Mental models of food security in rural Mali

Abstract

Recent estimates indicate that 12% of the global population is likely to have suffered from chronic hunger, due to lack of enough food for an active and healthy life. West Africa, specifically across the Sahel countries, is acutely vulnerable to food insecurity concerns. Mail is emblematic of this problem with approximately 4.6 million citizens considered food insecure. Food security poses formidable challenges. Studies have shown that in order to understand food insecurity and identify steps for effective intervention, there is a need to apprehend the food systems and food in/security in a holistic way beyond production alone. Understanding the behavioral aspects of food security is critical in the African context where agriculture, while oriented toward basic subsistence, remains embedded in social system including the social dynamics of households, extended families, and communities. This exploratory work focuses on developing a nuanced understanding of food security and adaptive behaviors to current challenges to food security at the household level with a distinct focus on inter- and intra-family behavioral dynamics in rural, southern Mali. Using mental models methodology, we developed two influence diagrams and a set of sub-models that represent rural households’ mental models of food security under traditional conditions and under conditions of external pressures. These models suggest that food security in rural Mali is at considerable risk due to the influence of external challenges, such as climate change, on traditional behaviors and a lack of easily accessible corresponding behavioral adaptations.

Full text

Vol.:(0123456789)
1 3
Environment Systems and Decisions
https://doi.org/10.1007/s10669-017-9669-y
Mental models offood security inrural Mali
LouieRiversIII1 · UditaSanga2· AmadouSidibe3· AlexaWood1· RajivPaudel4· SandraT.Marquart‑Pyatt5·
ArikaLigmann‑Zielinska6· LauraSchmittOlabisi7· EricJingDu8· SawedaLiverpool‑Tasie2
© Springer Science+Business Media, LLC, part of Springer Nature 2017
Abstract
Recent estimates indicate that 12% of the global population is likely to have suffered from chronic hunger, due to lack of
enough food for an active and healthy life. West Africa, specifically across the Sahel countries, is acutely vulnerable to food
insecurity concerns. Mail is emblematic of this problem with approximately 4.6 million citizens considered food insecure.
Food security poses formidable challenges. Studies have shown that in order to understand food insecurity and identify
steps for effective intervention, there is a need to apprehend the food systems and food in/security in a holistic way beyond
production alone. Understanding the behavioral aspects of food security is critical in the African context where agriculture,
while oriented toward basic subsistence, remains embedded in social system including the social dynamics of households,
extended families, and communities. This exploratory work focuses on developing a nuanced understanding of food secu-
rity and adaptive behaviors to current challenges to food security at the household level with a distinct focus on inter- and
intra-family behavioral dynamics in rural, southern Mali. Using mental models methodology, we developed two influence
diagrams and a set of sub-models that represent rural households’ mental models of food security under traditional condi-
tions and under conditions of external pressures. These models suggest that food security in rural Mali is at considerable risk
due to the influence of external challenges, such as climate change, on traditional behaviors and a lack of easily accessible
corresponding behavioral adaptations.
Keywords Mental models· Food security· Climate change· Agriculture· Mali
1 Introduction
Recent estimates indicate that 12% of the global population
is likely to have suffered from chronic hunger, due to lack of
enough food for an active and healthy life (FAO 2013). West
Africa, specifically within the Sahel’s countries, is acutely
vulnerable to food insecurity. The West African Sahel has
experienced long and severe droughts in the 20th and 21st
centuries (Nicholson etal. 1998) which have had dramatic
environmental, economic, and social impacts in the region.
A particularly wet period in the 1950s was followed by a
persistent period of severe drought in the 1960s and the
1970s, affecting millions of people and creating a shift in the
climatic pattern and food insecurity in the region (Brooks
* Louie Rivers III
lrivers@ncsu.edu
1 Department ofForestry andEnvironmental Resources, North
Carolina State University, 22219 Jordan Hall Addition,
Raleigh, NC27695, USA
2 Department ofCommunity Sustainability, Michigan State
University, 480 Wilson Road, 151 Natural Resources
Building, EastLansing, MI48824, USA
3 International Crops Research Institute fortheSemi-Arid
Tropics, BP320Bamako, Mali
4 Department ofGeography, Michigan State University, 673
Auditorium Rd., 121 Geography Building, EastLansing,
MI48824-1117, USA
5 Department ofSociology, Environmental Science andPolicy
Program, Michigan State University, 509 E. Circle Dr., 316
Berkey Hall, EastLansing, MI48824-1111, USA
6 Department ofGeography, Environmental Science andPolicy
Program, Michigan State University, 673 Auditorium Rd.,
121 Geography Building, EastLansing, MI48824-1117,
USA
7 Department ofCommunity Sustainability, Environmental
Science andPolicy Program, Michigan State University, 480
Wilson Road, 151 Natural Resources Building, EastLansing,
MI48824, USA
8 College ofArchitecture, Texas A&M University, 3137
TAMU Langford Bldg. A, Room 122, CollegeStation,
TX77843-3137, USA

Environment Systems and Decisions
1 3
2004). According to the Harmonised Framework survey, 37
million people are food insecure in the Sahel, with (21%)
of them facing severe food insecurity and requiring emer-
gency assistance (ECHO 2016). Food insecurity in Mali is
therefore emblematic of that regional picture where up to
4.6 million Malians are food insecure (UN-OCHA 2012).
Since independence from France in 1960, food security in
Mali has been a pressing issue for successive ruling govern-
ments. In spite of repeated political promises by multiple
governments, food security remains a critical issue. Food
insecurity is exacerbated by poverty, biophysical conditions,
and the political environment and conflict, specifically in
Northern Mali. Furthermore, erratic production of primar-
ily rain-fed agriculture owing to rainfall variability is one of
the major drivers of food insecurity in the region. Increase
in climate variability, higher temperatures, and incidences
of droughts are likely to exacerbate the vulnerability of rural
subsistence farming households in the region (Roudier etal.
2011).
Most studies associate food insecurity with structural
causes. It is believed that political commitment, effective
institutions, good policies, a comprehensive approach in
governance, and adequate levels of investment can contrib-
ute to winning the fight against hunger and poverty (FAO
2013). The structural response to food insecurity in Mali
consisted of policy reforms ostensibly focused on enabling
increased agriculture production. The Loi d’Orientation
Agricole (LOA), a comprehensive agricultural development
framework enacted in 2006, regulates agricultural produc-
tion in the country to promote sustainable and competitive
agriculture. The LOA advocates for land tenure reform that
enables long-term investment in agriculture lands. It sup-
ports farmers’ access to inputs (fertilizer), equipment, and
government subsidies, with an emphasis on farmer’s organi-
zations and collectives as a central element in the agriculture
development process.
Alongside the aggregate policy reform, specific responses
to food insecurity include: food assistance consisting of food
or cash transfers; emergency school feeding and nutrition
programs in the conflict-prone Northern part of the country;
and donors’ support to households with basic agriculture
inputs for kitchen garden as well as children’s treatment of
malnutrition.
While these specific responses to food insecurity are
known to contribute to enhancing food security, they remain
largely unsustainable. Furthermore, national policies to
achieve food security are full of promise for agriculture pro-
duction, but lack the specificity needed to address the com-
plex granular socioeconomic contextual reality, which goes
beyond the biophysical aspect of food production. There
is now ample evidence that neither trickle-down effects in
high-potential areas nor structural adjustment measures initi-
ated in many African countries have sufficiently generated
the intended results of increased farm incomes and sustain-
able food security (Kuyvenhoven 2004).
Overall, sustainable food security poses formidable chal-
lenges and requires comprehensive efforts and special meas-
ures. Studies have shown that to understand food insecurity
and identify steps for effective intervention, there is a need
to apprehend the food systems and food insecurity/security
in a holistic way which goes beyond production alone (Staatz
1994). Most farmers in West Africa rely on subsistence lev-
els of production and are unable to invest in agricultural
technology and machineries to boost their yields (Brown
el al. 2009). Food security is the outcome of food system
processes all along the food chain, from the producers/farm-
ers who cultivate food, to global/national/local markets who
distribute food, to consumers who buy, utilize, and consume
food. Understanding the behavioral aspects of food security
is critical in the African context where agriculture, while
oriented toward basic subsistence, remains embedded in
social system including the social dynamics of households,
extended families and communities (Richards 1990). In
light of these considerations, this study focuses on assess-
ing the different aspects of food security qualitatively at the
household level in Mali. This exploratory work focuses on
developing a nuanced understanding of food security and
adaptive behaviors to current challenges to food security
at the household level with a distinct focus on inter- and
intra-family behavioral dynamics in rural, southern Mali (the
Koutiala district). Methodologically, we make use of mental
modeling techniques (Wood etal. 2012; von Winterfeldt and
Edwards 2007) to guide our research, and theoretically we
build on previous work that examines food security, climate
change, and farming practices in this region of the world.
While the Malian context is unique, it does provide potential
insight into broader food security issues found across the
Sahel region.
2 Theoretical background
2.1 Food security
Food security is often defined in terms of four intercon-
nected parts that operate at multiple scales: food availability
(whether enough food is produced for a given population);
food access (whether a population is able to purchase or
obtain adequate food); food utilization (whether the food is
prepared and consumed in optimal ways to ensure safety and
proper nutrition); and food stability (whether a population
is able to obtain food over time) (Barrett 2010; FAO 2009).
A fifth dimension of food security that informs the other
parts is the concept of culturally appropriate foods, which
is tightly linked to the idea of food sovereignty (Menezes
2001). Beyond proper food availability, access, utilization,

