BGBiogeosciencesBGBiogeosciences1726-4189Copernicus PublicationsGöttingen, Germany10.5194/bg-13-5821-2016Ideas and perspectives: Heat stress: more than hot airDe BoeckHans J.hans.deboeck@uantwerp.beVan De VeldeHelenaDe GrooteToonNijsIvanCentre of Excellence PLECO (Plant and Vegetation Ecology), Department
of Biology, Universiteit Antwerpen (Campus Drie Eiken), Universiteitsplein
1, 2610 Wilrijk, BelgiumTerrestrial Ecology Unit, Department of Biology, Universiteit Gent,
K. L. Ledeganckstraat 35, 9000 Ghent, BelgiumThese authors contributed equally to this work.Hans J. De Boeck (hans.deboeck@uantwerp.be)24October201613205821582517March201621March201621September201612October2016This work is licensed under a Creative Commons Attribution 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by/3.0/This article is available from https://bg.copernicus.org/articles/13/5821/2016/bg-13-5821-2016.htmlThe full text article is available as a PDF file from https://bg.copernicus.org/articles/13/5821/2016/bg-13-5821-2016.pdf
Climate models project an important increase in the frequency and intensity
of heat waves. In gauging the impact on plant responses, much of the focus
has been on air temperatures, while a critical analysis of leaf temperatures
during heat extremes has not been conducted. Nevertheless, direct physiological
consequences from heat depend primarily on leaf rather than on air
temperatures. We discuss how the interplay between various environmental
variables and the plants' stomatal response affects leaf temperatures and the
potential for heat stress by making use of both an energy balance model and
field data. The results demonstrate that this interplay between plants and
environment can cause leaf temperature to vary substantially at the same air
temperature. In general, leaves tended to heat up when radiation was high and
when stomates were closed, as expected. But perhaps counterintuitively,
high air humidity also raised leaf temperatures, while humid conditions are
typically regarded as benign with respect to plant survival since they limit
water loss. High wind speeds brought the leaf temperature closer to the air
temperature, which can imply either cooling or warming (i.e. abating or
reinforcing heat stress) depending on other prevailing conditions. The
results thus indicate that heat waves characterized by similar extreme air
temperatures may pose little danger under some atmospheric conditions but
could be lethal in other cases. The trends illustrated here should give
ecologists and agronomists a more informed indication about which
circumstances are most conducive to the occurrence of heat stress.
Introduction
Modelled leaf-to-air temperature difference depending on
type of heat wave and stomatal conductance (gs). Type of heat wave:
high (a) or low (b) incident shortwave radiation (800 or 100 W m-2),
high or low relative humidity of the air (RH = 0.90 or 0.45), and calm or
windy weather (wind speed = 0.1 or 6 m s-1). Air temperature was set to
40 ∘C in all simulations, and leaf width to 0.005 m.
Current climate change has made heat waves more likely as both the
temperature mean and variability are increasing (Schär et al., 2004).
Several well-documented heat waves have occurred during the past years, such
as those of 2003 (Europe), 2010 (Russia) and 2012 (North America), and the
likelihood of such major events is expected to increase 5- to 10-fold within
the next 40 years (Barriopedro et al., 2011). Heat stress in plants is
usually observed when tissue temperatures exceed 40 ∘C, a threshold
that is fairly stable across biomes (Larcher, 2003). Such excessive
temperatures affect plant metabolism in multiple ways, ultimately reducing
growth and economic yield (Bastos et al., 2014; Chung et al., 2014). This
seems at odds with the reported lack of significant single-factor effects in
several ecological studies on heat waves (Poirier et al., 2012; Hoover et
al., 2014; De Boeck et al., 2016). We examine here how these seemingly
contrasting notions can be reconciled. The fundamental issue is that air
temperature (Ta) is often considered as an important indicator of
heat stress, while metabolic rates and physiological processes are affected
much more directly by leaf (tissue) temperatures (Tl). Many
studies on heat wave effects do not measure leaf or canopy temperatures and
report only on air temperatures (e.g. Bauweraerts et al., 2013; Filewod and
Thomas, 2014; Fernando et al., 2014), which suggests an underestimation of
the importance of Tl and the variables that influence it. Furthermore, in
models used to predict heat stress effects, air temperatures are still often
used instead of tissue temperatures, as noted by Webber et al. (2016)
regarding crop modelling, which can lead to inaccurate predictions of crop
yields (Siebert et al., 2014). From the literature on environmental biophysics
(e.g. Campbell and Norman, 1998; Jones, 2013), we know that leaf and tissue
temperatures are determined by a number of environmental conditions (apart
from Ta, primarily through radiation, wind speed and air
humidity) and the stomatal response of the plants. The extent to which these
variables can decouple leaf from air temperatures and therefore increase or
decrease the potential for heat stress during a heat wave of similar
magnitude (in terms of air temperature, as it is usually considered) is
discussed here by making use of both an energy balance model based on
established physical equations and field data.
