bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Throwing shade: Physiological responses to light explain
competition and facilitation in a tree diversity experiment
Shan Kothari,1* Rebecca Montgomery,2 Jeannine Cavender-Bares1,3
1
Department of Plant and Microbial Biology, University of Minnesota, 140 Gortner Laboratory, 1479
Gortner Ave., St. Paul, MN, 55108
2
Department of Forest Resources, University of Minnesota, St. Paul, MN, 55108
3
Department of Ecology, Evolution, and Behavior, University of Minnesota, St. Paul, MN, 55108
*corresponding author
Email: kotha020@umn.edu
Phone: (734) 502-2817
Author contributions: SK conceived the project with JCB, conducted physiological measurements,
analyzed and interpreted the data with JCB and RM, and wrote the first draft of the paper. JCB and RM
designed the FAB experiment, coordinated annual biomass surveys, and contributed to subsequent
revisions of the paper.
Summary
Interspecific facilitation is often invoked in explanations of biodiversity-ecosystem function
22
relationships in plant communities, but it is seldom clear how it occurs. Physiological
23
experiments show that excess light causes stress that may depress long-term carbon assimilation.
24
If shading by a plant’s neighbors reduces light stress, it may facilitate that plant’s growth. On the
25
other hand, when light is a limiting factor for growth, shading will often have a net negative,
26
competitive effect.
27
In a tree diversity experiment, we measured growth rates and photosynthetic physiology of
28
broadleaf tree species across a gradient of light availability imposed by their neighbors. At the
29
extremes, trees experienced nearly full sun (monoculture), or were shaded by nearby fast-growing
30
conifers (shaded biculture).
31
Although most species had lower growth with larger neighbors, implying a net competitive effect,
32
the two most shade-tolerant species (Tilia americana and Acer negundo) had positive responses
33
to neighbor size. Compared to the others, these two species were especially susceptible to
34
photoinhibition (reduced dark-acclimated Fv/Fm) in full sun. While most species had lower
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
35
assimilation rates in shaded bicultures, T. americana had carbon assimilation rates up to 25%
36
higher. T. americana also dropped its leaves 3-4 weeks later in the shaded biculture, extending its
37
growing season. We conclude that although large neighbors can cause light limitation in shade-
38
intolerant species, they can also increase growth through abiotic stress amelioration in shade-
39
tolerant species.
40
Both positive and negative species interactions in our experiment can be explained by the
41
photosynthetic responses of trees to the light environment created by their neighbors. We show
42
that physiological measurements can deepen our understanding of the species interactions that
43
underlie biodiversity-ecosystem function relationships.
44
45
Keywords: Photoprotection; photoinhibition; photosynthetic light-response curves; competition;
46
facilitation; tree diversity; biodiversity and ecosystem function
47
48
49
50
51
52
53
54
55
56
57
58
59
60
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
61
Introduction
62
The sun’s light powers photosynthesis, the chain of reactions through which plants turn carbon
63
dioxide into the living matter they use to grow and reproduce. A dominant view in plant ecology is that
64
light is often a limiting resource, and competition for scarce light shapes the fate of plant species (Braun-
65
Blanquet 1932; Canham et al. 1994; Monsi & Saeki 2005; Dybzinski & Tilman 2007). While this is often
66
true, physiologists have also shown that, as other factors begin to limit photosynthesis, plants often absorb
67
more light than they are able to use. Beyond an initial linear rise with increasing light availability,
68
photosynthesis begins to saturate, such that additional light contributes little to the plant’s carbon gain.
69
Moreover, external factors that limit photosynthesis—including water limitation, cold temperatures, or
70
nutrient-poor conditions—can exacerbate this excess of light (Fig. 1; Long et al. 1991; Demmig-Adams
71
& Adams 1992; Demmig-Adams & Adams 2018).
72
Light is thus both an essential resource and a potential stressor. A chronic excess of light can
73
cause lasting oxidative damage to plant cells, especially Photosystem II (PSII), and this damage reduces
74
the efficiency of photosynthesis. Plants can avoid damage through mechanisms of photoprotection. These
75
include biochemical pathways that shunt off excess light safely as heat—most notably, the xanthophyll
76
cycle. They also include structural means of avoiding excess light interception, such as self-shading,
77
reflective leaves, or steep leaf angles (Lovelock et al. 1992; Pearcy et al. 2005; Kothari et al. 2018).
78
However, once induced, many forms of biochemical photoprotection are slow to reverse. Even when light
79
conditions are no longer stressful, these mechanisms may continue to suppress photosynthesis for some
80
time (Demmig-Adams and Adams 2006; Kromdijk et al. 2016). Similarly, structural mechanisms like
81
self-shading may be costly under low ambient light. In general, photoprotective ability often increases
82
with exposure to conditions that put plants at risk of damage in the habitat of origin, including high light
83
(Montgomery et al. 2008) and cold or dry conditions (Cavender-Bares 2007; Savage et al. 2009; Wujeska
84
et al. 2013; Ramírez-Valiente et al. 2013). In these environments, photoprotection may be necessary to
85
prevent even more costly damage.
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
86
Depending on a plant’s level of investment in photoprotection, these mechanisms may lack the
87
capacity to protect against prolonged exposure to excess light, which can reduce net carbon assimilation
88
rates (Egerton et al. 2000; Howell et al. 2002; Murchie & Niyogi 2011). Consistent with Long et al.
89
(1991), we use the term “photoinhibition” to describe a potential drop in carbon assimilation caused by
90
either photodamage or slow-reversing biochemical photoprotection. Aside from its direct effects, light can
91
also alter other microclimatic factors, like temperature, soil moisture, and vapor pressure deficit (VPD),
92
which can compound photoinhibition (Björkman & Powles 1984). For example, high VPD can lower
93
carbon assimilation by causing stomatal closure, which reduces the potential to use light for
94
photosynthesis; this effect, in turn, may increase the risk of photoinhibition, reducing photosynthetic
95
capacity further. Competition may also may increase a plant’s risk of photoinhibition if it reduces its
96
access to resources in a way that limits photosynthesis. Thus, photodamage and photoprotective
97
investments are likely to affect plants’ fitness and distribution across environmental gradients (Külheim et
98
al. 2002).
99
Given that light can cause stress, it may be possible for one plant species to facilitate another
100
through shading. Whether shading by one species has a positive or negative overall effect on another
101
depends on the balance of two effects:
102
103
104
1. Potential reduction of carbon gain through declines in light available for photosynthesis at
any given instant; and,
2. Avoidance of chronic stress, which may increase photosynthetic rates at any given light level.
105
Monteith et al. (1977) defined light-use efficiency (LUE) as net photosynthesis divided by the amount of
106
photosynthetically active radiation (PAR) absorbed. In terms of this formalism, an individual plant will
107
have higher photosynthetic rates in the shade when the proportional benefit of (2) is greater than the
108
proportional loss caused by (1). Changes in instantaneous irradiance alter the position along a
109
photosynthetic light-response curve; changes in LUE alter the parameters of the curve such that plants
110
assimilate more or less carbon at a given irradiance (Fig. 1).
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
111
Here, we test whether the shade created by trees can facilitate the carbon assimilation and growth
112
of neighboring trees in a biodiversity experiment. Experimental and observational research alike has
113
shown that biodiversity often promotes ecosystem functions such as productivity (Tilman et al. 2014). In
114
forests, this trend has been shown through both experimental (Potvin & Gotelli 2008; Tobner et al. 2016;
115
Grossman et al. 2017; Williams et al. 2017; Huang et al. 2018; Zemp et al. 2019) and observational
116
(Gamfeldt et al. 2013; Liang et al. 2016; Oehri et al. 2017) data, across temperate (Tobner et al. 2016;
117
Grossman et al. 2017; Williams et al. 2017), subtropical (Huang et al. 2018), and tropical (Potvin &
118
Gotelli 2008; Zemp et al. 2019) forests. These positive relationships are robust to variation in climate,
119
successional stage, and other factors (reviewed in Grossman et al. 2018; Ammer 2018).
120
One of the many factors that could contribute to positive biodiversity-productivity relationships is
121
interspecific facilitation through reduction of abiotic stress (Wright et al. 2017). Some research has
122
marshaled indirect evidence that facilitative interactions promote these relationships by linking patterns of
123
productivity to species traits (Fichtner et al. 2017). Nevertheless, the physiological mechanisms that
124
underlie these effects are poorly understood, despite their potential relevance for projecting species
125
interactions under environmental change.
126
In this paper, we examined the interactions among tree species in the Forests and Biodiversity
127
(FAB) experiment at Cedar Creek Ecosystem Science Reserve (East Bethel, MN). We ask: Do certain
128
species aid the growth of others by shielding them from stresses caused by high irradiance? In particular,
129
given that conifer species had higher initial growth rates (Table 1), we focused on whether they could
130
facilitate the growth of broadleaf species through shading.
131
We addressed this question by measuring the physiology of carbon gain among broadleaf species
132
across neighborhood environments. We examined whether photosynthetic physiology might explain
133
patterns in aboveground woody growth. If shading by neighbors does have a net positive effect for some
134
species, we would expect larger neighbors to cause focal individuals to have higher growth and carbon
135
assimilation rates. We also posed the hypothesis that trees exposed to high light would use structural and
136
biochemical photoprotective mechanisms more than those in lower light environments. If true, this
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
137
finding would imply that trees growing in high light must invest in photoprotection to avoid damage,
138
reinforcing our proposition that light stress influences the interactions within these communities.
139
140
Methods
141
Experimental design
142
The Forests and Biodiversity (FAB) experiment began in 2013 and comprises 142 4 × 4 m plots,
143
each planted with 64 trees in a 0.5 × 0.5 m grid. These plots form three blocks, each with 49 plots
144
arranged in a 7 × 7 square. Neighboring plots within blocks are immediately adjacent to one another.
145
Each plot is planted with 1, 2, 5, or 12 species from a pool of 12 total species (Table 1). These species
146
include four evergreen conifers with needle-like leaves (two distinct families) and eight deciduous,
147
broadleaf angiosperms (four distinct families). Except for five-species plots, each distinct species
148
composition has a replicate in each block. All species in a plot have approximately equal frequency, and
149
the placement of species within a plot is random. Further details about the experimental design can be
150
found in Grossman et al. (2017).