Environment Systems and Decisions
1 3
and stability, there is a need for food systems to provide
access to food that is culturally appropriate. Despite some
level of conceptual clarity, food security can still be difficult
to measure. Staatz etal (1990) stressed the need to develop
standards to measure food security for people in the Sahel
zone. In the years following the famines of the 1970s, gov-
ernments stressed the need to address food security at the
national level. However, these early studies of food secu-
rity in Southern Mali often failed to capture food stress at
household and individual scales. A gap in knowledge of
food security on small scales prevents policymakers from
fully understanding and addressing the issues households
face. Hussein (2002) advocated for multi-level collaboration
between stakeholders and policy makers and stressed the
need for policy makers to view food as a right, and to deepen
understanding of the “root causes” of food insecurity. How-
ever, as suggested by prior articles (Staatz etal 1990; Hus-
sein 2002), gathering detailed information on the locations
and identities of food-stressed people is very challenging.
Hanjra and Qureshi (2010) discussed food security
through a natural resources lens. The authors briefly dis-
cussed climate change as a stressor, as it is expected to
impact water resources and agriculture yield. Reduced
access to suitable water resources will be a barrier to pro-
viding enough food to feed an expanding human population.
Butt etal. (2005) used climate projections to develop impact
scenarios upon Malian crops. The authors found that in the
absence of actions to cope with the most extreme effects of
climate change, large swaths of the population will be at risk
of hunger and malnutrition. Some of the potential pathways
to adaptation include development and utilization of heat-
tolerant crops in addition to improving land quality.
More recently, Traore etal. (2013) explored climate vari-
ability in southern Mali over a period of four decades, from
1965–2005. During that time, a slight increase in minimum
daily temperature was observed, though little change in max-
imum temperatures was noted. Seasonal rainfall changed
from year-to-year; an increase in dry days during growing
season led to a shift in the annual distribution of precipita-
tion. Among the crops studied (cotton, sorghum, groundnut,
and maize), cotton crops were the most sensitive to the pre-
cipitation shifts observed during growing season. That said,
interruptions to regular precipitation events during rainy sea-
son, along with decreased soil quality, may negatively affect
agricultural activities in the region (Traore etal. 2013). A set
of scenarios used to model crop sensitivity to climate change
across West Africa found that regardless of rainfall amounts,
temperature increases equaling or exceeding 2°C will likely
have a strongly negative impact on crop yields. Given the
greater sensitivity of crops grown in the Sudanian region,
farmers in southern Mali may experience more dramatic
crop decreases as climate change advances. Some research
shows more modernized cultivars may help counteract the
negative effects of increased temperatures (Sultan etal.
2013). These studies on the effect of climate variability on
food security allude to the importance of climate change in
understanding food security in this part of the world.
2.2 Climate change
Large-scale climate shifts were first noticed in the 1970s
and 1980s in the form of increased persistence of droughts,
but the trend has continued, putting traditional lifeways at
risk. In recent years, precipitation shifts, fauna migration,
and land use changes across Burkina Faso help illustrate
the trends much of Western Africa is experiencing, as the
Sahel expands southward into the Sudanian zone (Wittig
etal. 2007). Sylla etal. (2016) evaluated three climate pro-
jections to explore potential changes across West Africa.
The models reflected an increase in dry conditions across the
region, in addition to increased temperatures. An increase
in aridity across the Sahel is expected to develop by the
end of the twenty-first century. Increased temperatures,
which drive aridity, are expected to place increased stress
on water resources, leading to a more challenging agricul-
tural context.
There is a limited set of research examining the climate
change perceptions of people living in the Sahel. Exist-
ing research has focused on farmers in Senegal. Tschakert
(2007) explored the impact of climate change on subsistence
farmers in Senegal. Surveys indicate climate adaptation is
hindered by health challenges, difficulty in acquiring jobs,
and lack of infrastructure in villages. Tschakert (2007) also
used focus groups and “conceptual mapping activities” to
develop models of Senegalese farmers’ understanding of the
drivers and risks of climate change. The models suggest a
poor understanding of the complexity of climate change.
Past studies on adaptation of Malian farmers to climate vari-
ability and droughts show that farmers have a high percep-
tion of these climate trends (Mertz etal. 2009) and have been
using various adaptation strategies to adapt to these changes.
Village grain banks, labor sharing, informal lending institu-
tions, etc., form a collective social capital which the farm-
ers draw from during times of need. Similarly, Mertz etal.
(2009) noted that farmers in the Sahel have always faced
the need to adapt to climate variability. Studies have shown
that the resilience of rural communities in Africa to climate
change and variability is influenced by social learning and
communications across multiple institutional scales, com-
munity reorganization, and adaptive capacities (Osbahr etal.
2008).
The increasing effects of climate change are expected to
make culturally significant activities (like farming or herd-
ing practices) more difficult. Exploring the cultural threats
posed by climate change is essential to ensure policies
to promote climate resilience are feasible and relevant to

