Académique Documents
Professionnel Documents
Culture Documents
C) Tmax (
a +b
n
N
R
a
(1)
where R
a
is extraterrestrial SRD (MJ m
2
d
1
); a and b are the AP
coefcients; andnandNis actual andtheoretical sunshineduration,
respectively.
2.2. Analyses
Statistical models that relied on information frommultiple sta-
tions, namely panel models, were documented to be better at
predicting crop responses to temperature change than time-series
statistical model at eachstation(Lobell andBurke, 2010). Therefore,
for each zone, the time-series data from the respective weather
stations were combined into a panel dataset. A panel analysis was
subsequently carried out for each zone.
To investigate climate change and its impact on crop yield at
eachgrowthstage, the whole growthperiod (GPw) of winter wheat
were divided into four growth periods, i.e., VGP1 (from emer-
gence to winter dormancy), VGP2 (fromwinter dormancy to green
up), VGP3 (from green up to anthesis), and RGP (from anthesis to
94 F. Tao et al. / Agricultural and Forest Meteorology 189190 (2014) 91104
Fig. 2. Trends in Tmean, precipitation, and SRD during different growth periods of wheat from 1981 to 2009 in each wheat production zone. The error bar represents the
standard error of the estimates. The trends with a mark a are signicant at 0.05 level, and with a mark b are signicant at 0.01 level.
F. Tao et al. / Agricultural and Forest Meteorology 189190 (2014) 91104 95
Table 3
Trends in wheat phenological dates and the lengths of each growth period during 19812009, in the major wheat production zones of China. Trends with * and ** are
signicant at the 0.05 and 0.01 level, respectively.
Zone I Zone II Zone III Zone IV Zone V Zone VI Zone VII Zone VIII
Emergence date (days/decade) 1.45* 0.81 3.17** 1.37** 4.03** 1.46 1.22 0.65
Dormancy date (days/decade) 1.51* 0.78 1.63 2.37**
Greenup date (days/decade) 1.40 1.44** 0.88 2.05
Anthesis date (days/decade) 2.93** 4.50** 4.85** 2.84** 4.78** 0.60 0.07 3.27**
Maturity date (days/decade) 2.72** 3.61** 3.11** 2.41** 3.75** 2.28 0.57 2.84**
VGP1 length (days/decade) 0.03 1.07 0.38 1.66*
VGP2 length (days/decade) 2.93* 2.32* 1.75 4.39**
VGP3(VGP) length (days/decade) 2.55** 3.13** 4.68** 1.39** 2.73** 0.87 1.33*
RGP length (days/decade) 0.21 0.70** 1.73** 0.35 1.03** 1.69** 0.67 0.43
GPw length (days/decade) 4.19** 4.58** 0.34 1.04 7.75** 0.82 0.65 2.16*
maturity). The GPw of spring wheat was divided into two growth
periods, i.e., vegetative growth period (VGP from emergence to
anthesis) and reproductive growth period (RGP, from anthesis to
maturity).
For each zone, time trends in wheat emergence dates, anthe-
sis dates and maturity dates during 19812009, as well as time
trends in T
mean
, precipitation and SRD during each growth period,
i.e., VGP1, VGP2, VGP3, RGP, and GPw for winter wheat, VGP, RGP
and GPw for spring wheat, were analyzed using linear regression
method.
To investigate the correlations between annual yields with cli-
mate variables during each growth period across the stations for
each zone, the yields were rstly linearly de-trended to get the
de-trended yield series that were mainly affected by seasonal cli-
mate variability. Then the climate variables were also linearly
de-trended, and the partial correlation analysis was applied to
investigate the correlation between the de-trended yields series
andthe de-trendedclimate variables series during 19812009. Sta-
tistical signicance was tested using the two-tailed t-test.
