Presented by IWMI’s Yvan Altchenko at the 26th General Assembly of the International Union of Geodesy and Geophysics (IUGG), held in Prague - Czech Republic, on June 25, 2015. Session - Societal Relevance of Groundwater: Ever Increasing Demands on a Limited Resource

1. Mapping Irrigation Potential from Renewable Groundwater in Africa: A
Quantitative Hydrological Approach
Yvan Altchenko & Karen G. Villholth
PhotobySteveMcCurrach/IWMI

2. o Background
 Huge GW volumes in Africa (MacDonald et al, 2012) but not all
available for abstraction, and unevenly distributed
 1 % of cultivated land is equipped for irrigation with GW while it is
14% in Asia, which has similar climate (Siebert et al., 2010)
 Only approx. 2 mill. ha irrigated with GW presently
 Groundwater provides an important buffer to climate variability and
change
o Questions
 Where is the GW irrigation potential (GWIP) in Africa?
 How much cropland can GW irrigate?
o Objective
 To estimate the cropland area that renewable GW can irrigate in Africa (using Pavelic’ et al. (2012) water balance
approach)
o Some assumptions in computations
 GW is the only water source for irrigation (no conjunctive use with SW)
 GW is usable and accessible (no quality, yield, or socio-economic constraints)
 GW is locally available
(106 ha)
0
100
200
300
400
500
600
Asia Africa
Cultivated land
Total area equipped
for Irrigation
Area equipped for
irrigation irrigated
with groundwater

3. Methodology
o Hydrological data
• Recharge
• Water available for crop from rain (green water)
o Other GW uses
• human activities (domestic, livestock, industrial)
• environment
o Crop data
• Surface of cropland
• Crop distribution
• Crop water demand
• Irrigation efficiency
𝑮𝑾𝑰𝑷 (m2) =
𝐺𝑊 𝐴𝑣𝑎𝑖𝑙𝑎𝑏𝑙𝑒 (m3 year−1)
𝐼𝑟𝑟𝑖𝑔. 𝑊𝑎𝑡𝑒𝑟 𝐷𝑒𝑚𝑎𝑛𝑑 (𝑚 year−1)
𝑮𝑾 𝑨𝒗𝒂𝒊𝒍𝒂𝒃𝒍𝒆 = 𝐺𝑊 𝑅𝑒𝑐ℎ𝑎𝑟𝑔𝑒 – 𝐻𝑢𝑚𝑎𝑛 𝐺𝑊 𝐷𝑒𝑚𝑎𝑛𝑑 – 𝐸𝑛𝑣𝑖𝑟𝑜𝑛. 𝐺𝑊 𝑅𝑒𝑞.
Different geographical data compiled in GIS
𝑰𝒓𝒓𝒊𝒈. 𝑾𝒂𝒕𝒆𝒓 𝑫𝒆𝒎𝒂𝒏𝒅 =
{ 𝑖=1
𝑛
𝑗=1
𝑚
𝐶𝑟𝑜𝑝 𝑊𝑎𝑡𝑒𝑟 𝐷𝑒𝑚𝑎𝑛𝑑−𝐺𝑟𝑒𝑒𝑛 𝑊𝑎𝑡𝑒𝑟 𝑗 × % 𝑜𝑓 𝐴𝑟𝑒𝑎 𝑖 }
𝐼𝑟𝑟𝑖𝑔. 𝐸𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦
=
𝑁𝑒𝑡 𝐼𝑟𝑟𝑔.𝑊𝑎𝑡𝑒𝑟 𝑑𝑒𝑚𝑎𝑛𝑑
𝐼𝑟𝑟𝑖𝑔. 𝐸𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦
n= crop
Cell size :
0.5 x 0.5 degree
(≈ 50 km x 50 km)

