This is the accepted version of the article: Verryckt, Lore T.; Van Langenhove, Leandro; Ciais, Philippe; [et al.]. «Coping with branch excision when measuring leaf net photosynthetic rates in a lowland tropical forest». Biotropica, vol. 52, issue 4 (July 2020), p. 608-615. DOI 10.1111/btp.12774 This version is avaible at https://ddd.uab.cat/record/232206 under the terms of the license 1 LRH: Verryckt et al. 2 RRH: gas exchange measurements on excised branches 3 Coping with branch excision when measuring leaf net photosynthetic rates in a lowland 4 tropical forest 5 Lore T. Verryckt1, Leandro Van Langenhove1, Philippe Ciais2, Elodie A. Courtois3, Sara Vicca1, Josep 6 Peñuelas4, Clément Stahl5, Sabrina Coste6, David S. Ellsworth7, Juan M. Posada8, Michael Obersteiner9, 7 Jérôme Chave10, Ivan A. Janssens1 8 1 Department of Biology, University of Antwerp, Universiteitsplein 1, 2610 Wilrijk, Belgium; 2 Laboratoire des 9 Sciences du Climat et de l’Environnement, CEA-CNRS-UVSQ, Gif-sur-Yvette, 91191, France; 3 Laboratoire 10 Ecologie, évolution, interactions des systèmes amazoniens (LEEISA), Université de Guyane, CNRS, IFREMER, 11 97300 Cayenne, French Guiana; 4 CSIC, Global Ecology CREAF-CEAB-CSIC-UAB, Cerdanyola del Valles, 08193 12 Barcelona, Spain; 5 INRA, UMR Ecofog, AgroParisTech, CNRS, Cirad, Université des Antilles, Université de 13 Guyane, 97310 Kourou, France; 6 Université de Guyane, UMR Ecofog, AgroParisTech, CNRS, Cirad, INRA, 14 Université des Antilles, 97310 Kourou, France; 7 Hawkesbury Institute for the Environment, Western Sydney 15 University, Locked Bag 1797, Penrith, NSW 2751, Australia; 8 Programa de Biología, Universidad del Rosario, Cr. 16 24 No. 63C-69, 111221 Bogotá, D.C., Colombia; 9 International Institute for Applied Systems Analysis (IIASA), 17 Laxenburg, Austria; 10 UMR 5174 Laboratoire Evolution et Diversité Biologique, Université Paul Sabatier, CNRS, 18 Toulouse, France 19 Author for correspondence: 20 L. T. Verryckt 21 Email: lore.verryckt@uantwerpen.be 22 Received: ; Revised: Accepted: 1 23 ABSTRACT 24 Measuring leaf gas exchange from canopy leaves is fundamental for our understanding of 25 photosynthesis and for a realistic representation of carbon uptake in vegetation models. Since 26 canopy leaves are often difficult to reach, especially in tropical forests with emergent trees up to 27 60 meters at remote places, canopy access techniques such as canopy cranes or towers have 28 facilitated photosynthetic measurements. These structures are expensive and therefore not very 29 common. As an alternative, branches are often cut to enable leaf gas exchange measurements. 30 The effect of branch excision on leaf gas exchange rates should be minimised and quantified to 31 evaluate possible bias. We compared light-saturated leaf net photosynthetic rates measured on 32 excised and intact branches. We selected branches positioned at three canopy positions, 33 estimated relative to the top of the canopy: upper sunlit foliage, middle canopy foliage, and 34 lower canopy foliage. We studied the variation of the effects of branch excision and transport 35 amongst branches at these different heights in the canopy. After excision and transport, light- 36 saturated leaf net photosynthetic rates were close to zero for most leaves due to stomatal closure. 37 However, when the branch had acclimated to its new environmental conditions – which took on 38 average 20 minutes –light-saturated leaf net photosynthetic rates did not significantly differ 39 between the excised and intact branches. We therefore conclude that branch excision does not 40 affect the measurement of light-saturated leaf net photosynthesis, provided that the branch is 41 recut under water and is allowed sufficient time to acclimate to its new environmental 42 conditions. 43 44 Keywords: branch cutting, canopy physiology, French Guiana, gas exchange, photosynthesis, 45 rainforest, stomatal conductance 2 46 1. INTRODUCTION 47 The measurement of leaf gas exchange is important for our understanding of photosynthesis (e.g. 48 Dang, Margolis, Coyea, Sy & Collatz, 1997, Schaberg, Shane, Cali, Donnelly & Strimbeck, 49 1998, Domingues et al., 2010, Cavaleri, Reed, Smith & Wood, 2015, Heberling, Kichey, Decocq 50 & Fridley, 2016), to understand processes of plant evolution and ecological strategies (Wright et 51 al., 2004), and to improve the representation of carbon uptake in vegetation models (Rezende et 52 al., 2016). In areas such as Amazonia, where the number of plants exceeds 14,000 about half 53 being trees (Cardoso et al., 2017), the challenge of measuring leaf gas exchange across the region 54 is enormous. Even at accessible field sites, reaching canopy leaves is difficult, because they are 55 situated between 30 and 60 m above the ground (Perry, 1978). 