To avoid the confounding effects of highly correlated climate
variables, for each zone and each growth period, seven panel
regression models with different predictors were estimated to
quantifytheuncertainties inestimatingcropresponses tomajor cli-
mate variables. The sevenpanel regressionmodels were as follows:
Yd
i,t
=
i,0
+
1
t +
2
T
mean
i,t
+
i,t
(2)
Yd
i,t
=
i,0
+
1
t +
2
T
mean
i,t
+
3
P
i,t
+
i,t
(3)
Yd
i,t
=
i,0
+
1
t +
2
T
mean
i,t
+
4
SRD
i,t
+
i,t
(4)
Yd
i,t
=
i,0
+
1
t +
2
T
mean
i,t
+
3
P
i,t
+
4
SRD
i,t
+
i,t
(5)
Yd
i,t
=
i,0
+
1
t +
3
P
i,t
+
i,t
(6)
Yd
i,t
=
i,0
+
1
t +
3
P
i,t
+
4
SRD
i,t
+
i,t
(7)
Yd
i,t
=
i,0
+
1
t +
4
SRD
i,t
+
i,t
(8)
where Yd
i,t
is annual yield observations (not detrended) at station
i in year t.
i,0
represents an intercept for each station i.
1
rep-
resents the linear time trend of observed yields mainly due to
the long-term climatic and no-climatic trends including improve-
ments in varieties, technology, management and policy during the
study period.
14
are model parameters to be t, and
i,t
is an
error term. T
meani,t
, P
i,t
, and SRD
i,t
represent the linearly de-
trended growing season averages for T
mean
, precipitation, and SRD,
respectively, at station i in year t.
The parameter
2
in four panel regress models, i.e., Eqs. (2)(5),
represents different estimates of yields sensitivity to T
mean
during
a growth period. Likewise, the parameter
3
in four panel regress
models, i.e., Eqs. (6), (3), (7) and (5), represents different estimates
of yields sensitivity to precipitation during a growth period. The
parameter
4
in four panel regress models, i.e., Eqs. (8), (4), (7)
and (5), represents different estimates of yields sensitivity to SRD
during a growth period.
For each zone, the sensitivity of yield change to T
mean
, precipi-
tation and SRDchange during a growth period, i.e. the panel model
parameter
2
,
3
, and
4
, respectively, was estimated using multi-
ple regression method based on the trial data on yield and climate
from 1981 to 2009 across the stations in the zone. The sensitivity
was further expressed in percentage of actual mean yield across
the stations during the study period as
i
Ydmean
100%,where
i
is the panel model parameter
2
,
3
,
and
4
. Yd
mean
is the actual mean yield across the stations in a
zone during 19812009.
For each zone and each growth period, the impact of change
in a climate variable (i.e., T
mean
, precipitation and SRD) on crop
yield during 19812009 (expressed in percentage of actual mean
yield) was estimated by multiplying the sensitivity of yield change
to the climate variable with the magnitude of change in the climate
variable across the stations in the zone during the study period. The
later was estimated by a linear trend.
For each zone and each growth period, the joint impact of
climate change on crop yield during 19812009 (expressed in per-
centage of actual mean yield) was computed by summing the
impacts of changes in T
mean
, precipitation and SRD on crop yield
during the study period.
Table 4
Correlations between length of each growth period and Tmean, as well as between length of each growth period and yield. The correlations with a mark * are signicant at
0.05 level, and with a mark ** are signicant at 0.01 level.
Zone I Zone II Zone III Zone IV Zone V Zone VI Zone VII Zone VIII
VGP1 length and Tmean 0.02 0.07* 0.14 0.38**
VGP2 length and Tmean 0.44** 0.58** 0.29** 0.62*
VGP3 (VGP) length and Tmean 0.62** 0.46** 0.77** 0.69** 0.36** 0.71** 0.22**
RGP length and Tmean 0.45** 0.49** 0.49** 0.36** 0.05 0.80** 0.45** 0.30**
GPw length and Tmean 0.60** 0.44** 0.01 0.65** 0.52** 0.86** 0.14 0.36**
VGP1 length and yield 0.08 0.003 0.01 0.19**
VGP2 length and yield 0.25** 0.14** 0.08 0.26**
VGP3 (VGP) length and yield 0.04 0.13** 0.11 0.05 0.04 0.02 0.05
RGP length and yield 0.19** 0.06 0.10 0.22** 0.22** 0.19** 0.05 0.06
GPw length and yield 0.27** 0.11** 0.08 0.15** 0.33** 0.13 0.08 0.08
96 F. Tao et al. / Agricultural and Forest Meteorology 189190 (2014) 91104
Fig. 3. Correlations betweenwheat yield and Tmean, precipitation, and SRDduring different growthperiods from1981 to 2009 ineachwheat productionzone. The correlations
with a mark a are signicant at 0.05 level, and with a mark b are signicant at 0.01 level.
