Main

Agricultural Water Management 288 (2023) 108485
Contents lists available at ScienceDirect
Agricultural Water Management

journal homepage: www.elsevier.com/locate/agwat
Resilience assessment of blue and green water resources for staple crop
production in China
Hongrong Huang a, c, La Zhuo b, c, d, *, Wei Wang c, d, Pute Wu b, c, d, *
a
College of Water Resources and Architectural Engineering, Northwest A&F University, Yangling 712100, China
b
Institute of Soil and Water Conservation, Northwest A&F University, Yangling 712100, China
c
Institute of Water-saving Agriculture in Arid Regions of China, Northwest A&F University, Yangling 712100, China
d
Institute of Soil and Water Conservation, Chinese Academy of Sciences and Ministry of Water Resources, Yangling 712100, China
A R T I C L E I N F O A B S T R A C T
Handling Editor: Dr. B.E. Clothier Blue water (irrigation water) and green water (soil moisture), as natural bases in crop production, are vulnerable
to natural and socioeconomic disturbances. However, quantitative resilience assessments of blue and green water
Keywords: for crop production are still lacking. Here we constructed a social-ecological based general water resilience
Water footprint assessment framework that distinguishes blue and green water for crop production in four dimensions, including
Social-ecological system
resistance, absorptive capacity, adaptive capacity and transformative capacity. The framework was applied to
Resistance
quantify the spatiotemporal characteristics of water resource resilience for three staple crops (rice, wheat, and
Absorptive capacity
Adaptive capacity maize) at provincial level in mainland China over 2000–2017. Results indicate that water resilience related to
Transformative capacity crop production of China as a whole was at the lower-middle level over the study period. The green water
resilience had higher initial performance, while the blue water resilience had a faster growth rate (>50%). The
performance of adaptive and transformative capacity was 8.9~153.7% better than resistance and absorptive
capacity. There were significant heterogeneities in the water resilience score among crops and in time and space.
Wheat had the highest resilience of green water, while its blue water resilience was the lowest among the three
crops. The water resilience for crop production varied by up to 2.4-fold across provinces. Synergies and trade-offs
among the four dimensions of water resilience were further evaluated. The current study provides an effective
approach for assessing the state and progress of resilience of regional water resources for crop production.
1. Introduction shock and uncertainty on agricultural water resource systems, that