4. Hydrological data
from the PCR-GLOBWB model (Utrecht University, the Netherlands, Wada et al., 2011)
• Leaky bucket type of model, applied on a cell-by-cell base
• cell size of 0.5 x 0.5 degree ≈ 50 x 50 km
• Daily step time
• Data from 1960 to 2001
(41 years)
QDR = Direct Runoff
QSf = Interflow
QBf = baseflow
Store 1 = soil 1 (0 – 0.3 m deep)
Store 2 = soil 2 (0.3 – 1.2 m deep)
Store 3 = GW storage
Green water (monthly value)
= Transpiration from soil 1 + soil 2
Recharge (annual value)
Methodology

5. Other GW uses
• Human activities from FAO (geonetwork) using density of population and livestock
• Environmental water need according to three different scenarios:
• Scenario (a) : 70 % of recharge goes to environment
• Scenario (b) : 50 % of recharge goes to environment
• Scenario (c) : 30 % of recharge goes to environment
Methodology
Uses/unit
Daily water need
(L)
Portion assumed to come from GW
(%)
Domestic Inhabitant 50 75
Industrial Inhabitant 25 75
Livestock
Big ruminant 40 100
Small ruminant 20 100
Pig 30 100
Poultry 0.2 100

6. Groundwater Availability
Annual calculations over 41 years, then average
From 692 to 1644
km3 available for
irrigation per year,
depending on
scenario
Average groundwater availability for irrigation (1960–2000), expressed in 106 m3 year−1 cell−1 (0.5° ×0.5°)
𝑮𝑾 𝑨𝒗𝒂𝒊𝒍𝒂𝒃𝒍𝒆 = 𝐺𝑊 𝑅𝑒𝑐ℎ𝑎𝑟𝑔𝑒 – 𝐻𝑢𝑚𝑎𝑛 𝐺𝑊 𝐷𝑒𝑚𝑎𝑛𝑑 – 𝐸𝑛𝑣𝑖𝑟𝑜𝑛. 𝐺𝑊 𝑅𝑒𝑞.
3 scenario: environmental groundwater requirements as
70 % of recharge 50 % of recharge 30 % of the recharge

7. Crop data
• Cropland area: Crop distribution from Center for
Sustainability and the Global Environment (SAGE), University
of Wisconsin, Madison, Wisconsin, USA
• World map
• Resolution: 5 minutes
• 11 major crops
• Geographic crop distribution
for the year 2000
• Crop water demand from compiled FAO database
Monthly crop water demand calendar for 6 crop groups
Methodology
Proportion of cropland per cell (0.5° x 0.5°) in 2000

8. Monthly crop water demand based on
• Crop Water Demand j = Crop Group Coefficient × E0,maxj
With E0,maxj
= maximum reference evapotranspiration for each calendar month
• Length of crop growth stage (initial, development, mid-season, end)
• Crop coefficient for crop growth stage (Kc initial , Kc development , Kc mid-season , Kc end )
Example: tomatoes
• Single crop coefficient and growing length
• Calculation of monthly crop coefficient
• Planting and harvesting calendar
Growing stage initial development mid-season end
Growing length (days) 35 45 40 25
Crop coefficient 0.45 0.75 1.15 0.8
Month 1 2 3 4 5
Water need (mm) 0.45 0.7 0.88 1.15 0.67
Methodology

9. Zone
code
Zone name
Irrigation
Efficiency
( %)
1 Mediterranean coastal zone 60
2 Sahara oases 70
3 Semi-arid to arid savannah West-East Africa 50
4 Semi-arid/arid savannah East Africa 50
5 Niger / Senegal rivers 45
6 Gulf of Guinea 50
7 Southern Sudan 50
8 Madagascar tropical lowland 50
9 Madagascar highland 50
10 Egyptian Nile and Delta 80
11 Ethiopian highlands 50
12 Sudanese Nile area 80
13 Shebelli-Juba river area in Somalia 50
14 Rwanda - Burundi - Southern Uganda highland 50
15 Southern Kenya - Northern Tanzania 50
16 Malawi - Mozambique - Southern Tanzania 45
17 West and Central African humid areas 45
18 Central African humid areas below equator 45
19 River affluents on Angola / Namibia / Botswana border 50
20 South Africa - Namibia - Botswana desert & steppe 65
21 Zimbabwe highland 60
22 South Africa - Lesotho - Swaziland 60
23 Awash river area 50
Irrigation cropping pattern zones and irrigation efficiency (FAO, 1997)
Methodology

