IRRIGATION AND DRAINAGE

Irrig. and Drain. (2013)

Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/ird.1799

PREDICTING CLIMATE CHANGE IMPACTS ON MAIZE CROP PRODUCTIVITY AND

WATER USE EFFICIENCY IN THE LOESS PLATEAU†

YINHONG KANG1,2, XIAOYI MA2* and SHAHBAZ KHAN3

1
Department of Information and Engineering Technology, Sichuan Agricultural University, Ya’an 625014, China
2
College of Water Resources and Architectural Engineering, Northwest A & F University, Yangling 712100, China
3
Regional Science Bureau for Asia and the Pacific UNESCO, Jakarta 12110, Indonesia

ABSTRACT
Climate change impacts on food supply and water use efficiency have become an urgent issue for the agricultural researchers
and governments. With climate change, water resources will be a restricting factor for agriculture, especially in arid and semi-
arid areas. This paper mainly discusses the changing tendency of water balance components, maize yield and water use indices
under rainfed and irrigated conditions in subareas of the Loess Plateau under the A2 and B2 scenarios with the SWAGMAN®
Destiny model. The results show that evapotranspiration will increase in future for both rainfed and irrigated maize under both
scenarios. The water use indices mentioned in this paper have the same tendency under both scenarios in the same subareas, but
variability under the A2 scenario is higher than that under B2. The crop yield and water use indices will decrease for rainfed
maize production in the whole Loess Plateau. With irrigation, crop yield will increase in Clay Loam-2 and Clay Loam-1 while
it will decrease in Light Loam and Sandy Loam under the A2 scenario and B2 scenarios in 2020, 2050 and 2080 respectively.
The evapotranspiration efficiency and ratio of evapotranspiration and irrigation under irrigated maize will increase in Clay
Loam-2, Clay Loam-1, Light Loam and Sandy Loam under both scenarios. Crop water use efficiency, irrigation water use
efficiency and total water use efficiency will decrease under the A2 and B2 scenarios during irrigated maize growth. Therefore,
for the rainfed crop, it is necessary to improve water harvesting technology so as to meet the increasing crop water require-
ments with climate change; while, for the irrigated crop, it is vital to enhance soil water capacity in order to make good use
of the irrigated water. Copyright © 2013 John Wiley & Sons, Ltd.
key words: climate scenarios; water use indices; semi-arid area; yield; maize

Revised 27 July 2013; Accepted 29 July 2013

RÉSUMÉ
Les impacts du changement climatique sur l’approvisionnement alimentaire et l’efficacité d’utilisation de l’eau sont devenus un
problème urgent pour les chercheurs dans le domaine de l’agriculture et les gouvernements. Avec le changement climatique, les
ressources en eau seront un facteur limitant pour l’agriculture, en particulier dans la zone aride et semi-aride. Ce document traite
principalement de la tendance de l’évolution des composants du bilan hydrique, le rendement du maïs et des indices de
consommation d’eau dans des conditions pluviales et irriguées dans les sous-zones de plateau de loess sous les scénarios A2 et
B2 avec le modèle Destiny SWAGMAN ®. Les résultats montrent que l’évapotranspiration va augmenter à l’avenir à la fois pour
le maïs pluvial et irrigué dans les deux scénarios. Les indices de consommation d’eau mentionnés dans le présent document ont la
même tendance dans les deux scénarios dans les mêmes sous-zones, mais la variabilité dans le scénario A2 est plus élevée que dans
le cadre du B2-ci. Le rendement des cultures et des indices de consommation d’eau va diminuer la production de maïs pluvial dans
l’ensemble du Plateau de Loess. Avec l’irrigation, le rendement des cultures augmentera dans le limon argileux-1 et 2, alors qu’il
diminuera en limon léger et limon sableux dans le scénario scénarios A2 et B2 en 2020, 2050 et 2080, respectivement. L’efficacité
de l’évapotranspiration et le taux d’évapotranspiration et d’irrigation en maïs irrigué augmenteront dans tous les limons testés pour
les deux scénarios. L’efficacité de la collecte de l’eau, l’efficacité d’utilisation de l’eau d’irrigation et l’efficacité totale de

* Correspondence to: Xiaoyi Ma, College of Water Resources and Architectural Engineering, Northwest A & F University, Yangling 712100. Tel.: +86 130
8895 8810; Fax: +86 29 8708 2133. E-mail: [email protected]
†
Prédire les impacts du changement climatique sur la productivité des cultures de maïs et l’efficacité d’utilisation de l’eau dans le plateau de Lœss.

Copyright © 2013 John Wiley & Sons, Ltd.

 Y. KANG ET AL.

l’utilisation de l’eau vont diminuer dans les scénarios A2 et B2 au cours de la croissance de maïs irrigué. Par conséquent, pour la
culture pluviale, il est nécessaire d’améliorer la technologie de récupération de l’eau de manière à répondre aux besoins croissants
en eau des cultures avec le changement climatique, tandis que, pour l’irrigation-ci, il est essentiel de renforcer la capacité en eau du
sol afin de faire une bonne l’utilisation de l’eau d’irrigation. Copyright © 2013 John Wiley & Sons, Ltd.
mots clés: scénarios climatiques; indices de consommation d’eau; zone semi-aride; rendement; maïs