Materials and methods
The model used to calculate leaf temperature is based on the energy balance
equation (Eq. 1):
Rs, in+Rl, in-Rl, out-H-λE= 0.
The equation states that an equilibrium is reached under a certain set of
environmental conditions (the flux of sensible heat H can be either incoming
or outgoing), whereby the sum of incoming energy (via shortwave radiation
Rs, in and longwave radiation Rl, in absorbed by the
leaf) and outgoing energy (outgoing longwave radiation Rl, out
and latent heat λE) is 0. The different terms are derived from
other equations, which feature both environmental variables such as wind
speed (u) and relative humidity (RH) of the air, leaf-scale parameters such
as stomatal conductance (gs) and characteristic leaf dimension
(d), and constants such as the Stefan–Boltzmann σ
(5.67e-8 W m-2 K-4). For more details, we refer to De Boeck
et al. (2012).
The leaf temperature is calculated in an iterative manner: as a starting
situation it is assumed that leaf and air temperature are equal, in which
case the energy budget equals 0. In any other situation, the model will
assume Tl to be lower (higher) than Ta if the energy
budget is negative (positive). The iteration proceeds in a stepwise manner
until a precision of 0.01 ∘C is achieved. The model was validated previously
(De Boeck et al., 2012), demonstrating a deviation between measured
and modelled leaf temperatures of less than 1.5 ∘C for over 90 %
of the cases. The model is freely available (see Supplement).
In this study, we set Ta at 40 ∘C to approximate the
general threshold for heat stress. Atmospheric pressure (which has limited
influence) was kept constant at 100 kPa. Emissivity, reflectivity and
absorptivity parameters for leaves and soil were used like in De Boeck et
al. (2012). In the main analyses, major inputs – namely incident shortwave
energy, stomatal conductance, wind speed and relative humidity of the air –
were varied in a dichotomous manner (high or low) to clearly illustrate the
direction of responses. More detailed analyses pairing input variables to
better illustrate interrelations are presented as supplementary material. We
focus on vegetation represented by species that have narrow leaves (like
those found in many grasses) with a characteristic dimension of 0.5 cm, but
we also consider the opposite end of the spectrum, namely very broad leaves
with a diameter of 20 cm.
The modelled results are supported by data recorded on 5 sunny days during
a heat wave in Belgium in 2015 (1–5 July). These data were collected at an
experimental site in Wilrijk, Belgium, on two homogeneous 10 cm tall young
grass stands sown 5 weeks earlier on homogenized soils (Fig. S1 in
Supplement). The grass was irrigated daily (ca. 5 L m-2), with the
exception of 1 day to test the impact of surface drying on the difference
between Ta and Tl. Radiation sensors (SR03-05,
Hukseflux Thermal Sensors, Delft, the Netherlands) had been installed
approximately 30 cm above the vegetation, with one sensor directed upwards
and one sensor directed downwards to measure absorbed radiation (the
difference between the two readings). At the same height, canopy temperature
was recorded with a non-contact thermometer (custom-made with a MLX90416ESF
sensor, Melexis, Tessenderlo, Belgium). Air temperature and relative humidity
were measured at 15 cm height (i.e. just above the canopy) in each plot
using a custom-made system (with a SHT75 RH/Ta sensor,
Sensirion AG, Staefa, Switzerland) shielded from the sun by a thin wooden
panel. To ensure that mostly data from times when direct sunshine reached the
plots was used (generally between 09:00 and 19:00 CET), we omitted
data points with absorbed radiation below 100 W m-2. This was done to
prevent artefacts from dew or times when stomates were still closed.