56 Canopy cranes or towers have facilitated the in-situ measurement of leaf gas exchange in 57 forest canopies, but these structures are rare (Basset, Horlyck & Wright, 2003, Nakamura et al., 58 2017). Therefore, branch excision is often necessary prior to leaf gas exchange measurement. 59 The excision of branches using a single-rope climbing technique (Perry, 1978), pole pruners, or 60 shotguns (e.g. Messier, Mcgill & Lechowicz, 2010) are different methods that can give access to 61 leaves in the canopy. The effect of excision of branches on leaf gas exchange rates should 62 therefore be minimised and quantified to evaluate possible bias. 63 Tropical studies measuring gas exchange on excised branches have reported different 64 effects of excision on the leaf net photosynthetic rate (An). Some species are not affected 65 (Santiago & Mulkey, 2003, Rowland et al.,2015), whereas for other species An decreases within 66 a few minutes to several days (Santiago & Mulkey, 2003). The reported effects can be caused by 67 alterations in the xylem stream after excision, influencing stomatal response (Sperry, Alder & 68 Easttlack, 1993, Williamson & Milburn, 1995). Therefore, it has been advised that prior to 3 69 measurement, excised branches should be recut under water, to remove embolised xylem vessels 70 and to restore hydraulic conductivity for maintaining an adequate supply of water to the leaves. 71 Santiago and Mulkey (2003) and Rowland et al. (2015) studied upper-canopy, sunlit leaves. 72 However, several leaf physiological traits of tropical trees, such as Asat and gs, change with 73 height (Kenzo et al., 2015). The effect of excision might thus also vary with tree height. To the 74 best of our knowledge, vertical differences in the excision effect have not yet been investigated. 75 In this study, we therefore assessed whether the effect of excision varied with depth in the 76 canopy. 77 Measurement can be conducted at a nearby campsite (Rowland et al., 2015), or in a 78 laboratory further away (Ellsworth & Reich, 1993, Dang et al.,1997), implying different timings 79 between excision and measurement and different environmental conditions. The excised 80 branches are thus treated in different ways prior to leaf gas exchange measurements, even though 81 the studies cited tried to minimise the time between excision and the actual measurement and to 82 maintain the original light conditions for the excised branch. 83 We assessed the effects of branch excision and transport on leaf net photosynthetic rate at 84 light saturation (Asat) and the variation of these effects amongst branches at different heights in 85 the canopy. We hypothesised that the effects of excision and transport would be smaller in the 86 canopy, because these branches are acclimated to darker conditions than the branches at the top 87 of the canopy. We assumed that they would be less stressed after excision and transport through 88 a dark forest. We therefore tested for a faster recovery of Asat for branches lower in the canopy. 89 2. METHODS 90 2.1 Study site 4 91 This study was conducted at the Nouragues Research Station (4°02’ N, 52°41’ W), an old- 92 growth, tropical lowland forest site in French Guiana, South America. The climate is tropical 93 humid, with a dry season lasting from August to November and a short dry period in March. 94 Mean annual temperature is 26.3°C and mean annual rainfall is about 3000 mm (Bongers, 95 Charles-Dominique, Forget & Théry, 2001). 