Environment Systems and Decisions
1 3
vulnerable communities (Adger etal. 2013). Nielsen and
Reenberg (2010) found culture also impacts a community’s
ability to adapt to climate change. Several factors impact
a group’s ability to adapt, given differences in gender and
socioeconomic level. Within rural households, families
often adopt a variety of adaptation strategies. Nielsen and
Reenberg’s (2010) study focused on the factors that pre-
vented the development of climate resilience among the
Fulbe people, who are located in Northern Burkina Faso.
Though the results cannot be generalized beyond the Fulbe,
it highlights the need to understand the cultural factors that
enhance or inhibit climate resilience in sub-Saharan com-
munities (Nielsen and Reenberg 2010). A gender-focused
study conducted by Aberman etal. (2015) explored climate
vulnerability through the lens of gender. Given that men and
women have differing levels of access to resources, a per-
son’s ability to navigate climate shifts is affected by gender.
Malians were included in a cross-cultural study that explored
gender dynamics within groups involved in agricultural
activities that are threatened by climate change. Across the
four countries included in the study, authors found women’s
access to adaptive resources are often constrained by gender
norms, but climate vulnerability can be moderated through
community support and pooled resources. The impact of
climate change on food security is also closely tied to local
agricultural practices.
2.3 Malian agriculture
Mali experienced rapid agricultural growth throughout the
second half of the twentieth century, but in recent years, the
southern cotton-producing region has experienced lower cot-
ton yields, along with increased use of fertilizer. Soil analy-
ses were not able to confirm degradation, as the quality of
samples from cultivated fields and fallow fields were simi-
lar. Benjaminsen etal. (2010) suggested declines in yields
can be attributed to expansion into lower-quality tracts of
land. Laris etal. (2015) challenged this notion, through an
analysis of satellite data and interviews with farmers. Since
the early 2000s, Malian farmers intentionally switched from
cotton to maize. Farmers discussed a perceived need to focus
on maize over cotton, as a lack of fertilizer has forced a shift
in priorities.
Furthermore, dramatic shifts in precipitation are expected
to alter subsistence farmers’ abilities to provide food. Farm-
ers who are unable to produce and sell excess food are at an
increased economic risk during droughts, when their abili-
ties to provide for their households’ needs become stressed.
Low financial reserves prohibit farmers from purchasing
supplies during times of food stress (Brown and Funk 2008).
An analysis suggested traditional farming methods maxi-
mized outcomes for food crops (Jolly and Gadbois 1996).
This work complements Tappan and McGahuey’s (2007)
assessment of Malian communities that implemented soil
and water management practices with the support of a gov-
ernment agency. The analysis demonstrated villages that
received help experienced better outcomes. Though the
research is over a decade old, it may point to best practices.
Traore etal. (2013) analyzed the planting patterns of the
major crops in Mali. Over a period of three seasons, planting
dates and crop duration of cotton, millet, maize, and sor-
ghum were studied. A comparison showed some crops (mil-
let, maize, and sorghum) were more tolerant to late planting,
while others (cotton) produced higher yields when planted
early. Additionally, crops had differing levels of sensitivity
to weather patterns. These findings indicate a potential for
climate adaptation; adjusting planting patterns may assist
farmers’ efforts to adjust to shifting rainfall patterns while
maximizing agricultural output.
Beyond agricultural practices, cultural practices signifi-
cantly influence Malian agriculture. Becker (1990) con-
ducted a study of a Bambara village in the late 1980s. Across
the region, a chief, who determines how land is distributed
among the heads of household, leads villages. Each head
of household is then responsible for managing the land and
directs the tasks of younger brothers, sons, and wives. In the
mid-1990s, Wooten (2003) explored the gender dynamics of
agricultural production in a small Bambara community near
Bamako. Though women technically had access to income-
generating activities such as gardening, men often domi-
nated the higher income-generating tasks, such as cultivating
cash crops like cotton. Given structural restrictions within
the community, women are forced into labor-intensive and
low-earning activities, like producing charcoal. This imbal-
ance perpetuates women’s reliance upon men and prevents
women from acquiring resources to expand their income-
generating activities.
The research presented in this paper addresses the need
to understand food security at household and individual
levels. Small-scale knowledge is viewed as an essential
part of developing sound policy (Hussein 2002; de Haen
etal. 2011). This work builds upon existing research con-
cerning the climate-related stresses Sahelian farmers face,
but expands the knowledge through modeling how farmers
respond to a variety of climate-related threats. Furthermore,
differences in gender and age were also explored, adding
depth to the understanding of how culture impacts climate
adaptation strategies.
3 Methodology
3.1 Mental models methodology
The research presented in this article adheres to the appli-
cation of mental model theory and tools using influence

Environment Systems and Decisions
1 3
diagrams to represent key variables and their relation to key
actors’ decision behavior in a complex decision environ-
ment, food security in Mali for subsistence farmers (Wood
etal. 2012; von Winterfeldt and Edwards 2007). Men-
tal models are “internal conceptual representations of an
external system” (Wood etal. 2012) that allow people to
know, perceive, make decisions, and construct behavior in
a variety of environments (Gentner 1983). They go beyond
“schema” or “images,” however, by providing constructs
that are “manipulatable,” enabling people to predict sys-
tem states via mental manipulation of model parameters
(Cannon-Bowers etal. 1993). The theoretical foundations
of mental models are found in the cognitive sciences and
the examination of how humans make sense of their envi-
ronment (Craik 1943). Over the past two decades, there has
been growing use of such models in the field of risk commu-
nication to develop a greater understanding of complex sys-
tems that feature multiple stakeholders, particularly systems
that include interactions between individuals and potentially
risky hazards (Zaksek and Arvai 2004; Allen etal 2002). In
this study, the stakeholders are subsistence Malian farmers
and the complex decision environment in question is house-
hold food security, which is intimately linked to agriculture
and at risk from a number of contextual variables, such as
climate change and variability.
The primary source of data for mental model methodolo-
gies is in-depth, semistructured interviews of individuals
that have either experienced or could potentially experi-
ence the phenomenon being examined (Wood etal 2012).
It is suggested in the literature that this style of interviews
reveals more about the major beliefs, values and percep-
tions concerning an issue in a given population compared to
more structured, close-ended types of interviews (Morgan
etal 2002). Interviews are subsequently coded for recur-
ring themes and concepts related to the phenomenon being
examined. Themes and concepts are then used to develop
an influence diagram that represents the collective mental
model of the behavior in question.
3.2 Study area
The interviews for this study were collected in the Koutiala
district in Mali, an administrative subdivision of the Sikasso
Region of southern Mali with a population of about 575,000
people (2009 census) in an area of approximately 8730 km2.
Most farmers in Southern Mali belong to the Bambara tribal
group, who are mostly subsistence farmers within a rain-fed
agro-pastoral system. Cotton is the main cash crop and is
grown in rotation with sorghum, pearl millet, maize, ground-
nuts, and cowpeas during the wet season (Traore etal 2013).
In Mali, which is a polygynous society, large households,
with an average family size of 60–90 household members,
are common. The household consists of the entire family
with the head of the household, his wives, and their chil-
dren. In turn, the children have their own subfamily units
of wives and children. Most household structures in Mali
are composed of subfamily units (husband, wives and chil-
dren) living in coexistence and sharing family resources. The
farming system in Southern Mali consists of mainly cere-
als, especially millet, sorghum, and maize, which provide
approximately 70 percent of total calorific consumption of
Malians (Bingen etal 2000). Household food production
among the Bambara is a dynamic process where individual
household members farm in the collective plot and have the
right to consume from the common granary (Adams 1993).
The collective plot is owned by the head of the household
and the patriarch. Select individual men (sons of household
head) and women (wives of household head) may have user
rights to individual plots to produce additional grains and
vegetables to provide supplemental income. Active members
of the household provide labor to collective plots to cultivate
cash and cereal crops which is either sold or stored in the
granary for future consumption by the family. The head of
household, usually the elder male in the family, is the pri-
mary decision maker around food production and allocation
and controls the land use and food allocation rights. The
head/chief of the household also controls the production and
allocation of household labor to produce subsistence con-
sumption goods. Allocation of food is usually done by the
head of the household who decides the amount of grains to
be given to the women to prepare food. Elder women decide
the menu for the meals and the allocation of labor among
themselves to prepare the food. Young women in the house-
hold are responsible for food preparation, fetching water for
household use, collecting shea nuts to make shea butter and
firewood in the field, and forming an integral part in farm
labor (specially for weeding and harvesting). The younger
men and women in the household are also responsible for
the labor in the collective and individual plots of land, while
young boys are responsible for herding cattle.
We selected six villages for our study based on their dis-
tance from the Koutiala city center (Table1). We did not
want to go outside of a 50-km radius of the city center to
Table 1 Study villages in Koutiala district
Village name Distance from Koutiala Number of
interviews
Sincina 3km 2
Wolobougou 5km 6
Kaniko 14km 5
Nampossela 15km 7
Molobala 34km 5
M’pessoba 45km 6
Konsseguela 50km 6

