.

Accepted Article

forests can have important implications for the climate sensitivity of tree growth in these forests.

Rainfall effects In our simulations with historical climate, rainfall was the second most important factor driving inter-annual variation in stem growth. The rather low sensitivity of stem growth to variation in rainfall was largely offset by the large year-to-year variation in rainfall (CV = 20%), which was higher than that of any other climate variable (Table 2). Combined, the low sensitivity and high variability in rainfall yielded a simulated growth pattern that matched the observed growth patterns quite well (r2 = 0.47). This result is consistent with findings from tree-ring and plot-based studies, which in most cases reveal (strong) positive relations (Table 4, (Rozendaal & Zuidema, 2011). Interestingly, such positive relations are even found in forests with an annual precipitation over 4000 mm, probably as a high-rainfall year reduced water stress during the dry season (Clark et al., 2013).

CO2 effects As in many tree growth models (Cramer et al., 2001; Sitch et al., 2008) simulated tree growth in IBTREE was highly sensitive to changes in CO2 pressure: an increase of 1% caused a 1.29% increase in stem growth. This increase was largely caused by the direct effect of CO2 on photosynthesis, and only marginally by an effect of CO2 on stomatal conductance (Fig. 2, see also (Holtum & Winter, 2010). The direct effect on photosynthesis results from higher internal CO2 concentrations in the leaves in line with the assumption that the ratio of CO2 pressure in leaves and the atmosphere is constant (at 0.7) under conditions without water stress (Lloyd & Farquhar, 2008). This higher internal CO2 concentration increases photosynthesis rates (van de Sleen et al., 2014). In addition, increased CO2 pressure results in

This article is protected by copyright. All rights reserved.

Accepted Article

a higher water use efficiency (Cernusak et al., 2008), and thus increases photosynthesis rates under water stress. We expected a gradually increase of productivity over time due to the 14% CO2 increase in atmospheric CO2 concentration over the 30 years of study. Remarkably, despite the high sensitivity of stem growth to CO2, the 14% increase in CO2 did not lead to a gradual increase in growth in the observations, nor in the simulations (Fig. 4f). Apparently, the strong inter-annual fluctuations in rainfall and temperature governed the temporal pattern in growth rates to a much larger extent than the relatively small change in CO2 pressure over time (van de Sleen et al., 2014). A 12-year study on stand-level wood production in a Costa Rican wet forest site also showed that changes in CO2 pressure contributed little (but significantly) to explaining temporal patterns in tree growth (Clark et al., 2013). In terms of physiological mechanisms and pathways, it appears that the direct positive effect of increased CO2 pressure on photosynthesis and the indirect positive effect on water relations cannot be materialized if other drivers (rainfall, temperature) determine tree growth through the same or other pathways. Overall, we thus did not find evidence for a fertilizing effect of increased CO2 pressure on stem growth of Toona during the period of analyses.

Interactive effects of climatic variables on tree growth Interactions between climatic variables indicate that a change in one variable changes the response of tree growth to another climatic variable (Norby & Luo, 2004). (Luo et al., 2008) reported two-way interactive effects of climate warming with elevated [CO2] and precipitation, using Dynamic Global Vegetation Models. In our model we found two-way interactions between maximum temperature on the one hand and CO2, rainfall, and minimum temperature on the other. The regression coefficients of these interactions were, however, very low compared to single factors, indicating that the impacts of single factors were more

This article is protected by copyright. All rights reserved.