10. Crop group coefficient per month
Example: zone 15 (en mm)
Crop group Description Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Cereals
Maize/Rice/Sorghum 1.03 1.08 1.18 1.18 1.03 0.83
Maize/Rice/Sorghum 1.18 1.03 0.83 1.03 1.08 1.18
Oil crops
Sunflower/Groundnut/Soybean 0.58 0.47 0.85 1.13 1.08
Sunflower/Groundnut/Soybean 0.47 0.85 1.13 1.08 0.58
Roots
Potatoes 0.45 0.75 1.08 1.1 0.71
Potatoes 0.71 0.45 0.75 1.08 1.1
Pulses
Bean 0.2 0.47 0.83 1.1
Bean 0.47 0.83 1.1 0.2
Vegetables
Onion/Tomato/Eggplant/Cucumb
er
0.53 0.74 1 1.13 0.65
Onion/Tomato/Eggplant 0.52 0.74 1 1.13 0.65
Sugarcane Sugarcane 0.4 0.9 1 1.13 1.25 1.25 1.25 1.25 1.25 1 0.75 0.88
Methodology
𝑰𝒓𝒓𝒊𝒈. 𝑾𝒂𝒕𝒆𝒓 𝑫𝒆𝒎𝒂𝒏𝒅 =
{ 𝑖=1
𝑛
𝑗=1
𝑚
𝐶𝑟𝑜𝑝 𝑊𝑎𝑡𝑒𝑟 𝐷𝑒𝑚𝑎𝑛𝑑−𝐺𝑟𝑒𝑒𝑛 𝑊𝑎𝑡𝑒𝑟 𝑗 × % 𝑜𝑓 𝐴𝑟𝑒𝑎 𝑖 }
𝐼𝑟𝑟𝑖𝑔. 𝐸𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦
=
𝑁𝑒𝑡 𝐼𝑟𝑟𝑔.𝑊𝑎𝑡𝑒𝑟 𝑑𝑒𝑚𝑎𝑛𝑑
𝐼𝑟𝑟𝑖𝑔. 𝐸𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦
n= crop
𝐶𝑟𝑜𝑝 𝑊𝑎𝑡𝑒𝑟 𝐷𝑒𝑚𝑎𝑛𝑑 𝑗 = Crop Group Coefficient × E0,maxj
(maximum reference evapotranspiration for each calendar month)
Monthly calculations, then sum to get annually irrigation water demand over 41 years

11. Average Net Irrigation Water demand (1960-2000)
Results
Cropland

12. Results
𝑮𝑾𝑰𝑷 (m2) =
𝐺𝑊 𝐴𝑣𝑎𝑖𝑙𝑎𝑏𝑙𝑒 (m3 year−1)
𝐼𝑟𝑟𝑖𝑔. 𝑊𝑎𝑡𝑒𝑟 𝐷𝑒𝑚𝑎𝑛𝑑 (𝑚 year−1)
Total area irrigable with groundwater inside a cell (0.5° ×0.5°) in 103 ha
GWIP calculated annually, then averaged over 41 years
GW Available entered as average over 41 years to consider buffer effect of GW
70 % of recharge to environment 50 % of recharge to environment 30 % of the recharge to environment

13. AFRICA Environmental requirements represent
30% of recharge 50% of recharge 70% of recharge
Area (106 ha) 105.3 74.9 44.6
% of cropland 48.5% 34.5% 20.5%
Proportion of cropland irrigable with groundwater
Results
A factor of 20
increase in overall
GWI possible (from
2 to ≈ 40 mill ha.)
70 % of recharge to environment 50 % of recharge to environment 30 % of the recharge to environment

14. Comparison with GW irrigated cropland in 2005 (Siebert et al. 2010)
Results
Area irrigated with groundwater in 2005 expressed
in ha. per cell adapted from Siebert et al. (2010)
Groundwater irrigation potential for the year 2000 expressed as
the percentage of the area irrigated with groundwater in 2005
scenario (b): 50 % of recharge goes to environment

