Agricultural Water Management 216 (2019) 284–293
Contents lists available at ScienceDirect
Agricultural Water Management
journal homepage: www.elsevier.com/locate/agwat
Paying for privately installed irrigation services in Northern Ghana: The case T
of the smallholder Bhungroo Irrigation Technology
N.A. Akro afi , D.B. Sarponga,⁎, H.A.S. Somuaha, Y. Osei-Owusub
a Department of Agricultural Economics & Agribusiness, University of Ghana, Legon, Ghana
b Conservation Alliance, Ghana
A R T I C L E I N F O A B S T R A C T
Keywords: We assess factors influencing farmers’ willingness to pay (WTP) amount for privately installed smallholder ir-
Water conservation rigation systems in Northern Ghana that conserve water for farming. Northern Ghana has a unimodal rainfall
Smallholder irrigation pattern that last for three months with a prolonged dry season. We provide insights into smallholder farmers’
Privately installed affordability of simple irrigation systems fashioned on build-operate and transfer and how long it would take to
Willingness to pay breakeven to own the system. We add to the largely paucity in the literature in Northern Ghana on smallholder
Payback period
farmer access to irrigation services. Primary data and focus group discussions from farmers was collected be-
tween November 2015 and July 2016 in communities where the technology is installed as pilots and in areas
noted for floods during the single rainy season and drought in the dry season. The Contingency Valuation
Method was used to elicit information on farmers Willingness To Pay Amount. The Tobit regression model
estimated the key factors that would influence their WTP amount whilst the simple investment appraisal
technique, payback period, estimated the breakeven point. The mean WTP amount from the farmers was
GHS180 ($45) per two seasons (a year) per hectare. Given this mean willingness to pay amount, at baseline, the
typical group of five (5) maize farmers can afford to pay to own one BIT after 22 years of life of the technology.
That notwithstanding, for sustainability of such a project, there is the need for strong farmer based groups to
operate and manage the BIT and for the farmers to access farm inputs, access markets and be able to sell their
produce at adequate prices to cover costs and to increase their income. Recommendations directed at NGOs,
extension institutions, and local government authorities are to intensify training given to farmers on adoption of
irrigation technologies as well as strengthening farmer-based organizations for sustainability and management of
such technologies.
1. Introduction Globally, irrigation development is seen as key in mitigating cli-
matic change effects on crop production and important in increasing
Water for agricultural production has become dire in dry and semi- the utilization of the same piece of land several times in a year, for land-
arid areas with climatic changes and the need to conserve water for scarce smallholder farm households, thus increasing production and
farming has become germane. Farmers in sub-Saharan Africa are ex- productivity to increase incomes and improving livelihoods. Hasnip
periencing climatic changes with associated changes in rainfall patterns et al. (2001) provide a comprehensive literature review on the con-
accompanied by frequent floods and droughts, leading to consequences tribution of irrigation to farm households in sustaining rural livelihoods
for already vulnerable people in terms of declining crop yields (FAO, in Bangladesh and Nepal through the security of increased levels of
2008; Umesh et al., 2015). Food crop farmers’ dependence on rainfall farm productivity, employment and incomes and a fostering of the
for food production has become erratic, sometimes with accompanying linkage effects of farm-level irrigation development for the local
flooding, threatening food security and efforts to combat poverty. In- economy and beyond. Kamwamba-Mtethiwa et al. (2016) and Mango
deed, the scarcity of water for smallholder farmer’s crop production is a et al. (2018) all assert to the improvements that smallholder farm ir-
cause in agriculture’s underperformance in sub-Saharan Africa and is rigation systems impart to farm level crop productivity, impart to
one of the main reasons that Africa lags behind other regions on most of farmer adaptation to climate variability and in enhancing household
the Millennium Development Goals (MDGs). food security at the farm level.
⁎ Corresponding author.
E-mail address: dbsarpong@ug.edu.gh (D.B. Sarpong).
https://doi.org/10.1016/j.agwat.2019.02.010
Received 31 October 2018; Received in revised form 7 February 2019; Accepted 8 February 2019
Available online 16 February 2019
0378-3774/ © 2019 Elsevier B.V. All rights reserved.
N.A. Akrofi, et al. Agricultural Water Management 216 (2019) 284–293
Namara et al. (2013) discuses the types of irrigation technologies it employs almost 40% of the economically active population (SGER,
that are used at the smallholder farmer level to include human pow- 2018). The service and the industry sector comprise 50% and 31.7% of
ered, rope and treadle pumps to liquid fuel engine-driven systems and GDP, respectively. With relatively low levels of farm productivity
solar powered pumps as well as gravity and river diversion methods. arising from low farm technology uptake and the effects of drought and
The FAO (2014) provides practical examples of these common irriga- climate change, government policies have been to commit to investing
tion techniques (watering cans, treadle pumps, motorized pumps, solar in technologies to protect the agricultural sector.
pumps, shallow wells, canal and pipe conveyance systems, sprinkler Weather uncertainties such as droughts, floods and extreme tem-
irrigation systems, drip irrigation systems, small-scale community irri- peratures are having adverse impacts on Ghana's agriculture, worsening
gation systems, among others) and their cost implications in operations food shortages, food insecurity and livelihoods. To attain the
and maintenance to small-scale farmers, and the key principles and Millennium Development Goals (MDGs) associated to food security and
practices for selection and installation. The FAO (2014) discussions of poverty reduction (FAO, 2008), poverty reduction has become a major
these common irrigation techniques’ effectiveness are related to the global agenda (Kassie et al., 2010).
diverse conditions of climate change, soil and water resources avail- In Ghana, a large proportion of the rural population rely on agri-
ability and the cost implications of installation, operation and main- culture (82%, Ghana Statistical Service (GSS, 2016) as their means of
tenance. The listed cost estimates of some of the common techniques, livelihood. Several agricultural strategies are in place to reduce poverty
most often excluding operations and maintenance are: watering can (US of farmers by increasing productivity of crops including staple crops
$500/ha); treadle pumps (US$500/ha and labour cost of US$600/ha); with a primary objective of linking farmers to markets to increase
solar pumps (US$10-15,000/ha) and most small scale community irri- productivity and income levels. For example, the Ghana Shared Growth
gation systems (simple inlet structures US$500-1000/ha to US$4000- and Development Agenda (GSGDA) and Food and Agriculture Sector
6000/ha of standardized quality structures). Development Policy (FASDEP II), the policy documents of the Gov-
Scarcity of water therefore has been a restraining factor for crop ernment of Ghana on agriculture, focus on using cash crops and food
production in the drier regions in sub-Saharan Africa including the crops as a poverty reduction instrument to increase farmer income in
northern regions of Ghana, and to address declining food crop pro- Ghana. Crop productivity expansion is a main concern and staple crops
ductivity, water conservation and irrigation management issues needs defined in FASDEP II for support are cassava, maize, rice, yam and
addressing for sustainable livelihoods. Irrigated agriculture is important cowpea. Smallholder farmers primarily suffer limited access to appro-
but it is yet to be significant in Ghana’s agriculture production land- priate inputs for production including irrigation access to enhance crop
scape. Irrigated agriculture contributes minimally (about 0.5%) to the productivity.