Accepted Article

important than possible interactive effects. Nevertheless, the magnitude of interactive effects may be larger if climatic variables are perturbed by more than 5% as we did in our analyses

Outlook In this study we showed that the inter-annual variation in stem growth of the Toona trees in a Thai forest is mainly driven by temporal variation in maximum day temperature and rainfall, but not to minimum day temperature and CO2. Moreover, we showed that the impacts of maximum day temperature resulted from higher tree respiration, reduced crown photosynthesis and reduced stomatal conductance in response to an increased water demand, whereas rainfall affected the stomatal conductance through its influence on the soil water availability. Our approach of combining tree ring records with a tree growth model covering a climate factor limits on a carbon balance with reserves, stomatal response, and biomass allocation principles helps to disentangle impacts of climatic variables on tree stem growth through different physiological pathways. As far as we know this is a new approach, which overcomes the problem of long census intervals in most plots studies and of co-varying climatic variables in correlative analyses of climate sensitivity of (annual) tree growth. Applying this approach to other tree species and forest types will improve insight into the effect of temporal variation in climatic drivers on tree growth and forest carbon dynamics in response to climatic variation.

Acknowledgements We thank the European Research Council (ERC, grant #242955) for providing financial support for this study. The National Research Council Thailand (NRCT) and the Department of National Parks (DNP) provided permission for accessing the Huai Kha Khaeng Wildlife Sanctuary and the collection of wood samples. Field staff of the Huai Kha Khaeng forest

This article is protected by copyright. All rights reserved.

Accepted Article

dynamics plot provided logistical support and assistance with species identification in the field. Sarayudh Bunyavejchewin and Somboon Kiratiprayoon provided invaluable support in arranging the conditions for our fieldwork. Finally, we thank three reviewers for their helpful comments.

Supporting Information legends ESM Fig. S1 Average rainfall, minimum and maximum temperature of the location. ESM Fig. S2 Standardized regression coefficients of climatic variables and their interactions. ESM Table S1 Pearson correlation coefficients among climatic variables and time. ESM Hierarchical regression analysis. ESM Model description Comprehensive model description .

This article is protected by copyright. All rights reserved.

Accepted Article

Table 1. Parameters used in the model Symbol Unit Photosynthesis k m2ground m-2 leaf LUE (kg CO2 Mj PAR-1 )

Description

Value Obtained 1 Source

Extinction coefficient of the canopy Light use efficiency at optimal conditions at Ca of 350ppm (C3 species)

0.61 0.007

V V

(Maass, 1995) (Schippers, 2001)

Km C ai Cc

ppm (-) ppm

The Michaelis Menten constant for the carboxylation Ratio between internal and external CO2 concentration Compensation point of the carboxylation process

404 0.7 37

V V V

(Lambers, 1998) (Grant, 2001) (Tuzet, 2003)

Respiration Rl Rs Rr R rs CVF l CVF s CVF r CVF r T ref T dbl

(kgCH2 0 kgDM-1 d-1) Respiration rate of leaves (kgCH2 0 kgDM-1 d-1) Respiration rate of total sapwood -1 -1 (kgCH2 0 kgDM d ) Respiration rate of root -1 -1 (kgCH2 0 kgDM d ) Respiration rate of reserves (kg DM/kg carbohydrates -1 ) Conversion factor of leaves (kg DM/kg carbohydrates -1 ) Conversion factor of sapwood (kg DM/kg carbohydrates -1 ) Conversion factor of root (kg DM/kg carbohydrates -1 ) Conversion factor of root (oC) Reference temperature at which respiration is measured (oC) Temperature increase at which the respiration doubles

0.032 0.0008 0.015 0.0008 0.68 0.63 0.68 1 25 10

E V V V V V V V V V

from leaf N and (Poorter, 2006) (Sterck, 2011) (Penningdevries, 1975) (Sterck, 2011) (Penningdevries, 1974) (Penningdevries, 1974) (Penningdevries, 1974) (Penningdevries, 1974) (Penningdevries, 1975) (Goudriaan, 1994)

(m2 kg-1 ) (m2 kg-1 )

Specific leaf area Ratio between sapwood cross sectional area and leaf weight Mass ratio between root and leaves Mass ratio between reserves and sapwood

13.6 0.0054 1.4 0.2

S F E V

(Ares, 2000) estimated from field values and (Poorter, 2006 ) (Malhi, 2011;Sanz-Perez, 2009) (Veneklaas, 2005; VanNieuwstadt, 2002)