Environment Systems and Decisions
1 3
ensure similarity between villages’ agricultural practices.
Forty-one interviews were conducted between July 23 and
30, 2015, with 20 male (12 elder and 10 younger) and 19
female (10 elder and 9 younger) respondents who were
recruited based on their gender and decision-making roles
within the household. The respondents were placed in four
categories: (1) Elder men, the primary decision makers
within the household, and the owners of the plots within
the household; (2) Elder women, the wives of the elder men
and managers of individual plots within the household; (3)
Younger men, the secondary decision makers of the house-
hold and managers/laborers of the collective and individual
plots; (4) Younger women, the primary household workers,
they help elder men and women and young men within the
household with the household activities, and they tradition-
ally have limited say in agricultural decisions. The study
sample was determined through purposive snowballing
technique (Biernacki and Waldorf 1981) where the initial
contact in a village was made through the chief of the vil-
lage who then referred to eligible respondents from other
households in the village who fit our criteria. Interviews
were conducted through this referral process until the point
of data saturation was reached, where no new information or
themes was obtained from the interviews. The respondents
were interviewed individually with the average length of the
interviews being approximately 60–75min.
Respondents were asked a set of broad open-ended ques-
tions such as: “What is your household day to day experi-
ence with food?”; How does this experience change dur-
ing the seasons?”; “What challenges around food does
your household face today as compared to when you were
younger?”; and “What is your perspective on agriculture?”.
Respondents were also asked questions about their specific
roles and responsibilities while farming within their house-
holds and the kind of decisions they take on a daily/seasonal
basis. This open-ended question design adheres to principles
suggested by Morgan etal. (2002). The open-ended nature
of the interviews also allowed respondents to talk about their
perceptions of other household member’s role in terms of
food security. The goal of these broad interview narratives
was to develop insight into the individual’s cognition and
perceptions of agriculture, climate change and food secu-
rity. Additionally, the interviews focus on individual and
household-level decision-making in terms of adaptation to
climate risks and food insecurity. The interview narratives
are used to develop two influence diagrams of the farmers’
mental models of farming and household food security.
3.3 Interview analysis
All interviews were recorded, transcribed, and then trans-
lated from Bambara to French and English by a transcriber/
translator who was a part of the study team’s interview pro-
cess in the field. This ensured that the transcription pro-
cess was completed with respect to the study context. The
transcriptions were crosschecked and reviewed by other
team members to ensure consistency and accuracy in the
translations.
The research team used Nvivo content analysis software
to develop a coding rubric that consisted of 23 concepts
that emerged from the interview narratives. The specific
coding rubric for “External Challenges” developed from
the narratives and their definitions is outlined in Table2.
A team of three coders coded the interviews separately
based on the given rubrics, and the interviews were recoded
twice to ensure intercoder reliability. Table3 indicates how
many interviews contained each of the challenges by coder
(Sources), and how often each challenge was coded across
the interviews by each coder (References). For example,
Coder #1 coded for “Food Related Challenges” in 26 of
Table 2 Definition of external
challenges to food security that
were used by the interview
coders
External challenge Definition
Food related Regarding any changes in food consumption and distribution based on time/season
Technological Developments in seed varieties, equipment, pesticide use, etc
Environmental How soil degradation, lack of rainfall, etc., affects food production
Sociocultural Challenges in way people think, act, and relate to others within the household, and
in relation to society
Table 3 Sources indicates how
many interviews contained the
code for each coder. References
indicates how many times each
code was identified by a coder
Sources References
Coder #1 Coder #2 Coder #3 Avg. Coder #1 Coder #2 Coder #3 Avg.
Food-related challenges 26 20 21 22.3 35 31 25 30.3
Technological challenges 27 26 28 27 36 35 52 41
Environmental challenges 32 27 30 29.7 41 47 41 43
Sociocultural challenges 21 16 15 17.3 36 28 22 28.7

Environment Systems and Decisions
1 3
the interviews, and across those 26 interviews Coder #1
coded 35 phrases or sentences as referring to “Food Related
Challenges.”
3.4 Model development
We collectively developed two influence diagrams using an
iterative process once all coding was completed representing
participant’s collective mental models of food security. The
influence diagram is a holistic collection of all concepts,
decisions, and contexts that were described by the interview
respondents and as such is a comprehensive qualitative over-
view of the behavioral aspects around food security of all
respondents (Whitley etal. 2017; Zaksek and Arvai 2004;
Morgan etal. 2002). There was not a threshold measure for
the inclusion of a concept,
Consider the following quote from an elder woman, who
described her day-to-day experience with food:
“Every day the head of the family gives a gourd of
cereal to the cook. She divides the cereal and gives
each half to both women who do not cook in the
daytime to crush/pound. She divides then the flour
between the breakfast, the lunch and the dinner.”
Based upon reviewing this and other interview transcripts,
it became clear that elder women manage the daily dietary
needs of their respective households. In Fig.5, this respon-
sibility was noted in a node labeled “food consumption.”
The node was connected to three sub-nodes meant to illus-
trate the tasks that comprised “food consumption,” and
were labeled “menu,” “preparation,” and “distribution.”
The above quote most closely follows the tasks related to
“preparation” and “distribution.” In addition to discussing
the work they perform with regard to food consumption,
other elder women described their roles related to household
food production and income generation.
The initial diagram was developed by the two of the
coders in consultation with one of the project PIs using the
LucidChart1 program. This first model was referred to as the
“Status Quo without challenges” model and represents the
participants’ collective mental model of food security under
status quo conditions in the absence of external pressures
(Fig.1). The first model focused on the individual level tasks
that contributed to the four facets of food security (avail-
ability, utilization, access, and stability) at the household
level. The model was split into four sections (elder men,
elder women, young men, and young women). For each of
these sections, the realm of decision-making for which a
given class of individuals is responsible for was then linked
to specific tasks that were linked to one of the four facets of
food security at the household level. For example, elder men
are responsible for decisions in the realm income genera-
tion for the extended family. A potential way for generating
cash is to grow cash crops, which contributes to greater food
access and stability at the household level.
The “Status Quo without challenges” model condi-
tions were contrasted with the present reality for interview
respondents in the second influence diagram (Fig.2). The
second diagram is referred to as the “Status Quo with chal-
lenges” model and represented participants’ collective men-
tal model for food security conditions with the perceived
presence of external challenges. This model focused on the
same individual level behavioral actions as the first model;
however, they were filtered through external challenges
identified during the coding process (environmental, tech-
nological, food related, and sociocultural) (See Table2). In
cases where adaptations were difficult to achieve or do not
currently exist, the challenge represented an insurmountable
barrier to achieving food security. The behavioral actions
that terminated in the external challenges were still pos-
sible, but participants indicated they have become signifi-
cantly more difficult or time consuming. However, a lim-
ited number of adaptations that had the potential to lead to
food security were identified, such as the practice of sell-
ing cattle to pay for food that is in the realm of elder male
decision-making.
4 Results
The collected mental models of food security for the house-
hold are represented in the influence diagrams of Figs.1
and 2, under status quo conditions, and the presence of sig-
nificant external challenges presents a complex system. For
clarity, we also present two individual sub-models for each
category of family member under status quo conditions and
under the presence of significant external challenges. The
following section of the paper presents a short description of
each model and sub-model (excluding young women, whom
are covered in greater detail in Sect.5) (Sect.4.1) and an
illustration of the external challenges to family food security
using direct quotes from the interviews (Sect.4.2).
4.1 Model andsub‑model descriptions
4.1.1 Model descriptions
Status Quo without challenges influence diagram (Fig.1):
This model features each category of family member and
their primary responsibilities in terms of food security dur-
ing normal conditions. Each sub-node is linked to an aspect
1 Lucidchart is an online model creation software package that allows
multiple individuals to work on the same model collectively or indi-
vidually.

