Environmental Research Letters

LETTER • OPEN ACCESS Related content

- Intensity of heat stress in winter
Impact of heat stress on crop yield—on the wheat—phenology compensates for the
adverse effect of global warming
importance of considering canopy temperature Ehsan Eyshi Rezaei, Stefan Siebert and
Frank Ewert

- Heat stress is overestimated in climate

To cite this article: Stefan Siebert et al 2014 Environ. Res. Lett. 9 044012 impact studies for irrigated agriculture
Stefan Siebert, Heidi Webber, Gang Zhao
et al.

- Irrigation offsets wheat yield reductions

from warming temperatures
View the article online for updates and enhancements. Jesse Tack, Andrew Barkley and Nathan
Hendricks

Recent citations
- Engineering Meteorological Features to
Select Stress Tolerant Hybrids in Maize
Gordan Mimi et al

- Combined influence of soil moisture and

atmospheric evaporative demand is
important for accurately predicting US
maize yields
A. J. Rigden et al

- Response of wheat grain quality to low

temperature during jointing and booting
stages—On the importance of considering
canopy temperature
Leilei Liu et al

This content was downloaded from IP address 112.205.198.94 on 28/02/2020 at 02:54

 Environmental Research Letters

Environ. Res. Lett. 9 (2014) 044012 (8pp) doi:10.1088/1748-9326/9/4/044012

Impact of heat stress on crop yield—on the

importance of considering canopy
temperature
Stefan Siebert1, Frank Ewert1, Ehsan Eyshi Rezaei1, Henning Kage2 and
Rikard Graß2
1
Institute of Crop Science and Resource Conservation, University of Bonn, Katzenburgweg 5, D-53115
Bonn, Germany
2
Institute of Crop Science and Plant Breeding, Agronomy and Crop Science Department,
Christian-Albrechts-Universität zu Kiel, Hermann-Rodewald-Str. 9, D-24118 Kiel, Germany

E-mail: s.siebert@uni-bonn.de

Received 21 February 2014, revised 25 March 2014

Accepted for publication 28 March 2014
Published 1 May 2014

Abstract
Increasing crop productivity while simultaneously reducing the environmental footprint of crop
production is considered a major challenge for the coming decades. Even short episodes of heat
stress can reduce crop yield considerably causing low resource use efficiency. Studies on the
impact of heat stress on crop yields over larger regions generally rely on temperatures measured
by standard weather stations at 2 m height. Canopy temperatures measured in this study in field
plots of rye were up to 7 °C higher than air temperature measured at typical weather station
height with the differences in temperatures controlled by soil moisture contents. Relationships
between heat stress and grain number derived from controlled environment studies were only
confirmed under field conditions when canopy temperature was used to calculate stress thermal
time. By using hourly mean temperatures measured by 78 weather stations located across
Germany for the period 1994–2009 it is estimated, that mean yield declines in wheat due to heat
stress during flowering were 0.7% when temperatures are measured at 2 m height, but yield
declines increase to 22% for temperatures measured at the ground. These results suggest that
canopy temperature should be simulated or estimated to reduce uncertainty in assessing heat
stress impacts on crop yield.
S Online supplementary data available from stacks.iop.org/ERL/9/044012/mmedia

Keywords: heat stress, crop yield, temperature, soil moisture, modelling, wheat, rye

1. Introduction tunnels controlling temperature during crop growth [4–9], the

As sustainably meeting increasing world food demand is a
major challenge for humanity [1–3], it is important to
understand factors constraining crop yields and resource use
efficiency. Evidence of the negative impact of heat stress on
major world food crop yields is considered increasingly
robust. Formerly studied in climate chambers or plastic
Content from this work may be used under the terms of the
Creative Commons Attribution 3.0 licence. Any further
distribution of this work must maintain attribution to the author(s) and the
title of the work, journal citation and DOI.
negative impacts of heat stress on yield has recently been
confirmed for actual production conditions at large spatial
scales by the interpretation of field trials, statistical analyses
of large scale observational datasets and the application of
crop models [10–20].
Heat and drought have multiple, negative impacts on
crop yields, including reducing leaf photosynthesis and
enhancing leaf senescence rates. More critically for yield
determination, however, are the reported effects of decreasing
grain number when heat stress occurs before or around
anthesis and reduced grain weight when it occurs during grain

