-
World Development 122 (2019) 597–613
Contents lists available at ScienceDirect
World Development
journal homepage: www.elsevier .com/locate /wor lddev
Waking a sleeping giant: Realizing the potential of
groundwaterin Sub-Saharan Africa
https://doi.org/10.1016/j.worlddev.2019.06.0240305-750X/� 2019
Elsevier Ltd. All rights reserved.
⇑ Corresponding author at: PO Box 77000, Port Elizabeth 6031,
South Africa.E-mail addresses: [email protected] (J. Cobbing),
[email protected]
(B. Hiller).1 We largely adopt the World Bank definition of SSA,
comprising 48 countries,
extending from Mauritania, Mali, Niger, Chad and Sudan in the
north, to South Africain the south.
2 Almost half of the SSA population lives below the
international poverty line.
3 The United Nations Office for the Coordination of
Humanitarian(UNOCHA). December 2016 (2016) regarded a drought in
2016 as the woyears, with 38 million people at risk across eastern
and southern Africa. Furtvon Uexkull (2014) and Couttenier and
Soubeyran (2014) find at least a weaklink between drought and civil
conflict/war.
4 Including urbanization, population growth, increasing per
capita consrates, increasing average temperatures, reduced rainfall
in dry months, and ifrequency and magnitude of extreme events.
Jude Cobbing a,⇑, Bradley Hiller baAfrica Earth Observatory
Network (AEON), Nelson Mandela University, South AfricabGlobal
Sustainability Institute, Anglia Ruskin University, UK
a r t i c l e i n f o a b s t r a c t
Article history:Accepted 19 June 2019
Keywords:GroundwaterSub Saharan AfricaPolitical economy
factorsIrrigationUrban and rural water securityResilience
Unlike many global regions, Sub-Saharan Africa (SSA) has yet to
undergo a groundwater revolution. Inthis paper we confirm that for
most SSA countries current groundwater use remains under 5% of
nationalsustainable yield. This is likely to be a constraint on
wider economic development and on addressing vul-nerabilities to
climate change and other shocks. Groundwater use has supported the
process of economicstructural change in other global regions; hence
we derive an empirical model for groundwater use tosupport economic
development, comprising trigger, boom and maturation phases. We
identify that thetrigger phase depends on political and economic
(‘secondary’) factors, in addition to resource character-istics.
The boom phase is described as ‘semi anarchic’, while the
maturation phase is characterized byslowing abstractions but
continued economic benefits. In SSA, we posit that the predominance
of limitingsecondary factors, coupled with a discourse of caution
and focus on the maturation phase (more appro-priate for other
regions), is constraining the use of groundwater for economic
development. We suggestthat groundwater has the potential to be a
foundational resource to support irrigated agriculture, urbanand
rural water security, and drought resilience across the region, as
it has in many other global regions.We argue that overcoming the
current barriers and costs to groundwater development can be offset
bythe benefits of regional socioeconomic development and increased
resilience. In the context of enduringpoverty and recurrent
humanitarian crises in SSA, this new synthesis of information
suggests that such anunderutilization of sustainable groundwater is
unjustifiable. Stakeholders active in the region should pri-oritize
groundwater development to help facilitate a transition to higher
value-added activities andgreater regional prosperity and
resilience, and ensure that measures are put in place for this to
be donesustainably. We conclude with some ideas to help trigger
such development in SSA.
� 2019 Elsevier Ltd. All rights reserved.
1. Introduction
Sub-Saharan Africa (SSA)1 (Fig. 1) suffers chronic
developmentchallenges2, some of which are related to the
availability of, andaccess to, water resources. Economic and/or
absolute water scarcityin SSA manifests in 315 million people
remaining without access toimproved drinking water (UNDESA, 2015),
endemic food insecurityand low levels of irrigated agriculture
(Siebert et al., 2010; Pavelicet al., 2012) and recurrent drought
events (Besada & Werner,
2015; Baro & Deubel, 2006), all of which can contribute to
humani-tarian crises, environmental migration and civil
instability3. Further-more, major climate and non-climate drivers4
are predicted to placeincreasing pressure on regional water
resources (Thompson,Berrang-Ford, & Ford, 2010; UNDP, 2012,
Gizaw & Gan, 2017).
To remedy current scarcities and meet future demands,
waterresource development in SSA should diversify away from a
pre-dominant investment focus on surface waters (Foster, Hirata,
&Howard, 2011) towards an integrated range of alternatives,
Affairsrst in 35hermore,positive
umptionncreased
http://crossmark.crossref.org/dialog/?doi=10.1016/j.worlddev.2019.06.024&domain=pdfhttps://doi.org/10.1016/j.worlddev.2019.06.024mailto:[email protected]:[email protected]://doi.org/10.1016/j.worlddev.2019.06.024http://www.sciencedirect.com/science/journal/0305750Xhttp://www.elsevier.com/locate/worlddev
-
Fig. 1. Sub-Saharan Africa, with drylands highlighted (after
Ward et al., 2016).
598 J. Cobbing, B. Hiller /World Development 122 (2019)
597–613
including groundwater5. Conditions in SSA are suitable for
potentialgroundwater contribution: rainfall volatility is the
highest of anyregion globally (Calow & MacDonald, 2009), much
of the region’spopulation lives distant from a perennial surface
water source(Kummu et al., 2011), and 40% of the region is
classified as drylands6
(Fig. 1) (Ward, Torquebiau, & Xie, 2016). Groundwater may
also offerflexibility, timeliness and resilience of use and the
potential to avoidsome of the centralized technical and
institutional burdens associ-ated with surface water development
(Pahuja et al., 2010). However,despite such inherent potential
benefits in many parts of SSA7,groundwater remains a poor sibling
of surface water, due to – atleast in part – its hidden nature,
both physically and institutionally(Wijnen et al., 2012; Braune
& Xu, 2010).
The record of groundwater development in SSA to date is one
ofnuance and contrast. On one hand, accessible and often
shallowgroundwater resources are utilized by millions of private
wellowners and communities who apply low technology and
tradi-tional methods (such as hand dug wells) for local
drinking/live-stock water and small-scale irrigation. Abstraction
of suchaccessible groundwater resources has doubled in the last two
dec-ades across much of SSA (Pavelic et al., 2012) and an estimated
50%(Carter & Parker, 2009) to 75% (Goulden et al., 2009) of the
regionalpopulous now relies on these resources. In short, such
resourcesare used widely, but often not intensively, at a regional
scale. How-ever, they can be prone to local mismanagement
(including over-exploitation and pollution), their seasonal
fluctuations can bevulnerable to drought events and their
utilization is often not opti-mized. On the other hand, less
accessible and often deeper ground-water resources remain poorly
understood and largely unutilizedat scale. Such resources require
greater technology and financingto access and pump despite being
typically more resilient, reliable
5 Groundwater is water contained in interconnected pores in the
saturated sub-surface, accessible via wells and boreholes and which
can emerge naturally at thesurface as springs, baseflow to rivers,
and other discharges.
6 Defined based on the Aridity Index (Trabucco & Zomer,
2014) and including arid,semi-arid and dry sub-humid areas. Climate
change is expected to exacerbate thearidity of drylands in SSA
(Haensler, Hagemann, & Jacob, 2011).
7 Including large storage volumes, lower development costs,
lower treatmentrequirements, lower evaporation losses, greater
resilience (buffer) to climate varia-tions / change and extreme
events, protection from pollution, local availability
andincremental development potential.
and able to support greater yields. As an example, few large
urbanwater utilities in the region favor groundwater as a
permanentsource of supply (Foster, Hirata, Misra, & Garduno,
2010).
The relative absence of strategic and large-scale formal
invest-ment in SSA’s groundwater, and of formal policy and
institutionalsupport (Braune & Xu, 2010), lies in contrast to
many globalregions (such as South Asia, People’s Republic of China
(PRC), Mid-dle East and west coast USA), where groundwater has
underpinnedimpressive development outcomes (Ward et al., 2016).
We believe that multiple demand- and supply-side factors
areconverging to make now the right time to address
increasinglyurgent humanitarian, socioeconomic and climate
imperatives viaan exploration of sustainable groundwater
development potential.Up until now, there has been limited, and at
times contradictory,information on the potential and accessibility
of groundwaterresources across SSA. For example, warnings about
SSA’s limitedgroundwater potential (e.g. Edmunds, 2012) contrast
with reportsof ‘megawatersheds’ and substantial underexplored
resources (e.g.Bisson & Lehr, 2004). There has also been little
research into factorsresponsible for the current levels of
under-development. We aimto understand some of the conditions which
could help promptgreater groundwater utilization at scale,
beginning with an explo-ration of the availability and
accessibility of SSA groundwaterresources as a basis before moving
forward.
2. The groundwater resources of Sub-Saharan Africa
2.1. Assessment
The groundwater resources of SSA are among the least under-stood
globally (Tuinhof et al., 2011; Villholth, 2013), with a
fewwell-studied aquifers contrasted with large areas where
detailedknowledge of local conditions remains poor. The sheer scale
ofSSA presents both opportunity and challenge: on one hand it is
agrowing regional economy with a diverse population of
approxi-mately one billion; on the other, the SSA drylands alone
areroughly the size of Europe and India combined, which
increasesthe complexity of assessing and utilizing regional
groundwaterresources. Yet, SSA contains a small fraction of Europe
and India’sborehole records and hydrogeological studies for
characterizingits groundwater resources8. Continental scale
assessments ofgroundwater resources exist but rely on remotely
sensed data com-bined with global model outputs (e.g. Döll &
Flörke, 2005; Siebertet al., 2010). Likewise, contemporary
(post-1980) hydrogeologicalmaps are often available at national or
regional scale (using geolog-ical data of similar scale and
supplemented by hydrogeological dataif available, e.g. MacDonald,
Bonsor, Dochartaigh, & Taylor, 2012;SADC, 2009).
However, groundwater is a resource that depends on local
geol-ogy, topography and climate, and its potential is spatially
highlyvariable. Local hydrogeological investigation is vital to
confirmparameters such as sustainable yield, water quality and
currentusers.