INTRODUCTION Plateau (Li et al., 2002). Wang and Liang (2002) have pro-
Climate change impacts on agriculture have been paid great posed that maize has impressive production potential in the
attention by researchers, governments, policy makers and Loess Plateau because of the suitable rainfall and temperature
the public in recent years. With climate change and increas- conditions. However, drought is a restraining factor for crop
ing population, food production and food security are be- production in this area because of the shortage and unequal
coming a bigger concern than ever (De Silva et al., 2007). distribution of water resources there (Liu et al., 2010). As cli-
Meanwhile, it is a big challenge for governments to provide mate change has uncertain impacts on crop growth and food
sufficient food for ever-increasing populations due to the in- production (Reddy and Hodges, 2000), it is therefore impor-
equality of food production and population distribution in tant to quantify climate change effects on crop growth and
the world. There are many factors related to food production yield and provide countermeasures to mitigate the negative
such as climate change impacts, extreme weather events, effects of climate change on food production.
farmland degradation, water availability and escalating food This paper aims to analyse climate change impacts on
prices. Of all the factors, climate change is a complex issue maize productivity and water use efficiency using the SWAG-
for crop production since it can alter soil water storage MAN® Destiny model under the A2 and B2 scenarios with
(Eitzinger et al., 2001), soil fertility (Sirotenko et al., rainfed and irrigated conditions in subareas of the Loess
1997) and the soil hydro-ecology system through multiple Plateau. The results in this paper may be of great value for
pathways (Zhang and Liu, 2005). agriculture production in other arid and semi-arid areas.
Meanwhile, water availability and enhanced atmospheric
CO2 levels also pose uncertainties for crop production MATERIALS AND METHODS
(Eitzinger et al., 2003). Anwar et al. (2007) used CropSyst
version-4 to predict climate change impacts on wheat yields The subareas in the Loess Plateau
in south-eastern Australia. Their results showed that ele- There are five subareas in the Loess Plateau, and each has a
vated CO2 levels can reduce the median wheat yield by different soil type. From south to north, the soil types are
about 25%. Akpalu et al. (2008) studied the climate change successively Heavy Clay Loam, Clay Loam-2, Clay
impacts on maize yield in the Limpopo Basin of South Loam-1, Light Loam and Sandy Loam (Li et al., 1985),
Africa and showed that increased rainfall can have a more shown in Figure 1. Since the Heavy Clay Loam soil only
positive effect for crop yield than temperature. Tojo et al. occupies a relatively small area compared with the other
(2007) analysed the impacts of planting dates and different four types, it is not studied in this paper. For each subarea,
weather on maize production in Brazil with CERES-Maize, one typical site was chosen for the study, namely Wugong
and the result shows that a later planting date will decrease for Clay Loam-2, Luochuan for Clay Loam-1, Ansai for
average yield by 55% under rainfed conditions and 21% un- Light Loam and Yulin for Sandy Loam respectively. The
der irrigated conditions. Yao et al. (2007) analysed CO2 soil hydraulic parameters of typical soil for each subarea
level impacts on rice yield with the CERES-Rice model in are shown in Table I (Kang et al., 2008).
the main Chinese rice production areas, which shows that
rice yield will increase with CO2 effect otherwise it will de-
Historical climate analysis
crease. The crop yield can be enhanced with irrigation appli-
cation and increasing precipitation during the crop growth Historical climate data are used to project and analyse future
period; what is more, crop yield is more sensitive to precip- climate scenarios. The historical climate analysis used in
itation than temperature (Kang et al., 2009). this paper is based on the scientific database of the Loess
Maize is planted in most of China owing to its adaptation Plateau from the Institute of Soil and Water Conservation,
to a wide range of temperature and precipitation (Xiong Chinese Academy of Sciences. Figure 2 shows the historical
et al., 2007a). The Loess Plateau is one of the main maize- mean monthly rainfall, Tmax and Tmin (1961–1990), in four
growing areas in China. The maize planting area in the subareas of the Loess Plateau respectively.
Loess Plateau is about 1.9 × 106 ha, almost 17.9% of the to- Figure 2 indicates that the highest temperature (including
tal planting area and 30.8% of crop production in the Loess Tmax and Tmin) in the four subareas all occurred in June, July

Copyright © 2013 John Wiley & Sons, Ltd. Irrig. and Drain. (2013)
 PREDICTING CLIMATE CHANGE IMPACTS ON MAIZE CROP PRODUCTIVITY

Figure 1. The soil texture subarea of the Loess Plateau

Table I. Soil hydraulic parameters of typical soils for each subarea in the Loess Plateau

Bulk density Saturated soil water Residual soil water Saturated hydraulic
Soil type (g cm–3) content (cm3 cm–3) content (cm3 cm–3) α– n– conductivity (cm d–1)

Clay Loam-2 in Wugong 1.30 0.095 0.4905 0.01616 1.3083 0.2598

Clay Loam-1 in Luochuan 1.35 0.078 0.5094 0.04364 1.2586 24.2588
Light Loam in Ansai 1.35 0.065 0.4905 0.09615 1.2842 7.8392
Sandy Loam in Yulin 1.45 0.035 0.4528 0.15836 1.2404 650

Source: Kang et al. (2008).

and August, and both of the highest values are in July. The temperature as well. Both temperature and rainfall are cru-
mean monthly rainfall gradually increased from January to cial for maize growth.
September, then decreased. The highest precipitation oc-
curred in July, August and September, while the lowest data
Future climate analysis
were from December to February. The average mean rainfall
is 602 mm in Wugong, 588 mm in Luochuan, 533 mm in Climate scenarios may be a reasonable description of the fu-
Ansai and 380 mm in Yulin. It is obvious that the rainfall ture climate, based on a range of climatological relationships
amount in the Loess Plateau decreases from south to north. (Intergovernmental Panel on Climate Change (IPCC),
However, the wet season in the Loess Plateau happens to 2007). Future climate characteristics are dependent on the
be at the key growing stage of maize, and the high emission scenarios published by the IPCC as the Special

Copyright © 2013 John Wiley & Sons, Ltd. Irrig. and Drain. (2013)
 Y. KANG ET AL.