1748-9326/14/044012+08$33.00 1 © 2014 IOP Publishing Ltd

Environ. Res. Lett. 9 (2014) 044012 S Siebert et al

filling [6, 21–24]. While drought represents a process that

develops and intensifies slowly, heat stress can occur very
abruptly and even short episodes of high temperature can
cause a severe decline in grain yields. This reality that both
heat stress and drought occur together, makes it difficult to
assess the unique impact of heat stress on crop yields under
field conditions and bares the risk of confounding effects [23].
Previous work, has used statistical methods to analyze the
relationship between high temperature and crop yields for
different crops and regions [15, 20], though such methods are
not able to identify nor explain the specific processes
responsible for yield losses [16, 25]. Process-based crop
models are increasingly used to quantify the impact of heat
and drought on crop yield but differ with respect to the
stresses and processes considered [26–28]. A remaining
challenge is to implement relationships into crop models
describing the impact of heat stress around flowering [29, 30].
Reported relationships between heat stress and grain yield or
grain number indentified under controlled environment con-
ditions may not hold under field conditions as temperatures in
the canopy close to the flowers may diverge considerably
from weather station measurements [31], where air tempera-
ture is typically measured above a grass canopy at 2 m height.
Further, the degree of deviation of canopy and flower tem- Figure 1. Schematic illustration of the temperature measurements
peratures from weather station measurements is likely used in this study. Tc is canopy temperature in rye measured by two
dependent on crop type, cultivar and crop management. As infrared radiometers placed about 50 cm above the canopy while the
other measurements refer to air temperature measured at a height of
such, the uncertainties in assessments of heat stress impacts 350 cm (T350), 200 cm (T200), 20 cm (T20) above ground or at the
on crop yield due to air temperature measurements are not ground (Tg).
well reported. The objectives of this study were (i) to deter-
mine differences between air temperature at 2 m height and
canopy temperature under field conditions, (ii) to evaluate the
relationship between stress thermal time around flowering replications (1920 test plots, SI figure 1(c)) and accounted
calculated from temperatures measured at different heights therefore for a considerable part of the heterogeneity in
and grain number or grain yield, and (iii) to estimate the varieties, soils and water management as observed under real
uncertainty in heat stress assessments caused by the missing field conditions. The test plots of 1.25 m times 4 m size were
information on canopy temperature. The study was performed arranged in a rectangular setup with gaps of 0.1 m between
in Germany but the findings are of general relevance for the plots. Measurements were conducted under clear sky
cereal cultivation in temperate regions. conditions in the period 10 AM to 2 PM. Plant height varied
between 70 cm and 150 cm with the majority of plants with a
height of about 115 cm (irrigated) and 95 cm (rainfed). In
addition, air temperature (T200), humidity, wind speed and
2. Materials and methods
global radiation were measured at 200 cm height above
ground on a grass canopy by a nearby standard weather sta-
2.1. Experimental setup to determine differences between
tion. Temperatures measured at the standard weather station
canopy temperature and air temperature above the canopy
at 200 cm height were in good agreement to the measure-
To determine the range of differences between canopy tem- ments taken at 350 cm height at the rye fields but on average
perature and air temperature under field conditions, tem- 0.6 K lower during the measurement period (SI figure 1(b)).
perature was measured in 1920 rye plots on field trials of Mean wind speed during the measurement period was 2.0
KWS LOCHOW GmbH located at Wohlde (52.80 °N, m s−1, global radiation ranged between 526 W m−2 and 969
10.02 °E, 80 m a.s.l.) in northern Germany on 30th of May W m−2 and relative humidity between 40% and 54%. The soil
2012 (anthesis). Canopy temperature Tc was measured by two properties differed between an area in the southeast char-
infrared radiometers (Campbell IR 120) fixed to a carrier at acterized by sandy soil texture (Su3, 69% sand, 26% silt, 5%
50 cm on average above the canopy while air temperature clay) and a loamy part in the northwest (Su4, 53% sand, 41%
T350 was measured using a band-gap temperature sensor silt, 6% clay). Soil water content was measured by an FDR
(Campbell CS 215) at 350 cm above ground or about 250 cm probe (Sentek Diviner 2000) to a depth of 85 cm (measure-
above the rye canopy (figure 1, SI figure 1(a)). The test site ments every 10 cm, 8 different depths) for 19 loamy rainfed
consisted of 480 inbred lines and hybrids of rye grown under plots, 20 loamy irrigated plots, 12 sandy rainfed plots, and 20
rainfed and irrigated conditions on loamy or sandy soil in 2 sandy irrigated plots. Mean plant available water content in