continuously pushing the global or regional water resources system out
In the Anthropocene, anthropogenic activities have become the of the “safe operating space” (Gleeson et al., 2020; Kummu et al.,2020).
overwhelming driver affecting the hydrological system (Steffen et al., Therefore, it is urgent and essential to comprehensively understand and
2015). Agriculture accounts for the largest use of water resources and is evaluate the unpredictable risks from all aspects and propose corre
a part of typical social-ecological systems that are closely associated sponding policies and actions to improve the water resilience, i.e., the
with nature and human activities; however, water resources are facing ability of regional water resources system to respond to these pressures
various severe challenges (e.g., shortage of surface water resources, and shocks (Falkenmark and Wang-Erlandsson, 2021).
decline of groundwater level, water pollution, and increased risk of The ecologist Holling first introduced the concept of resilience into
flood and drought caused by extreme weather events) that threaten the field of ecology research to describe “the ability of an ecosystem to
regional water supplies and food security (Falkenmark and recover to a stable state in the face of disturbance” (Hoiling, 1973). Over
Wang-Erlandsson, 2021; Piao et al., 2010; Rodina et al., 2018; time, the theory of resilience has been widely used in a broad range of
Vörösmarty et al., 2000). As such, population growth and urbanization disciplines, such as climate change (Kahiluoto et al., 2019), food systems
led to the increasing water demand. Climate change, aging infrastruc (Ansah et al., 2019; Rockström et al., 2009; Seekell et al., 2017; Suweis
ture, public health-related pandemics, and local conflicts around the et al., 2015), and water resource systems (Dewulf et al., 2019; Falken
world have placed unprecedented long-term pressure or short-term mark et al., 2019; Krueger et al., 2019; Varis and Kummu, 2019). With
* Corresponding authors at: Institute of Soil and Water Conservation, Northwest A&F University, Yangling 712100, China.
E-mail addresses: zhuola@nwafu.edu.cn (L. Zhuo), gjzwpt@vip.sina.com (P. Wu).
https://doi.org/10.1016/j.agwat.2023.108485
Received 10 July 2023; Received in revised form 14 August 2023; Accepted 15 August 2023
Available online 17 August 2023
0378-3774/© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
H. Huang et al. Agricultural Water Management 288 (2023) 108485
the continuous extension of application fields, the dimensions of resil intrinsic attributes and capabilities of the system itself to cope with and
ience have been gradually extended and improved (Table 1). The early respond to all uncertain shocks and pressure sources (e.g., rivers to cope
dimension of resilience principally emphasized the system’s resistance with all shocks including water resource shortage, floods, droughts, and
to shock and the ability to absorb impact and maintained the structure water quality degradation). Unlike special resilience, general resilience
and function (Holling, 2005). Subsequently, the resilience of can be measured no matter whether the shocks occur or not. Currently,
social-ecological systems has attracted widespread attention. (Béné and there are two ways for measuring the two types of resilience: core var
Doyen, 2018; Biggs et al., 2012; Cinner and Barnes, 2019; Cumming and iable simulation (for special resilience) and comprehensive index eval
Peterson, 2017; Folke et al., 2016; Grêt-Regamey et al., 2019; Sinclair uation (for general resilience). The approach of core variable simulation
et al., 2017). Researchers have reached a consensus that the resilience of is applied in the engineering field (i.e., engineering resilience), and is
social-ecological systems is a complex and multidimensional concept. In often represented by the change in core variables sensitive to distur
addition to resistance and absorption, the characteristics of adaptability, bance (e.g., degradation of system performance, recovery time or
transformation, feedback, diversity, and persistence were mentioned probability of shocks) (Asefa et al., 2014; Cottrell et al., 2019; Grafton
(Table 1) (Béné and Doyen, 2018; Biggs et al., 2012; Dewulf et al., 2019; et al., 2019; Kahiluoto et al., 2019). However, this approach only
Folke et al., 2010; Sharifi and Yamagata, 2016). Social-ecological system measures specific impact and resilience targets and therefore fails to
resilience emphasizes the ability to dynamically adapt and transform in comprehensively consider other types of shocks and uncertainties.
the face of disturbance (Cumming et al., 2017; Folke et al., 2010). The Moreover, the system performance will not decline immediately due to
classic theories of social-ecological system resilience include dynamic the existence of resistance when facing shocks, resulting in certain errors
evolution and adaptive cycles, highlighting the connotation and attri (Grafton et al., 2019; Zampieri et al., 2020). In contrast, other scholars
butes of social-ecological system continuous development through pre adopt the methodology of comprehensive assessment of the general
venting, resisting, absorbing, adapting, and even restructuring when resilience, especially the social-ecological resilience, due to the
facing disturbances and shocks (Béné et al., 2018; Biggs et al., 2012; complexity and multi-dimensional attributes of the social-ecological
Folke et al., 2016). Therefore, we summarized the generally accepted systems (Cutter and Derakhshan, 2018; Herrera de Leon and Kopain
four dimensions of social-ecological system resilience based on the sky, 2020; Ifejika Speranza et al., 2014; Liu et al., 2019; Seekell et al.,
connotation of social-ecological system resilience from literature re 2017). Ifejika Speranza et al. (2014) established an evaluation frame
view: resistance, absorption, adaptation and transformation. work of livelihood resilience in the context of social-ecological systems
Agricultural water consumption process is one of the socio- based on three dimensions: buffer capacity, self-organization capacity
hydrological system that transcends ecology, society, and economy and learning capacity. Herrera de Leon and Kopainsky (2020) assessed
(Dewulf et al., 2019; Di Baldassarre et al., 2019; Sivapalan et al., 2012), the resilience of the food system from three dimensions: stability,
which is highly susceptible to disturbances and shocks from anthropo adaptability and transformation.
genic activities and nature. Therefore, we define regional agricultural To date, as a subset of resilience, the research on water resilience also
water resource resilience as the ability of the communities using the includes two aspects: engineering-based water resilience and social-
water resources to cope with shocks, adapt to changes and transform in ecological based water resilience. The former accounts for the major
crisis following the connotation of social-ecological resilience. The ity (Dewulf et al., 2019; Diao et al., 2016; Li and Lence, 2007; Roach
quantification of resilience varies according to the resilience classifica et al., 2018; Rodina, 2018). While there is relatively little information on
tion. The Resilience Alliance (2010) divides resilience into special the resilience of agricultural water resources, which account for a larger
resilience and general resilience. The special resilience is the targeted share of human water appropriation, especially considering its
response of the system to specific types of disturbances (e.g., water social-ecological attribute (Lu et al., 2022). The water footprint (Hoek
shortage on rivers), and the measurement can only be conducted after stra, 2003) is an effective metric to distinguish the sources and types of
the change of system performance (e.g., the decrease of river runoff) water resource consumption—blue water (irrigation water), green water
occurs. While the general resilience does not consider any specific types (soil moisture) and gray water (water pollution degree)—especially in
of disturbances, and hold the idea that the general resilience is the the field of agricultural water resources. Water for crop production ac
counts for 92% of global water footprint (Hoekstra and Mekonnen,
Table 1 2012). Green water is the major water resource for crop production,
Summary of representative literature on social-ecological system resilience and accounting for ~87% of the total water consumption for crop produc
associated dimensions. tion, and 60% of the world’s food is produced under rain-fed conditions
References Application field Dimensions of resilience
(Hoekstra and Mekonnen, 2012; Rockström et al., 2009). Currently,
agricultural water efficient utilization and sustainable evaluation based
Folke et al. Social-ecological system Persistence, adaptability &
on water footprint has made considerable progress (Cao et al., 2020.,
(2010) transformability
Biggs et al. Social-ecological system Diversity, redundancy & 2023), while the water resilience urgently need to be integrated into the
(2012) connectivity, feedbacks, up-to-date concepts of water resource management. However, only a
adaptability & learning few studies have evaluated the social-ecological based resilience of blue
Sharifi and Urban energy system Resistance, absorptivity, water resources, such as basins or lakes (Behboudian et al., 2021; Varis
Yamagata recoverability & adaptability
(2016)
and Kummu, 2019; Varis et al., 2019). Varis and Kummu (2019) and
Béné and Doyen General social-ecological Resistance, absorptivity, Varis et al. (2019) constructed six indicators from the perspective of
(2018) system adaptability & transformability environmental vulnerability and human adaptability to evaluate the
Ansah et al. Food system Absorptivity, adaptability, resilience of global and Chinese basin-based river systems. They found
(2019) transformability
that river resilience shows significant spatial heterogeneity at the con
Meuwissen et al. Farming systems Resistance, absorptivity &
(2019) adaptability tinental scale. Behboudian et al. (2021) developed an evaluation
Dewulf et al. Socio-hydrological Absorptivity, adaptability & framework for the resilience of lake water resource systems with a focus
(2019) systems transformability on resistance, redundancy, resource endowments and recoverability.
Varis et al. River basins Ecological vulnerability & Rockström et al. (2009) demonstrated that the green water has sub
(2019) adaptability
Behboudian et al. Lake system Robustness, redundancy,
stantial potential for resilience in crop production impacted by an
(2021) Water consumption resourcefulness, rapidity & extreme climate. Falkenmark and Rockström (2010) introduced a water
This study system related to crop durability resource management framework that transitioned from considering
production Resistance, absorptivity, only blue water resilience to considering blue–green water resilience in
adaptability & transformability
crop production. Although these two studies call for the necessity and
2
urgency of increasing the resilience of blue water and green water Therefore, the purpose of this study is to (1) develop a consistent and
related to crop production, neither provides a specific manner for transparent social- ecological (i.e., general resilience) water resilience
measuring resiliency. Given the grand contribution of water consump assessment framework (WRAF) to differentiate blue water (the executor
tion in crop production to human water use, constructing a common of resilience is communities or farmers managing irrigated agriculture)
resilience assessment framework for water consumption in crop pro and green water resources (the executor of resilience is communities or
duction that distinguishes blue and green water would provide an farmers managing rain-fed agriculture) based on water footprint ac
essential foundation for assessing the risks of agricultural water re counting of crop production and (2) quantify the provincial spatiotem
sources to facilitate the differentiated policies. poral progress of the resilience of different types of water resources
