







































 
 

 

1 
© 2022 by the authors; licensee Asian Online Journal Publishing Group 
 

Agriculture and Food Sciences Research 
Vol. 9, No. 1, 1-8, 2022 

ISSN(E) 2411-6653/ ISSN(P) 2518-0193 
DOI: 10.20448/aesr.v9i1.3758 

© 2022 by the authors; licensee Asian Online Journal Publishing Group 

 
 

 
 
 
Agriculture Adaptation Strategies of Tunisian Oasis Households to Climate Change 

 
Amira Abdelhamid1   

Houcine Jeder2    

Ahmed Salah3   

 
 

( Corresponding Author)  
 
1,3Faculty of Economics and Management, University of Tunis El-Manar, Tunisia. 
1Email: amira.abdelhamid83@gmail.com  
3Email: ahmedtun@yahoo.fr 
2Regional Researches Center on Horticulture and Organic Agriculture, Sousse, and Laboratory of Economy and 
Rural Communities, Arid Regions Institute, University of Gabes, Medenine, Tunisia. 
2Email: dejderhoucine@yahoo.fr  

  

 

Abstract 
Climate change should have impacts on Tunisian arid regions oasis households are likely to bear 
the most significant negative impacts, the case of the oasis of "Metouia" in the governorate of 
"Gabes". This research is based on an econometric analysis through cross-sectional probit models 
involving 50 oasis households. The binary probit models showed that certain factors contribute 
significantly to the adaptation strategies identified, such as: the age of the agricultural household 
head, agriculture as a main activity, the agricultural production system adopted the mode of 
ownership of agricultural land, extension for farmers. All actions aimed at improving the 
resilience of Tunisian oasis agricultural households to climate change focus mainly on the 
strategies adopted by farmers in terms of water management, the technical choices and the 
production systems adopted combined with the experience and local know-how. 

 
Keywords: Climate change, Adaptation, Agricultural, Households, Econometric analyses, Tunisia. 

 
Citation | Amira Abdelhamid; Houcine Jeder; Ahmed Salah (2022). 
Agriculture Adaptation Strategies of Tunisian Oasis Households to 
Climate Change. Agriculture and Food Sciences Research, 9(1): 1-8. 
History:  
Received: 29 December 2021 
Revised: 15 February 2022 
Accepted: 1 March 2022 
Published: 8 March 2022 
Licensed: This work is licensed under a Creative Commons 

Attribution 4.0 License  
Publisher:  Asian Online Journal Publishing Group 
 

Funding: This study received no specific financial support. 
Authors’ Contributions: All authors contributed equally to the conception 
and design of the study. 
Competing Interests: The authors declare that they have no conflict of 
interests. 
Transparency: The authors confirm that the manuscript is an honest, 
accurate, and transparent account of the study; that no vital features of the 
study have been omitted; and that any discrepancies from the study as planned 
have been explained. 
Ethical: This study followed all ethical practices during writing.    

 

 

Contents 
1. Introduction ......................................................................................................................................................................................... 2 
2. Theoretical and Methodological Framework ............................................................................................................................... 2 
3. Results and Discussions ..................................................................................................................................................................... 5 
4. Conclusion and Recommendations .................................................................................................................................................. 7 
References ................................................................................................................................................................................................. 7 
 

 
 
 
 
 
 

 

 

 

mailto:amira.abdelhamid83@gmail.com
mailto:ahmedtun@yahoo.fr
mailto:dejderhoucine@yahoo.fr
https://creativecommons.org/licenses/by/4.0/
https://creativecommons.org/licenses/by/4.0/
https://www.doi.org/10.20448/aesr.v9i1.3758
https://orcid.org/0000-0001-7958-4894
https://orcid.org/0000-0003-1953-8209
https://orcid.org/0000-0002-3735-7092


Agriculture and Food Sciences Research, 2022, 9(1): 1-8 

2 
© 2022by the authors; licensee Asian Online Journal Publishing Group 

 

 

 

 
1. Introduction 

Agriculture lays a heavy burden on the environment in the process of providing humanity with food and fibers 
[1]. However, agriculture and food systems as well as the rural economies in the Maghreb and North Africa 
regions have been experiencing major drastically reduced agricultural production through extreme weather events, 
such as recurrent droughts and floods in these recent decades [2, 3].In these regions, the climate variability causes 
severe impacts on agriculture through long drought periods. Recurrent droughts often affect entire countries over 
multiyear periods and can result in serious social problems caused by water scarcity and the intensive demand of 
water for agriculture. Impacts anticipated under projected climate change such as increasing rainfall variability, 
increasing temperature, increasing evaporation rate and water deficit pose a significant challenge to the Maghreb 
region. Mean temperatures of Morocco, Algeria and Tunisia are expected to rise by between 2 and 4°C until 2100. 
Already by 2020, rainfall is expected to drop by between 5 and 20% [4]. Tunisia is one of the Maghreb countries 
and is very vulnerable to the water shortage. Most of the water resources have medium to poor quality, and the 
salinity is often high. Water deficits and droughts are ongoing risks for Tunisian agriculture. The agricultural 
sector that pro vides approximately 13% of the annual Gross Domestic Product to Tunisia is very vulnerable to 
climatic changes also due to poor soils, limited ground and surface water, low rainfall and recurrent droughts 
particularly in the arid regions of Tunisia In these regions, the consumption of irrigation water in irrigated areas 
and oases continues to increase to ensure the sustainability of the farming activity and guarantee the income of 
several households agricultural [5]. 