160 Wugong
Luochuan
Table II. Rainfall and temperature change tendency under future
140
Ansai scenarios (compared with baseline 1961–1990)
120
Yulin
100
80
Scenario A2 Scenario B2
60
40 Time Temperature Rainfall Temperature Rainfall
20 slices increase (°C) increase (%) increase (°C) increase (%)
0
Jan Feb Mar Apr May June July Aug Sep Oct Nov Dec
2020 1.4 3.3 0.9 3.7
Month 2050 2.6 7.0 1.5 7.0
Wugong
2080 3.9 12.9 2.0 10.2
30
Luochuan
25 Ansai
Source: Xiong et al. (2007b).
20 Yulin

15
10
rainfall season vary in the subareas of the Loess Plateau. Cli-
5
0
mate change also will further change the crop growth period
-5
Jan Feb Mar Apr May June July Aug Sep Oct Nov Dec and harvesting time. What is more, it is summer maize in
Month Wugong and spring maize in other subareas. The growth pe-
20 Wugong riod of spring maize is longer than that of summer maize.
Luochuan
15
Ansai
10
SWAGMAN® Destiny model description
Yulin
5
0
-5 Jan Feb Mar Apr May June July Aug Sep Oct Nov Dec SWAGMAN® Destiny is a one-dimensional computer
-10 model to simulate crop growth and water balance in the
-15 soil–plant–atmosphere system, developed by the Griffith
-20 Laboratory, Commonwealth Scientific and Industrial
Research Organization (CSIRO) Land and Water Division,
Figure 2. Historical monthly mean rainfall, Tmax and Tmin in the typical soil
sites of the Loess Plateau (1961–1990). This figure is available in colour Australia. It is used to examine the impacts of particular crop
online at wileyonlinelibrary.com/journal/ird and irrigation management scenarios on soil water and salt
status, groundwater table depth and crop productivity. It can
simulate crop growth and yield in response to shallow ground-
Report on Emissions Scenarios (SRES). The IPCC provides water table levels, available soil water, waterlogging and
four climate scenarios to assess climate change, depending prevailing weather conditions. It enables formulation of
on different governance and orientation towards social and strategies to maximize productivity based on managing
environmental concerns, and other aspects (Xiong et al., groundwater tables and avoiding salinization. With long-term
2007b). In this paper we have chosen A2 (scenario of rapid re- weather data, different simulation sequences can be used to
gional economic growth and a heterogeneous world) and B2 assess a particular strategy by probabilistic analysis (Timsina
(scenario of environmental priority, clean and efficient technol- and Humphreys, 2003). The SWAGMAN® Destiny model-
ogies and a heterogeneous world) to evaluate climate change ling framework is shown in Figure 3.
impacts on maize production in subareas of the Loess Plateau. The simulation of crop growth in SWAGMAN® Destiny
The tendency of rainfall and temperature to change under is influenced by the amount of available radiant energy, tem-
future scenarios is based on the regional climate scenarios perature and the prevailing stresses. The crop growth rou-
which were generated following the methodology described tines of Destiny comprise simple procedures for estimating
by the United Kingdom Climate Impacts Programme daily increments in leaf area index (LAI), above-ground bio-
(UKCIP) (Xiong et al., 2007b) and shown in Table II. mass and root biomass and their response to aeration, nitro-
Future rainfall during the maize growth period in the Loess gen, water deficit and salt stress (Godwin et al., 2006). For
Plateau is presented in Table III. annual grain and fibre crops, yield is determined from an in-
Table III indicates that rainfall in each subarea of the Loess put potential yield modified by the rate at which simulated
Plateau will increase in 2020, 2050 and 2080 under the A2 dry matter production is reduced by prevailing stresses.
and B2 scenarios. Meanwhile, the incremental percentage Biomass is estimated through the amount of radiant energy
increase of rainfall will be higher from 2020 to 2080. intercepted by actual LAI and the conversion efficiency of
There are some differences in the amplitude of rainfall in- intercepted radiation into biomass for each species.
crease between Tables II and III from 2020 to 2080 for both SWAGMAN® Destiny has three modes to be chosen from
scenarios. It is because the maize growth period and the simulation control, namely continuous detail, continuous with

Copyright © 2013 John Wiley & Sons, Ltd. Irrig. and Drain. (2013)
 PREDICTING CLIMATE CHANGE IMPACTS ON MAIZE CROP PRODUCTIVITY

Table III. Comparison of rainfall under future climate scenarios in subareas of the Loess Plateau with the baseline (1961–1990) during maize
growth

2020 2050 2080

Scenario Subarea Baseline rainfall (mm) Rainfall (mm) Δ (%) Rainfall (mm) Δ (%) Rainfall (mm) Δ (%)

A2 Wugong 320 331 3.30 343 6.92 361 12.81

Luochuan 417 421 0.81 435 4.17 459 9.91
Ansai 397 399 0.28 412 3.62 433 8.92
Yulin 297 301 1.55 308 3.92 326 9.65
B2 Wugong 320 332 3.58 343 6.92 353 10.10
Luochuan 417 422 0.81 435 4.17 449 7.52
Ansai 397 402 1.03 412 3.62 426 7.11
Yulin 297 299 0.71 306 2.99 318 7.03

Δ, rainfall increase.

Graphics and files

Crop Data
Weather Data Destiny Shell

Soil Data
Output Files
(include soil water Data
balance, plant growth Files
and development)
Crop Growth Model

Figure 3. Framework of SWAGMAN® Destiny

statistics and strategic mode. Continuous detail mode allows Water use indices
the specified crop growth scenario to be run over a chosen
number of years, and it is a useful method to examine the de- Water use indices in this paper are defined in Table IV
tailed functions of crop growth and water movement. Contin- referred to Edraki et al. (2003). Water use indices in this pa-
uous mode with statistics is used to estimate what may happen per include evapotranspiration efficiency, ratio of evapo-
if the same irrigated soil and crop scenario is repeated transpiration and irrigation, crop water use efficiency,
throughout the simulation period. The strategic mode is irrigation water use efficiency and total water use efficiency.
designed to test the effect of yearly weather conditions on The evapotranspiration efficiency is an important concept
the irrigated soil and crops, and thus decide which strategy for understanding how much of the total water (rain and
will be the best and most stable from year to year (Xevi irrigation) is being used by crop evapotranspiration and the
et al., 2008). balance that goes to deep drainage and runoff. In this paper,

Table IV. Definitions of water use indices in this paper

Water use indices Definition

Evapotranspiration efficiency (mm mm–1) ETc./(irrigation + rain)

Ratio of evapotranspiration and irrigation (mm mm–1) ETc./irrigation
Crop water use efficiency(kg mm–1 ha–1) kg ha–1 of yield per mm evapotranspired
Irrigation water use efficiency(kg mm–1 ha–1) kg ha–1 of crop yield per mm irrigation water
Total water use efficiency(kg mm–1 ha–1) kg ha–1 of crop yield per mm (irrigation + rain)

Source: Edraki et al. (2003).