151
The mean relative growth rate in monoculture during the first three years varied dramatically
152
among species, being highest in conifers and B. papyrifera (Table 1). During these first three years,
153
Grossman et al. (2017) found that diverse plots overyield in productivity—their growth is greater than a
154
weighted average of the constituent species’ growth in monoculture. Using the partition of Loreau &
155
Hector (2001), most of this overyielding was due to complementarity effects, which are often interpreted
156
as a result of synergistic interactions among species, rather than selection effects.
157
158
159
Tree growth
We surveyed tree growth in late fall of each year. For all living trees, we measured diameter and
160
height to the tallest leader. We measured basal diameter (5 cm from the ground) for trees less than 137 cm
161
tall; and diameter at breast height for trees more than 137 cm tall. For a subset of individuals, we
162
measured both diameters. To predict basal diameter when it was not measured, we used linear
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
163
relationships predicting basal diameter from the height of the same individual in the same year and the
164
basal diameter in the previous year.
165
We estimated woody stem volume as 𝑉 = 𝜋𝑟 2 ℎ, where h is height and r is the basal radius (half
166
the basal diameter). This assumes each tree stem is a cylinder, an assumption that has been used and
167
justified in other tree diversity studies (Tobner et al. 2016; Williams et al. 2017), in the absence of
168
system-specific allometric equations. We then calculated the yearly relative growth rate (RGR) between
169
2016 and 2018 as 𝑅𝐺𝑅 = (ln 𝑉2018 − ln 𝑉2016 )/2 (Hoffman & Poorter 2002). We chose 2016 as a
170
171
starting point because no replanting of dead trees occurred after that point.
For each tree, we also calculated the average stem volume of its eight neighbors (in cardinal and
172
intercardinal directions) in 2018 as a proxy for the intensity of aboveground interactions. We assessed
173
neighbor effects on focal individual growth using ordinary least-squares (OLS) regression. When a
174
neighbor was missing, either due to mortality or because the focal individual was on the edge of the
175
experiment, we assigned that neighbor a volume of zero.
176
177
178
Photosynthetic physiology
We measured dark- and light-acclimated chlorophyll fluorescence parameters and reflectance
179
spectra across all eight broadleaf species. We also measured photosynthetic light-response curves across
180
four focal species (A. rubrum, B. papyrifera, Q. ellipsoidalis, and T. americana).
181
For measurements of physiology, we focused on three kinds of plots: (1) monocultures of
182
broadleaf species (“monoculture”); (2) bicultures comprising one broadleaf and one conifer species
183
(“shaded biculture”); and (3) twelve-species plots (“twelve-species”) (see Fig. S1 for sample images). For
184
each species in each treatment, we measured physiological parameters on six individuals—two in each of
185
three plots. Neither A. negundo nor Q. macrocarpa was planted with a conifer in any biculture plots. We
186
used B. papyrifera as a shaded biculture partner for Q. macrocarpa because it is a fast-growing species
187
that draws down light availability. We omitted the shaded biculture treatment for A. negundo.
188
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
189
190
Photosynthetic light-response curves
We measured photosynthetic-light response curves from the four focal species during July 2018
191
using an LI-6400 gas exchange system (LI-COR BioSciences, Lincoln, NE) with a 6400-40 Leaf
192
Chamber Fluorometer head. From each tree (n = 72 total), we selected a fully expanded upper leaf with
193
no visible sign of disease or herbivory. We noted the angle of each leaf relative to horizontal to the
194
nearest 15˚ before beginning measurements. Each curve had nine steps in descending order of brightness:
195
2000, 1500, 1000, 500, 200, 100, 50, 20, and 0 µmol m-2 s -1. We also measured relative electron transport
196
rate (ETR) using chlorophyll fluorescence at each light level. We maintained favorable conditions inside
197
the chamber during each measurement. Following each curve, we removed the leaf to measure leaf mass
198
per area (LMA). Further details about light-response curve procedures and ETR calculations can be found
199
in the Supplemental Methods. Finally, we fitted a non-rectangular hyperbolic model (Lobo et al. 2013) to
200
each light-response curve using R code written by Nick Tomeo, available at:
201
https://github.com/Tomeopaste/AQ_curves.
202
We aimed to determine how carbon assimilation rates vary with the ambient light level, which we
203
define as the photosynthetic photon flux density (PPFD) one might measure in an open field adjacent to
204
the experiment. For most leaves we measured, in situ light availability is lower than the ambient light
205
level, mainly because each tree is shaded by its neighbors. To estimate each tree’s carbon assimilation as
206
a function of ambient light, we have to transform its measured photosynthetic light-response curve along
207
the x-axis according to its light availability.
208
For example, if the topmost leaves of a particular tree only receive 50% of ambient light, we
209
would stretch its light-response curve two-fold along the x-axis. For this tree, the realized assimilation
210
rate we estimate at an ambient (or open-field) PPFD of 2000 µmol photons m-2 s -1 would be the
211
assimilation rate at an leaf-level PPFD of 1000 µmol m-2 s -1. The exact scaling factor is estimated for
212
each individual as RLA × cos(θ), where RLA is relative light availability (see Light availability) and θ is
213
the leaf angle. (Taking the cosine of θ approximately corrects for the reduction in horizontally projected
214
leaf area in steeply inclined leaves when the sun is directly overhead, as simplified from Ehleringer &
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
215
Werk [1986].) We used this procedure on each fitted curve in order to compare across treatments in a way
216
that accounts for the variable light intensity each individual receives. We refer to photosynthetic light-
217
response curves that describe a function of the leaf chamber PPFD as chamber light-response curves, and
218
the rescaled light-response curves as a function of ambient PPFD as ambient light-response curves.
219
Finally, we used ambient light-response curves to estimate carbon assimilation by top leaves
220
throughout July. We used an hourly averaged time series of ambient solar radiation from Carlos Avery
221
Wildlife Management Area in Columbus, MN, 13.6 km away from the study site. Following Carruthers et
222
al. (2001), we assumed a conversion of 0.217 μmol photons m-2 s-1 of PAR per Watt m-2 of solar radiation.
223
By using these data as inputs to the ambient light-response curves, we could estimate integrated carbon
224
assimilation across the month of July. This procedure assumes that the photosynthetic response to light
225
remains constant, and is not affected by factors like leaf temperature and stomatal closure. This
226
assumption is unrealistic, but we nevertheless consider it a useful way of estimating the consequences of
227
realistic fluctuations in the light environment. We compared treatments within species using ANOVA.
228
Throughout the paper, we express assimilation per unit of leaf dry mass because we aim to
229
determine the plants’ return on investment in biomass. As expected, leaves growing in low light
230
environments tended to have lower LMA in most species (Poorter et al. 2019), and expressing data on a
231
mass basis accounts for these large differences in construction cost. We also express stomatal
232
conductance (gs) and electron transport rate (ETR) on a mass basis for comparison with assimilation rates,
233
with the caveat that both of these processes typically occur near the leaf surface.
234
235
236
Instantaneous chlorophyll fluorescence, pigments, and spectral reflectance
Across all eight broadleaf species, we measured chlorophyll fluorescence parameters using an
237
FMS2 pulse-modulated fluorometer (Hansatech Instruments Ltd., Norfolk, UK) over two days in late July
238
(n = 138 total). We measured both dark- and light-acclimated parameters at the same spot on the same
239
leaves. For dark-acclimation, we placed opaque leaf clips on the leaf the evening before measurement.
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
240
We then measured Fv/Fm within two hours after sunrise; this parameter is a general index of
241
photoinhibition (including sustained photoprotection).
242
We took light-acclimated measurements between 12:00 and 14:00 each day. The protocol
243
involved the following steps: actinic light at 1000 µmol m-2 s-1 for 15 seconds, a saturating pulse, and two
244
seconds of far-red light. We exposed all leaves to the same actinic light because we aimed to assess
245
photoprotective capacity under comparable light conditions, even though individual trees had different
246
light environments.
247
From these data, we estimated qN, a parameter that indicates the extent of non-photochemical
248
quenching under the imposed actinic light; this parameter is correlated with de-epoxidation of
249
xanthophyll cycle pigments (Watling et al. 1997; Cavender-Bares and Bazzaz 2004). Following Kramer
250
et al. (2004), we also calculated the quantum yields ϕPSII, ϕNPQ, and ϕNO, which sum to 1. These three
251
parameters are the proportions of light energy dissipated through photochemistry, non-photochemical
252
quenching, and non-regulated energy dissipation, respectively. The last is a crucial parameter because it
253
represents the inability of a plant to dissipate light in a safe way; this portion of the absorbed light may
254
contribute to photodamage. The Supplemental Methods contain formulas for all parameters and
255
justifications for our choices in data analysis.
256
On separate top leaves of the same trees, we measured reflectance spectra (350-2500 nm) using a
257
PSR+ 3500 field spectroradiometer (Spectral Evolution, Lawrence, MA). We used these spectra to
258
calculate the photochemical reflectance index (PRI), a two-band index that most strongly indicates
259
carotenoid : chlorophyll ratios (Gamon et al. 1992; Wong & Gamon 2015; Gitelson et al. 2017). This
260
index is calculated as PRI = (R531 - R570) / (R531 + R570), where Rn is reflectance at a wavelength of n nm.
261
Low PRI tends to indicate high carotenoid : chlorophyll ratios. We estimated species-specific OLS
262
regression slopes of PRI and of each chlorophyll fluorescence parameter against relative light availability
263
(see Light availability).
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
264
To understand how photoprotective allocation varies with adaptations to shade, we used shade
265
tolerance values compiled by Niinemets & Valladares (2006) on a five-point scale based on prior
266
literature. Higher values along this scale indicate that plants are able to grow in lower light conditions.
267
268
269
Phenology
We also monitored the timing of leaf abscission in monoculture and shaded biculture plots for T.
270
americana to assess whether shade could delay senescence and abscission, extending the period for
271
carbon gain (Cavender-Bares et al. 2000). In early August, we surveyed 120 T. americana trees (60 in
272
monoculture, 60 in shaded biculture) to determine the proportion of leaves that had fallen based on how
273
many of the top five axillary buds lacked a leaf. From each of these trees, we marked a single leaf and
274
returned every 1-2 weeks to monitor its senescence. We assessed each leaf on a binary basis (senesced or
275
not) based on whether it had at least 50% remaining green leaf area. We chose this criterion as a simple
276
proxy for the continuous scale used by Cavender-Bares et al. (2000). Based on the initial survey and the
277
repeated surveys of marked leaves, we tracked the proportion of leaves that had senesced over time. To
278
ensure that leaves were photosynthetically active until near the point of senescence, we collected a one-
279
time measurement of dark-acclimated Fv/Fm in mid-September among remaining leaves.