For example, in the arid regions of south-eastern Tunisia, the Gabes oases is known since antiquity for their 
vegetable crops that historically conserved the seeds, their greatest source of resilience facing difficult climate 
conditions (dry land). Howewer, since the past few decades, the oases are getting very damaged (over-exploitation 
of resources, urbanisation…) as well as multiple factors have been behind this degradation [6]. The majority of 
young people and women are not interested in agriculture and the older populations that are, are turning to more 
profitable field crops. Producing local seeds is forsaken for the benefit of improved hybride seeds that are often 
combined with the use of chemical fertilizers (highly polluting the soil). The lack of valorisation of local indigenous 
seeds, the absence of a seed market or even laws restricting the commercialisation of local seeds. This all explains 
why farmers are discouraged and have difficulties guaranteeing their own production. The risk is that future 
generations that loses their capacity to adapt to climate change and no longer is able to guarantee food security for 
the population surrounding the oases [7]. Faced with climatic risks on agricultural activity in oases of irrigated 
perimeters, reviewing the concept of adaptation through a multidimensional vision touching different economic, 
social, agronomic, hydrological and political aspects is important to know and detect the main intervening factors 
in adaptation strategies. The adoption of a bottom-up approach that focuses on autonomous adaptation behavior 
seems to me more adequate to draw recommendations for adaptation strategies appropriate to oasis production 
systems in south-eastern Tunisia, case of the Metouia oasis in the governorate of Gabes. 

Therefore, this article aims to study the perception and choice of appropriate measures among smallholders in 
the oases of south-eastern Tunisia for adaptation to climate change. The rest of the article is structured as follows: 
Section two describes the theoretical framework. Section three presents the methodology. Section Four discusses 
the results and Section Five provides conclusions and policy recommendations. 
 

2. Theoretical and Methodological Framework 
Farmers' behavior towards adaptation to climate change is shaped by socio-economic, physical and behavioral 

factors [8]. Institutional arrangements for the farmers and their working environment, development for access to 
markets and climatic factors are useful in shaping the behavior of smallholders [3, 9, 10]. The study of behavior 
adapting to climate change has many angles of analysis. There is both theoretical and empirical analysis. Some 
questions are positive, many others are prescriptive. There are microeconomic issues and more macroeconomic 
problems. Within this diversity, several approaches have been developed in particular: integrated assessment 
approaches, empirical (econometric) analysis, economy-wide simulation with models and decision support tools. 
Each of these approaches can help shed light on different aspects of the adaptation problem [11]. 

For the problem of adopting certain strategies or measures to adapt to climate change in agriculture, the 
behavioral study of the perception or the choice of strategies requires understanding the reaction of economic 
agents to current climate and weather events. Much of this evidence is provided by often interdisciplinary studies 
[12, 13]. However, more and more researchers are using large data sets at the household or farm level to explore 
how economic agents adapt fully and rationally. The evidence is particularly rich for the agricultural sector. 

Such an economic approach in relation to adaptive behavior such as "the econometric approach to climate" has 
been reviewed by Dell, et al. [14] and Hsiang [15]. These two works mainly focus on assessing the impact and 
effect of climatic factors on economic variables such as labor productivity, production and growth, rather than on 
the benefits, costs or extent of the economy adaptation. However, many of the ideas of Hsiang and Dell et al also 
apply to the econometrics of adaptation. 

Several researchers have sought to identify climatic effects both in transversely and chronologically, by 
comparing the impacts and / or adaptation behaviors across different climatic regimes by measuring the impact of 
particular meteorological events, such as floods, over time. Increasingly, they have access to panel data. Cross-
sectional studies are closely associated with the "Ricardian approach" developed by Mendelsohn, et al. [16]; 
Kurukulasuriya, et al. [17]; Seo and Mendelsohn [18]; Wang, et al. [19]. Given the great diversity of climates 
around the world, these studies provide ample evidence of adaptive behavior. The approach of nominal and ordered 
econometric models can simultaneously model the influence of all the explanatory variables on each of the different 

Contribution of this paper to the literature 
Research has shown that certain cultural practices can contribute to the resilience of agriculture 
in arid regions of Tunisia to climate change through adequate water management, technical 
choices and production systems adopted combined with the experience and local know-how of 
farmers.  



Agriculture and Food Sciences Research, 2022, 9(1): 1-8 

3 
© 2022by the authors; licensee Asian Online Journal Publishing Group 

 

 

adaptation practices, while allowing the potential correlation between the unobserved disturbances as well as the 
relationship between the adoptions of different adaptation practices [20]. Consequently, an agricultural household 
is confronted with the decision to adopt or not an adaptation strategy taking into account the parameters 
perceptions of climate change. Necessarily, this decision to adopt or not depend on the characteristics of 
agricultural households explained using an "ad-hoc" approach "through several factors: socio-economic, climatic, 
endowment of available resources (land, water and labor) and policies related to extension services as well as access 
to information at optimal time. Principal Component Analysis Method and the ordered and binary probit 
econometric model were mobilized in this work to study the behavior of small farmers in the Methaoui oasis in 
south-eastern Tunisia in the face of the challenges of climate change. 
 

2.1. Ordered Econometric Models  
Several studies have used various methodological approaches to analyze the determinants of adaptations to 

climate change and the choice of adaptation strategies. Most commonly used analytical approaches in the literature 
include discrete choice regression models like binary probit or logit [21, 22], multinomial probit or logit and 
multivariate probit [2, 3, 23-26]. Other empirical studies used principal component analysis and the Ricardian 
model [27].  Thus, the decisions of perception and adaptation to climate change are intrinsically multivariate and 
the attempt at univariate modeling excludes the useful economic information contained in the interdependent and 
simultaneous adoption decisions. On the basis of this argument, the study adopted the econometric technique of 
ordered and binary probit models to simultaneously model the perception and the influence of the set of 
explanatory variables on the main adaptation strategies [20, 28].  
 