96 2.2 Measurement of leaf gas exchange 97 We selected seven trees (Table 1) belonging to four dominant families at this field site: 98 Fabaceae, Lecythidaceae, Malvaceae, and Burseraceae. Asat (measured at 1300 µmol/m2/s) and 99 stomatal conductance (gs; µmol/m2/s) were measured on mature leaves using an infrared gas 100 analyser (LI-6400XT, LI-COR, Lincoln, NE, USA). The measurements were carried out at three 101 canopy positions, estimated relative to the top of the canopy: upper sunlit foliage, middle canopy 102 foliage, and lower canopy foliage. We measured Asat at each position for at least three leaves on 103 an intact branch attached to the tree, with the assistance of a tree climber. This 2 m-long branch 104 was subsequently excised, lowered gently from the canopy to a person who then immediately 105 recut the branch under water to reconstitute the water supply to the leaves, and transported 106 (within about ten minutes) to a nearby camp site in an opaque plastic bag to avoid further water 107 loss due to transpiration, where at least two leaves were measured on the excised branch. The 108 base of the branch always remained under water during transport. The branch was again recut 109 under water at the camp site to prevent from potential disruption of hydraulic conductivity that 110 may have arisen during transport. An at light saturation from these excised branches was 111 continuously logged. An was then defined as Asat when An remained stable, i.e. did not vary more 112 than five percent for at least two minutes. Relative humidity was maintained as near ambient as 113 possible (70-85%), and the rate of air flow was 500 mL/min during these measurements, which 5 114 were carried out between 0930 h and 1500 h. The CO2 level in the chamber was constant at 400 115 ppm, and the chamber temperature was controlled by maintaining the block temperature at 30 ± 116 0.2°C. 117 We tested an additional tree species at the camp site, Eperua falcata, with sunlit leaves 118 reachable from the ground without climbing. We tested the effect on Asat of transporting the 119 excised branch in an opaque plastic bag. We continuously measured An at light saturation before 120 excision and continued measuring it after excision without transporting the branch in a plastic 121 bag. We also excised a branch, after measuring Asat, and transported it similarly to the other 122 excised branches and again continuously measured An at light saturation to compare both 123 methods. These measurements were carried out between 1000 h and 1130 h. 124 2.3 Data analyses 125 The effect of excision on Asat was tested with a paired t-test, comparing pre- and post-cutting 126 measurements for each leaf. We also tested if the slope of the linear regression between Asat for 127 non-excised and excised branches differed from 1. This analysis was carried out for all data, but 128 also for upper, middle and lower canopy foliage separately. We analysed our Asat data using a 129 two-way ANOVA, with excision (pre- or post-cutting), tree ID and canopy position (upper, 130 middle, lower foliage) as fixed factors. We specifically explored the two-way interactions 131 between these factors to assess the (interaction) effects of tree ID and canopy position on 132 excision. We used P < 0.05 to identify significant differences. Additionally we performed a 133 bootstrap analysis (10,000 replicates) for this two-way ANOVA for which we allowed 134 resampling within the same canopy position (upper, middle, lower foliage). We assumed that the 135 interaction effects would not be significant if the bias-corrected and accelerated (BCa) 95% 6 136 confidence intervals included zero. All analyses were performed in R 3.3.3 (R Core Team, 2017) 137 using the ggplot2 package (Wickham, 2009) for visualising the data. 138 3. RESULTS 139 3.1 Effect of excision 140 In general, Asat did not differ significantly between the excised and non-excised branches (paired 141 t-test: P = 0.10), and the slope of the linear regression did not differ significantly from 1 (P = 142 0.21; Figure 1). When analysing the upper, middle and lower canopy foliage separately, the 143 intercept and the slope of the linear regression did not differ significantly from 0 (Pupper = 0.79, 144 Pmiddle = 0.64, Plower = 0.54) and 1 (Pupper = 0.79, Pmiddle = 0.86, Plower = 0.19), respectively. The 145 two-way ANOVA showed no interaction between excision and tree (P = 0.59) or canopy 146 position (P = 0.55). The interaction between tree ID and canopy position was significant (P = 147 0.03), with different responses of Asat between tree IDs and canopy positions. Excision was not a 148 significant factor in the two-way ANOVA (P = 0.49), indicating no differences in Asat before and 149 after excision. Individual differences between pre- and post-cutting measurements of Asat are 150 shown in Figure S1. The bootstrap analysis showed a tendency of an excision effect for tree 7, 151 with BCa 95% confidence intervals approaching zero [0.144, 4.317], whereas it did not show any 152 effect for the other tree species, nor of canopy position, since these BCa 95% confidence 153 intervals all included zero. Intrinsic water-use efficiency (iWUE) did not differ significantly 154 between the excised and non-excised branches, with the slope of the linear regression not 155 differing significantly from 1 (P = 0.66; Figure S2). 