1 Running head: leaf-water δ18O in atmospheric CAM epiphytes · This shift lead to morphological...

1

Running head: leaf-water δ18O in atmospheric CAM epiphytes 1

2

Brent R. Helliker 3

Department of Biology 4

University of Pennsylvania 5

433 S. University Ave. 6

Philadelphia, PA 19104 7

Tel: 215-746-6217 8

Fax: 215-898-8780 9

Email: [email protected] 10

11 12 Whole Plant and Ecophysiology 13 14

15

Plant Physiology Preview. Published on February 7, 2011, as DOI:10.1104/pp.111.172494

Copyright 2011 by the American Society of Plant Biologists

www.plantphysiol.orgon May 8, 2020 - Published by Downloaded from Copyright © 2011 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

2

On the controls of leaf-water δ18O in the atmospheric CAM epiphyte Tillandsia 16

usneoides. 17

18

19

Brent R. Helliker 20

Department of Biology, University of Pennsylvania, 415 S. University Ave, Philadelphia, 21

PA 19104. 22

23



3

Financial source: 24

This work was supported by the National Science Foundation under award number IOB-25

0615501. Any opinions, findings and conclusions or recommendations expressed in this 26

material are those of the author(s) and do not necessarily reflect those of the National 27

Science Foundation. 28

29

Corresponding author: Brent R. Helliker, [email protected] 30

31



4

32

Abstract 33

Previous theoretical work showed that leaf-water isotope ratio (δ18OL) of CAM 34

epiphytes was controlled by the δ18O of atmospheric water vapor (δ18Oa), and 35

observed δ18OL could be explained by both a non-steady state model and a 36

‘maximum enrichment’ steady-state model (δ18OL-M); the latter requiring only δ18Oa 37

and relative humidity (h) as inputs. δ18OL should therefore contain an extractable 38

record of δ18Oa. Previous empirical work supported this hypothesis, but raised 39

many questions: How does changing δ18Oa and h affect δ18OL? Do hygroscopic 40

trichomes affect observed δ18OL? Are observations of changes in water content (W) 41

required for prediction of δ18OL? Does the leaf need to be at full isotopic steady state 42

for observed δ18OL to equal δ18OL-M? These questions were examined with a climate-43

controlled experimental system capable of holding δ18Oa constant for several weeks. 44

Water adsorbed to trichomes required a correction ranging from 0.5 to 1 ‰. δ18OL 45

could be predicted using constant values of W and even total conductance (gtot). 46

Tissue rehydration caused a transitory change in δ18OL but the consequent increase 47

in gtot led to a tighter coupling with δ18Oa. The non-steady-state leaf water models 48

explained observed δ18OL (y = 0.93*x - 0.07, R2 = 0.98) over a wide range of δ18Oa 49

and h. Predictions of δ18OL-M agreed with observations of δ18OL (y = 0.87*x - 0.99, R2 50

= 0.92), and when h > 0.9, the leaf did not need to be at isotopic steady state for the 51

δ18OL-M model to predict δ18OL in the CAM epiphyte Tillandsia usneoides. 52

53



5

Introduction 54

Tropical and subtropical epiphytic higher plants have long been a curiosity to plant 55

physiologists because of the multiple and unique constraints on physiology that the 56

epiphytic lifeform represents (Mez, 1904; Benzing, 1970; Medina and Troughton, 1974). 57

While the competition for light has no doubt lead the plants to the tree tops, the 58

concomitant loss of roots effectively removed the plants from an environment of 59

abundant rainfall to one of arid conditions (Griffiths et al., 1986; Smith et al., 1986; 60

Winter and Smith, 1996). In a sense, the plants shifted biomes by immigrating vertically 61

instead of horizontally. This shift lead to morphological adaptations like the development 62

of modified leaf hairs to absorb water and the formation of tanks via the overlapping of 63

leaf bases as well as physiological adaptations such as maintaining high leaf osmotic 64

potential and, perhaps most notably, the evolution of the CO2 concentrating mechanism 65

known as Crassulacean acid metabolism (CAM) photosynthesis (Medina and Troughton, 66

1974; Benzing et al., 1976; Griffiths and Smith, 1983; Smith et al., 1986; Martin and 67

Schmitt, 1989; Martin et al., 2004; Ohrui et al., 2007). This unique physiology of tropical 68

and subtropical CAM epiphytes also presents an intriguing contrast to the general way we 69

view oxygen isotope ratios (δ18O) in plant water and organic material (Helliker and 70

Griffiths, 2007; Helliker and Noone, 2009). 71

72

The enrichment of 18O in leaf water during plant transpiration leads to a suite of 73

physiology-based tracers that inform us of current and past plant-environment 74

interactions. Leaf water δ18O (δ18OL) labels CO2 and allows for partitioning of the gross 75

components of net CO2 flux from ecosystem to regional scales (Yakir and Wang, 1996; 76

Ciais et al., 1997; Styles et al., 2002; Cuntz et al., 2003; Ogee et al., 2004; Helliker et al., 77

2005). The production of isotopically distinct O2 by photosynthesis allows for global-78

scale estimates of productivity over millennia (Guy et al., 1993; Luz et al., 1999; 79

Hoffmann et al., 2004). The isotopic record of δ18OL in leaf and tree ring cellulose allows 80

for the reconstruction of growth environment and/or physiological responses to that 81

growth environment. (Epstein et al., 1977; Anderson et al., 1998; Switsur and 82

Waterhouse, 1998; Barbour and Farquhar, 2000; Barbour et al., 2000; Roden and 83

Ehleringer, 2000; Ferrio and Voltas, 2005; Poussart and Schrag, 2005; Helliker and 84



6

Richter, 2008). Much of the error associated with the above δ18O applications could be 85

decreased with better estimates of the isotope ratio of atmospheric water vapor (δ18Oa), a 86

primary control on leaf-water 18O enrichment (Farquhar and Cernusak, 2005; Helliker 87

and Griffiths, 2007), and the study of δ18OL in CAM epiphytes may offer better estimates 88

of δ18Oa through time and space (Helliker and Griffiths, 2007). 89

90

The epiphyte— and specifically the CAM vascular epiphyte— offers an extreme in the 91

way δ18O in plant water and organic material is viewed. For most vascular plants, δ18OL 92

is controlled by a balance of soil water δ18O and δ18Oa; soil water comes into the leaves 93

via the root system and atmospheric vapor diffuses into to leaf through open stomata. The 94

balance of the two water sources is determined by relative humidity (h); if h is high then 95

δ18OL is controlled proportionally more by δ18Oa, if h is unity then δ18OL is at equilibrium 96

with δ18Oa. If these plants happen to lose water during a humid night, then leaf water 97

exchange can lead to control by— and possibly equilibrium with— δ18Oa (Lai et al., 98

2008). Typically, the dramatic increase of transpiration during the daytime moves the 99

δ18OL away from equilibrium with δ18Oa. Hence, the nocturnal movement of δ18OL 100

towards equilibrium with δ18Oa in most vascular plants is not likely recorded in plant 101

organic material and is only important to the isotopic composition of CO2 during 102

nocturnal respiration (Cernusak et al., 2004). In the CAM epiphyte, rainwater and dewfall 103

rehydrates epiphyte tissue, and due to stomata being open at night when relative humidity 104

is near unity, rehydration water is quickly exchanged with atmospheric water vapor. This 105

leads to the situation, hypothesized by Helliker and Griffiths (2007), where rainfall and 106

dewfall (which are in equilibrium with δ18Oa; (Gat, 1996)) control tissue water status, and 107

the δ18O of this tissue water is continuously controlled by δ18Oa through subsequent 108

nocturnal vapor exchange. Because stomata are typically closed during the day, organic 109

material synthesized by the photosynthetic carbon reduction cycle should obtain the δ18O 110

signature of δ18Oa-controlled leaf water (Helliker and Griffiths, 2007). It is this situation 111

that yields new applications for δ18O that is unique to epiphytes: the reconstruction of 112

physiological responses to the epiphytic growth habit (Reyes-Garcia et al., 2008) and the 113

reconstruction of δ18Oa (Helliker and Griffiths, 2007). It was recently shown through 114



7

empirical and theoretical work that lichen thalli obtain equilibrium with δ18Oa even under 115

non-saturating conditions (Hartard et al., 2009). The lichen system of equilibrium with 116

δ18Oa represents an important contrast to that of CAM epiphytes, and this is discussed 117

later. The theoretical work presented by Hartard et al., however, is extremely helpful for 118

the interpretation of isotopic signals in CAM epiphytes. 119

120

Helliker and Griffiths (2007) developed the theoretical underpinnings for δ18Oa as a 121

control on CAM epiphyte δ18OL. They extended the thoughts of Farquhar and Cernusak 122