Within the region, there is also high variability in the
availabil-ity of hydrogeological data (and data products such as
maps andanalyses). For example, in some countries (such as the
DemocraticRepublic of Congo) routine monitoring of groundwater
levels and
8 Limited knowledge and poor data coverage sometimes mean
Africa’s geology(and groundwater resources) is over-simplified. A
recent book on world groundwaterstates: ‘‘Sub-Saharan Africa’s
geology consists of four basic rock types: Precambrianbasement
rocks, consolidated sedimentary rocks, unconsolidated sediments,
andvolcanic rocks” (Alley & Alley, 2016:72). Foster et al.
(2012) classify Africangroundwater into three simple categories.
Such assessments can provide a usefuloverview but may also obscure
the fact that Africa’s geology and groundwaterresources are at
least as diverse and complex as those of other continents.
-
J. Cobbing, B. Hiller /World Development 122 (2019) 597–613
599
quality is rare, and in some others (such as Zimbabwe), it is
indecline (Robins, Davies, Hankin, & Sauer, 2002). Where data
isavailable, it can be difficult to access (Cobbing & Davies,
2011). Avicious cycle often exists: poor groundwater data
contributes tolow levels of groundwater development, which in turn
produceslittle data (Robins et al., 2002).
Recent water resources studies (both surface water and
ground-water) show that, on average, the region withdraws about 121
bil-lion cubic meters (BCM) per annum9, which corresponds to
lessthan a quarter of the region’s total internally renewable
waterresources10. This is about 170 m3 per capita, less than a
third ofthe world average of 600 m3 per capita (Ward et al., 2016).
In com-parison, India currently withdraws approximately 761 BCM of
water,or about 60 times more per unit area (FAO, 2016). Whilst
total with-drawals in SSA for agriculture doubled between 1960 and
2008, theregional population has more than tripled during the same
time(from about 229 million people to 830 million people, according
toWorld Bank data) and neither (irrigated) food production nor
watersupply and sanitation services are keeping up with
populationgrowth. However, the constraint is not an absolute lack
of water.
The total volume of groundwater in storage in Africa
(includingnon-renewable groundwater) has been estimated at 0.66
mil-lion km3 (i.e. 660,000 BCM) with a range in uncertainty of
between0.36 and 1.75 million km3 (MacDonald et al., 2012). This is
not thesame as the total renewable or practically accessible
volume, but itmeans that groundwater is by far the largest stock of
fresh wateron the continent. In contrast, freshwater storage in
African lakesis estimated at about 0.03 million km3 (MacDonald et
al., 2012).Despite this, international organizations sometimes
imply thatgroundwater is considerably smaller in volume compared
withsurface water. For example, a report on African water
resourcesby the United Nations Economic Commission on Africa
(UNECA,2004:2) states: ‘‘The continent has large rivers, big lakes,
vastwater lands and limited, but widespread, ground
waterresources”11.
The total volume of theoretically renewable groundwater in SSAis
calculated to be about 1400 cubic kilometers per year (km3/yr)(FAO,
2016; Margat & van der Gun, 2013). Of this volume, it is
esti-mated that only about 20 km3/yr (±0.14%) is being abstracted.
Thecomparable figures for India are about 432 km3/yr total
(Frenken,2011), with about 251 km3/yr of this groundwater being
with-drawn (±58%)12. In SSA, average per capita annual abstraction
ofgroundwater is 28 m3/yr, compared to 208 m3/yr for India.
WithinSSA, per-capita volume withdrawn in the drylands is higher
thanthe regional average, but still well below sustainable levels
and thefigure for India (Fig. 2).
9 Approximately 87% (105 BCM) of total SSA water withdrawal is
used foragriculture, with the remainder used for domestic (10%) and
industrial (3%) uses(Ward et al., 2016). Worldwide, agriculture
accounts for approximately 70% for waterresource consumption.10
Internal renewable water resources is that part of the water
resource (surfacewater and groundwater) generated from endogenous
precipitation. We distinguishbetween ‘renewable’ and
‘non-renewable’ groundwater resources. Renewablegroundwater is
replenished under current climatic conditions, often seasonally
orevery few years. Non-renewable groundwater, sometimes called
‘fossil groundwater’,is groundwater that receives little or no
modern recharge. Although non-renewablegroundwater cannot be used
indefinitely, it can sometimes be used for decades orlonger. For
example, Libya’s ‘great man-made river’ project uses fossil
groundwaterfrom the Nubian Sandstone aquifer to provide water to
coastal cities, in theknowledge that this resource is finite (Voss
& Soliman, 2014), analogous to the miningof a mineral
resource.11 Rebouças (1999:235) reported similar misconceptions of
groundwater resourcesin Latin America, describing it as
‘‘frequently approached as something mystic ormetaphysical by the
public in general, and even by professionals”.12 These figures for
groundwater include overlap with surface water resources
(i.e.pumping groundwater would reduce surface water flows), but do
not include non-renewable groundwater resources, which are
substantial in SSA.
While average regional figures mask significant variationbetween
countries (Table 1), it is also evident that only seven of43 (16%)
SSA countries are currently using more than 10% of theirrenewable
groundwater resources, and 26 of 43 (60%) countriesuse less than 5%
of their renewable groundwater resources. Onlytwo countries
(Djibouti and Mauritania) exceed their renewablegroundwater limits.
Only Mauritania uses more groundwater percapita than India (220
m3/yr) – the next three heaviest per capitausers (Botswana, Namibia
and South Africa) all use less than onethird of that.
Based on a synthesis of available information by the
BritishGeological Survey (MacDonald et al., 2012), Fig. 3a links
spatialvariation with estimates of depth to groundwater across
SSA.Fig. 3a illustrates that many areas contain groundwater
resourcesat depths of less than 50 or 100 m, where development
could con-tribute to resilience and economic opportunities. While
some ofthese groundwater depths may be outside the range of
accessibil-ity for many individual smallholder farmers and
household watersupplies, they are within range of more
collaborative developmentfor similar purposes. Furthermore, linking
spatial and geologicalvariation with estimates of recharge to
groundwater, work by theBundesanstalt für Geowissenschaften und
Rohstoffe (BGR) andUNESCO (BGR & UNESCO, 2008) – as presented
in Fig. 3b – synthe-sizes available information on groundwater
recharge and aquifertype at regional scale across SSA. Both Fig. 3a
and b suggest thatin many parts of SSA, including dryland areas,
groundwater is oftenavailable (and recharged) where it is most
needed.
2.2. Potential for sustainable use
Hence, for 41 of 43 countries in SSA where there are estimatesof
renewable groundwater resources, there is significant
unutilizedpotential. This is despite well-documented benefits of
groundwateracross key sectors, such as: (i) irrigated agriculture,
(ii) urban andrural water security, and (iii) drought resilience,
each of which isbriefly discussed below.
Agricultural productivity is generally a strong driver of
struc-tural change (McArthur & McCord, 2017). More specifically
forSSA, authors such as Van Loon and Van Lanen (2003) andWijnen,
Barghouti, Cobbing, Hiller, and Torquebiau (2018) high-light the
salience of agricultural water development for economicgrowth and
poverty eradication in SSA. Growth in GDP in SSAdue to agriculture
is estimated to be multiple (�11) times moreeffective in reducing
poverty than growth from other sectors(Shah, Verma, & Pavelic,
2013). It is estimated that irrigated agri-culture could boost SSA
agricultural yields by more than 50%, sup-port diversification to
higher-value crops13 and achieve beneficialsupply-chain effects
which help to catalyze local virtuous circles ofeconomic growth and
development14. Some experts contend thatincreasing agricultural
productivity, including specialist horticulture,is part of a new
pattern of structural change in SSA, different to
themanufacturing-led boom in South Asia (Page, 2018). The total
area ofcultivated land in SSA is about 237 million hectares (MHa),
of whichonly about six to seven MHa, or approximately �3%, is
thought to beirrigated (the comparable figure for irrigated land
for South Asia is42% (Ward et al., 2016)). Despite comprising such
a small proportionof total agricultural land in SSA, Shah et al.
(2013) found that irri-gated land currently contributes 25% of the
region’s agricultural out-put. Only eleven SSA countries have
irrigated areas larger than0.1 MHa, and only two (Madagascar and
South Africa) have irrigatedareas larger than one MHa (Table 2). In
comparison, the US state of
13 Supported by a series of studies by FAO, IFPRI, IWMI,
Villholth 2013 and severalregional and national research and
development agencies.14 Results from a groundwater-based irrigation
study by Lejars, Daoudi, and Amichi(2017) in North Africa.
-
Fig. 2. Comparison between renewable groundwater and groundwater
abstracted: for SSA, SSA drylands (countries with >40% land area
classified as dry sub-humid or drier)and India (Source: FAO,
2016).
600 J. Cobbing, B. Hiller /World Development 122 (2019)
597–613
Nebraska alone has about 3.4 MHa under irrigation (much of
thisfrom groundwater), about the same as the top three SSA
countriescombined (Johnson, Thompson, Giri, & Van NewKirk,
2011). Criti-cally, the percentage of cultivated land irrigated by
groundwater inSSA is lower still – only about 1% of cultivated
land, or about twoMHa. This is despite multiple recent studies
(Xie, You, Wielgosz, &Ringler, 2014; Villholth, 2013; Pavelic,
Villholth, & Verma, 2013;Wijnen et al., 2018) confirming
significant groundwater irrigationpotential in SSA. For example,
Pavelic et al. (2013) estimated that a120-fold increase in
groundwater-supported irrigable area was pos-sible across a sample
of 13 SSA countries (taking into account waterdemands from other
sectors, including the environment). Sub-regions of SSA, such as
the Sahel, exhibit great potential for irrigationfrom groundwater
to support higher value agricultural and horticul-tural products
(Wijnen et al., 2018).