Wood density Scaling parameter of height-DBH relation Scaling parameter of height-DBH relation Scaling parameter of crown area-DBH relation

450 2.71 0.564 1.3

S F F E

(Nock CA 2009) estimated from field values and (Poorter, 2006) estimated from field values and (Poorter, 2006) estimated from (Poorter 2006)

Leaf turnover Sapwood turnover Root turnover

1.04 0.08 0.59

F F V

from deciduous period of Toona measured from tree ring records of Toona

Relative soil moisture content at wilting point Critical relative soil moisture content below which stomata are closed The water content of the soil at field capacity Ratio between water diffusivity and CO2 diffusivity molar based

0.15 0.23 0.32 1.6

E E E V

(Bunyavejchewin, 2009; Saxton, 1986) (Bunyavejchewin, 2009; Saxton, 1986) (Bunyavejchewin, 2009; Saxton, 1986) (Sterck, 2011)

Architecture SLA Bs Br B r/s

(-) (-)

WD (kg m-3 ) a (-) b (-) c (-) Turnover of organs year Ml year Ms year Mr Water relations (mm mm-1) Hw (mm mm-1) Hc (mm mm-1) H fc (-) R H2O,CO2

(Finer, 2011)

F= measured in the field locally, C= calibrated value, V= value from reference, E= estimated value from reference

Table 2 Climatic variables and the average sapwood productivity over the period (1981-2010) as used in the analysis. mean minimum maximum stdev. CV (%) Source Description Abbreviation Unit o C 29.19 28.33 30.12 0.473 1.6 Nakhon Sawan (NS) meteorological station Tmax Mean maximum daily temperature over a yea o C 18.48 17.8 19.5 0.413 2.2 Nakhon Sawan (NS) meteorological station Mean minimum daily temperature over a yea Tmin -1 mm y Annual rainfall Rain 1487 777 2085 298 20.0 Nakhon Sawan (NS) meteorological station Carbon dioxide concentration in the air CO2 ppm 363 340 388 14.4 4.0 Mauna Loa Observatory Estimated sapwood productivity of Toona

Prod

-1 -1

Mg ha y

16.6

4.4

31.3

7.6

45.8 Tree-ring sequence of 26 canopy trees

Table 3 The effect of introducing year to year variation in a growth factor on the fit of IBTREE to data. 1 1 1 1 2 Tmin Tmax Rain CO2 r F value P value 1 0 0 0 -0.004 0.008 0.92 0 1 0 0 0.53 31.31 1.00E-05 0 0 1 0 0.47 24.7 1.00E-04 0 0 0 1 -0.0002 0 1 0 1 1 0 0.6962 64.92 1.00E-07 1 1 1 0 0.686 61.28 1.00E-06 1 1 1 1 0.682 60.08 1.00E-06 1 measured variation is introduced, 0 = mean values are used, abbreviations according to table 1.

This article is protected by copyright. All rights reserved.

Accepted Article

Table 4 Overview on published effects of temporal climatic variation on stem growth in tropical forests. Country Data source Analysis Latitude Annual Mean Rainfall Mean Minimum Maximum CO2 rainfall temp. temp. temp. temp. Bolivia Tree rings Correlation 11 S 1690-1760 27 + Costa Rica Plots Regression 10 N 4537 25.1 + − ns + Congo Tree rings Correlation 5N 1180 24.6 + ns Panama, Tailand, Malaysia Plots Correlation 2N-15N 1473-3200 23.5-27.5 − Thailand, this site Tree rings Reg. & Cor. 16N 1473 23.5 + − − Ivory coast/ Benin Tree rings Correlation 8N 1150/1230 26.5-27 + Panama Plots Correlation 9N 2551 26 + − ns Malaysia Plots Correlation 3N 1788 27.5 ns − − Venezuela Tree rings Correlation 7N 1700 24.6 + + Brazil Tree rings Reg. & Cor. 10S 3000 22.9 Thailand Tree rings Sim. model 16N 1473 23.5 + ns − ns + = positive correlation, - = negative correlation, ns = no significant correlation found, empty = not tested