Table 2
The resilience dimensions and specific indicators involved in this study.
Dimensions Major aspects Indicators Reasons for indicator selection Unit Distinguish Distinguish Reference
crops blue and green
water
Resistance Resource Per capita blue water resources The total amount of blue or green water ha Y Y Liu et al.
endowment (Per capita arable area for resources reflects the ability of water (2019)
green water) (+) resources system to resist unknown
shocks.
Blue water stress index during Water resources shortage Y Y Varis et al.
crop growth period comprehensively represents the (2019)
(precipitation anomaly index mismatch between water supply and
during crop growth period for demand.
green water) (-)
Economy Per capita disposable income of Wealthier people can resist shocks by USD/ N N FAO (2016)
rural residents (+) purchasing insurance, investing in y
equipment, and participating in
specialized risk-resistant learning and
skill training.
Infrastructure Major agricultural The irrigation and agricultural power N N Author
infrastructures (+) infrastructures ensure the stable supply constructed
of water resources related to crop
production.
Absorptive Redundancy Blue (green)water redundancy Reserve water resources to cope with Mm3/ Y Y Fader et al.
capacity (+) long-term demand pressure and y (2016)
disturbance.
Virtual blue (green) water Reserve virtual water resources to cope Mm3/ Y Y Ruess, Konar
storage (+) with short-term and sudden shocks. y (2019)
Buffering Reservoir storage (+) Buffer and regulate too much or too little Gm3/ N N Author
water resources caused by drought or y constructed
flood from the two processes of “storage
and release”.
Diversity Diversity of income sources (+) Higher diversity is conducive to the % N N Resilience
substitution of similar structures and Alliance
functions to mitigate impacts. (2010)
Adaptive Farmers’ Net crop income (+) The profitable production process can USD/ Y N FAO (2016)
capacity wellbeing indirectly promote the water resources kg
system to take adaptive measures and
establish the basis for long-term
sustainable operation.
Government Disaster adaptability (+) The indicators of the probability and % N N Varis et al.
intensity of disasters can reflect the (2019)
experience and technology of water
resources system to improve the
adjustment and adaptability to disasters.
Learning Education of agricultural Workers with higher education years can y N N FAO (2016)
laborers (+) better respond shocks through
knowledge acquisition and adaptive
technology application.
Self- Blue (green) water self- The higher water self-sufficiency rate can % Y Y Author
organization sufficiency rate (+) better coordinate the internal structure constructed
and function to adapt shocks.
Transformative Production Crop production share (-) The higher proportion of crop production % Y N Author
capacity structure signifies the heavier responsibility of constructed
production, and the less easy for the
transformation.
Research R & D expenditure (+) The higher R & D expenditure means the USD/ N N Meuwissen
greater scientific and technological y et al., (2019)
innovation possibilities for water
transformation.
Urbanization Composite urbanization rate The socio-economic-cultural basis of N N Wu et al.
(+) transformative potential. (2020)
3
Technology & The variability of the water Technology, management improvement m /t Y Y Author
management footprint per unit mass of crop and innovation potential of water constructed
production (-) resources transformation related to crop
production.
Note: the + and - in parentheses indicate the positive and negative roles of specific indicators, respectively
3
consumed by three staple crops (rice, wheat and maize) in China from representative literature (Table 1). According to the consensus of
2000 to 2017. Based on the concept of social-ecological resilience, we multidimensional attributes of resilience, we discussed and defined the
address four universal dimensions of resilience: resistance, absorptive four major dimensions of social-ecological system resilience: resistance,
capacity, adaptive capacity and transformative capacity. Sixteen repre absorptive capacity, adaptive capacity and transformative capacity.
sentative indicators are selected to multidimensional assess the water Additionally, the resilience assessment must be within the specific
resilience related to crop production (Table 2). The natural, economic, boundary (Schlüter et al., 2019), so the WRAF scope of this study is the
social and governance aspects related to the consumption of water re blue and green water resources consumed by the staple crop production
sources in crop production were considered. Compared with the existing at the provincial level in mainland China.
literature, the current study has improvements in three aspects: Step 2: Identifying the cardinal impact aspects or core properties of
each dimension of resilience. Considering the driving factors or internal
(I) this study focuses on freshwater consumption in crop production attributes of each dimension, we discussed and determined the main
(the largest water user), and proposes a social-ecological system aspects affecting the water resources system of crop production based on
resilience evaluation framework for the communities using the a literature review and expert opinions from the perspective of nature,
water resources for crop production; economy, society, infrastructure, technology and governance.
(II) this study, to our knowledge, is the first study assessing resilience Step 3: Listing candidate indicators. For each impact aspect, we
of green water (a larger share than blue water) in the crop pro preliminarily identified and determined corresponding candidate in
duction based on the tool of water footprint; dicators, some of which refer to the literature on social-ecological sys
(III) the spatiotemporal evolution of the resilience of the communities tem resilience assessment (Fader et al., 2016; Meuwissen et al., 2019;
using the water resources for crop production in China has been Ruess, Konar, 2019; Varis et al., 2019; Wu et al., 2020); other indicators
elucidated. were drawn on authoritative resilience reports (FAO, 2016; Resilience
alliance, 2010). The remaining indicators were created by our under
2. Methods and data standing of the connotation of the specific dimension of resilience. These
candidate indicators were designed to be interdisciplinary and repre
2.1. Construction of water resilience assessment framework (WRAF) in sented the social, economic, and ecological systems that were directly or
crop production indirectly linked to the resilience of the communities using the water
resources systems related to crop production in a comprehensive,
The development of the WRAF follows eight steps (Fig. 1): concise and consistent manner.
Step 1: Defining the dimension range and scope of WRAF. We Step 4: Formulating the principles of screening indicators. Candidate
reviewed the connotation and measurement of resilience across indicators should be further screened against the following five criteria:
Fig. 1. Schematic diagram of the process of constructing the water resilience assessment framework (WRAF) in crop production.
4
(I) Relevance. The indicators should have direct or apparent indirect where Ej represents the entropy of index j. Ej was obtained from the
impacts on the water resources for crop production and have following equation:
obvious correlations with specific dimensions or major aspects of n (
∑ )
resilience; fi,j × ln fi,j
(II) Typicality. The indicators selected should be the typical and most Ej = − i=1
(5)
ln(n)
representative indicators adopted by well-known international
organizations or relevant literature to enable comparisons with
where fi,j is the proportion of the normalized value of index j in province
previous studies to some extent;
i in all provinces.
(III) Simplicity. The indicators selected should be simple and trans
Finally, the overall resilience score was the sum of the resilience
parent as much as possible to ensure the universal applicability of
scores of the four dimensions:
the framework;
(IV) Availability. The indicators should have available data for mul ∑
4
tiple years and regions to enable comparisons or track their RESi,c = Dm,i,c (6)
m=1
progress across regions;
(V) Quantifiable. The indicators selected should be measurable where RESi,c is the total resilience score of the communities using the
(including direct acquisition directly from statistics, indirect water resources produced by crop c in province i. Dm,i,c represents the
calculation or the literature) for the purpose of objectivity of the resilience score of crop c in province i in the specific dimension. The
WRAF results. resilience score of each dimension is between 0 and 1, i.e., the total
resilience score is 4. A higher score indicates more resilient communities
Step 5: Evaluating and determining the indicators of the first version using the water resources in crop production. In this study, total resil
of the WRAF. After the above candidate indicators were subjected to the ience is divided into four grades: low resilience (0− 1), lower-middle
screening principles, the final indicators for the first version of the resilience (1− 2), upper-middle resilience (2− 3) and high resilience
WRAF were evaluated and determined from over 50 candidate in (3− 4).
dicators, in which there was at least one specific indicator in each major We obtained the WARF framework through the strict eight steps. The
aspect. construction of the WRAF adopted the latest concept from the dynamic
Step 6: Verifying the feasibility of indicators. We conducted an evolution of social-ecological system resilience assessment. For the
empirical feasibility verification on all the indicators in the first version complex social-ecological system of water resources related to crop
of the WRAF. Specifically, we repeatedly discussed the interpretation production, the capacity of the system to flexibly respond and adapt to
ability of the indicators for the resilience dimension and again verified disturbances through adaptation, learning and self-organization is of the
the feasibility of data collection and resolution. Note that the indepen utmost importance, as is its ability to resist impacts while creating
dence of candidate indicators was ensured through collinearity diag redundancy and diversity to cushion against pressures. Moreover,
nosis, and the results indicated that there was no statistical collinearity transformative capacity is increasingly considered to be a crucial
(VIF<10) between indicators (Table S1). dimension of social-ecological resilience (Table 2). Note that the shock
Step 7: Determining of the final indicators and then normalize the and the formation mechanism of blue water and green water resilience
indicators. We determined the final indicators after all the above has either similarities (e.g., the same water function for crop production)
screening processes and repeated verification were completed. We or differences that blue water resilience is often affected by human and
normalized each raw indicator to the range of 0–1 to make a cross irrigation infrastructure, while green water is mainly determined by
comparison between indicators. climate change and land endowment. Therefore, some indicators are
xi,j − xmin,i,j consistent with blue and green water, and the rest distinguish these
The positive index : Zi,j + = (1)
xmax,i,j − xmin,i,j differences.
Specifically, resistance is the ability of resilience that the regional
The negative index : Zi,j − =
xmax,i,j − xi,j
(2) agricultural water system first shows to maintain a stable state when
xmax,i,j − xmin,i,j facing disturbance. It is often expressed as several synonyms (e.g.,
robustness, persistence and stability) (Béné and Doyen, 2018). Water
where Zij is the normalized value of indicator j (j = 1, 2. 4) in provincial i resource resistance in crop production is related to water and land
(i = 1, 2. 31) under each dimension, and xmax,i,j and xmin,i,j are the resource endowment, economic level and agricultural infrastructures