280
We also performed a one-time measurement of 60 (30 per treatment) A. rubrum plants on October
281
1, using the same protocol we used to do our initial early August survey of T. americana. We aimed for
282
this survey to help test the validity of our results across species.
283
284
Water potential
285
Water deficits—even moderate ones—can reduce carbon gain by causing stomata to close
286
(Brodribb et al. 2003). To assess how our treatments affected water status, we measured leaf water
287
potential using a Scholander pressure bomb. We measured pre-dawn water potential (ΨPD) in all eight
288
broadleaf species (n = 96) and midday water potential (ΨMD; between 12:00 and 14:00) in only the four
289
focal species (n = 48). In these measurements, we included monoculture and shaded biculture treatments,
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
290
except in A. negundo, where we used the twelve-species treatment in place of the absent shaded biculture
291
treatment. For all water potential measurements, we removed leaves and immediately placed them in
292
impermeable plastic bags with wet paper towels, which we stored in a cooler. We then transported leaves
293
indoors to be measured within 90 minutes.
294
Because we only intended to test whether there was a general tendency, across species, for water
295
potential to vary with treatment, we included species identity as a random effect and treatment as a fixed
296
effect in both models.
297
298
299
Leaf angles
Leaf angle is a key control on light exposure because more vertical leaves intercept less light
300
when the sun is directly overhead, and allow light to percolate deeper into the canopy. As a result, steep
301
leaf angles can be a key structural strategy for avoiding light, temperature, and water stress (Valladares &
302
Pugnaire 1999; Werner et al. 1999; Pastenes et al. 2004; Valiente-Banuet et al. 2010). We predicted that
303
trees in monoculture would have steeper leaf angles than those in other treatments because they need to
304
avoid excess light. From each of three species (A. rubrum, Q. ellipsoidalis, and T. americana), we
305
randomly selected five individuals from each of six plots—three from monoculture treatments, three from
306
shaded bicultures (n = 15 per species × treatment). On these individuals, we measured the leaf angle from
307
horizontal, to the nearest 15˚, of the top five leaves in the crown.
308
Because angles of leaves within individuals may be non-independent, we tested differences
309
among treatments using a mixed-effects model with treatment as a fixed effect and individual tree nested
310
within plot as a random effect. Because deviations from horizontal in either direction can reduce light
311
interception, we used the absolute value of leaf angle as the dependent variable in statistical analyses. We
312
calculated p-values using Satterthwaite’s degrees of freedom method in the lmerTest package
313
(Kuznetsova et al. 2017), with the caveat that such values should be interpreted conservatively.
314
315
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
316
Environmental data
317
Light availability
318
319
On a cloudy day, under diffuse light conditions, we measured light interception above each tree
320
selected for this study (n = 138) using an AccuPAR LP-80 ceptometer (METER Group, Pullman, WA).
321
We took the mean of two to four PAR measurements in the open and the mean of two measurements
322
directly above the topmost leaf of each tree. By calculating a ratio of these values, we could calculate the
323
percent of light transmitted to the top of each tree. We called this value relative light availability. This
324
value is typically well-correlated with the percentage of ambient light available over much longer time-
325
scales (Parent & Messier 1996).
326
327
328
Soil water content
We collected a one-time measurement of soil volumetric water content in mid-August using a
329
FieldScout Time-Domain Reflectometer 350 (Spectrum Technologies, Inc., Aurora, IL). This single
330
measurement followed six days from the last major rain. At midday, we collected and averaged three
331
measurements in each of 32 plots, spanning monocultures of the four core broadleaf species, shaded
332
bicultures, and twelve-species plots.
333
334
335
Air temperature
For nine days during early September, we set out four Thermochron iButtons (model DS1921G;
336
Maxim Integrated Products, Sunnyvale, CA, USA) in shaded plots and two in twelve-species plots,
337
setting them to log air temperature hourly. Each iButton was suspended on mesh inside a 1 m-tall PVC
338
pipe that was capped with an elbow to shield against solar irradiance. We compared these data to air
339
temperature measured hourly under open, sunny conditions at the Cedar Creek LTER weather station
340
about 0.77 km away. We assumed that data from this station would be representative of air temperature in
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
341
non-shaded areas. We ignored nighttime and early morning (20:00-08:00 h) readings because iButtons
342
may have been influenced by evaporative cooling of condensed water droplets.
343
344
Results
345
Tree size and growth
346
Neighborhood-scale relationships between focal individual RGR and total neighbor stem volume
347
were noisy but highly significant in most species (Fig. 2). The sign and strength of the relationships
348
varied among species. In most species, individuals with larger neighbors had lower RGR, including in A.
349
rubrum (R2 = 0.033, p <10-5), Q. alba (R2 = 0.038, p <10-6), Q. ellipsoidalis (R2 = 0.030, p <10-5), Q.
350
macrocarpa (R2 = 0.027, p <10-5), and Q. rubra (R2 = 0.056, p <10-9). On the other hand, some species
351
grew faster with larger neighbors, including A. negundo (R2 = 0.018, p < 0.005) and T. americana (R2 =
352
0.019, p <10-4). B. papyrifera showed no significant relationship with neighbor size. These results are
353
qualitatively unchanged by eliminating trees on the edge of the experiment or incorporating plot as a
354
random effect.
355
Although these individual-level effects are noisy, they show strong tendencies in average stem
356
growth across the full range of potential neighbor size. Imagine two T. americana trees that begin at the
357
same size in 2016: One in monoculture, where the average neighbor in 2018 is 573 cm3, and one growing
358
in a shaded biculture with Pinus strobus, where the average neighbor is 2175 cm3. The neighborhood-
359
level regression predicts that after two years, the average T. americana in the shaded biculture would be
360
39.7% larger due to the difference in RGR. Or, consider Q. rubra, which had the strongest negative
361
response to neighbor size: We predict that an average tree in the shaded biculture would be 39.1% smaller
362
than in monoculture. Because these figures assume that treatments show no average size differences by
363
2016, they may underrepresent the true differences in size that have accumulated and may continue to
364
accumulate with time.
365
366
We can also aggregate these observations to the plot scale—for example, by considering the RGR
of summed stem volume of each species in each plot (Fig. 2). (This approach is distinct from an average
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
367
of individual RGRs, which assigns equal weight to small and large individuals.) Although aggregation
368
leaves less power to detect significant relationships, the relationships we find here explain much more of
369
the variation in growth across plots, including negative RGR responses to average neighbor biomass in Q.
370
alba (R2 = 0.311, p < 0.001) and Q. macrocarpa (R2 = 0.174, p < 0.01) and a positive response in T.
371
americana (R2 = 0.282, p < 0.001).
372
Trees that died between 2016 and 2018 dropped out of our individual-level analyses of RGR
373
because their stem volume was 0 in 2018, making it impossible to calculate RGR. To see whether
374
mortality could alter our conclusions, we also performed statistical analyses on untransformed growth in
375
stem volume from 2017 to 2018, in which we alternately (1) treated mortality as a decline in volume to 0,
376
or (2) removed trees that died from analyses. The rate of mortality was low between 2016 and 2018
377
(~7.5%), and accounting for mortality makes no qualitative difference when considering absolute growth
378
as the response variable (Fig. S2).
379
380
Photosynthetic physiology
381
Photosynthetic light-response curves
382
In three out of the four focal species (Q. ellipsoidalis, A. rubrum, and T. americana),
383
photosynthetic rates (per unit dry mass) were higher in shaded biculture and twelve-species treatments
384
than in monoculture at any given chamber light level (Fig. 3, left). In the lone pioneer species, B.
385
papyrifera, photosynthetic rates in shaded biculture were lower than in the other two treatments.
386
The rate of photosynthesis is often limited by either ETR or by RuBP carboxylation rate, which in
387
turn is often limited by stomatal diffusion of CO2 (Farquhar et al. 1980). In our data, both ETR and
388
stomatal conductance (gs) increase with light availability (Fig. S3). (ETR tends to decline at very high
389
light levels, especially in shaded treatments, perhaps due to acute photoinhibition.) Within each species,
390
the rank-order of treatments in photosynthetic rates, ETR, and gs tend to be broadly congruent. One
391
exception is that T. americana in shaded biculture has much higher gs than in twelve-species or
392
monoculture plots, despite having slightly lower photosynthetic rates than in twelve-species plots.
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
393
Compared to chamber light-response curves, the picture that emerges from ambient light-
394
response curves is more complex (Fig. 3, right). In B. papyrifera, the shaded biculture treatment still had
395
the lowest assimilation rate across light levels. In late-successional T. americana, the shaded biculture and
396
twelve-species treatments had higher assimilation rates across most light levels, by up to 25%. In the
397
other two species, assimilation rates were highest in monoculture at low light availability; as top-of-
398
canopy light increases, assimilation rates for twelve-species and (in Q. ellipsoidalis) shaded biculture
399
plots eventually intersect and surpass monoculture rates.
400
Using a time series of solar radiation, we estimated total assimilation rates in July 2018 based on
401
each ambient light-response curve (Fig. S4). We found that trees would have lower total assimilation in
402
shaded biculture than in monoculture in A. rubrum (ANOVA; p = 0.012) and B. papyrifera (p = 0.003); in
403
both species, trees in twelve-species plots did not differ from those in monoculture. In the other two focal
404
species, there were no significant differences among treatments. In these integrated estimates, the costs of
405
shade seem larger than one might suppose from looking only at instantaneous assimilation rates under
406
high ambient light. We attribute these high costs to the fact that light levels were frequently low; PPFD
407
was under 500 µmol m-2 s-1 more than 60% of the time. It is under these low-light conditions that trees in
408
monoculture most outperform those in shaded biculture.
409
410
411
Instantaneous chlorophyll fluorescence, pigments, and spectral reflectance
Among all eight species except the early-successional B. papyrifera, dark-acclimated Fv/Fm
412
declined as relative light availability increased (Fig. 4), showing that high light exposure causes
413
photoinhibition. Across all trees of the four focal species, dark-acclimated Fv/Fm is positively correlated
414
with light-saturated photosynthetic rates (Asat), as estimated from the non-rectangular hyperbolic model
415
(OLS regression; R2 = 0.44; p < 10-9; Fig. 5). Slopes are not significantly different among species. This
416
effect suggests that photoinhibitory declines in PSII efficiency may contribute to variation in light-
417
response curves.