country's agricultural production with about 11,000 ha of land from an Several irrigation technologies and waterbodies are in Northern
identifiable irrigable area of 500,000 ha that has been developed for Ghana. However, these irrigation facilities are limited to specific loca-
public irrigation, albeit, highly underutilized as a result of institutional, tions and are not able to reach large number of farmers in remote areas
management, input and other constraints (Breisinger et al., 2008; to supplement the needs of the large smallholder farmers. There is also
Namara et al., 2011). Public investments in irrigation development has the problem of broken irrigation schemes. There is therefore the need to
declined considerably over the past two decades in Ghana until recently enhance the availability of supplementary water for crop farming
but there is an increasing skepticism about the returns to the huge among smallholder farmers to enhance food security.
public investment in irrigation development that was witnessed in the The performance of smallholder farmer’s engagement in irrigation
1970s (Namara et al., 2011). Meanwhile, private sector involvement in water management however has been mixed in Ghana, with several
irrigation development has been limited, and this has given rise to the instances of default payments, resulting in stalled public-private irri-
need to promote private sector involvement in irrigation development gation initiatives (Regassa et al., 2010). Public irrigation initiatives
and management (MoFA, 2012). closed down due to breakdown of pumping machine, high electricity
Water demand in Ghana will increase from 40,780 million m3 in cost, old and choked canals and poor service repayments. Recent gov-
2010 to 42,969 million m3 in 2020, with the main uses for consump- ernment investments in rehabilitating some existing irrigation schemes
tion, irrigation and livestock rearing. Superficial water resources are has brought some of these irrigated area back to production. None-
not adequate to meet present and future consumptive water demand theless, given the poor experiences of public-private water management
(WRC, 2012). It is important, within these challenges, that the Gov- of irrigation services in Ghana and the promotion of private sector
ernment of Ghana regards and enhances smallholder farmer access to participation, it is important to understand whether smallholder
irrigated agricultural infrastructure to increase farm productivity. farmers are willing to adopt and pay for tailored privately installed
Ghana’s Irrigation Policy (MoFA, 2012) addresses private sector irrigation schemes.
involvement in providing irrigated water for agricultural production. The Bhungroo Irrigation Technology (BIT) is a private-sector led
Similarly, private-public participation (PPP) arrangements in the agri- smallholder irrigation system installed in Northern Ghana and cham-
cultural sector also provides opportunities for private sector partnering pioned by two private entities in Ghana, Conservation Alliance in as-
of the public sector institutions in delivering irrigation services to the sociation with International Water Management Institute (IWMI). The
agricultural sector (ADB, 2013). The Irrigation Policy and private sector BIT was introduced in Ghana in 2015. Currently, three of the BIT
participation in agriculture are fully aligned with the goals of the system is sited on selected farmer’s farm as a demonstration to access
governments’ Growth and Poverty Reduction Strategy (GPRS I & II). the patronage and feasibility of the technology in farming communities
Specifically, private sector participation in irrigation was set up to in the Builsa South and West Mamprusi districts, which are noted for
improve the livelihoods and potential irrigators and farmer-based or- maize, rice, cowpea, and groundnut production, but experience severe
ganizations, ensuring that service providers are given opportunities to water-logging during the single raining season and prolonged drought
perform by opening up investments for expanded and strengthened in the dry period (Owusu et al., 2017). These demonstration sites were
crop production. This makes room for enhanced services that delivers installed in 2015. The principle behind the private irrigation service
cost-effective, high irrigation services demand driven to private and provision is a build-operate and recover cost over 30 years and transfer
public irrigation. to group of farmers whilst ensuring farm income increases for the
Ghana’s economy is dependent on agriculture and its growth is farmers.
important to total economic growth and development. Agriculture is The BIT is a water conservation technology that is able to sip large
fundamental to Ghana’s economy. Although its contribution to Gross quantities of water from the land surface and store in unsaturated layers
Domestic Product (GDP) has been declining (29.9% in 2010, 18.9% in in the soils as groundwater (Biplap, 2013; Owusu et al., 2017). The
2016 and 18.3% in 2017) it continues to be vital to economic growth as uniqueness is the ability of the technology to absorb water, store, and
285
N.A. Akrofi, et al. Agricultural Water Management 216 (2019) 284–293
Fig. 1. Location Map of the study area.
Source: Remote Sensing/GIS Lab, Department of Geography, University of Ghana, 2016
bring back the water for various uses when needed. The BIT has a ca- test were good and were within the values known for the study area.
pacity to infiltrate as much as 40,000m3 of water underground with a The BIT has the potential to provide smallholder farmers with all year
recovery potential of 8–12 ha irrigable area based on the water stored round irrigation for farming. Maize is a major food and cash crop in the
and crop water requirements (Owusu et al., 2017). The BIT is a new BIT catchment areas, and has a productivity of about 0.897 tons/ha as
irrigation technology that are employed in India by private irrigation at 2013 (Amanor-Boadu et al., 2015) compared to the average
service providers to make more farmlands available for agriculture by 5–6 tons/ha potential. In northern Ghana, about 80 per cent of the
freeing waterlogged and flooded lands of excess water during the wet production of maize is consumed by the production households (Wiredu
season. This ensures availability of all year round water for a dual et al., 2010) and is a commercial crop with the proportion of crop sold-
season in a cropping year. to-total amount produced of 20–40% (Amanor-Boadu et al., 2015). The
The BIT has been very successful in most parts of India especially in business plan estimates that five (5) smallholder farmers in the study
the Gujarat state. Owusu et al. (2017) indicates that the BIT came to area will collectively pay about $5000 once in a lifetime installation
flourish after the serious drought that affected the western Indian state cost of the BIT that has a lifespan of 20–25 years. Such costs could be as
of Gujarat in 2000, and at the same time, flash floods in the region led high as $10,000 (Paul et al., 2013, cited in Owusu et al., 2017) for a 30-
to waterlogging during the wet season. It is estimated that a-one BIT year lifespan for surface storage and basin infiltration but still estimated
sunk can free 5–7 acres (2–3 ha) of waterlogged land, and can irrigate to be the lowest of cost effective smallholder installed irrigation sys-
20–30 acres (8–12 ha) conversion from a one-time BIT installation tems. Anticipated also are that farmers in the waterlogged areas of
without maintenance and operational cost. About five (5) resource-poor Northern Ghana who are unable to produce maize on the waterlogged
farmers in India jointly pay $4930 for 20–25 years for this once in a lands during the wet season would adopt the BIT services. Other crops
lifetime installation cost. In Haryana, Gujarat, India, water for irriga- that require adequate or muddy fields will also benefit. For crops that
tion is valued at US$2.50 /ha (Cornish and Perry, 2003), and charge do not require muddy fields for cultivation, by installing the private
price ranges from $2 to $8/ha in Pakistan which covers part of op- irrigation technology on the waterlogged land to inject water under-
erational and maintenance costs (Ahmad, 2002). The underground re- ground at the same time lifting the water for periods of severe drought,
servoir can hold as much as 40 million litres of rainwater. Most bene- farm productivity could increase. In effect, the BIT will increase food
ficiary farmers using the technology in India are experiencing at least and cash crop of smallholders’ and thus enhance food security and li-
150 percent increase in farm productivity. The BIT adoption in India velihoods.