Copyright © 2013 John Wiley & Sons, Ltd. Irrig. and Drain. (2013)
 Y. KANG ET AL.

the water use indices will be analysed for rainfed and Table V shows that potential and actual evapotranspira-
irrigated maize. Therefore when the maize is rainfed, total tion will increase in the four subareas of the Loess Plateau
water use efficiency means how much rainfall is used by under the A2 and B2 scenarios, although the rates of in-
the crop. crease are different in each zone. In Clay Loam-2, the in-
creased rates of potential evapotranspiration are around
30% under the A2 scenario and around 20% under the B2 sce-
RESULTS AND DISCUSSION nario from 2020 to 2080 respectively; the increase in the rate
of actual evapotranspiration is a little less than the potential
Crop yield and water use indices indicated for rainfed
evapotranspiration in the study time slices under both scenar-
maize
ios; the ratio of actual transpiration and evapotranspiration
Based on the SWAGMAN® Destiny model, rainfed maize will decrease in the future; the decrease rate is 7.04% under
yield and evapotranspiration in the subareas of the Loess the A2 scenario and 9.86% under the B2 scenario in 2080.
Plateau under future climate scenarios in 2020, 2050 and In Clay Loam-1, potential and actual evapotranspiration will
2080 are shown in Table V. Because of the deep groundwa- also increase while the rate of increase is less than that in Clay
ter conditions in the Loess Plateau, a free-draining bottom Loam-2. The increased rate of potential evapotranspiration is
boundary condition is considered. around 10% under the A2 scenario and 2% under the B2

Table V. Water balance components and rainfed maize yield prediction with the SWAGMAN® Destiny model under future climate scenarios
in four subareas of the Loess Plateau

Scenario A2 Scenario B2
Water balance components
Subarea and crop yield Baseline 2020 2050 2080 2020 2050 2080

Clay Loam-2 Precipitation mm 320 331 343 361 332 343 353
Δ(%) 3.30 6.92 12.81 3.58 6.92 10.10
ETpot mm 346.7 383.8 430.5 456.2 371.5 393.1 420.1
Δ(%) 10.70 24.17 31.58 7.15 13.38 21.17
ETact mm 318.6 351 376.1 402 339.1 352.5 368.6
Δ(%) 10.17 18.05 26.18 6.43 10.64 15.69
T/ETact mm mm–1 0.71 0.70 0.70 0.66 0.70 0.66 0.64
Δ(%) 0.00 0.00 7.04 0.00 7.04 9.86
Crop yield kg ha–1 5770 5529 5448 5308 5583 5488 5423
Δ(%) 4.18 5.58 8.01 3.24 4.89 6.01
Clay Loam-1 ETpot mm 828.4 842.4 881.4 920.6 830.8 836.8 844.2
Δ(%) 1.69 6.40 11.13 0.29 1.01 1.91
ETact mm 466.9 475.7 493.2 519.8 470.9 487.7 505.1
Δ(%) 1.88 5.63 11.33 0.86 4.45 8.18
T/ETact mm mm–1 0.62 0.63 0.63 0.63 0.62 0.62 0.63
Δ(%) 1.61 1.61 1.61 0.00 0.00 1.61
Crop yield kg ha–1 4795 4757 4724 4666 4763 4734 4717
Δ(%) 0.79 1.48 2.69 0.67 1.27 1.63
Light Loam ETpot mm 819.6 845.5 880.5 919.5 830.6 848.6 863.6
Δ(%) 3.16 7.43 12.19 1.34 3.54 5.37
ETact mm 482.1 484.1 492 505.6 485.8 489.1 499.3
Δ(%) 0.41 2.05 4.87 0.77 1.45 3.57
T/ETact mm mm–1 0.71 0.69 0.68 0.68 0.69 0.69 0.69
Δ(%) 2.82 4.23 4.23 2.82 2.82 2.82
Crop yield kg ha–1 2978 2708 2631 2579 2799 2730 2735
Δ(%) 9.07 11.65 13.40 6.01 8.33 8.16
Sandy Loam ETpot mm 887.9 978 1022.6 1067.2 959.8 982.6 1000.2
Δ(%) 10.15 15.17 20.19 8.12 10.69 12.67
ETact mm 465.2 475.1 486.3 514.7 470.1 488.8 509.4
Δ(%) 2.13 4.54 10.64 1.05 5.07 9.50
T/ETact mm mm–1 0.69 0.70 0.70 0.70 0.69 0.70 0.70
Δ(%) 1.45 1.45 1.45 0.00 1.45 1.45
Crop yield kg ha–1 2496 1672 1592 1564 1679 1678 1617
Δ(%) 33.01 36.22 37.34 32.73 32.77 35.22