2.1.1. Ordered and Binary Probit Models 
Ordered probit model is widely used approach to estimate models of ordered types. The ordered probit model 

is built around a latent regression in the same manner as the binomial probit model [29]: 
 

yi
∗ =  β′xi + εi                                                                                                                         (1) 

Equation 1 presents the latent variable (farmers' perceptions) in this study exhibits itself in ordinal categories 
which were coded as 0, 1, 2…j. The response of category j is thus observed when the underlying continuous 
response falls in the j-th interval as: 

𝑦 (𝑓𝑎𝑟𝑚𝑒𝑟𝑠′𝑝𝑒𝑟𝑐𝑒𝑝𝑡𝑖𝑜𝑛𝑠) = {

0                            𝑖𝑓 𝑦𝑖
∗  ≤ 𝛿0  ∶ 𝑙𝑜𝑤 𝑝𝑒𝑟𝑐𝑒𝑝𝑡𝑖𝑜𝑛

1     𝑖𝑓 𝛿0 ≤ 𝑦𝑖
∗  ≤ 𝛿1  ∶ 𝑚𝑜𝑑𝑒𝑟𝑎𝑡𝑒 𝑝𝑒𝑟𝑐𝑒𝑝𝑡𝑖𝑜𝑛

2                𝑖𝑓 𝛿1 ≤ 𝑦𝑖
∗  ≤ 𝛿2  ∶ ℎ𝑖𝑔ℎ 𝑝𝑒𝑟𝑐𝑒𝑝𝑡𝑖𝑜𝑛

    (2) 

Where in Equation 2, the variable Y* (i = 0, 1, 2) is the unobservable threshold parameters that were estimated 
together with other parameters in the model. When an intercept coefficient is included in the model, Yo* is 
normalized to a zero value and hence only j-1 additional parameters are estimated with Xs. As binary data models 
adopt or not adopt (0/1) adaptation strategies, the probabilities for each of the observed ordinal response, that is, 
farmer’s perception to climate change in this study had 3 responses which could be low, moderate and high with 
ordinal values of 0, 1, 2. 

For adaptation of such a strategy j, the latent variable in this case is a binary dependent variable with yi = 1 to 
adopt strategies j or yi = 0 not adopt strategies j. Binary probit models can also be motivated by an underlying 

continuous latent variable yi * which depends on β‘xi and an error term εi (for i = 1,…, n) as in the case of 
Equation 3. If the latent variables would be observable, this would lead to linear regression models. However, 
latent variables are not observable. But they can be related to the observed binary dependent variables yi: 

 

𝑦 (𝐹𝑎𝑟𝑚𝑒𝑟′𝑠 𝑠𝑡𝑟𝑎𝑡𝑒𝑔𝑖𝑒𝑠) = {
1 𝑖𝑓  𝑦𝑖

∗  ≥ 0 , 𝑎𝑑𝑜𝑝𝑡 𝑠𝑡𝑟𝑎𝑡𝑒𝑔𝑖𝑒𝑠 𝑗 

0 𝑖𝑓  𝑦𝑖
∗ < 0 , 𝑛𝑜𝑡 𝑎𝑑𝑜𝑝𝑡 𝑠𝑡𝑟𝑎𝑡𝑒𝑔𝑖𝑒𝑠 𝑗

                                        (3) 

 

If follows for the probability 𝑝(𝑥𝑖 , 𝛽) that 𝑦𝑖 takes the value one in the following Equation 4: 
 

𝑝𝑖(𝑥𝑖, 𝛽) = 𝑃(𝑦𝑖 = 1|𝑥𝑖, 𝛽) = 𝑃(𝑦𝑖
∗ ≥ 0|𝑥𝑖, 𝛽) = 𝑃(𝛽′𝑥𝑖 +  휀𝑖  ≥ 0) = 𝑃(휀𝑖  ≥ −𝛽′𝑥𝑖)      (4) 

 

Different binary response models can be derived by different distribution assumptions for휀𝑖 .If the error term 휀𝑖 
has a standard normal distribution (with expected value of zero and variance one), this leads to binary probit 

models. In this case 휀𝑖 is symmetrically distributed around zero so that it follows in binary probit model Equation 
5: 
 

𝑝𝑖(𝑥𝑖, 𝛽) = 𝑃(𝑦𝑖 = 1|𝑥𝑖 , 𝛽) = Φi(𝛽′𝑥𝑖) = 1 − Φi(−𝛽′𝑥𝑖) (5) 
  

The assumption of a known variance of 휀𝑖 is not problematic since this variance is not identified in binary 

response models and thus cannot be estimated besides 𝛽 so that it has to be normalized[29]. Interpretation of a 

parameter 𝛽ℎ in binary response models with respect to the (partial) effect of the corresponding explanatory 

variable 𝑥𝑖ℎ  (ℎ =  2, … , 𝑘) on the probability 𝑝𝑖(𝑥𝑖 , 𝛽). The parameter 𝛽ℎ cannot be interpreted as simply as in the 

linear probability model, i.e. it cannot be interpreted as the change in𝑝𝑖(𝑥𝑖 , 𝛽). If 𝑥𝑖ℎ increases by one unit (for a 

quantitative explanatory variable). Instead, the (partial) marginal probability effects of 𝑥𝑖ℎ in binary response 
models are as follows (for i = 1… n) in the following Equation 6: 
 

𝜕𝑝𝑖(𝑥𝑖, 𝛽)

𝜕𝑥𝑖ℎ
=

𝜕𝐹𝑖(𝛽′𝑥𝑖)

𝜕𝑥𝑖ℎ
=

𝑑𝐹𝑖(𝛽′𝑥𝑖)

𝑑(𝛽′𝑥𝑖)

𝜕(𝛽′𝑥𝑖)

𝜕𝑥𝑖ℎ
= 𝑓𝑖(𝛽′𝑥𝑖)𝛽ℎ                                                    (6)  

 



Agriculture and Food Sciences Research, 2022, 9(1): 1-8 

4 
© 2022by the authors; licensee Asian Online Journal Publishing Group 

 

 

While 𝐹𝑖(𝛽′𝑥𝑖) is the distribution function and 𝑓𝑖(𝛽′𝑥𝑖) is the density function both of the standard normal 

distribution in binary probit models. Partial marginal probability effects of 𝑥𝑖ℎ in binary probit models with 

𝜑𝑖(𝛽′𝑥𝑖) as the density function of the standard normal distribution in the following Equation 7: 
 