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
418
In most species, qN rose with in situ light availability, indicating that trees that grow in high light
419
have a higher capacity to dissipate excess light with non-photochemical quenching (Fig. 4). However, two
420
species were clear exceptions: In T. americana and A. negundo, qN was nearly invariant across the range
421
of light availability they experienced, which suggests that these species are unable to upregulate
422
photoprotection in full sun.
423
Extracting the species-specific slopes of the change in dark-acclimated Fv/Fm with relative light
424
availability, we find that that the species with the greatest decline in Fv/Fm are the most shade-tolerant
425
(Fig. 5; R2 = 0.601; p = 0.025). Similarly, more shade-tolerant species have less (or no) increase in qN at
426
high light availability (R2 = 0.619; p = 0.022). These results suggest that shade-tolerant species may fail to
427
protect themselves from excess light when growing in light-saturated environments, making them
428
susceptible to photoinhibition. This suggestion is borne out in patterns of ϕNO across light environments,
429
which show that T. americana and A. negundo exposed to 1000 μmol photons m-2 s-1 are just as
430
susceptible to non-regulated dissipation of light when growing in full sun as in the shade (Fig. 4).
431
On a ternary plot, we plotted ϕPSII, ϕNPQ, and ϕNO—the quantum yields of PSII photochemistry,
432
non-photochemical quenching, and non-regulated dissipation, respectively—to illustrate how light energy
433
is dissipated differently across treatments (Fig. S5). Our results broadly show that across species, when
434
exposed to an equivalent, high level of light per unit area, trees that develop in the sun vary in ϕPSII and
435
ϕNPQ but have lower ϕNO. However, because they regularly experience higher light levels, they may still be
436
at high risk of damage.
437
PRI declined with increasing light availability in all species, which supports the idea that trees in
438
monoculture allocate more to carotenoids (Fig. 4). Although this relationship had remarkably consistent
439
slopes and intercepts across nearly all species, B. papyrifera had higher PRI at all light levels within its
440
narrow domain.
441
442
443
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
444
Phenology
445
In T. americana, leaf abscission in monoculture began in early August, and more than half of all
446
leaves had senesced by September 3 (Fig. 6). In shaded bicultures, more than 90% of leaves remained by
447
September 3, and no survey found greater than 50% senescence until October 1. We assigned the
448
senescence date of each leaf as the date of the first survey by which it had senesced; using this response
449
variable, leaves in monoculture senesced much earlier than those in shaded cultures (t-test; p < 10-11).
450
These results suggest that leaves in shaded bicultures had, on average, one additional month for carbon
451
gain. Here, we assume that leaves in shaded biculture did not downregulate photosynthesis long before
452
abscission, which was supported by mid-September measurements of dark-acclimated Fv/Fm. In shaded
453
biculture, dark-acclimated Fv/Fm averaged 0.745 at this point, which was higher than the mean among
454
remaining monoculture leaves, 0.642 (t-test; p = 0.061).
455
Our one-time measurement of A. rubrum leaves in early October suggested that this pattern is not
456
limited to T. americana. A. rubrum trees in shaded biculture plots retained about 56% of their leaves,
457
while those in monoculture retained only 10% (t-test; p < 10-4).
458
459
460
Water potential
For ΨPD, we removed two data points—one in shaded biculture, one in monoculture—with values
461
more negative than -0.8 MPa on the grounds that they may have been caused by measurement error.
462
Across treatments, the mean ΨPD and ΨMD were -0.15 and -1.29 MPa, respectively. ΨPD in shaded
463
bicultures was more negative than in monocultures by about 0.07 MPa (p = 0.004). This result holds even
464
when including the outliers (p = 0.049), with a mean difference of 0.08 MPa between treatments (Fig.
465
S6). By midday, this pattern reversed; ΨMD was more negative in monocultures than shaded bicultures,
466
with a mean difference of 0.33 MPa (p = 0.004).
467
468
469
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
470
471
Leaf angles
Among each of the three species in which we measured leaf angles, leaves were more horizontal
472
in the shaded biculture treatment (Fig. 7). The size of this effect varied among species: 9.0˚ in Q.
473
ellipsoidalis (p = 0.037), 22.0˚ in A. rubrum (p = 0.022), and 25.0˚ in T. americana (p = 0.002). The
474
direction of this shift varied among species. Because T. americana leaves usually drooped downward in
475
monoculture, leaf angles became less negative in shade. In contrast, Q. ellipsoidalis, whose leaves were
476
often inclined upwards in monoculture, had less positive leaf angles in the shade. In A. rubrum, leaf
477
angles in monoculture were inclined both up and down, while in shade, leaves had a tendency to become
478
more horizontal.
479
480
481
Environmental factors
In our survey, soil moisture varied among treatments (ANOVA; p < 0.01); it was lowest (5.7%)
482
in shaded biculture plots and highest (7.2%) in broadleaf monoculture plots. We found that daytime air
483
temperature in the open was higher than in shaded bicultures or twelve-species plots at 62.5% of hourly
484
log-points during the nine-day logging period. The air temperature in monocultures averaged 1.03˚ C
485
higher over this period.
486
487
Discussion
488
We investigated how species interactions in a tree diversity experiment might emerge from the
489
physiological responses of trees to the light environment created by their neighbors. We found that two
490
species (A. negundo, T. americana) grew faster on average with larger neighbors, while five (A. rubrum,
491
Q. alba, Q. ellipsoidalis, Q. macrocarpa, and Q. rubra) grew slower. The two species with positive
492
responses to neighbor size were shade-tolerant and faced a high risk of photoinhibition in high light,
493
which could cause carbon assimilation to be lower in monoculture. In contrast, the five species with
494
negative responses increased non-photochemical quenching in high light and tended to have lower carbon
495
gain in shaded bicultures. These results suggest that the divergent responses of these two groups of
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
496
species were driven, at least in part, by their tolerance to excess light. Here, we interpret these patterns in
497
further detail.
498
499
Physiology of carbon gain
500
We chose our four focal species along a gradient from early-successional, shade-intolerant B.
501
papyrifera, to mid-successional A. rubrum and Q. ellipsoidalis, to late-successional, shade-tolerant T.
502
americana (Braun 1954). This successional gradient aligns with the varied consequences of shade for
503
carbon gain across species. Growing in a shaded biculture can either: (1) decrease assimilation rates
504
(compared to monoculture) at all ambient light levels, as in B. papyrifera; (2) decrease assimilation under
505
low ambient light, but increase it under high light, as in A. rubrum and Q. ellipsoidalis; or (3) increase
506
assimilation rates at nearly all ambient light levels, as in T. americana. Our finding that carbon gain in T.
507
americana increases under shade is reinforced by Carter and Cavaleri (2018), who find that its
508
photosynthetic rates may increase down vertical gradients from the upper canopy to the sub-canopy.
509
The untransformed light-response curves can help us identify the causes of these patterns. As a
510
function of leaf chamber PPFD—without penalizing for lowered light availability—three of our four
511
species (A. rubrum, Q. ellipsoidalis, and T. americana) had higher photosynthetic rates in the shaded
512
biculture and twelve-species treatments than in monoculture (Fig. 3, left). In all species except early-
513
successional B. papyrifera, dark-acclimated Fv/Fm declines with light availability. Because dark-
514
acclimated Fv/Fm correlates with Asat, this result suggests a role for photoinhibition in determining
515
photosynthetic rates. Because photoinhibition most often occurs through damage or regulation of PSII, it
516
is usually thought to depress photosynthesis by lowering ETR. However, both ETR and stomatal
517
conductance seem to reflect the patterns in photosynthesis across treatments in chamber light-response
518
curves (Fig. S3), suggesting that both have some role in explaining why most species showed lower
519
photosynthesis in monoculture.
520
521
Given that light availability fluctuates—diurnally, seasonally, and with cloud cover—the
treatment that performs best may also vary from moment to moment. We produced a more integrated
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
522
picture by estimating total assimilation during the month of July, assuming that each tree’s photosynthetic
523
light-response curve remains constant (Fig. S4). While trees in shaded bicultures had lower estimated
524
assimilation than in monoculture in two of our focal species (B. papyrifera and A. rubrum), those in
525
twelve-species plots had similar assimilation compared to those in monoculture in all four species. It
526
seems plausible that the intermediate light environment in these twelve-species plots allowed trees to
527
avoid photoinhibition without becoming frequently light-limited. We caution that these measurements are
528
only based on top leaves, such that we cannot directly evaluate whole-plant carbon gain.
529
These estimates of assimilation during July cannot explain certain features in our stem biomass
530
data—including, for instance, the clearly positive effect of neighbor size on growth in T. americana. We
531
next discuss two kinds of phenomena that could contribute to variation in growth rates, but which
532
analyses based on light-response curves do not account for: Delayed senescence and microclimatic
533
effects. Both may contribute to potential positive effects of shading by neighbors.
534
T. americana in shade dropped their leaves nearly a month later into the fall compared to in
535
monoculture (Fig. 6). We also found accelerated senescence in full sun in A. rubrum. Given that most
536
leaves remained photosynthetically active until shortly before abscission (as in Mattila et al. 2018), shade
537
can allow plants a longer period of carbon gain. The pattern that the light environment affects senescence
538
timing admits at least two explanations: (1) that plants in the sun accumulate phenological cues for
539
senescence related to temperature, or (2) that excess light and other stresses cause damage that accelerates
540
senescence (Cavender-Bares et al. 2000), especially as the temperature declines, rendering leaves more
541
susceptible to photodamage (Juvany et al. 2013; Renner & Zohner 2019). Contrary to the first
542
explanation, air temperatures in full sun remained higher than in shaded plots.
543
Although we emphasize the role of light stress, other microclimatic factors could contribute to
544
variation in photosynthetic rates among treatments. Because we aimed to control the chamber
545
microclimate during light-response curves, we may have excluded some effects of the ambient
546
microclimate from our photosynthesis estimates. For example, midday leaf water potential tended to be
547
somewhat more negative in monoculture—perhaps due to greater VPD—which could make stomata close
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
548
more often (Fig. S6). On the other hand, pre-dawn leaf water potential was slightly more negative in
549
shaded bicultures, which may owe to lower soil moisture. Our water potential measurements show that
550
water stress was not very severe in any treatment, but even modest differences could affect stomatal
551
behavior (Brodribb et al. 2003). In addition, it seems plausible that greater light exposure may make leaf
552
temperature higher in monoculture (Schymanski et al. 2013), which could push leaves above thermal
553
optima for photosynthesis.