Copyright © 2013 John Wiley & Sons, Ltd. Irrig. and Drain. (2013)
 PREDICTING CLIMATE CHANGE IMPACTS ON MAIZE CROP PRODUCTIVITY

scenario in 2020, 2050 and 2080 respectively; the increased B2 scenarios. The transpiration and evapotranspiration rates
rate of actual evapotranspiration is around 10% under the A2 will remain stable in future since the increase is just 2% under
and B2 scenarios. The ratio of actual transpiration and evapo- both scenarios in 2020, 2050 and 2080.
transpiration will increase slightly and the increase rate is only The rainfed maize yield in the Loess Plateau will decrease
around 1.61% under both scenarios in the future. under both scenarios in 2020, 2050 and 2080. The maize
In Light Loam, the potential and actual evapotranspiration yield in Clay Loam-2 will decrease within 10% under the
will increase in the future and the increase rate will also be A2 scenario and around 5% under the B2 scenario compared
higher. The increased rate of potential evapotranspiration is to the baseline in 2020, 2050 and 2080. The maize yield in
around 10% under the A2 scenario and 5% under the B2 sce- Clay Loam-1 will reduce within 3% under the A2 and B2
nario in 2020, 2050 and 2080 respectively. The increase per- scenarios; that in Light Loam will decrease within 15% un-
centage of actual evapotranspiration is within 5% under the der the A2 scenario and within 10% under the B2 scenario;
A2 and B2 scenarios in 2020, 2050 and 2080 respectively, the maize yield in Sandy Loam is around 35% under the A2
while the ratio of actual transpiration and evapotranspiration and B2 scenarios in 2020, 2050 and 2080 respectively.
will decrease in the future compared to the baseline and the Therefore, the yield decrease will be largest in Sandy Loam,
decrease rate will increase, but not by greater than 5%. followed by Light Loam and Clay Loam.
In Sandy Loam, potential and actual evapotranspiration, the Figure 4 shows the water use indices of rainfed maize
ratio of transpiration and evapotranspiration will increase and with the SWAGMAN® Destiny model under future climate
the increase rate will also be higher in 2020, 2050 and 2080 in four subareas of the Loess Plateau. It shows that evapo-
under both the A2 and B2 scenarios compared to the baseline. transpiration efficiency will increase in Clay Loam-2, Clay
The increased rate of potential evapotranspiration is around Loam-1 and Sandy Loam while decreasing in Light Loam
20% under the A2 scenario and within 15% under the B2 sce- under the A2 and B2 scenarios in 2020, 2050 and 2080
nario in 2020, 2050 and 2080 respectively. The increased rate respectively. Crop water use efficiency and total water use
of actual evapotranspiration is around 10% under the A2 and efficiency will decrease in the four subareas of the Loess

2 Clay Loam-2 Clay Loam-1 2 Clay Loam-2 Clay Loam-1

efficiency (mm/mm)
Evapotranspiration

Light Loam Sandy Loam Light Loam Sandy Loam

1.5 1.5

1 1

0.5 0.5

0 0
Baseline 2020 2050 2080 Baseline 2020 2050 2080
Time slices Time slices
Crop water use efficiency

Clay Loam-2 Clay Loam-1 20 Clay Loam-2 Clay Loam-1

20
Light Loam Sandy Loam Light Loam Sandy Loam

15 15
(kg/mm/ha)

10 10

5 5

0 0
Baseline 2020 2050 2080 Baseline 2020 2050 2080
Time slices Time slices
Total water use efficiency

Clay Loam-2 Clay Loam-1 20 Clay Loam-2 Clay Loam-1

20
Light Loam Sandy Loam Light Loam Sandy Loam
15 15
(kg/mm/ha)

10 10

5 5

0 0
Baseline 2020 2050 2080 Baseline 2020 2050 2080

Figure 4. Water use indices of rainfed maize with the SWAGMAN® Destiny model under future climate scenarios in four subareas of the Loess Plateau. This
figure is available in colour online at wileyonlinelibrary.com/journal/ird

Copyright © 2013 John Wiley & Sons, Ltd. Irrig. and Drain. (2013)
 Y. KANG ET AL.

Plateau. In Clay Loam-2, the increase in the rate of evapo- efficiency and total water use efficiency in Clay Loam-2 is
transpiration efficiency is around 10% under the A2 scenario within 30% under the A2 scenario and 20% under the B2
and 5% under the B2 scenario in 2020, 2050 and 2080 re- scenario in 2020, 2050 and 2080 respectively. The decrease
spectively. In Clay Loam-1 and Sandy Loam, the increased in the rate in Clay Loam-1 is within 15% under the A2 sce-
rate of evapotranspiration efficiency is around 2% under nario and 10% under the B2 scenario; the decrease range in
both scenarios in 2020, 2050 and 2080. The decrease in Light Loam is around 20% under the A2 scenario and 15%
the rate of evapotranspiration efficiency in Light Loam is under the B2 scenario; that in Sandy Loam is around 40%
within 5% under the A2 and B2 scenarios in 2020, 2050 under the A2 and B2 scenarios in 2020, 2050 and 2080
and 2080 respectively. respectively.
The variable ranges of crop water use efficiency and total Evapotranspiration efficiency in Clay Loam-2 is the low-
water use efficiency are similar in the four subareas of the est and in Sandy Loam the highest; however, crop water use
Loess Plateau. The decrease in the rate of crop water use efficiency and total water use efficiency in Clay Loam-2 is

Table VI. Evapotranspiration and yield predictions during irrigated maize growing season under future climate scenarios in Loess Plateau