has had significant positive impacts on agribusinesses and has reduced In conserving farmer land and water for irrigation under such si-
poverty in Bhungroo beneficiary areas by at least 30 percent (Biplap, tuation of waterlogged lands, the use of Bhungroo Irrigation
2013). Technology is a promising tool. It is therefore imperative to understand
Although the BIT is new in Ghana, it is a very promising irrigation the demand-side constraints for effective adoption of the technology.
technology suited to smallholder farmers in Ghana given its success in Given the increasing threats from climate variability and change, the
India among smallholder women farmers (Owusu et al., 2017). The effectiveness of farm-level irrigation adoption to sustainably increase
borehole success was rated as good (50–65%) for two sites and much productivity and farmer incomes should be paramount. Can the
higher (75–85%) for one site (Owusu et al., 2017) and that the borehole smallholder farmer in Northern Ghana pay for the privately installed
yield and the time taken for the Bhungroo to recover after the pumping irrigation technology? This paper derives information from the
286
N.A. Akrofi, et al. Agricultural Water Management 216 (2019) 284–293
smallholder farmers in the catchment areas of the piloted BIT demon- drain the district. There is a yearly flooding caused by overflow of the
stration sites and non-piloted catchment areas in Northern Ghana, to White Volta. Occasionally, flash floods are caused by spilling of waters
address the study question. We estimate the mean amount the small- from the Bagre Dam, further up-stream in Burkina Faso. The rainfall
holder farmer is willing to pay (WTP) for the BIT, analyze the factors and the nature of the underlying rock formations determines to a large
influencing this amount, and estimate the payback period for a typical extent the ground and surface water potential for the district. The
group of farmers to pay for and own the BIT. We then provide policy present combination of heavy run-off, high evaporation and transpira-
directions. tion and low infiltration rates to recharge aquifers in some areas in the
district, contribute to water deficiencies.
2. Materials and methods Mamprugu Moagduri District, carved from West Mamprusi, forms
part of the new districts and municipalities created in the year 2012.
2.1. Study area The district is between longitudes 0°35′W and 1°45′W and Latitude
9°55′N and 10°35′N. The White Volta River and its main tributaries
The study was conducted in Northern Ghana in three districts: such as Sissili and the Kulpawn rivers are major drainage in the district.
Builsa South (Upper East Region) and West Mamprusi and Mamprugu There are rivers along large arable land, good for the cultivation of rice
Moaduri (Northern Region), where agriculture employs majority of the and other cereals. The wet season spans from May to October, peaking
workforce (see Fig. 1) and the BIT are championed. Northern Ghana from July to September, when there can be floods. The rest of the year
particularly have been designated as the ‘food basket’ of the nation is dry with average rainfall between 1000mm and 1400mm with an
given the dry condition and the vast landscape for drying cereals and average monthly temperature of about 25.5 °C. The District is within
other crops and the potential for increased mechanized farming. Pre- the savannah climatic belt. Temperatures are generally high all year
dominantly an agricultural region, Northern Ghana has one single round with the hottest month being March.
rainfall regime and lacks sufficient irrigation infrastructure to support
dry season farming and have a relatively higher poverty status. 2.2. Data and data analysis
A major reason attributed to the relatively high poverty status in
Northern Ghana is the single raining season. The dry season period We used both quantitative and qualitative data in a mixed method
tends to be distinct and extended to almost 8–9 months (Namara et al., approach. The quantitative survey using structured questionnaires tar-
2011). This implies that with rain-fed agriculture, crop farming in geted at farmers at the plot level was undertaken in Northern Ghana in
northern Ghana can be suitable for a period of 3–4 months and culti- November-December, 2015. The qualitative study involved focus group
vation for longer periods as well as double cropping of short gestation discussions using checklists and conducted separately for male, female
crops are risky without irrigation. Most smallholder farmers in and youth farmers in July 2016 to validate and triangulate farmer in-
Northern Ghana who are predominantly rain-fed farmers generally re- formation.
main idle for the greater period of the year and are mainly not involved The survey covered farmer’s socio-economic characteristics and
in other non-farm economic activities, which worsen their poverty si- farm production information in major staple crops: maize, sorghum,
tuation, or temporarily migrate to Southern Ghana to engage in other rice, vegetables and other crops. On the willingness to adopt and the
farm (cocoa) activities. Many poverty alleviation programs such as willingness to pay amount, farmers were asked several single-bounded
small-scale irrigation, Savana Agricultural Development Authority dichotomous choice questions. The question was asked if the farmer
(SADA), amongst others, are in Northern Ghana in an effort to over- will be willing to adopt (WTA) and willing to pay (WTP) for BIT water
come poverty. service charge and whether the farmer would pay to use the BIT? They
In West Mamprusi for example, the one irrigation facility the district were asked further to state the amount they will be willing to pay to use
has is a small-undeveloped irrigation facility at Nasia. Maize, cowpea the service per unit area for the cropping season if they responded
and rice doubles as staple grain and cash crops and therefore has im- positively to the WTP question. The study therefore adopted the
plications for food security and poverty reduction in the district. Contingency Valuation Method (CVM) in eliciting the WTP information
However, the districts’ perennial flooding during the wet rainy season from the smallholder farmers for the irrigation services for their stated
and dry land with inadequate access to water source during the dry preferences (Birol et al., 2006) since the market for the demand of the
season make food cultivation a challenge. Nevertheless, maize, cowpeas BIT services has only been provided through installed demonstration
and sorghum does not do well in waterlogged areas and suffers wilting units. In natural resource extraction (environmental goods), the em-
during the dry season due to lack of adequate moisture for growth pirical economic literature concerns the economic benefit of changes in
(Ajiegbe et al., 2010). Although, some variety of rice can yield heavily the levels of use of these resources. Services deriving from the small-
when grown in waterlogged lands, paddy rice production is not the holder irrigation supply was therefore not generally traded in markets
major crop for the district, but maize. in northern Ghana and information on market demand and competitive
Builsa South District is part of the nine districts in the Upper East market prices are not available to value benefits. The CVM creates a
Region, which is located South-west of the region. It is between lati- hypothetical market for such services and seeks to elicit the value that
tudes 100 15′N and 100 20′N and longitudes 10 05′W and 10 35′W. It is farmers attach to them by asking how much they would be prepared or
to the South by West Mamprusi District. Cereals such as maize, rice, WTP to obtain the benefits of such services. The CVM captures both use
sorghum, millet and pulses such as cowpea and groundnuts are main and non-use value attached to the service and more superior than the
staple crops in the district. The communities are predominantly Travel Cost and Hedonic pricing methods as these only captures used
smallholders who grow an array of rain-fed food crops. The district falls values of the services (Koundouri et al., 2016). The methods for esti-
within the Guinea Savannah zone. The district has a single maximum mating WTP models have encompassed maximum likelihood estima-
rainfall regime (unimodal) expanding over a period of 5 months with tors. This paper followed the CVM by explaining to the respondents, the
annual totals of 700-1,000mm. The climatic condition in the district is nature of the BIT, the expected cost and benefits including the avail-
tropical and semi-arid making the area relatively dry. The dry period ability of all year round water for crop cultivation upon the adoption of
extends for 7 months with a mean temperature of 25–30 °C. the technology.