Results and discussion
Our results show that high radiation loads are an important prerequisite for
heat stress, unless air temperatures exceed the tissue heat stress threshold
significantly. Without the energy provided by significant amounts of
sunshine, plant tissues will almost always be cooler than the surrounding
air, regardless of other conditions (Figs. 1, S2–S4). In reality, heat waves
usually feature clear skies (De Boeck et al., 2010), implying that high
radiation loads during hot weather are probable. This also means that
experiments in which high air temperatures are imposed in low-radiation
environments, like under laboratory conditions or during overcast days, may
underestimate impacts.
Differences between leaf (Tl) and air (Ta)
temperature as a function of relative air humidity (RH) measured on a
homogeneous grass stand during 5 heat wave days (1–5 July 2015, Belgium).
The grass was irrigated daily (white circles), with the exception of 1 day
(black circles). The linear regression (white data points only) was
significant at p<0.001 (R2= 0.13). The difference between
regressions (white vs. black) was significant (ANCOVA; F1257= 10.3;
p= 0.001; GraphPad Prism). In contrast to the model runs, which focus
on one peak air temperature (40 ∘C) to obtain clean comparisons
between differing conditions, the relationship presented here contains more
scatter because of factors varying throughout the day such as air
temperature, incident radiation, stomatal conductance and wind speed.
As highlighted in earlier studies, water availability or lack thereof is
greatly relevant in gauging whether a heat wave will give rise to heat stress
(Salvucci and Crafts-Brandner, 2004). If drought prompts a plant to conserve
water by lowering stomatal conductance (gs), it warms up as
energy dissipation shifts from latent fluxes (providing cooling) to sensible
fluxes (increasing temperatures). Because heat and drought often co-occur
naturally (De Boeck et al., 2010), this effect is very relevant in assessing
heat wave impacts (Idso, 1982; De Boeck et al., 2016). The potentially
misleading nature of Ta in predicting heat stress under varying
stomatal conductance is clearly highlighted in our results (Figs. 1, S2,
S5–S6).
Whenever other conditions alleviate some amount of heat stress (e.g. less
radiation, higher gs), more wind counteracts such beneficial
effects (Figs. 1, S4–S5, S7) through closer coupling between the plant and
the air. This may seem counterintuitive as windiness is generally associated
with heat dissipation, but the same process also works in the opposite case:
when other environmental conditions exacerbate heat stress, more wind
reduces the increase of leaf temperatures. In other words, windy conditions
lead to avoidance of the most extreme cases of overheating. Obviously, higher
wind speeds promote evapotranspiration, resulting in faster depletion of soil
water reserves. This could subsequently lead to lower gs and thus
indirectly promote overheating. As wind speeds in laboratory conditions
and/or enclosures are often far below those observed outside (De Boeck et
al., 2012), canopy warming may be significantly different from outside as
calm conditions tend to exacerbate other effects (Figs. 1, S4–S5, S7).
For relative air humidity, the results are also counterintuitive, with higher
humidity more likely to give rise to heat stress (Figs. 1, S2, S6–S7). This
is caused by slower heat dissipation via transpiration as the water vapour
gradient between leaf and air is smaller than in the case of drier air. In
fact, the combination of low stomatal conductance and high air humidity
causes the greatest warming of leaves above the air temperature (Fig. 1). A
5-day period featuring air temperatures at vegetation height exceeding
30 ∘C every day provided us with an opportunity to test whether
increasing air humidity diminishes the cooling capacity of leaves. We indeed
found a significant relationship between RH and
Tl–Ta (Fig. 2), with ±0.84 ∘C change
per 0.1 increase in RH (excluding the dry day). This is comparable to the
slope (0.72 ∘C per 0.1 increase in RH) found with a model run using
conditions similar to the heat wave period (Fig. S8). Leaf cooling seemed to
be reduced on the only day during which irrigation was withheld (Fig. 2):
leaves were warmer than the air 32 % of the time on the dry day vs.
4 % on days with irrigation (even though incident radiation was ca. 15 %
lower on the dry day, while wind speed was similar). We attribute
this relative warming to stomatal closure (leaf wilting observed) resulting
from drying of the top soil and subsequent lower transpiration.