Environment Systems and Decisions
1 3
Fig. 1 “Status Quo without
challenges” influence diagram
represents the participants’
collective mental model of food
security under optimal condi-
tions

Environment Systems and Decisions
1 3
Fig. 2 “Status Quo with challenges” influence diagram represents the participants’ collective mental model of food security with the presence of external challenges

Environment Systems and Decisions
1 3
of food security (availability, utilization, access, or stability)
model illustrates how redundant actions on the part of mul-
tiple family members lead to a resilient system. This notion
is further examined in Sect.5 of the paper.
Status Quo with challenges influence diagram (Fig.2):
This model introduces a set of external challenges identi-
fied by participants that significantly hamper their normal
responsibilities in terms of food security. This model illus-
trates how these challenges significantly reduce the redun-
dancy and subsequent resiliency of food security for a fam-
ily. This assertion is further explored in Sect.5 of the paper.
4.1.2 Sub‑model descriptions
Elder Men, Status Quo without challenges influence dia-
gram (Fig.3): This model shows the set of responsibilities
that elder men control the decisions for in terms of food
security under normal conditions. Elder men are responsible
for decisions concerning crop production, food allocation,
and income generation. It is important to note that many of
the decisions that they make directly influence other fam-
ily members. For example, they make decisions about what
crops will be grown, how much and what type of inputs
(fertilizer) will be used, and what equipment will be utilized;
however, young men and women provide the labor to imple-
ment these decisions.
Elder Men, Status Quo with challenges influence diagram
(Fig.4): This diagram shows how external challenges influ-
ence elder men food security responsibilities. Of notice,
is how responsibilities related to income generation all
terminate in food-related or sociocultural challenges. Cash
crop production (primarily cotton) is often halted to focus
on crops to feed the family, livestock is sold to purchase
food, and it can be difficult to obtain credit from local banks.
These decisions lead to reduced redundancy for activities
that secure food access and stability. However, this group
can overcome many challenges that it faces, often by shift-
ing greater responsibility to other family members for whom
they make decisions. Examples of this process are illustrated
in Sect.5.
Elder Women, Status Quo without challenges influence
diagram (Fig.5): This model shows the set of responsibili-
ties that elder women control the decisions for in terms of
food security under normal conditions. Elder women are
responsible for decisions concerning food consumption,
food production, and income generation. Similar to elder
men, many of the decisions that elder women make directly
affect other family members. For example, elder women are
responsible for most decisions concerning food consumption
(setting menus, planning the preparation and distribution
of food to other family members); however, young women
provide most of the labor for these tasks and are primarily
responsible for fetching water and pounding millet, two key
aspects of food consumption in this system.
Elder Women, Status Quo with challenges influence dia-
gram (Fig.6): This diagram shows how external challenges
influence elder women food security responsibilities. Of note
is how income-generating activities are ceased when faced
with food challenges. The production of shea butter and cash
crops is often halted to focus on more core activities, such
Fig. 3 “Status Quo without challenges” influence diagram with a focus on elder men collective mental model of food security

Environment Systems and Decisions
1 3
as the production of supplements (sauces and additives to
make food more appealing and nutritious for children). The
cessation of these activities reduces the redundancy of tasks
that contribute to food availability and access.
Young Men, Status Quo without challenges influence dia-
gram (Fig.7): This model shows the set of responsibilities
that young men control the decisions for in terms of food
security under normal conditions. Young men are primar-
ily responsible for food production via field labor, livestock
rearing, and gardening. Gardening refers to the cultivation
of a small plot of land sometimes giving by the head of the
household to a young male to grow food for his immedi-
ate family. Young women often provide additional labor for
garden plots.
Fig. 4 “Status Quo with challenges” influence diagram with a focus on elder men collective mental model of food security
Fig. 5 “Status Quo without challenges” influence diagram with a focus on elder women collective mental model of food security

Environment Systems and Decisions
1 3
Young Men, Status Quo with challenges influence
diagram (Fig.8): This diagram shows how external chal-
lenges influence young men food security responsibilities.
Field labor is key for young men. Poor soil or sporadic rain
can pose significant environmental challenges to working
their own land; however, they are still able to work on
other farms or to migrate to urban areas to provide remit-
tances for food to the family, despite the sociocultural
challenge of elder men and women preferring that young
men focuses their labor on the family’s land.
Fig. 6 “Status Quo with challenges” influence diagram with a focus on elder women collective mental model of food security
Fig. 7 “Status Quo without challenges” influence diagram with a focus on young men collective mental model of food security

Environment Systems and Decisions
1 3
4.2 Inuence diagrams andthreats tofood security
The models developed by this work illustrate the acute
threat of food insecurity to the well-being of Malian farm-
ers. Specifically, the challenges identified in Fig.2 at the
group level and Figs.4, 6, 8, and 10 at the individual level
(environmental, sociocultural, food related, and technolog-
ical) have exacerbated the risk of food insecurity for par-
ticipants in our study. Our work suggests that these chal-
lenges loom large in the mental models of our participants
as demonstrated by the influence diagrams (Figs.1–10)
and the quotes found below. These challenges to food secu-
rity have largely been examined in previous work; particu-
larly food-related (Hatløy etal. 2000; Torheim etal. 2004),
technological (Tappan and McGahuey 2007; Brown and
Funk 2008), and environmental challenges (Wittig etal.
2007; Sylla etal. 2016) are well covered in the extant
literature.
Food-related challenges often occur seasonally. At
present, households are unable to produce sufficient food
to cover an entire year. The following quotes summarize
these challenges:
“The food shortage is our big problem. Our produce
can feed the family just seven month in the year” –
Elder female (This challenge is reflected in Fig.6,
as the food production node and related sub-nodes
encounter a food-related challenge, specifically a
lack of food during part of the year. To address these
challenges requires greater labor in the field or small
gardens on the parts of young men and women in
the family)
“It is the periods of food shortage which affect us a lot.
Last year we harvested the corn to consume it imme-
diately. The yield is low to meet our entire needs. We
buy cereal subsidized by the PAM [UN world food
program] but the quantity subsidized also is little to
cover the request of the population.” – Young male
(see Fig.8)
In response to these stresses, households have been forced
to find ways to generate additional income to provide food.
Several interview participants described the need to sell live-
stock to mitigate food loss.
“When we left at Ziéla head of the family sold its
cattle and its equipment, we encountered equipment
problems and work but because we were small and
we hardly received help.” – Young female (See Fig.4,
the young women that gave this quote is referring to
the actions of an elder male to face food-related chal-
lenges)
One young man cited “Insufficient equipment. [And]
insufficient oxen for plowing” as the primary barriers to
cultivating sufficient food. The lack of access to technology
increases the burdens placed upon household members:
“Weeding [by hand] is painful.” – Young female
Both of these challenges are reflected in Fig.4, since
elder men control access to farm equipment and the selling
Fig. 8 “Status Quo with challenges” influence diagram with a focus on young men collective mental model of food security