2
Environ. Res. Lett. 9 (2014) 044012
the root zone was 17% for the rainfed sandy soil, 24% for the
rainfed loamy soil, 67% for the irrigated sandy soil, and 59%
for the irrigated loamy soil. The experiment at Wohlde was
however not suitable to test relationships between temperature
measured at different height and grain number or grain yield
because the temperature differences were measured on one
day only.
2.2. Impact of heat stress on winter wheat grain number and
yield
We tested the relationship between stress thermal time around
anthesis STT (°C minutes), computed as the accumulated
temperature sum above a threshold of 31 °C [6], and grain
number per ear or grain yield at winter wheat variety trials in
Saxony (Eastern Germany) during period 2001–2011. The
trials, performed by the State Office of Environment, Agri-
culture and Geology of the federal state of Saxony (http://
www.smul.sachsen.de/lfulg), are designed to test the perfor-
mance of cultivars registered for commercial growing in
Germany at test sites representing the range of climate and
soil conditions found in this federal state. Annual reports have
been available for the period 2001–2011 and contained data
for 7 test sites in Saxony. The number of varieties tested in
these trials varied between locations and years with a mini-
mum of 37 cultivars grown in year 2009 and a maximum of
45 cultivars tested in year 2005. For each year and test site the
mean of cultivar specific heading dates, grain number per ear
and grain yield was computed from the observations. While
heading date and grain yield was available for every year,
data on grain number per ear were missing for the years 2004
and 2011. Hourly mean of air temperature measured by agro-
climatic stations at the variety test sites at 20 cm height (T20,
°C) and at 200 cm height (T200, °C) were extracted from the
database of the agro-meteorological service of the state of
Saxony [32]. The automated weather stations were placed
above a short grass canopy. The sites Christgrün (50.58 °N,
12.21 °E, 430 m), Nossen (51.06 °N, 13.27 °E, 255 m), and
Roda (51.05 °N, 12.62 °E, 224 m) were selected for further
analysis because there was complete data coverage for both,
weather data and crop data, within the period 2001–2011. The
soil type at the test sites was sandy loam (Christgrün) or loam
(Nossen, Roda). Long-term annual precipitation sum was
722 mm yr−1 at Christgrün, 711 mm yr−1 at Roda and 643
mm yr−1 at Nossen, while long-term annual mean temperature
was 7.4 °C at Christgrün, 8.6 °C at Roda and 8.1 °C at Nos-
sen. Stress thermal time STT (°C minutes) was computed
from hourly mean temperature Th (°C) measured at 20 cm
height (STT20, °C minutes) and at 200 cm height (STT200, °C
minutes) by accumulating temperature sums above a thresh-
old Tcrit set to 31 °C for an 18-day period starting at the day of
heading:
STT = ∑ 60 MAX ( Th − Tcrit ; 0) (1)
Experiments undertaken before in climate chambers or
plastic tunnels showed that temperatures above this threshold
cause a reduction of grain number and grain yield of wheat
[6, 33]. The length of the period around flowering in which
3
S Siebert et al
heat stress was applied was 5–12 days in some experiments
[5, 34] but other reports indicate that the sensitive period may
be longer [24, 35, 36]. Teixeira et al [36] for example,
assumed a thermal-sensitive period of 30 days centered in the
mid-point of the reproductive phase. To account for the cul-
tivar differences in heading date and the typical length of the
period between heading and flowering at the test sites it was
decided to use a constant sensitive period of 18 days for this
study. STT20 and STT200 was then compared to mean grain
number per ear and relative yields Yrel (-) computed as ratio
between the yield in a specific year and maximum yield
observed at the research site during the period 2001–2011.
2.3. Uncertainty of heat stress impact around anthesis on
winter wheat in Germany
To estimate the uncertainty in STT arising from unknown
canopy temperatures for other locations in Germany, we
computed STT around flowering of winter wheat by using
hourly mean temperatures measured at the ground (Tg, °C)
or at 200 cm height (T200, °C) for 78 weather stations located
across Germany and period 1994–2009. We estimated the
corresponding relative crop yield by applying an equation
obtained from data collected under a controlled environment
Yrel ‒ heat = 1 − 0.000164 STT (2)
where Yrel-heat was grain yield reduced by the effect of heat
stress around flowering [37]. Tg and T200 were extracted
from a database providing observations for all weather sta-
tions that are maintained by the German Meteorological
Service [38]. We selected 78 weather stations located on—or
close to—areas classified as cropland according to the
Corine land cover classification 2006 [39] with complete
data coverage for the thermal sensitive phase around flow-
ering (18 days starting with heading of winter wheat) within
the period 1994–2009 and calculated stress thermal time at
200 cm height (STT200, °C min) and stress thermal time at
the ground (STTg, °C min). Heading day at each weather
station and each year was calculated by linear interpolation
(inverse distance weighting method with 12 neighbours
and power of 2) of phenological observations provided by
the WebWerdis database [38]. The observations of the
beginning of heading of winter wheat extracted from the
data base of the German Meteorological Service were first
filtered to exclude potential outliers or suspect data from the
analysis [40]. The number of heading day observations used
as input into the interpolation after application of this fil-
tering procedure varied between 502 in year 2007 and 1373
in year 1995. From the STT computed at the 2 different
heights, 78 stations and the period 1994–2009 we then
computed the annual mean across observation sites and the
mean across years for each station. In addition, the fre-
quency of years with heat stress around flowering of winter
wheat was compared.
Environ. Res. Lett. 9 (2014) 044012 S Siebert et al