The aforementioned trends were observed both for simulations using narrow
leaves (Fig. 1) and for simulations using bigger leaves (Fig. S9). Any variable
increasing the heat load (high radiation) or decreasing heat dissipation
(high RH, low wind and gs) led to higher temperature increases in
big leaves than in small leaves, however. This is no surprise as larger
surfaces result in increased decoupling from air temperatures, which can lead
to extreme temperature deviations. In cushion plants, which physically act as
a giant leaf, increases of tissue temperatures of 20 ∘C and more
above the air temperature have been observed (Gauslaa, 1984), illustrating
the importance of physical dimensions in energy balances.
Calculations of leaf temperatures are possible at well-equipped sites by
applying a model such as the one used here. However, increasing quality and
decreasing costs of infrared imaging also enable direct quantification of
leaf temperatures and variability thereof. Infrared cameras allow the user to
select those pixels or zones deemed most appropriate (e.g. excluding bare
soil, focusing only on fully developed leaves), improving control and
versatility. Automated measurements and batch image processing can render the
entire process more efficient and allow for a high temporal resolution with
limited workload. Moreover, simultaneous measurements of incoming shortwave
radiation enable data filtering (e.g. clear sky, completely overcast),
further improving possibilities during data analysis. More technical
background information on extrapolation from leaves to canopies – dealing with
temperature variability, improving temperature accuracy and automated image
recognition – can be found in Jones et al. (2009), Jones and Vaughan (2010) and
Wang et al. (2010).
In conclusion, we clearly demonstrated that exceedance of critical
temperatures in plants depends on more variables than air temperature alone.
Radiation, wind speed and relative humidity all affect tissue temperatures,
depending on plant water status. This implies that heat waves characterized
by the same extreme air temperatures may cause little plant damage under
some conditions but could be detrimental to plant growth and survival in
other cases. Although heat stress also depends on other factors, like
hardening (Neuner and Buchner, 2012) and development stage (Fischer, 2011),
the results from this study can help predict when the probability of heat
stress occurring is most likely and can stimulate ecologists and
agronomists to shift the focus beyond merely air temperatures when
considering heat waves.
The Supplement related to this article is available online at doi:10.5194/bg-13-5821-2016-supplement.
Hans J. De Boeck and Ivan Nijs conceptualized the study. Toon De Groote developed
the model code, and Helena Van De Velde and Hans J. De Boeck performed the
simulations. Hans J. De Boeck carried out the field measurements. All authors
worked together to prepare the manuscript.
Acknowledgements
Hans Van De Velde was supported by FWO Vlaanderen. We thank Fred Kockelbergh for
technical assistance and the referees for valuable suggestions.
Edited by: A. Rammig
Reviewed by: two anonymous referees
References
Barriopedro, D., Fischer, E. M., Luterbacher, J., Trigo, R., and
Garcia-Herrera, R.: The Hot Summer of 2010: Redrawing the Temperature Record
Map of Europe, Science, 332, 220–224, 2011.Bastos, A., Gouveia, C. M., Trigo, R. M., and Running, S. W.: Analysing the
spatio-temporal impacts of the 2003 and 2010 extreme heatwaves on plant
productivity in Europe, Biogeosciences, 11, 3421–3435,
10.5194/bg-11-3421-2014, 2014.
Bauweraerts, I., Wertin, T. M., Ameye, M., McGuire, M. A., Teskey, R. O., and
Steppe, K.: The effect of heat waves, elevated [CO2] and low soil water
availability on northern red oak (Quercus rubra L.) seedlings, Glob. Change
Biol., 19, 517–528, 2013.
Campbell, G. S. and Norman, J. M.: An Introduction to Environmental
Biophysics, 2nd Edn. Springer, New York, 1998.
Chung, U., Gbegbelegbea, S., Shiferawe, B., Robertson, R., Yun, J. I.,
Tesfayea, K., Hoogenboom, G., and Sonder, K.: Modeling the effect of a heat
wave on maize production in the USA and its implications on food security in
the developing world, Weather Clim. Extremes, 5/6, 67–77, 2014.