F. Tao et al. / Agricultural and Forest Meteorology 189190 (2014) 91104 97
Fig. 4. Estimated wheat yield changes by four panel regression models for each 1
C increase in Tmean during different growth periods in each wheat production zone. The
error bar represents the standard error of the estimates.
3. Results
3.1. Climate trends in the major wheat production zones from
1981 to 2009
From 1981 to 2009, climate during wheat growth period had
changed signicantly in the major wheat production zones. In
general, T
mean
increased signicantly while precipitation and SRD
changed diversely across the zones (Fig. 2). During GPw, in Zone
I, Zone II, Zone III, Zone IV and Zone VIII, T
mean
increased sig-
nicantly while precipitation decreased. In Zone V and Zone VII,
both T
mean
and precipitation increased. In Zone VI, T
mean
decreased
slightly while precipitationincreasedsignicantly. Inaddition, pre-
cipitation change little during VGP1 and VGP2, however decreased
signicantly during VGP3(VGP) and RGP in Zone I, Zone II, Zone
III and Zone IV. SRD during GPw did not change signicantly
in most of the zones, except in Zone VII where it decreased
signicantly by 0.63MJ m
2
day
1
decade
1
. Nevertheless, SRD
during VGP1 and VGP2 decreased notably in Zone I, Zone II, Zone
III and Zone V.
3.2. Changes in wheat phenology and the relations to
temperature change
Wheat sowing date and subsequently emergence date was
delayed in Zone I, Zone II, Zone V, Zone VI and Zone VII, however
advanced signicantly in Zone III and Zone IV (Table 3). Anthe-
sis date and maturity date advanced signicantly in all the zones
except Zone VI and Zone VII (Table 3). As a result, the length of each
growth period before anthesis was generally shortened, however
the length of RGP was prolonged in all the zones (Table 3). Finally
the GPw was signicantly shortened in Zone I, Zone II, Zone V and
Zone VIII.
98 F. Tao et al. / Agricultural and Forest Meteorology 189190 (2014) 91104
Fig. 5. Estimated wheat yield changes by four panel regression models for 10% increase in precipitation during different growth periods in each wheat production zone. The
error bar represents the standard error of the estimates.
T
mean
was signicantly negatively correlated with the length
of each growth period at most of the cases (Table 4); suggesting
increase in T
mean
reduced the length of each growth period.
3.3. Correlations between wheat yields and climate variables
during each growth period
Winter wheat yield was positively correlated with T
mean
in
northern China including Zone I and Zone II, however was nega-
tively correlated with T
mean
in southern China including Zone III
and Zone VIII (Fig. 3). In northwestern China, spring wheat yield
was negatively correlated with T
mean
in Zone VI, but positively
correlated with T
mean
in high altitude Zone VII. The positive (neg-
ative) correlations suggested that yield was positively (negatively)
affected by increase in T
mean
.
Wheat yield was generally negatively correlated with pre-
cipitation in southern China, was insignicantly correlated with
precipitation in northern China (Fig. 3).
Wheat yield was positively correlated with SRD for winter
wheat in eastern China including Zone II and Zone III, and for spring
wheat innorthwesternChina including Zone IVandZone VI (Fig. 3).
However it was negatively correlated with SRDfor winter wheat in
Zone I and Zone V, as well as for spring wheat in high altitude Zone
VII.
3.4. Sensitivity of wheat yield to climate variables during each
growth period
The estimates on sensitivity of wheat yield to climate variables
from the four panel regression models were generally consistent
(Fig. 4). For each 1
C increase in T
mean
during GPw, yield increased
for winter wheat in Zone I, Zone II and Zone V of northern China
by 9.4%, 1.7% and 0.7%, respectively (Fig. 4). By contrast, yield
decreased for winter wheat in Zone III and Zone VIII of southern
China by 3.1% and 2.3%, respectively (Fig. 4); as well as for spring
wheat in Zone IV and Zone VI of northwestern China by 1.3% and
6.1%, respectively. Spring wheat yield in high altitude Zone VII
increased by 6.8% (Fig. 4).