𝜕𝑝𝑖(𝑥𝑖, 𝛽)

𝜕𝑥𝑖ℎ
=

𝜕Φi(𝛽′𝑥𝑖)

𝜕𝑥𝑖ℎ
=

𝑑Φi(𝛽′𝑥𝑖)

𝑑(𝛽′𝑥𝑖)

𝜕(𝛽′𝑥𝑖)

𝜕𝑥𝑖ℎ
= 𝜑𝑖(𝛽′𝑥𝑖)𝛽ℎ                                           (7)   

 
The discussion so far has implicitly assumed quantitative explanatory variables. However, if explanatory 

variables are discrete or even binary, to calculate of the (partial) effect of an infinitesimal change of an explanatory 

variable 𝑥𝑖ℎ can be very inaccurate. Therefore, a discrete change of𝑝𝑖(𝑥𝑖, 𝛽)due to a discrete change ∆𝑥𝑖ℎ is (for i = 
1…, n) for the following Equation 8: 

∆𝑝𝑖(𝑥𝑖, 𝛽) = Φi(𝛽′𝑥𝑖 + 𝛽ℎ∆𝑥𝑖ℎ) −  Φi(𝛽′𝑥𝑖)                                                                     (8) 

If 𝑥𝑖ℎ is a dummy variable,∆𝑥𝑖ℎ = 1 implies the change from zero to one. Again it is possible to calculate 

(partial) average (discrete) effects and (discrete) effects at the mean �̅� =
1

𝑛
∑ 𝑥𝑖

𝑛
𝑖  of the explanatory variables. 

Where, Φ is the normal density function, h the threshold parameter and 𝑥𝑖ℎ the j the explanatory variables. The 
farmer's perceptions of climate change and the adoption of an adaptation strategy are specified as follows Equation 
9: 
 

𝑌 ∗ =  𝛽0 + 𝛽1𝑥1 + + 𝛽2𝑥2 +  𝛽3𝑥3 + + 𝛽3𝑥3 +  𝛽4𝑥4 + + 𝛽5𝑥5 + + 𝛽6𝑥6 + 𝛽7𝑥7 +  𝛽8𝑥8  (9) 
 

Y*(farmer’s perceptions) =1 (No perception), 2 (Average perception) or 3 (Good perception ( Ordered 
probit model). 
Y* (Farmer's strategies) = 1 (adopt strategies j), 0 (not adopt strategies j). (Binary probit models). 
x1= Ages of farmer (years), continuous (in number) . 
x2 = Level of education (ordered), 1 (literate), 2 (primary), 3 (secondary), 4 (university). 
x3 = Main agricultural activity, binary (1 if agriculture, 0 other). 
x4 = Place of residence, binary (1 if on the farm, 0 outside) . 
x5 = Farm size, continuous (in hectare). 
x6 = Type of agricultural production system, ordered (3-stage system, 2 classic system, 1 otherwise). 
x7 = Agricultural land owner, binary (1 if farm owner, 0 other). 
x8 = Membership of the Agricultural Development Group, binary (1 if yes, 0 no). 
 

2.2. The Study Area 
The Métouia oasis is one of the coastal oases of the governorate of Gabès. It is located 12 km north of the city 

of Gabès (south-eastern Tunisia) and covers an area of approximately 270 ha Figure 1. The Métouia oasis is 
characterized by low rainfall. The monthly distribution of precipitation is characterized by a period without rain 
(June, July and August) and a period with rainfall irregularly distributed over the other months and an annual 
water balance which is highly deficient. The oasis farms cover very small areas of around 1.5 ha on average. The 
crops are staged there in height, the palm trees are on the first floor, the pomegranate trees are on the second floor 
and alfalfa and market gardening are on the third floor. The drainage network is ineffective and moderately 
maintained. The Métouia oasis is characterized by the presence of a very shallow water table which closely 
conditions the evolution of the soil throughout the oasis [30, 31]. 
 

 
Figure 1. Location map of the Métouia oasis, South-Eastern Tunisia. 

 



Agriculture and Food Sciences Research, 2022, 9(1): 1-8 

5 
© 2022by the authors; licensee Asian Online Journal Publishing Group 

 

 

2.3. Data Type and Sources 
Data used in this analysis were collected from a household survey conducted in Métouia oasis of the 

governorate of Gabes on south-eastern Tunisia. The zones have been chosen to take a representative sample of 50 
farms for this case study. Data were gathered at the household level on socio-economic characteristics, agricultural 
production system characteristics, extension institutions and climate change perceptions Table 1.  

 
Table 1. Summary of household characteristics. 

Variable (type variable)  Description Number (%) Means 

Perception (Ordered variable)  

Climate change perceptions 50(100%)  

No perception 
Average perception 
Good perception 

6(20%) 

5(17%) 

19(63%) 
Age (discrete variable) Ages of farmer 50(100%) 50 
Education (Ordered variable) Level of education 50(100%)  
 Literate =1 

Primary =2 
Secondary =3 
University=4 

3(6%) 
37(73%) 

0(0%) 
10(20%) 

Main activity (Binary variable) Main agricultural activity 50(100%) 
if agriculture = 1 
other = 0 

25(50%) 
25(50%) 

Residence (Binary variable) Place of residence 50(100%) 
if on the farm = 1 
Outside = 0 

45(90%) 
5(10%) 

Farm size (Continuous variable)  Farm size 50(100%) 4.46 
Agricultural system (Binary variable)   Type of agricultural production system 50(100%)  

If stage system = 1 
If other system = 0 

14(26%) 
36(74%) 

Land owner (Binary variable) Agricultural land owner 50(100%) 
if farm owner = 1 
If other =0 

18(26%) 
32(74%) 

Membership (Binary variable) Membership of (GDA) 50(100%) 
If membership = 1 
If no =0 

23(26%) 
37(74%) 

Source: Data survey. 