156 3.2 Effect of darkness in E. falcata: to bag or not to bag? 7 157 Branch excision of the E. falcata tree in the camp had no effect on the continuously measured An 158 at light saturation (Figure 2a). When the excised branch was first transported in a plastic bag and 159 An at light saturation was measured 10 min after recutting the branch under water, An at light 160 saturation was more than 50 percent lower than the original value (Figure 2b). An at light 161 saturation increased again after about 30 minutes of stabilisation, and Asat was similar to the 162 measurement on the intact branch in the tree. 163 3.3 Stabilisation time 164 An at light saturation after branch excision and transport to the camp was near zero (<1 165 µmol/m2/s) for 65 percent of the leaves and decreased by 30-80 percent for the remainder of the 166 leaves. An increased over time after the leaf was placed in the leaf chamber and exposed to 1300 167 µmol/m2/s photosynthetically active radiation (Figure 3a), and Asat did not differ significantly 168 from Asat for the leaf measured on the intact branch in the tree (Figure 1). The time to 169 stabilisation across all trees averaged 22 min, but ranged between 6 and 44 min (Figure 3b, 170 Figure S3, Figure S4). The stabilisation time was longer for tree 7 than the other trees. The 171 stabilisation time was longer for upper canopy foliage (27 ± 9 min) compared to middle (22 ± 10 172 min) and lower (19 ± 7 min) canopy foliage (Figure 3c). 173 4. DISCUSSION 174 We quantified the effects of branch excision on Asat as differences between Asat measured on an 175 intact branch in the tree and Asat measured on the same leaf after branch excision under water. An 176 at light saturation decreased soon after the branch was excised and transported to the camp site 177 (Figure 2b, Figure S5a). This immediate decrease in An at light saturation was caused by stomatal 178 closure (Figure 2b, Figure S5b), but the stomata reopened on average 22 min after the branch 8 179 was re-exposed to light (Figure 3b, c). An at light saturation, however, did not decrease in the 180 excised E. falcata branch kept in the light (Figure 2a), suggesting that the stomatal closure was 181 due to light limitation rather than dehydration caused by excision. This requires confirmation by 182 testing more tree species, because this was an unreplicated experiment. We only tested one 183 species that we kept at constant light while excising the branch. 184 The cut branches were transported from the canopy to the forest floor, where less than 185 two percent of the incoming light at the top of the canopy is typically available (Yoda, 1974, 186 Théry, 2001, Montgomery & Chazdon, 2002). We then cut the excised branch under water and 187 transported it in an opaque bag to the camp site, where we measured leaf gas exchange. Light is a 188 major determinant of stomatal conductance (Motzer, Munz, Küppers, Schmitt & Anhuf, 2005), 189 so both the dark conditions at the forest floor and the opaque bag likely induced stomatal closure. 190 Stomata respond quickly to changing light conditions (Roelfsema & Hedrich, 2005, Devireddy, 191 Zandalinas, Gómez-Cadenas, Blumwald & Mittler, 2018), allowing the plant to regulate its 192 water-use efficiency. Stomata of the leaves can thus be expected to close after excision due to 193 light limitation to maintain their water balance and to reopen when the leaves are re-exposed to 194 light in the leaf chamber. The recovery of stomatal conductance and An was faster for leaves 195 lower in the canopy and thus typically exposed to less light (Figure 3c), supporting our 196 hypothesis that branches lower in the canopy would recover faster after excision. The differences 197 in recovery time along a height gradient in the canopy might also be explained by the differences 198 in leaf water potential. Upper-canopy leaves typically have a lower water potential and the 199 branches at the top of the canopy might thus experience more stress when excised compared to 200 middle- and lower-canopy branches (Koch, Sillett, Jennings & Davis, 2004, Olson et al., 2018). 