De Boeck, H. J., Dreesen, F. E., Janssens, I. A., and Nijs, I.: Climatic
characteristics of heat waves and their simulation in plant experiments,
Glob. Change Biol., 16, 1992–2000, 2010.
De Boeck, H. J., De Groote, T., and Nijs, I.: Leaf temperatures in
glasshouses and open-top chambers, New Phytol., 194, 1155–1164, 2012.
De Boeck, H. J., Bassin, S., Verlinden, M., Zeiter, M., and Hiltbrunner, E.:
Simulated heat waves affected alpine grassland only in combination with
drought, New Phytol., 209, 531–541, 2016.Fernando, N., Panozzo, J., Tausz, M., Norton, R. M., Neumann, N., Fitzgerald,
G. J., and Seneweera, S.: Elevated CO2 alters grain quality of two bread
wheat cultivars grown under different environmental conditions, Agric.
Ecosyst. Environ., 185, 24–33, 2014.
Filewod, B. and Thomas, S. C.: Impacts of a spring heat wave on canopy
processes in a northern hardwood forest, Glob. Change Biol., 20, 360–371,
2014.
Fischer, R. A.: Wheat physiology: a review of recent developments, Crop
Pasture Sci., 62, 95–114, 2011.
Gauslaa, Y.: Heat resistance and energy budget in different Scandinavian
plants, Holarctic Ecol., 7, 1–78, 1984.
Hoover, D. L., Knapp, A. K., and Smith, M. D. Resistance and resilience of a
grassland ecosystem to climate extremes, Ecology, 95, 2646–2656, 2014.
Idso, S. B. Non-water-stressed baselines: a key to measuring and interpreting
plant water stress, Agr. Meteorol., 27, 59–70, 1982.
Jones, H. G.: Plants and Microclimate: A Quantitative Approach to
Environmental Plant Physiology: Third Edition. Cambridge University Press,
Cambridge, 2013.
Jones, H. G. and Vaughan, R. A.: Remote sensing of vegetation: principles,
techniques and applications. Oxford University Press, Oxford, 2010.
Jones, H. G., Serraj, R., Loveys, B. R., Xiong, L., Wheaton, A., and Price,
A. H.: Thermal infrared imaging of crop canopies for the remote diagnosis and
quantification of plant responses to water stress in the field, Funct. Plant
Biol., 36, 978–989, 2009.
Larcher, W.: Physiological plant ecology, 4th Edn., Springer-Verlag, Berlin,
2003.Neuner G. and Buchner, O.: Dynamics of tissue heat tolerance and
thermotolerance of PS II in alpine plants, in: Plants in alpine regions: cell
Physiology of adaptation and survival strategies, edited by: Lütz, C.,
Springer-Verlag, Vienna, 61–74, 2012.
Poirier, M., Durand, J. L., and Volaire, F.: Persistence and production of
perennial grasses under water deficits and extreme temperatures: importance
of intraspecific vs. interspecific variability, Glob. Change Biol., 18,
3632–3646, 2012.
Salvucci, M. E. and Crafts-Brandner, S. J.: Inhibition of photosynthesis by
heat stress: the activation state of Rubisco as a limiting factor in
photosynthesis, Physiol. Plantarum, 120, 179–186, 2004.
Schär, C., Vidale, P. L., Lüthi, D., Frei, C., Häberli, C.,
Liniger, M. A., and Appenzeller, C.: The role of increasing temperature
variability in European summer heatwaves, Nature, 427, 332–336, 2004.Siebert, S., Ewert, F., Rezaei, E. E., Kage, H., and Graß, R.: Impact of
heat stress on crop yield – on the importance of considering canopy
temperature, Environ. Res. Lett., 9, 044012,
10.1088/1748-9326/9/4/044012/meta, 2014.
Wang, X., Yang, W., Wheaton, A., Cooley, N., and Moran, B.: Efficient
registration of optical and IR images for automatic plant water stress
assessment, Comput. Electron. Agric., 74, 230–237, 2010.
Webber, H. Ewert, F., Kimball, B. A., Siebert, S. White, J. W., Wall, G. W.,
Ottman, M. J., Trawally, D. N. A., and Gaiseret, T.: Simulating canopy
temperature for modelling heat stress in cereals, Environ. Modell. Softw.,
77, 143–155, 2016.