(2005) to show that at the high nocturnal relative humidity (h) experienced by CAM 123

epiphytes, gross exchange fluxes of water vapor into the leaf from the atmosphere can be 124

several times the transpirational flux out of the leaf. Model simulations showed that 125

constant environmental conditions and continuous water loss lead to an isotopic steady 126

state where both δ18OL and transpired water (δ18OE) converged to a single isotopic value 127

controlled by the exchange of δ18Oa and described by the following ‘maximum 128

enrichment’ equation (Farquhar and Gan, 2003; Helliker and Griffiths, 2007; Hartard et 129

al., 2009): 130

131

RL−M ≈ Rah1

α *− α K + α K h

132

133

Where the steady-state δ18OL (RL) is determined solely by δ18Oa (Ra), h, the temperature 134

dependent equilibrium fractionation factor α*, and the balance of the ratio of diffusivities 135

of light to heavy water molecules through the stomata and through the leaf boundary 136

layer, αK (Flanagan et al., 1991; Farquhar and Lloyd, 1993; Farquhar and Cernusak, 137

2005). The non-steady state analytical solution for δ18OL developed in by Hartard et al. 138

(2009; Equation 6 in Materials and Methods, hereafter referred to as the Hartard-Cuntz 139

solution) clearly demonstrates that through time in the non-steady state δ18OL is 140

continuously moving towards the value of RL-M (δ18OL-M). In a similar manner, the 141

simulations of Helliker and Griffiths suggested that even at values of nocturnal h of 0.8— 142

which is low for a CAM epiphyte (Garth, 1964; Smith et al., 1986)— δ18OL was 143



8

controlled almost entirely by the isotope ratio of atmospheric water vapor and that the 144

maximum enrichment equation above would ultimately predict δ18OL. The simulations 145

also showed that the approach to steady was faster and occurred through less water loss 146

as h increased. In general, their simulations showed that δ18OL approached the steady-147

state value, or δ18OL-M, much faster than δ18OE and this yields an important prediction 148

that is tested by current study: the leaf does not need to be at full isotopic steady state for 149

δ18OL to be at or near the steady-state, maximum enrichment value (δ18OL-M). 150

151

There are many unresolved questions as to the agreement between models and 152

observations that underpin the efficacy of using δ18OL to reconstruct δ18Oa. The 153

empirical work of Helliker and Griffiths (2007) did show support for the modeling 154

exercises, but only over a narrow range of conditions. Their experimental setup was 155

limited as it was developed to demonstrate a preliminary proof of concept. Also, the 156

invariably high h and noisy δ18Oa (± 1.2 ‰) of Helliker and Griffiths could lead one to 157

erroneously conclude that δ18OL was in direct equilibrium with δ18Oa even when h was 158

less than unity, which should not be the case. The primary controls of CAM epiphyte 159

δ18OL are h and δ18Oa and the manner in which changes in δ18Oa and h affect δ18OL 160

through time must be assessed. Like many CAM epiphytes, the study species Tillandsia 161

usneoides has a heavy covering of hygroscopic trichomes. While the water adsorbed to 162

these trichomes does not lead to water uptake by living cells at sub-saturating conditions 163

(Martin and Schmitt, 1989), the water is inevitably sampled and extracted for δ18OL 164

determination. Hence, the isotopic offset caused by these trichomes must be determined. 165

Total plant conductance to water loss (gtot; stomatal and boundary layer conductance) 166

controls the rate of exchange of tissue water with water vapor, and the relative water 167

content (RWC) of the plant. Therefore the effect of changes in RWC on δ18OL — through 168

both water loss and rehydration— must be determined. Previous work has shown that 169

changes in water volume are not important to predictions of δ18OL (Cernusak et al., 2002; 170

Cuntz et al., 2007) and a similar finding in CAM epiphytes could greatly simplify our 171

interpretation of observed δ18OL. In summary, the goal of this study was to construct 172

controlled-environment experiments to assess how well observed δ18OL could be 173



9

predicted over a range of conditions through both steady state and non-steady state 174

approaches of Helliker and Griffiths (2007) and Hartard et al. (2009). 175

176

Results 177

Representative chamber measurements of h, air temperature, and δ18Oa over a 16 day 178

experimental period are presented in Table I. Mean values of δ18Oa, h and the number of 179

days that T. usneoides strands were in the chamber for each experimental run are 180

presented in Table II. An image of the experimental setup and representative diurnal 181

cycles are presented in the Supplementary Material. Data from the 4 initial experimental 182

runs that were used to test the chamber setup were not included. Throughout this paper, h 183

= ea/ei and it is assumed that h equaled relative humidity in all chamber studies because 184

the chamber fans assured little boundary layer development around the T. usneoides 185

ramets. Support for this assumption is provided from several early runs of the chamber 186

setup where T. usneoides leaf temperatures were measured and they were not 187

significantly different from measures of air temperature (data not shown), so ei was equal 188

to esat at air temperature. The chamber held temperature and h relatively constant 189

(typically standard deviation for temperature = 0.08 °C and for h = 0.015; Table 1). 190

Maintaining constant h above 0.9 for long periods of time is difficult and, to my 191

knowledge, has not been attempted in many plant gas exchange studies. Condensation in 192

tubing and within the chamber, and directly on any chamber temperature control device 193

like the peltier unit, can systematically and dynamically alter h within the chamber. There 194

were occasional, unexplained changes in h as seen on days 12 and 13 (Table I). These 195

changes were manifest as a systematic drop in h that was caused by a slight drop in ea and 196

not changes in temperature (thus, not a change in ei). Fortunately such variability in h did 197

not appear to compromise the long-term stability of δ18Oa. The source for δ18Oa was the 198

23 L capacity temperature-controlled water tower, the volume and specific heat of which 199

lead to a low multi-day standard deviation for δ18Oa of ± 0.15 ‰. 200

201

To quantify the amount of water adsorbed to trichomes (Figure 1) and to determine how 202

the δ18O of this trichome water should affect observed δ18OL (Figure 2) the gain in mass 203

from water adsorption to T. usneoides trichomes was determined at 23 separate settings 204



10

of h from 0.34 to unity. The data in Figure 1 represent the water adsorbed to trichomes 205

expressed as RWC, therefore this value is comparable to RWC values elsewhere in this 206

study. Below an h of about 0.8, the water absorbed to trichomes was less than 8 % RWC, 207

however, above 0.8 the trichome water increased to more than 40% at h =1. While the 208

latter value may seem high, these results are consistent with water hydration studies of 209

non-lipid plant cuticle components (Dominguez and Heredia, 1999). These data were 210

used to parameterize the Hailwood and Horrobin model which was then used to predict 211

the ‘trichome offset’ for a variety of T. usneoides relative water contents (RWC) and h 212

(Figure 2). For the model values of Figure 2, RWC includes both tissue water and water 213

adsorbed to trichomes. The expected isotopic offset can be as high as 2 ‰ if RWC is low 214

and the h during sampling is high, however, for most of the sampling scenarios 215

encountered in this study, the trichome offset was between 0.5 and 1 ‰, which is still 216

well above the measurement precision of 0.2 ‰. These results show that the trichome 217

offset must be considered for this study because sampling occurred during nocturnal 218

periods for the plants, when h was highest. All steady and non-steady-state predicted 219

values of δ18OL were corrected for trichome water adsorption based on these observations 220

(see Materials and Methods). In field sampling conditions, trichome offsets should be less 221

of a worry because of typically low h during the day. For example, at an h of 0.65 and 222

RWC of 100 %, the trichome offset is near the sampling precision for δ18O (~ 0.2 ‰). If 223

the RWC is decreased to 50 % (thereby increasing the proportional representation of 224

trichome water) the offset increases to 0.36 ‰. 225

226

To examine how δ18OL changed as a consequence of rehydration, T. usneoides strands 227

that were not watered for one-week were placed in enriched water and removed at set 228

times from 10 minutes to 24 hours (Figure 3). A mass balance using the δ18O of wetting 229

water (11.5 ‰), the initial δ18OL (-3.3 ± 0.31 ‰) and the final δ18OL (3.7 ± 0.57 ‰) 230

showed that approximately 47% of the final water volume was the wetting water, or that 231

plant water volume nearly doubled due to the rewetting. After 5 hours of submersion in 232

water, the δ18OL was statistically indistinguishable from δ18OL after 24 hours submersed 233

(Figure 3). These results are similar to previous work showing that T. usneoides attains 234

saturated water content in approximately 4 hours. After only 20 minutes submerged, 235



11

δ18OL became significantly enriched by 2.5 ‰ and approximately 70 % of the “new” 236

water was absorbed. Therefore, substantial changes in δ18OL can occur over relatively 237

short time spans due to water absorption. 238

239

Figure 4 shows total plant conductance (gtot; mmol m-2 s-1; stomatal and boundary layer 240

conductance) for water vapor from T. usneoides versus RWC for average chamber h 241

ranging from 0.556 to 0.997. There was a highly significant (R2 = 0.78, p < 0.001) 242

correlation between RWC and gtot. Across treatments there was also a significant effect of 243

vapor pressure deficit (VPD) on gtot, but VPD explained only about 17 percent of the 244

variation in gtot (data not shown). 245

246

A general empirical sense of how the control of δ18OL by δ18Oa is regulated by h, and that 247

as h approaches unity δ18OL approaches equilibrium with δ18Oa can be seen in Figure 5. 248