We argue that comments made above regarding the availabilityof
groundwater for irrigation also often hold true for potable
watersupplies in SSA, since the volumes and quality required can be
sim-ilar. The potential of groundwater for improving urban and
ruralwater security in SSA – both for municipal water-supply
systemsand for direct in situ water supply – has long been
recognized(Foster et al., 2011, 2012), along with its potential for
local over-abstraction and contamination (Braune & Xu, 2010;
Nlend et al.,2018; Tuinhof, Foster, van Steenbergen, Talbi, &
Wishart, 2011).While some large SSA cities are partially or mostly
dependent ongroundwater (e.g. Abidjan, Addis Ababa, Dar Es Salaam,
Dodoma,Lusaka, Pretoria, Windhoek), the resource is still not
widely seenas a strategic asset by most SSA water utilities
(Foster, vanSteenbergen, Zuleta, & Garduno, 2010). However, as
witnessed in2018 in the South African city of Cape Town, many
growing urbansettlements in SSA are outgrowing their increasingly
vulnerablesurface water sources and face the possibility of ‘Day
Zero’, whentaps run dry (Wijnen et al., 2018). With half of the SSA
cities in2035 not yet built and the secondary15 cities of today
expected tobecome the mega cities of tomorrow (Jacobsen, Webster,
&Vairavamoorthy, 2013), a quadrupling in water supply service
rateswill be required just to maintain current levels16 (let alone
to
15 Currently with populations under 1 million people.16 A recent
World Bank report confirms that although water delivery services
inseveral SSA urban centers have improved, the portion of
households connected toclean water supplies have declined by more
than 15 percent (IEG, 2017).
increase proportional coverage), much of which will need to
comefrom groundwater (Banerjee & Morella, 2011). UPGro
(2017:64) sug-gests that wherever high-yielding aquifers exist
within 30 km of anurban demand center in SSA, their managed and
staged developmentby water utilities ‘‘can significantly increase
water-supply security”.Pavelic et al. (2012) found that drinking
water supplies sourced fromgroundwater in rural/small towns in SSA
can be high in some coun-tries (e.g. 92 percent in Niger, 70
percent in Nigeria). Indeed, Tuinhofet al. (2011:21) found that
many SSA towns ‘‘depend on groundwa-ter for their municipal
water-supply over a wide range of hydrogeo-logical settings”. SSA
does contain some world-leading examples ofgroundwater innovation,
such as in Windhoek, the capital of Namib-ia, which operates a
managed aquifer recharge (MAR) scheme wheresurplus water is stored
underground in aquifers during times ofplenty, to be extracted as
high-value water during droughts(Murray, van der Merwe, Peters,
& Louw, 2016). The phased capitalcost of groundwater
development may permit greater water sourceand supply coverage to
poorer consumers (UPGro, 2017) and MARschemes could be coordinated
with improved sanitation/wastewatermanagement (Lapworth et al.,
2017; Adelana, Tamiru, Nkhuwa,Tindimugaya, & Oga, 2008).
UNISDR (2009) report that droughts in SSA account for less
than20% of natural disasters but account for over 80% of the
affectedpopulation. Shiferaw et al. (2014:67) describe the
economic, socialand environmental impacts of drought in SSA as
‘‘huge”, withnational costs and losses incurred threatening to
‘‘undermine thewider economic and development gains made in the
last few dec-ades in the region”. In recent history, SSA has
suffered numerousdroughts, most notably across the Sahel, southern
Africa and theHorn of Africa (Sheffield et al., 2014). For example,
between 1970and 2017, more than 30 countries in SSA experienced at
least eightdroughts, with sub-regions, such as the Horn of Africa,
experienc-ing frequently recurring and widespread events (Burney et
al.,2013; Pavelic et al., 2012) and countries, such as Kenya,
experienc-ing associated economic costs of up to 1% of annual
GDP(Demombynes & Kiringai, 2011; Stockholm
EnvironmentInstitute, 2009). However, there has typically been
little sustainedinterest in drought mitigation measures in SSA
(Benson & Clay,1998) and the current paradigm (globally and in
SSA), is focusedpredominantly on disaster response, rather than
prevention andpreparedness. For example, only four percent of
global humanitar-ian aid goes towards prevention and preparedness,
despite it being
-
Table 1Groundwater resources and groundwater use in SSA
(supplementary material1).
SSA country Renewablegroundwater (km3/yr)
Groundwaterabstraction (km3/yr)
Proportion of renewablegroundwater that is used (%)
Groundwater abstractionper capita (m3/yr)
Angola 58 0.41 0.7 21.5Benin 1.8 0.17 9.4 19.2Botswana 1.7 0.14
8.2 69.8Burkina Faso 9.5 0.39 4.1 23.7Burundi 7.47 0.16 2.1
19.1Cameroon 100 0.37 0.4 18.9Central African Rep. 56 0.08 0.1
18.2Chad 11.5 0.45 3.9 40.1Congo Dem. Rep. 421 1.22 0.3 18.5Congo
Brazzaville 122 0.03 0.0 7.4Cote d’Ivoire 37.84 0.37 1.0
18.7Djibouti 0.015 0.02 133.3 22.5Eq. Guinea 10 0.01 0.1
14.3Eritrea 0.5 0.09 18.0 17.1Ethiopia 20 1.49 7.5 18.0Gabon 62
0.03 0.0 19.9Gambia 0.5 0.03 6.0 17.4Ghana 26.3 0.51 1.9 20.9Guinea
38 0.09 0.2 9.0Guinea-Bissau 14 0.03 0.2 19.8Kenya 3.5 0.62 17.7
15.3Lesotho 0.5 0.02 4.0 9.2Liberia 45 0.07 0.2 17.5Madagascar 55
0.38 0.7 18.3Malawi 2.5 0.28 11.2 18.8Mali 20 0.34 1.7
22.1Mauritania 0.3 0.76 253.3 219.7Mozambique 17 0.44 2.6
18.8Namibia 2.1 0.15 7.1 65.7Niger 2.5 0.14 5.6 9.0Nigeria 87 3.44
4.0 21.7Rwanda 7 0.2 2.9 18.8Senegal 3.5 0.74 21.1 59.5Sierra Leone
25 0.11 0.4 18.7Somalia 3.3 0.28 8.5 30.0South Africa 4.8 3.14 65.4
62.6Sudan and S. Sudan 7 0.59 8.4 13.3Swaziland 0.66 0.04 6.1
33.7Tanzania 30 0.98 3.3 21.9Togo 5.7 0.11 1.9 18.2Uganda 29 0.62
2.1 18.5Zambia 47 0.3 0.6 22.9Zimbabwe 6 0.43 7.2 34.2
1 Figures after Margat and van der Gun (2013), Döll and Fiedler
(2008), Siebert et al. (2010), and FAO (2016 – AQUASTAT database).
Continental models calculategroundwater recharge using available
rainfall data, and take topography, soil type and geology into
account. They use a monthly time-step function, often with a
spatialresolution of around 0.5� by 0.5�. Inherent uncertainty in
the data is acknowledged, but the point remains that large
increases in sustainable use of groundwater, on average,are
possible for most SSA countries.
J. Cobbing, B. Hiller /World Development 122 (2019) 597–613
601
more cost-effective, helpful in mitigating the worst effects
ofhumanitarian emergencies, and protecting development gains
invulnerable communities (Kellet & Sweeney, 2011; Cabot
Venton,Fitzgibbon, Shitarek, Coulter, & Dooley, 2012). While
water man-agement cannot prevent drought17, ex-ante management
strategies(Shiferaw et al., 2014), such as strategic development of
groundwa-ter resources could help mitigate both the acute and
chronic socialand economic impacts for many SSA countries. We
propose develop-ing a strategic network of deep groundwater bores
in drought ‘hot-spots’ to improve resilience for future emergency
events,particularly in the context of climate variability and
change.
Finally, environmental sustainability of groundwater resourcesis
a concern (Villholth, 2013; Braune & Xu, 2010). Locally,
aquifersmay be over-exploited, with a range of disbenefits,
reinforcing theneed for better local data, and for local-level
investigations prior togroundwater development. Fig. 4 summarizes
work by Altchenkoand Villholth (2015), who calculated potential
irrigable areas per
17 Drought is a natural hazard caused by large scale climate
variability (Van Loon &Van Lanen, 2013; Sadoff, 2016).
SSA country, whilst also reserving a proportion of
groundwater(30%, 50% or 70%) for environmental purposes. It is
evident that,for nearly all SSA countries, groundwater abstraction
could begreatly increased on average, even with stringent
environmentalprovisions.
3. An empirical model of groundwater development
Confirmation that most countries in SSA use less than 5% oftheir
renewable groundwater, coupled with the knowledge thatgroundwater
contributes to structural development in many otherglobal regions,
poses the question of why SSA’s groundwater devel-opment trajectory
should be any different. An empirical model(Fig. 5) helps to
understand the process of groundwater develop-ment, with the goal
to help facilitate progress in SSA.
The model is built around three key phases: (i) Trigger
phase:where multiple factors combine to reach a tipping point to
‘trigger’intensive groundwater utilization; (ii) Boom phase: a
rapid scalingup of increasingly sophisticated groundwater
utilization (andassociated economic development); and (iii)
Maturation phase:where (volumetric) utilization peaks before either
plateauing or
-
Fig. 3. (a) Depth to groundwater across SSA (after MacDonald et
al., 2012) and (b) Groundwater recharge across SSA (after BGR &
UNESCO, 2008).
Table 2Cultivated land and irrigation in India and Sub-Saharan
Africa (FAO, 2016 and World Bank data).
Pop.(billions)
Total area of cultivableland (Mha)
Area of land that iscultivated (Mha)
Area of land that isirrigated (Mha)
Cultivated land that isirrigated (%)
Surface waterirrigation (Mha)
Ground- waterirrigation (Mha)
SSA 0.9 400 237 7 3% 5 2India 1.3 173 160 66.3 39% 24.5 41.8
Fig. 4. Potential areas in SSA irrigable with groundwater,
whilst reserving a percentage of groundwater (30%, 50% or 70%) for
environmental requirements (Altchenko &Villholth, 2015).
18 Arguably, select other global regions, such as parts of South
America, also havenot undergone significant groundwater
development, although they have advancedfurther than SSA (Rebouças,
1999; International Groundwater Resources AssessmentCentre (IGRAC),
2014).
602 J. Cobbing, B. Hiller /World Development 122 (2019)
597–613
moderating, conservation and/or remediation measures
becomeinfluential factors, and economic advancement continues via
inno-vation and efficiency gains.