Source Brienen and Zuidema (2005) Clark, et al. (2013) Couralet et al. (2010) Dong et al. (2012) Vlam et al. (2014) Schongart et al. (2006) Feeley et al. (2007) Feeley et al. (2007) Worbes (1999) Dunisch et al. (2003) This study

Figure legends Fig. 1 Scheme of the IBTREE model describing the effect of climatic variables on tree biomass growth. Continuous arrows indicate matter flows; dashed arrows indicate information flows. [CO2]a = atmospheric CO2 concentration (ppm), [CO2]i = CO2 concentration in the leaf (ppm), [H2O]a = atmospheric water vapour pressure (Pa), [H2O]i = water vapour pressure in the leaf (Pa), Tmax = the maximum temperature of the day (oC), Tmin is the minimum temperature of the day (oC), Light = photosynthetic active radiation (Mj m-2), LAI = leaf area index (m2 leaf m-2 ground surface), W = weight of different organs (kg dry m2

). Wres= reserve weight (kg carbohydrates m-2), Temp.= temperature of the atmosphere oC.

Respiration* indicates the loss of reserves due to assimilate shortage when photosynthesis is smaller than the respiration. Fig. 2 Results of the hierarchical regression analysis on sensitivity data generated by the IBREE model parameterized for Toona ciliata and climate at the study site. Tmax = the maximum temperature of the day (oC), Tmin is the minimum temperature of the day (oC), Rainfall = annual rainfall (mm year-1), CO2 = the atmospheric CO2 concentration (ppm), Fw = is the water availability as perceived by the plant or relative stomatal conductivity. Photosynthesis = annual rate of photosynthesis (kg CH2O m-2 y-1; CH2O = carbohydrates),

This article is protected by copyright. All rights reserved.

Accepted Article

Respiration = annual respiration per unit biomass (kg CH2O m-2 y-1), Sapwood production = annual production of sapwood biomass (kg DM m-2 y-1; DM = dry matter). Three pathways are distinguished: (1) the water availability pathway that influences photosynthesis through stomatal conductance (solid arrows; blue), (2) the direct effect of various meteorological factors on photosynthesis (dashed arrows; green) and (3) the respiration pathway (coarsely dashed arrows; red). Numbers give the sensitivity (% change) of the variable to which the arrow points for produced by a 1%-change in the variable from which the arrow departs. Fig. 3 Summary of sensitivity analysis in which the effect of a 1% increase in climatic variables on stem growth of Toona ciliata was assessed. Results are presented for each of three pathways (a) and as net result of these pathways (b). Climatic variables: minimum temperature (Tmin), maximum temperature (Tmax), rainfall and atmospheric CO2 concentration ([CO2]a). Fig. 4 Matching simulated (bold line; red) and measured (fine line; blue) temporal variation in stem growth of Toona ciliata. Tree growth was simulated using annual variation of the climatic variable indicated for each panel, without any annual variation in all other climatic variables.