maximum and the minimum values, respectively. To avoid the error (Falkenmark and Wang-Erlandsson, 2021; Gleeson et al., 2020).
caused by extreme outlying values, the values of the 95th percentile and Therefore, we selected four specific indicators, per capita blue water
the 5th percentile of the raw value were taken as the maximum and the resources (per capita arable area for green water), blue water resource
minimum values, respectively. stress index during crop growth period (precipitation anomaly index
Step 8: Calculating the resilience score. We gave equal weight to during crop growth period for green water), per capita disposable in
each dimension, as the four dimensions of resilience are of equal come of rural residents and agricultural infrastructures as proxies to
importance. The weight of specific indicators was determined by the characterize the natural, economic and engineering resistance of the
widely used objective weighting methodology, the entropy method (Liu communities using the water resources related to crop production to
et al., 2019; Lu et al., 2022): shocks in terms of natural resource endowment and scarcity, economy
∑
4
( ) and infrastructure, respectively (Table 2). The per capita disposable
Dm = Zi,j × wj (3) income of rural residents is an indicator that has been proven to more
j=1 truly reflect residents’ economic ability to cope with shocks than per
capita GDP (Wu et al., 2020).
where Dm is the four different dimensions of resilience, m = 1, 2.4; wj Absorptive capacity of resilience defines the ability of a regional
indicates the weight of each indicator, which is calculated as: agricultural water system to buffer, absorb and recover from shocks. Its
1 − Ej major characteristic is that the system can sustain temporary damage
wj = 4 (
(4) without changing the current state of performance if shocked (Dewulf
∑ )
j=1
1 − Ej et al., 2019). Absorptive capacity is a primary strategy for a system to
deal with a shock and ensures the durability of the system function (Béné
5
and Doyen, 2018). Generally, the redundancy or diversity of the water productivity.
system is increased to improve the ability of the system to absorb shocks.
Therefore, we selected the blue (green) water resources redundancy and
virtual blue (green) water storage to represent the water system’s 2.2. Quantification of key indicators
redundancy capacity to long-term and short-term shocks, respectively.
Additionally, the reservoir effectively regulates water resources via en The majority of indicators were directly obtained from statistical
gineering for “storage and release” to mitigate the impact of the shock. data, while certain specific indicators (including the blue and green
Finally, income diversity measures the composition of income sources to water footprints of crop production, unused water resource redundancy,
ensure the ability to obtain income from other channels or industries in major agricultural infrastructures, diversity of income sources,
case of shock. comprehensive urbanization rate, and virtual water storage) needed to
Adaptive capacity of resilience refers to the ability of regional agri be specifically quantified.
cultural water systems to adapt to environmental changes through The blue and green water footprints of crop production were simu
learning, adjusting, and self-organizing their own structure (Folke et al., lated by the AquaCrop water-driven crop model, which is launched by
2010). Adaptive capacity emphasizes the learning ability of the system, FAO and has been widely used to estimate the water footprints of crop
combining experience and knowledge to adjust its response to changing production (Wang et al., 2019; Yang et al., 2021; Zhu et al., 2016).
external drivers and internal processes (Folke et al., 2010). Favorable Specifically, the AquaCrop model is based on root zone water dynamic
adaptability means adapting to shocks without significantly reducing balance. The proportion of blue water and green water in daily tran
key functions and reorganizing and configuring (Béné and Doyen, spiration (mm/d) can be further estimated by discerning the contribu
2018). Adaptive capacity is usually related to the concepts of workers’ tion of daily rainfall and irrigation water to the soil water balance
income, learning ability, and self-organization (Folke et al., 2010). For (Hukalla et al., 2015; Zhuo et al., 2016). More details on the calculation
agricultural water consumption social-ecological system, the adaptive of the water footprint of crop production can be found in Wang et al.
capacity of the communities is achieved by the adaptive measures such (2019).
as learning and investing in the use of more water-saving technologies Water resource redundancy is one of the key indicators of this study.
and equipment, and advanced management models. Therefore, we The estimation of water resource redundancy related to crop production
finally determined the net income of crops, disaster adaptability, the was slightly improved following Fader et al. (2016) and Mueller et al.
average years of education of rural residents and the blue (or green) (2012). Here, we assumed that the proportion allocated to crop c in the
water self-sufficiency rate to measure the adaptability of the water redundant blue water resources was the same as the share of utilized
resource system in crop production. The sustainable and stable income blue water. The green water surplus was related to the unused arable
cropping system is particularly important for producers to take adaptive land area. It was assumed that the share of green water surplus allocated
measures actively, which can indirectly promote the water resource to crop c is equal to the amount of green water used. The specific
system to adapt to the shocks and is the basis for establishing long-term calculation is as follows:
sustainable operation. Disaster adaptability is the integrated ability to ( ) WF blue [i, c]
use experience and technology to adjust, respond and adapt disasters Rblue [i, c] = Qi − EFRi − AW i − IW i − DW i − Pvar,i × (7)
WFblue [i, total]
and disturbances. The average years of education of rural residents is the
intuitive embodiment of workers’ learning and self-organization ability. Arae[i, c]
Rgreen [i, c] = (TAi − ALi − CLi − PLi − NALi ) × Pe,i × a × (8)
The blue (green) water self-sufficiency rate shows the ability of regional Area[i, total]
agricultural water system self-sufficiency, self-coordination, and man
agement. A higher self-sufficiency rate of the crop water footprint cor where Rblue [i, c] and Rgreen [i, c] represent the blue water and green water
responds to a stronger ability of the system to adapt to external shocks redundancy of crop c in province i, in m3y-1. Qi is the quantity of
(e.g., drought, flood, and other extreme shocks). renewable water resources in province i, in m3y-1. EFRi is the environ
Transformative capability of resilience evaluates the ability to create mental flow requirement, in m3y-1. Here, we estimated EFR as 60% of
a new system if the structure of the existing system cannot be maintained the total renewable water resources following Li et al. (2020); AWi ,IWi
(Walker et al., 2004). Transformative capability abandons the idea of a and DWi were the quantities of agricultural, industrial and domestic
single steady state and is the necessary condition for changing the basic water, respectively, in m3y-1; and Pvar,i indicated the variability in pre
configuration of the system to create long-term resilience (Elmqvist cipitation. If the precipitation exceeds 10% and 20% of the multi-year
et al., 2019). The transformation of the communities using the water average precipitation, respectively, take 5% and 10% of Qi as
resources related to crop production occurs when the long-term viability the Pvar,i . TAi ，ALi ，CLi , PLi and NALi were the total land area, the
of the system cannot be maintained, such as reallocating the functional area of agricultural land (including arable, garden, forest and grassland
value of water resources to obtain alternative opportunities or exiting land), construction land (urban and rural construction, industrial and
irrigation and even agricultural production (Gleeson et al., 2020). The mining, transportation, and water conservancy facilities land), protected
strength of transformation mainly depends on the current crop pro areas, and unsuitable arable land (water area, Gobi Desert and bare rock
duction state, science and technology investment, urbanization degree, land) of province i, in ha; Pe,i referred to the effective precipitation of
technology and management and other factors. In this study, four in province i, in mm; a was the utilization coefficient of unused arable land,
dicators were selected to characterize the transformation ability of water which was taken as 0.7 following Fader et al. (2016); WFblue [i, c] was the
resource resilience related to crop production: the proportion of crop blue water footprint of crop c in province i; and WF blue [i, total] was the
production in the national production, R&D expenditure, comprehen total blue water produced by 22 crops in province i. The data were
sive urbanization rate and the variability of the per-unit-mass water currently the most available crop types in terms of blue water footprints
footprint of crop production. The first indicator reflects the share and and sourced from Zhuo et al. (2016). Area[i, c] and Area[i, total] are the
status of cropping structure in the nation. The higher the share, the more sown area of crop c in province i and the total sown area of crops in
difficult the transformation will be. The middle two indicators reflect the province i, respectively, in ha.
imperative conditions for the water system transformation from the Major agricultural infrastructures refer to irrigation and agricultural
perspective of science and technology and social prosperity, which are power infrastructures that affects the transmission and utilization of
the basis for evaluating the potential of the system transformation, water resources required for crop production. The irrigation investment
respectively (Béné and Doyen, 2018). The last indicator represents the of per unit effective irrigation area and agricultural power consumption
improvement and innovation potential of technology and management were used to characterize agricultural irrigation and agricultural power
of water system related to crop production from the perspective of water infrastructures, respectively, following to Huang et al. (2021). Finally,
6
equal weights were assigned to them to represent the major agricultural economic-related data have been converted to constant prices based on
infrastructures. the year 2000 to eliminate the impact of price fluctuations.
The diversity of income sources was expressed as a percentage of the
added value of the secondary and tertiary industries to the provincial 3. Results
GDP.
Disaster adaptability only calculated water-related disasters, i.e., the 3.1. Temporal progress of water resource resilience in crop production
ratio of the cause disaster area to the total disaster area related to floods
and droughts. On the national scale, the resilience of blue water and green water in
The comprehensive urbanization rate took into account the multi the production of the three crops over the period of 2000–2017 showed
dimensional attributes of urbanization, including population urbaniza a higher “basic value” for green water and better “growth” for blue water
tion (urban population proportion), economic urbanization (percentage (Fig. 2a). The overall resilience score of the communities using the water
of non-agricultural GDP), spatial urbanization (per capita urban con resources related to crop production varied among crops. The resilience
struction area), and social urbanization (per capita consumption of blue water in wheat production was significantly lower than that of
expenditure of urban residents). Equal weights were assigned to four other crops, as wheat is a typical water-intensive crop, and blue water