Scenario A2 Scenario B2

Water balance components and crop yield Baseline 2020 2050 2080 2020 2050 2080

Clay Loam-2 Irrigation mm 206 218 225 228 212 220 222
Δ(%) 5.83 9.22 10.68 2.91 6.80 7.77
ETpot mm 346.7 383.8 430.5 456.2 371.5 393.1 420.1
Δ(%) 10.70 24.17 31.58 7.15 13.38 21.17
ETact mm 328.6 365.2 398.6 427 355.7 374.2 389.3
Δ(%) 11.14 21.30 29.95 8.25 13.88 18.47
T/ETact mm mm–1 0.69 0.69 0.67 0. 65 0.66 0.65 0.64
Δ(%) 0.00 2.90 5.80 4.35 5.80 7.25
Crop yield kg ha–1 10016 10393 10412 10529 10132 10346 10439
Δ(%) 3.76 3.95 5.12 1.16 3.29 4.22
Clay Loam-1 Irrigation mm 220 226 234 238 225 231 235
Δ(%) 2.73 6.36 8.18 2.27 5.00 6.82
ETact mm 479.2 496.8 518.3 541.5 492 505.3 521.1
Δ(%) 3.67 8.16 13.00 2.67 5.45 8.74
ETpot mm 828.4 842.4 881.4 920.6 830.8 836.8 844.2
Δ(%) 1.69 6.40 11.13 0.29 1.01 1.91
T/ETact mm mm–1 0.57 0.57 0.58 0.59 0.57 0.57 0.57
Δ(%) 0.00 1.75 3.51 0.00 0.00 0.00
Crop yield kg ha–1 4817 4845 4924 5004 4833 4868 4897
Δ(%) 0.58 2.22 3.88 0.33 1.06 1.66
Light Loam Irrigation mm 240.9 241.6 242.2 241.5 240.4 240.8 239.2
Δ(%) 0.29 0.54 0.25 0.21 0.04 0.71
ETpot mm 819.6 845.5 880.5 919.5 830.6 848.6 863.6
Δ(%) 3.16 7.43 12.19 1.34 3.54 5.37
ETact mm 617 622.3 630.7 644.9 622.3 627.5 635.1
Δ(%) 0.86 2.22 4.52 0.86 1.70 2.93
T/ETact mm mm–1 0.73 0.73 0.73 0.73 0.73 0.73 0.73
Δ(%) 0.00 0.00 0.00 0.00 0.00 0.00
Crop yield kg ha–1 4202 4148 3950 3892 4166 4154 4136
Δ(%) 1.29 6.00 7.38 0.86 1.14 1.57
Sandy Loam Irrigation mm 456 434 434 433.4 434 434 433.4
Δ(%) 4.82 4.82 4.96 4.82 4.82 4.96
ETpot mm 887.9 978 1022.6 1067.2 959.8 982.6 1000.2
Δ(%) 10.17 15.20 20.22 8.12 10.69 12.67
ETact mm 686.6 692.4 707.1 723.2 693 701.1 710.5
Δ(%) 0.84 2.99 5.33 0.93 2.11 3.48
T/ETact mm mm–1 0.69 0.68 0.68 0.68 0.67 0.67 0.67
Δ(%) 1.45 1.45 1.45 2.90 2.90 2.90
Crop yield kg ha–1 4269 3815 3672 3619 3920 3809 3797
Δ(%) 10.63 13.98 15.23 8.18 10.78 11.06

Copyright © 2013 John Wiley & Sons, Ltd. Irrig. and Drain. (2013)
 PREDICTING CLIMATE CHANGE IMPACTS ON MAIZE CROP PRODUCTIVITY

highest and in Sandy Loam lowest. Therefore, evapotranspi- Crop yield and water use indices for irrigated maize
ration efficiency of rainfed maize will increase in the Loess
Plateau from south to north, while crop water use efficiency Fengjiashan, Baojixia and Heihe irrigation areas are major
and total water use efficiency will decrease from north to south irrigation systems in the Loess Plateau. The irrigation appli-
under the A2 scenario and B2 scenarios in 2020, 2050 and 2080. cation rate is kept the same as the present irrigation rates in

1.2 Clay Loam-2 Clay Loam-1 1.2 Clay Loam-2 Clay Loam-1

efficiency (mm/mm)
Evapotranspiration
Light Loam Sandy Loam 1 Light Loam Sandy Loam
1
0.8 0.8
0.6 0.6
0.4 0.4
0.2 0.2
0 0
Baseline 2020 2050 2080 Baseline 2020 2050 2080
Time slices Time slices

Ratio of ET and irrigation

Clay Loam-2 Clay Loam-1 Clay Loam-2 Clay Loam-1
3.5 3 Light Loam Sandy Loam
3 Light Loam Sandy Loam
2.5
2.5

(mm/mm)
2
2
1.5
1.5
1 1
0.5 0.5
0 0
Baseline 2020 2050 2080 Baseline 2020 2050 2080
Time slices Time slices
Crop water use efficiency

35 Clay Loam-2 Clay Loam-1 35 Clay Loam-2 Clay Loam-1

Light Loam Sandy Loam
30 Light Loam Sandy Loam 30
(kg/mm/ha)

25 25
20 20
15 15
10 10
5 5
0 0
Baseline 2020 2050 2080 Baseline 2020 2050 2080
Time slices Time Slices

60 Clay Loam-2 Clay Loam-1 Clay Loam-2 Clay Loam-1

efficiency (kg/mm/ha)

60
Irrigation water use

Light Loam Sandy Loam Light Loam Sandy Loam

50
50
40 40
30 30
20 20
10 10
0 0
Baseline 2020 2050 2080 Baseline 2020 2050 2080
Time slices Time slices

25 Clay Loam-2 Clay Loam-1 25 Clay Loam-2 Clay Loam-1

Light Loam Sandy Loam Light Loam Sandy Loam
20 20
15 15
10 10
5 5
0 0
Baseline 2020 2050 2080 Baseline 2020 2050 2080

Figure 5. Water use indices of irrigated maize with the SWAGMAN® Destiny model under future climate scenarios in four subareas of the Loess Plateau. This
figure is available in colour online at wileyonlinelibrary.com/journal/ird

Copyright © 2013 John Wiley & Sons, Ltd. Irrig. and Drain. (2013)
 Y. KANG ET AL.