3. Results

3.1. Comparison of canopy and air temperatures measured at

anthesis

Canopy temperature Tc was 19.5 °C–33.5 °C while air tem-

perature measured above the rye canopy T350 varied between
22.0 °C and 26.0 °C (figure 2(a)). In irrigated treatments, Tc
was on average 2 °C lower than T350 while in rainfed treat-
ments Tc was up to 7.5 °C higher than T350. The difference
between Tc and T350 was mainly controlled by soil moisture
(figure 2(a)) and net radiation (figure 2(b)). On sandy rainfed
plots, which had the lowest soil moisture content, Tc was
higher than T350 for almost all measurements. The tempera-
ture difference was low in the morning hours with low net
radiation and increased then to 3–7 °C in the hours with the
highest sun elevation and highest net radiation (figure 2(b)).
The critical temperature threshold of 31 °C reported for wheat
and other temperate cereals [5, 6, 41], was already surpassed
at a T350 of about 25 °C (figure 2(a)). In contrast, Tc was lower
than T350 for almost all measurements at irrigated plots with
largest differences of up to 6 °C in the morning hours
(figure 2(b)). In the period around midday with net radiation
between 400 and 700 W m−2, the temperature difference at
irrigated plots was 0–2 °C (figure 2(b)). Under irrigated
conditions, the canopy temperature Tc was always lower than
26 °C and thus well below the critical threshold of 31 °C.

3.2. Relationship between stress thermal time and grain yield

The decline in grain number per ear with increasing STT,

reported from experiments under controlled conditions
[5, 22, 41], was reproduced well when temperatures mea-
sured at 20 cm height were used to compute STT (figure 3)
while there was no relationship when standard air tempera-
ture measurements at 200 cm height were used (figure 3). A
similar result was obtained when STT was related to relative
crop yield (SI figure 2), which indicates that the decline in
grain number caused by heat stress around anthesis was not
compensated for in later growing stages. T20 at the three
sites in Saxony was up to 7 °C higher than T200 with annual
means of the temperature difference of 1.0–4.7 °C in periods
with a T20 larger than 31 °C (SI figure 2). Consequently,
STT20 was much higher than STT200 (figure 3 and SI figure
2). Air temperature in the period around anthesis of winter
wheat exceeded 31 °C in 5–7, depending on the site, out of
12 years when it was measured at 20 cm height but only in
2–3, depending on the site, out of 12 years when it was
measured at 200 cm height (SI figure 2). Linear trends of
relative yield on STT20 were in good agreement to relative
yield calculated by applying equation (2) which was derived
from experiments under controlled conditions [37]. The
slope of the trend lines for the three sites in Saxony, indi-
cating the yield reduction per °C minutes of STT, was
−0.000 134 for Christgrün, −0.000 224 for Nossen, and
−0.000 335 for Roda.
Figure 2. Relationship between air temperature 350 cm above
ground T350 and canopy temperature Tc (a) and between the
difference Tc − T350 and net radiation at the time of measurement (b)
in irrigated (filled symbols) and rainfed (unfilled symbols)
treatments, for sandy (squares) and loamy (circles) soil. Measure-
ments are shown for 1920 test plots of 480 inbred lines and hybrids
of rye at Wohlde (Germany) on 30 May 2012 (flowering).
Regression lines: loamy irrigated (solid), sandy irrigated (dashed),
loamy rainfed (dotted), sandy rainfed (dotdashed). Depending on
plant height, T350 was measured 200–280 cm above the canopy.
3.3. Uncertainty in estimates of heat stress impact on wheat
yield from air temperatures
We found that mean STT across all stations and years was
only 44 °C min when temperature measurements at 200 cm
height were used but 1366 °C min when temperature was
measured at the ground, with corresponding declines in crop
yield estimated by 0.7% and 22.4%, respectively. The mean
number of years with heat stress across all stations was 2.8
(out of 16 years) when temperature at 200 cm height was used