9 201 Although we did not measure the leaf water potentials, our data on stomatal conductance and 202 intrinsic water use efficiency suggest that this did not play a major role in our study. 203 Asat did not differ significantly before and after excision at the camp site after acclimating 204 to the new light conditions in the leaf chamber. The immediate decrease in An at light saturation 205 was restored after, on average, 22 min, which is in contrast to other studies where excision had 206 no effect or a delayed decrease in Asat was shown (Lange, Führer & Gebel, 1986, Meng & Arp, 207 1992, Ellsworth & Liu, 1994, Dang et al., 1997, Ambrose, Sillett & Dawson, 2009, Dusenge et 208 al., 2015, Rowland et al., 2015). Measurements of Asat on sugar maple leaves, immediately after 209 excision and rehydration of the branch, were always within five percent of the rate measured on 210 intact branches (Ellsworth & Liu, 1994). Excision did not affect Asat in Pinus banksiana or Picea 211 mariana, either immediately or during the next 14 h (Dang et al., 1997). Rowland et al. (2015) 212 reported that excision did not affect photosynthesis in tropical trees. Our branches were allowed 213 to stabilise in a canopy opening in full ambient irradiance for at least 30 min after recutting 214 underwater prior to the measurements. Dusenge et al. (2015) found no difference in Vcmax, the 215 maximum carboxylation rate, and only a small negative effect on Jmax, the maximum electron 216 transport rate, for tropical trees ten minutes before and after excision. An in twigs of Picea abies 217 was constant for a minimum of 15 min after excision (Lange et al.,1986), and net photosynthesis 218 of Picea rubens twigs did not vary significantly within about six minutes after detachment 219 (Meng & Arp, 1992). Asat did not decrease after excision in Sequoia sempervirens and S. 220 giganteum in California, USA (Ambrose et al., 2009). The leaves for all these studies were 221 exposed to full sunlight prior to measuring photosynthesis or were kept in the chamber under 222 saturating light conditions, in contrast to our study. 10 223 Santiago and Mulkey (2003) reported that some species retained more than 85 percent of 224 the photosynthetic rate for intact branches (i.e. pre-excision) 60 min after excision, whereas the 225 rates for other species were less than 40 percent of the rates for intact branches within three 226 minutes. The effect of excision can vary among species, and was shown to depend on the 227 production of latex or resin since these exudates might clog the xylem, reduce sap flow, and 228 thereby reduce foliar gas exchange (Santiago & Mulkey, 2003). Excision did not generally affect 229 Asat in all seven of our species, although Asat decreased significantly after excision for sunlit 230 upper-canopy foliage for tree 2 (Tetragastris altissima) and middle-canopy foliage for trees 3 231 (Eschweilera coriacea) and 7 (E. falcata) (Figure S1). Asat may have decreased because the time 232 needed for the leaves to acclimate to the settings of light and humidity at the camp was not long 233 enough to reach Asat. For tree 2 (Tetragastris altissima), belonging to the Burseraceae family that 234 is known to produce latex or resin (Plowden, Uhl & De Assis Oliveira, 2004, Weeks, Daly & 235 Simpson, 2005), the decrease in Asat after excision for sunlit, upper-canopy leaves might have 236 been influenced by the resin produced by this species, although we did not observe any resin 237 production. No decrease in Asat after excision was observed for middle and lower canopy leaves 238 of this tree. Asat measured in the camp on the excised branch was larger than Asat measured on the 239 intact branches for some trees (Figure S1), perhaps because xylem water potentials were higher 240 in the rehydrated excised branches than the intact branches (Comstock & Ehleringer, 1984). 241 Since there was no significant decrease in Asat, we assume that there was no effect of excision on 242 leaf photosynthesis, although there is large variation among individually sampled leaves. 243 Additionally, given that we only sampled seven species, it is still recommended to test the effect 244 of excision before measuring leaf gas exchange for other species than those studied here or in 11 245 previous studies (e.g. Santiago & Mulkey, 2003, Ambrose et al., 2009, Dusenge et al., 2015, 246 Rowland et al., 2015). 