each subarea of the Loess Plateau. Table VI shows potential irrigation is within 5% under the A2 scenario and 2% under
and actual evapotranspiration and the maize yield with irri- the B2 scenario. The decreased rate of crop water use effi-
gation under future climate scenarios. The precipitation is ciency is within 10% under the A2 and B2 scenarios in
the same as that in Table V, which is omitted in Table VI. 2020, 2050 and 2080 respectively; that of irrigation water
Potential and actual evapotranspiration will increase in use efficiency is within 5% under both scenarios; that of total
the four subareas of the Loess Plateau and the increase rate water use efficiency is around 5% under the A2 and B2
will be higher under future climate scenarios for irrigated scenarios in 2020, 2050 and 2080 respectively.
maize as compared with the rainfed. In Light Loam, the decrease in the rate of evapotranspiration
The ratio of transpiration and actual evapotranspiration will efficiency is around 1% under the A2 and B2 scenarios in
decrease in Clay Loam-2 and Sandy Loam, increase in Clay 2020, 2050 and 2080 respectively. The increased rate of ratio
Loam-1 and remain stable in Light Loam. The decrease rate of evapotranspiration and irrigation is within 5% under both
in Clay Loam-2 is around 5% under the A2 and B2 scenarios scenarios. The decreased rate of crop water use efficiency is
in 2020, 2050 and 2080 respectively but the decrease rate in around 10% under the A2 scenario and within 5% under the
Sandy Loam is within 3% under both scenarios. The increase B2 scenario in 2020, 2050 and 2080 respectively; that of irriga-
rate in Clay Loam-1 is within 3% under both scenarios. There- tion water use efficiency is within 10% under the A2 scenario
fore, climate change does not have a strong impact on the ratio and 5% under the B2 scenario; that of total water use efficiency
of transpiration and evapotranspiration. is within 15% under the A2 scenario and around 5% under the
Table VI also indicates that irrigated maize yield will in- B2 scenario in 2020, 2050 and 2080 respectively.
crease in Clay Loam-2 and Clay Loam-1 but decrease in Light In Sandy Loam, the increase in the rate of evapotranspira-
Loam and Sandy Loam under the A2 and B2 scenarios in tion efficiency is within 10% under the A2 and B2 scenarios
2020, 2050 and 2080 respectively. The increase rates of yield in 2020, 2050 and 2080 respectively. The increasd rate of
in Clay Loam-2 and Clay Loam-1 are around 5% under A2 ratio of evapotranspiration and irrigation is around 10% un-
and B2 scenarios in 2020, 2050 and 2080 respectively. How- der both scenarios. The decreased rate of crop water use ef-
ever, the decrease in maize yield in Light Loam is around 5% ficiency is within 20% under the A2 scenario and 15% under
under the A2 and B2 scenarios in 2020, 2050 and 2080 respec- the B2 scenario in 2020, 2050 and 2080 respectively; that of
tively; that in Sandy Loam is around 15% under both scenarios. irrigation water use efficiency is around 10% under the A2
Figure 5 presents water use indices of irrigated maize and B2 scenarios; that of total water use efficiency is around
under future climate scenarios by SWAGMAN® Destiny 15% under the A2 scenario and 10% under the B2 scenario
in the four subareas of the Loess Plateau. in 2020, 2050 and 2080 respectively.
Evapotranspiration efficiency and the ratio of evapotrans-
piration and irrigation will increase in Clay Loam-2, Clay
Loam-1, Light Loam and Sandy Loam under the A2 and
CONCLUSIONS
B2 scenarios. However, the other indices, including crop
water use efficiency, irrigation water use efficiency and total This paper mainly discusses evapotranspiration, maize yield
water use efficiency, will decrease under the A2 and B2 and water use indices with and without irrigation under the
scenarios during irrigated maize growth. A2 and B2 scenarios in 2020, 2050 and 2080 in four subareas
In Clay Loam-2, the increased rate of evapotranspiration of the Loess Plateau. The above analysis shows that the indi-
efficiency is around 15% under the A2 scenario and within ces referred to in this paper are higher under the A2 scenario
10% under the B2 scenario in 2020, 2050 and 2080 respec- than those under the B2 scenario, although the indices have
tively. The increased rate of ratio between evapotranspira- the same variable tendency under both scenarios.
tion and irrigation is within 20% under the A2 scenario Although precipitation will increase under future scenarios,
and 10% under the B2 scenario. The decreased rate of crop rainfed maize yield will decrease in the four subareas of the
water use efficiency is within 20% under the A2 scenario Loess Plateau while there are some differences in the irrigated
and 15% under the B2 scenario in 2020, 2050 and 2080 re- maize yield under future climate scenarios. Namely, maize
spectively; that of irrigation water use efficiency is around yield will increase in Clay Loam but is likely to decrease in
5% under the A2 scenario and 3% under the B2 scenario; Light Loam and Sandy Loam. Under future climate scenarios,
that of total water use efficiency is around 5% under the evapotranspiration will increase under both rainfed and irri-
A2 and B2 scenarios in 2020, 2050 and 2080 respectively. gated conditions. Crop water use efficiency and total water
In Clay Loam-1, both the increased rate of evapotranspira- use efficiency in the future will decrease in 2020, 2050 and
tion efficiency and the increased rate of ratio of evapotranspi- 2080, although the rate of decrease varies across the Loess
ration and irrigation are within 5% under the A2 scenario and Plateau under both climate scenarios.
2% under the B2 scenario in 2020, 2050 and 2080 respec- In the case of rainfed maize, as the evapotranspiration
tively. The increased rate of ratio of evapotranspiration and goes up, the corresponding increasing rainfall is not enough

Copyright © 2013 John Wiley & Sons, Ltd. Irrig. and Drain. (2013)
 PREDICTING CLIMATE CHANGE IMPACTS ON MAIZE CROP PRODUCTIVITY