In summary, we contend that shade has both costs and benefits. Being shaded reduces the light
554
555
available for photosynthesis, but it may also prevent chronic stress due to excess light, allowing the plant
556
to use the light it absorbs more efficiently. Both factors must be jointly considered to determine whether
557
shade benefits or harms a plant.
558
559
Interpreting patterns of stem growth
The stem growth data are consistent with the hypothesis that larger neighbors can facilitate
560
561
growth in T. americana and A. negundo, the most shade-tolerant species. In light of the physiological
562
data, we interpret this pattern as a result of abiotic stress amelioration. Because we lacked measures of
563
root allocation, we cannot rule out the hypothesis that this trend is caused by a shade avoidance response
564
that increases allocation to shoots compared to roots (Schall et al. 2012). Nevertheless, the idea that much
565
of this effect is facilitative is supported both by the physiological data and by other studies that have
566
found that shade can increase total (root and shoot) biomass (e.g. Ryser & Eek 2000; Semchenko et al.
567
2012).
568
In the majority of species, including all four Quercus species and A. rubrum, larger neighbor size
569
caused focal individual size to be smaller, which we interpret as the result of net competitive interactions,
570
potentially both aboveground and belowground. The occurrence of belowground competition is suggested
571
by the fact that soil moisture was lowest in our dense-growing shaded biculture plots. B. papyrifera, a
572
shade-intolerant early-successional species whose leaves had lower photosynthetic rates in the shaded
573
biculture, had a neutral response to neighbor size, perhaps because its rapid height growth allowed it to
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
574
overtop neighbors rather than be overtopped. In contrast to all the other broadleaf species, its top leaves
575
suffered no reduction of light availability in shaded biculture, compared to monoculture.
576
577
578
Mechanisms of photoprotection
Plants use multiple means to avoid damage from excess light, which can be classified broadly
579
into biochemical and structural strategies. Our results show that trees in full sun allocate more to pigment-
580
based methods of photoprotection, as shown by lower PRI, and higher qN in most species. The two most
581
shade-tolerant species—A. negundo and T. americana—do not show a greater capacity for non-
582
photochemical quenching when grown in full sun. Across a gradient of shade to sun, these species may
583
experience large increases in ϕNO, and may suffer from greater photodamage as a result, as supported by
584
patterns of dark-acclimated Fv/Fm. This result is consistent with other research that finds that shade-
585
adapted species often fail to develop strong photoprotective abilities in the sun (Montgomery et al. 2008).
586
Our survey of leaf angles is also consistent with the idea that plants may steeply incline their
587
leaves to avoid intercepting excess light—a form of structural photoprotection. Leaves that are steeply
588
inclined tend to intercept less light per unit area, particularly during midday, when solar radiation is
589
otherwise most intense. The dramatic trend we observe in leaf angles suggests that these species are able
590
to use gross structural characteristics to partially regulate their light absorption.
591
Plants—especially those that grow in the shade—may lack perfect photoprotective mechanisms
592
in part because they are costly. Building pigments and proteins for photoprotection uses nutrients that
593
could otherwise be allocated to chlorophyll, RuBisCO, or structural tissue components. Furthermore,
594
slow relaxation of NPQ may hinder photosynthesis even when periodic declines in light availability
595
remove the immediate threat of damage (Zhu et al. 2004; Murchie & Niyogi 2011; Kromdijk et al. 2016).
596
Steep leaf angles may also reduce carbon assimilation by reducing light interception even under low light.
597
However, plants in full sun may need these mechanisms to avoid even greater long-term costs caused by
598
photodamage. Our measurements of the mechanisms of photoprotection suggest that these species may
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
599
incur some risk of damage under high light, reinforcing our conclusions about the role of light stress in
600
shaping species interactions.
601
602
603
The case for facilitation
An important result of our study is that shade from large neighbors increases both net
604
photosynthetic rates and stem growth rates in shade-tolerant species. This finding has precedents: Ball et
605
al. (1991) and Egerton et al. (2000) found that shade enhanced photosynthesis and growth of evergreen
606
Eucalyptus seedlings during the winter, when cold temperatures impose a risk of photoinhibition. Howell
607
et al. (2002) also found that the leafless outer branches of New Zealand’s divaricating shrubs reduce
608
photoinhibition by shading inner leaves.
609
Much of the work on facilitation in plant communities is guided by the stress-gradient hypothesis
610
(SGH), which proposes that facilitative interactions are more common in stressful environments (Bertness
611
& Callaway 1994; Maestre et al. 2009). We can understand photoinhibition in light of the SGH as
612
follows: Stressful conditions limit the use of light for photochemistry, causing high light exposure to
613
become a potential stressor. Under these conditions, shade from neighbors can ameliorate stress, causing
614
facilitation. Under benign conditions, where plants can use more of the available light for photochemistry,
615
high light exposure may not cause damage. Here, shading by neighbors can cause light limitation,
616
resulting in an adverse, competitive effect on growth. As we show here, species’ physiological traits also
617
influence whether they are stressed by their environment.
618
Ever since the SGH was first described, the role of facilitative interactions in structuring plant
619
communities has received more attention. For example, Bimler et al. (2018) suggest that models of
620
species occurrence will be more accurate when they incorporate positive interactions. Nevertheless, the
621
role of facilitation in structuring plant communities is still often overlooked (Wright et al. 2017), and its
622
mechanisms are seldom assessed by measuring physiological rates (but see, for example, Barron-Gafford
623
et al. 2017).
624
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
625
Competition and facilitation in biodiversity-ecosystem function relationships
626
Overyielding—or biodiversity-enhanced growth—is a hallmark of positive biodiversity-
627
ecosystem function relationships. A whole community can only overyield if at least one of its constituent
628
species overyields. Although we show how species growth changes with neighbor size, not neighbor
629
diversity, our results show that shading allows some species to achieve greater biomass in shaded
630
bicultures or twelve-species plots. In the same experiment, Grossman et al. (2017) found that T.
631
americana has the greatest overyielding in multi-species plots of any broadleaf species, especially when
632
growing with conifers or B. papyrifera. Grossman et al. (2017) further showed that conifer partners in
633
these species combinations also grow faster than in monoculture, resulting in community overyielding.
634
In general, the species that faced the strongest negative effects from shade had physiological
635
tolerance to high light and experienced shading as a form of competition. On the other hand, species that
636
were facilitated by shade had low tolerance to excess light and likely benefited from the milder
637
microclimate their neighbors created. We can place these findings in the context of the three classes of
638
facilitative mechanisms distinguished by Wright et al. (2017):
639
640
641
642
643
644
1. Biotic facilitation, which is mediated by the activity of higher-order trophic interactions, such as
promotion of shared mutualists or dilution of natural enemy loads;
2. Abiotic facilitation through resource enrichment, in which some species increase resource
availability for others; and,
3. Abiotic facilitation through stress amelioration, in which some species create a favorable
microclimate for others.
645
In the FAB experiment, enemy-mediated biotic facilitation is likely to have a limited role; Grossman et al.
646
(2018) found that diversity had a mild influence on damage by herbivores and pathogens, and the
647
direction of the relationship varied among tree species and herbivore guilds. Nutrient enrichment is
648
unlikely to be a major mechanism for facilitation here because, although nitrogen is the main limiting
649
mineral nutrient at Cedar Creek (Tilman 1987), none of the planted tree species associates with nitrogen-
650
fixing microbes. As a result, stress amelioration is likely to be the main class of facilitation in this
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
651
experiment, perhaps acting in concert with niche partitioning and (to a limited extent) selection effects to
652
drive overyielding.
653
In much of the literature on biodiversity and ecosystem function, the attempt to parse out
654
explanations—such as niche partitioning and facilitation—often depends on ever-finer statistical
655
partitions of biomass data (Loreau & Hector 2001; Fox & Kerr 2012), whose interpretation is often
656
contested (Carroll et al. 2011; Pillai et al. 2019). Because the processes that drive biomass accumulation
657
are ultimately physiological, physiological measurements can guide a mechanistic interpretation of
658
biomass patterns. These techniques can help us determine what specific kinds of plant-plant and plant-
659
environment interactions produce biodiversity-productivity relationships, and how these interactions
660
emerge from basic aspects of plant function.
661
In this study, the survey-based biomass data and the physiological measurements generally
662
converge in showing that certain species respond positively to the size of their neighbors, while others
663
respond negatively. However, the physiological measurements allow us to reach a deeper level of
664
explanation, showing how these interactions emerge in part from the photosynthetic responses of trees to
665
the light environment created by their neighbors. Thus, we demonstrate that community ecology can
666
benefit by incorporating mechanistic approaches from physiology.
667
668
669
670
671
672
673
674
675
676
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
677
Acknowledgements
678
We acknowledge that University of Minnesota, including Cedar Creek ESR, lies on the ancestral,
679
traditional, and contemporary Land of the Dakota. This research was inspired in large part by
680
conversations with Jake Grossman and Laura Williams. Beth Fallon, German Vargas G., Artur Stefanski,
681
Daniel Stanton, and Danielle Way all gave valuable advice about measuring photosynthesis. Cathleen
682
Nguyen and Chris Buyarski provided tremendous logistical support for data collection in FAB; Troy
683
Mielke, Kally Worm, Jim Krueger, Pam Barnes, Mark Saxhaug, Dan Bahauddin, and others supported
684
other crucial aspects of Cedar Creek infrastructure. Many Cedar Creek interns helped in the stem volume
685
survey. Sarah Hobbie and Peter Reich contributed to the design of the FAB experiment. The Cavender-
686
Bares lab (especially Gerard Sapès), Daniel Stanton, German Vargas G., Artur Stefanski, Jamie Mosel,
687
José Lopez, and David Kramer all provided feedback on the results or manuscript. The FAB experiment
688
is maintained with support from National Science Foundation under DEB #1234162 to Cedar Creek
689
LTER. Spectral measurements were conducted as part of NSF/NASA DEB #1342778 to J.C.B. and R.M..
690
S.K. has been supported by an NSF Graduate Research Fellowship (Grant No. 00039202) and a UMN
691
Doctoral Dissertation Fellowship. R.M. is also supported by Minnesota Agricultural Experiment Station
692
project MIN-42-060.
693
694
695
696
697
698
699
700
701
702
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
References
Ammer C. Diversity and forest productivity in a changing climate. New Phytologist 221, 50–66 (2019).
Ball MC, Hodges VS, Laughlin GP. Cold-Induced Photoinhibition Limits Regeneration of Snow Gum at
Tree-Line. Functional Ecology 5, 663–668 (1991).