4
Environ. Res. Lett. 9 (2014) 044012
Figure 3. Relationship between mean grain number per ear and stress
thermal time calculated from hourly temperature measurements at
20 cm height STT20 (°C min) and at 200 cm height STT200 (°C min)
of winter wheat at three sites in Saxony (Germany). Stress thermal
time was computed for a period of 18 days starting at heading of
winter wheat by using a temperature threshold of 31 °C, only years
with heat stress during flowering (stress thermal time > 0 °C min) are
shown.
but 11.1 when temperature was measured at the ground
(figure 4). Inter-annual variability in STT was large with the
highest STT computed for year 2003 and the lowest in year
2001. The relative yield reduction due to heat stress estimated
for year 2003 was 75% when using data for temperature
measured at the ground and 3% when temperature measured
at 200 cm height was used (figure 4(a)). In addition, we found
large spatial differences in STT with almost no heat stress
along the coastline and in higher altitudes and hot spots of
heat stress with highest STT in the river valleys in Southern
Germany and in the lowlands of the southern part of Eastern
Germany (figure 4(b)). In contrast to the large uncertainty in
the magnitude of STT, its spatial pattern was found to be in
good agreement to means of daily maximum temperature in
months May and June, measured at 200 cm height
(figure 4(b), SI figure 3).
4. Discussion
Due to the thermal gradient in the canopy and typical crop heights of
German winter wheat between 70 cm and 100 cm, STT computed by
using ground temperature measured in a short grass canopy (e.g.
used in section 3.3) will likely represent an overestimate while using
air temperature measured at 200 cm height will underestimate STT
and corresponding yield losses, as shown in the previous section.
According to the German agricultural statistics [42], the lowest winter
wheat yield in period 1994–2009 was recorded in year 2003 (6.55
t ha−1) representing a 20% decline to the highest yield reported for
year 2004 (8.21 t ha−1), while mean yield reduction across all sites
due to heat stress estimated in this study for year 2003 was 75%
when using data for Tg and 3% when T200 was used (figure 3).
However, the weather stations, considered in section 3.3, were
selected based on the completeness of hourly temperature data
resulting in dense station coverage in heat affected regions of Ger-
many but low station coverage in the more humid and less heat
affected north-western part of Germany (SI figure 3). Therefore the
5
S Siebert et al
derived stress thermal time and corresponding yield reduction may
not be representative for the whole wheat growing area of Germany.
Heat stress with a canopy temperature of more than 31 °C
typically occurs in the early afternoon on warm days with
clear sky. Standard weather stations are placed on a short
grass so that the ground will heat up in this period resulting in
a temperature gradient of high temperature at the ground (Tg)
to lower temperature at 200 cm height (T200) and intermediate
temperature at 20 cm height (T20) which is also reduced by
the cooling effect of transpiration of the grass vegetation. The
results presented in sections 3.2 and 3.3 suggest that T20
measured on grass might be a better indicator of heat stress in
wheat than T200 or Tg. Unfortunately, T20 is only measured by
agro-meteorological stations in Germany so that the data
coverage for regional scale analyses is very limited.
Our results from the field experiment at Wohlde (sec-
tion 3.1) show that canopy temperature can differ sub-
stantially from the temperature measured at 200 cm height,
which is typically used to assess the impact of the heat stress
on agricultural crops. This is consistent to results of previous
research [13, 43, 44]. The magnitude of the difference
between these temperatures depends on weather variables like
incoming radiation, wind speed or humidity, on canopy
properties like canopy roughness, plant height or LAI, and on
actual soil moisture [13, 43, 44]. Under rainfed conditions,
that are typical for Germany, we propose that the impact of
heat stress during flowering would be underestimated for
major cereals like rye and wheat when air temperature mea-
sured at 200 cm height was used.
We suggest that crop models should be improved by
simulating canopy temperature instead of relying on 200 cm
air temperature measured at standard weather stations. An
alternative might be to use surface temperatures derived by
satellite imagery [45]. Interestingly, remote sensing products
are used to predict patterns of actual evapotranspiration and
corresponding drought stress based on differences in surface
temperature, e.g. by using the SEBAL or METRIC methods
[46–48]. However, currently available sensors do not have the
temporal and spatial resolution required for heat stress
assessments at field scale but may be used to determine the
soil moisture status around flowering at large extent. Because
of the dependency of canopy temperature on soil moisture this
information may be well suited to inform models simulating
canopy temperature.
Another alternative to account for the difference between
canopy temperature and 200 cm air temperature may be to use
air temperature measured at 200 cm height but to reduce the
critical temperature threshold determined by pot experiments
under controlled conditions (Tcrit in equation (1)) when cal-
culating the heat stress effect. Previous heat stress assess-
ments performed at the large scale used a wide range of
thresholds for the critical temperature when heat stress starts,
e.g. for wheat a range of 27–31 °C when daily maximum
temperatures were considered [5, 14, 49] or lower thresholds
of 22–28 °C when mean daytime temperature or mean tem-
perature in a season were considered [18, 36]. Reducing the
temperature threshold in the calculation of stress thermal time
represents however a simplification which does not account
Environ. Res. Lett. 9 (2014) 044012 S Siebert et al