The West Mamprusi District also faces a rainy season that starts in A multi-stage sampling technique was adopted for the study. The
late April with little rainfall, rising to its peak in July-August and de- multi-stage procedure was a three-stage, purposive sampling and stra-
clining sharply thereafter, and coming to a complete halt in October- tification approach. In the first stage, there was the purposive selection
November. Mean annual rainfall ranges between 950mm-1,200mm. of three districts (Builsa South, Mamprugu Moaduri and West
The White Volta and its tributaries (Sissili and the Kulpawn rivers) Mamprusi) in the Upper East and Northern regions of Ghana. The
287
N.A. Akrofi, et al. Agricultural Water Management 216 (2019) 284–293
Table 1
Socio-economic and Farm characteristics of the respondents.
(A)
Variable Northern Region (%) Upper East (%) Total
West Mamprusi Mamprugu/Moaduri Builsa South
Jagsi/Duu Kpasenkpe Kubori Wiese Gbedemblisi
Gender Treatment Treatment Control Treatment Control
Male 100.0 100.0 96.9 88.3 83.3 N=228: 94.2%
Female 0.0 0.0 3.1 11.7 16.7 N=14: 5.8%
Total 100.0 100.0 100.0 100.0 100.0 N=242: 100.0%
Age (years)
15–35 58.0 73.0 41.0 67.0 50.0 N=139: 58.0%
>35 42.0 27.0 59.0 23.0 50.0 N=101: 42.0%
Member of FBO 61.0 60.0 45.0 78.0 67.0 N=149: 62.0%
(% YES)
Farm Size Mean (Ha)
Maize 3.61 2.19 2.27 2.4 2.5 2.59
Cowpea 1.6 – – 2.22 2.8
Maize seed cultivated (All)
Certified/improved seed/purchased 4.5% 10.5% 21.1% 11.7%
Own seeds 95.5% 89.5% 78.9% 89.3%
(B)
Crops grown (all) %
Maize 75
Cowpea 6
Other crops 19
fertilizer used (all)
NPK 49.4
SA 1.1
Urea 0.6
None 48.9
Land Tenure (all)
Family land 90.5
Own land/purchased 9.5
Educational status (all)
Primary education 20
Middle/Junior Secondary School (JSS) 15
Secondary/Senior High School (SHS) 4.2
Higher than Secondary/SHS 2.1
Arabic Education 1.0
No Education 55
Non-formal Education 1.0
Tertiary 1.7
Marital status (all)
Married 92
Single 8
Widowed 0
irrigation technology are to be installed due to the drought and per- The survey covered 260 smallholder farmers. The BIT-installed
ennial flooding of communities in these districts. The project, however, communities are labelled as treatment communities in this paper. The
purposively sited the demonstration of the Bhungroo System, based on 260 sampled respondents are made up of 160 treatment and 100 control
technical feasibility on initial assessments by soil scientists and hy- farmers. The Statistical Package for Social Sciences (SPSS) software
drogeologist (Owusu et al., 2017); and also on the approval by the version 16 was employed in analyzing and describing the data by
community leadership. A purposive sample of five (5) farming com- identifying the mean, minimum and maximum amount smallholder
munities that gets flooded from the three districts were selected in the farmers are willing to pay by age, and gender, among others. The SPSS
second stage (Wiesi, Gbedembilisi, Kpasenkpe, Kubori-Yagaba and was used in describing the distribution in the estimated amount
Jagsi/Duu) guided by the location of the installed BIT. Two (2) of the smallholders are willing to pay for the service in the districts.
farming communities served as control, where there were no BIT The tobit estimation approach following from Maddala (1983) was
(Gbedemblisi (Builsa South) and Kubori-Yagaba (Mamprugu Moaduri). employed in the maximum likelihood estimation approach using Stata
The third stage involved randomly sampling smallholder farmers version 10 to analyze the factors influencing the willingness to pay
growing food crops in the communities stratified by gender. The ra- amount. The choice of the tobit model arises because whilst all the
tionale of the stratification was to ensure a representation of male and respondents were willing to adopt the BIT, some farmers were not able
female farmers. The actual selection of respondents was made difficult to indicate how much they will pay, hence predicting an outcome that
as a result of the unavailability of a comprehensive list of farmers. is censored from below. The tobit model is expressed as:
There were no population figures nor farmer lists in the communities.
The study provided its own household listing in the field with the help Y=Xi β+ εi, I = 1, 2… N
of community leaders for the sampling. where N is the number of observations, Yi is the dependent variable
288
N.A. Akrofi, et al. Agricultural Water Management 216 (2019) 284–293
(willingness to pay amount to use the BIT services per hectare per farmers indicated they practice such. The water saving technologies and
season), Xi is a vector of independent variables, β is a vector of estim- practices were crop rotation (1.2%), mixed cropping with leguminous
able parameters, and εi is a standard normal independently distributed crops (1.2%) and 0% practice for earth bunding, stone bunding, rid-
error term. The estimated parameters (β) provide indications of how a ging, mulching, and cover-cropping.
one-unit change in an independent variable (X) affect the uncensored All the respondents were willing to adopt the irrigation system
observation (Y). The estimated equation is (see Table 5 for variable under the Bhungroo. However, not all the respondents were willing to
definitions and units): pay an amount for use of the system on their fields (about 19% could
β β β β β β β β β not offer any amount, perhaps on the belief that the irrigation should beY= 0 + 1X1 + 2X2+ 3X3 + 4X4 + 5X5 + 6X6 + 7X7+ 8X8
β β free). The median amount respondents in any of the communities were+ 9X9 +…+ 17X17 + εi willing to pay to use the BIT per year per hectare of two farming season
The simple investment appraisal technique, the Payback period, is is GHS 100 (48.1% of all respondents). See Table 3. Although the mean
used in estimating the breakeven point in the irrigation project. The willingness to pay amount in Jagsi/Duu where a BIT is installed is the
Payback period, expressed in years, is the length of time required to highest, the control communities are relatively more willing to pay
recover cost of the BIT investment. The payback period is calculated by higher amounts for the BIT use.