Barron-Gafford GA, Sanchez-Cañete EP, Minor RL, Hendryx SM, Lee E, Sutter LF, Tran N, Parra E,
Colella T, Murphy PC et al. Impacts of hydraulic redistribution on grass–tree competition vs facilitation
in a semi-arid savanna. New Phytologist 215, 1451–1461 (2017).
Bertness MD, Callaway R. Positive interactions in communities. Trends in Ecology & Evolution 9, 191–
193 (1994).
Bimler MD, Stouffer DB, Lai HR, Mayfield MM. Accurate predictions of coexistence in natural systems
require the inclusion of facilitative interactions and environmental dependency. Journal of Ecology 106,
1839–1852 (2018).
Björkman O, Powles SB. Inhibition of photosynthetic reactions under water stress: interaction with light
level. Planta 161, 490–504 (1984).
Braun-Blanquet J. Plant Sociology (trans. Fuller GD, Conard HS). New York: McGraw-Hill (1932).
Brodribb TJ, Holbrook NM, Edwards EJ, Gutiérrez MV. Relations between stomatal closure, leaf turgor
and xylem vulnerability in eight tropical dry forest trees. Plant, Cell & Environment 26, 443–450 (2003).
Canham CD, Finzi AC, Pacala SW, Burbank DH. Causes and consequences of resource heterogeneity in
forests: interspecific variation in light transmission by canopy trees. Can. J. For. Res. 24, 337–349 (1994).
Carroll IT, Cardinale BJ, Nisbet RM. Niche and fitness differences relate the maintenance of diversity to
ecosystem function. Ecology 92, 1157–1165 (2011).
Cavender-Bares J. Chilling and freezing stress in live oaks (Quercus section Virentes): Intra- and
interspecific variation in PS II sensitivity corresponds to latitude of origin. Photosynthesis Research 94,
437–453 (2007).
Cavender-Bares J, Bazzaz FA. From Leaves to Ecosystems: Assessing Photosynthesis and Plant Function
in Ecological Studies. In Papageorgiou GC and Govindjee (Eds.), Chlorophyll fluorescence: A Signature
of Photosynthesis (pp. 737–755). Dordrecht, Netherlands: Kluwer Academic Publishers (2004).
Cavender‐Bares J, Potts M, Zacharias E, Bazzaz FA. Consequences of CO2 and light interactions for leaf
phenology, growth, and senescence in Quercus rubra. Global Change Biology 6, 877–887 (2000).
Demmig-Adams B, Adams, WW III. Photoprotection and Other Responses of Plants to High Light Stress.
Annual Review of Plant Physiology and Plant Molecular Biology 43, 599–626 (1992).
Demmig-Adams B, Adams, WW III. An integrative approach to photoinhibition and photoprotection of
photosynthesis. Environmental and Experimental Botany 154, 1–3 (2018).
Dybzinski R, Tilman D. Resource Use Patterns Predict Long‐Term Outcomes of Plant Competition for
Nutrients and Light. The American Naturalist 170, 305–318 (2007).
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
Egerton JJG, Banks JCG, Gibson A, Cunningham RB, Ball MC. Facilitation of Seedling Establishment:
Reduction in Irradiance Enhances Winter Growth of Eucalyptus pauciflora. Ecology 81, 1437–1449
(2000).
Ehleringer J, Werk KS. Modifications of solar radiation absorption patterns and the implications for
carbon gain at the leaf level. In T. Givnish (ed.), On the Economy of Plant Form and Function.
Cambridge Univ. Press, London. 57-82 (1986).
Farquhar GD, von Caemmerer S, Berry JA. A biochemical model of photosynthetic CO2 assimilation in
leaves of C3 species. Planta 149, 78–90 (1980).
Fichtner A, Härdtle W, Li Y, Bruelheide H, Kunz M, von Oheimb G. From competition to facilitation:
how tree species respond to neighbourhood diversity. Ecology Letters 20, 892–900 (2017).
Fox JW, Kerr B. Analyzing the effects of species gain and loss on ecosystem function using the extended
Price equation partition. Oikos 121, 290–298 (2012).
Gamfeldt L, Snäll T, Bagchi T, Jonsson M, Gustafsson L, Kjellander P, Ruiz-Jaen MC, Fröberg M,
Stendahl J, Philipson CD et al. Higher levels of multiple ecosystem services are found in forests with
more tree species. Nature Communications 4, 1340 (2013).
Gamon JA, Peñuelas J, Field CB. A narrow-waveband spectral index that tracks diurnal changes in
photosynthetic efficiency. Remote Sensing of Environment 41, 35–44 (1992).
Gitelson AA, Gamon JA, Solovchenko, A. Multiple drivers of seasonal change in PRI: Implications for
photosynthesis 1. Leaf level. Remote Sensing of Environment 191, 110–116 (2017).
Grossman JJ, Cavender‐Bares J, Reich PB, Montgomery RA, Hobbie SE. Neighborhood diversity
simultaneously increased and decreased susceptibility to contrasting herbivores in an early stage forest
diversity experiment. Journal of Ecology 107, 1492–1505 (2019).
Grossman JJ, Vanhellemont M, Barsoum N, Bauhus J, Bruelheide H, Castagneyrol B, Cavender-Bares J,
Eisenhauer N, Ferlian O, Gravel D et al. Synthesis and future research directions linking tree diversity to
growth, survival, and damage in a global network of tree diversity experiments. Environmental and
Experimental Botany 152, 68–89 (2018).
Hoffmann WA, Poorter H. Avoiding Bias in Calculations of Relative Growth Rate. Annals of Botany 90,
37 (2002).
Howell CJ, Kelly D, Turnbull MH. Moa ghosts exorcised? New Zealand’s divaricate shrubs avoid
photoinhibition. Functional Ecology 16, 232–240 (2002).
Huang Y, Chen Y, Castro-Izaguirre N, Baruffol M, Brezzi M, Lang A, Li Y, Härdtle W, von Oheimb G,
Yang V et al. Impacts of species richness on productivity in a large-scale subtropical forest experiment.
Science 362, 80–83 (2018).
Juvany M, Müller M, Munné-Bosch S. Photo-oxidative stress in emerging and senescing leaves: a mirror
image? J Exp Bot 64, 3087–3098 (2013).
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
Kothari S, Cavender-Bares J, Bitan K, Verhoeven AS, Wang R, Montgomery RA, Gamon J. Communitywide consequences of variation in photoprotective physiology among prairie plants. Photosynthetica 56,
455–467 (2018).
Kramer DM, Johnson G, Kiirats O, Edwards GE. New Fluorescence Parameters for the Determination of
QA Redox State and Excitation Energy Fluxes. Photosynthesis Research 79, 209–218 (2004).
Kromdijk J, Głowacka K, Leonelli L, Gabilly ST, Iwai M, Niyogi KK, Long SP. Improving
photosynthesis and crop productivity by accelerating recovery from photoprotection. Science 354, 857–
861 (2016).
Külheim C, Ǻgren J, Jansson S. Rapid Regulation of Light Harvesting and Plant Fitness in the Field.
Science 297, 91–93 (2002).
Liang J, Crowther TW, Picard N, Wiser S, Zhou M, Alberti G, Schulze E-D, McGuire AD, Bozzato F,
Pretzsch H et al. Positive biodiversity-productivity relationship predominant in global forests. Science
354, aaf8957 (2016).
Lobo FA, de Barros MP, Dalmagro HJ, Dalmolin ÂC, Pereira WE, de Souza ÉC, Vourlitis GL,
Rodríguez Ortíz CE. Fitting net photosynthetic light-response curves with Microsoft Excel — a critical
look at the models. Photosynthetica 51, 445–456 (2013).
Long SP, Humphries S, Falkowski PG. Photoinhibition of Photosynthesis in Nature. Annual Review of
Plant Physiology and Plant Molecular Biology 45, 633-62 (1994).
Loreau M, Hector A. Partitioning selection and complementarity in biodiversity experiments. Nature 412,
72–76 (2001).
Lovelock CE, Clough BF. Influence of solar radiation and leaf angle on leaf xanthophyll concentrations
in mangroves. Oecologia 91, 518–525 (1992).
Maestre FT, Callaway RM, Valladares F, Lortie CJ. Refining the stress-gradient hypothesis for
competition and facilitation in plant communities. Journal of Ecology 97, 199–205 (2009).
Mattila H, Valev D, Havurinne V, Khorobrykh S, Virtanen O, Antinluoma M, Mishra KB, Tyystjärvi E.
Degradation of chlorophyll and synthesis of flavonols during autumn senescence—the story told by
individual leaves. AoB Plants 10, (2018).
Monsi M, Saeki T. On the Factor Light in Plant Communities and its Importance for Matter Production.
Ann Bot 95, 549–567 (2005).
Monteith JL, Moss CJ, Cooke GW, Pirie NW, Bell GDH. Climate and the efficiency of crop production
in Britain. Philosophical Transactions of the Royal Society of London B: Biological Sciences 281, 277–
294 (1977).
Montgomery RA, Goldstein G, Givnish TJ. Photoprotection of PSII in Hawaiian lobeliads from diverse
light environments. Functional Plant Biol. 35, 595–605 (2008).
Murchie EH, Niyogi KK. Manipulation of Photoprotection to Improve Plant Photosynthesis. Plant
Physiology 155, 86–92 (2011).
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
Niinemets Ü, Valladares F. Tolerance to Shade, Drought, and Waterlogging of Temperate Northern
Hemisphere Trees and Shrubs. Ecological Monographs 76, 521–547 (2006).
Oehri J, Schmid B, Schaepman-Strub G, Niklaus PA. Biodiversity promotes primary productivity and
growing season lengthening at the landscape scale. PNAS 114, 10160–10165 (2017).
Parent S, Messier C. A simple and efficient method to estimate microsite light availability under a forest
canopy. Can. J. For. Res. 26, 151–154 (1996).
Pastenes C, Porter V, Baginsky C, Horton P, González J. Paraheliotropism can protect water-stressed
bean (Phaseolus vulgaris L.) plants against photoinhibition. Journal of Plant Physiology 161, 1315–1323
(2004).
Pearcy RW, Muraoka H, Valladares F. Crown architecture in sun and shade environments: assessing
function and trade-offs with a three-dimensional simulation model. New Phytologist 166, 791–800
(2005).
Pillai P, Gouhier TC. Not even wrong: the spurious measurement of biodiversity’s effects on ecosystem
functioning. Ecology 100, e02645 (2019).