indicators to represent the comprehensive urbanization rate. consumption by wheat occurs mainly in the water-scarce winter.
The provincial data for the virtual water reserve were not available. Therefore, the blue water in wheat production is more vulnerable to the
The National Grain Reserve data in the OECD database (OECD-FAO, risk of water shortages. Although the water resource resilience related to
2022) were downscaled to a specific province according to the pro the production of the three crops was still at the lower-middle level
duction share in the corresponding province. Finally, the virtual water (1− 2) in 2017, it increased significantly during the study period. In
storage of blue water and green water for each crop was estimated ac particular, blue water resilience in rice and wheat production in 2017
cording to the unit mass water footprint of the crop production in the surged by over 50% compared with 2000, while that of green water
producing province. increased less than 35%, which was due to the substantial achievements
of blue water management during the study period.
2.3. Sensitivity analysis The different resilience dimension score of blue and green water
consumption in crop production in 2000 and 2017 was shown in Fig. 2b,
To explore the sensitivity of water resilience scores related to crop c. We found that in 2000 and 2017, both the resistance and absorptive
production to different indicators, we conducted sensitivity analysis. We capacities of water resilience of the communities were lower than the
employed the widely used sensitivity evaluation approach following Xu adaptive and transformative capacities. Compared with 2000 (Fig. 2b),
et al. (2020). Each indicator is changed one-by-one to detect the varia the scores of different dimensions of water resource resilience in crop
tion range of resilience score. The specific principle is as follows: production in 2017 showed that the absorptive capacity had no
/ detectable change but significantly improved in resistance and trans
Sx =
ΔX ΔP
(9) formative capacity of resilience (Fig. 2c). In particular, the resistance of
X P crops to blue water increased by 123.4% in 2017 compared with 2000,
followed by a transformative capacity of 81.6%, indicating that China
where ΔX and X is the change range and the initial value of the raw has made substantial progress toward socioeconomic development and
indicator; ΔP and P is the variation range and the initial value of the sustainable blue water utilization and management in recent years.
water resilience score caused by ΔX; Sx is the sensitivity coefficient. We further selected the four indicators with the largest weight of
Taking the year of 2017 as an example, we ran the sensitivity analysis for each resilience dimension: per capita disposable income of rural resi
blue and green water resilience of three crops. The value of each indi dents, redundancy of crop blue and green water resources, disaster
cator was set ± 10% of the initial value as the change range, and their adaptability and R&D expenditure to better explain the resilience score
water resilience score was recalculated to obtain the Sx . results for each crop dimension (Fig. S1). The per capita disposable in
come of rural residents increased by 6.3 times during the study period
2.4. Data sources (Fig. S1a), which promoted the ability of crops to resist shocks from an
economic perspective. The redundant quantity of blue and green water
The data for the water footprint quantitation of crop production, is the most representative indicator of crop absorptive capacity (i.e.,
including the monthly precipitation on the grid scale and temperature buffering long-term disturbance). The blue and green water redundancy
required for AquaCrop simulation, were sourced from the CRU-TS 4.03 of the three crops in 2017 decreased by 85.5% and 35.7%, respectively,
dataset (Harris et al., 2020). The data for AquaCrop simulation on compared with 2000, indicating that the buffer capacity of water re
annual average CO2 concentration data were obtained from the Monaloa sources related to crop production is dropping (Fig. S1b and Fig. S1c).
Observatory in Hawaii (NOAA, 2022). Soil texture data were taken from Disaster adaptability is the integrated ability of human and water con
the Israeli soil and terrain database (Dijkshoorn et al., 2008). The sumption systems to adapt to disasters and disturbances. During the
cropping structures of the three crops were taken from the Mirca 2000 study period, the disaster adaptability of crops fluctuated and increased
database and were adjusted in proportion to meet the annual statistical by 13% points (Fig. S1d), indicating that the adaptability of the water
data of each crop within the provincial scope (NBSC, 2022). The data resource system associated with crop production was enhanced during
related to resilience indicators, including per capita disposable income, the study period. R&D expenditure can directly reflect the scientific
per capita arable land area, water withdrawal of each industry, total research capital investment and transformative potential of the system.
water resources, disaster area, proportion of nonagricultural GDP, R&D R&D expenditure increased by 23.1 times over 2000–2017, resulting in
expenditure, and composite urbanization, were extracted from the na the transformative capacity of the communities using the water re
tional data network of the National Bureau of Statistics of China (NBSC, sources for crop production markedly increasing. More details on spe
2022). The reservoir capacity was from the Water Resources Bulletin of cific indicators were given in Table S2-S7.
the Ministry of Water Resources of China (MWRC, 2021). The data of net
crop income were derived from the compilation of cost–benefit of 3.2. Spatial heterogeneity of water resource resilience in crop production
agricultural products (CNKI, 2022). The data for the composite index of
major agricultural infrastructures were sourced from our previous study Fig. 3 further elaborates the spatial pattern of the overall resilience
(Huang et al., 2021). The crop storage data were refined from the score of the communities using the water resources in crop production
OECE-FAO Agricultural Outlook 2019–2028 (OECE, 2022). All on the provincial scale in 2017 and the relative change rate in 2017
7
Fig. 2. Resilience scores of blue and green water consumptions by the communities for rice, wheat and maize production from 2000 to 2017 (a), and scores of each
dimension in 2000 (b) and 2017 (c) in China. A higher value means a better performance for the specific dimension.
compared with 2000. On the whole, the resilience of blue water and the relatively undeveloped economy in these provinces (Fig. S4).
green water resources used by the communities was at the level of lower- We further exhibited the spatial distribution characteristics of resil
medium (≤2) and presented similar spatial distribution characteristics, ience of the communities using the water resources in various di
i.e., the highest water resilience in the main production areas or rela mensions to explore the composition and formation mechanism of water
tively developed economic regions (Fig. S2). The resilience of blue and resource resilience (Fig. 4). The provinces with the highest resistance to
green water resources used by the communities in rice production was blue water for the three crops were located in Shanghai, which was
highest in the provinces in east and central China (Jiangsu, Guangdong, relatively economically developed (Fig. 4a). The highest green water
Shanghai and Hunan), and the provinces with upper-medium to high resistance of resilience was in the main production areas of various
resilience of blue water resources used by the communities in rice pro crops, such as Heilongjiang (0.71) for rice, Henan (0.67) for wheat, and
duction were more than those for wheat and maize (Fig. 3a). The blue Jilin (0.49) for maize as a higher share of crop production in these
and green water resilience in wheat production was the highest in provinces. In contrast, the resistance of resilience to blue and green
Shandong and Jiangsu, respectively, while for maize, it was the highest water was the weakest in northern China, except Beijing and Tianjin.
in Shanghai and Heilongjiang, which is generally consistent with the This is probably because the resistance of resilience of the communities
spatial distribution trend of crop water footprint (Fig. S3). Meanwhile, depends not only on the water endowment but also on the economic
there was apparent spatial heterogeneity in the resilience scores of the development (Fig. S2).
communities using the water resources. The provinces with the highest In terms of the absorptive capacity of the communities using the
resilience of water resources were 1.7–2.4 times higher than those with water resources for crop production, the provinces with the strongest
the lowest resilience. The region with the lowest water resource resil absorptive capacity of blue and green water were in the main production
ience score for crop production was concentrated in the region with areas of crops and provinces with richer water and arable land endow
scarce water and a relatively undeveloped economy. For instance, the ments (Fig. 4b). Specifically, the blue water absorptive capacity for rice,
region with the lowest blue and green water resilience scores for rice and wheat and maize was the highest in Hunan, Tibet and Hubei, respec
maize was in the northwest, while the region with the lowest water tively, as water resources are relatively abundant to cushion the shock in
resource resilience score for wheat was in Jilin. Additionally, the resil these provinces. The green water absorptive capacity also showed a
ience of blue and green water resources used by the communities in most similar trend to that of blue water.
provinces increased by varying degrees during the study period, except With respect to the adaptive capacity of the communities using the
for the resilience of green water resources in wheat production in a few water resources, we found that the adaptive capacity of resilience was
provinces (e.g., Tibet, Gansu and Ningxia). This was primarily attributed concentrated in advantageous production areas (Fig. 4c). The adaptive
to the lack of suitable meteorological conditions or water resources and capacity of the communities using the blue and green water resources in
8
Fig. 3. The provincial spatial pattern of the overall resilience scores of rice, wheat and maize in China in 2017 (background), the changes in 2017 compared with
2000 (red triangle or yellow circle), and the number of provinces with overall resilience scores at different levels (frequency histogram).
rice was the highest in Heilongjiang, which was the main rice-producing communities using the blue and green water resources is closely related
area. Similarly, the highest adaptive capacity of blue and green water in to regional meteorological conditions, water and soil resource endow
maize was in the northeastern region (Heilongjiang, Jilin and Liaoning) ment and technological conditions.
(Fig. S3). These discoveries revealed that the adaptive capacity of the The transformative capacity of resilience of the communities using
9
Fig. 4. The spatial pattern of the (a) resistance, (b) absorptive capacity, (c) adaptive capacity, and (d) transformative capacity of blue and green water resilience
associated with the production of rice, wheat and maize in China in 2017 (background) and the composition of the specific indicators of the provinces with the
highest score (pie chart).
the water resource in crop production is closely interrelated with rice, which reached 0.94 because the policy of the reform and opening-
regional socioeconomic development (Fig. 4d). The province with the up was first implemented in Guangdong, resulting in Guangdong being
highest blue and green water transformative capacity for the three crops at the forefront of transformation compared with other provinces.
was Guangdong, especially the blue water transformative capacity for
10
4. Discussion and Lu et al. (2022).