Helliker and Griffiths (2007) showed that, under reasonably constant conditions, δ18OL 249

tended towards maximum enrichment (Equation 4 in Materials and Methods) after four 250

days. Figure 1 shows the difference between δ18OL and δ18Oa (δ18OL – δ18Oa) plotted 251

against h for every δ18OL sample that had been in the chamber for at least 4 days (n = 252

169). The data showed a tight relationship (y = -44.9x + 52.8, p < 0.0001, R2 = 0.87) 253

where h explained 87 percent of observed δ18OL. However, when the range of h was 254

restricted to 0.8 and above— conditions more in line with natural field conditions— h 255

explained the observed data scatter less well (y = - 59x + 66.0, p < 0.0001, R2 = 0.67). 256

257

Figures 6 and 7 show day to day observations and predictions of δ18OL for representative 258

chamber studies at three high (0.96 and 0.92 in Figure 6, and 0.97 in Figure 7A), one mid 259

(0.84; Figure 7B) and one low (0.56; Figure 7C) values of mean h. The leaf water models 260

(Equations 1, 4 and 6) were parameterized with observed nocturnal h, temperature, δ18Oa, 261

and gtot data for each day and the solutions were corrected for water adsorbed to 262

trichomes (δ18Otrichome) using observed RWC, and the empirically derived offset between 263

δ18Oa and δ18Otrichome (see Materials and Methods). The models were initialized with 264

observed Day 0 data for δ18OL, so that both observed and predicted values were equal at 265



12

the start. For the Hartard-Cuntz solution (Equation 6) I used the following schemes: 266

observed gtot and changing W, observed gtot and constant W (9 mol/m-2) and constant gtot 267

(0.002 mmol m-2 s-1) and W. Both Figures show the march of non-steady-state predictions 268

and observed δ18OL towards maximal enrichment (RL-M, Equation 4). Figure 6 further 269

shows that the non-steady-state predictions from the Hartard-Cuntz solution match 270

observations well, whether or not W, or gtot and W, were kept constant. Where h was 271

above 0.9, maximal enrichment was achieved in 4 to 9 days. Within similar levels of h 272

(Figure 6A vs. Figure 7A), the length of time to reach δ18OL-M was determined by the 273

isotopic distance between δ18OL-M and δ18OL at Day 0. Finally, δ18OE (RE; Figure 6 and 274

Figure 7A) slowly approached and occasionally reached the value of δ18OL-M as would be 275

expected at isotopic steady state. In contrast, observed δ18OL quickly matched δ18OL-M 276

and generally followed δ18OL-M even with slight changes in h (e.g. Days 16-18 in Figure 277

6A). 278

279

For all experimental runs, predicted δ18OL using the full non-steady-state leaf-water 280

model, Equation (1), agreed very well with measured δ18OL (Figure 8A). Every sampling 281

period of the chamber studies with the exception of Days 0 are presented, therefore 282

Figure 8 includes all of the within-experiment dynamism shown in Figures 6 and 7, and 283

the broad range of h and δ18Oa shown in Table II. The gray line represents a perfect 284

correlation between predicted and observed values. The solid line represents the 285

relationship between predicted and observed, the slope of which was near 1 with an 286

intercept near zero (y = 0.93*x - 0.07, R2 = 0.975, p < 0.0001). δ18OL-M predictions from 287

the maximum enrichment model, Equation (5), also agreed well with all observed δ18OL 288

that were sampled at h > 0.8 and after four days in the chamber (Figure 8B). That data in 289

Figure 8B include all of the data in Figure 1 in which h > 0.8. 290

291

Rehydration of plants during experimental runs did cause a significant change in δ18OL, 292

but the change was short-term. In 4 out of 12 of the high h experimental runs (h > 0.9), 293

two populations of T. usneoides were maintained within the chamber, one wetted and one 294

non-wetted. The frequency of the wetting varied from 2 to 7 days, and the rehydration 295



13

water was enriched to artificially enhance the difference between the wetted and non-296

wetted δ18OL. Representative data from these comparisons are presented in Table III. As 297

in all experiments, all plants were fully hydrated on Day 0. From Day 1 to Day 6 all 298

plants received no rehydration water. On Day 7, δ18OL samples were taken and half of the 299

plants were submerged in the wetting water for an hour. When all plants were sampled 300

for δ18OL on day 10, there was no significant difference between the wetted and non-301

wetted plants. The wetted subset of plants were again rehydrated on Day 10 (and again on 302

Day 14) and in each subsequent sampling period there were no significant differences 303

between the wetted and non-wetted plants. δ18OL for the wetted plants was sampled 304

shortly after rehydration on Day 10 to show that the wetting water did in fact cause a 305

significant enrichment of δ18OL, yet this difference was not apparent on Day 14. With one 306

exception, the results in Table III are comparable to all the wetted versus non-wetted 307

comparisons. In only one part of one trial was there a significant difference between 308

wetted and non-wetted strands. In this trial, the wetted plants were rehydrated every two 309

days and the non-wetted plants were very dry (RWC below 80 %) and exchanging 310

(losing) very little water with chamber water vapor (gtot was very low). 311

312

Discussion 313

It is a long-term goal to use the δ18O of epiphytic CAM plants— both the δ18O of leaf 314

water (δ18OL) and organic material that is labeled by δ18OL— to illuminate plant 315

physiological responses to the unique epiphytic growth habit (Reyes-Garcia et al., 2008) 316

and to reconstruct the δ18O of atmospheric water (δ18Oa; Helliker and Griffiths 2007); the 317

latter goal being much more broadly important to the use of stable isotopes in plant 318

physiological and ecophysiological research. The largely abiotic factors h and δ18Oa are 319

the primary controls of δ18OL (Figure 1) and their control under both constant and 320

dynamical conditions and must be considered. In addition, epiphyte morphology and 321

physiology act as important secondary controls on δ18OL and not accounting for both 322

could potentially obscure any δ18O interpretation. Plant physiology affects δ18OL 323

primarily through total plant conductance (gtot; stomatal and boundary layer conductance) 324

which controls the rate of exchange of tissue water with water vapor, and the relative 325



14

water content (RWC) of the plant. Plant morphology should affect epiphytic δ18OL 326

through water trapped in tanks formed at leaf bases (although this is not a concern for T. 327

usneoides), and through water adsorbed to hygroscopic trichomes. Therefore, the goal of 328

this study was to use controlled experiments on the CAM epiphyte T. usneoides to assess 329

how well observed δ18OL could be explained through controlled variation in environment 330

and physiology. 331

332

Water adsorbed to trichomes was shown to have a substantial potential impact (Figures 2 333

and 3) on observed and predicted δ18OL, but the actual impact and the need for 334

accounting in field and lab situations needs clarification. For all predictions shown in 335

Figure 7, the relationship between predicted δ18OL using the trichome offset and 336

predictions not using the offset was y = 0.99x - 0.81 (R2 = 0.997), so the non-corrected 337

predictions of δ18OL were about 0.8 ‰ more enriched than the predictions using the 338

trichome offset. This overall difference of about 1 ‰ is clearly a substantial offset. 339

However, the experimental setup exacerbated this offset because we sampled δ18OL 340

during the nocturnal period when h was highest. This setup was chosen so that δ18Oa 341

could be sampled during the period when most vapor exchange was occurring. Under 342

field conditions, δ18OL would likely be sampled during the low humidity day then the 343

trichome offset would be less of a problem, and perhaps not a problem at all. The 344

exception to this would be sampling leaves for δ18OL near dawn or dusk when h would 345

likely be high. It should be pointed out that while most of the adsorbed water was likely 346

onto the trichomes, adsorption to the cuticle could not be ruled out. This is somewhat 347

inconsequential as discussed below. Also, the apparent isotopic offset between the 348

trichome water and δ18Oa was 7.4 ± 0.1 ‰ while the equilibrium fractionation factor for 349

condensation was around 9.2 ‰. It is unclear at this time how similar adsorption of 350

water to leaf cuticles and trichomes of more common plant types (i.e. non-epiphytic) 351

could bias observations of samples taken near dawn or dusk. Similarly, this is an unlikely 352

problem in field studies with the exception of diurnal sampling schemes. In the latter 353

case, water adsorption to leaf cuticles/trichomes could lead to lower than expected 354

enrichment during high humidity morning and late afternoon sampling. 355



15

356

When h is below 1, The direct adsorption of atmospheric water vapor by the dead 357

trichomes of T. usneoides does not lead to any apparent absorption by the living cells of 358

the plants (Martin and Schmitt, 1989). Several studies have reported direct uptake of 359

water by atmospheric epiphyte leaves, but this is under conditions of saturating vapor 360

and/or leaf wetness (Benzing et al., 1976; Martin and Schmitt, 1989; Andrade, 2003). The 361

trichomes of T. usneoides are highly hygroscopic and can, within gas exchange 362

chambers, lead to what appears as a ‘negative’ transpiration while equilibrating to an 363

increase in h. This negative transpiration is the process of water adsorbing onto the 364

trichomes and it stops once equilibrium is reached with h. This process is also what was 365

quantified in isolation here in with dead T. usneoides (Figure 2). The dead trichomes of 366

