Land Use Policy 91 (2020) 104395

Contents lists available at ScienceDirect

Land Use Policy

journal homepage: [Link]/locate/landusepol

T
Farm households’ risk perception, attitude and adaptation strategies in
dealing with climate change: Promise and perils from rural Pakistan
Imran Khana,*,1, Hongdou Leia,1, Irshad Ali Shahb, Imad Alia, Inayat Khanc, Ihsan Muhammadd,
Xuexi Huoa,*, Tehseen Javede
a
College of Economics & Management, Northwest A&F University, 3 Taicheng Road, Yangling 712100, Shaanxi, China
b
Department of Agriculture Economics and Management, School of Management, Zhejiang University Hangzhou, China
c
School of Management, Northwestern Polytechnical University, Xi’an China
d
Shaanxi Key Laboratory of Earth Surface System and Environmental Carrying Capacity, College of Urban and Environmental Sciences, Northwest University, Xi’an
710127 China
e
College of Water Resources and Architectural Engineering, Northwest A&F University, Yangling 712100, Shaanxi, China

A R T I C LE I N FO A B S T R A C T

Keywords: Pakistan is highly vulnerable to extreme climatic events, such as floods and droughts. This study determines the
Adaptation measures farmers’ risk perception, risk attitude, adaptation measures and various aspects of vulnerability to climate
Climate variability change (e.g. floods, droughts, heavy rainfalls, pests and disease) at farm level in rural Pakistan. The risk per-
Flood and drought risks ception and attitude of farm households are crucial factors that influence farm productivity, investment and
Risk attitude
management decisions at this level. A well-designed questionnaire was used to interview 720 farm households
Risk perception
from six districts of the Khyber Pakhtunkhwa province. A binary logit model was used to determine the main
factors that affect the choice of adaptation strategies of the farm household. The findings revealed that crop
diversification, changing crop varieties, altering the crop calendar, varying the fertilizer used, mulching and
farm insurance were the main adaptation strategies followed by farm households. The results of the binary logit
model revealed that age, education, farm size, household size, credit accessibility, annual income and the
perception on the increase in temperature and decrease in rainfall had significant influence on the selection of
the adaption strategies. The findings of this study can provide guidance, policy recommendations and reference
for future researchers.

1. Introduction of several climatic hazards, such as floods, storms, droughts, etc. (Field
et al., 2012).
Climate change is an environmental hazard with effects that have Climate change is one of the major environmental problems faced
been increasingly perceived over the past several decades (Abid et al., by the modern world today. This phenomenon can generate fluctua-
2016; Patt and Schröter, 2008). In low-income states, the projections of tions, such as sea-level rise, changes in rainfall series and shifts in cli-
weather and climate variabilities predict changing climate with high matic regions because of rising temperatures. The degree of droughts,
vulnerability (Easterling et al., 2000; McCarthy et al., 2001). Balancing storms and floods is projected to increase because of the changes in
the costs and benefits of mitigation is challenging because the value of climate patterns. The risks associated with weather and extreme climate
estimating climate change damage depends on a number of parameters events can increase significantly due to the upswing in climatic varia-
with values that have been debated for decades. In terms of vulner- bility (Ahmad et al., 2007; Akcaoz and Ozkan, 2005), consequently
abilities, climate change has strong effects in poor agricultural and rural changing the frequency, magnitude and spatial amount of disasters,
communities of developing countries because of their low income and such as floods and droughts (Udmale et al., 2014). Such such changes
deprived adaptive capabilities (IPCC, 2007; Kurukulasuriya and will cause the the manifestation of these extreme events with con-
Rosenthal, 2013; Skoufias et al., 2011). In addition, deviations in siderable impact (Hay and Mimura, 2010; IPCC, 2007), which will
weather situations are likely to upswing the frequency and magnitude generate immense pressure on communities, natural systems, societies

⁎
Corresponding authors.
E-mail addresses: imran@[Link] (I. Khan), xuexihuo@[Link] (X. Huo).
1
These authors contributed equally to this research work.

[Link]

0264-8377/ © 2019 Elsevier Ltd. All rights reserved.

Received 10 August 2019; Received in revised form 22 October 2019; Accepted 27 November 2019
I. Khan, et al. Land Use Policy 91 (2020) 104395