counting the number of years it will take to recover the cash invested in The mean willingness to pay amount (WTPA) by male respondents
the BIT before it becomes the property of the “farmer” trough the ser- was about GHS 186 and about GHS 78 amongst the female respondents
vice charges: with a total mean of GHS 180. See Table 4. These responses are con-
Payback Period= Initial Investment / Cash Inflow (service charges) sistent with that from the focus group discussion (FGD), as the male
per Period respondents indicated paying much relatively higher amounts than the
The project is acceptable or recommended only if its payback period females. In the FGD, some communities were prepared to forgo 2 bags
is less than or equal to the target payback period. (100 kg per bag) of maize per the season (equivalent to GHS 150–180 at
The qualitative data collected from the focus group discussions the time of the interview) to pay for the BIT services. The mean will-
validated the information obtained from the quantitative survey of ingness to pay amount amongst the youth is GHS 164.9 with a max-
farmers. Issues discussed were on cropping patterns, the extent of their imum amount of GHS 1500 for the farming seasons. The mean will-
experiences to flooding, their willingness to pay for the irrigation ser- ingness to pay amount amongst the adult is GHS 196.9, which is
vices and the amounts they are willing to pay and finally constraints consistent with the responses from the focus group discussions.
that could inhibit their willingness to pay for the private irrigation The mean willingness to pay amount in this study of GHS180/ha/
service charge in the communities. year (US$45) is higher than what other studies report. Alemayehu
(2014) reports the mean WTP amount of smallholder farmers in Koga in
Ethiopia for improved irrigation water of US$6.78/ha/year. Alhassan
3. Results and discussions et al. (2013) use CVM for their study in the Bontanga irrigation scheme
in Northern Ghana and report a mean WTP/ha/year of GHS 16.32 (US$
3.1. Characteristics of the respondents 8.48). Zakaria et al. (2014) from the same irrigation scheme found a
mean WTP amount for improved irrigation service delivery per hectare
There are more male (94%) than female farmers in the sample (see per season to be GHS 22.92 (US$11.92) with a maximum bid of GHS
Table 1a). This suggests the domination of male farmers in the com- 100 (US$52) and the minimum, GHS 6 (US$3.12).
munities (see also, Alhassan et al., 2013 who also had a male proportion
of 94% in their sample). There are, however, more youth (defined as 3.2. Estimating the magnitudes in the factors influencing farmers’
15–35 years in Ghana’s Youth Policy document) than adults in the willingness to pay amount for BIT services and discussion
sample. The mean farm size under maize is 2.6 ha. The main crops
cultivated are maize (75%) and cowpea (6%). See Table 1b Commu- The determinants of smallholder farmers’ willingness to pay amount
nities in the Builsa South appear to engage more in maize and cowpea for BIT services are influenced by several socio-economic, institutional
relative to communities in the West Mamprusi and Mamprugu/Moaduri and community-specific factors. Table 5 presents the definitions and
who grow more of maize. About 42% of the sampled farmers have at units of measurements of the variables in the regression equation and in
least primary education (compared to 55% with no education). Almost Table 6, the tobit maximum likelihood estimates of the factors influ-
all the sampled respondents are married (92%). The use of own-stored encing the household’s willingness to pay amount, corrected for po-
seeds in crop production is pronounced (about 89% of respondents) and tential heteroscedasticity problems, are reported.
the use of inorganic fertilizers are relatively limited to the use of NPK The tobit regression result is evaluated at 1%, 5% and 10% level of
(49% of respondents). significance. The estimated log-likelihood of -935.82 shows that the
On the other hand, almost half of the farmers do not use any in- overall model is evaluated as fitting the data well and is statistically
organic fertilizers. Land for farming is mainly family owned (91%) and significant. The likelihood ratio with a chi-square value of 145.59
hired labour is predominant in field crop cultivation. Among maize (df= 20) and a p- value of 0.0000 indicates our model as a whole, fit
farmers, about 43% hire labour for planting; 54% for weeding; 40% for significantly better than an empty model.
harvesting and 55% for transporting crops to the house. Out of the explanatory variables comprising of socio-economic, in-
Smallholder farmer perceptions on water-use, water savings and the stitutional and community specificity variables, twelve variables were
advantage they take when fields are flooded suggest that in the midst of statistically significant at least at the 10% level (marital status, yield of
the availability of water, some farmers become helpless (see Table 2). maize crop, farm income, farmer taking flood advantage, member of
Whilst in general about 73% of the respondents take advantage of the FBO, family land tenure, hired labour for harvesting maize, cultivating
flooded fields to grow rice and beans, about 27% do nothing. There are maize crop, radio for farm production asset and locations (Wiese,
also spatial differences even among the communities that take ad- Jagduu, Kpasenkpe)). Thirteen variables (age, marital status, sex, se-
vantage of the flooding. In Kubori (Mamprugu/Moaduri district) and vere food shortage, farm income, access to credit, extension contact,
Gbedemblisi (Builsa district), almost half of the smallholder farmers are member of FBO, family land tenure, hired labour for harvesting, maize
unable to take any advantage of the floods. The perception of severe crop, Jagduu and radio asset) have positive effects on willingness to pay
food shortages is high (92%) in the districts and are attributed among amount for the BIT services. Seven variables (primary education, maize
others, to poor harvest due to weather conditions (95%). In terms of the yield, flood advantage, hired labour for weeding, fertilizer use (NPK),
use of water-saving technologies on their farms, only 2.5% of the Wiese and Kpasenkpe community-specificities) are negative in their
289
N.A. Akrofi, et al. Agricultural Water Management 216 (2019) 284–293
Table 2
Farmer Perceptions on Water use and water-saving technologies.
Variable Category Northern Region (%) Upper East (%) Total
West Mamprusi Mamprugu/ Builsa South
Moaduri
Jagsi/Duu Kpasenkpe Kubori Wiese Gbedemblisi
Perception on Practices of Water Saving Technologies on NO 97.2 98 93.8 100 100 N=234 97.5%
farm
Availability of Natural Water source in community for YES 100 96.2 9.8 100 66.7 N=171 70.7%
agricultural purposes
Experienced flooding on their land YES 97.2 94.1 100 93.3 96.7 N=230 96.2%
Experience with flooding and the advantage farmers Crop production 94.4 67.4 56.3 93.4 50.0 N=170 73.0%
take for production (Beans/Rice)
Do Nothing 5.6 32.6 43.7 6.7 50.0 N=63 27.0%
(Field destroyed)
Perceptions on severe food shortage YES 94.4 92.2 81.7 98.3 100 N=219 92.4%
Perceptions on difficulty in satisfying food needs the Poor harvest due to weather N=224 94.9%
previous year conditions
Food prices high N=5
2.1%
Table 3
Amount smallholder farmers are willing to pay (per year of two cropping).