The model is based on empirical evidence from various
regionsglobally that have experienced the
trigger-boom-maturationphases (see Fig 6a and B in the United
States and the PRC respec-tively, and Fig. 7 in India, as
examples). Authors such as Shah(2009) have reported similar
empirical trends in local groundwaterdevelopment.
However, the most critical component of Fig. 5 is the
trajectoryof groundwater development in SSA, which, in general, has
notexperienced the groundwater trigger-boom-maturation
phasingevident in many other global regions18. Below we analyze
each of
-
Fig. 5. Empirical model of groundwater development – trigger,
boom and maturation phases.
Fig. 6. a – United States national groundwater withdrawals
(billions of gallons per day) (United States Geological Survey
(USGS). December 2017, 2017), and b – PRC nationalgroundwater
withdrawals (billions of cubic meters per year).
J. Cobbing, B. Hiller /World Development 122 (2019) 597–613
603
the three phases of groundwater development, with a focus on
theprevailing conditions in SSA relative to other global regions
and onidentifying potential factors to trigger SSA’s groundwater
boom.
The observed correlation between more intensive groundwateruse
and economic or structural development is not evidence for acausal
relationship. Nevertheless, we demonstrate the potentialof untapped
groundwater resources in SSA to impact positivelyon at least three
sectors (irrigated agriculture, urban and ruralwater security, and
drought resilience), all of which are likely tobe important
components of the required structural economicchange in SSA. The
secondary or political-economy factors whichwe outline in the
section below influence the relationship betweengroundwater use and
economic development in complex ways.Different political-economy
considerations apply to the three sec-tors, and the presence of a
groundwater resource is only one of sev-eral necessary attributes.
The successful utilization of groundwaterin one area does not
necessarily imply that the conditions will be
right for success in others. Instead, we use the empirical
modelto extract principles of practice associated with different
phasesof groundwater development and draw upon examples from
differ-ent global regions to help understand factors contributing
to phasechanges. We acknowledge that each region’s experience
comprisesa unique combination of physical, social, political and
economicconditions (taken at specific junctures) and is
illustrative and infor-mative rather than directly comparable with
other regions.
3.1. The trigger phase
While groundwater resource availability is determined by
phys-ical factors, the dynamics and sustainability of groundwater
useare determined by socioeconomic and institutional factors(Pahuja
et al., 2010). As we described in Section 2, the
theoreticalavailability of renewable groundwater in SSA does not
appear tobe the main reason for the region’s low groundwater
utilization
-
Fig. 7. Tube wells are increasingly the main source for
irrigation in India (irrigated area, ‘000 ha) (after Indian
Ministry of Agriculture and PRS Legislative Research.
2014,2014).
604 J. Cobbing, B. Hiller /World Development 122 (2019)
597–613
(Chokkakula & Giordano, 2013). Similarly, Pahuja, Tovey,
Foster,and Garduno (2010:79) found in India that there is ‘‘almost
no cor-relation between groundwater availability and groundwater
use”.Whilst poor local groundwater availability clearly
precludesgroundwater-based development, once this threshold is
passed itappears that factors other than the physical availability
of ground-water control the ‘triggering’ of development.
Several authors have identified and discussed ‘secondary’
orpolitical economy factors that combine in various ways to
con-strain groundwater exploration, drilling, borehole installation
andmechanized pumping (e.g. Villholth, 2013; Chokkakula
&Giordano, 2013, Shah et al., 2013; van Koppen, 2003; Wijnenet
al., 2018; Foster et al., 2012). Secondary factors include a
diverserange of issues (discussed below) and may be more important
tooverall water supply viability and sustainability than
physicalgroundwater availability. DeFries and Nagendra (2017)
concludedthat groundwater physical conditions and secondary factors
inter-act in any groundwater economy to give rise to complex
or‘wicked’ problems with non-linear components and feedback
atvarious levels.
In SSA, it is likely that secondary factors are collectively the
lar-gest obstacle to a step change in groundwater use (Shah et
al.,2013; Chokkakula & Giordano, 2013; Tuinhof et al., 2011).
Never-theless, and as their name implies, such factors are often
seen assubordinate or ‘secondary’ to primary groundwater
availability,which is presumed to be the controlling variable. In
other globalregions, some critical secondary factors contributing
to triggeringof groundwater development included irrigation
technology inno-vations in the USA (Ashworth, 2006), energy
subsidies in SouthAsia (Shah, 2009), and macro policy in the PRC
(de Marsily &Abarca-del-Rio, 2016). However, comparatively
little is knownabout secondary factors generally, despite their
apparent impor-tance in influencing groundwater development. The
interactionof possible secondary factors that may play a critical
role in trigger-ing groundwater development in SSA is explored
below, drawingon a relatively scant base of available
information.
3.2. Secondary factors
Empirical observation suggests that hydrogeological data doesnot
automatically lead to better or more widespread
groundwaterdevelopment; in fact, it is often widespread groundwater
develop-
ment that leads to better data. Increased groundwater use in
turnhelps shape the wider political-economy, such as subsidies
andother policy structures for agricultural products in Midwest
USA(Ashworth, 2006) or the nature of electricity pricing in
India(Shah, 2009; Shah, Verma, & Durga, 2014). These outcomes
wouldhave been difficult to foresee prior to widespread
groundwaterdevelopment since they are the products of iterative
political inter-actions in a particular social context. These
political-economydevelopments have still wider backward and forward
linkageswith further implications for the pace and characteristics
of devel-opment. The nature and trajectory of economic development
basedon groundwater development is therefore likely to be a
complexsystem rather than an arrangement of predictable
cause-and-effect linkages.
Secondary factors that initially appear relatively minor
canderail a groundwater project as effectively as the physical
absenceof groundwater. For example, lack of access to credit
facilities orcollateral may make drilling even shallow boreholes
impossible(Colenbrander & van Koppen, 2013). Alternatively,
more affordableenergy can make pumping from greater depths viable,
with thesame overall outcome on a groundwater project as a
shallowerwater table. Hence, a favorable political economy can
bring anexisting physical groundwater resource within easier reach,
whilsta political economy indifferent to groundwater development
canplace a viable resource out of bounds.
Here, we draw on the limited literature on secondary or
politi-cal economy factors which may affect the triggering (and
ongoingnature) of groundwater development, broadly categorizing
themas: (i) material, financial and technical factors, and (ii)
social, legaland institutional factors. The factors we identify are
not exhaustivebut serve to illustrate the types of secondary
factors that could beinfluential in SSA.
3.2.1. Material, financial and technical factorsEnergy
availability and price: determine pumping (and treat-
ment) potential of groundwater, via electricity and
alternativeenergy sources, such as diesel or kerosene. According to
WorldBank data, less than 25% of SSA’s rural population has access
toelectricity in all but eight countries. Penetration of energy
servicesand price were critical factors in triggering groundwater
irrigationin India (Shah et al., 2014) and indeed, groundwater
irrigationconsumes up to 31 percent of India’s electricity, with
government
-
J. Cobbing, B. Hiller /World Development 122 (2019) 597–613
605
subsidies helping shield farmers from the full cost of
pumping(Pahuja et al., 2010). Authors such as Shah et al. (2013)
state thatmotorized irrigation is needed to truly transform
agricultural out-put, as evidenced from a study of smallholders in
nine SSA coun-tries where irrigation by motorized pumps more than
doubledoutput. The advent of increasingly affordable solar pumping
sys-tems presents intriguing possibilities in accessing
groundwaterresources, as well as new challenges for management
(Shahet al., 2014; World Bank, 2015).
Cost and availability of drilling equipment, pumps, spare parts,
andrelated equipment: tend to be high in SSA, relative to other
globalregions (Shah et al., 2013). Real drilling costs may be
higher thanequipment costs imply, due to large distances, rugged
conditions,need for prompt return on investment, lack of
competition,and/or poor regulation. The full benefits of cheaper
foreign pumpsand related equipment have yet to be realized in many
parts ofSSA, partly due to import restrictions (Colenbrander &
vanKoppen, 2013). Pahuja et al. (2010) identified the increase in
avail-ability of modular well and pump technologies (coupled
withaccess to credit) as an important contributing factor to
triggeringIndia’s groundwater revolution. Similarly, Shah et al.
(2014:10)point out that ‘‘in effect, electric and diesel water
extraction mech-anisms have become the engine of India’s
agricultural and ruraleconomy”. These costs link to other factors:
for example, a studyin Zambia (Colenbrander & van Koppen, 2013)
found a lack ofinformation and spare parts, coupled with transport
costs, can con-stitute a third of delivered pump costs.
Technical data and information: extends beyond
hydrogeologicaldata to include information on climate and
hydrometeorologicalsystems, technical options for equipment such as
pumps, clearrules on import and export procedures, specialist
extension ser-vices, local technical advice, regulatory regimes and
incentives,social/cultural norms, prices for commodities and
existing ground-water usage/s. Small improvements in (for example)
local radioprogramming, cellphone coverage, or internet access may
have dis-proportionately large downstream benefits in facilitating
access totechnical data and information for groundwater
development.Information and communication technology extension is
stillemerging across much of SSA and is tied to other factors such
aselectricity availability, transport networks and private sector
activ-ity and investment.
Access to capital and credit: is important for all
groundwaterinstallations, particularly when larger and deeper
groundwaterinstallations are envisaged (e.g. city supplies). Access
to credit,banking systems, currency stability and interest rates
all impacton groundwater development feasibility and in many parts
ofSSA, access to simple banking facilities is complex and
onerous.Without access to credit (perhaps secured using land title
deedsas collateral), it is impossible for poor farmers to
‘bootstrap’themselves into the irrigation economy or for
pastoralists or poorurban/peri-urban dwellers to drill their own
boreholes. Moreresearch is needed to understand the collective
impacts offinancing and banking services on facilitating or
retarding an SSAgroundwater revolution.
Transport infrastructure: facilitates movement of
machinery,product delivery to market, movement of extension
officials,migration of labor, provision of supplies, and numerous
otherinputs essential to modern business practices. For
agriculture,transporting goods to market requires road, rail or air
infrastruc-ture and potable groundwater supply schemes require
similarinfrastructure to access areas to drill boreholes and
deliver assetssuch as pumps and conveyance piping.