For precipitation increase by 10% during GPw, yield decreased
for winter wheat in Zone I, Zone II and Zone V of northern China
by 0.4%, 0.4% and 0.2%, respectively; and in Zone III and Zone VIII
of southern China by 1.9%, and 1.7%, respectively. In northwestern
China, spring wheat yielddecreasedinZone IVandZone VII by 0.6%,
however increased in Zone VI by 1.3% (Fig. 5).
F. Tao et al. / Agricultural and Forest Meteorology 189190 (2014) 91104 99
Fig. 6. Estimated wheat yield changes by four panel regression models for 10% increase in SRD during different growth periods in each wheat production zone. The error bar
represents the standard error of the estimates.
For SRD increase by 10% during GPw, yield increased for winter
wheat in Zone II and Zone III of eastern China by 6.5% and 23.1%,
respectively; as well as for spring wheat in Zone IV and Zone VI of
northwestern China by 4.9% and 6.9%, respectively (Fig. 6). By con-
trast, yield decreased for winter wheat in Zone V of western China
and Zone VIII of southwestern China by 5.4% and 0.7%, respectively.
In addition, yield decreased for winter wheat in Zone I by 12.2%,
and for spring wheat in high altitude Zone VII by 7.6%.
3.5. Yield change due to climate change over each growth period
from1981 to 2009
Climate change from 1981 to 2009 had caused measurable
impacts on wheat yield in most of the zones. Due to increase in
T
mean
during GPwin the period, yield increased for winter wheat in
Zone I, Zone II andZone Vof northernChina by 13.3%, 6.5%and0.4%,
respectively. By contrast, yield decreased for winter wheat in Zone
III and Zone VIII of southern China by 3.8% and 4.0%, respectively
(Fig. 7). For spring wheat in northwestern China, yield decreased in
Zone IV by 2.3% but increased in high altitude Zone VII by 5.9% due
to increase in T
mean
during GPw; yield increased in Zone VI by 2.6%
due to decrease in T
mean
during GPw.
Due to decrease in precipitation during GPwfrom1981 to 2009,
yield increased for winter wheat in Zone I and Zone II of northern
China by 1.2% and 0.5%, respectively; in Zone III and Zone VIII of
southernChina by 1.5%and3.1%, respectively; andfor spring wheat
in Zone IVof northwestern China by 1.6% (Fig. 8). Due to increase in
precipitation during GPw, yield increased for spring wheat in Zone
VI by 8.3%, however decreased for winter wheat in Zone V and for
spring wheat in Zone VII by 0.3% and 3.3%, respectively.
Due to increase in SRD during GPw from 1981 to 2009, yield
increased for spring wheat in Zone IV and Zone VI of northwest-
ern China by 1.6% and 1.3%, respectively (Fig. 9). By contrast, yield
deceasedfor winter wheat inZone I by 1.4%. Due todecrease inSRD,
yield increased for winter wheat in Zone V and for spring wheat in
Zone VII of western China by 1.0% and 6.2%, respectively; however,
deceased for winter wheat in Zone II and Zone III by 3.9% and 7.9%,
respectively.
From 1981 to 2009, changes in T
mean
, precipitation and SRD
jointlyincreasedwheat yieldinnorthernChina, however decreased
100 F. Tao et al. / Agricultural and Forest Meteorology 189190 (2014) 91104
Fig. 7. Estimated wheat yield changes by four panel regression models due to observed changes in Tmean during different growth periods from1981 to 2008 in each wheat
production zone. The error bar represents the standard error of the estimates.
wheat yield in southern China. For example, wheat yield in Zone I,
Zone II, Zone IV, Zone V, Zone VI and Zone VII of northern China
increased by 12.9%, 7.3%, 0.9%, 1.1%, 12.3% and 8.9%, respectively;
however, decreased in Zone III and Zone VIII of southern China by
10.2% and 1.2%, respectively (Fig. 10).