Figure 4. Stress thermal time (STT) during flowering of winter wheat at 78 weather stations across Germany in the period 1994–2009 (°C
minutes). Mean STT across sites and respective mean relative yield reduction according to Ferris et al (2000) [37] computed from hourly
temperatures measured at 200 cm height T200 (open circles) and temperature at the ground Tg (filled circles) (a), mean STT across years and
number of years with a STT > 0 computed from T200 (left) or Tg (right) (b).

for the soil moisture status and other impacts on canopy temperature threshold of 31 °C and STT200 with a threshold
temperature and will thus tend to produce less precise results. ranging between 27 °C and 31 °C for the site Christgrün (SI
We demonstrate this by comparing STT20 calculated with a figure 4). To derive a STT200 similar to STT20 it would be

6
Environ. Res. Lett. 9 (2014) 044012
required to reduce the temperature threshold in the STT200
calculations at this location from 31 °C to 27 °C. The coef-
ficient of determination r2 is then, however, lower because a
high STT200 and corresponding yield loss is calculated for two
years when using this temperature threshold although STT20
and observed yield loss in these two years was small (SI
figure 4). This may point to the effect of other factors such as
soil moisture on canopy temperature which are less important
for air temperature at 200 cm height.
The strong interaction between heat and drought, implied
by our results, indicates that the impact of these stressors
needs to be assessed in future research simultaneously.
Because soil moisture was not measured at the agro-climatic
stations in Saxony, we cannot exclude that there was also
some drought stress in years with heat stress around flowering
at the sites used here to analyze heat stress effects on grain
number and grain yield (section 2.2). This could also explain
the different slopes of the trends shown in SI figure 2.
However, because of the suitable soil texture at the sites in
Saxony, a high water storage capacity can be expected which
facilitates storage of winter rains. Therefore drought stress
should be less frequent in the spring season when flowering
typically occurs in Germany. Our measurements at the site
Wohlde showed that the magnitude of heat stress differed for
the same temperature measured at the weather station at
200 cm height depending on the soil moisture status
(figure 1(a)). Similarly, the exposure to drought stress is
assumed to be higher on hot days with a high evaporative
demand of the atmosphere as compared to colder days, even
when the soil moisture content is the same [50, 51]. Previous
research indicated that also the effects of heat and drought
stress on grain yield are hypo-additive, the effect of combined
stress was higher than the individual effects but lower than
their sum [7]. In regional assessments it will therefore become
very difficult to disentangle occurrence and effects of heat and
drought, in particular when these stresses occur more than
once in a growing season. Here it may be suggested to define
a two dimensional stress function describing effects of heat
and drought simultaneously. Alternatively, crop models
should be able to account for these different mechanisms and
possible interactions. Another implication of our results is that
increased soil moisture (and corresponding increased evapo-
transpiration) due to irrigation has two positive effects on
crop growth and crop yield: reduction of drought stress and
reduction of heat stress. The cooling effect of irrigation has
been assessed and quantified before [31, 52–55], but the direct
positive impact on crop growth by reducing heat stress
requires more attention in further research.
Acknowledgements
Research support was provided by grants from the German
Research Foundation DFG (grant no. EW 119/5-1) and by the
Federal Ministry of Food and Agriculture BMEL via the
Agency for Renewable Resources FNR (grant no. FKZ
22013509).
7
S Siebert et al
References
[1] Foley J A et al 2011 Solutions for a cultivated planet Nature
478 337–42
[2] Godfray H C J, Beddington J R, Crute I R, Haddad L,
Lawrence D, Muir J F, Pretty J, Robinson S,
Thomas S M and Toulmin C 2010 Food security: the
challenge of feeding 9 billion people Science 327 812–8
[3] Tilman D, Balzer C, Hill J and Befort B L 2011 Global food
demand and the sustainable intensification of agriculture
P. Natl Acad. Sci. USA 108 20260–4
[4] Djanaguiraman M, Prasad P V V, Boyle D L and