We also explored the structural complexity of water resource resil
4.1. Integrating blue and green water resilience management in crop ience in crop production. The resistance and absorptive capacities were
production lower than the adaptive and transformative capacities for water resource
resilience in China. In recent decades, China’s progress in socioeconomic
This study utilizes, for the first time, the water footprint as a tool to aspects has attracted global attention, though to a certain extent this
develop a feasible quantitative framework for assessing the social- progress comes at the cost of accelerating resource consumption and
ecological resilience (i.e., general resilience) of the communities using environmental degradation (Liu, 2010). Moreover, we observed that the
the blue and green water resource related to crop production. Subse four dimensions of water resource resilience are mainly synergistic,
quently, a case study at the provincial level in China is verifying the while trade-offs between individual dimensions (e.g., resistance and
applicability of this framework in the assessment of the resilience of the absorption capacity of resilience in green water for rice vs. its trans
communities using the blue and green water resource in staple crop formative capacity, as well blue water resource adaptive vs. trans
production. The result of sensitivity analysis shows that the sensitivity of formative capacity) (Fig. 5). Therefore, the progress of each dimension
the blue and green water resilience of the three crops to the change of of resilience and the potential synergy or trade-off relationship between
specific indicators is small (all less than 0.3) (Fig. S5), which further dimensions need ongoing attention and integration to enhance water
proves the robustness of the framework. resource resilience. For China, it is particularly urgent to enhance the
Currently, the majority of the literature on ecological water resil resistance and absorptive capacities of water resource system resilience
ience focuses on blue water resources (e.g., rivers, wetlands and lakes) related to crop production. Efforts in terms of the economy (increasing
(Behboudian et al., 2021; Li et al., 2016; Varis and Kummu, 2019). the per capita disposable income of rural residents), technology
However, several recent studies have shown that green water has been (improving water productivity) and policy (optimizing cropping pat
disturbed by human activities, such as desertification and soil erosion in terns) are clearly warranted.
the Loess Plateau of China, and hydrological cycle changes caused by
widespread deforestation (Falkenmark and Wang-Erlandsson, 2021; 4.2. Potential solutions to enhancing water resilience
Gleeson et al., 2020). Our results emphasize the essential contribution of
green water to water resource resilience of communities associated with Our results also highlight the spatial heterogeneity of blue and green
crop production, which is an improvement on the existing research. water resource resilience. Overall, China’s water resource resilience in
Eighty-three percent of human water consumption is sourced from green crop production is currently at a lower-middle level and still has much
water (Hoekstra and Mekonnen, 2012). Therefore, green water plays a room for improvement. However, significant regional heterogeneity in
decisive role in ensuring regional water resources and food security. We the resilience of water resources associated with crop production is
found that green water resilience in crops is generally higher than blue observed through cross-provincial comparison. The resilience of water
water resilience in crops. Moreover, green water resources mainly resources in crop production involves extensive aspects, including the
depend on precipitation and arable land area, which is controlled by natural, social, economic, infrastructure and governance of agricultural
climate change, agricultural production and national policies. China has production and water resource management. Different provinces have
adhered to the arable land area protection red line of 0.12 billion ha to advantages in one or several aspects of these factors, but no province has
ensure food security (The State Council of China, 2007), which is advantages in all aspects, i.e., all provinces need to be supplemented in a
beneficial to the green water resilience of crops to a certain extent. Given certain dimension. For instance, Beijing and Hebei in the North China
the crucial role of green water consumption in crop production, it is Plain are weak in the resistance and absorptive capacity of water re
urgent to better understand and evaluate it so that we can continue to sources resilience for rice and wheat production as their limited water
maintain a reasonable arable land area to guarantee resilient green resources while lager water demand, which is similar to the results
water resources. observed by Varis and Kummu (2019) in the resilience of rivers in China.
Additionally, as the traditional object of water resource manage Therefore, several targeted measures by the communities or institutions
ment, blue water resources are under intense pressure due to the on various scales (from local water management authorities to the na
competitive consumption of agriculture, industry, households and tional policies mentioned above) to improve water resilience can be
ecology, all of which also increase the demand for blue water in crop taken (e.g., adopting water-saving irrigation equipment and technology,
production areas. We found that the crop-related blue water redundancy more diversified water resources utilization and irrigation infrastructure
decreased by 85.5% from to 2000. Therefore, blue water is more maintenance and update, and increasing physical or virtual water re
vulnerable than green water. Hence, the protection and efficient utili sources reserves through the South-to-North Water Transfer Project or
zation of blue water resources are still the key measures to build a virtual water trade) to improve the resilience of water resources related
resilient blue water system for crop production. Efforts can be made to crop production. Note that the resilience of water resources in
chiefly via two aspects. The first is the relevant policy. Over the past Northwest China is relatively weak in all dimensions since natural re
decade, the Chinese government has promulgated a series of policies to sources, economy, society and technology are relatively underdeveloped
safeguard blue water resources, of which strict control of water re for these northwest provinces. There are some differences between our
sources is the most influential (The State Council of China, 2012). The findings and the results obtained by Lu et al. (2022) as inconsistencies in
Ministry of Water Resources in China has successively imposed 105 the evaluation objects and spatial scale of water resilience. Therefore,
water quotas involving agriculture, industry, construction and service measures such as the polycentric governance of water resources,
industries since 2019 (MWRC, 2021). These policies have further including the broader forms of civil society, community organizations
strengthened the excellent management of blue water resources and are and local self-government to improve water resilience are required.
expected to improve their resilience. The second is engineering mea Although Tibet and Qinghai have advantages in water resources
sures. China’s investment in irrigation infrastructure has increased endowment, their water resources are particularly sensitive to climate
34-fold in the last 20 years (Huang et al., 2021) and includes the con change. Therefore, establishing long-term or short-term climate pre
struction of canal and irrigation water redistribution projects and the diction and early warning system to increase the adaptability of water
vigorous expansion of equipment and advanced water-saving technol resources to the changing environment are needed. In contrast, the most
ogies (e.g., sprinkler irrigation and micro irrigation). Owing to these resilient provinces for blue water in rice are in East provinces (Zhejiang,
efforts, the resilience of China’s blue water resource consumption of Jiangsu and Shanghai), which profit from their excellent adaptive ca
crop production has shown an increasing trend over time, which is pacity and transformative capacity. Table 3 lists the main pressures or
consistent with the conclusions conducted by Varis and Kummu (2019) shocks, the current state of resilience, and the corresponding resilience
11
Fig. 5. Correlation between each dimension of resilience in blue (a) and green (b) water of rice, blue (c) and green (d) water of wheat, and blue (e) and green (f)
water of maize. Blue and red indicate positive and negative correlations, respectively, and the darker the color is, the higher the degree of correlation.
12
Table 3 incorporated in in-depth research. Third, we only consider the water