Sector-specific conditions and access to technologies:
contribute tothe viability of groundwater-led development. For
example, whileSouth Asia’s ‘green revolution’ and groundwater-led
irrigationdepended partly on the availability of cheap fertilizers
and high
yielding crop varietals (Pingali, 2012), such inputs in SSA are
cur-rently expensive and their availability is poor, hence their
use isnot popular (Druilhe & Barreiro-Hurlé, 2012; de Marsily
&Abarca-del-Rio, 2016; Dethier & Effenberger, 2012).
Furthermore,assets such as product storage facilities are
relatively rare in SSA,which leads to crop losses, as well as the
need to sell crops soonafter harvesting, when prices are low and
transport costs may behigh. Lack of refrigerated facilities
precludes storage of high valueagricultural commodities (e.g. meat,
cut flowers, vegetables) andprevents local processing of these
commodities. Low control overinputs, sales timing, pricing and
value-adding contribute to a reluc-tance to invest in upstream
infrastructure, such as groundwaterdevelopment. For potable
groundwater supply systems, decentral-ized infrastructure
approaches incorporating new and increasinglyaffordable
technologies (such as solar pumping) may overcomesome traditional
sectoral challenges (Cherunya, Janezic, &Leuchner, 2015), but
may require greater public-private collabora-tion to scale. For
drought resilience, improved hydrometeorologicalforecasting
technologies may contribute to increased resilienceand better
conjunctive management of water resources.
Groundwater knowledge and state of existing essential
services:can influence community demand for groundwater
development.For example, Pahuja et al. (2010) reported in India
that increasedpublic awareness of the availability of groundwater
(particularlyin areas where resources are more accessible), as well
as moresite-specific benefits – such as realization that
groundwater pump-ing could help alleviate challenges of
waterlogging and salinity –contributed to groundwater development.
Additionally, the poorstate of some water utility services in India
(which is oftenreflected in parts of SSA), prompted farm and
non-farm users toseek out their own local source and supply
systems, using ground-water. Hence, the appetite for improved
groundwater developmentmay be high in some areas in SSA to help
improve local drinkingwater, health and sanitation conditions.
3.2.2. Social, legal and institutional factorsRule of law and
regional stability: in the forms of physical secu-
rity, and predictable legal and political regimes, are
fundamentalrequirements for investment, including in groundwater
infrastruc-ture (Wijnen et al., 2012). As at 2018, there was poor
or deteriorat-ing civil stability in parts of several SSA
countries, includingSomalia, Mali, Chad, South Sudan, Central
African Republic, andthe DRC. This factor can be roughly quantified
at national level,but local resolution is dynamic and less well
understood.
Policy and regulation: analyses relating to groundwater are
rarein SSA and Tuinhof et al. (2011) suggest that institutional
capacityto implement existing policy is poor. Chokkakula and
Giordano(2013:796) provide one of the few available studies,
arguing‘‘clearly there is limited policy or institutional support
for thedevelopment of groundwater irrigation in SSA”. They suggest
thatexisting regimes may even hinder groundwater development
byover-emphasizing regulation and by favoring surface water.
Household income and resilience: contributes to
decision-makingon investment in groundwater extraction technologies
and broaderrisk management. Household resilience is precarious in
many partsof SSA, especially in ‘dryland’ areas threatened by
worseningdrought and regional instability. Many local SSA incomes
maynot permit savings accumulation sufficient for the purchase
ofeven basic technologies. Low resilience may discourage
invest-ment, which in turn can expose households to shocks,
potentiallyfurther lowering resilience.
Local community and institutional structures: may
embedgroundwater use in practice and tradition, and in turn
facilitatespecific solutions to local groundwater problems. They
can helpestablish critical mass for repair, maintenance and
drilling servicesand thereby reduce the risk and cost of extracting
groundwater.
-
606 J. Cobbing, B. Hiller /World Development 122 (2019)
597–613
Conversely, in areas where no local groundwater development
hasoccurred, effort may be required to remove institutional
obstaclesto overcome the ‘first mover’ disadvantage of using
groundwater.
Land rights, land tenure and collateral: are essential for
privategroundwater developers to access loans, particularly for
individu-als whose land is their largest asset (Chokkakula &
Giordano,2013). Conversely, uncertain land tenure makes long-term
invest-ments unattractive (USAID, 2016). Land tenure is often
closelyrelated to access to capital and credit, and security of
tenure is typ-ically associated with sets of rights (user-,
transfer-, exclusion- andenforcement-rights) (Feder & Feeny,
1991). However, land tenuresystems in large parts of SSA remain
tied to structures vested intraditional leaders or the state (which
often deters lenders) andspecific studies of the impacts of
existing land tenure arrange-ments on capital-raising at local
levels in SSA are rare. One excep-tion is in Niger, where a process
of securing land tenure for existingand future irrigation schemes
jointly recognizes both state andfarmer rights (Niger National
Office for Irrigation Schemes, 2017).
Additional secondary factors, not explored here, may
include:disease burden (humans and livestock), cultural beliefs
regardingwater use, changing gender roles, and the impact of
urbanization.While the relative importance of each secondary factor
will varywith context, we posit that the most influential limiting
factorswill need to be identified and overcome before more
intensivegroundwater development can be triggered at scale in
SSA.Evidence suggests that addressing these political economy
factorshas contributed to the relatively advanced
groundwater-irrigation economies in South Africa and in parts of
North Africa(Massuel et al., 2017; Vegter, 2001), despite these
regions havinglower overall groundwater availability than many
parts of SSA.
3.3. The boom phase
Acknowledging the importance of political economy factors
intriggering groundwater development has implications for theway in
which ‘management’ of groundwater development is envis-aged. It is
often assumed that groundwater development can beregulated or
controlled, however empirical data suggests that thisoften doesn’t
reflect the reality (Shah, 2017). Economic develop-ment, enabled or
catalyzed by a step change in groundwater-based development, is a
process of dynamic structural change insociety, rather than a
series of isolated adjustments to existing(static) water use
arrangements. Political economy factors will pro-gress in
unpredictable ways as local economies develop anddiverse new
economic relationships are established (Lejars et al.,2017).
Contemporary political economy factors, even if data wasavailable,
would evolve as development progresses and requireongoing data
gathering and interpretation (Manghee & Poole,2012).
Some authors (e.g. Pahuja et al., 2010; Shah,
Molden,Sakthivadivel, & Seckler, 2000) have described the
initial growthin groundwater use as ‘explosive’ – reflecting its
pace and scaleof utilization. The challenge of management during
the boomphase can be illustrated by global experiences in
groundwater irri-gation and water security. The difficulty of
trying to control orunderstand, ex-ante, the characteristics of a
step change in SSAgroundwater irrigation is illustrated by the fact
that even in Cali-fornia, where extensive groundwater irrigation
takes place in asophisticated institutional and data-rich context,
most of the 445groundwater basins in the state have inadequate
management,despite serious overdraft in some of them (Ayres,
Edwards, &Libecap, 2017). The ‘anarchy’ of groundwater
irrigation growth inIndia provides a comparable lesson (Shah, 2009)
and similarly inthe PRC, the ability of the state to control
groundwater abstractionsduring this phase can be limited (Shah,
2017). Nevertheless,enormous and transformative socioeconomic
benefits linked to
groundwater development have accrued to California, India andthe
PRC, and continue to do so.
Hence, the real value of a groundwater boom is the conversionof
resource utilization into socioeconomic development, which is
ashift in the relative importance of different sectors and
activities,for example, moving from low-productivity agriculture
and lowvalue-added extractive activities towards higher
productivityactivities (McMillan & Rodrik, 2011; McArthur &
McCord, 2017),with benefits accruing to society. Often, this
process of changehas been supported by a modernizing agricultural
sector whichboosted labor productivity, increased agricultural
surplus to accu-mulate capital, and increased foreign exchange via
exports(McArthur & McCord, 2017). In many cases, use of
groundwaterhas supported improved agricultural productivity, and
subsequenteconomic structural change.
Closer analysis of the challenges India has faced in taming
itssemi-anarchic groundwater boom can be contrasted with the
greatsocial and economic benefits derived. India is now the largest
userof groundwater in the world, with more than 60% of irrigated
agri-culture and 85% of drinking water supplies nationally
beinggroundwater dependent (Pahuja et al., 2010). The economic
valueof groundwater irrigation in India (in 2002) was
conservativelyestimated at US$8 billion annually – a figure greater
than allgovernment expenditures on poverty reduction and rural
develop-ment programs (Shah, 2008). Despite the presence of major
surfacewaters such as the Ganges, Indus and Brahmaputra Rivers,
Pahujaet al. (2010:91) describe groundwater in India as ‘‘arguably
themost critical water resource”, supporting irrigated
agriculturalproduction and rural livelihoods, as well as urban and
rural watersupplies. India experienced a boom period between �1960
and2010, which saw the area of irrigated agriculture supported
bytube wells grow from effectively zero to reach greater than30
million hectares – an area at least double that supported byany
other water source (Fig. 7). This boom period coincided withthe
Green Revolution, which was supported by intensive inputsof
(ground)water and fertilizer (Quinlan, Sen, & Nanda,
2014;Suhag, 2016). The increased socioeconomic resilience
providedby groundwater is evidenced by a rainfall deficit in
1963–66 (priorto wide tube well use) causing a 20% reduction in
national foodproduction, but a similar drought in 1987–88 (when
tube wellirrigation was far more widespread) had much lower impact
onfood production (Pahuja et al., 2010).