T. usneoides likely equilibrate with the water potential of ambient air much like what is 367

seen in the lichen thallus (Lange et al., 2001; Hartard et al., 2009) and isotopically the 368

adsorption of water onto the trichomes may be very similar to that seen by the thallus of a 369

lichen. This is, however, where the similarities end. Lichens are poikilohydric and some 370

algal-symbiotic species can become physiologically active solely through this water 371

adsorption (yet only at relatively high h); thus, the adsorption of water vapor by the 372

thallus matrix can lead to a cellular absorption (Lange et al., 2001). Such an absorption of 373

atmospheric water vapor does not occur in homoiohydric T. usneoides and is unlikely to 374

occur in other vascular plant atmospheric-type epiphytes (e.g. Bromeliads, Orchids, 375

Lycophytes, and Ferns; (Martin et al., 2004)). 376

377

T. usneoides, like most epiphytic higher plants that have been observed, maintains very 378

high (near zero) plant water potentials with osmotic potentials maintained consistently in 379

the realm of aquatic plants (Martin et al., 2004). In general, the measured water potential 380

of atmospheric epiphytes, T. usneoides included, has rarely been observed below -1 MPa 381

(Martin, 1994). In one experiment with T. ionantha, total plant water potential did not 382

drop below -0.75 MPa after 60 days of imposed drought (Nowak and Martin, 1997). 383

Considering that the water potential of air at h = 0.99 is -1.4 MPa (and -2.8 MPa at h = 384

0.98, both at 20 °C), it is hard to imagine a scenario in which water adsorbed onto 385

trichomes that are in equilibrium with subsaturated air could move against the water 386



16

potential gradient into the living cells of T. usneoides, and the work of Martin and 387

Schmitt (1989) corroborates this. Therefore, there should be no consideration of water 388

vapor uptake by trichomes affecting δ18OL. Once saturation is realized and the plant 389

tissue is covered with liquid water, the trichomes become appressed against the cuticle 390

and channel the absorption of liquid water into the underlying living cells (Benzing et al., 391

1976). This sort of precipitation event is explicitly considered through Equation 1. 392

Furthermore, the maintenance of extremely high water and largely invariant water 393

potential in the face of drought is likely the driving force behind the austere physiology 394

of epiphytes (Martin et al., 2004) in terms of both carbon assimilation and water loss. For 395

T. usneoides, well-watered measurements of CO2 uptake as low as 0.2 μmol m-2 s-1 396

(Martin & Schmitt, 1989) and transpiration of 0.002 mmol m-2 s-1 are common (this study 397

and (Martin, 1994)). The tight coupling of plant water potential and gtot likely led to the 398

rapid decrease in gtot as RWC decreased (Figure 5) and could explain why gtot was more 399

highly correlated with RWC than with VPD. The inherently low levels of water loss 400

meant that estimates of gtot could be determined once per day or even once every other 401

day, thus maintaining constant chamber conditions for longer periods of time (see 402

Supplementary Material). 403

404

Transpiration forced δ18OL towards the value of maximum enrichment, δ18OL-M, and the 405

closer the value of h to unity, the faster was the approach to δ18OL-M. This overall 406

tendency can be seen in Figures 6 and 7 where all values of δ18OL moved towards the 407

predicted δ18OL-M. Yet only when values of h were greater than 0.9 (Figures 6 and 7A), 408

did observed δ18OL actually reach δ18OL-M. At higher h, the gross flux of water into the 409

leaf (gtot • wa) was proportionally much higher than at low h and therefore the 410

replacement of leaf water with atmospheric water was faster. At lower h, a considerably 411

greater volume of leaf water must be lost before δ18OL becomes equal to δ18OL-M 412

(Helliker and Griffiths, 2007) and the response of gtot to RWC tends stop any great 413

reduction in leaf water volume (Figure 5). A quick comparison of Figures 6A and 7A 414

shows two experimental runs at similar h that took drastically different amounts of time 415

for δ18OL to equal δ18OL-M. For Figure 7A, the apparent leaf-water turnover time was 416

about 4 days, similar to the results of Helliker and Griffiths (2007), whereas in Figure 6A 417



17

the apparent leaf-water turnover time was 9 days. The reason for this discrepancy was 418

the different starting points of δ18OL. In Figure 6A, the initial δ18OL was approximately 419

15 ‰ more enriched than δ18OL-M and for Figure 7A the starting δ18OL was 420

approximately 7 ‰ more enriched than δ18OL-M. 421

422

In general, as water is lost from CAM epiphytes δ18OL is more and more likely to be 423

described by the maximum enrichment model, δ18OL-M. Yet it should be recognized that 424

as the period between rehydration events increases, RWC decreases and so does gtot, 425

which leads to an inevitable decrease in the water vapor exchange between the leaf and 426

the atmosphere. So, as the plant dries, δ18Oa still controls δ18OL, but the turnover of leaf 427

water and consequently any affect that changes in ambient δ18Oa has on δ18OL are 428

minimized. This can be seen distinctly in Figure 6C where even after several days at low 429

humidity, the enrichment of 18O in leaf water ceased. As T. usneoides dries out, the 430

overall tissue water exchange with atmospheric water vapor slows until δ18OL becomes 431

decoupled from δ18Oa. 432

433

The largest potential hurdle in using δ18OL-M to describe observed δ18OL appears to be the 434

effect that the isotope ratio of rehydration water (simulated precipitation events) has on 435

δ18OL and there are two components of this: the absolute difference in isotopic space 436

between the rehydration water and δ18OL-M and the total amount of rehydration water 437

absorbed by the plant. As a general response, water gain through precipitation has the 438

initial tendency to move δ18OL farther away from δ18OL-M, but then— due to increased 439

gtot—rapidly trends toward δ18OL-M. The more specific response of T. usneoides δ18OL to 440

precipitation events was examined three ways: The first was through the rehydration 441

study (Figure 4) which showed that the absorption of precipitation can lead to a rapid and 442

significant change in δ18OL after only 20 minutes in liquid water and after five hours the 443

change in δ18OL was complete. The second was starting all experimental runs by soaking 444

plants in an enriched water source, thereby ensuring that each experimental run began 445

after a simulated precipitation event. This method showed that, at high h, the 446

precipitation isotope label could exchange quickly with atmospheric water vapor and the 447



18

speed of exchange depended on the size of disparity between the initial label leaf water 448

and δ18OL-M. The final assessment of how precipitation inputs affect T. usneoides δ18OL 449

was designed to examine the differential effect of rehydrating one subset of strands while 450

keeping the other dry during the same experimental runs (Table III). This approach 451

explored a potential situation in the field where rainfall or dew formation could vary 452

spatially either across a field site or within a single clump of T. usneoides depending on 453

the intensity of a precipitation event. This method was performed in several of the high h 454

experimental runs, the frequency of the wetting varied from 2 to 7 days, and the 455

rehydration water was enriched. In general, there was no sustained, significant difference 456

between the wetted and non-wetted plants. Even though significant enrichment occurred 457

in as little as 20 minutes (Figure 4), it appears that this water was quickly turned over 458

through nocturnal exchange of water vapor. This rapid turnover is likely due to the fact 459

that a rehydration allows for an uptick in gtot which then allows for increased gross 460

exchange of vapor between the plant and the atmosphere. The end result being that both 461

wetted and non-wetted strands tracked δ18Oa, although it seems likely that strands with 462

higher RWC would respond to changes in δ18Oa more quickly than drier strands, as 463

mentioned above. It appears that a relatively consistent input of dew or precipitation 464

water could lead to a tighter coupling of δ18OL and δ18Oa because of enhanced vapor 465

exchange. Additionally, as discussed below, I hypothesize that under field conditions 466

precipitation itself should lead to a greater coupling between δ18OL-M. 467

468

The use of enriched rehydration water allowed for easier observation of how precipitation 469

affected δ18OL, but also led to a potential misinterpretation of the coupling between δ18OL 470

and δ18Oa. Both the rapid change in δ18OL during rehydration (Figure 4) and the multi-471

day turnover of δ18OL could lead one to conclude that a rehydration event pulls δ18OL too 472

far away from δ18OL-M to use δ18OL as a decent proxy for δ18Oa. However the δ18O of the 473

wetting water was typically 15 ‰ more enriched in 18O than the δ18Oa-equilibrated of 474

δ18OL and 25 ‰ to 30 ‰ more enriched than δ18Oa. While this isotopic span allowed for 475

a clearer picture of rehydration effects and tissue-water turnover times, it is highly 476

unlikely that such a span would occur in the field. Precipitation (dewfall or rainfall) 477