Community Mean GHS Median Maximum offer GHS Minimum amount Total respondents
GHS GHS [N]
Treatment Kpasenkpe 98.9 100 300 5 50
Wiesi 105.5 100 600 10 55
Jagsi/Duu 446.7 100 1200 100 36
Control Kubori 181.4 100 3000 10 62
Gbedemblisi 128.3 100 1000 5 29
Total Respondent 112 [48.1%] 232 [100%]
GHS=Ghana Cedis (currency) 1 GHS=USD 0.25.
Table 4 maize on their plots all year round will increase the willingness to pay
Distribution in the WTPA by Gender and Age (all communities). amount for the BIT water charge by GHS 375 (US$ 94), all other
Gender Mean (GHS) Minimum (GHS) Maximum (GHS) Median (GHS) variables held constant. The implication we draw is that household
respondents that cultivate maize are more likely to participate and pay
Male 186.2 5 3000 100 for the water service charge in the communities. This could be due to
Female 78.1 5 120 80 the expected opportunity to cultivate maize twice in the year to in-
Total 180.1 5 3000 100
Age crease productivity, ceteris paribus, a consequence of freeing the land
15–35 164.9 5 1500 100 from flooding for cultivation. Being married and having access to credit
> 35 196.9 6 3000 100 could increase the willingness to pay amount by about GHS 92 (US$ 23)
and GHS 140 (US$ 35), respectively.
The maize yield was negative and statistically significant at 5%.
effects. This is not consistent with the expectation and the explanation in
The need to take advantage of the flood for cultivation variable is support that farmers with high-expected yield tend to have more in-
negatively related to the willingness to pay amount but significant at terest in paying for BIT services than farmers with less expected yield.
the 10% level. The negative sign intuitively makes sense as most Farm income was statistically significant at 5% and consistent with the
farmers abandon their fields during the flooding but one would have expected positive effect.
expected that farmers would be willing to pay more to take advantage Family land (secure land tenure) for maize farming was significant
of the BIT services. (5%) and consistent with the expected positive effect. Farmers on fa-
The study also showed that membership of farmer based organiza- mily lands are more likely to adopt the BIT and pay for water services.
tion (FBO) was significant at 5% with a positive influence on farmer’s The result contradicts the arguments of Nimoh et al. (2011) who sug-
willingness to pay amount for the BIT services. The estimated marginal gested that farmers who farm on family lands are less likely to parti-
effect of 28.8 indicate that being a member of an FBO will increase the cipate in investing in cocoa insurance.
willingness to pay amount for the BIT water charge by GHS 29 (US$ The community-specificity variables have significant effects on
7.3), all other variables held constant. This implies that farmers who are willingness to pay amounts for BIT services, although their signs in the
part of farmer-based groups learn from each other and receive other regression varies. Jagsi/Duu, which is a BIT community, was positive
trainings on acquiring improved technologies and other inputs. This is and statistically significant at 1%, consistent with the expectation in
consistent with studies conducted by Foster and Rosenzweig (1995), support of farmer responses from the enthusiasm showed in their
and Conley and Udry (2001) who report positive and significant effects willingness to pay from the focus group discussion. This is also con-
of a farmer’s membership to an agricultural association on the adoption sistent with earlier studies that reported significant but varying signs
of new agricultural technologies in Ghana. with the location variables (Alebel et al., 2009) depending on com-
Maize as a food crop variety is significant at 5 percent and positively munity specific factors.
influenced the willingness to pay amount of the BIT services. The es-
timated marginal effect of 375.04 indicate that being able to cultivate
290
N.A. Akrofi, et al. Agricultural Water Management 216 (2019) 284–293
Table 5
Definition and units of measurement of variables included in the tobit model.
Variables Variable Type Unit of Measurement A-priori sign expectationa
Dependent variable
WTP Amounts for BIT Water Service Charge
Independent variables
Age of farmer (X1) Continuous Measured in years +
Marital Status (X2) Dummy Married= 1, 0therwise, 0 +
SEX of farmer (X3) Dummy Male= 1, 0therwise, 0 +
Education level (X4) Dummy Primary=1, 0therwise, 0 +
Crop Yield (X5) Continuous Maize output/ha +
Farm Income (X6) Continuous Total income from farm activity +
Flood Advantage (X7) Dummy Take advantage= 1, 0therwise, 0 +
Food Shortage (X8) Dummy Shortage= 1, 0therwise, 0 +
Extension Contact (X9) Dummy Ext contact= 1, 0therwise, 0 +
Member of FBO (X10) Dummy Member= 1, 0therwise, 0 +
Access to Credit (X11) Dummy Access= 1, 0therwise, 0 +
Land Tenure (X12) Dummy Family land= 1, 0therwise, 0 +/-
Hired labour (X13) Dummy Yes= 1, 0therwise, 0 +
Access to Fertilizer (X14) Dummy Access= 1, 0therwise, 0 +
Food Crop Variety (X15) Dummy Maize farm=1, 0therwise, 0 +
Community-specific (X16) Dummy Community dummies +/-
Farmer Asset(X17) Dummy Asset aiding farming (eg Radio) +/-
a These represent the expected direction of the effects of the independent variables on the dependent variable.
3.3. Estimating the Payback period and discussion amount of $45/ha/year, it appears that 5 typical smallholder farmers in
the BIT community could afford to pay and own the BIT only after 22
The cost of installing one BIT, its life span and its expected benefits years. At the median amount of $25/ha/year, the payback period can
to farmers and the estimated payback period are presented in Table 7. be as long as 40 years. With the expected increased income from the
The maximum amount of flooded land that a BIT could free for farming adoption of the BIT of farmers’ growing maize, the experience in an
is approximately two (2) hectares per flooded area. On the other hand, increased average income per hectare because of efficient irrigation,
the stored water is estimated to irrigate 8 ha per season. The estimated may allow the farmers to pay to recover the cost of the equipment much
payback period is based on the estimated willingness to pay amounts. faster than the baseline estimated periods.
Given a mean farmer farm size under maize in the study areas of
2.59 ha and also given that the average farmer is willing to pay a mean
Table 6
Tobit Results for WTP Amounts for BIT Water Service Charge.