4. Discussion
4.1. Spatial pattern of wheat yield response to climate change
from1981 to 2009
Impacts of climate change from 1981 to 2009 on wheat yield
had an explicit spatial pattern. The impacts were positive in north-
ern China including Zone I, Zone II, Zone IV, Zone V, Zone VI and
Zone VII, however negative in southern China including Zone III
and Zone VIII (Fig. 10). The spatial pattern is well consistent with
Tao et al. (2012a) that used the census yield data at county scale
andspatiallyextrapolatedphenological data basedonstations data.
Although the nal impacts were dependent on the joint roles of
the changes in all climate variables, the roles of one or two climate
variables dominated in a zone. Yield increase in Zone I was mainly
ascribed to increase in T
mean
, where temperature was generally
less than the optimumtemperature for wheat (Table 2). The mean
of optimum temperatures for wheat was 22.0
C, 4.9
C, 10.6
C,
21.0
Cand20.7
C, above
the optimum temperature. Increase in temperature reduced the
duration of RGP at some stations in the zone. From 1981 to 2009,
SRD during RGP increased signicantly which offset the negative
impacts of temperature increase on nal yield. Winter wheat yield
F. Tao et al. / Agricultural and Forest Meteorology 189190 (2014) 91104 101
Fig. 8. Estimated wheat yield changes by four panel regression models due to observed changes in precipitation during different growth periods from1981 to 2008 in each
wheat production zone. The error bar represents the standard error of the estimates.
in Zone V was most sensitive to SRD, where SRD and T
mean
during
RGP was the largest among all the zones; however SRD and T
mean
changed slightly during the study period and had slight impacts on
wheat yield. As in Zone IV, irrigated spring wheat in Zone VI was
alsosensitivetoSRDandT
mean
, whereT
mean
duringRGPwas 22.9
C.
During the study period, slight decrease inT
mean
andslight increase
in SRD, together with signicant increase in precipitation, jointly
increased wheat yield in the zone. In high altitude Zone VII, climate
was characterized by high irradiance, low atmospheric pressure
and low temperature, photo-inhibition on plant photosynthesis
occurred sometimes (Zhang and Tang, 2005). Spring wheat yield
in the zone increased due to increase in temperature and decrease
in SRD, however decreased due to increase in precipitation. In Zone
VIII, high temperature during RGP, water-logging, insects and dis-
ease were the major problems with wheat production (Jin, 1961).
Increase in temperature reduced yield however decrease in pre-
cipitation increased yield, and nally climate change during the
study period had slight negative impacts on wheat yield in the
zone.
The results showed that precipitation had relative less impact
on yields than SRD and temperature. Wheat at the stations was
well irrigated. Yields werenot sensitivetoprecipitationbecausethe
majority of the
3
in the multiple regression models were not sta-
tistically signicant inZones IVII (Supporting Information). Wheat
was pronetowater-logging, insects anddiseaseduringrainyseason
and high precipitation reduced crop yields particularly during RGP.
In Zone VI, temperature and SRD were quite high, which caused
high evapotranspiration and water requirements. The results also
showed that yields decreased with SRD increase in Zone I, Zone V
and Zone VII. Excess SRDcan photoinhibit photosynthesis and may
lead to photooxidative destruction of the photosynthetic appara-
tus (Long et al., 1994), such as in Zone V and high altitude Zone
VII (Zhang and Tang, 2005). In Zone I, there was a negative corre-
lation between SRD and minimum temperature, decreases of SRD
beneted yields because of the associated reduction in frost occur-
rence. In addition, there is experimental evidence that yields of
some crops can rise if small reductions in total radiation coincide
with increases in diffuse radiation (Stanhill and Cohen, 2001).
102 F. Tao et al. / Agricultural and Forest Meteorology 189190 (2014) 91104
Fig. 9. Estimated wheat yield changes by four panel regression models due to observed changes in SRD during different growth periods from 1981 to 2008 in each wheat
production zone. The error bar represents the standard error of the estimates.
The impacts of long-termclimate change we investigated here
do not explicitly account for the impact of extreme weather.
Extreme weather such as high temperature at crop critical growth
stages can play a critical role in affecting wheat growth and yield
(Challinor et al., 2010; Asseng et al., 2011; Lobell et al., 2012), which
should be investigated in further study.