and economies struggling to cope with other non-climatic threats that ranked 7th among the topmost regions of the biosphere exposed to the
seriously challenge the sustainable development efforts of the govern- twists of weather and climate-related risks (Kreft et al., 2013). The
ment, organisations and other parties (Hay and Mimura, 2010). severe droughts from 1999 to 2003 and the successive floods on 2010,
Despite the technologically advanced period, the weather is still a 2011, 2012 and 2014 are common examples of natural disasters in
key factor in determining agricultural productivity (Jha, 2015). Rainfall Pakistan (Abid et al., 2016).
and temperature act as the core drivers in crop production and rural Studies regarding the risks associated with extreme weather and
food security (Akhtar et al., 2019; Wheeler and Von Braun, 2013). In- climate events, farm-level vulnerability and adaptation strategies are
creased climate variability and extreme weather conditions affect crop rare in Pakistan, especially in the Khyber Pakhtunkhwa (KP) Province,
and livestock production (Chaudhary and Aryal, 2009; Howden et al., which is highly vulnerable and exposed to natural disasters. Therefore,
2007; Schiermeier, 2008) as well as an increase in the incidence of to address this issue, this study is designed to potentially shed light on
diseases, insects and pests and the deterioration of soil metabolism and the level of farmers’ perception, knowledge and awareness on the
soil water content (Liverman, 2008; Paudel et al., 2014). The techno- weather and climate variabilities and the associated vulnerabilities.
logically modified and improved ways of production are highly influ- Moreover, this study recognises the indigenous adaptation measures
enced by the change in climate conditions (e.g. rainfall, temperature, followed by farm household and the factors influencing these measures.
humidity, etc.). The crop productivity in a specific area, region or lo- Finally, the outcome of this research will serve as a benchmark for
cation is influenced substantially by local rather than global climatic future research and reference point for policymakers in devising the
conditions (Jha, 2015). Climate variation will disturb crop production appropriate adaptation policies to facilitate rural communities in sus-
considerably and lead to significant changes in agricultural yields. The taining their livelihoods against future climate-related risks.
extreme weather conditions and presence of insects, pests and diseases
might have supplementary adverse effects on the farming sector (IPCC, 2. Methodology
2014). Various factors, such as topography, socio-economic, biophy-
sical and environmental factors, substantially influence farming sys- 2.1. Study sites and sample procedure
tems. Moreover, climate change and weather variability shape the
natural hazards and influence the distribution of income, consequently The study sites are located in the KP province, which shares the
affecting the livelihood of agricultural communities (Paudel et al., international border with Afghanistan (Fig. 1). KP province is one of the
2014). Crop failures and production-related risks are the top risks as- coldest regions in Pakistan, with an average daily temperature of only
sociated with climate change (Drollette, 2009). The rural farming 30 °C. The temperature during several months of the year is warm to
community has almost no counteractions with natural disasters (e.g. hot, with temperatures that continuously exceed 25 °C and even
floods, droughts, hail storms, cyclones and storm surges) because of the reaching 40 °C. Despite being geographically the smallest province, KP
uncertainty of these events. To cope with the unpredictable conditions is the third largest province in Pakistan in terms of population and
of weather and climate change, farmers should adjust their farming economy. Based on the agricultural importance, prior vulnerability
practices and water resources and change their existing cropping pat- history, extreme weather and climatic change-related natural hazards
terns (IPCC, 2014; Jha, 2015). and vulnerability, the KP province was purposely selected for this
The literature revealed that climate change with significant world- study. As an important sector in Pakistan, agriculture plays a vital role
wide phenomenon. The effect of climate change in developing countries in the country’s economy by contributing 18.9 % to the GDP and en-
are severe because of the low level of adaptive capabilities (Bello et al., gaging 42.3 % of the labour force (GoP, 2017). The province is a risk-
2013). Floods and droughts are the most common and frequent natural prone zone that faces many climate and weather-related hazards, such
hazards that substantially cause economic and social risks for humans, as droughts, heavy rainfalls, floods and hurricanes. According to the
especially in developing countries with meagre adaptive capabilities International Disaster Database, 25 hurricanes occurred in Pakistan
and vulnerable residents. Further increases in these extreme events can from 1950 to 2019, six of which occurred in the KP province in the form
shape global warming and might impose additional hurdles to sus- of convective storms and tropical cyclones. During the past 25 years,
tainable development, which will cause food insecurity and hinder the province has experienced eight major flood incidences, including
poverty alleviation. Countries located at low latitudes, especially low- the 2010 flood incident, which is the most severe incident in the pro-
income countries, are more susceptible to extreme weather and climate vince. This incident affected 24 out of the 25 districts of the province
events than countries in high latitudes. A rise in surface temperature in (Ullah et al., 2015).
the future might push the former away from optimum temperatures
zones (Mendelsohn and Dinar, 2009). By 2080, the risk of hunger can 2.2. Sample techniques and data collection
seriously affect over 170 million people worldwide (Schmidhuber and
Tubiello, 2007). Implementing appropriate and relevant adaptation In this study, a multistage sampling technique was used to collect
strategies (e.g. disaster risk-reduction measures) into developmental the primary data from rural households. In the first sampling stage, the
plans and policies will give developing countries the opportunity to KP province was selected because of its agricultural importance and
reduce the natural hazards and risks associated with extreme weather prior vulnerability history. Six districts of the province, namely, Swat,
and climate events. Dir, Malakand, Charsadda, Mardan and Peshawar, were then chosen in
Developing countries, including Pakistan, are the most affected the second stage. The province is divided into four agro-ecological
areas of climate change. These countries are highly exposed to climatic zones based on the climate, temperature, rainfall, and topography. Zone
hazards in the form of an increase in surface temperature and fluctua- A is characterised by high northern mountains, including Buner,
tion in precipitation (Ali and Erenstein, 2017; IPCC, 2014). Pakistan is Shangla, Dir (Lower and Upper), Swat and Chitral districts. Zone B can
one of the susceptible countries that are highly exposed to natural ca- be described by the sub-humid eastern and wet mountains, including
lamities because of the dry topographical background and scarce re- Haripur, Batagram, Mansehra, Abbottabad, Kohistan and Torghar dis-
sources (Schilling et al., 2013). Over the last decade, Pakistan has ex- tricts. Zone C is located at the central valley plain, which covers
perienced a rise in incidence, magnitude and severity of natural Peshawar, Mardan, Charsadda, Malakand, Nowshehra, Swabi, Kohat
hazards, including floods, thrilling temperatures, extreme water scar- and Hangu districts. Lastly, Zone D is described as the piedmont plain
city, droughts in various areas and high occurrence of diseases and (i.e. Suleiman piedmont) consisting of Bannu, Karak, Lakki Marwat,
pests (Smit and Skinner, 2002). Pakistan was ranked as the 29th, 16th, Tank and Dera Ismail Khan districts. In the final sampling stage, using
12th and 8th most vulnerable country on 2009–2010, 2010–11, 2012 the equidistance principle from the preselected urban centre, a strati-
and 2015, respectively (Abid et al., 2016; Khan and Fee, 2014), and fied random sampling technique was used to select four to five villages