Poorter H, Niinemets Ü, Ntagkas N, Siebenkäs A, Mäenpää M, Matsubara S, Pons TL. A meta-analysis
of plant responses to light intensity for 70 traits ranging from molecules to whole plant performance. New
Phytologist 223, 1073–1105 (2019).
Potvin C, Gotelli NJ. Biodiversity enhances individual performance but does not affect survivorship in
tropical trees. Ecology Letters 11, 217–223 (2008).
Ramírez-Valiente JA, Koehler K, Cavender-Bares J. Climatic origins predict variation in photoprotective
leaf pigments in response to drought and low temperatures in live oaks (Quercus series Virentes). Tree
Physiology 35, 521–534 (2015).
Renner SS, Zohner CM. The occurrence of red and yellow autumn leaves explained by regional
differences in insolation and temperature. New Phytologist 224, 1464–1471 (2019).
Ryser P, Eek L. Consequences of phenotypic plasticity vs. interspecific differences in leaf and root traits
for acquisition of aboveground and belowground resources. American Journal of Botany 87, 402–411
(2000).
Savage, J., J. Cavender-Bares, and A. Verhoeven. Habitat generalists and wetland specialists in the genus
Salix vary in their photoprotective responses to drought. Functional Plant Biology 36, 300–309 (2009).
Schall P, Lödige C, Beck M, Ammer C. Biomass allocation to roots and shoots is more sensitive to shade
and drought in European beech than in Norway spruce seedlings. Forest Ecology and Management 266,
246–253 (2012).
Schymanski SJ, Or D, Zwieniecki M. Stomatal Control and Leaf Thermal and Hydraulic Capacitances
under Rapid Environmental Fluctuations. PLOS ONE 8, e54231 (2013).
Semchenko M, Lepik M, Götzenberger L, Zobel K. Positive effect of shade on plant growth: amelioration
of stress or active regulation of growth rate? Journal of Ecology 100, 459–466 (2012).
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
Tilman D. Secondary Succession and the Pattern of Plant Dominance Along Experimental Nitrogen
Gradients. Ecological Monographs 57, 189–214 (1987).
Tilman D, Isbell F, Cowles JM. Biodiversity and Ecosystem Functioning. Annual Review of Ecology,
Evolution, and Systematics 45, 471–493 (2014).
Tobner CM, Paquette A, Gravel D, Reich PB, Williams LJ, Messier C. Functional identity is the main
driver of diversity effects in young tree communities. Ecol Lett 19, 638–647 (2016).
Valiente-Banuet A, Verdú M, Valladares F, García-Fayos P. Functional and evolutionary correlations of
steep leaf angles in the mexical shrubland. Oecologia 163, 25–33 (2010).
Valladares F, Pugnaire FI. Tradeoffs Between Irradiance Capture and Avoidance in Semi-arid
Environments Assessed with a Crown Architecture Model. Annals of Botany 83, 459–469 (1999).
Watling JR, Robinson SA, Woodrow IE, Osmond CB. Responses of Rainforest Understorey Plants to
Excess Light during Sunflecks. Australian Journal of Plant Physiology 24, 17–25 (1997).
Werner C, Correia O, Beyschlag W. Two different strategies of Mediterranean macchia plants to avoid
photoinhibitory damage by excessive radiation levels during summer drought. Acta Oecologica 20, 15–23
(1999).
Williams LJ, Paquette A, Cavender-Bares J, Messier C, Reich PB. Spatial complementarity in tree crowns
explains overyielding in species mixtures. Nature Ecology & Evolution 1, 0063 (2017).
Wong CYS, Gamon JA. Three causes of variation in the photochemical reflectance index (PRI) in
evergreen conifers. New Phytol 206, 187–195 (2015).
Wright AJ, Wardle DA, Callaway R, Gaxiola A. The Overlooked Role of Facilitation in Biodiversity
Experiments. Trends in Ecology & Evolution 32, 383–390 (2017).
Wujeska A, Bossinger G, Tausz M. Responses of foliar antioxidative and photoprotective defence
systems of trees to drought: a meta-analysis. Tree Physiology 33, 1018–1029 (2013).
Zemp DC, Gérard A, Hölscher D, Ammer C, Irawan B, Sundawati L, Teuscher M, Kreft H. Tree
performance in a biodiversity enrichment experiment in an oil palm landscape. Journal of Applied
Ecology 56, 2340–2352 (2019).
Zhu X-G, Ort DR, Whitmarsh J, Long SP. The slow reversibility of photosystem II thermal energy
dissipation on transfer from high to low light may cause large losses in carbon gain by crop canopies: a
theoretical analysis. J Exp Bot 55, 1167–1175 (2004).
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
957
Tables
958
Table 1: Characteristics of species in FAB. Except for B. papyrifera, broadleaf species tended to grow
959
slower than needleleaf conifers. Shade tolerance is drawn from Niinemets and Valladares (2006), who
960
evaluated it along a 1-5 scale where higher values indicate an ability to grow under lower light.
961
962
963
964
965
966
967
968
Scientific
name
Species
code
Common
name
Family
Leaf form
Acer
negundo
Acer
rubrum
Betula
papyrifera
Juniperus
virginiana
Pinus
banksiana
Pinus
resinosa
Pinus
strobus
Quercus
alba
Quercus
ellipsoidalis
Quercus
macrocarpa
Quercus
rubra
Tilia
americana
ACNE
Box elder
Sapindaceae
ACRU
PIBA
Red
maple
Paper
birch
Eastern
redcedar
Jack pine
PIRE
PIST
BEPA
JUVI
QUAL
QUEL
QUMA
QURU
TIAM
Broadleaf
Mean RGR in
monoculture,
2013-6 (/year)
0.039
Shade tolerance Partner in
shaded
biculture
3.47
None
Sapindaceae
Broadleaf
0.379
3.44
Betulaceae
Broadleaf
1.272
1.54
Cupressaceae
Needleleaf 0.975
1.28
Pinus
banksiana
Pinus
banksiana
NA
Pinaceae
Needleleaf 2.288
1.36
NA
Red pine
Pinaceae
Needleleaf 1.619
1.89
NA
White
pine
White oak
Pinaceae
Needleleaf 1.722
3.21
NA
Fagaceae
Broadleaf
0.735
2.85
Northern
pin oak
Bur oak
Fagaceae
Broadleaf
0.722
NA
Fagaceae
Broadleaf
0.547
2.71
Northern
red oak
American
basswood
Fagaceae
Broadleaf
0.505
2.75
Tiliaceae
Broadleaf
0.724
3.98
Pinus
strobus
Pinus
strobus
Betula
papyrifera
Pinus
strobus
Pinus
strobus
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
969
Figure captions
970
Fig. 1: (A) Consider two hypothetical photosynthetic light-response curves of unstressed (blue) and
971
photoinhibited (red) plants, where the unstressed plant’s neighbors reduce the amount of light it absorbs.
972
Given a light-response curve, an instantaneous reduction in light tends to reduce photosynthetic rates
973
(represented by the small black arrow along the red curve). However, if a long-term reduction in light
974
spares the plant from chronic photoinhibition, it can alter the parameters of the curve such that carbon
975
assimilation at any given instantaneous light level is greater (represented by the small black vertical
976
arrow). The balance (represented by the large dashed arrow) may either increase or reduce net carbon
977
assimilation.
978
(B) To allow us to compare photosynthetic rates visually, we can put the two light-response curves on a
979
common x-axis that represents light arriving at the top of the canopy. To do this, we need to scale the blue
980
light response curve horizontally to account for the fact that when the stressed plant receives the amount
981
of light corresponding to the red dashed line in (A), the unstressed plant receives a proportionally lower
982
amount, here represented by the blue dashed line. By stretching the blue curve proportionally such that
983
these two lines coincide, we can compare the true estimated photosynthetic rates at a given light
984
availability.
985
986
Fig. 2: Relative growth rate (RGR) of woody stem volume in individuals of each species between fall
987
2016 and 2018 as a function of the total stem volume of all neighbors. Gray dots represent individuals,
988
and the regression line is fit through the individual-level data. Large colored dots are aggregated to the
989
plot scale, as described in the main text, and color-coded by treatment. A shaded biculture is a plot in
990
which the focal broadleaf species grows with either a conifer or B. papyrifera (unless B. papyrifera is the
991
focal broadleaf species). An open biculture is a plot in which the focal broadleaf grows with another
992
broadleaf species (other than B. papyrifera).
993
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
994
Fig. 3: Mass-based photosynthetic light-response curves for four broadleaf species. The left panels
995
display photosynthetic rates as a function of chamber light intensity; the right panels represent an estimate
996
of realized photosynthetic rates as a function of light intensity at the top of the canopy, after accounting
997
for light interception by neighbors (as in Fig. 1B). Species are arranged from the least shade-tolerant (B.
998
papyrifera) to the most (T. americana). Error bars in left panels are ± 1 SE.
999
1000
Fig. 4: Dark-acclimated Fv/Fm, qN, ϕNO, and PRI as a function of relative light availability across species
1001
and treatments.
1002
1003
Fig. 5: Compared to shade-intolerant species, shade-tolerant species show greater declines in dark-
1004
acclimated Fv/Fm and smaller increases in qN as light availability increases. Asat derived from light-
1005
response curves correlates positively and strongly with dark-acclimated Fv/Fm; slopes are statistically
1006
indistinguishable among species.
1007
1008
Fig. 6: Timing of fall leaf abscission in monoculture and shaded biculture treatments of T. americana.
1009
1010
Fig. 7: Top leaf angles of A. rubrum (left), Q. ellipsoidalis (middle), and T. americana (right) across
1011
treatments.
1012
1013
Fig. S1: Example images of Tilia americana in shaded biculture (top) and monoculture (bottom). Red
1014
scale bars represent about 1 m in the foreground.
1015
1016
Fig. S2: Growth of woody stem volume in individuals of each species between fall 2017 and 2018 as a
1017
function of the total stem volume of all neighbors. Gray dots represent individuals, while colored dots are
1018
aggregated to the plot scale and color-coded by treatment. Two regression lines are fit: One (blue) to all
1019
data, and one (red) excluding trees that died between 2017 and 2018.
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1020
1021
Fig. S3: Responses of ETR and stomatal conductance to light availability for the four focal broadleaf
1022
species. Error bars are ± 1 SE.
1023
1024
Fig. S4: An estimate of total carbon assimilation across species and treatments throughout the month of
1025
July, calculated by applying a time series of solar radiation from a nearby weather station as inputs to our
1026
ambient light-response curves.