Environment Systems and Decisions
1 3
of livestock. As alluded to in Sect.4.1, these examples
illustrate how a challenge in the decision-making realm of
one category of the family can have sever impacts for other
members of the family that do not have control of decisions
related to this challenge.
In response, some older household members explain chal-
lenges through the lens of belief, as described below:
“Rain Failure and bad distribution in time: we organize
prayer sessions for rain, it rains often and often nothing
happens.” – Elder male
Many participants discussed the role of environmental chal-
lenges. The lack of reliable rainfall was a widely discussed
concern:
“Last year rainfall was insufficient, the yield has
decreased. Wells in which we collect water quickly
dried up during the dry season we had a serious prob-
lem of water for domestic use and for gardening. We
need water to sustain the family we go collecting water
to the pump to 3km from home, it puts us behind
the entire program of the day.” – Elder female (See
Fig.10, the elder woman who provide this quote is
referring to actions that are the responsibility of young
women to face environmental-related challenge. In
order to address this challenge, young women need
to exert a significant amount of labor and time and are
unable to delegate it to other members of the family.
This is why the dashed arrow ends in the challenge.)
“Yes, rainfall is a challenge. This year for example, we
started sowing late. Because the rain was delayed. It
started raining on the 10th of July, so insufficient rain-
fall is a big problem.” –Young male (This challenge
is reflected in Fig.8, where field labor is terminated
into the environmental challenge node since labor can-
not address the challenge of inadequate or sporadic
rainfall)
“We rely on rainfall. When there is no rain, we cannot
proceed with the sowing. Fields are also invaded by
weeds.” –Young female
This is in contrast to past years, when rainfall occurred on a
more predicable scale. One elder female discussed the ease
of cultivation in years past:
“There were no major problems, soils were fertile and
it rained heavily. We had beef and tillage equipment.
We did all the work by hand and yields were good.”
Other environmental concerns, namely soil degradation,
were also discussed:
“Land degradation: it is necessary for us to use the
manure every two years to have a good yield, our
labor is mainly women and do not like collecting the
manure.” –Elder female
This quote illustrates how the need for increased fertilizer,
which is in the decision-making realm of elder males (See
Fig.3), is addressed via the labor of another category of
family member, young women. This is how elder males are
able to overcome the environmental and technological chal-
lenges posed by poor soil quality.
However, there is less coverage of the effects of sociocul-
tural changes on food security in the literature. For example,
our interviews suggest a paradoxical perception of educa-
tion. On the one hand, education is highly valued, and on
the other, there is concern that education makes children
less suitable agricultural workers. This is illustrated in the
following quote from an interview with an elder male:
“Children do not like to cultivate. Schooling has
spoiled them, they do not study well and they do not
work in the field.”
There is a similar worry about the spread of liberal ideals
among young people negatively affecting their contributions
to household-level food security.
“With more democracy we can[not] educate our chil-
dren as we have been by our parents or you risk going
to jail. In our time at 4 am, our old awoke and spent
water on the face and went directly to the field. Never
did the sun rose find us at home and we never were
back before sunset. We left at dawn and go back at
dusk. The person who refused was beaten very hard.”
–Elder male
While these observations can be partially attributed to tra-
ditional inter-generational conflict (i.e., young people are
often perceived as lazy by older generations across cultures),
they also represent anxiety about the breakdown of tradi-
tional social structures. This anxiety is also expressed by the
younger generations:
“Today the heads of families ensure [adequate food]
for a period of the year, its every man for himself for
the rest [of the year, this includes] the problems of
health, clothing, wedding second wife. We often diso-
bey when the leaders do not provide [for] our needs.”
–Young Male
“The family heads before was supporting all the prob-
lems of the family; everyone obeyed them by what they
ensured the well-being of everyone. All properties of
the family were collective and under management of
the family head. Nobody could make personal sav-

Environment Systems and Decisions
1 3
ings. Nobody disobeyed, myself I was chastised by
my grandfather when I refused to get into the granary
to bring out the millet.”
–Young Male
5 Discussion
The models developed by this research project have pro-
vided an in-depth perspective on food security at the
household level in rural Mali. Furthermore, this work
has highlighted the precarious nature of young women in
Fig. 9 “Status Quo without challenges” influence diagram with a focus on young women collective mental model of food security
Fig. 10 “Status Quo with challenges” influence diagram with a focus on young women collective mental model of food security

Environment Systems and Decisions
1 3
relation to household food security in these large families.
Essentially, young women provide a great deal of labor
to establishing a food secure household, but are also the
most vulnerable to external challenges to household food
security. In this section of the paper, we examine this para-
doxical situation.
5.1 Young women andfood security
Current literature has not thoroughly documented the cli-
mate adaptation efforts of young women in Mali. Figures9
and 10 provide a more detailed look at how young Malian
women navigate food provisioning, along with the barriers
they face in their efforts to feed their households. Our work
suggests that young women are key to food security in this
system.
In the “Status Quo without challenges” model, young
women are responsible for food production, household
needs, and income generation (Fig.9). Each of these respon-
sibilities is further divided into smaller tasks. Food produc-
tion consists of pounding millet and engaging in field labor.
Collecting firewood and water is the main task associated
with household needs. It is important to note that each of
these four tasks are directly influenced by decisions made by
other members of the household. Additionally, some young
women generate additional income through commerce and
off-farm jobs. Several of these tasks benefit the household
through supporting one of the pillars of food security (Bar-
rett 2010; FAO 2009). However, in times of stress, the afore-
mentioned tasks often encounter challenges (environmen-
tal, technological, food related, and sociocultural) (Fig.10).
Interviews with young women indicated these challenges
were often difficult to navigate by conventional means or
without exerting significantly more labor. The influence
diagram in Fig.10 reflects these barriers through dashed
links that start at a task and end at a barrier. Furthermore, as
indicated in the discussion of the sub-models in Sect.4.1.2,
young women are often responsible for the completion of
tasks for which other members of the family control the rel-
evant decision-making process. In short, even under status
quo conditions, young women carry a heavy burden for the
maintenance of day-to-day household food security. This is
in addition to their responsibility as the primary caregiver for
children. The paradoxical position faced by young women
can be contextualized by examining how food security oper-
ates across the larger family structure.
5.2 Food security acrossthefamily structure
As suggested in earlier sections of the paper, household
food security is very much a collective task, and decisions
made by one individual has significant ramifications for
other household members. Unlike young women, other
segments of the household (see Figs.3–10) are often able
to utilize alternatives in the decision-making processes to
adapt to food stress. When faced with environmental chal-
lenges, such as sporadic rainfall, elder men are able to adapt
to these pressures by changing the type of crops the house-
hold will cultivate (Fig.4). In turn, a change in crop type
places food-related stress upon elder women, who are tasked
with preparing daily menus. However, many of the elder
women discussed their ability to navigate these challenges
by directing the younger women in the household to prepare
meals that utilize the food that is available (Fig.6). Young
men reported having to find work away from their house-
hold’s farm during times of food-related or environmental
stress (Fig.8). Due to decreased decision-making authority,
adaptation options are not readily available to young women
(Fig.10). Environmental stresses, like deforestation, have
forced young women to spend more time foraging for fire-
wood. Rather than delegating firewood collection to other
members of the household, young women must respond to
this stress by spending limited funds on firewood or reducing
the use of firewood for everyday tasks. As a result, their abil-
ities to contribute to daily household functions decreases.
Furthermore, unlike young men, young women face socio-
cultural barriers to migrating, such as safety concerns and a
lack of off-farm employment opportunities (Fig.10). In any
household, the decisions of one member will affect the deci-
sions and activities of other members, and this is definitely
the case for the participants in our study. However, our work
suggests that in term of food security young women’s labor
and time are disproportionately affected by the decisions of
other household members.
5.3 Resiliency offood security systems
Our work suggests that these challenges to food security
across the family structure, particularly for young women,
create risk by significantly reducing the resiliency of food
security systems. The concept of resilience originated in
ecology as the degree to which a system can continue to
function after exposure to stress (Miller etal. 2010). Another
perspective on resilience as defined by Holling (1973) is
“the persistence of relationships within a system and meas-
ure of the ability of these systems to absorb change of state
variables, driving variables and parameters, and still persist.”
In both approaches, resilience is the capacity of a system
to absorb disturbance and retain essentially the same func-
tion, structure, identity and feedbacks (Walker etal. 2004;
Alliance 2009). Figure2 represents a system that lacks
resiliency largely by a severe reduction in redundancy. The
external challenges reduce the set of successful behaviors
that farmers can make use of to achieve household food
security, exacerbating an already precarious position. Fur-
thermore, as suggested in earlier in this section, the majority