Atkin OK, Tjoelker MG (2003) Thermal acclimation and the dynamic response of plant respiration to temperature. Trends in Plant Science, 8, 343-351. Atkinson LJ, Hellicar MA, Fitter AH, Atkin OK (2007) Impact of temperature on the relationship between respiration and nitrogen concentration in roots: an analysis of scaling relationships, Q(10) values and thermal acclimation ratios. New Phytologist, 173, 110-120. Battipaglia G, Saurer M, Cherubini P, Calfapietra C, Mccarthy HR, Norby RJ, Cotrufo MF (2013) Elevated CO2 increases tree-level intrinsic water use efficiency: insights from carbon and oxygen isotope analyses in tree rings across three forest FACE sites. New Phytologist, 197, 544-554. Beer C, Reichstein M, Tomelleri E et al. (2010) Terrestrial Gross Carbon Dioxide Uptake: Global Distribution and Covariation with Climate. Science, 329, 834-838. Bonan GB (2008) Forests and climate change: Forcings, feedbacks, and the climate benefits of forests. Science, 320, 1444-1449. Brienen RJW, Zuidema PA (2005) Relating tree growth to rainfall in Bolivian rain forests: a test for six species using tree ring analysis. Oecologia, 146, 1-12. Bunyavejchewin S, Lafrankie JV, Baker PJ, Davies SJ, Ashton PS (2009) Forest Trees of Huai Khaeng Wildlife Sanctuary, Tailand: Data from the 50-Hectare Forest Dynamics Plot, Thailand, National Parks, Wildlife and Plant Conservation Department. Cernusak LA, Winter K, Aranda J, Turner BL (2008) Conifers, angiosperm trees, and lianas: Growth, wholeplant water and nitrogen use efficiency, and stable isotope composition (delta C-13 and delta O-18) of seedlings grown in a tropical environment. Plant Physiology, 148, 642-659.

This article is protected by copyright. All rights reserved.