19

should also be near equilibrium with δ18Oa (Gat 1996). Therefore, isotopic differences 478

between rain and/or dew δ18O and δ18OL (and δ18OL-M) on the scale of 15 ‰ would not be 479

expected— not even considering large-scale episodic events that can dramatically alter 480

δ18O of precipitation (Miller et al., 2006)— and precipitation should enhance the 481

coupling of δ18OL and δ18Oa. 482

483

When variation in δ18Oa, h, gtot, RWC and trichome water adsorption were considered 484

through the leaf water models (Equations 1, 4 and 6) and through a wide-range of 485

conditions (Table II), observed δ18OL was explained very well (Figures 6, 7 and 8). Both 486

variations of the non-steady state model (Equations 1 and 6) predicted observed δ18OL 487

well and this included runs when gtot and W were held constant (Figure 6). Previous work 488

on non-epiphytic plants has also shown measurements of dW

dt are not crucial for accurate 489

predictions of non-steady-state models (Cernusak et al., 2002; Farquhar and Cernusak, 490

2005; Cuntz et al., 2007; Ogee et al., 2007; Kahmen et al., 2008). The modeled estimates 491

of δ18OE (RE; Figure 6 and Figure 7A) approached and only occasionally matched the 492

value of δ18OL-M as would be expected at isotopic steady state. It can be concluded then 493

that δ18OL was rarely at true isotopic steady state, however observed δ18OL often matched 494

δ18OL-M at values of h > 0.9 and generally followed δ18OL-M even with slight changes in h. 495

This apparent contrast leads to one of the more important conclusions of this study: that, 496

in agreement with the predictions of Helliker and Griffiths (2007), leaf water does not 497

need to be at full isotopic steady state for observed δ18OL to equal δ18OL-M. 498

499

The extreme sensitivity of δ18OL to h > 0.9 means that Equation 7— the full equilibrium 500

equation— is rarely useful. This may seem contradictory to the some of the data 501

presentation of Helliker and Griffiths (2007) which showed a prediction range for 502

Equation 5 on some figures. The experimental setup of Helliker and Griffiths had an 503

invariably high h and noisy δ18Oa (± 1.2 ‰) and we used the range of predictions 504

stemming from the variability in δ18Oa to show that observations were near these values. 505

Unfortunately, this approach could lead one to erroneously conclude that δ18OL was in 506



20

direct equilibrium with δ18Oa even when h was less than unity, which is not the case. The 507

value of h should be known to decide which equation to use when extracting δ18OL from 508

measures of δ18Oa using CAM epiphytes or non-epiphytic plants as in the pre-dawn 509

sampling approach developed by Lai et al. (2008). While predictions of δ18OL-M 510

(Equation 4) matched observed δ18OL at h> 0.9, the predictions from this model showed 511

good agreement with observed δ18OL over the broader range of h > 0.8 (y = 0.87x - 0.99, 512

R2 = 0.92; Figure 8B) after 4-5 days. The primary need to exclude data under four days 513

results from the artifact of using an enriched water source for rehydration, as mentioned 514

above. A more complete test of Equation 4, the maximum enrichment model δ18OL-M, has 515

been performed under field conditions and is the subject of subsequent publications. 516

517

518

Conclusions 519

Helliker and Griffiths (2007) proposed that the δ18OL of CAM atmospheric epiphytes 520

could be used to reconstruct δ18Oa. An inherent assumption of this proposal was that the 521

environmental controls on δ18OL— aside from δ18Oa — are known and can be accounted 522

for in the framework of a mechanistic model. The solid agreement between observations 523

of δ18OL and predictions from Equations 1, 4 and 6 appears to satisfy this assumption and 524

shows a fair understanding of the controls on δ18OL in T. usneoides, at least under 525

laboratory conditions. Hygroscopic trichomes can affect interpretations of observed 526

δ18OL but only when leaf water is sample under high h; a condition that is not often 527

satisfied in the field. Observed δ18OL could be predicted well through the non-steady 528

state equations even by using constant values of W and gtot. Therefore, the non-steady-529

state models can potentially be applied more easily. Also, the leaf does not need to be at 530

full isotopic steady state for observed δ18OL to equal δ18OL-M (Equation 4) so only δ18Oa 531

and relative humidity (h) are required as predictive inputs. The poor performance of 532

Equation 4 in the first 4-5 days of experimentation was likely an artifact of using 533

artificially enriched rehydration water to start each experiment. Tissue rehydration using 534

enriched water caused a significant, but transitory change in δ18OL because the 535

consequent increase in gtot ultimately led to a faster coupling with δ18Oa. The inherent 536



21

coupling between precipitation water δ18O and δ18Oa in the field should mean that δ18OL 537

is more tightly coupled to δ18Oa after rehydration. This enhanced understanding about the 538

controls of δ18OL in CAM epiphytes under controlled laboratory conditions will improve 539

the interpretation of δ18OL observations in the more varied natural environment. 540

541

542

Materials and Methods 543

Modeling leaf water isotopes 544

The isotope ratio of leaf water (δ18OL) for a CAM epiphyte such as T. usneoides can be 545

explained by the following non-steady state equation (Flanagan et al., 1991; Farquhar and 546

Lloyd, 1993; Farquhar and Cernusak, 2005; see Helliker and Griffiths 2007 for a full 547

derivation): 548

549

RL = α * RpJp − d(WRL )

dt

⎛ ⎝ ⎜

⎞ ⎠ ⎟

α K

gtotwi

+ Ra

wa

wi

⎡

⎣ ⎢

⎤

⎦ ⎥ (1). 550

551

where R is the molar isotope ratio the subscripts L and P represent leaf and precipitation 552

(rainfall or dewfall) water, respectively. JP is the flux of water into the plant, W is leaf 553

water volume (mol m-2), gtot is total conductance (stomatal and boundary layer; mol m-2 s-554 1) and wi and wa represent the mole fraction of water vapor in the substomatal cavity and 555

ambient air, respectively. the mole fraction of water vapor, w, is equal to the partial vapor 556

pressure e divided by total atmospheric pressure. The temperature dependent equilibrium 557

fractionation factor is represented by α*, and αK is the balance of the ratio of diffusivities 558

of light to heavy water molecules through the stomata and through the leaf boundary 559

layer (Merlivat, 1978; Luz et al., 2009). Between rewetting events, when RPJP is zero, RL 560

can be explained by: 561

562

RL = α * − d(WRL )

dt

⎛ ⎝ ⎜

⎞ ⎠ ⎟

α K

gwi

+ Ra

wa

wi

⎡

⎣ ⎢

⎤

⎦ ⎥ (2) 563

564



22

In between rain events d(WRL )

dt= −RE E , where E is the transpiration flux from the leaves 565

(E = g (wi –wa)). If we define h = wa/wi, then equation 2 can be rewritten as: 566

567

RL = α * α K RE (1− h) + Rah[ ] (3) 568

569

Throughout this study, h = ambient relative humidity, so in Equation (3) the effect of h 570

on RL can be seen easily where, as h increases towards unity, Ra assumes a greater control 571

over RL. Under constant conditions and water loss, steady state is achieved and RE in 572

Equation 3 can be replaced by RL arriving at maximum enrichment, RL-M: 573

574

RL−M ≈ Rah1

α *− α K + α K h

(4). 575

Equation 4 is similar to the derivation of maximum enrichment of Farqhar and Gan 576

(2003) and Hartard et al. (2009): 577

578

RL−M ≡ α * hRa

1−α Kα * 1− h( ) (5). 579

Hartard et al. (2009) further developed an analytical solution to the non-steady state 580

equation, referred to throughout as the Hartard & Cuntz solution: 581

RL = RL−M + R0 − RL−M[ ]exp− 1

α K α*(1−h )−1

⎡

⎣ ⎢

⎤

⎦ ⎥

E

Wt

⎧ ⎨ ⎩

⎫ ⎬ ⎭ (6) 582

583

Lastly, when h = 1, RL is equal to the isotope composition of atmospheric water vapor, 584

corrected for the equilibrium fractionation factor, 585

586

RL = Raα * (7). 587

588

In this paper, the majority of model solutions were obtained through iterative solution of 589

Equation (2). The Hartard & Cuntz solution was compared to outputs of equation (2) 590



23

using changing and constant values of gtot and W. Equation (5) was examined as a 591

simplified method of predicting RL for eventual field applications. 592

593

594

Plant Material 595

The study species for this work was the CAM epiphytic Bromeliad Tillandsia usneoides. 596

T. usneoides is a subtropical-to-tropical plant that ranges from coastal Virginia, USA 597

through the tropics to Argentina. The broad distribution is determined by high nocturnal 598

relative humidity (Garth, 1964) and makes it a model organism for the reconstructing 599

δ18Oa. In October 2006 and 2008, plants were obtained from the northernmost point of 600

distribution for T. usneoides in North America, First Landing State Park in Virginia 601