Pressures or shocks, the current state of resilience, and the corresponding resources in the process of crop production, ignoring the water resource
resilience improvement strategies for agricultural water resources in typical consumption in subsequent links such as crop transportation and pro
provinces. cessing. Future research on the resilience of water consumption in the
Provinces Main disturbance State of water Water resilience whole supply chain of crops from production to consumption is highly
or shock resilience and improvement recommended.
characteristics of strategy
water and economy
5. Conclusions
Beijing/ Transition water Low resistance and Adopting water-
Hebei consumption and absorptive capacity of saving irrigation
groundwater level resilience; The water equipment and A social-ecological WRAF related to crop production in four di
drop caused by resources are limited technology, more mensions of resistance, absorptive capacity, adaptive capacity and
irrigation and the severe water diversified water transformative capacity and distinguishing blue and green water is
scarcity while the resources utilization
proposed and applied in the case for staple crops in China. Results show
economy is developed (water recycling,
renewable water that the resilience of water resources in crop production in China has
irrigation) and been outstanding in terms of adaptive capacity and transformative ca
irrigation pacity in the last 20 years, which are 8.9~153.7% higher than resistance
infrastructure and absorptive capacities of resilience. The resilience of green water in
maintenance and
crop production generally has a higher basic value, while the resilience
update, and increase
physical or virtual of blue water has increased by over 50% over the study period; these
water resources findings emphasize the importance of integrating the resilience of blue
reserves through the water and green water resources in crop production, including efforts
South-to-North
toward awareness, policy and engineering. Improving blue water use
Water Transfer
Project or virtual
efficiency and protecting arable land area are key policy recommenda
water trade tions for promoting the resilience of blue and green water resources
Gansu Water shortage Low resistance and Strengthening related to crop production. Our findings provide opportunities for the
caused by land use adaptive capacity of polycentric risk assessment of blue water and green water resources related to
change and resilience; The water governance of water
regional crop production and broaden the scope of regional agricultural
irrigation resources are limited, resources, including
and the severe water broader forms of civil water resources, especially the sustainability assessment of water re
scarcity and the society, community sources associated with crop production.
economy is organizations and
undeveloped local self-
government Declaration of Competing Interest
Tibet/ Climate change Low adaptability of Establishing long-
Qinghai water resources term or short-term
resilience; Abundant climate prediction The authors declare that they have no known competing financial
in water resources, and early warning interests or personal relationships that could have appeared to influence
while sensitive to system to increase the work reported in this paper.
climate change and the adaptability of
backward in economy water resources to
the changing
Data Availability
environment
Yunnan Hydrological cycle Weak to adaptive and Intensifying the Data will be made available on request.
changes caused by transformative ecosystem services
deforestation capacity of resilience; accounting, and
Abundant in water enhance the Acknowledgments
resources, but weak awareness and policy
awareness of publicity of water The study is financially supported by the Program for Cultivating
protection and resources resilience Outstanding Talents on Agriculture, Ministry of Agriculture and Rural
backward economy investment
Affairs, People’s Republic of China (13210321), and Cyrus Tang Foun
dation to L. Zhuo.
improvement strategies for agricultural water resources in typical
provinces. Appendix A. Supporting information
Supplementary data associated with this article can be found in the