 

3. Results and Discussions 
3.1. Perception of Climate Change for Oasis Farming Households 

The objective of this section is to identify the determining factors of perception of climate change for the oasis 
farming households. The ordered probit regression model was used to find out the contributing factors implicitly 
(ad-hoc) to the perception of the phenomenon of climate change and which can play in the development of 
adaptation strategies. The results of the ordered probit regression model are presented in Table 2. 
 

Table 2. Results of ordered probit regression model for perception of climate change. 

Perception of climate change via ordered probit regression model 

 Coef. Std. Err. P>|z| 

X1: AGE 0.036 0.038 0.349 
X2: EDUC 0.808 * 0.437 0.064* 
X3: AGR 0.692 0.773 0.371 
X4: RESID 3.364 *** 0.732 0.000*** 
X5: SIZE 0.611** 0.193 0.002** 
X6: SYSTEM 0.467 0.544 0.391 
X7: OWNER 1.512 * 0.788 0.055* 
X7: MEMBERSHIP 1.991*** 0.723 0.006*** 
 Number of obs.   =            50 

Wald chi2(8) =                  504.42                               Prob > chi2  =        0.000 
Pseudo R2   =                    0.353 
Log pseudo likelihood =   -17.636                

Notes: *** significant at 1%, ** significant at 5%, * significant at 10%. 
Source: Model results. 

 
The results of the probit regression model ordered in Table 2 show an overall significance of the model at the 

1% level (Prob> chi2). The positive contribution and the level of significance of the independent variables also 
determine the importance of these variables in the functioning of the oasis production system. Indeed, a significance 
level of 1% for the residence variable (RESID) clearly reflects that the farmer, who is installed in the oasis, felt the 
change of the environment and the deterioration of the oasis agricultural systems from day to day other than a 
farmer who is outside the oasis. The other variable that is significant at the 1% level is membership of an 
agricultural development group (MEMBERSHIP) which is responsible for the activity of agriculture in the oasis, 
among other things the distribution of irrigation hours. Being a member of this group means ease of access to 
information and extension and staffing of main irrigation hours or additional hours. Therefore, we can deduce that 
these two variables are of the first order from the point of view of perception of climate change among oasis 
farming households. Then, in second order, we find the variable area of the agricultural holding (SIZE) significance 
at the 5% level. This variable shows that the perception can be perceived at the level of large farms whose activities 
require a lot of production inputs such as water, labor, etc. In third order, at the level of positive significance of 



Agriculture and Food Sciences Research, 2022, 9(1): 1-8 

6 
© 2022by the authors; licensee Asian Online Journal Publishing Group 

 

 

10%, we find variables like the level of education (EDUC) as well as the property of the land (OWNER) which 
plays an important role in the understanding of the phenomenon of climate change and their impact on oasis 
systems when the farmer owns their agricultural land. The land ownership variable (OWNER) also reflects a 
socio-cultural aspect among some farmers, beyond that oasis activity is a source of agricultural income, but it is 
also a natural heritage characterizing the region which must be preserved for biodiversity, sustainable development 
and for future generations. 
 

3.2. Adaptation Strategies on Oasis Farming Households 
The results of the ordered probit regression model showed the positive and significant contribution of certain 

key variables that can be used firstly to understand the behaviors of oasis households with respect to the 
phenomenon of climate change and secondly, avenues for reflection to develop adequate strategies adaptation. in 

the Table 3, the Principal Component Analysis method (PCA) was used according to a set of classification criteria 
in relation to the variables of positive perception of climate change identified by the ordered probit model. Among 
these criteria are: CSD: change of sowing date; ADD_IRRIG: additional irrigation; PURCH_IRRIG: purchase of 
additional irrigation hours; SAL_AN: sale of animals to finance agricultural activity; ADOPT_CROP: adoption of 
other crops; POLITICAL_INSTR: State intervention through subsidies and encouragement; ACCES_CREDIT: 
Access to credit to invest and finance agricultural activity. 

 
Table 3. Component matrix of principal component analysis (APC). 

Classification Criteria 
Components 

1 2 3 

CSD -0.394 -0.244 0.729 
ADD_IRRIG 0.799 0.404 0.254 
SAL_AN -0.593 0.542 -0.051 
ADOPT_CROP -0.630 -0.392 0.577 
POLITICAL_INSTR -0.206 0.628 0.534 
ACCES_CREDIT 0.714 0.175 -0.315 
PURCH_IRRIG 0.887 0.305 0.160 

Note: Extraction method: Principal component analysis. a. 3 components extracted. 

 
Component 1 characterized by the positive contribution of criteria (ADD_IRRIG, PURCH_IRRIG and 

ACCES_CREDIT) can be interpreted as Strategy 1: Adaptation strategy in terms of water saving policy. Component 2 
characterized by the positive contribution of criteria (SAL_AN and POLITICAL_INSTR) can be interpreted as 
Strategy 2: Autonomous adaptation incentive strategy and component 3 characterized by the positive contribution of the 
criteria (CSD and ADOPT_CROP) can be interpreted as Strategy 3: Technical strategy and production system. 

These three adaptation strategies reflecting the behaviors of oasis farming households which are deduced by 
the main component method on the data from the survey questionnaires are also confirmed by local actors, whether 
they are experts or agents of the service agricultural extension in the region. 
 

3.3. Determinants factors of Adaptation Strategies of Oasis Farming Households to Climate Change  
The results of ordered binary probit models in the Table 4 show that the overall significance of a level 5% 

(Prob> chi2) for the Strategy 1 and Strategy 2; at level 1% (Prob> chi2) for the Strategy 3 reflecting acceptance of the 
choice of adaptation variables to describe the strategies identified. The results also show the significant 
contribution of certain variables to express adaptive behavior among farmers in the oasis of Methouia.  

 
Table 4. Binary probit models results. 