Beach, VA, USA. The plants were subsequently maintained in a high-humidity 602

greenhouse at the University of Pennsylvania until experimentation. Approximately 5 603

days prior to the start of an experiment, clumps of T. usneoides were separated into 604

strands of 10 to 20 ramets (ramets consist of 2 to 5 leaves that were 3 –6 cm in length) 605

each and placed in the growth chamber under a reversed 12h photoperiod (i.e. forcing the 606

plants nocturnal cycle to coincide with the researchers normal diurnal cycle). Prior to this 607

acclimation period all strands were submersed in enriched water (ca. 11 ‰) for 2-4 hours 608

and plants were watered daily thereafter until the start of an experiment. 609

610

Experimental Conditions 611

For experiments, plants were maintained in a climate-controlled chamber coupled to a 612

large volume dew-point generator (the ‘water tower’) that was capable of maintaining a 613

constant δ18Oa for up to 3 weeks (s.d. ± 0.15 ‰). The chamber was fabricated from a 614

cubic 22.2 L acrylic desiccator cabinet (Fisher Scientific) which was back-fitted with 615

tubing inlets and a 75 W, 24 VDC Peltier type thermoelectric air conditioner (INB260-616

24-AA, Watronix Inc., CA, USA). A temperature controller (model CN77322-C2; 617

Omega Engineering, Stamford, CT) signaled a relay to change the DC polarity to either 618

heat or cool the chamber based on the prescribed set point; separate day/night 619

temperature setpoints were controlled by a Grässlin Digi-20E timer (Intermatic, IL, 620

USA). 20 cm long strands of T. usneoides were placed horizontally in the chamber on top 621



24

of a wire mesh shelf. A 400 W metal-halide light source (Sun System 3, Sunlight Supply 622

Inc., Vancouver, WA, USA ) was placed above the chamber; photosynthetically active 623

radiation was 500 μmol m-2 s-1 at plant height. 624

625

The water tower consisted of two 132 cm long, 0.64 cm wall thickness, extruded acrylic 626

cylinders— one inside of the other (outer cylinder ID = 20.3 cm, inner cylinder ID = 15.2 627

cm) — which were sealed off with 1 cm thick Aluminum plates at either end. The effect 628

was to create an outer cavity for a temperature-controlled water jacket that surrounded 629

the inner cylinder. The inner cylinder was filled with 23 L of water which was the source 630

for water vapor in the chamber. The water jacket temperature was controlled by pumping 631

water from a circulating water bath (water circulated between the outer and inner 632

cylinders). Ambient air was pumped (KNF Neuberger Inc., Freiberg, Germany) through 633

the length of the inner cylinder at 2 L min-1 (MKS flow controller, Andover, MA, USA) 634

to create δ18Oa and dewpoint values that were fed to the plant chamber. The temperature 635

of the water tower controlled the ambient vapor pressure (ea) values for the climate 636

controlled chamber and the temperature of the chamber set the saturated vapor pressure 637

(esat) values. Because of the high heat capacity of the water tower and the tight 638

temperature control of the chamber, a fairly constant ea, and therefore h (ea/esat), could be 639

maintained (see supplementary data for picture of chamber/water tower setup and a 640

representative diurnal cycles of h and air temperature). 641

642

Five days prior to each experiment, 30 to 50, 30 cm long strands of T. usneoides were 643

removed from the greenhouse and placed in a growth chamber set on reversed, 12h 644

photoperiod to acclimate the plants. The reversed photoperiod allowed for easy sampling 645

of δ18Oa during the period of T. usneoides water loss (and isotopic exchange). Each 646

experimental run was started (Day 0) by submerging the strands in a tub of enriched 647

water (typically 10 ‰ to 20 ‰) for two hours. This submersion assured hydration of the 648

strands and assured that the initial δ18O of leaf water was considerably different from any 649

chamber-induced δ18OL. The strands were surface-dried and then placed in the 650

experimental chamber. 4 of the strands were sealed in glass vials immediately to sample 651

for initial δ18OL and 4 of the strands were tagged for measurements of relative water 652



25

content and water loss (through loss of mass) during every sampling period. Typically 653

every third day, 3 to 5 strands of T. usneoides were removed from the chamber and 654

placed into glass vials for eventual water extraction and determination of δ18O. 655

656

Isotope sampling and analysis 657

Water vapor was sampled from the chamber exhaust tubing for at 2.5 to 3 hours prior to 658

opening the chamber to sample leaf water in a 200 ml internal volume condenser inserted 659

into a dry ice/ethanol slush (-69 °C). The air exiting the chambers was split and 660

monitored with a rotometer to assure that flow rates of air through the condenser never 661

exceeded 0.5 L m-1 (Helliker et al., 2002). Leaf water was extracted cryogenically 662

(Ehleringer et al., 2000). Water samples (0.5 mL) were equilibrated for at least 48 hours 663

in 3 ml Exetainer® vials (Labco Limited, UK) with 10/90 mixture of CO2/He. 664

Approximately 100 μL of the headspace gas were injected into a GC and carried in a 665

helium air stream to a Delta Plus (Thermo-Finnigan, Germany) isotope ratio mass 666

spectrometer. All molar isotope ratios were expressed in the standard ‘delta’ notation on a 667

per mil basis by δ = (Rsample/Rstandard) * 1000, where R = 18O/16O and the standard was 668

Standard Mean Ocean Water (V-SMOW R = 0.002005). 669

670

Estimating relative water content and conductance to water loss 671

After each sampling period for δ18OL, three to four strands were placed in plastic bags to 672

assure no immediate loss of water from the trichomes and were removed from the 673

chamber. One by one the strands were taken out of the bags and the mass measured. 674

Relative water content was determined in the same manner as Martin and Schmitt (1989). 675

676

RWC = [(FW – DW)/DW] * 100 (8) 677

678

Where FW is the measured fresh weight and DW is the measured dry weight after 3 days 679

in a 60 °C drying oven. Plant conductance to water loss (gtot; mmol m-2 s-1) was 680

determined by: 681

682

gtot = E/(wi – wa) (9) 683



26

684

Plant transpiration (E) was estimated from the measures of fresh weight above, wi and wa 685

were obtained from measurements of chamber h and temperature, pressure was assumed 686

constant at 100 kPa. Leaf area was determined using the fresh weight to leaf area 687

conversions presented in Martin and Schmitt (1989). Based on direct measurements, leaf 688

temperature and air temperature were not different, therefore wi was equal to wsat at 689

chamber temperature. 690

691

Correcting for rehydration and water adsorbed to trichomes 692

The effect that rehydration had on T. usneoides δ18OL was determined by fully 693

submerging 45, 30 cm long strands for several hours to rehydrate. The strands were then 694

placed in the greenhouse and not watered for one-week. After a week the strands were 695

placed in enriched water (11.5 ‰) and 5 strands were removed from the rehydrating 696

water at intervals of 10 min, 20 min, 40 min, 1 hour, 1.5 hours, 2.5 hours, 5 hours, 7 697

hours and 24 hours. 698

699

In prior work we assumed water was adsorbed to trichomes at the time of sampling and 700

that this water was inevitably extracted with tissue water when bulk leaf water was 701

cryogenically extracted. In our prior attempt to correct for the isotope ratio of this water 702

to obtain tissue water (δtissue) several untested assumptions were made (Helliker and 703

Griffiths 2007): (i) that water adsorption by the trichomes from the atmosphere did not 704

fractionate and (ii) the humidity-driven proportion of adsorbed water on trichomes 705

(ƒtrichome) could be explained by a simple parameterization based on two measured 706

humidities (Martin and Schmitt 1989). The first assumption was examined by placing 707

several clumps of dead T. usneoides (about 2g of material, killed via desiccation at 60 ° 708

C) into the growth chamber for two days and then extracting the water adsorbed onto the 709

dead tissue for isotope analysis. This yielded an apparent fractionation factor (εtrichome) 710

between atmospheric water vapor and the water adsorbed to trichomes of 7.4 ± 0.1 ‰. 711

712

δ18Otrichome =δ18Oa + εtrichome (10) 713

714



27

The second assumption was examined by building an empirical relationship between the 715

mass of dead T. usneoides and h ranging from 0.34 to saturation. This was accomplished 716

by placing dead T. usneoides of a variety of masses into nylon pouches (made from 717

pantyhose, L’eggs®, Hanes, Winston-Salem, NC) and comparing oven-dried mass to the 718

mass after placement in a growth chamber at a known humidity for 24 hours. Empty 719

nylon pouches were also put through this exercise to correct for any water adsorption 720

onto the nylon (the maximum for nylon adsorption was less than 6 % of dry nylon mass 721

at saturating humidity). The relationship between trichome water content (ƒtrichome) and h 722

was used to parameterize the Hailwood and Horrobin (Hailwood and Horrobin, 1946) 723

model for predicting water adsorption of solid polymer solutions as a function of h: 724

725

f t = k1

k2h

1+ k2h+ k3h

1− k3h

⎡

⎣ ⎢

⎤

⎦ ⎥ ×100 (11) 726

727

Where k1 is the number of binding sites, k2 is the affinity of the material for adsorption, k3 728

is the water activity. The values for k1, k2 and k3 were changed so that the adsorption 729

curve fit the data over a range of h. These values were k1 = 0.02, k2 = 10 and k3 = 0.95. 730