Harms done by overdraft and other problems associated
withover-abstraction are undeniable, but they must be balanced byan
assessment of the benefits derived during that process. It maybe
more constructive to understand enabling and/or
facilitatinggroundwater development by addressing political economy
bottle-necks or hurdles, than of controlling or planning such
develop-ment. In this context it is useful to consider the
economist AlbertHirschman’s perspective on economic development –
one that isincremental, local, and builds on ‘what works’
(Ellerman, 2001).This is opposed to more generic continental
overviews and associ-ated sweeping recommendations. Hirschman’s
‘Theory of Unbal-anced Growth’ requires an element of
disequilibrium in thepursuit of a step-change in economic
development. In the sameway, a step-change in SSA groundwater
development is unlikelyto happen ‘naturally’, since large parts of
the region remain in acycle of low investment, low returns and low
expectations (triggerfactors have not coalesced as needed). Such
groundwater develop-ment may also not respond well to a
predetermined top-downplanning or management approach, since so
many unknowns exist(mainly secondary factors, but also
hydrogeological factors). Themost productive intervention therefore
may be to work to incre-mentally remove bottlenecks to groundwater
development, likelyto be mainly secondary factors but also
including hydrogeologicalfactors (such as the lack of data on depth
to groundwater) where
-
J. Cobbing, B. Hiller /World Development 122 (2019) 597–613
607
necessary. Such pragmatic interventions will of course
dependheavily on the country, and on the sector (e.g. irrigated
agricultureversus urban/rural supply versus drought
resilience).
Different governance and management instruments tend tobecome
more prominent as the boom phase progresses, including:regulatory
measures (requiring sound legislation and capacity tomonitor and
enforce), economic measures (pricing mechanismssuch as volumetric
charges, taxes, and user fees), tradable ground-water rights (to
help users reach optimal outcomes), and commu-nity management of
groundwater (with local users as custodians ofwater resources via
regulation, property rights, pricing, etc.)(Pahuja et al., 2010).
These can help transition to the maturationphase.
3.4. The maturation phase
The maturation phase of groundwater development occurswhere
(volumetric) utilization peaks and then either plateaus
ormoderates; efforts towards conservation and/or remediation
pre-vail; and economic and structural development continues and
par-tially de-couples from groundwater use through innovation
andefficiency gains. It is ideally characterized by a shift from
develop-ment of groundwater to management (Shah, 2009).
This phase may be transitioned to consciously and
proactivelyviameasures described in the boomphase – regulatory or
economicmeasures, tradeable rights, community management – but in
otherglobal regions it has usually been prompted as a response to
casesof overexploitation or mismanagement. Many global regions
havealready reached this maturation phase. For example, Pahuja et
al.(2010:3) describes India’s ‘‘era of seemingly endless reliance
ongroundwater for both drinking water and irrigation purposes isnow
approaching its limit. . ..”, as almost one-third of
groundwaterblocks are in the ‘semi-critical’, ‘critical’, or
‘overexploited’ cate-gories. Better knowledge of secondary factors
could help proac-tively manage the boom and maturation phase,
rather thanreactively limit the costs of overexploitation and
mismanagement.
This maturation phase can also be characterized by innovationand
efficiency gains to permit continued economic development,via: (i)
demand-side measures (e.g. water efficient supply and con-sumption
schemes, behavioral changes and water reuse to reduceconsumptive
groundwater use), (ii) conjunctive use (achieving sav-ings by
aligning surface and groundwater use/management), and(iii)
groundwater recharge enhancement (physical interventionsto
concentrate and encourage infiltration). For example, drought-prone
areas in the Indian state of Andhra Pradesh have achievedlarge
scale success in self-regulation of groundwater use, wherefarmers
have doubled their income and continue to safeguard theircrops, all
the while reducing their groundwater use close to sus-tainable
levels (Pahuja et al., 2010). Similarly, evidence from NorthAfrica
suggests that better irrigation techniques can realize multi-ple
times more benefit for the same volume of groundwater used(Kuper et
al., 2017). Shah et al. (2014) and others have proposedthe
innovative roll-out of ‘electricity farming’ in areas of
ground-water overdraft – i.e. small farmers would be paid by
electricityutilities to ‘grow’ solar power as a remunerative ’cash
crop’, ratherthan consume electricity to pump dwindling
groundwater. Suchstrategies could be available to policymakers in
SSA for possibleearlier intervention.
19 This is analogous to the health sector, where more easily
measurable variableslike stunting or weight are used as proxies for
nutrition or overall health.
4. Discussion
Having confirmed significant groundwater potential in SSA
andderived an empirical model to chart a notional three-phase
path-way for groundwater development, we now pose three key
ques-tions for discussion:
(i) A range of complex interacting political economy
factorsappear to be the predominant barrier to triggering
SSA’sgroundwater development revolution, hence is there a wayto
better identify, understand and quantify these?
(ii) Given experiences in other global regions, what are
theadvantages and drawbacks in adopting a pro-developmentapproach
to groundwater in SSA?
(iii) Is the prevailing international discourse on
groundwaterconservation and remediation limiting SSA’s
developmenttrajectory?
4.1. Proxies for secondary factors
The empirical model suggests that SSA needs to tackle an (asyet)
unconfirmed mix of secondary factors to trigger a groundwa-ter
boom. There may be opportunity to use ‘proxy factors’ – indi-rectly
or loosely related to groundwater use, but which cannevertheless
shed light on groundwater use trends – to helpimprove resolution
and understanding of both hydrogeologicaland secondary factors. For
example, proxy factors for groundwaterirrigated agriculture could
include data on fertilizer sales, time ofelectricity use,
electricity grid coverage, crop export figures, satel-lite
estimates of groundwater irrigation clusters, import enquiriesby
pump manufacturers, profitability of drilling contractors, andmany
others. Data on proxy factors may be much more widelyavailable (and
possibly more useful) than data on actual ground-water use
(Mayer-Schönberger & Cukier, 2014). Analysis of proxyfactors is
one way in which the large data gaps on SSA groundwateruse and
secondary conditions may be bridged efficiently
andeffectively19.
4.2. The sustainability challenge
If a step-change in groundwater use in SSA – once catalyzed
bythe removal of political-economy impediments – cannot be
easilymanaged or controlled, this raises the possibility of falling
watertables and contamination, and their disproportionate impacts
onthe poor or vulnerable. There is a growing and influential
literaturethat points out the negative effects of the boom in
groundwateruse elsewhere in the world (e.g. Sekhri, 2014), and a
concern thatadvocates for increased groundwater use may
inadvertently beresponsible for future hardship. As such, a
cautious or managedapproach to SSA groundwater development is often
recommendedby specialists, stressing better up-front management as
a way ofhedging against the possibility of overdraft (e.g. Edmunds,
2012;Foster et al., 2012). There are several points relevant to
this debatein SSA.
Firstly, whilst it is sometimes assumed that a boom in
ground-water use will be ephemeral or unsustainable, the
synthesized fig-ures quoted in Section 2 show that, on average,
large increases insustainable groundwater use are possible in SSA
even when allow-ing for conservative environmental requirements
(see Fig. 4).Whilst the average figures do mask possible hotspots,
and thereare already small areas in SSA where unsustainable use of
ground-water occurs, in general large increases in renewable
groundwaterutilization are feasible across most countries in the
region. In short,SSA is nowhere near unsustainability of resource
use, at a regionalscale.
Secondly, surveys of the disbenefits of groundwater
abstractionfor development in other global regions cannot easily
compare theoverall economic impacts, including growing overdrafts,
with acounterfactual in which no groundwater development in the
same
-
22 Defined as projects which may potentially involve the
utilization of groundwaterresources and/or impact them directly or
indirectly.23 A total of 254 projects in SSA with relevance to
groundwater were reviewedbetween 1997 and 2017. The number of
projects approved annually did not exceed 10between 1997 and 2002
and peaked in 2013 and 2014, at 19 and 22 respectively. 85project
completion reports were reviewed.24 Current includes all pipeline,
lending, ongoing and approved projects.25 For example, the
Millennium Development Goals (which transitioned into
theSustainable Development Goals), the Kyoto and Paris climate
agreements, and the
608 J. Cobbing, B. Hiller /World Development 122 (2019)
597–613
region has occurred. It is likely that overall economic and
struc-tural benefits are considerably higher, despite the clear
drawbackslinked to areas of overdraft, than if no groundwater
developmenthad taken place at all. Indeed, India’s ‘green
revolution’ was largelypredicated on groundwater (Shah, 2009) and
the PRC’s grain pro-duction has shown correlation with groundwater
abstraction ratesat a national level for approximately 60 years
(Fig. 8). Where theexploitation of unsustainable or ‘fossil’
groundwater is considered,arguments like those made by mining
companies for the catalyticimpact of economic development based on
a finite resource aremade by some authors (e.g. Maliva &
Missimer, 2012). In suchcases, authors such as Collier (2014, 2017)
justify the use of non-renewable resources only where the income
from their use con-tributes to improvements to future public social
goods.
Thirdly, there are powerful humanitarian and social justice
ele-ments to consider. The real choice in SSA may be between the
cur-rent situation in which little groundwater is being used
andtherefore it is not significantly contributing to economic
develop-ment, and future groundwater development where only
partialcontrol may be possible (semi-anarchic) until the
maturationphase is reached. We argue that the potential benefits of
muchwider groundwater development in the region are too importantto
delay, and that potential disbenefits due to possible incidentsof
over-abstraction are preferable to the current challenges of
pov-erty and vulnerability in many SSA countries while
availablegroundwater resources lay dormant. There may be moral
hazardin advocating for caution in the use of groundwater, just as
thereis in promoting the growth in its use.
Finally, economic dynamism (potentially supported by
ground-water use, for example) is likely to be essential to the
political andsocial stability needed to overcome long-term
environmentalproblems. Economic development today should be built
on pro-gressive and sustainable environmental policies, but a
narrowemphasis on the latter may preclude the former and thereby
riskboth. A commentary on the current broader discourse in the
inter-national development sector is provided below.
4.3. Elevating groundwater up the development agenda in SSA
Our model in Fig. 5 shows SSA at an earlier phase of
groundwa-ter development compared to India, California, the PRC,
and otherglobal regions of higher intensity groundwater use. The
emphasisin these latter regions has rightly shifted from the boom
tomaturation phase, giving rise to a large contemporary
literatureon over-abstraction (e.g. Shah et al., 2000) and
conservation andremediation (Wada et al., 2010; Famiglietti, 2014).