Explanatory 
variables 

Strategy 1 Strategy 2 Strategy 3 

Coef. P>|z| Coef. P>|z| Coef. P>|z| 

X1: AGE 0.259*** 0.003 0.065 0.250 0.155** 0.027 
X2: EDUC 0.583 0.147 0.504 0.114 0.395 0.380 
X3: AGR 1.863** 0.045 1.299** 0.047 1.747*** 0.008 
X4: RESID 1.828 0.191 -1.177 0.430 -2.879 0.060 
X5: SIZE 0.050 0.795 0.044 0.854 0.075 0.755 
X6: SYSTEM -1.033 0.186 1.767*** 0.003 1.731*** 0.001 
X7: OWNER 2.345*** 0.001 1.366* 0.069 0.841 0.249 
X7: MEMBERSHIP 2.472** 0.016 1.617* 0.079 2.051** 0.034 

Constante -19.642 0.002 -8.700 0.052 -13.740 0.015 
 Number of obs.   =     50 

Wald chi2(8) =  16.15                                             
Prob > chi2  =          0.040 
Pseudo R2   =           0.388 

Number of obs.   =       50 
Wald chi2(8)=  23.02                                             
Prob > chi2   =         0.003 
Pseudo R2  =            0.413 

Number of obs.  =        50 
Wald chi2(8)=  30.83                                             
Prob > chi2   =            0.000 
Pseudo R2   =              0.468 

Notes: *** significant at 1%, ** significant at 5%, * significant at 10%. 

 
The positive and significant contribution at the 1% level for variables such as age (AGE) and land ownership 

(OWNER) and also other variables that are significant at the 5% level as the main activity variable which is 
agricultural activity (AGR) and the variable member of an agricultural development group (MEMBERSHIP). 
These variables explain and justify the adoption of strategy 1 (Adaptation strategy in terms of water saving policy) for 
some farmers in the oasis of Methouia. Indeed, an oasis household head who owns a farm, his main activity is 
agriculture and member of an agricultural development group where their access to water is possible, all these 
conditions allow him to access credit to invest in water saving for the purchase of drip irrigation or to build a water 
basin to store rainwater or additional irrigation water. 

For strategy 2 (Incentive strategy for autonomous adaptation), the results of the binary probit models in Table 3 
show the positive and significant contribution at the 1% level for the classification variable of the production 



Agriculture and Food Sciences Research, 2022, 9(1): 1-8 

7 
© 2022by the authors; licensee Asian Online Journal Publishing Group 

 

 

system (SYSTEM), at the level of 5% for the variable (AGR) and at the level of 10% for the two variables 
(OWNER and MEMBERSHIP). This strategy explains that when agriculture is the main activity for the oasis 
household head who is also the owner of a farm and member of an agricultural development group for access to 
irrigation water, contribute to the choice of orientation of oasis agricultural production system. This orientation 
towards a new oasis production system can be interpreted as a kind of strategy of adaptation to climate change 
which is carried out for the benefit of the activity of the breeding in particular the cattle breeding which has known 
a significant deterioration due to of their significant cost. This autonomous adaptation strategy is achieved through 
the sale of heads of cattle to finance agricultural activity and also through the incentive procedure of public actors 
to encourage investment in profitable crops with high added value such as pomegranate trees. This strategy 
implicitly reflects the transformation and the dynamics of the functioning of oasis agricultural households in the 
south-eastern region, in particular in the oasis of Methouia. 

For strategy 3 (Technical strategy and production system), the results of the binary probit models in Table 3 show 
the positive and significant contribution at the 1% level of the principal active variable (AGR) and the production 
system orientation (SYSTEM), and at the 5% level for age (AGE) and member of an agricultural development 
group (MEMBERSHIP) variables. In fact, in recent years, we have noticed a change in the classic oasis production 
system in three stages (palm, arboriculture, market garden and fodder crops) associated with the activity of cattle 
and goat breeding towards a new production system oasis in two stages (arboriculture, vegetable and fodder crops) 
associated only with goat farming. This strategy is imposed by the phenomenon of urbanization, the change in 
lifestyle of oasis households and also the degradation of natural resources due to climate change. This positive 
contribution explains that these variables together play an important role in the strategy of technical adaptation 
and orientation of the production system, whether through the change of the date for certain vegetable crops 
thanks to the experience of older farmers although the orientation towards less costly agricultural production 
systems that consume less water. 

The perception of oasis households of climate change in the study area was consistent with the findings of 
other researchers around the world. Indeed, the regression analysis of the ordered probit model revealed that 
certain variables such as education, agricultural area, residence, owner of agricultural land and membership of an 
agricultural development group have influenced the perception of climate change by Farmers. The same 
interpretations for these variables are justified by the work developed in the central agricultural zone of the state of 
Delta, Nigeria [32]. For the different adaptation strategies and their determining factors which are identified by 
the binary probit regression model, for example, the variables: age, agriculture is the main activity, owner of 
agricultural land and membership of a development group agricultural, reflect the adaptive behavior of oasis 
households in Methouia. The motivation for this adaptive behavior is based almost on four key terms: experience 
for autonomous adaptation and orientation of production systems, the owner of agricultural land for access to 
credit and membership in a group of agricultural development for access to water and information and an extension 
service. These same key adaptation terms also summarize the adaptation of farmers' livelihoods to environmental 
changes in the case of the Minqin oasis, northwest China [33]. 
 