1027
1028
Fig. S5: A ternary plot depicting variation among species and treatments in ϕPSII, ϕNPQ, and ϕNO, as
1029
calculated using the lake model derivations of Kramer et al. (2004). These three quantities describe light
1030
energy dissipation through photochemistry, through non-photochemical quenching, and through non-
1031
regulated dissipation, respectively. A thick dashed line represents 0.83 the theoretical maximum
1032
efficiency of PSII posited in Tietz et al. (2017). Deviations from this line represent increasingly high ϕNO,
1033
which can result in photodamage.
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
Fig. S6: Water potential before dawn (top) and at midday (bottom) among species and treatments.
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
Figures
Fig. 1
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1085
1086
Fig. 2
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1087
1088
Fig. 3
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
Fig. 4
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
Fig. 5
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1120
Fig. 6
1121
1122
1123
Fig. 7
1124
1125
1126
1127
1128
1129
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1130
Supplement to: “Throwing shade: Physiological responses to light mediate competition and
1131
facilitation in a tree diversity experiment”
1132
1133
Materials and Methods
1134
Photosynthetic light-response curve methods
1135
We measured photosynthetic-light response curves from the four focal species (Acer negundo,
1136
Betula papyrifera, Quercus ellipsoidalis, and Tilia americana) during July 2018 using an LI-6400 gas
1137
exchange system (LI-COR BioSciences, Lincoln, NE) with a 6400-40 Leaf Chamber Fluorometer head.
1138
From each tree (n = 72 total), we selected a fully expanded, sun-exposed leaf with no visible sign of
1139
disease or herbivory. We noted the angle of each leaf relative to horizontal to the nearest 15˚ before
1140
beginning measurements. Each curve had nine steps in descending order of brightness: 2000, 1500, 1000,
1141
500, 200, 100, 50, 20, and 0 µmol m-2 s -1.
1142
During each curve, we controlled environmental conditions to reduce the immediate effects of
1143
environmental conditions, other than light, on stomatal conductance. We maintained relative humidity at a
1144
favorable level, typically between 65 and 75%, and kept block temperature as close as possible to 25 ˚C.
1145
The reference CO2 concentration remained at 400±10 ppm. We kept each leaf in the chamber for at least
1146
five minutes before beginning each curve. We checked that outputs were stable and matched IRGAs
1147
before each measurement.
1148
While we sought to control environmental conditions in the chamber, any remaining differences
1149
among treatments could, in theory, contribute to the differences we observe in photosynthetic function.
1150
During light-response curves, we found no difference among treatments in leaf-to-air VPD within the
1151
chamber. There were significant but slight differences in leaf temperature inside the chamber; leaves in
1152
monoculture, shaded biculture, and twelve-species plots had average temperatures of 27.7 ˚C, 26.5 ˚C,
1153
and 27.0 ˚C, respectively. Previous studies on some of the same species in Minnesota have typically
1154
found thermal optima below 26 ˚C (Sendall et al. 2015). However, while cooler leaves may be closer to
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1155
thermal optima, the small differences in temperature we observed seem unlikely to explain much of the
1156
observed variation in photosynthetic rates within the leaf chamber.
1157
In all but five light-response curves, we also used a saturating pulse to measure light-acclimated
1158
chlorophyll fluorescence at each light level. These data allowed us to calculate ϕPSII, a measure of the
1159
efficiency of Photosystem II (PSII) at each actinic light level, as:
𝜙𝑃𝑆𝐼𝐼 =
1160
𝐹𝑚′ − 𝐹𝑠
𝐹𝑚′
1161
where Fs is the steady-state fluorescence yield under actinic light, and Fm’ is the maximum fluorescence
1162
yield of the actinic light-acclimated sample following a saturating light pulse, which temporarily closes
1163
all PSII reaction centers. From this parameter, we could calculate electron transport rate as:
1164
𝐸𝑇𝑅 = 𝜙𝑃𝑆𝐼𝐼 × 𝑃𝑃𝐹𝐷 × 0.42 × 0.5
1165
where PPFD is photosynthetic photon flux density (the light level inside the chamber, in µmol m-2 s -1),
1166
0.42 is an estimate of leaf absorptance, and 0.5 is an estimate of the fraction of photons that are captured
1167
by PSII rather than PSI. Electron transport rate is a key parameter because electron transport is one of the
1168
main steps that is capable of limiting photosynthesis (Farquhar et al. 1980), especially when light
1169
availability or PSII efficiency are low.
1170
1171
1172
Chlorophyll fluorescence procedure
As described in the main text, we used dark- and light-acclimated measurements of chlorophyll
1173
fluorescence to calculate several derived parameters, from which we make inferences about
1174
photosynthetic and photoprotective function. Most importantly, we wanted to assess: (1) capacity for non-
1175
photochemical quenching, as measured through the parameters qN or NPQ; and (2) the partitioning of
1176
light energy among different processes, as assessed using the quantum yields ϕPSII, ϕNPQ, and ϕNO.
1177
These derived chlorophyll fluorescence parameters are calculated from a few basic measurements
1178
taken during the dark- and light-acclimated measurements, including: Fo, the minimum fluorescence yield
1179
of a dark-acclimated sample with all PSII reaction centers open; Fo’, the minimum fluorescence yield of a
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1180
light-acclimated sample with all PSII reaction centers open (following far-red light exposure); and Fm, the
1181
maximum fluorescence yield of a dark-acclimated sample following a saturating pulse. Fv, the variable
1182
fluorescence yield of a dark-acclimated sample, is calculated as Fm – Fo.
1183
The parameters qN and NPQ are meant to indicate the rate constant of the thermal dissipation of
1184
energy from PSII (Bilger and Schreiber 1986; Bilger and Björkman 1990). These parameters increase
1185
with the use of NPQ to dissipate light—in this case, under exposure to 1000 µmol m-2 s -1. They are
1186
calculated, respectively, as:
𝑞𝑁 = 1 −
1187
𝐹𝑚′ − 𝐹𝑜′
𝐹𝑚 − 𝐹𝑚′
; 𝑁𝑃𝑄 =
𝐹𝑚 − 𝐹𝑜
𝐹𝑚′
1188
In our dataset, qN is highly correlated (R2 = 0.96) with the natural logarithm of NPQ; we chose to report
1189
qN to easily visualize variation among leaves with low NPQ and to help satisfy the error term normality
1190
assumption of OLS regression.
1191
The quantum yields ϕPSII, ϕNPQ, and ϕNO represent the proportions of absorbed light energy that are
1192
dissipated through photochemical processes, non-photochemical quenching, and non-regulated
1193
dissipation, respectively. As proportions, these quantum yields sum to 1. We calculated these parameters
1194
following the “lake model” derivation of Kramer et al. (2004), which treats photosynthetic reaction
1195
centers as connected by shared antennae. We achieve qualitatively similar results using the simpler
1196
analogues found in Hendricksen et al. (2005).
1197
The parameter ϕPSII is as calculated above (Supplemental Methods, Photosynthetic light-response
1198
curve procedure). In the derivation from Kramer et al. (2004), ϕNPQ and ϕNO are calculated using the
1199
parameter qL, an estimate of the fraction of open PSII reaction centers. qL is calculated as:
𝑞𝐿 =
1200
1201
1202
Given this parameter, we can calculate ϕNO as:
𝜙𝑁𝑂 =
𝐹𝑚′ − 𝐹𝑠 𝐹𝑜′
𝐹𝑚′ − 𝐹𝑜′ 𝐹𝑠
1
𝐹
𝑁𝑃𝑄 + 1 + 𝑞𝐿 ( 𝐹𝑚 − 1)
𝑜
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1203
and given the other two quantum yields, we can calculate ϕNPQ as:
1204
𝜙𝑁𝑃𝑄 = 1 − 𝜙𝑃𝑆𝐼𝐼 − 𝜙𝑁𝑂
1205
Many researchers, including Murchie and Lawson (2013) and Tietz et al. (2017) treat ~0.83 as a
1206
theoretical maximum of ϕPSII or dark-acclimated Fv/Fm after a period of dark-acclimation long enough to
1207
reduce non-photochemical quenching to zero. Similarly, one may treat ~0.83 as a theoretical maximum of
1208
ϕPSII + ϕNPQ, deviations from which indicate high ϕNO, and thus, a declining ability to dissipate light
1209
energy in a regulated way. We indicate this putative maximum in our ternary plot of the quantum yields
1210
(Fig. S5).
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
References
Bilger W, Björkman O. Role of the xanthophyll cycle in photoprotection elucidated by measurements of
light-induced absorbance changes, fluorescence and photosynthesis in leaves of Hedera canariensis.
Photosynthesis Research 25, 173–185 (1990).
Bilger W, Schreiber U. Energy-dependent quenching of dark-level chlorophyll fluorescence in intact
leaves. Photosynthesis Research 10, 303–308 (1986).
Farquhar GD, von Caemmerer S, Berry JA. A biochemical model of photosynthetic CO2 assimilation in
leaves of C 3 species. Planta 149, 78–90 (1980).
Hendrickson L, Förster B, Pogson BJ, Chow WS. A simple chlorophyll fluorescence parameter that
correlates with the rate coefficient of photoinactivation of Photosystem II. Photosynthesis Research 84,
43–49 (2005).
Kramer DM, Johnson G, Kiirats O, Edwards GE. New Fluorescence Parameters for the Determination of
QA Redox State and Excitation Energy Fluxes. Photosynthesis Research 79, 209–218 (2004).
Murchie EH, Lawson T. Chlorophyll fluorescence analysis: a guide to good practice and understanding
some new applications. Journal of Experimental Botany 64, 3983–3998 (2013).
Sendall, KM, Reich PB, Zhao C, Jihua H, Wei X, Stefanski A, Rice K, Rich RL, Montgomery RA.
Acclimation of photosynthetic temperature optima of temperate and boreal tree species in response to
experimental forest warming. Global Change Biology 21, 1342–1357 (2015).
Tietz S, Hall CC, Cruz, JA, Kramer DM. NPQ(T): a chlorophyll fluorescence parameter for rapid
estimation and imaging of non-photochemical quenching of excitons in photosystem-II-associated
antenna complexes. Plant, Cell & Environment 40, 1243–1255 (2017).
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
Fig. S1
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1259
1260
Fig. S2
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1261
1262
Fig. S3
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1263
Fig. S4
1264
1265
Fig. S5
1266
1267
bioRxiv preprint doi: https://doi.org/10.1101/845701; this version posted November 18, 2019. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
under aCC-BY 4.0 International license.
1268
1269
Fig. S6