Variable Coefficient Std. Error z-Statistic Prob. Marginal Effects
C −336.899 181.361 −1.858 0.063 −384.837
Socio-econ. characteristics
Age in Years 0.526 1.222 0.431 0.667 1.595
Marital Status 92.351** 46.619 1.981 0.048 74.325
Sex 10.580 57.961 0.183 0.855 23.031
Prim Education −39.018 35.942 −1.085 0.278 19.028
Crop (Maize) Yield −0.097** 0.039 −2.519 0.012 −0.084
Farm income 0.019** 0.008 2.352 0.019 0.024
Food shortage 81.529 73.729 1.106 0.269 73.472
Flood advantage −237.398* 132.029 −1.799 0.072 −64.207
Institutional factors
Member of FBO 78.409** 34.899 2.247 0.025 28.847
Extension Contact 27.814 41.822 0.665 0.506 46.066
Access to credit 140.118 145.031 0.966 0.334 −24.201
Land tenure under Maize 141.854** 58.629 2.419 0.016 100.850
Hired labour for weeding −55.457 48.261 −1.149 0.251 −41.435
Hired labour for harvesting 121.890** 49.126 2.481 0.013 103.875
NPK use for maize −21.216 39.746 −0.534 0.594 −21.182
Food (Maize) crop variety 375.039* 141.849 2.644 0.008 193.176
Community-specificity
Wiese −89.491* 51.171 −1.749 0.080 −29.005
Kpasenkpe −95.554* 52.979 −1.804 0.071 −14.265
Jagsi/Duu 419.516*** 94.589 4.435 0.000 269.006
Asset: Radio 68.381* 40.737 1.679 0.093 226.631
Mean dependent var 190.3333 S.D, Dep variable 256.3207
Akaike info criterion 3.88146 Schwarz criteria 14.34812
Log likelihood ratio 145.59 Hannan-Quinn
Log likelihood -935.8204 criteria 14.07110
Dependent Variable: WTPA for BIT water service charge.
* Significant at 10%.
** Significant at 5%.
*** Significant at 1% levels respectively. Source: Field Data.
291
N.A. Akrofi, et al. Agricultural Water Management 216 (2019) 284–293
Table 7 crop cultivation is critical in the face of climatic changes that sub-
Payback Period for Suppliers Investments (baseline estimations). sidizing such technologies for smallholder farmers must be a priority for
Cost Item Actual development agencies.figures on Estimates for
BIT* Analysis To help improve smallholder farmer’s participation in privately in-
stalled irrigation water to increase smallholder farmer productivity in
Installation cost of BIT $4930 $5000 Ghana, farmer capacity building to enhance on-farm skills and on how
Maintenance Cost ($) 0 0
Expected Life span 30 years 25 years the BIT works to enhance their productivity and income is important.
Expected Land freed from flooding 5–7 acres (2–3 ha) 5 acres (2 ha) These notwithstanding, for sustainability of such projects, there is the
Expected Land irrigated 20–30 acres 20acres (8 ha) need for strong farmer based groups to operate and manage the BIT and
(8–12 ha) for the farmers to access farm inputs, access markets and be able to sell
No. of farmers per unit installed 5 persons 5 persons their produce at adequate prices to cover costs and to increase their
Mean willingness to pay amount/ha/ $45/ha/year (GHS 180)
year $ 25/ha/year (GHS 100) income. NGOs, extension institutions, and local government authorities
Median willingness to pay $5000 ÷ ($45×5 farmers) = 22 must intensify trainings given to farmers on the adoption of irrigation
amount/ha/year 22 years technologies as well as strengthening farmer based organizations for
Cost of installation ÷ mean cash $5000 ÷ ($25×5 farmers) = 40 sustainability and management of such technologies. Public and Private
flow/year 40 years
Payback period (mean) Sector Partnerships (PPP) to develop infrastructures at community,
Cost of installation ÷ median district and regional levels to enhance farmer productivity are neces-
cash flow/year sary.
Payback period (median)
Acknowledgments
Note:* Installation cost, expected life span and technical details of the BIT are
courtesy of Conservation Alliance.
Financial support was provided by the International Water
4. Conclusions Management Institute-Water, Lands and Ecosystems (WLE) Program,
Ghana. Project Number V7 (Volta Basin).
The main aim of this paper was to assess smallholder farmers’
willingness and ability to pay for and own a privately installed small- References
holder irrigation system, the Bhungroo Irrigation Technology (BIT) in
Northern Ghana. From our ndings, all the farmers were willing to ADB, 2013. Exploring Public-private Partnership in the Irrigation and Drainage Sector infi India - A Scoping Study. Asian Development Bank, Manila, Phillipines July. ISBN:
adopt the technology. The mean willingness to pay amount was GHS 978-92-9254-061-6.
180 ($45) with a median willingness to pay amount of GHS 100 ($25) Ahmad, B., 2002. Implications of Water Pricing in Pakistan. FAO Regional Office for the
per hectare per year. The signi cant and most in uencing factors on the Near East, Cairo, Egypt.fi fl
Ajiegbe, H.A., Mohammed, S.G., Ieosun, J.O., Ihedioha, D., 2010. Farmers’ Guide to
smallholder farmers’ willingness to pay amount identified were the Increased Productivity of Improved Legume–Cereal Cropping Systems in the
need to take advantage of the flooding for cultivation, membership of Savannas of Nigeria. IITA, Ibadan, Nigeria, pp. 104 pp.
an FBO, maize as the food crop to intensify under the irrigation scheme, Alebel, B.W., Fror, O., Eline, B., Stephan, D., 2009. The economic value of improved
wastewater irrigation: a contingent valuation study in Addis Ababa, Ethiopia. J.
and increased farm income. Farmers’ need to take advantage of the Agric. Resour. Econ. 34 (3), 428–449.
flood for cultivation is contrary to expectation, as one would have ex- Alemayehu, T., 2014. Smallholder farmers’ willingness to pay for improved irrigation
pected that farmers would be willing to pay more to take advantage of water: a contigent valuation study in koga irrigation project, Ethiopia. J. Econ.
Sustainable Dev. 5 (19), 5–15.
the BIT services. Membership of farmer based organization increases Alhassan, M., Loomis, J., Frasier, M., Davies, S., Andales, A., 2013. Estimating farmers’
willingness to pay amount and so is the expectation of those maize willingness to pay for improved irrigation: an economic study of the Bontanga irri-
farmers who expect opportunity to cultivate maize twice in the year to gation scheme in northern Ghana. J. Agric. Sci. 5 (4), 31–42.
increase productivity, a consequence of freeing the land for cultivation. Amanor-Boadu, V., Zereyesus, Y., Ross, K., 2015. Agricultural Production Survey for theNorthern Regions of Ghana: 2013-2014 Results. METSS/USAID and Kansas State
The payback period for the installed irrigation technology is about 22 University Final Report, April.
years at the mean cash flow rates from the smallholder farmers. This Biplap, K.P., 2013. Soil Water Rejuvenation Desertification and Agro Productivity
means the smallholder farmer group can only own fully the facility after Augmentation. Field Experience through Innovation, Gujarati, India.Birol, E., Karousakis, K., Koundouri, P., 2006. Using economic valuation techniques to
22 years. inform water resources management: a survey and critical appraisal of available
With the expected increased income from the adoption of the BIT techniques and an application, techniques and an application. Sci. Total Environ.
from farmers’ growing maize, the experience in an increased average 365, 105–122.