247 We conclude that branch excision does not affect the measurement of Asat when the 248 branch is recut under water after excision and the measurement is carried out for a sufficiently 249 long time (at least 20 minutes) after the branch has acclimated to its new environmental 250 conditions for reopening the stomata. Light is necessary to reopen the stomata when the branch is 251 transported in dark conditions, either on the forest floor or inside an opaque bag. This process 252 could be facilitated by transporting the excised branch in a bag containing LED lights, or the 253 branch could be exposed to sunlight before measurement. 254 ACKNOWLEDGEMENTS 255 L. T. Verryckt is funded by a PhD fellowship from the Research Foundation Flanders (FWO). I. 256 A. Janssens acknowledges support from the European Research Council Synergy Grant; ERC- 257 2013-SyG-610028 IMBALANCE-P. We thank the staff of the Nouragues Natural Reserve and 258 The Nouragues Ecological Research Station, supported by USR mixte LEEISA (CNRS; 259 Cayenne, French Guiana). The research station has benefited from financial support by French 260 Investissement d’Avenir programs managed by the ANR (AnaEE-France ANR-11-INBS-0001; 261 Labex CEBA ANR-10-LABX-25-01). 262 We are grateful to Stefan van Beveren, Valentine Alt, Jean-Loup Touchard and Anthony 263 Percevaux for climbing the trees to collect the selected branches. We thank the rest of the team 264 for helping us with the field measurements. 12 265 CONFLICT OF INTEREST 266 The corresponding author confirms on behalf of all authors that there have been no involvements 267 that might raise the question of bias in the work reported or in the conclusions, implications, or 268 opinions stated. 269 DATA AVAILABILITY STATEMENT 270 The data that support the findings of this study are openly available in Dryad Digital Repository 271 at https://doi.org/10.5061/dryad.v9s4mw6rg. 13 272 LITERATURE CITED 273 AMBROSE, A. R., SILLETT, S. C., & DAWSON, T. E. (2009). Effects of tree height on branch 274 hydraulics, leaf structure and gas exchange in California redwoods. Plant, Cell & 275 Environment, 32(7), 743-757. doi:10.1111/j.1365-3040.2009.01950.x 276 BASSET, Y., HORLYCK, V., & WRIGHT, S. J. (2003). Studying Forest Canopies from Above: The 277 International Canopy Crane Network. Smithsonian Tropical Research Institute and 278 UNEP. 279 BONGERS, F., CHARLES-DOMINIQUE, P., FORGET, P.-M., & THERY, M. (2001). Nouragues: 280 Dynamics and Plant-Animal Interactions in a Neotropical Rainforest. Kluwer Academic 281 Publishers. 282 CARDOSO, D., SÄRKINEN, T., ALEXANDER, S., AMORIM, A. M., BITTRICH, V., CELIS, M., … 283 FORZZA, R. C. (2017). Amazon plant diversity revealed by a taxonomically verified 284 species list. Proceedings of the National Academy of Sciences, 114(40), 10695-10700. 285 doi:10.1073/pnas.1706756114 286 CAVALERI, M. A., REED, S. C., SMITH, W. K., & WOOD, T. E. (2015). Urgent need for warming 287 experiments in tropical forests. Global Change Biology, 21, 2111-2121. 288 doi:10.1111/gcb.12860 289 COMSTOCK, J., & EHLERINGER, J. (1984). Photosynthetic responses to slowly decreasing 290 leaf water potentials in Encelia frutescens. Oecologia, 61, 241-248. 291 doi:10.1007/BF00396767 292 DANG, Q. L., MARGOLIS, H. A., COYEA, M. R., SY, M., & COLLATZ, G. J. (1997). Regulation of 293 branch-level gas exchange of boreal trees: roles of shoot water potential and vapor 294 pressure difference. Tree Physiology, 17(8-9), 521-535. doi:10.1093/treephys/17.8-9.521 14 295 DEVIREDDY, A. R., ZANDALINAS, S. I., GÓMEZ-CADENAS, A., BLUMWALD, E., & MITTLER, R. 296 (2018). Coordinating the overall stomatal response of plants: Rapid leaf-to-leaf 297 communication during light stress. Science Signaling, 11(518), eaam9514. 298 doi:10.1126/scisignal.aam9514 299 DOMINGUES, T. F., MEIR, P., FELDPAUSCH, T. R., SAIZ, G., VEENENDAAL, E. M., SCHRODT, F., … 300 LLOYD, J. (2010). Co-limitation of photosynthetic capacity by nitrogen and phosphorus in 301 West Africa woodlands. Plant, Cell & Environment, 33, 959-980. doi:10.1111/j.1365- 302 3040.2010.02119.x 303 DUSENGE, M. E., WALLING, G., GÅRDESTEN, J., NYRONZIMA, F., ADOLFSSON, L., NSABIMANA, D., 304 & UDDLING, J. (2015). Photosynthetic capacity of tropical montane tree species in 305 relation to leaf nutrients, successional strategy and growth temperature. Oecologia, 117, 306 1183-1194. doi:10.1007/s00442-015-3260-3 307 ELLSWORTH, D. S., & LIU, X. (1994). Photosynthesis and canopy nutrition of four sugar maple 308 forests on acid soils in northern Vermont. Canadian Journal of Forest Research, 24, 309 2118-2127. doi:10.1139/x94-272 310 ELLSWORTH, D. S., & REICH, P. B. (1993). Canopy structure and vertical patterns of 311 photosynthesis and related leaf traits in a deciduous forest. Oecologia, 96, 169-178. 