4.2. Cultivar turnover, phenology change and yield change during
1981-2009, as well as the implications to adaptation
In the study, the adaptations including shifts in sowing date
and changes in crop cultivars photo-thermal traits were explicitly
accounted for when quantifying climate impacts on crop growth
and yields. The contributions of other adaptations suchas improve-
ments in varieties, technology, management and policy to crop
yield were implicitly described by a non-climatic factor linear
trend, i.e.,
1
in Eqs. (2)(8). Using data on management prac-
tices from experimental stations may have impacted the results
and obscure our ability to detect adaptation to climate change in
an explicit manner.
Wheat cultivars were shifted frequently, which, together with
improvement of agronomic management practices, contributed
notably toyieldincrease inthe past fewdecades (Zhang et al., 2005;
Zhou et al., 2007). For example, Zhang et al. (2005) indicated that
winter wheat yield at an experiment station in the North China
Plain (i.e., Zone II in this study) increased by 50% from 1982 to
2002 or 2.38% per year, and yield increase was associated with the
increase in kernel numbers per unit area without alternation of the
weight of kernels. Field trials using 47 leading common wheat cul-
tivars released during 19602000 in the North China Plain showed
that average annual genetic gainingrainyieldrangedfrom32.07 to
72.11kg/ha/year or from0.48% to 1.23% (Zhou et al., 2007). Largely
because of successful utilization of dwarng genes and the 1B/1R
translocation, the genetic improvement in grain yield was primar-
ily attributed to increased grain weight per spike, reduced plant
height, and increased harvest index (Zhou et al., 2007). These stud-
ies support our results that wheat yields generally increased from
F. Tao et al. / Agricultural and Forest Meteorology 189190 (2014) 91104 103
Fig. 10. Estimated wheat yield changes by four panel regression models due to joint changes in Tmean, precipitation and SRD during different growth periods from1981 to
2008 in each wheat production zone. The error bar represents the standard error of the estimates.
50.81kg/ha/year to 147.92kg/ha/year or from 1.19% to 3.20% per
year, with an average of 85.6kg/ha/year or 1.98% per year, across
the major wheat production zones except Zone VIII where yield
decreased by 2.16kg/ha/year (Table 2).
We found that wheat anthesis date and maturity date advanced
generally, andlengthof growthperiodbeforeanthesis andGPwwas
shortened, however length of RGP was signicantly prolonged. The
length of each growth period was signicantly negatively corre-
latedwithtemperature, suggesting wheat growthwas signicantly
affected by temperature change. In most cases, increase in tem-
perature reduced the length of growth period in the past three
decades, nevertheless the length of RGP increased signicantly due
to adoption of cultivars with high thermal-requirement and late-
maturity cultivars, as well as slight decrease in temperature during
RGP resulted fromadvancement of anthesis date (see also Tao et al.,
2012b). This adaptation option indeed worked since length of RGP
was signicantly correlated with wheat yield in Zone I, Zone IV,
Zone V and Zone VI (Table 4). The length of GPw was signicantly
shortenedinZone I, Zone II, Zone VandZone VIII (Table 3), however
the shortening was benecial to yield because there was a negative
correlation between the length of GPw and yield in these zones
(Table 4). The reasons underlying may be that early maturity can
prevent grain-lling period from high temperature stress or rainy
season (Tao and Zhang, 2013). The length of GPw was signicantly
positively correlated with yield in Zone IV, however the length of
GPw was shortened because spring wheat in the zone was more
sensitive to temperature increase (Tao et al., 2012b).
5. Conclusion
The experiment observations at 120 agricultural meteorological
stations spanningfrom1981to2009across China were usedtolook
insights into the response andadaptationof wheat growthandpro-
ductivity to climate change in different climate zones. Our studies
better represented crop system dynamics by using detailed phe-
nological records, consequently better accounted for adaptations
such as shifts in sowing date and crop cultivars, when quantifying
climate impacts on wheat yield. We found that climate change had
caused notable impacts on wheat growth and productivity across
the major wheat production regions in China although agronomic
management and cultivars turnover were continuing to play an
important role in increasing productivity and adapting to climate
104 F. Tao et al. / Agricultural and Forest Meteorology 189190 (2014) 91104
change. Our ndings suggest the changes in crop systemdynamics
and cultivars photo-thermal traits in the past decades, as well as
their possible changes in future; have to be sufciently taken into
account to improve the prediction of climate impacts and to plan
adaptations for future.