4. Conclusion and Recommendations 
The study aimed to assess the perceptions and adaptation strategies of farmers to climate change in the oasis of 

Methouia in south-eastern Tunisia. It was found that the perception was raised among the majority of oasis 
farmers who were well aware that the climate was changing. The majority of farmers noted that there was an 
increase in temperature, decrease in rainfall, changes in the timing of rains and an increase in the frequency of 
droughts. The most common adaptation strategies among farm households were: crop diversification, change of 
production system, increase in water conservation practices, adjustment and management of livestock, the 
abundance of cattle breeding for the benefit of oasis agriculture and the increased use of irrigation technology 
through access to credit. The results of the study also show that certain variables such as level of education, 
residence on the farm, agricultural area, land owner and membership in the agricultural development group are 
crucial factors in influencing the probability oasis farmers to perceive climate change. Likewise, factors such as the 
age of the head of household, education, the system of the land owner and membership in the agricultural 
development group, facilitate access to credit and also to extension and information on change climate. These 
variables can be considered as factors to trace the most adequate adaptation strategies for oasis farmers of 
Methouia to climate change. Any policy aimed at strengthening the adaptive capacity of farmers in the study area 
should consider the use of the factors mentioned above in developing adaptation strategies. The importance of 
these socio-economic and technical factors of production in the perception and strategies of adaptation to climate 
change in the case of agriculture is justified by several studies in the world [24, 33-39]. Indeed, this study was an 
example to show that the bottom-up approach going from the individual scale for the case of the farmer to the 
global (community or society) to forecast the perception and ideas of the autonomous adaptation. This approach 
can be interpreted as the most effective methodological process in the design of adequate adaptation strategies 
which takes into account all the economic, social and ecological characteristics of a given region. Today, it is time 
to rethink the development of adaptation strategies to climate change by strengthening the adoption of the bottom-
up approach on scientific and participatory bases with the actors concerned, first and foremost the farmer and their 
concerns for the internal and external environment of their activity. 
 

References 
[1] IPCC, Impacts, adaptation and vulnerability, Cambridge. Cambridge UK: University Press, 2007. 
[2] R. M. Hassan and C. Nhemachena, "Determinants of African farmers’ strategies for adapting to climate change: Multinomial choice 

analysis," African Journal of Agricultural and Resource Economics, vol. 2, pp. 83-104, 2008. 
[3] T. Deressa, R. Hassan, C. Ringler, T. Alemu, and M. Yesuf, "Analysis of the determinants of farmers’ choice of adaptation methods 

and perceptions of climate change in the Nile basin of Ethiopia," IFPRI Discussion Papers No. 798, International Food Policy 
Research Institute, Washington, DC2008. 

[4] METT (Ministry of the Environment and Spatial Planning Tunisia), Vulnerability of the Maghreb region to climate change, and needs 
for adaptation. Tunis: UNDP/GEF & Ministry of Environment and Spatial Planning Tunisia, 2002. 



Agriculture and Food Sciences Research, 2022, 9(1): 1-8 

8 
© 2022by the authors; licensee Asian Online Journal Publishing Group 

 

 

[5] R. Mougou, M. Mansour, A. Iglesias, R. Z. Chebbi, and A. Battaglini, "Climate change and agricultural vulnerability: A case study 
of rain-fed wheat in Kairouan, Central Tunisia," Regional Environmental Change, vol. 11, pp. 137-142, 2011. 

[6] A. Marshall, E. Lavie, J.-L. Chaléard, M. Fort, and J. D. Lombard, "Proceedings of the international colloquium Oases in the 
globalization: Ruptures and continuities, Paris, 16 and 17 December 2013," in Proceedings of the international colloquium Oases in the 
globalization: Ruptures and continuities, Paris, 2013-December 16-17, 2014. 

[7] S. Abdedaiem and B. V.-B. Ahmed, "Le paradoxe des oasis du sud tunisien: rénovation ou mort d'un concept?," in Colloquium "Oasis 
in Globalization: Ruptures and Continuities", Colloquium "Oases in Globalization: Ruptures and continuities", Colloquio "Los Oasis en la 
globalización: Rompimientos y Continuidades", 2013, pp. 21-28. 

[8] U. Doerr, T. Mahmoud, and U. Schmidt, "Overconfidence and risk management of Ethiopian farmers," Working Paper, University 
of Kiel2011. 

[9] S. Devereux, "Famine in the Twentieth Century," Working Paper 105, Brighton; Institute of Development Studies, University of 
Sussex2000. 

[10] T. Deressa, R. Hassan, and C. Ringler, "Factors affecting the choices of coping strategies for climate extremes," IFPRI Discussion 
Paper No. 01032, Washington, DC2010. 

[11] K. Fisher-Vanden, I. Sue Wing, E. Lanzi, and D. Popp, "Modeling climate change feedbacks and adaptation responses: Recent 
approaches and shortcomings," Climatic Change, vol. 117, pp. 481-495, 2013.Available at: https://doi.org/10.1007/s10584-012-
0644-9. 

[12] E. C. Penning-Rowsell, P. Sultana, and P. M. Thompson, "The ‘last resort’? Population movement in response to climate-related 
hazards in Bangladesh," Environmental Science & Policy, vol. 27, pp. S44-S59, 2013.Available at: 
https://doi.org/10.1016/j.envsci.2012.03.009. 

[13] N. Ranger and S. Surminski, "A preliminary assessment of the impact of climate change on non-life insurance demand in the 
BRICS economies," International Journal of Disaster Risk Reduction, vol. 3, pp. 14-30, 2013.Available at: 
https://doi.org/10.1016/j.ijdrr.2012.11.004. 

[14] M. Dell, B. F. Jones, and B. A. Olken, "What do we learn from the weather? The new climate-economy literature," Journal of 
Economic Literature, vol. 52, pp. 740-98, 2014. 

[15] S. Hsiang, "Climate econometrics," NBER Working Paper 22181, National Bureau of Economic Research, Cambridge MA2016. 
[16] R. Mendelsohn, W. D. Nordhaus, and D. Shaw, "The impact of global warming on agriculture: A Ricardian analysis," The American 

Economic Review, vol. 84, pp. 753-771, 1994. 
[17] P. Kurukulasuriya, N. Kala, and R. Mendelsohn, "Adaptation and climate change impacts: A structural Ricardian model of 

irrigation and farm income in Africa," Climate Change Economics, vol. 2, pp. 149-174, 2011. 
[18] S. N. Seo and R. Mendelsohn, "An analysis of crop choice: Adapting to climate change in South American farms," Ecological 

Economics, vol. 67, pp. 109-116, 2008. 
[19] J. Wang, R. Mendelsohn, A. Dinar, and J. Huang, "How Chinese farmers change crop choice to adapt to climate change," Climate 