731

Finally, modeled values of δ18OL from Equations (3) (4) (5) and (6) were corrected for 732

trichome adsorbed water as a proportion of RWC by: 733

734

δ18Opredicted =δ18OEqns3,4,5,6 • 1− f t

RWC

⎛ ⎝ ⎜

⎞ ⎠ ⎟ +δ18Otrichome • f t

RWC

⎛ ⎝ ⎜

⎞ ⎠ ⎟ (12) 735

736

Acknowledgements 737

M. Suplick, F. Letterio and J. Andrews-Labinski offered a great deal of assistance in 738

constructing the chamber and water tower. M. Vergari, C. Hulshof, B. Waring, C. 739

Policastro and A. Dinerstein helped with water extractions. D. Vann was instrumental in 740

performing isotopic analyses. J. Lloyd-Helliker assisted in obtaining sample material. I 741

am grateful to First Landing State Park, Virginia State Parks and in particular Erik 742

Molleen for permission and assistance in obtaining sample material of T. usneoides. Two 743



28

anonymous reviewers gave extremely helpful comments to improve this manuscript. 744

745



29

References 746

747

Anderson W, Bernasconi S, McKenzie J (1998) Oxygen and carbon isotopic record of 748 climatic variability in tree ring cellulose (Picea abies): An example from central 749 Switzerland (1913-1995). Journal of Geophysical Research 103: 31,625-631,636 750

Andrade JL (2003) Dew deposition on epiphytic bromeliad leaves: an important event in 751 a Mexican tropical dry deciduous forest. Journal of Tropical Ecology 19: 479-488 752

Barbour MM, Farquhar GD (2000) Relative humidity- and ABA-induced variation in 753 carbon and oxygen isotope ratios of cotton leaves. Plant Cell and Environment 754 23: 473-485 755

Barbour MM, Fischer RA, Sayre KD, Farquhar GD (2000) Oxygen isotope ratio of 756 leaf and grain material correlates with stomatal conductance and grain yield in 757 irrigated wheat. Australian Journal of Plant Physiology 27: 625-637 758

Benzing DH (1970) Foliar Permeability and Absorption of Minerals and Organic 759 Nitrogen by Certain Tank Bromeliads. Botanical Gazette 131: 23-& 760

Benzing DH, Henderson K, Kessel B, Sulak J (1976) The absorptive capacities of 761 bromeliad trichomes. American Journal of Botany 63: 1009-1014 762

Cernusak LA, Farquhar GD, Wong SC, Stuart-Williams H (2004) Measurement and 763 interpretation of the oxygen isotope composition of carbon dioxide respired by 764 leaves in the dark. Plant Physiology in press 765

Cernusak LA, Pate JS, Farquhar GD (2002) Diurnal variation in the stable isotope 766 composition of water and dry matter in fruiting Lupinus angusifolius under field 767 conditions. Plant, Cell and Environment 25: 893-907 768

Ciais P, Tans PP, Denning AS, Francey RJ, Trolier M, Meijer HAJ, White JWC, 769 Berry JA, Randall DA, Collatz GJ, Sellers PJ, Monfray P, Heimann M 770 (1997) A three-dimensional synthesis study of δ18O in atmospheric CO2 2. 771 Simulations with the TM2 transport model. 102: 5873-5883 772

Cuntz M, Ciais P, Hoffmann G, Allison CE, Francey RJ, Knorr W, Tans PP, White 773 JWC, Levin I (2003) A comprehensive global three-dimensional model of ∂18O 774 in atmospheric CO2: 2. Mapping the atmospheric signal. Journal of Geophysical 775 Research 108: 4528 776

Cuntz M, Ogee J, Farquhar GD, Peylin P, Cernusak LA (2007) Modelling advection 777 and diffusion of water isotopologues in leaves. Plant, Cell & Environment 30: 778 829-909 779

Dominguez E, Heredia A (1999) Water hydration in cutinized cell walls: a physico-780 chemical analysis. Biochimica Et Biophysica Acta-General Subjects 1426: 168-781 176 782

Ehleringer JR, Roden JR, Dawson TE (2000) Assessing ecosystem-level water 783 relations through stable isotope ratio analyses. In OE Sala, RB Jackson, HA 784 Mooney, RW Howarth, eds, Methods in ecosystem science. Springer, New York, 785 pp 181-198 786

Epstein S, Thompson P, Yapp CJ (1977) Oxygen and hydrogen isotopic ratios in plant 787 cellulose. Science 198: 1209-1215 788

Farquhar GD, Cernusak LA (2005) On the isotopic composition of leaf water in the 789 non-steady state. Functional Plant Biology 32: 293-303 790



30

Farquhar GD, Gan KS (2003) On the progressive enrichment of the oxygen isotope 791 composition of water along a leaf. Plant Cell & Environment 26: 1579-1597 792

Farquhar GD, Lloyd J (1993) Carbon and oxygen isotope effects in the exchange of 793 carbon dioxide between terrestrial plants and the atmosphere. In JR Ehleringer, 794 AE Hall, GD Farquhar, eds, Stable Isotopes and Plant Carbon/Water Relations. 795 Academic Press, San Diego, pp 47-70 796

Ferrio JP, Voltas J (2005) Carbon and oxygen isotope ratios in wood constituents of 797 Pinus halepensis as indicators of precipitation, temperature and vapour pressure 798 deficit. Tellus B 57: 164-173 799

Flanagan LB, Comstock JP, Ehleringer JR (1991) Comparison of modeled and 800 observed environmental influences on the stable oxygen and hydrogen isotope 801 composition of leaf water in Phaseolus vulgaris L. Plant Physiol 96: 588-596 802

Garth RE (1964) Ecology of Spanish Moss ( Tillandsia Usneoides ) - Its Growth + 803 Distribution. Ecology 45: 470-& 804

Gat JR (1996) Oxygen and hydrogen isotopes in the hydrologic cycle. Annu Rev Earth 805 Planet Sci 24: 225-262 806

Griffiths H, Luttge U, Stimmel KH, Crook CE, Griffiths NM, Smith JAC (1986) 807 Comparative ecophysiology of CAM and C3 Bromeliads. 3. Environmental 808 influences on CO2 assimilation and transpiration. Plant Cell and Environment 9: 809 385-393 810

Griffiths H, Smith JAC (1983) Photosynthetic pathways in the Bromeliaceae of 811 Trinidad: relations between life-forms, habitat preference and the occurrence of 812 CAM. Oecologia 60: 176-184 813

Guy RD, Fogel ML, Berry JA (1993) Photosynthetic Fractionation of the Stable 814 Isotopes of Oxygen and Carbon. Plant Physiol 101: 37-47 815

Hailwood AJ, Horrobin S (1946) Absorption of water by polymers: analysis in terms of 816 a simple model. Transactions of the Faraday Society 42B: 84-101 817

Hartard B, Cuntz M, Maguas C, Lakatos M (2009) Water isotopes in desiccating 818 lichens. Planta 231: 179-193 819

Helliker BR, Berry J, Bakwin PS, Davis KJ, Denning AS, Ehleringer JR, Miller JB, 820 Butler MP, Ricciuto DM (2005) Regional-scale estimates of forest CO2 and 821 isotope flux based on monthly CO2 budgets of the atmospheric boundary layer. In 822 H Griffiths, PG Jarvis, eds, The Carbon Balance of Forest Biomes Taylor & 823 Francis, New York, pp 77-92 824

Helliker BR, Griffiths H (2007) Toward a plant-based proxy for the isotope ratio of 825 atmospheric water vapor. Global Change Biology 13: 723-733 826

Helliker BR, Noone D (2009) Novel approaches to spatial monitoring of water vapor 827 isotope ratios. . In GJ Bowen, JB West, TE Dawson, eds, Isocapes: Understanding 828 movement, pattern, and process on Earth through isotope mapping, 829

Helliker BR, Richter SL (2008) Subtropical to boreal convergence of tree-leaf 830 temperatures. Nature 454: 511-514 doi: 510.1038/nature07031 831

Hoffmann G, Cuntz M, Weber C, Ciais P, Friedlingstein P, Heimann M, Jouzel J, 832 Kaduk J, Maier-Reimer E, Seibt U, Six K (2004) A model of the earth's dole 833 effect. Global Biochemical Cycles 18 834

Kahmen A, Simonin K, Tu KP, Merchant A, Callister A, Siegwolf R, Dawson TE, 835 Arndt SK (2008) Effects of environmental parameters, leaf physiological 836



31

properties and leaf water relations on leaf water delta O-18 enrichment in 837 different Eucalyptus species. Plant Cell and Environment 31: 738-751 838

Lai CT, Ometto J, Berry JA, Martinelli LA, Domingues TF, Ehleringer JR (2008) 839 Life form-specific variations in leaf water oxygen-18 enrichment in Amazonian 840 vegetation. Oecologia 157: 197-210 841