Environment Systems and Decisions
1 3
of adaptation options are afforded to elder men and women,
which increases the vulnerability of younger members of
society. However, it should be noted the traditional practices
are not always the best practices, as modernity has benefits.
This is reflected in the precarious position that young women
find themselves in relation to food security. The patriarchal
nature of traditional Malian society has limited the decision-
making authority of young women and subsequently the set
of related tasks that they control. Figure10 illustrates the
lack of adaption options afforded to young women in com-
parison with other household members.
This observation resonates with previous studies (Aber-
man etal. 2015; Wooten 2003), young women (and by
extension children, which are largely under their care) are
more vulnerable to the challenges of food security than other
demographic groups. This suggests the need for future pol-
icy programs to specifically focus on the needs of young
women and to expand their access to resources in order to
increase the range of adaptation behaviors available to them.
Additionally, it is important to consider the need to maintain
culturally appropriate foods when addressing the resiliency
of food security systems. While the food choices in the food
systems we examined, even under external pressure, remain
culturally appropriate, global trends in food delivery systems
challenge the cultural appropriateness and biodiversity of
food selections (Johns and Sthapit 2004). These develop-
ments also have the potential reduce the overall resiliency
of food production systems.
6 Conclusion
Efforts to broaden adaptation policies can benefit immensely
from the integration of nuanced data that indicates how
farmers—differentiated across multiple axes of social,
demographic, political, and ecological factors—make deci-
sions on how to adapt to social–ecological change and
uncertainty. Factors which are interlinked to each other in
determining the adaptive capacity of households (such as
resource entitlements, institutions, gender, socioeconomic,
behavioral and cultural) require an approach which embraces
the complexity of embedded systems. It requires looking
at biophysical risks and exposure, sensitivity, and adaptive
capacities (sociological, ecological, behavioral and insti-
tutional) not in silos but as an integrated complex system
where there is a scope for a range of adaptation and resilient
outcomes to emerge.
Beyond the policy level, adaptive behavior to climate
risks requires decision-making in an environment, which
is new and unfamiliar (de Suarez etal. 2012). The adaptive
process represents not only a change in the behavior but also
a change in the sociocognitive risk perception, mental mod-
els, and appraisal of the capacity to adapt (Grothmann and
Patt 2005). In addition, farmer adaptation behaviors depend
heavily on changes in the sociocognitive risk perception and
behavior, their mental models of risk and impact as well as
their appraisal of the capacity to adapt (Grothmann and Patt
2005). These factors are in turn affected by farmers’ interac-
tions and social learning among social groups in conjunction
with changes in biophysical and climate conditions (Adger
2010; Pelling and High 2005).
Various integrated assessment modeling of climate
change has been conducted at the global, regional, and
national levels to analyze the consequences of mitigation
and adaptation policies and behavior (Fisher-Vanden etal.
2013), but they do not capture the nuances of human adap-
tive responses to climate change at the local level. The com-
plex interaction of natural and social systems coupled with
uncertainty in human responses to climate change warrants
further explorations in the ways mental modeling can help us
understand climate change adaptation. Our work represents
an early, albeit limited, step in developing this more nuanced
conception of food security.
Compliance with ethical standards
Ethical approval All procedures performed in studies involving human
participants were in accordance with the ethical standards of the insti-
tutional and/or national research committee and with the 1964 Dec-
laration of Helsinki and its later amendments or comparable ethical
standards.
References
Aberman N, Ali S, Behrman JA, Bryan E, Davis P, Donnelly A, Gath-
aara V, Kone D, Nganga T, Ngugi J, Okoba B, Roncoli C (2015)
Climate change adaptation assets and group-based approaches:
gendered perceptions from Bangladesh, Ethiopia, Mali, and
Kenya. http://ebrary.ifpri.org/cdm/ref/collection/p15738coll2/
id/128950. Accessed 19 April 2017
Adams A (1993) Food insecurity in Mali: exploring the role of the
moral economy. IDS Bull 24(4):41–51
Adger WN (2010) Climate change, human well-being and insecurity.
New Political Econ 15(2):275–292
Adger WN, Barnett J, Brown K, Marshall N, O’Brien K (2013) Cul-
tural dimensions of climate change impacts and adaptation. Nat
Clim Change 3:112–117
Alliance R (2009). Adaptive capacity
Allen CD, Savage M, Falk DA, Suckling KF, Swetnam TW, Schulke T,
Stacey PB, Morgan P, Hoffman M, Klingel JT (2002) Ecological
restoration of southwestern ponderosa pine ecosystems: a broad
perspective. Ecol Appl 12:1418–1433
Barrett CB (2010) Measuring food insecurity. Science 327:825–828.
https://doi.org/10.1126/science.1182768
Becker LC (1990) The collapse of the family farm in West Africa?
Evidence from Mali. Geogr J 156:313–322
Benjaminsen TA, Aune JB, Sidibé D (2010) A critical political ecology
of cotton and soil fertility in Mali. Geoforum 41:647–656
Biernacki P, Waldorf D (1981) Snowball sampling: problems and tech-
niques of chain referral sampling. Soc Methods Res 10:141–163