312 doi:10.1007/BF00317729 313 HEBERLING, J. M., KICHEY, T., DECOCQ, G., & FRIDLEY, J. D. (2016). Plant functional shifts in 314 the invaded range: a test with reciprocal forest invaders of Europe and North America. 315 Functional Ecology, 30(6), 875-884. doi:10.1111/1365-2435.12590 15 316 KENZO, T., INOUE, Y., YOSHIMURA, M., YAMASHITA, M., TANAKA-ODA, A., & 317 ICHIE, T. (2015). Height-related changes in leaf photosynthetic traits in diverse Bornean 318 tropical rain forest trees. Oecologia, 177, 191-202. doi:10.1007/s00442-014-3126-0 319 KOCH, G. W., SILLETT, S. C., JENNINGS, G. M., & DAVIS, S. D. (2004). The limits to tree 320 height. Nature, 428, 851-854. doi:10.1038/nature02417 321 LANGE, O. L., FÜHRER, G., & GEBEL, J. (1986). Rapid field determination of photosynthetic 322 capacity of cut spruce twigs (Picea abies) at saturating ambient CO2. Trees, 1, 70-77. 323 doi:10.1007/bf00197027 324 MENG, F.-R., & ARP, P. A. (1992). Net photosynthesis and stomatal conductance of red spruce 325 twigs before and after twig detachment. Canadian Journal of Forest Research, 23, 716- 326 721. doi:10.1139/x93-093 327 MESSIER, J., MCGILL, B. J., & LECHOWICZ, M. J. (2010). How do traits vary across ecological 328 scales? A case for trait-based ecology. Ecology Letters, 13(7), 838-848. 329 doi:10.1111/j.1461-0248.2010.01476.x 330 MONTGOMERY, R., & CHAZDON, R. (2002). Light gradient partitioning by tropical tree seedlings 331 in the absence of canopy gaps. Oecologia, 131(2), 165-174. doi:10.1007/s00442-002- 332 0872-1 333 MOTZER, T., MUNZ, N., KÜPPERS, M., SCHMITT, D., & ANHUF, D. (2005). Stomatal conductance, 334 transpiration and sap flow of tropical montane rain forest trees in the southern Ecuadorian 335 Andes. Tree Physiology, 25, 1283-1293. doi:10.1093/treephys/25.10.1283 336 NAKAMURA, A., KITCHING, R. L., CAO, M., CREEDY, T. J., FAYLE, T. M., FREIBERG, M., … 337 ASHTON, L. A. (2017). Forests and Their Canopies: Achievements and Horizons in 16 338 Canopy Science. Trends in Ecology & Evolution, 32(6), 438-451. 339 doi:10.1016/j.tree.2017.02.020 340 OLSON, M. E., SORIANO, D., ROSELL, J. A., ANFODILLO, T., DONOGHUE, M. J., 341 EDWARDS, E. J., … MÉNDEZ-ALONZO, R. (2018). Plant height and hydraulic 342 vulnerability to drought and cold. PNAS, 115(29), 7551-7556. 343 doi:10.1073/pnas.1721728115 344 PERRY, D. R. 1978. A method of access into the crowns of emergent and canopy trees. 345 Biotropica, 10(2), 155-157. doi:10.2307/2388019 346 PLOWDEN, C., UHL, C., & DE ASSIS OLIVEIRA, F. (2004). Alipumilio (Diptera: Syrphidae) 347 fly association with Burseraceae tree resins in the eastern Brazilian Amazon. Journal of 348 Natural History, 38(15), 1941-1947. doi:10.1080/00222930310001618886 349 R CORE TEAM. 2017. R: A Language and Environment for Statistical Computing. R Foundation 350 for Statistical Computing, Vienna, Austria. 351 REZENDE, L. F. C., ARENQUE, B. C., AIDAR, S. T., MOURA, M. S. B., VON RANDOW, C., 352 TOURIGNY, E., … OMETTO, J. P. H. B. (2016). Evolution and challenges of dynamic 353 global vegetation models for some aspects of plant physiology and elevated atmospheric 354 CO2. International Journal of Biometeorology, 60(7), 945-955. doi:10.1007/s00484-015- 355 1087-6 356 ROELFSEMA, M. R., & HEDRICH, R.. 2005. In the light of stomatal opening: new insights into 'the 357 Watergate'. New Phytologist, 167, 665-691. doi:10.1111/j.1469-8137.2005.01460.x 358 ROWLAND, L., LOBO-DO-VALE, R. L., CHRISTOFFERSEN, B. O., MELEM, E. A., KRUIJT, B., 359 VASCONCELOS, S. S., … MEIR, P. (2015). After more than a decade of soil moisture 17 360 deficit, tropical rainforest trees maintain photosynthetic capacity, despite increased leaf 361 respiration. Global Change Biology, 21(12), 4662-4672. doi:10.1111/gcb.13035 362 SANTIAGO, L. S., & MULKEY, S. S. (2003). A test of gas exchange measurements on excised 363 canopy branches of ten tropical tree species. Photosynthetica, 41(3), 343-347. 364 doi:10.1023/B:PHOT.0000015457.92479.eb 365 SCHABERG, P. G., SHANE, J. B., CALI, P. F., DONNELLY, J. R., & STRIMBECK, G. R. (1998). 366 Photosynthetic capacity of red spruce during winter. Tree Physiology, 18, 271-276. 