Change Economics, vol. 1, pp. 167-185, 2010.Available at: https://doi.org/10.1142/s2010007810000145. 
[20] R. Belderbos, M. Carree, B. Diederen, B. Lokshin, and R. Veugelers, "Heterogeneity in R&D cooperation strategies," International 

Journal of Industrial Organization, vol. 22, pp. 1237-1263, 2004.Available at: https://doi.org/10.1016/j.ijindorg.2004.08.001. 
[21] H. Acquah-de Graft and E. Onumah, "Farmers’ perceptions and adaptations to climate change: An estimation of willingness to 

pay," Agris, vol. 3, pp. 31-39, 2011. 
[22] B. Fosu-Mensah, P. Vlek, and M. Manschadi, "Farmers’ perceptions and adaptations to climate change: A case study of 

Sekyedumase District in Ghana," presented at the A Contributed Paper Presented at World Food Systems Conference, Tropentag, 
Zurich, 14-16 September,2010, 2010. 

[23] ACCCA, "Advancing capacity to support climate change adoptation, project No. 093. Improving Decision- Making Capacity of 
Small Holder Farmers in Response to Climate Risk Adaptation in Three Drought-Prone Districts of Tigray, Northern Ethiopia 
Farm—Level Climate Change Perception and Adaptation in Drought Prone Areas of Tigray, Northern Ethiopia," 2010. 

[24] N. Sofoluwe, A. Tijani, and O. Baruwa, "Farmers perception and adaptation to climate change in Osun State, Nigeria," African 
Journal of Agricultural Research, vol. 6, pp. 4789-4794, 2011. 

[25] T. C. Nzeadibe, C. L. Egbule, N. Chukwuone, and V. Agu, "Farmers’ perceptions of climate change governance and adaptation 
constraints in Niger delta region of Nigeria," African Technology Policy Network, Research Paper No. 72011. 

[26] T. Aemro, K. Mengistu, and T. Beyene, "Climate change adaptation strategies of smallholder farmers: The case of Babile District, 
East Hararghe Zone of Oromiya Regional State of Ethiopia," Journal of Economics and Sustainable Development, vol. 3, pp. 1-12, 2012. 

[27] P. Kurukulasuriya and R. Mendelsohn, "Crop selection: Adapting to climate change in Africa," CEEPA Discussion Paper Number 
26, Centre for Environmental Economics and Policy in Africa, University of Pretoria2006. 

[28] C.-T. J. Lin, K. L. Jensen, and S. T. Yen, "Awareness of foodborne pathogens among US consumers," Food Quality and Preference, 
vol. 16, pp. 401-412, 2005. 

[29] W. Greene, Econometric analysis, 5th ed. New Jersey: Prentice Hall, 2003. 
[30] A. Hatira, M. Grira, B. Ben Mansour, and T. Gallali, Impact of irrigation water on the evolution of gypsum accumulations. Case of the oasis 

of Metouia (Gabes N). Watmed, Monastir: International Symposium on Water in the Mediterranean Basin: Resources and 
Sustainable Development, 2002. 

[31] M. Grira, A. Hatira, B. Benmansour, and T. Gallali, Hydrodynamics and hydrochemistry of the superficial aquifer of the Metouia oasis 
(North Gabes, Tunisia). Watmed, Monastir, Tunisia: International Symposium on Water in the Mediterranean Basin: Resources and 
Sustainable Development, 2002. 

[32] A. Ofuoku and C. Chukwuji, "Farmers’ group growth trend in Delta State, Nigeria," Global Journal of Science Frontier Research: 
Agriculture and Biology, vol. 12, pp. 62-68, 2012. 

[33] J. Chen, S. Yin, H. Gebhardt, and X. Yang, "Farmers’ livelihood adaptation to environmental change in an arid region: A case 
study of the Minqin Oasis, Northwestern China," Ecological Indicators, vol. 93, pp. 411-423, 2018.Available at: 
https://doi.org/10.1016/j.ecolind.2018.05.017. 

[34] C. Chen, M. Doherty, J. Coffee, T. Wong, and J. Hellmann, "Measuring the adaptation gap: A framework for evaluating climate 
hazards and opportunities in urban areas," Environmental Science & Policy, vol. 66, pp. 403-419, 2016.Available at: 
https://doi.org/10.1016/j.envsci.2016.05.007. 

[35] M. Abid, J. Schilling, J. Scheffran, and F. Zulfiqar, "Climate change vulnerability, adaptation and risk perceptions at farm level in 
Punjab, Pakistan," Science of the Total Environment, vol. 547, pp. 447-460, 2016.Available at: 
https://doi.org/10.1016/j.scitotenv.2015.11.125. 

[36] K. Alam, "Farmers’ adaptation to water scarcity in drought-prone environments: A case study of Rajshahi District, Bangladesh," 
Agricultural Water Management, vol. 148, pp. 196-206, 2015.Available at: https://doi.org/10.1016/j.agwat.2014.10.011. 

[37] C. Nhemachena and R. Hassan, "Micro-level analysis of farmers’ adaptation to climate change in Southern Africa," IFPRI 
Discussion Paper No. 00714. International Food Policy Research Institute, Washington, D.C2007. 

[38] D. Maddison, "The perception of and adaptation to climate change in Africa," CEEPA Discussion Paper No.10. Centre for 
Environmental Economics and Policy in Africa, University of Pretoria, South Africa2006. 

[39] J. B. Smith, T. Dickinson, J. D. Donahue, I. Burton, E. Haites, R. J. Klein, and A. Patwardhan, "Development and climate change 
adaptation funding: Coordination and integration," Climate Policy, vol. 11, pp. 987-1000, 2011.Available at: 
https://doi.org/10.1080/14693062.2011.582385. 

  
Asian Online Journal Publishing Group is not responsible or answerable for any loss, damage or liability, etc. caused in relation to/arising out of the use of the content. 
Any queries should be directed to the corresponding author of the article. 
 