Environment Systems and Decisions
1 3
Bingen RJ, Robinson D, Staatz JM (2000) Democracy and develop-
ment in Mali. Michigan State University Press, East Lansing.
https://doi.org/10.14321/j.ctt15hvwv2
Brooks N (2004) Drought in the African Sahel: long term perspec-
tives and future prospects. Tyndall Centre for Climate Change
Research, Norwich, Working Paper, No. 61, p. 31
Brown ME, Funk CC (2008) Food security under climate change.
Science 319:580–581
Butt TA, McCarl BA, Angerer J, Dyke PT, Stuth JW (2005) The
economic and food security implications of climate change in
Mali. Clim Change 68:355–378
Cannon-Bowers JA, Salas E, Converse SA (1993) Shared mental
models in expert team decision making. In: Castellan NJ Jr (ed)
Current issues in individual and group decision making. Psy-
chology Press, New York, pp 221–246
Craik K (1943) The nature of explanation. Cambridge University
Press, Cambridge
de Haen H, Stephan Klasen S, Matin Qaim M (2011) What do we
really know? Metrics for food insecurity and undernutrition.
Food Policy 36:760–769
de Suarez JM, Suarez P, Bachofen C, Fortugno N, Goentzel J, Gon-
çalves P, Grist N, Macklin C, Pfeifer K, Schweizer S, van Aalst
M, Virji H (2012) Games for a new climate: experiencing the
complexity of future risks. Pardee Center Task Force Report
European Commission’s Directorate—General for European Civil
Protection and Humanitarian Aid Operations (2016) ECHO
factsheet—Sahel: food & nutrition crisis. http://ec.europa.eu/
echo/files/aid/countries/factsheets/sahel_en.pdf. Accessed 19
April 2017
FAO (2009) Declaration of the world summit on food security:
FAO. http://www.fao.org/wsfs/world-summit/en/?no_cache=1.
Accessed 24 August 2017
FAO, IFAD, WFP (2013) The state of food Insecurity in the world
2013: the multiple dimensions of food security. http://www.
fao.org/docrep/018/i3434e/i3434e.pdf. Accessed 19 April 2017
Fisher-Vanden K, Wing IS, Lanzi E, Popp D (2013) Modeling
climate change feedbacks and adaptation responses: recent
approaches and shortcomings. Clim Change 117(3):481–495
Gentner D (1983) Structure-mapping: a theoretical framework for
analogy. Cognit Sci 7:155–170
Grothmann T, Patt A (2005) Adaptive capacity and human cognition:
the process of individual adaptation to climate change. Glob
Environ Chang 15(3):199–213
Hanjra MA, Qureshi ME (2010) Global water crisis and future food
security in an era of climate change. Food Policy 35:365–377
Hatløy A, Hallund J, Diarra MM, Oshaug A (2000) Food variety,
socioeconomic status and nutritional status in urban and rural
areas in Koutiala (Mali). Public Health Nutr 3:57–65
Holling CS (1973) Resilience and stability of ecological systems.
Ann Rev Ecol Evol Syst 4(1):1–23
Hussein K (2002) Food security: rights, livelihoods and the World
Food Summit—five years later. Soc Policy Adm 36:626–647
Johns T, Sthapit BR (2004) Biocultural diversity in the sustainability
of developing-country food systems. Food Nut Bul 25:143–155
Jolly CM, Gadbois M (1996) The effect of animal traction on labour
productivity and food self-sufficiency: the case of Mali. Agric
Syst 51:453–467
Kuyvenhoven A (2004) Creating an enabling environment: policy
conditions for less-favored areas. Food Policy 29:407–429
Laris P, Foltz JD, Voorhees B (2015) Taking from cotton to grow
maize: the shifting practices of small-holder farmers in the cot-
ton belt of Mali. Agric Syst 133:1–13
Menezes F (2001) Food sovereignty: a vital requirement for
food security in the context of globalization. Development
44(4):29–33
Mertz O, Mbow C, Reenberg A, Diouf A (2009) Farmers’ perceptions
of climate change and agricultural adaptation strategies in rural
Sahel. Environ Manag 43:804–816
Miller F, Osbahr H, Boyd E, Thomalla F, Bharwani S, Ziervogel G,
Walker B, Birkmann J, Van der Leeuw S, Rockström J, Hinkel J,
Downing T, Folke C, Nelson D (2010) Resilience and vulnerabil-
ity: complementary or conflicting concepts? Ecol Soc 15(3):1–25
Morgan MG, Fischhoff B, Bostrom A, Atman CJ (2002) Risk com-
munication: a mental models approach. Cambridge University
Press, Cambridge
Nicholson SE, Tucker CJ, Ba MB (1998) Desertification, drought, and
surface vegetation: an example from the West African Sahel. Bull
Am Meteorol Soc 79(5):815–829
Nielsen JØ, Reenberg A (2010) Cultural barriers to climate change
adaptation: a case study from Northern Burkina Faso. Glob Envi-
ron Chang 10:142–152
Osbahr H, Twyman C, Adger WN, Thomas DS (2008) Effective liveli-
hood adaptation to climate change disturbance: scale dimensions
of practice in Mozambique. Geoforum 39(6):1951–1964
Pelling M, High C (2005) Social learning and adaptation to climate
change. Benfield Hazard Research Centre, Disaster Studies Work-
ing Paper, No. 11, pp. 1–19
Richards P (1990) Local strategies for coping with hunger: central
Sierra Leone and Northern Nigeria compared. Afr Aff 89:265–275
Roudier P, Sultan B, Quirion P, Berg A (2011) The impact of future
climate change on West African crop yields: what does the recent
literature say? Glob Environ Change 21(3):1073–1083
Staatz JM (1994) The strategic role of food and agricultural systems in
fighting hunger through fostering sustainable economic growth.
Michigan State University. http://fsg.afre.msu.edu/scans/mali/
sp%2094-39.pdf. Accessed 19 April 2017
Staatz JM, D’Agostino VC, Sundberg S (1990) Measuring food security
in Africa: conceptual, empirical, and policy issues. Am J Agric
Econ 72:1311–1317
Sultan B, Roudier P, Quirion P, Alhassane A, Muller B, Dingkuhn M,
Ciais P, Guimberteau M, Traore S, Baron C (2013) Assessing cli-
mate change impacts on sorghum and millet yields in the Sudanian
and Sahelian savannas of West Africa. Environ Res Lett 8:1–9
Sylla MB, Elguindi N, Giorgi F, Wisser D (2016) Projected robust
shift of climate zones over West Africa in response to anthro-
pogenic climate change for the late 21st century. Clim Change
134:241–253
Tappan G, McGahuey M (2007) Tracking environmental dynamics and
agricultural intensification in southern Mali. Agric Syst 94:38–51
Torheim LE, Ouattara F, Diarra MM, Thiam FD, Barikmo I, Hatløy A,
Oshaug A (2004) Nutrient adequacy and dietary diversity in rural
Mali: association and determinants. Eur J Clin Nutr 58:594–604
Traore B, Corbeels M, van Wijk MT, Rufino MC, Giller KE (2013)
Effects of climate variability and climate change on crop produc-
tion in southern Mali. Eur J Agron 49:115–125
Tschakert P (2007) Views from the vulnerable: understanding climatic
and other stressors in the Sahel. Glob Environ Change 17:381–396
UN-OCHA (2012) Office for the coordination of humanitarian affairs
(OCHA) annual report. https://www.unocha.org/publication/
ocha-annual-report/ocha-annual-report-2012
Walker B, Holling CS, Carpenter S, Kinzig A (2004) Resilience, adapt-
ability and transformability in social–ecological systems. Ecol
Soc. https://doi.org/10.5751/ES-00650-090205
von Winterfeldt D, Edwards W (2007) Advances in decisanalysis:
from foundations to applications. Cambridge University Press,
Cambridge
Whitley CT, Rivers III L, Mattes S Marquart-Pyatt ST, Ligmann-Zie-
linska A, Olabisi LS, Du J (2017) Climate-induced migration:
using mental models to explore aggregate and individual decision-
making. J Risk Res 20:1–17

Environment Systems and Decisions
1 3
Wittig R, König K, Schmidt M, Szarzynski J (2007) A study of climate
change and anthropogenic impacts in West Africa. Environ Sci
Pollut Res 14:182–189
Wood MD, Bostrom A, Bridges T, Linkov I (2012) Cognitive map-
ping tools: review and risk management needs. Risk Anal
32:1333–1348
Wooten S (2003) Women, men, and market gardens: gender relations
and income generation in rural Mali. Hum Organ 62:166–177
Zaksek M, Arvai JL (2004) Toward improved iommunication about
wildland fire: mental models research to identify information
needs for natural resource management. Risk Anal 24:1503–1514