4.3. Limitations online version at doi:10.1016/j.agwat.2023.108485.
This study has several limitations. First, the current WRAF does not References
include the indicators in terms of several essential aspects (e.g., cultural,
institutional and government governance indicators) as data limitations, Ansah, I.G.K., Gardebroek, C., Ihle, R., 2019. Resilience and household food security: a
and some indicators are not specifically formulated for water resources. review of concepts, methodological approaches and empirical evidence. Food Secur.
11 (6), 1187–1203. https://doi.org/10.1007/s12571-019-00968-1.
This set of 16 indicators may be expanded or improved in the future, but Asefa, T., Clayton, J., Adams, A., Anderson, D., 2014. Performance evaluation of a water
we believe that they are currently the best and most advanced frame resources system under varying climatic conditions: Reliability, resilience,
work available for assessing the resilience of water resource systems for vulnerability and beyond. J. Hydrol. 508, 53–65. https://doi.org/10.1016/j.
jhydrol.2013.10.043.
crop production. With the availability of data and a deepening under
Behboudian, M., Kerachian, R., Pourmoghim, P., 2021. Evaluating the long-term
standing of resilience, the WRAF will be further improved and perfected resilience of water resources systems: Application of a generalized grade-based
in the future. Second, only three staple crops are considered in this combination approach. Sci. Total Environ. 786 https://doi.org/10.1016/j.
study. Although the three crops account for more than 85% of the sown scitotenv.2021.147447.
Béné, C., Doyen, L., 2018. From resistance to transformation: A generic metric of
area in China, they cannot reflect the resilience of the whole water re resilience through viability. Earth’s Future 6 (7), 979–996. https://doi.org/10.1002/
sources system related to crop production. More crops should be 2017ef000660.
13
Biggs, R., Schlüter, M., Biggs, D., Bohensky, E.L., BurnSilver, S., Cundill, G., Dakos, V., virtual water flows in China. J. Clean. Prod. 309, 127455 https://doi.org/10.1016/j.
Daw, T.M., Evans, L.S., Kotschy, K., Leitch, A.M., Meek, C., Quinlan, A., Raudsepp- jclepro.2021.127455.
Hearne, C., Robards, M.D., Schoon, M.L., Schultz, L., West, P.C., 2012. Toward Ifejika Speranza, C., Wiesmann, U., Rist, S., 2014. An indicator framework for assessing
principles for enhancing the resilience of ecosystem services. Annu. Rev. Environ. livelihood resilience in the context of social–ecological dynamics. Glob. Environ.
Resour. 37 (1), 421–448. https://doi.org/10.1146/annurev-environ-051211- Change 28, 109–119. https://doi.org/10.1016/j.gloenvcha.2014.06.005.
123836. Kahiluoto, H., Kaseva, J., Balek, J., Olesen, J.E., Ruiz-Ramos, M., Gobin, A.,
Cao, X., Zeng, W., Wu, M., Guo, X., Wang, W., 2020. Hybrid analytical framework for Kersebaum, K.C., Takac, J., Ruget, F., Ferrise, R., Bezak, P., Capellades, G.,
regional agricultural water resource utilization and efficiency evaluation. Agric. Dibari, C., Makinen, H., Nendel, C., Ventrella, D., Rodriguez, A., Bindi, M.,
Water Manag. 231, 106027 https://doi.org/10.1016/j.agwat.2020.106027. Trnka, M., 2019. Decline in climate resilience of European wheat. Proc. Natl. Acad.
Cao, X., Bao, Y., Li, Y., Li, J., Wu, M., 2023. Unravelling the effects of crop blue, green Sci. USA 116 (1), 123–128. https://doi.org/10.1073/pnas.1804387115.
and grey virtual water flows on regional agricultural water footprint and scarcity. Krueger, E.H., Borchardt, D., Jawitz, J.W., Klammler, H., Yang, S., Zischg, J., Rao, P.S.C.,
Agric. Water Manag. 278, 108165 https://doi.org/10.1016/j.agwat.2023.108165. 2019. Resilience dynamics of urban water supply security and potential of tipping
Cinner, J.E., Barnes, M.L., 2019. Social dimensions of resilience in social-ecological points. Earth’s Future 7 (10), 1167–1191. https://doi.org/10.1029/2019ef001306.
systems. One Earth 1 (1), 51–56. https://doi.org/10.1016/j.oneear.2019.08.003. Li, G., Kou, C., Wang, Y., Yang, H., 2020. System dynamics modelling for improving
CNKI). (2022). China Yearbooks Full-Text Database. Retrieved from http://epub.cnki.net/ urban resilience in Beijing, China. Resour. Conserv. Recycl. 161, 104954 http://doi.
kns/brief/result.aspx?dbPrefix=CYFD. org/10.1016/j.resconrec.2020.104954.
Cottrell, R.S., et al., 2019. Food production shocks across land and sea. Nat. Sustain. 2 Li, Y., Lence, B.J., 2007. Estimating resilience for water resources systems. Water Resour.
(2), 130–137. https://doi.org/10.1038/s41893-018-0210-1. Res. 43 (7), W07422. https://doi.org/10.1029/2006wr005636.
Cumming, G.S., Peterson, G.D., 2017. Unifying research on social-ecological resilience Li, Y., Li, Y., Wu, W., 2016. Threshold and resilience management of coupled
and collapse. Trends Ecol. Evol. 32 (9), 695–713. https://doi.org/10.1016/j. urbanization and water environmental system in the rapidly changing coastal region.
tree.2017.06.014. Environ. Pollut. 208 (A), 87–95. https://doi.org/10.1016/j.envpol.2015.08.042.
Cutter, S.L., Derakhshan, S., 2018. Temporal and spatial change in disaster resilience in Liu, D., Qi, X., QiangFu, Li, M., Zhu, W., Zhang, L., Abrar Faiz, M., Khan, M.I., Li, T.,
US counties, 2010–2015. Environ. Hazard. 19 (1), 10–29. https://doi.org/10.1080/ Cui, S., 2019. A resilience evaluation method for a combined regional agricultural
17477891.2018.1511405. water and soil resource system based on Weighted Mahalanobis distance and a Gray-
Dewulf, A., Karpouzoglou, T., Warner, J., Wesselink, A., Vij, S., Buytaert, W., 2019. The TOPSIS model. J. Clean. Prod. 229, 667–679. https://doi.org/10.1016/j.
power to define resilience in social–hydrological systems: toward a power-sensitive jclepro.2019.04.406.
resilience framework. Wiley Interdiscip. Rev.: Water 6 (6), e1377. https://doi.org/ Liu, J., 2010. China’s road to sustainability, 50-50 Science 328 (5974). https://doi.org/
10.1002/wat2.1377. 10.1126/science.1186234.
Di Baldassarre, G., Sivapalan, M., Rusca, M., Cudennec, C., Garcia, M., Kreibich, H., Lu, C., Ji, W., Hou, M., Ma, T., Mao, J., 2022. Evaluation of efficiency and resilience of
Konar, M., Mondino, E., Mard, J., Pande, S., Sanderson, M.R., Tian, F., Viglione, A., agricultural water resources system in the Yellow River Basin, China. Agric. Water
Wei, J., Wei, Y., Yu, D.J., Srinivasan, V., Bloschl, G., 2019. Socio-hydrology: Manag. 266, 107605 https://doi.org/10.1016/j.agwat.2022.107605.
scientific challenges in addressing the sustainable development goals. Water Resour. Mueller, N.D., Gerber, J.S., Johnston, M., Ray, D.K., Ramankutty, N., Foley, J.A., 2012.
Res. 55, 6327–6355. https://doi.org/10.1029/2018WR023901. Closing yield gaps through nutrient and water management. Nature 490 (7419),
Diao, K., Sweetapple, C., Farmani, R., Fu, G., Ward, S., Butler, D., 2016. Global resilience 254–257. https://doi.org/10.1038/nature11420.
analysis of water distribution systems. Water Res. 106, 383–393. https://doi.org/ MWRC). (2021). 105 national water quotas. Retrieved from http://qgjsb.mwr.gov.cn/
10.1016/j.watres.2016.10.011. zcfg/bzde/slbwj/202112/t20211214_1555041.html.
Dijkshoorn, J.A., Engelen, V.W.P.V., Huting, J.R.M., 2008. Soil and landform properties for , 2022NationalOceanic and Atmospheric Administration (NOAA). (2022). Retrieved
LADA partner countries (Argentina, China, Cuba, Senegal, South Africa and Tunisia) from https://www.esrl.noaa.gov.
https://www.isric.org/. NBSC). (2022). China statistical Yearbook. Retrieved from http://data.stats.gov.cn/
Elmqvist, T., Andersson, E., Frantzeskaki, N., McPhearson, T., Olsson, P., Gaffney, O., english/easyquery.htm?cn=E0103.
Takeuchi, K., Folke, C., 2019. Sustainability and resilience for transformation in the Piao, S., et al., 2010. The impacts of climate change on water resources and agriculture in
urban century. Nat. Sustain. 2 (4), 267–273. https://doi.org/10.1038/s41893-019- China. Nature 467 (7311), 43–51. https://doi.org/10.1038/nature09364.
0250-1. Resilience Alliance. (2010) Assessing resilience in social-ecological systems: workbook for
Fader, M., et al., 2016. Past and present biophysical redundancy of countries as a buffer practitioners (Revised version 2.0). Retrieved from http://www.resalliance.org/srv/
to changes in food supply. Environ. Res. Lett. 11 (5), 055008 https://doi.org/ file.php/261.
10.1088/1748-9326/11/5/055008. Roach, T., Kapelan, Z., Ledbetter, R., 2018. Resilience-based performance metrics for
Falkenmark, M., Rockström, J., 2010. Building water resilience in the face of global water resources management under uncertainty. Adv. Water Resour. 116, 18–28.
change: from a blue-only to a green-blue water approach to land-water management. https://doi.org/10.1016/j.advwatres.2018.03.016.
J. Water Resour. Plan. Manag. 136 (6), 606–610, 10.1061/(asce)wr.1943- Rockström, J., Falkenmark, M., Karlberg, L., Hoff, H., Rost, S., Gerten, D., 2009. Future
5452.0000118. water availability for global food production: The potential of green water for
Falkenmark, M., Wang-Erlandsson, L., 2021. A water-function-based framework for increasing resilience to global change. Water Resour. Res. 45 (7), W00A12. https://
understanding and governing water resilience in the Anthropocene. One Earth 4 (2), doi.org/10.1029/2007wr006767.
213–225. https://doi.org/10.1016/j.oneear.2021.01.009. Rodina, L., 2018. Defining “water resilience”: Debates, concepts, approaches, and gaps.
Falkenmark, M., Wang-Erlandsson, L., Rockström, J., 2019. Understanding of water Wiley Interdiscip. Rev.: Water 6 (2), e1334. https://doi.org/10.1002/wat2.1334.
resilience in the Anthropocene. J. Hydrol. X 2, 100009. https://doi.org/10.1016/j. Ruess, P.J., Konar, M., 2019. Grain and virtual water storage capacity in the United
hydroa.2018.100009. States. Water Resour. Res. 55 (5), 3960–3975. https://doi.org/10.1029/
FAO). (2016). Resilience Index Measurement and Analysis—II. Retrieved from https:// 2018wr024292.
www.fao.org/emergencies/resources/documents/ resources-detail/it/c/416587/. Schlüter, M., Haider, L.J., Lade, S.J., Lindkvist, E., Martin, R., Orach, K., Wijermans, N.,
(Accessed 6 Jan. 2022). Folke, C., 2019. Capturing emergent phenomena in social-ecological systems: an
Folke, C., Carpenter, S.R., Walker, B., Scheffer, M., Chapin, T., Rockström, J., 2010. analytical framework. Ecol. Soc. 24, 11. https://doi.org/10.5751/ES-11012-240311.
Resilience thinking: integrating resilience, adaptability and transformability. Ecol. Seekell, D., Carr, J., Dell’Angelo, J., D’Odorico, P., Fader, M., Gephart, J., Kummu, M.,
Soc. 15 (4), 299–305. https://doi.org/10.1890/08-2244.1. Magliocca, N., Porkka, M., Puma, M., Ratajczak, Z., Rulli, M.C., Suweis, S.,
Folke, C., Biggs, R., Norström, A.V., Reyers, B., Rockström, J., 2016. Social-ecological Tavoni, A., 2017. Resilience in the global food system. Environ. Res. Lett. 12 (2),
resilience and biosphere-based sustainability science. Ecol. Soc. 21 (3), 210341 025010 https://doi.org/10.1088/1748-9326/aa5730.
https://doi.org/10.5751/es-08748-210341. Sharifi, A., Yamagata, Y., 2016. Principles and criteria for assessing urban energy
Gleeson, T., et al., 2020. Illuminating water cycle modifications and earth system resilience: a literature review. Renew. Sustain. Energy Rev. 60, 1654–1677. https://
resilience in the Anthropocene. Water Resour. Res. 56 (4) https://doi.org/10.1029/ doi.org/10.1016/j.rser.2016.03.028.
2019wr024957. Sinclair, K., Rawluk, A., Kumar, S., Curtis, A., 2017. Ways forward for resilience thinking:
Grafton, R.Q., et al., 2019. Realizing resilience for decision-making. Nat. Sustain. 2 (10), lessons from the field for those exploring social-ecological systems in agriculture and
907–913. https://doi.org/10.1038/s41893-019-0376-1. natural resource management. Ecol. Soc. 22 (4), 220421 https://doi.org/10.5751/
Grêt-Regamey, A., Huber, S.H., Huber, R., 2019. Actors’ diversity and the resilience of es-09705-220421.
social-ecological systems to global change. Nat. Sustain. 2 (4), 290–297. https://doi. Sivapalan, M., Savenije, H.H.G., Blöschl, G., 2012. Socio-hydrology: a new science of
org/10.1038/s41893-019-0236-z. people and water. Hydrol. Process. 26, 1270–1276. https://doi.org/10.1002/
Harris, I., Osborn, T.J., Jones, P., Lister, D., 2020. Version 4 of the CRU TS monthly high- hyp.8426.
resolution gridded multivariate climate dataset. Sci. Data 7 (1). https://doi.org/ Steffen, W., Richardson, K., Rockstrom, J., Cornell, S.E., Fetzer, I., Bennett, E.M.,
10.1038/s41597-020-0453-3. Biggs, R., Carpenter, S.R., de Vries, W., de Wit, C.A., Folke, C., Gerten, D., Heinke, J.,
Herrera de Leon, H.J., Kopainsky, B., 2020. Do you bend or break? System dynamics in Mace, G.M., Persson, L.M., Ramanathan, V., Reyers, B., Sorlin, S., 2015. Planetary
resilience planning for food security. Syst. Dyn. Rev. 35 (4), 287–309. https://doi. boundaries: guiding human development on a changing planet. Science 347 (6223),
org/10.1002/sdr.1643. 1259855. https://doi.org/10.1126/science.1259855.
Hoekstra, A.Y., Mekonnen, M.M., 2012. The water footprint of humanity. Proc. Natl. Suweis, S., Carr, J.A., Maritan, A., Rinaldo, A., D’Odorico, P., 2015. Resilience and
Acad. Sci. USA 109 (9), 3232–3237. https://doi.org/10.1073/pnas.1109936109. reactivity of global food security. Proc. Natl. Acad. Sci. USA 112 (22), 6902–6907.
Hoiling, C.S., 1973. Resilience and stability of ecological systems. Annu. Rev. Ecol. Evol. https://doi.org/10.1073/pnas.1507366112.
Syst. 4, 1–23. The State Council of China. (2007). Hold the red line of 1.8 billion mu of cultivated land.
Huang, H., Zhuo, L., Wang, R., Shang, K., Li, M., Yang, X., Wu, P., 2021. Agricultural Retrieved from http://www.gov.cn/2007lh/content_543917.htm.
infrastructure: the forgotten key driving force of crop-related water footprints and The State Council of China. (2012). Implement the most stringent water management system.
Retrieved from http://www.gov.cn/zhuanti/2015–06/13/content_2878992.htm.
14
Varis, O., Kummu, M., 2019. The demanding quest for harmony: China’s polarizing Wu, Y., Que, W., Liu, Y.-g, Cao, L., Liu, S.-b, Zhang, J., 2020. Is resilience capacity index
freshwater resilience map. Environ. Res. Lett. 14 (5), 054015 https://doi.org/ of Chinese region performing well? Evidence from 26 provinces. Ecol. Indic. 112,
10.1088/1748-9326/ab1040. 106088 https://doi.org/10.1016/j.ecolind.2020.106088.
Varis, O., Taka, M., Kummu, M., 2019. The planet’s stressed river basins: too much Xu, Z., et al., 2020. Assessing progress towards sustainable development over space and
pressure or too little adaptive capacity? Earths Future 7 (10), 1118–1135. https:// time. Nature 577 (7788), 74–78. https://doi.org/10.1038/s41586-019-1846-3.
doi.org/10.1029/2019EF001239. Yang, X., Zhuo, L., Xie, P., Huang, H., Feng, B., Wu, P., 2021. Physical versus economic
Vörösmarty, C.J., Green, P., Salisbury, J., Lammers, R.B., 2000. Global water resources: water footprints in crop production: a spatial and temporal analysis for China.
vulnerability from climate change and population growth. Science 289 (5477), Hydrol. Earth Syst. Sci. 25 (1), 169–191. https://doi.org/10.5194/hess-25-169-
284–288 http://doi.org/DOI 10.1126/science.289.5477.284. 2021.
Walker, B., Holling, C.S., Carpenter, S.R., Kinzig, A., 2004. Resilience, adaptability and Zampieri, M., Weissteiner, C.J., Grizzetti, B., Toreti, A., van den Berg, M., Dentener, F.,
transformability in social–ecological systems. Ecol. Soc. 9 (2), 5. https://doi.org/ 2020. Estimating resilience of crop production systems: From theory to practice. Sci.
10.5751/ES-00650-090205. Total Environ. 735, 139378 https://doi.org/10.1016/j.scitotenv.2020.139378.
Wang, W., Zhuo, L., Li, M., Liu, Y., Wu, P., 2019. The effect of development in water- Zhuo, L., Mekonnen, M.M., Hoekstra, A.Y., 2016. The effect of inter-annual variability of
saving irrigation techniques on spatial-temporal variations in crop water footprint consumption, production, trade and climate on crop-related green and blue water
and benchmarking. J. Hydrol. 577, 123916 https://doi.org/10.1016/j. footprints and inter-regional virtual water trade: a study for China (1978-2008).
jhydrol.2019.123916. Water Res. 94, 73–85. https://doi.org/10.1016/j.watres.2016.02.037.
15

Main

Uploaded by

Copyright:

Available Formats

Main

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Main

Uploaded by

Copyright:

Available Formats

Agricultural Water Management 288 (2023) 108485

Contents lists available at ScienceDirect

Agricultural Water Management

1. Introduction shock and uncertainty on agricultural water resource systems, that

4. Discussion and Lu et al. (2022).

Table 3 incorporated in in-depth research. Third, we only consider the water

Supplementary data associated with this article can be found in the

You might also like