Lange OL, Green TGA, Heber U (2001) Hydration-dependent photosynthetic 842 production of lichens: what do laboratory studies tell us about field performance? 843 Journal of Experimental Botany 52: 363 844

Luz B, Barkan E, Bender ML, Thiemens MH, Boering KA (1999) Triple-isotope 845 composition of atmospheric oxygen as a tracer of biosphere productivity. Nature 846 400: 547-550 847

Luz B, Barkan E, Yam R, Shemesh A (2009) Fractionation of oxygen and hydrogen 848 isotopes in evaporating water. Geochimica Et Cosmochimica Acta 73: 6697-6703 849

Martin CE (1994) Physiological Ecology of the Bromeliaceae. Botanical Review 60: 1-850 82 851

Martin CE, Lin TC, Lin KC, Hsu CC, Chiou W (2004) Causes and consequences of 852 high osmotic potentials in epiphytic higher plants. Journal of Plant Physiology 853 161: 1119-1124 854

Martin CE, Schmitt AK (1989) Unusual water relations in the CAM atmospheric 855 epiphyte Tillandsia usneoides L. (Bromeliaceae). Bot Gaz 150: 1-8 856

Medina E, Troughton JH (1974) Dark CO2 fixation and the carbon isotope ratio in 857 Bromeliaceae. Plant Science Letters 2: 357-362 858

Merlivat L (1978) Molecular diffusivities of H216O, HD16O and H2

18O in gases. J Chem 859 Phys 69: 2864-2871 860

Mez C (1904) Physiologische Bromeliaceen-Studien. I. Die wasser-Oekonomie der 861 extrem atmosphaerischen Tillandsien. Jahrbuecher fuer wissenschaftliched 862 Botanik 40: 157-229 863

Miller DL, Mora CI, Grissino-Mayer HD, Mock CJ, Uhle ME, Sharp Z (2006) Tree-864 ring isotope records of tropical cyclone activity. Proceedings of the National 865 Academy of Sciences of the United States of America 103: 14294-14297 866

Nowak EJ, Martin CE (1997) Physiological and anatomical responses to water deficits 867 in the CAM epiphyte Tillandsia ionantha (Bromeliaceae). International Journal of 868 Plant Sciences 158: 818-826 869

Ogee J, Cuntz M, P P, Bariac T (2007) Non-steady-state, non-uniform transpiration rate 870 and leaf anatomy effects on the progressive stable isotope enrichment of leaf 871 water along monocot leaves. Plant, Cell and Environment 30: 367-387 872

Ogee J, Peylin P, Cuntz M, Bariac T, Brunet Y, Berbigier P, Richard P, Ciais P 873 (2004) Partitioning net ecosystem carbon exchange into net assimilation and 874 respiration with canopy-scale isotopic measurements: An error propagation 875 analysis with (CO2)-C-13 and (COO)-O-18 data. Global Biogeochemical Cycles 876 18 877

Ohrui T, Nobira H, Sakata Y, Taji T, Yamamoto C, Nishida K, Yamakawa T, 878 Sasuga Y, Yaguchi Y, Takenaga H, Tanaka S (2007) Foliar trichome- and 879 aquaporin-aided water uptake in a drought-resistant epiphyte Tillandsia ionantha 880 Planchon. Planta 227: 47-56 881



32

Poussart PF, Schrag DP (2005) Seasonally resolved stable isotope chronologies from 882 northern Thailand deciduous trees. Earth and Planetary Science Letters 235: 752-883 765 884

Reyes-Garcia C, Mejia-Chang M, Jones GD, Griffiths H (2008) Water vapour 885 isotopic exchange by epiphytic bromeliads in tropical dry forests reflects niche 886 differentiation and climatic signals Plant, Cell and Environment 31: 828-841 887

Roden JS, Ehleringer JR (2000) Hydrogen and oxygen isotope ratios of tree ring 888 cellulose for field-grown riparian trees. Oecologia 123: 481-489 889

Smith JAC, Griffiths H, Luttge U (1986) Comparative ecophysiology of CAM and C3 890 Bromeliads. 1. The ecology of the Bromeliaceae in Trinidad. Plant Cell and 891 Environment 9: 359-376 892

Smith JAC, Griffiths H, Luttge U, Crook CE, Griffiths NM, Stimmel KH (1986) 893 Comparative ecophysiology of CAM and C3 Bromeliads. 3. Plant water relations. 894 Plant Cell and Environment 9: 395-410 895

Styles JM, Lloyd J, Zolotoukhine D, Lawton KA, Tchebakova N, Francey RJ, 896 Arneth A, Salamakho D, Kolle O, Schulze E-D (2002) Estimates of regional 897 surface carbon dioxide exchange and carbon and oxygen isotope discrimination 898 during photosynthesis from concentration profiles in the atmospheric boundry 899 layer. Tellus 54: 768-783 900

Switsur R, Waterhouse J (1998) Stable isotopes in tree ring cellulose. In H Griffiths, ed, 901 Stable Isotopes: integration of biological, ecological and geochemical processes. 902 BIOS Scientific Publishers Ltd, Oxford, pp 303-321 903

Winter K, Smith JAC, eds (1996) Crassulacean Acid Metabolism: Biochemistry, 904 Ecophysiology, and Evolution. , Vol 114. Springer Verlag, Berlin 905

Yakir D, Wang XF (1996) Fluxes of CO2 and water between terrestrial vegetation and 906 the atmosphere estimated from isotope measurements. Nature 380: 515-517 907

908 909

910



33

911

Table I. Representative chamber measurements of chamber air temperature, h, and δ18Oa over a 16 day 912

experimental period. Note that Day 0 was characterized by labeling the plants by soaking them in enriched 913

water for several hours and allowing both the chamber and water tower to reach the desired temperature 914

and h. Day 1 represents the first full nocturnal period that plants were in the chamber. 915

916

Experimental day

Chamber air temperature (°C)

h

δ18Oa

(‰) 1 27.5 ± 0.03 0.96 ± 0.01 2 27.5 ± 0.06 0.97 ± 0.01 -16.8 3 27.6 ± 0.20 0.96 ± 0.02 4 27.5 ± 0.04 0.98 ± 0.01 -17.0 5 27.5 ± 0.09 0.96 ± 0.01 6 27.5 ± 0.12 0.96 ± 0.01 7 27.4 ± 0.19 0.97 ± 0.01 8 27.5 ± 0.14 0.98 ± 0.01 -17.2 9 27.5 ± 0.13 0.98 ± 0.01 -17.1 10 27.5 ± 0.07 0.98 ± 0.01 11 27.4 ± 0.17 0.97 ± 0.01 -17.2 12 27.3 ± 0.19 0.93 ± 0.01 13 27.3 ± 0.20 0.94 ± 0.01 14 27.5 ± 0.14 0.96 ± 0.02 -17.0 15 27.5 ± 0.04 0.98 ± 0.01 16 27.5 ± 0.00 0.98 ± 0.01 -17.2 Multi-day mean 27.5 0.967 -17.1 Standard deviation 0.08 0.015 0.15

917

918



34

919

Table II. Mean values of h, δ18Oa, and the number of days that T. usneoides strands were in the chamber for 920

each experimental run. 921

922

h δ18Oa Days in

Chamber 0.98 -2.3 15 0.97 -17.1 16 0.96 -15.0 18 0.96 -16.8, -7.3, 2.5 21 0.96 -16.4 6 0.94 -16.4 6 0.93 -16.6 20 0.93 -2.9 10 0.92 -16.4 3 0.92 -15.0 9 0.92 -16.4 5 0.91 -16.5 15 0.88 -16.4 7 0.86 -16.4 7 0.84 -15.0 15 0.56 -12.4 9

923 924

925



35

926 Table III. The representative effect of rehydration on δ18OL through time during the chamber experiments. 927 On Day 10, one population of T. usneoides strands within the chamber were watered, while one population 928 remained dry. This scheme was repeated in 1/3 of the high h experiments in Table II with similar results. 929

Wetted plants δ18OL (‰)

Non-wetted plants δ18OL (‰)

Day 0 -3.0 ± 0.4 Day 7 -6.8 ± 0.6 Day 10 -7.4 ± 0.2 -6.8 ± 0.1 ns Day 10, 1 hour after rehydration 2.5 ± 0.6 Day 14 -6.1 ± 0.2 -6.3 ± 0.6 ns Day 21 -6.5 ± 0.4 -7.1 ± 0.7 ns

930 931



36

932

933

934

935

936

937

938

939

940

941

942

943

944



37

945

946

947



38

948

949

950

951

952

953

954



39

955

956

957

958

959

960 961



40

962 963 964 965

966 967 968 969

970



41

971

972



42

973

974



43

975

976 977



1 Running head: leaf-water δ18O in atmospheric CAM epiphytes · This shift lead to morphological...

Documents

Transcript of 1 Running head: leaf-water δ18O in atmospheric CAM epiphytes · This shift lead to morphological...