This discourseis often applied to SSA today (e.g. Xu, 2008; Braune
et al., 2008),although the region has not yet experienced the
groundwaterboom seen elsewhere. Where this facilitates better
groundwaterdevelopment it is appropriate, but if it adds to the
alreadyconsiderable inertia in initiating even small improvements
ingroundwater-based development in SSA, it may be
stronglycounter-productive. If groundwater is perceived as a
resource toconserve from the outset, rather than one to develop
sustainably,it may lie effectively dormant.
We observe evidence of this discourse permeating
groundwaterinvestment in SSA, which remains low relative to surface
waterand has even declined as a priority for some development
agencies.For example, while the World Bank has significantly
increasedinvestment in SSA’s water sector20, this is concentrated
in surfacewater infrastructure21. Over a twenty-year period
(1997–2017),
20 For example, World Bank lending to the SSA water sector
increased from US$820m in 2006/07 to US$1780m in 2016/17,
concentrated in water supply andsanitation and irrigation and
drainage (Wijnen et al., 2018).21 Dams and irrigation canals,
watershed and river basin management.
the number of World Bank funded projects with ‘relevance
togroundwater’22 increased23, however almost none had groundwateras
their central focus. Less than 1% contained groundwater in
theirproject title; only 3% contained reference to groundwater in
the pro-ject appraisal document abstract; and no project completion
reportabstracts referenced groundwater (Wijnen et al., 2018) –
indicatingthat groundwater is either rarely being integrated in
projects, or isbeing grossly under-reported, or both. Similarly, a
2018 review ofthe African Development Bank’s (AfDB) portfolio24 of
projects acrossall sectors (including water supply and sanitation,
agriculture andagro-industries, environment) revealed only one
project with‘groundwater’ in the title.
For the World Bank, trends in overlooking groundwater havebeen
recognized formally by its own Independent EvaluationGroup (IEG) at
both global and regional scales. For example, asillustrated in Fig.
9, the World Bank’s investment (globally) ingroundwater extraction
declined significantly between the mid-1990s and late-2000s – a
trend described by IEG (2010:79) as‘‘problematic”. IEG’s 2010
analysis of the World Bank’s globalwater portfolio (1997–2007)
found that groundwater was the leastcommon theme and a lower or
diminishing priority both at the glo-bal scale and for SSA, and for
which IEG (2010:27) recommendedthe World Bank be ‘‘more ambitious
in addressing issues criticalto the long-term use of
groundwater”.
To contextualize these decreasing and low levels of investmentin
groundwater, such trends have occurred against a backdrop
ofincreasing international and regional focus on climate
change,extreme events, resiliency and poverty alleviation25. Beyond
mini-mal strategic prioritization, other reasons for historical and
currentlow levels of groundwater investment in SSA may include:
poorunderstanding of groundwater resources and their sustainable
man-agement (IEG, 2010); the potentially complex and hidden nature
ofgroundwater, both politically and physically (Wijnen et al.,
2012);a general hesitancy in SSA to invest in groundwater
irrigation, basedon mixed results of past interventions (Ward et
al., 2016)26; anemphasis on River Basin Organizations (RBOs) as
regional bodiesfor freshwater governance inclining policy and
funding towards sur-face waters; relatively few dedicated
groundwater experts innational and international organizations
(Llamas and Custodio,2001); and groundwater cutting across multiple
sectors, meaningthat it is potentially everywhere but also nowhere
specifically27.While it can be argued for SSA country governments
to be moreproactive in their own groundwater development, we
believe that,as thought-leaders, organizations such as the
multilateral develop-ment banks and United Nations agencies do have
a role to play inhelping countries become more aware of the
potential benefits ofsustainable groundwater development, and
supporting them toachieve that.
Beyond a general dearth of groundwater investment in SSA,there
is also no strong evidence of groundwater being
developedstrategically and at scale to counter chronic water stress
or therecurrent and increasing threat of drought in hard-hit
regions of
Sendai Disaster Risk Reduction Framework.26 Irrigation
development during the 1970s and 1980s in SSA delivered low rates
ofreturn and many nationally financed schemes failed. Such poor
results deterredgovernments and donors in financing further
irrigation expansion (Ward et al., 2016).27 For example, at the
World Bank, projects in SSA with relevance to groundwater(between
1997 and 2017) cut across 62 different units in the
institution.
-
Fig. 8. Groundwater Exploitation and Total Grain Production from
1950 to 2011 in the PRC (after Liu & Zheng, 2016).
Fig. 9. The focus of groundwater projects across World Bank
global portfolio (1997–2007) (after Independent Evaluation Group
(IEG), 2010).
J. Cobbing, B. Hiller /World Development 122 (2019) 597–613
609
SSA. For example, in the Horn of Africa – a region of high
waterstress and repeated drought events (2018 was its third
consecutiveyear of drought) – only one of the three countries in
the WorldBank’s SSA Horn of Africa region has had more than one
nationalproject relevant to groundwater in the past 20 years28.
Other coun-tries which have suffered from recent severe drought,
and whichhave had very few (or no) national projects with relevance
togroundwater supported by the World Bank between 1997 and2017,
include Namibia (0), South Africa (2), South Sudan (1), Swazi-land
(1), and Zimbabwe (1). There is little evidence of national
orinternational actors explicitly linking strategic groundwater
invest-ments to areas experiencing frequent and recurrent drought
events.To put this lack of investment in focus (and building upon
Table 1),Table 3 highlights the strong dichotomy between the
serious state ofwater stress (current and future) and vulnerability
to climate changeand drought, contrasted with the dormant renewable
groundwaterresources available in most countries in SSA. As stated
previously,
28 These don’t include regional projects, which primarily relate
to transboundarysurface water basin interventions.
we recommend multilateral agencies collaborate with regional
andnational agencies to explore how a strategic network of
deepgroundwater bores in drought ‘hotspots’ could proactively build
resi-lience against future events.
Hence, there is an opportunity for many actors to support amore
balanced narrative by acknowledging the current predomi-nance of
surface water investments and proactively addressingthe low levels
of groundwater investment in SSA. Institutions suchas the
multilateral development banks span the technical and sec-toral
spectrum, have strong convening power and can play criticalroles as
advocates and knowledge brokers to help raise the profileof
groundwater as a reliable and climate resilient resource
aroundwhich countries can orient elements of economic and social
trans-formation and disaster prevention. Such institutions also
have thecapacity to work collaboratively with interested SSA
countries tohelp understand and address some of the limiting
secondary fac-tors in SSA and facilitate knowledge exchanges from
other globalregions to learn how to trigger and then navigate a
boom phasein SSA, all the while helping to better manage the
potential perilsof over-development from the outset.
-
Table 3National1 states of water stress and vulnerability
contrasted with low levels of renewable groundwater
utilization.
Water stressa Vulnerability of national freshwater supplies to
climatechangeb
Drought vulnerabilityindicatorc
Percentage renewablegroundwater usedd
Present Future
SSA Countries (Only high andmoderate-high risk countriesare
listed for water stress andvulnerability. Only lowutilization
countries are listedfor renewable groundwater).
High:Eritrea,Lesotho,SouthAfrica,Swaziland.
High:Botswana,Eritrea,Namibia,SouthAfrica.Moderate-High:Swaziland.
High: Benin, Chad, Congo(Democratic Republic), Congo(Republic
of), Eritrea, Ethiopia,Kenya, Madagascar, Mali,Mauritania, Niger,
Nigeria,Senegal, Somalia, Sudan,Swaziland.Moderate-High: Angola,
BurkinaFaso, Liberia, Mozambique,Sierra Leone, Tanzania, Togo.
High: Burundi, Chad, Ethiopia,Guinea-Bissau, Mali,
Niger,Nigeria, Sierra Leone, Somalia.Moderate-High: CentralAfrican
Republic, Congo(Democratic Republic),Liberia, Malawi,
Mauritania,Mozambique, Rwanda, SouthSudan, Sudan, Togo.
Low: Angola, Benin, Botswana,Burkina Faso, Burundi,
Cameroon,Central African Republic, Chad,Congo (Democratic
Republic),Congo (Republic of), Cote D’Ivoire,Equatorial Guinea,
Eritrea,Ethiopia, Gabon, Gambia (The),Ghana, Guinea,
Guinea-Bissau,Kenya, Lesotho, Liberia,Madagascar, Malawi,
Mali,Mozambique, Namibia, Niger,Nigeria, Rwanda, Senegal,
SierraLeone, Somalia, South Sudan,Sudan, Swaziland, Tanzania,
Togo,Uganda, Zambia, Zimbabwe.
High risk and moderate-high risk rating by the authors is based
on the following sources: aGassert, Reig, Luo, and Maddocks (2013),
Luo, Young, and Reig (2015); bNotre DameGlobal Adaptation
Initiative (www.gain.nd.edu); and cNaumann, Barbosa, Garrote,
Iglesias, and Vogt (2014). Low classification defined as country
currently using less than25% of national renewable groundwater
resources, sourced from: dFAO (2016) AQUASTAT data.
1 Caveats to the analysis in the matrix above include: (i) the
indicators come from different sources, and hence methodologies and
consideration of groundwater resourcesare not consistent; (ii)
national-level assessments do not capture the variability across
regions within countries (for example, a country assessment may
reveal low overallwater stress or vulnerability, but have select
areas where water stress is high); and (iii) sustainable
groundwater resource development potential is high across all
countries(except Mauritania), meaning groundwater development
should be considered as part of an integrated water resources
strategy for all SSA countries.
610 J. Cobbing, B. Hiller /World Development 122 (2019)
597–613
5. Conclusions and recommendations
We confirm, at regional scale, that SSA hosts significant
ground-water resources, often in areas where it could be most
impactful,and which have the capacity to contribute significantly
as a base-line resource for regional development. Despite the
wealth ofregional resources, groundwater use is thought to remain
under5% of sustainable yield for most countries in SSA, meaning
that sig-nificant renewable resources are currently dormant.
We substantiate that ‘secondary’ or political-economy factorsare
the primary impediments to further groundwater developmentin SSA.
There is further work to be conducted to understand polit-ical
economy factors, their complex interactions, and how to trig-ger
them. This work will allow appropriate policy to bedeveloped, to
remove bottlenecks.