Accepted Article

Clark DA (2004) Sources or sinks? The responses of tropical forests to current and future climate and atmospheric composition. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences, 359, 477-491. Clark DA, Clark DB, Oberbauer SF (2013) Field-quantified responses of tropical rainforest aboveground productivity to increasing CO2 and climatic stress, 1997–2009. Journal of Geophysical Research: Biogeoscience, 118, 1-12. Clark DA, Piper SC, Keeling CD, Clark DB (2003) Tropical rain forest tree growth and atmospheric carbon dynamics linked to interannual temperature variation during 1984-2000. Proceedings of the National Academy of Sciences of the United States of America, 100, 5852-5857. Clark DB, Clark DA, Oberbauer SF (2010) Annual wood production in a tropical rain forest in NE Costa Rica linked to climatic variation but not to increasing CO2. Global Change Biology, 16, 747-759. Clark DB, Olivas PC, Oberbauer SF, Clark DA, Ryan MG (2008) First direct landscape-scale measurement of tropical rain forest Leaf Area Index, a key driver of global primary productivity. Ecology Letters, 11, 163-172. Comins HN, Mcmurtrie RE (1993) Long-term response of nutrient-limited forest to CO2 enrichment - Equilibrium behavior of plant-soil models. Ecological Applications, 3, 666-681. Couralet C, Sterck FJ, Sass-Klaassen U, Van Acker J, Beeckman H (2010) Species-Specific Growth Responses to Climate Variations in Understory Trees of a Central African Rain Forest. Biotropica, 42, 503-511. Cramer W, Bondeau A, Woodward FI et al. (2001) Global response of terrestrial ecosystem structure and function to CO2 and climate change: results from six dynamic global vegetation models. Global Change Biology, 7, 357-373. Dong SX, Davies SJ, Ashton PS et al. (2012) Variability in solar radiation and temperature explains observed patterns and trends in tree growth rates across four tropical forests. Proceedings of the Royal Society B-Biological Sciences, 279, 3923-3931. Dunisch O, Montoia VR, Bauch J (2003) Dendroecological investigations on Swietenia macrophylla King and Cedrela odorata L. (Meliaceae) in the central Amazon. Trees-Structure and Function, 17, 244-250. Edwards NT, Hanson PJ (1996) Stem respiration in a closed-canopy upland oak forest. Tree Physiology, 16, 433-439. Farquhar GD, Buckley TN, Miller JM (2002) Optimal stomatal control in relation to leaf area and nitrogen content. Silva Fennica, 36, 625-637. Feeley KJ, Wright SJ, Supardi MNN, Kassim AR, Davies SJ (2007) Decelerating growth in tropical forest trees. Ecology Letters, 10, 461-469. Gharun M, Turnbull TL, Adams MA (2013) Validation of canopy transpiration in a mixed-species foothill eucalypt forest using a soil-plant-atmosphere model. Journal of Hydrology, 492, 219-227. Goudriaan J, Van Laar HH (1994) Modelling potential crop growth processes, Dordrecht, Kluwer Academic Publishers. Grant RF, Goulden ML, Wofsy SC, Berry JA (2001) Carbon and energy exchange by a black spruce-moss ecosystem under changing climate: Testing the mathematical model ecosys with data from the BOREAS experiment. Journal of Geophysical Research-Atmospheres, 106, 33605-33621. Haxeltine A, Prentice IC (1996) A general model for the light-use efficiency of primary production. Functional Ecology, 10, 551-561. Hickler T, Smith B, Sykes MT, Davis MB, Sugita S, Walker K (2004) Using a generalized vegetation model to simulate vegetation dynamics in northeastern USA. Ecology, 85, 519-530. Holmes RL (1983) Computer-assisted quality control in tree-ring dating and measurement. Tree-ring Bulletin, 43, 69-78. Holtum JaM, Winter K (2010) Elevated CO2 and forest vegetation: more a water issue than a carbon issue? Functional Plant Biology, 37, 694-702. Keenan TF, Baker I, Barr A et al. (2012) Terrestrial biosphere model performance for inter-annual variability of land-atmosphere CO2 exchange. Global Change Biology, 18, 1971-1987. Kirk JTO (1994) Light an photosynthesis in aquatic systems, Cambridge, Cambridge University Press. Kirschbaum MUF (1999) CenW, a forest growth model with linked carbon, energy, nutrient and water cycles. Ecological Modelling, 118, 17-59. Koerner C (2009) Responses of Humid Tropical Trees to Rising CO(2). In: Annual Review of Ecology Evolution and Systematics. pp Page. Leakey ADB, Press MC, Scholes JD (2003) High-temperature inhibition of photosynthesis is greater under sunflecks than uniform irradiance in a tropical rain forest tree seedling. Plant Cell and Environment, 26, 1681-1690. Lewis SL, Brando PM, Phillips OL, Van Der Heijden GMF, Nepstad D (2011) The 2010 Amazon Drought. Science, 331, 554-554. Lewis SL, Lopez-Gonzalez G, Sonke B et al. (2009) Increasing carbon storage in intact African tropical forests. Nature, 457, 1003-U1003. Liu N, Dang QL, Parker WH (2006) Genetic variation of Populus tremuloides in ecophysiological responses to CO2 elevation. Canadian Journal of Botany-Revue Canadienne De Botanique, 84, 294-302. Lloyd J, Farquhar GD (2008) Effects of rising temperatures and CO2 on the physiology of tropical forest trees. Philosophical Transactions of the Royal Society B-Biological Sciences, 363, 1811-1817. Luo YQ, Gerten D, Le Maire G et al. (2008) Modeled interactive effects of precipitation, temperature, and CO2 on ecosystem carbon and water dynamics in different climatic zones. Global Change Biology, 14, 1986-1999. Luyssaert S, Inglima I, Jung M et al. (2007) CO2 balance of boreal, temperate, and tropical forests derived from a global database. Global Change Biology, 13, 2509-2537.

This article is protected by copyright. All rights reserved.