Acknowledgements
This study was supported by the National Science Founda-
tion of China (Project No. 41071030) the strategic pilot scientic
projects of the Chinese Academy of Science (Project Number
XDA05090308), and the National Key Programme for Developing
Basic Science (Project Number 2010CB950902). We acknowledge
greatly the anonymous referees for their valuable comments onthe
early version of this manuscript.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.agrformet.
2014.01.013.
References
Asseng, S., Foster, I., Turner, N.C., 2011. The impact of temperature variability on
wheat yields. Glob. Change Biol. 17, 9971012.
Asseng, F., Ewert, I., Rosenzweig, C., Jones, J.W., Hateld, J.L., 2013. Quantifyinguncer-
tainties in simulating wheat yields under climate change. Nat. Clim. Change,
http://dx.doi.org/10.1038/nclimate1916.
Challinor, A.J., Simelton, E.S., Fraser, E.D.G., Hemming, D., Collins, M., 2010. Increased
crop failure due to climate change: assessing adaptation options using models
and socio-economic data for wheat in China. Environ. Res. Lett. 5, 034012.
FAO, 2012. Food and Agriculture Organization of the United Nations (FAO), FAO
Statistical Databases; available at http://faostat.fao.org
Jin, S.B., 1961. China Wheat Cultivation. Agricultural Publisher, Beijing, pp. 626.
Liu, Y., Tao, F., 2013. Probabilistic change of wheat productivity and water use in
China for global mean temperature changes of 1, 2, and 3
C. J. Appl. Meteorol.
Climatol. 52, 14129.
Licker, R., Kucharik, C.J., Dor, T., Lindeman, M.J., Makowski, D., 2013. Cli-
matic impacts on winter wheat yields in Picardy, France and Rostov, Russia:
19732010. Agric. For. Meteorol. 176, 2537.
Lobell, D.B., Asner, G.P., 2003. Climate and management contributions to recent
trends in U.S. agricultural yields. Science 299, 1032.
Lobell, D.B., Burke, M.B., 2010. On the use of statistical models to predict crop yield
responses to climate change. Agric. For. Meteorol. 150, 14431452.
Lobell, D.B., Schlenker, W., Costa-Roberts, J., 2011. Climate trends and global crop
production since 1980. Science 333, 616620.
Lobell, D.B., Sibley, A., Ivan Ortiz-Monasterio, J., 2012. Extreme heat effects on wheat
senescence in India. Nat. Clim. Change 2, 186189.
Long, S.P., Humphries, S., Falkowski, P.G., 1994. Photoinhibition of photosynthesis
in nature. Ann. Rev. Plant Physiol. Plant Mol. Biol. 45, 633662.
Olesen, J.E., Trnka, M., Kersebaum, K.C., Skjelvag, Seguin, B., Peltonen-Sainio, P.,
Rossi, F., Kozyra, J., Micale, F., 2011. Impacts and adaptation of European crop
production systems to climate change. Eur. J. Agron. 34, 96112.
Ortiz, R., Sayre, K.D., Govaerts, B., Gupta, R., Subbarao, G.V., Ban, T., Hodson, D., Dixon,
J.A., Ortiz-Monasterio, J.I., Reynolds, M., 2008. Climate change: can wheat beat
the heat? Agric. Ecosys. Environ. 126, 4658.
Porter, J.R., Gawith, M., 1999. Temperatures and the growth and development of
wheat: a review. Eur. J. Agron. 10, 2336.
Prescott, J.A., 1940. Evaporation froma water surface in relation to solar radiation.
Trans. Royal Soc. Sci. Aust. 64, 114115.
Qiu, J.J., Tang, H.J., Frolking, S., Boles, S., Li, C., Xiao, X., Liu, J., Zhuang, Y.H., Qin, X.G.,
2003. Mappingsingle-, double-, andtriple-cropagricultureinChinaat 0.5
0.5