Schapaugh W T 2013 Soybean pollen anatomy, viability and
pod set under high temperature stress J. Agron Crop Sci. 199
171–7
[5] Ferris R, Ellis R H, Wheeler T R and Hadley P 1998 Effect of
high temperature stress at anthesis on grain yield and
biomass of field-grown crops of wheat Ann. Bot. 82 631–9
[6] Porter J R and Gawith M 1999 Temperatures and the growth
and development of wheat: a review Eur. J. Agron. 10
23–36
[7] Pradhan G P, Prasad P V V, Fritz A K, Kirkham M B and
Gill B S 2012 Effects of drought and high temperature stress
on synthetic hexaploid wheat Funct. Plant Biol. 39 190–8
[8] Prasad P V V, Boote K J and Allen L H Jr 2006 Adverse high
temperature effects on pollen viability, seed-set, seed yield
and harvest index of grain-sorghum [Sorghum bicolor (L.)
Moench] are more severe at elevated carbon dioxide due to
higher tissue temperatures Agr. Forest Meteorol. 139
237–51
[9] Prasad P V V, Boote K J, Allen L H Jr, Sheehy J E and
Thomas J M G 2006 Species, ecotype and cultivar
differences in spikelet fertility and harvest index of rice in
response to high temperature stress Field Crop Res. 95
398–411
[10] Asseng S, Foster I and Turner N C 2011 The impact of
temperature variability on wheat yields Glob. Change Biol.
17 997–1012
[11] Deryng D, Conway D, Ramankutty N, Price J and Warren R
2014 Global crop yield response to extreme heat stress under
multiple climate change futures Environ. Res. Lett. 9 034011
[12] Gourdji S M, Sibley A M and Lobell D B 2013 Global crop
exposure to critical high temperatures in the reproductive
period: historical trends and future projections Environ. Res.
Lett. 8 024041
[13] Julia C and Dingkuhn M 2013 Predicting temperature induced
sterility of rice spikelets requires simulation of crop-
generated microclimate Eur. J. Agron. 49 50–60
[14] Liu B, Liu L, Tian L, Cao W, Zhu Y and Asseng S 2014 Post-
heading heat stress and yield impact in winter wheat of
China Glob. Change Biol. 20 372–81
[15] Lobell D B, Banziger M, Magorokosho C and Vivek B 2011
Nonlinear heat effects on African maize as evidenced by
historical yield trials Nat. Clim. Change 1 42–5
[16] Lobell D B, Hammer G L, McLean G, Messina C,
Roberts M J and Schlenker W 2013 The critical role of
extreme heat for maize production in the united states nature
Clim. Change 3 497–501
[17] Lobell D B, Sibley A and Ortiz-Monasterio J I 2012 Extreme
heat effects on wheat senescence in India Nat. Clim. Change
2 186–9
[18] Modhej A, Naderi A, Emam Y, Aynehband A and
Normohamadi G 2008 Effects of post-anthesis heat stress
and nitrogen levels on grain yield in wheat (T-durum and T-
aestivum) genotypes Int. J. Plant Prod. 2 257–67
[19] Moriondo M, Giannakopoulos C and Bindi M 2011 Climate
change impact assessment: the role of climate extremes in
crop yield simulation Clim. Change 104 679–701
Environ. Res. Lett. 9 (2014) 044012
[20] Schlenker W and Roberts M J 2009 Nonlinear temperature
effects indicate severe damages to US crop yields under
climate change P. Natl Acad. Sci. USA 106 15594–8
[21] Dolferus R, Ji X M and Richards R A 2011 Abiotic stress and
control of grain number in cereals Plant Sci. 181 331–41
[22] Prasad P V V, Pisipati S R, Momcilovic I and Ristic Z 2011
Independent and combined effects of high temperature and
drought stress during grain filling on plant yield and
chloroplast EF-Tu expression in spring wheat J. Agron Crop
Sci. 197 430–41
[23] Rajala A, Hakala K, Makela P, Muurinen S and
Peltonen-Sainio P 2009 Spring wheat response to timing of
water deficit through sink and grain filling capacity Field
Crop Res. 114 263–71
[24] Ugarte C, Calderini D F and Slafer G A 2007 Grain weight and
grain number responsiveness to pre-anthesis temperature in
wheat, barley and triticale Field Crop Res. 100 240–8
[25] Bakker M M, Govers G, Ewert F, Rounsevell M and Jones R
2005 Variability in regional wheat yields as a function of
climate, soil and economic variables: assessing the risk of
confounding Agr. Ecosyst. Environ. 110 195–209
[26] Asseng S et al 2013 Uncertainty in simulating wheat yields
under climate change Nat. Clim. Change 3 827–32
[27] Eitzinger J et al 2013 Sensitivities of crop models to extreme
weather conditions during flowering period demonstrated for
maize and winter wheat in Austria J. Agric. Sci. 151 813–35
[28] Rötter R P et al 2012 Simulation of spring barley yield in
different climatic zones of Northern and Central Europe: a
comparison of nine crop models Field Crop Res. 133 23–36