to meet the water demand of the crop. The soil water deficit impact of climate change on winter wheat production in northeastern
increases when the climate changes from 2020 to 2080. This Austria.Australian Journal of Agricultural Research 52: 199–212.
Eitzinger J, Stastna M, Zalud Z, Dubrovsky M. 2003. A simulation study of
results in higher water crop stress, leading to in increasing the effect of soil water balance and water stress on winter wheat produc-
yield losses. This result is consistent with IPCC findings in tion under different climate change scenarios. Agricultural Water
other regions of the world. Management 61: 195–217.
In the case of irrigated maize for heavier soil, as evapotrans- Godwin D, Meyer WS, Xevi E, White RGJ. 2006. SWAGMAN® Destiny:
piration increases, the increasing crop water demand can be a simulation model of irrigated crops as affected by irrigation practice,
water tables and salinity. 1. Model description. CSIRO Land and Water.
met by irrigation therefore there was no crop water stress, Unpublished.
and increasing crop water evapotranspiration results in higher Intergovernmental Panel on Climate Change (IPCC). 2007. Climate change
use. However, in the case of lighter soils, the above findings 2007: impacts, adaptation and vulnerability. In Contribution of Working
are not true because of low water-holding capacity and higher Group II to the Fourth Assessment Report of the Intergovernmental Panel
soil hydraulic conductivity. Because of low soil water capac- on Climate Change, Parry ML, Canziani OF, Palutikof JP, van der Linden
PJ, Hanson CE. Cambridge University Press: Cambridge, UK; 1–976.
ity and higher hydraulic conductivity it is difficult to maintain
Kang Y, Ma X, Li J, Wang B. 2008. Soil water movement model and
soil moisture close to field capacity. Therefore crop yields subarea irrigation parameter for gravity subsurface drip irrigation in the
may go down unless there is investment in improved water Loess Plateau. Transactions of the Chinese Society for Agricultural
irrigation technologies such as drip and trickle irrigation Machinery 39(3): 90–95.
systems to match soil hydraulic properties. Kang Y, Khan S, Ma X. 2009. Climate change impacts on crop yield, crop
water productivity and food security— a review. Progress in Natural
For rainfed cropping systems there is need to do future
Science 19: 1665–1674.
research on water-harvesting technologies for soil water Li Y, Han S, Wang Z. 1985. Soil water properties and its zonation in the
conservation to protect and enhance crop yields to meet Loess Plateau. Research of Soil and Water Conservation 2: 1–16.
increasing crop water demand due to climate change. For Li J, Shao M, Wang L. 2002. Developing strategy of grain crop production
irrigated cropping systems, future work is needed to link in the Loess Plateau region of China. Chinese Journal of Eco-Agriculture
irrigation scheduling with irrigation methods to provide 10(1): 118–120.
Liu Y, Li S, Chen F, Yang S, Chen X. 2010. Soil water dynamics and water
on-need soil water during low rainfall periods. This is espe- use efficiency in spring maize (Zea mays L.) fields subjected to different
cially required for lighter soils. CO2 impacts on crop growth water management practices on the Loess Plateau, China. Agricultural
were not considered during the study. There is a need to use Water Management 97(5): 769–775.
crop growth models with the future scenario data provided Reddy KR, Hodges HF. 2000. Climate change and global crop productivity:
in this work. an overview. In Reddy KR, Hodges HF (eds). Climate Change and Global
Crop Productivity, CABI Publishing: New York; 1–5.
Sirotenko OD, Abashina HV, Pavlova VN. 1997. Sensitivity of the Russian
agriculture to changes in climate, CO2 and tropospheric ozone concentra-
ACKNOWLEDGEMENTS tions and soil fertility. Climatic Change 36: 217–232.
Timsina J, Humphreys E. 2003. Performance and Application of CERES and
This work was supported by the National Natural Science SWAGMAN Destiny Models for Rice–Wheat Cropping Systems in Asia
Foundation of China (Grant No. 51279167), the National and Australia: a Review. CSIRO Land and Water, Griffith. Technical
’Twelfth Five-Year’ Plan for Science and Technology Report 16/03
Support (Grant No. 2011BAD25B03 and 2012BAD08B01) Tojo SCM, Sentelhas PC, Hoogenboom G 2007. Application of the CSM–
CERES–Maize model for planting date evaluation and yield forecasting
and the Scientific Research Fund of Sichuan Provincial
for maize grown off-season in a subtropical environment. European
Education Department (Grant No. 11ZA072). Journal of Agronomy 27: 165–177.
Wang Z, Liang Y. 2002. The application of EPIC model to calculate crop
productive potentialities in loessic yuan region. Journal of Natural
REFERENCES Resources 17(4): 481–487.
Xevi E, Godwin D, Meyer WS, White RGJ, Humphreys L. 2008.
Akpalu W, Hassan RM, Ringler C. 2008. Climate variability and maize SWAGMAN® Destiny: Crop Growth, Soil Water and Salt Balance Model.
yield in South Africa: results from GME and MELE methods. Environment Processes and Description. CSIRO Land and Water, Griffith Laboratory.
and Production Technology Division IFPRI discussion paper; 1–12. Xiong W, Lin E, Ju H, Xu Y. 2007a. Climate change and critical thresholds
Anwar MR, O’Leary G, McNeil D, Hossain H, Nelson R. 2007. Climate in China’s food security. Climatic Change 81: 205–221.
change impact on rainfed wheat in south-eastern Australia. Field Crops Xiong W, Matthews R, Holman I, Lin E, Xu Y. 2007b. Modelling China’s
Research 104: 139–47. potential maize production at regional scale under climate change.
De Silva CS, Weatherhead EK, Knox JW, Rodriguez JA. 2007.Predicting the Climatic Change 85: 433–451.
impacts of climate change: a case study of paddy irrigation water require- Yao F, Xu Y, Lin E, Yokozawa M, Zhang J. 2007. Assessing the impacts of
ments in Sri Lanka. Agricultural Water Management 93(1–2): 19–29. climate change on rice yields in the main rice areas of China. Climatic
Edraki M, Smith D, Humphreys E, Khan S, O’Connell N, Xevi E. 2003. Change 80: 395–409.
Validation of the SWAGMAN® Farm and SWAGMAN® Destiny Zhang XC, Liu WZ. 2005. Simulating potential response of hydrology, soil
Models. CSIRO Land and Water, Griffith. Technical Report erosion, and crop productivity to climate change in Changwu tableland
Eitzinger J, Zalud Z, Alexandrov V, van Diepen CA, Trnka M, Dubrovsky region on the Loess Plateau of China. Agricultural and Forest Meteorology
M, Semeradova D, Oberforster M. 2001. A local simulation study on the 131: 127–142.

Copyright © 2013 John Wiley & Sons, Ltd. Irrig. and Drain. (2013)