Accepted Article

Makela A, Pulkkinen M, Kolari P et al. (2008) Developing an empirical model of stand GPP with the LUE approach: analysis of eddy covariance data at five contrasting conifer sites in Europe. Global Change Biology, 14, 92-108. Meir P, Grace J (2002) Scaling relationships for woody tissue respiration in two tropical rain forests. Plant Cell and Environment, 25, 963-973. Monteith JL (1977) CLIMATE AND EFFICIENCY OF CROP PRODUCTION IN BRITAIN. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences, 281, 277-294. Nock CA, Geihofer D, Grabner M, Baker PJ, Bunyavejchewin S, Hietz P (2009) Wood density and its radial variation in six canopy tree species differing in shade-tolerance in western Thailand. Annals of Botany, 104, 297-306. Norby RJ, Luo YQ (2004) Evaluating ecosystem responses to rising atmospheric CO(2) and global warming in a multi-factor world. New Phytologist, 162, 281-293. Pan Y, Birdsey RA, Fang J et al. (2011) A Large and Persistent Carbon Sink in the World's Forests. Science, 333, 988-993. Penningdevries FWT (1975) Cost of maintenance processes in plant cells. Annals of Botany, 39, 77-92. Pepper DA, Mcmurtrie RE, Medlyn BE, Keith H, Eamus D (2008) Mechanisms linking plant productivity and water status for a temperate Eucalyptus forest flux site: analysis over wet and dry years with a simple model. Functional Plant Biology, 35, 493-508. Philip MS (1994) Measuring Trees and Forests (2nd ed), Wallingford (UK), CABI Publishing. Phillips OL, Aragao LEOC, Lewis SL et al. (2009) Drought Sensitivity of the Amazon Rainforest. Science, 323, 1344-1347. Phillips OL, Lewis SL (2014) Evaluating the tropical forest carbon sink. Global Change Biology, 20, 2039-2041. Phillips OL, Lewis SL, Baker TR, Chao K-J, Higuchi N (2008) The changing Amazon forest. Philosophical Transactions of the Royal Society B-Biological Sciences, 363, 1819-1827. Phillips OL, Van Der Heijden G, Lewis SL et al. (2010) Drought-mortality relationships for tropical forests. New Phytologist, 187, 631-646. Poorter L, Bongers L, Bongers F (2006) Architecture of 54 moist-forest tree species: Traits, trade-offs, and functional groups. Ecology, 87, 1289-1301. Rozendaal DMA, Zuidema PA (2011) Dendroecology in the tropics: a review. Trees-Structure and Function, 25, 3-16. Ryan MG, Hubbard RM, Clark DA, Sanford RL (1994) Woody-tissue respiration for Simarouba amara and Minquartia guianensis, 2 tropical wet forest trees with different growth habits. Oecologia, 100, 213220. Saxton KE, Rawls WJ, Romberger JS, Papendick RI (1986) Estimated generalized Soil-water Characteristics from texture. Soil Sci. Soc. Am. J., 50, 1031-1036 Schenk HJ, Jackson RB (2002a) The global biogeography of roots. Ecological Monographs, 72, 311-328. Schenk HJ, Jackson RB (2002b) Rooting depths, lateral root spreads and below-ground/above-ground allometries of plants in water-limited ecosystems. Journal of Ecology, 90, 480-494. Schippers P, Joenje W (2002) Modelling the effect of fertilizer, mowing, disturbance and width on the biodiversity of plant communities of field boundaries. Agriculture, Ecosystems and Environment., 93, 351-365. Schippers P, Kropff MJ (2001) Competition for light and nitrogen among grassland species: a simulation analysis. Functional Ecology, 15, 155-164. Schippers P, Vermaat JE, De Klein J, Mooij WM (2004) The effect of atmospheric carbon dioxide elevation on plant growth in freshwater ecosystems. Ecosystems, 7, 63-74. Schongart J, Orthmann B, Hennenberg KJ, Porembski S, Worbes M (2006) Climate-growth relationships of tropical tree species in West Africa and their potential for climate reconstruction. Global Change Biology, 12, 1139-1150. Shinozaki K, Yoda K, Hozumi K, Kira T (1964) A quantitative analysis of plant form. The pipe model theory. II. Further evidence of the theory and its application in forest ecology. Jpn J. Ecol., 14, 133-139. Sitch S, Huntingford C, Gedney N et al. (2008) Evaluation of the terrestrial carbon cycle, future plant geography and climate-carbon cycle feedbacks using five Dynamic Global Vegetation Models (DGVMs). Global Change Biology, 14, 2015-2039. Slot M, Rey-Sánchez C, Gerber S, J.W. L, Winter K, Kitajima K (2014) Thermal acclimation of leaf respiration of tropical trees and lianas: response to experimental canopy warming, and consequences for tropical forest carbon balance. Global Change Biology, 20, 2915-2926. Sterck F, Markesteijn L, Schieving F, Poorter L (2011) Functional traits determine trade-offs and niches in a tropical forest community. Proceedings of the National Academy of Sciences of the United States of America, 108, 20627-20632. Sterck F, Schieving F (2011) Modelling functional trait acclimation for trees of different height in a forest light gradient: emergent patterns driven by carbon gain maximization. Tree Physiology, 31, 1024-1037. Sterck FJ, Schieving F (2007) 3-D growth patterns of trees: Effects of carbon economy, meristem activity, and selection. Ecological Monographs, 77, 405-420. Tricker PJ, Trewin H, Kull O et al. (2005) Stomatal conductance and not stomatal density determines the longterm reduction in leaf transpiration of poplar in elevated CO2. Oecologia, 143, 652-660. Van De Sleen P, Groenendijk P, Vlam M et al. (2014) No growth stimulation of tropical trees by 150 years of CO2 fertilization but water-use effciency increased. Nature Geoscience, DOI: 10.1038/NGEO2313. Vannieuwstadt MGL (2002) Trail by fire - Postfire development of a tropical dipterocarp forest. Unpublished PHD University of Utrecht, Utrecht.