[29] Rötter R P, Carter T R, Olesen J E and Porter J R 2011 Crop-
climate models need an overhaul Nat. Clim. Change 1 175–7
[30] Siebert S and Ewert F Future crop production threatened by
extreme heat Environ. Res. Lett. 9 041001
[31] Durigon A and De Jong Van Lier Q 2013 Canopy temperature
versus soil water pressure head for the prediction of crop
water stress Agric. Water Manag. 127 1–6
[32] LfULG 2013 Agrarmeteorologisches Messnetz Sachsen -
Wetterdaten. Sächsisches Landesamt für Umwelt,
Landwirtschaft und Geologie http://www.landwirtschaft.
sachsen.de/Wetter09/asp/inhalt.asp?seite=uebersicht
accessed 25.04.2013
[33] Sanchez B, Rasmussen A and Porter J R 2014 Temperatures
and the growth and development of maize and rice: a review
Glob. Change Biol. 20 408–17
[34] Wheeler T R, Hong T D, Ellis R H, Batts G R,
Morison J I L and Hadley P 1996 The duration and rate of
grain growth, and harvest index, of wheat (Triticum
aestivum L) in response to temperature and CO2 J. Exp. Bot.
47 623–30
[35] Tashiro T and Wardlaw I F 1990 The response to high
temperature shock and humidity changes prior to and during
the early stages of grain development in wheat Funct. Plant
Biol. 17 551–61
[36] Teixeira E I, Fischer G, Van Velthuizen H, Walter C and
Ewert F 2013 Global hot-spots of heat stress on agricultural
crops due to climate change Agr. Forest Meteorol. 170
206–15
[37] Ferris R, Wheeler T R, Ellis R H, Hadley P, Wollenweber B,
Porter J R, Karacostas T S, Papadopoulos M N and
Schellberg J 2000 Climate Change, Climatic Variability and
S Siebert et al
Agriculture in Europe ed T E Downing et al (Oxford, UK:
Environmental Change Institute, University of Oxford))
pp 31–55
[38] DWD 2013 WebWerdis. Deutscher Wetterdienst (DWD)
https://werdis.dwd.de/werdis/start_js_JSP.do accessed
25.04.2013
[39] European Environment Agency 2009 Corine land cover 2006,
Version 12/2009 http://www.eea.europa.eu/data-and-maps/
data/ds_resolveuid/F2E46EFD-4FC6-4144-BE2A-
A57FD00F9C2E accessed 12.04.2011
[40] Siebert S and Ewert F 2012 Spatio-temporal patterns of
phenological development in Germany in relation to
temperature and day length Agr. Forest Meteorol. 152 44–57
[41] Wheeler T R, Craufurd P Q, Ellis R H, Porter J R and
Vara Prasad P V 2000 Temperature variability and the yield
of annual crops Agric. Ecosyst. Environ. 82 159–67
[42] Statistisches Bundesamt 2013 GENESIS-online database.
Statistisches Bundesamt https://www-genesis.destatis.de/
genesis/online accessed 25.04.2013
[43] Fuchs M 1990 Infrared measurement of canopy temperature
and detection of plant water-stress Theor. Appl. Climatol. 42
253–61
[44] Stockle C O and Dugas W A 1992 Evaluating canopy
temperature-based indices for irrigation scheduling Irrig.
Sci. 13 31–7
[45] Tomlinson C J, Chapman L, Thornes J E and Baker C 2011
Remote sensing land surface temperature for meteorology
and climatology: a review Meteorol. Appl. 18 296–306
[46] Long D and Singh V P 2012 A modified surface energy
balance algorithm for land (M-SEBAL) based on a
trapezoidal framework Water Resour. Res. 48 W02528
[47] Teixeira A H D C, Bastiaanssen W G M, Ahmad M D and
Bos M G 2009 Reviewing SEBAL input parameters for
assessing evapotranspiration and water productivity for the
low-middle Sao Francisco river basin, Brazil part A:
calibration and validation Agr. Forest Meteorol. 149 462–76
[48] Zwart S J and Bastiaanssen W G M 2007 SEBAL for detecting
spatial variation of water productivity and scope for
improvement in eight irrigated wheat systems Agric. Water
Manag. 89 287–96
[49] Semenov M A and Shewry P R 2011 Modelling predicts that
heat stress, not drought, will increase vulnerability of wheat
in Europe Sci. Rep-Uk 1 66
[50] Doorenbos J and Kassam A H 1979 Yield Response to Water
(Rome, Italy: FAO)
[51] Rao N H, Sarma P B S and Chander S 1988 A simple dated
water-production function for use in irrigated agriculture
Agric. Water Manage. 13 25–32
[52] Kueppers L M, Snyder M A and Sloan L C 2007 Irrigation
cooling effect: regional climate forcing by land-use change
Geophys. Res. Lett. 34 L03703
[53] Lobell D, Bala G, Mirin A, Phillips T, Maxwell R and
Rotman D 2009 Regional differences in the influence of
irrigation on climate J. Clim. 22 2248–55
[54] Lobell D B, Bonfils C J, Kueppers L M and Snyder M A 2008
Irrigation cooling effect on temperature and heat index
extremes Geophys. Res. Lett. 35 L09705
[55] Sacks W J, Cook B I, Buenning N, Levis S and Helkowski J H
2009 Effects of global irrigation on the near-surface climate
Clim. Dyn. 33 159–75