We identify a predominance of limiting, rather than
enablingconditions in groundwater development, which have
discouragedinvestment at scale. Based on the development cycle of
groundwa-ter in other parts of the world which have already
experiencedtrigger-boom-maturation phases, the international
discourse ongroundwater has shifted towards conservation and
remediation,which may inadvertently be denying SSA its opportunity
to expe-rience social and economic benefits derived from
groundwaterdevelopment.
We also argue that attention and funding for groundwater
pro-jects should be of similar scale to that afforded to surface
waterprojects in SSA. Groundwater has the potential to act as the
foun-dational resource to underpin regional development in
sectorssuch as irrigated agriculture, urban and rural water
security, anddrought resilience, just as it has in other global
regions. We arguethat it is now unconscionable and unjustifiable
not to developSSA’s groundwater resources.
Finally, while the primary intentions of this paper are to
informreaders about SSA’s renewable groundwater potential and
encour-age discussion on future groundwater development actions
foreconomic and humanitarian purposes, we draw on our
empiricalmodel to offer some non-prescriptive elements of a roadmap
tohelp support this process:
� Dissemination of study findings to decision-makers and
policy-makers in SSA countries, highlighting the significant
latentrenewable groundwater potential and the importance of
sec-ondary (political economy) factors in triggering wider
ground-water development.
� Encourage improved resolution and coverage of hydrogeologi-cal
data, including exploration of proxy indicators, asappropriate.
� Provide financial and technical resourcing for
adequategroundwater-specific investigations to be included in
nationalwater resource assessments in all SSA countries.
� Educate and encourage international institutions to
prioritizegroundwater development support in SSA, including
bolsteringinternal technical capacities.
� Establish a taskforce (comprising relevant stakeholders such
asthe UN agencies, multilateral development banks, disaster
relieforganizations, local community representatives, etc.) to
explorethe potential benefits of a strategic network of deep
groundwa-ter boreholes in recurrent (and predicted future)
droughthotspots.
� Explore linkages between groundwater development andemerging
climate (financing and convening) mechanisms topromote adaptation
and resilience building.
Simultaneously, we encourage countries to help themselveswhilst
also calling upon international and regional institutions toprovide
financial and technical support to help realize SSA’s pend-ing
groundwater revolution.
Acknowledgements
We extend thanks to our colleagues, Marcus Wijnen andShawki
Barghouti, for their guidance and advice on our joint
inves-tigation into groundwater in SSA in 2017, and to Cleo
Rose-Innesand Maarten de Wit for kindly providing feedback on a
draftversion of this paper. We also acknowledge the two
anonymousreviewers whose comments improved the final
manuscript.
http://www.gain.nd.edu
-
J. Cobbing, B. Hiller /World Development 122 (2019) 597–613
611
Declaration of Competing Interest
The authors have worked as consultants to the World Bank.
Funding sources
This research did not receive any specific grant from
fundingagencies in the public, commercial, or not-for-profit
sectors.
Author approval
Both authors have approved this version of the article.
References
Adelana, S. M. A., Tamiru, A., Nkhuwa, D. C. W., Tindimugaya,
C., & Oga, M. S. (2008).Urban groundwater management and
protection in sub-Saharan Africa. In S.Adelana & A. MacDonald
(Eds.), Chapter 14 in Applied groundwater studies inAfrica, IAH
Selected Papers on Hydrogeology (pp. 231–260).
Altchenko, Y., & Villholth, K. G. (2015). Mapping irrigation
potential from renewablegroundwater in Africa – A quantitative
hydrological approach. Hydrology andEarth System Sciences, 19,
1055–1067.
Ashworth, W. (2006). Ogallala Blue. Water and life on the high
plains. Woodstock,Vermont: The Countryman Press.
Ayres, A. B., Edwards, E. C., & Libecap, G. D. (2017). How
transaction costs obstructcollective action: Evidence from
California’s groundwater NBER Working Paper23382. Cambridge, MA,
USA: National Bureau of Economic Research.
Banerjee, S., & Morella, E. (2011). Africa’s water and
sanitation infrastructure access,affordability, and alternatives.
World Bank Publication.
Baro, M. A., & Deubel, T. F. (ro and Deubel, September
2006). Persistent hunger:Perspectives on vulnerability, famine, and
food security in Sub-Saharan Africa.Annual Review of Anthropology,
35, 521–538.
https://doi.org/10.1146/annurev.anthro.35.081705.123224.
Benson, C., & Clay, E. (March 1998). The impact of drought
on sub-saharan Africaneconomies: A preliminary examination. World
Bank Publication. https://doi.org/10.1596/0-8213-4180-4.
Besada, H., & Werner, K. (2015). An assessment of the
effects of Africa’s water crisison food security and management.
International Journal of Water ResourcesDevelopment, 31(1),
120–133. https://doi.org/10.1080/07900627.2014.905124.
BGR (Bundesanstalt für Geowissenschaften und Rohstoffe) and
UNESCO. 2008.Groundwater resources of the World 1:25 000 000 –
Hannover, Paris.
Bisson, R. A., & Lehr, J. H. (2004). Modern groundwater
exploration: Discovering newwater resources in consolidated rocks
using innovative hydrogeologic concepts,exploration, drilling,
aquifer testing and management methods. New Jersey: JohnWiley &
Sons.
Braune, E., Hollingworth, B., Xu, Y., Nel, M., Mahed, G., &
Solomon, H. (2008).Protocol for the assessment of the status of
sustainable utilization and managementof groundwater resources with
special reference to Southern Africa Report TT 318/08. Pretoria:
Water Research Commission.
Braune, E., & Xu, Y. (2010). The role of ground water in
Sub-Saharan Africa.Groundwater, 48(2), 229–238.
Burney, J., Naylor, R. L., & Postel, S. L. (2013). The case
for distributed irrigation as adevelopment priority in Sub Saharan
Africa. Proceedings of the National Academyof Sciences., 110,
no.31.
Cabot Venton, C., Fitzgibbon, C., Shitarek, T., Coulter, L.,
& Dooley, O. (2012). Theeconomics of early response and
disaster resilience: Lessons from Kenya andEthiopia Economics of
Resilience, Final Report. UK: DFID.
Calow, R., MacDonald, A. M., Nicol, A. L., & Robins, N. S.
(2009). Ground watersecurity and drought in Africa: Linking
availability, access and demand.Groundwater, 48(2), 246–256.
Carter, R. C., & Parker, A. (2009). Climate change,
population trends andgroundwater in Africa. Hydrological Sciences,
54(4), 676–689. https://doi.org/10.1623/hysj.54.4.676.
Cherunya, P. C., Janezic, C., & Leuchner, M. (2015).
Sustainable supply of safedrinking water for underserved households
in Kenya: Investigating the viabilityof decentralized solutions.
Water, 7(10), 5437–5457. https://doi.org/10.3390/w7105437.
Chokkakula, S., & Giordano, M. (2013). Do policy and
institutional factors explainthe low levels of smallholder
groundwater use in Sub-Saharan Africa? WaterInternational, 38(6),
790–808.
Cobbing, J. E., & Davies, J. (2011). Improving access to
southern Africa’s groundwater‘‘grey data”. Hydrogeology, 19(6),
1117–1120.
Colenbrander, W., & van Koppen, B. (2013). Improving the
supply chain of motorpumps to accelerate mechanized small-scale
private irrigation in Zambia.WaterInternational, 38(4),
493–503.
Collier, P. (2014). The ethics of natural assets. Global Ethics,
10(1), 45–52. https://doi.org/10.1080/17449626.2014.896573.
Collier, P. (2017). The institutional and psychological
foundations of naturalresource policies. Development Studies,
53(2), 217–228. https://doi.org/10.1080/00220388.2016.1160067.
Couttenier, M., & Soubeyran, R. (March 2014). Drought and
civil war in Sub-SaharanAfrica. The Economic Journal, 124(575),
201–244. https://doi.org/10.1111/ecoj.12042.
de Marsily, G., & Abarca-del-Rio, R. (2016). Water and food
in the twenty-firstcentury. Surveys in Geophysics, 37, 503–527.
DeFries, R., & Nagendra, H. (2017). Ecosystem management as
a wicked problem.Science, 356, 265–270.
Demombynes, G., & Kiringai, J. (2011). The drought and food
crisis in the horn ofAfrica: Impacts and proposed policy responses
for Kenya. Washington, DC:Economic Premise. No. 71 World Bank.
https://openknowledge.worldbank.org/handle/10986/10073.
Dethier, J. J., & Effenberger, A. (2012). Agriculture and
development: a brief reviewof the literature. Economic Systems,
36(2), 175–205.
Döll, P., & Fiedler, K. (2008). Global-scale modeling of
groundwater recharge.Hydrology and Earth System Sciences, 12,
863–885.
Döll, P., & Flörke, M. (2005). Global-scale estimation of
diffuse groundwater rechargeFrankfurt Hydrology Paper 03. Frankfurt
am Main, Germany: Institute ofPhysical Geography, Frankfurt
University.
Druilhe, Z., & Barreiro-Hurlé, J. (2012). Fertilizer
subsidies in sub-Saharan Africa ESAWorking paper No. 12-04. Food
and Agriculture Organization of the UnitedNations (FAO).
Edmunds, W. M. (2012). Limits to the availability of groundwater
in Africa.Environmental Research Letters, 7(2), 1–3.
Ellerman, D. (2001). Hirschmanian themes of social learning and
change World BankPolicy Research Working Paper No. 2591. Washington
DC: World Bank.
Famiglietti, J. S. (November 2014). The global groundwater
crisis. Nature ClimateChange, 4, 945–948.
FAO (Food and Agricultural Organization). 2016. AQUASTAT Main
Database, Foodand Agriculture Organization of the United Nations
(FAO).
http://www.fao.org/nr/water/aquastat/data/query/index.html?lang=en.
Accessed August 2017.
Feder, G., & Feeny, D. (January 1991). Land tenure and
property rights: Theory andimplications for development policy. The
World Bank Economic Review, 5(1),135–153
https://www.jstor.org/stable/3989973.
Foster, S., Hirata, R., Misra, S., & Garduno, H. (2010a).
Urban groundwater use policy:Balancing the benefits and risks in
developing nations Strategic Overview Se