Breisinger, C., Diao, X., Thurlow, J., Yu, B., Kolavalli, S., 2008. Accelerating Growth and
income per hectare because of efficient irrigation, may allow the Structural Transformation: Ghana’s Options for Reaching Middle-Income-Country
farmers to recover the cost of the equipment much faster than the Status. IFPRI Discussion Paper 750. International Food Policy Research Institute,
baseline estimated periods. Washington, DC.
Conley, T., Udry, C., 2001. Social learning through networks: the adoption of new agri-
This study provides useful insights into smallholder farmers’ af- cultural technologies in Ghana. Am. J. Agric. Econ. 83, 668–673.
fordability of privately installed irrigation systems fashioned on build- Cornish, G.A., Perry, C.J., 2003. Water Charging in Irrigated Agriculture: Lessons From
operate and transfer in sub-Saharan Africa and how long it would take the Field. Report OD 150. HR Wallingford Ltd, Wallingford, UK.
FAO, 2008. Climate Change and Food Security: A Framework Document. Food and
to break even to own the system. It appears from the findings that the Agriculture Organization of the United Nations, Rome, Italy.
installation of the privately installed smallholder irrigation intervention FAO, 2014. Irrigation Techniques for Smallscale Farmers—Key Practices for Disaster Risk
in the various communities to enhance and improve food security may Reduction (DDR) Implementers. Rome. .
well t in the farming systems of the communities. Farmers would pay Foster, A.D., Rosenzweig, M.R., 1995. Learning by doing and learning from others: humanfi capital and technical change in agriculture. J. Polit. Econ. 103 (6), 1176–1209.
for the BIT if they have relevant information on the perceived benefits Ghana Statistical Service (GSS) (2016). 2015 Labour Force Report, December.
that helps the farmer make informed decisions (cost of technology, need Hasnip, N., Mandal, S., Morrison, J., Pradhan, P., Smith, L., 2001. Contribution of
to take advantage of ood land, need to be in Farmer Based Irrigation to Sustaining Rural Livelihoods. Knowledge and Research (KAR) Project Rfl 7875. MDS 0537. September (Funded by DFID). .
Organization’s, amongst others). However, if farmer based groups do Kamwamba-Mtethiwa, J., Weatherhead, K., Knox, J., 2016. Assessing performance of
not position well in sourcing for farm inputs and its management, there small-scale pumped irrigation systems in Sub-Saharan Africa: evidence from a sys-
will be poor productivity resulting in low incomes among the small- tematic review. Irrig. Drain. 65, 308–318.
Kassie, M., Zikhali, P., Pender, J., Kohlin, G., 2010. The economics of sustainable land
holder farmers. Availability of finance to smallholder farmers could management practices in the Ethiopian highlands. J. Agric. Econ. 61 (605), -627.
enhance their WTP for farm investments. In addition, water for food Koundouri, P., Ker Rault, P., Pergamalis, V., Skianis, V., Souliotis, I., 2016. Development
292
N.A. Akrofi, et al. Agricultural Water Management 216 (2019) 284–293
of an integrated methodology for the sustainable environmental and socio-economic 176).
management of river ecosystems. Sci. Total Environ. 540, 90–100. Paul, B.K., 2013. Soil Water Rejuvenation, Desertification Control & Agro Productivity
Maddala, G.S., 1983. Limited-dependent and qualitative variables in economics. Society Augmentation: Field Experience Through Innovation. (Accessed by Owusu, et al.
Monographs No. 3. Cambridge University Press. (2017) on May 26, 2017). https://waterlandandecosystems.wikispaces.com/file/
Mango, N., Makate, Clifton, Tamene, Lulseged, Mponela, Powell, Ndengu, Gift, 2018. view/Bhungroo.pdf.
Adoption of small-scale irrigation farming as a climate-smart agriculture practice and Regassa, E.N., Horowitz, L., Kolavalli, S., Giordana, K.B., Busia, N.D., Barry, B., Giordano,
its influence on household income in the Chinyanja Triangle, Southern Africa. Land 7 M., 2010. Typology of irrigation systems in Ghana. IWMI Working. Paper 142.
(49). https://doi.org/10.3390/land7020049. pp. 4–8.
MoFA, 2012. Ghana, Ministry of Food and Agriculture: Pilot Program Based Budget (PBB) SGER, 2018. The State of the Ghanaian Economy in 2017. Institute of Statistical, Social
for 2013-2015. November. . and Economic Research, College of Humanities. University of Ghana September.
Namara, R.E., Horowitz, L., Nyamadi, B., Barry, B., 2011. Irrigation Development in Umesh, A., Nejadhashemi, A.P., Woznicki, S.A., 2015. Climate change and Eastern Africa:
Ghana: Past Experiences, Emerging Opportunities, and Future Directions. Ghana a review of impact on major crops. Food Energy Secur. 4 (Issue 2), 110–132.
Strategy Support Program (GSSP) Working Paper # 26. IFPRI March. Wiredu, N., et al., 2010. Characterization of Maize Producing Households in the Northern
Namara, R.E., Gebregziabher, G., Giordano, M., De Fraiture, C., 2013. Small pumps and Region of Ghana. Country Report – Ghana. Council for Scientific and Industrial re-
poor farmers in Sub-Saharan Africa: an assessment of current extent of use and search/ Savannah Agricultural research Institute and International Institute of
poverty outreach. Water Intern 38, 827–839. Tropical Agriculture, Ibadan.
Nimoh, F., Bah, K., Tham-Agyekum, E.K., 2011. Investigating the interest of farmers and WRC, 2012. Drivers for Wastewater Technology Selection – Assessment of the Selection of
insurance companies in farm insurance: the case of cocoa farmers in Sekyere West Wastewater Technology by Municipalities in Relation to the Management Capability
Municipal of Ghana. J. Agric. Sci. 3 (4). and Legislative Requirements. 1 December. .
Owusu, S., Cofie, O.O., Osei-Owusu, P.K., Awotwe-Pratt, V., Mul, M.L., 2017. Adapting Zakaria, H., Abujaja, A.M., Adam, H., Nabila, A.Y., Mohammed, I., 2014. Factors affecting
Aquifer Storage and Recovery Technology to the Flood-prone Areas of Northern farmer’s willingness to pay for improved irrigation service: A case study of Bontanga
Ghana for Dry-season Irrigation. International Water Management Institute (IWMI), irrigation scheme in Northern Ghana. Int. J. Agric. Econ. Exten. 2 (1), 68–76.
Colombo, Sri Lanka. https://doi.org/10.5337/2017.214. 35p. (IWMI Working Paper
293