This article is protected by copyright. All rights reserved.

Accepted Article

Veneklaas EJ, Den Ouden F (2005) Dynamics of non-structural carbohydrates in two Ficus species after transfer to deep shade. Environmental and Experimental Botany, 54, 148-154. Verboom J, Schippers P, Cormont A, Sterk M, Vos CC, Opdam PFM (2010) Population dynamics under increasing environmental variability: implications of climate change for ecological network design criteria. Landscape Ecology, 25, 1289-1298. Vlam M, Baker PJ, Bunyavejchewin S, Zuidema PA (2014) Temperature and rainfall strongly drive temporal growth variation in Asian tropical forest trees. Oecologia, 174, 1449-1461. Waring RH, Pitman GB (1985) Modifying lodgepole pine stands to change susceptibility to mountain pine-beetle attack. Ecology, 66, 889-897. Weinstein DA, Beloin RM, Yanai RD (1991) Modelling changes in Red Spruce carbon balance and allocation in response to interacting ozone and nutrient stresses. Tree Physiology, 9, 127-146. Worbes M (1999) Annual growth rings, rainfall-dependent growth and long-term growth patterns of tropical trees from the Caparo Forest Reserve in Venezuela. Journal of Ecology, 87, 391-403. Wright SJ (2013) The carbon sink in intact tropical forests. Global Change Biology, 19, 337-339. Yamori W, Suzuki K, Noguchi K, Nakai M, Terashima I (2006) Effects of Rubisco kinetics and Rubisco activation state on the temperature dependence of the photosynthetic rate in spinach leaves from contrasting growth temperatures. Plant Cell and Environment, 29, 1659-1670. Zotz G, Pepin S, Korner C (2005) No down-regulation of leaf photosynthesis in mature forest trees after three years of exposure to elevated CO2. Plant Biology, 7, 369-374. Zuidema PA, Baker PJ, Groenendijk P, Schippers P, Van Der Sleen P, Vlam M, Sterck F (2013) Tropical forests and global change: filling knowledge gaps. Trends in Plant Science, 18, 418-424. Zuidema PA, Brienen RJW, Schongart J (2012) Tropical forest warming: looking backwards for more insights. Trends in Ecology & Evolution, 27, 193-194.

This article is protected by copyright. All rights reserved.

Accepted Article This article is protected by copyright. All rights reserved.

Accepted Article This article is protected by copyright. All rights reserved.

Accepted Article This article is protected by copyright. All rights reserved.

Accepted Article This article is protected by copyright. All rights reserved.