367 SPERRY, J. S., ALDER, N. N., & EASTLACK, S. E. (1993). The Effect of Reduced Hydraulic 368 Conductance on Stomatal Conductance and Xylem Cavitation. Journal of Experimental 369 Botany, 44(6), 1075-1082. doi:10.1093/jxb/44.6.1075 370 THE PLANT LIST. 2013. Version 1.1. http://www.theplantlist.org/ (accessed November 2018). 371 THÉRY, M. (2001) Forest light and its influence on habitat selection. Plant Ecology, 153(1-2), 372 251-261. doi:10.1023/A:1017592631542 373 VERRYCKT, L. T., VAN LANGENHOVE, L., CIAIS, P., COURTOIS, E. A., VICCA, S., 374 PEÑUELAS, J., … JANSSENS, I. A. (2020). Data from: Coping with branch excision 375 when measuring leaf net photosynthetic rates in a lowland tropical forest. Dryad Digital 376 Repository. doi:10.5061/dryad.v9s4mw6rg 377 WEEKS, A., DALY, D. C., & SIMPSON, B. B. (2005). The phylogenetic history and 378 biogeography of the frankincense and myrrh family (Burseraceae) based on nuclear and 379 chloroplast sequence data. Molecular Phylogenetics and Evolution, 35, 85-101. 380 doi:10.1016/j.ympev.2004.12.021 381 WICKHAM, H. (2009). ggplot2: Elegant Graphics for Data Analysis. . Springer-Verlag New 382 York. 18 383 WILLIAMSON, V. G., & MILBURN, J. A. (1995). Cavitation events in cut stems kept in water: 384 implications for cut flower senescence. Scientia Horticulturae, 64(4), 219-232. 385 doi:10.1016/0304-4238(95)00842-X 386 WRIGHT, I. J., REICH, P. B., WESTOBY, M., ACKERLY, D. D., BARUCH, Z., BONGERS, F., … 387 VILLAR, R. (2004). The worldwide leaf economics spectrum. Nature, 428(6985), 821- 388 827. doi:10.1038/nature02403 389 YODA, K. (1974). Three-dimensional distribution of light intensity in a tropical rain forest of 390 West Malaysia. Japanese Journal of Ecology, 24(4), 247-254. 391 doi:10.18960/seitai.24.4_247 392 19 393 TABLES 394 Table 1. Overview of the tree species selected for this study. The Plant List (2013) was used to validate the scientific 395 names and to confirm authorship. The diameter at breast height (DBH) and total height are given for each tree. The 396 heights of the upper, middle, and lower canopy foliage indicate the heights above the ground surface at which the 397 branches were sampled. Tree Family Species name DBH Total Height Height middle Height ID (cm) height (m) upper foliage (m) lower foliage (m) foliage (m) 1 Malvaceae Sterculia cf. 39.5 32 30 25 20 multiovula E.L. Taylor 2 Burseraceae Tetragastris altissima 47.4 30 27 23 20 (Aubl.) Swart 3 Lecythidacea Eschweilera coriacea 53.2 38 36 32 25 e (DC.) S.A. Mori 4 Lecythidacea Lecythis persistens 47.4 42 42 34 29.5 e Sagot 5 Lecythidacea Eschweilera coriacea 54.7 35 33 29 23 e (DC.) SA. Mori 6 Fabaceae Inga sp. 36.6 31 28 22 17.5 7 Fabaceae Eperua falcata Aubl. 81.8 40 38 33 27 20 398 FIGURE LEGENDS 399 FIGURE 1 Leaf net photosynthetic rate at light saturation (Asat; µmol/m2/s) for non-excised 400 (intact Asat) and excised (excised Asat) branches (n = 21). Each point represents an average of 2-3 401 leaves for each canopy position and tree, with standard errors on the x- and y-axes. The slope of 402 the linear regression did not differ significantly from 1 (P = 0.63). The black line represents the 403 1:1 line and the grey shading shows the 95% confidence interval. 404 FIGURE 2 Leaf net photosynthetic rate (An; µmol/m2/s) at light saturation and stomatal 405 conductance (gs; µmol/m2/s) in Eperua falcata as functions of time of day, during an 406 unreplicated experiment to test the effect of transporting the excised branch in an opaque plastic 407 bag on Asat. (A) An was continuously measured before and after excision of the branch. The solid 408 black line represents the time of excision. (B) An was measured on the attached branch. The 409 excised branch was transported in an opaque plastic bag, and An was again continuously 410 measured after 10 min until An had stabilised and differed by less than five percent from the 411 original values before excision. The dashed line represents An before excision. 412 FIGURE 3 (A) Actual data of a representative response of continuously measured leaf net 413 photosynthetic rate (An) at light saturation after excision and transport of the branch. Time 0 414 corresponds to the placement of the leaf in the gas exchange chamber. (B) Average stabilisation 415 time for the seven trees. The error bars are the standard errors of the mean. No data are available 416 for tree 5, because we did not log An continuously. (C) Average stabilisation time for branches at 417 different positions in the canopy (U, upper; M, middle; L, lower) for the seven trees. The error 418 bars are the standard errors of the mean. 21