15. Recharge
• Uncertainty
Country
Recharge (mm/yr)
FAO, AQUAStat,
2009
Döll and Fiedler,
2008
This paper
Burkina Faso 34.6 39 39
Ethiopia 18.1 39 80
Ghana 110.3 105 127
Kenya 6.0 46 29
Malawi 21.1 164 170
Mali 16.1 22 23
Mozambique 21.3 104 82
Niger 2.0 12 4
Nigeria 94.2 163 154
Rwanda 265.8 68 78
Tanzania 31.7 93 90
Uganda 122.9 95 50
Zambia 62.4 108 117
Discussion

16. Recharge
• Uncertainty
• Variability
Discussion
both over the period 1960-2000 (data from Wada et al., 2011)
Average annual recharge (mm/year) coefficient of variation (%)

17. Recharge
• Uncertainty
• Variability
Groundwater available:
• 3 scenarios (in km3 year-1)
Limitations of approach
• Non-limiting condition for other fundamental physical properties
e.g. soil, water quality, terrain slope, groundwater accessibility, …
• Socio-economics constraints
e.g. investment capacity, infrastructure, …
• Groundwater irrigation potential for 2000
No consideration of climate change and population/livestock growth, change in cropping pattern,
improvement in irrigation efficiency
Scenario 1 Scenario 2 Scenario 3
Min. Average Max. Min. Average Max. Min. Average Max.
442.2 692.1 990.1 751.1 1168.3 1664.9 1006.1 1644.5 2339.7
Discussion

18. A continental-scale distributed map of GWIP has been produced for the first time.
Is there potential for irrigation with groundwater in Africa?
• Yes, there is, up to a factor of 20 increase, relative to present GWI area (2  40 mill ha.)
• Map shows large spatial variability across Africa, and even within countries
• GWIP from renewable GW strongly related to GW recharge, cropland availability, and present
demand on GW
• GWIP is particularly significant and relevant in the semi-arid Sahel and East African corridor,
especially for small-scale and smallholder irrigation, with huge poverty alleviation potential
• In areas like Northern Africa and South Africa, GWIP from renewable GW has already been
exhausted
• In central Africa, GWIP is huge, but largely irrelevant
• Further research required for more detailed and localised assessments, including knowledge of
socio-economic factors
• Actual GWIP will be greatly influenced by irrigation efficiency and crop choices
• Climate change might affect GW recharge and increase crop water demand
Conclusion

19. References
• Altchenko Y. and Villholth K.G (2015) Mapping irrigation potential from renewable groundwater in Africa – a quantitative
hydrological approach. Hydrol. Earth Syst. Sci., 19, 1-13, Doi: 10.5194/hessd-19-1-2015
• Doll P. and Fiedler K. (2008) Global-scale modeling of groundwater recharge, Hydrol. Earth Syst. Sci. 12, 863–885 pp
• MacDonald A, Bonsor H, Dochartaigh B, Taylor R (2012) Quantitative maps of groundwater resources in Africa. Environ. Res.
Lett. 7, doi:10.1088/1748-9326/7/2/024009
• Pavelic P., Smakhtin V., Favreau G., and Villholth K.G. (2012). Water-balance approach for assessing potential for smallholder
groundwater irrigation in Sub-Saharan Africa. Water SA, 38(3), 399-406. doi.org/10.4314/wsa.v38i3.18
• Siebert S., Burke J., Faures J.M., Frenken K., Hoogeveen J., Döll P., and Portmann F.T. (2010). Groundwater use for irrigation - a
global inventory. Hydrol. Earth Syst. Sci. Discuss., 7, 3977–4021.
• Wada Y., Van Beek, L, Viviroli D., Dürr H., Weingartner R. and Bierkens M.: Global monthly water stress: 2. Water demand and
severity of water stress, Water Resour. Res., 47, W07517, doi:10.1029/2010WR009792, 2011.

20. Thank You
Merci
PhotobyX.Cai/IWMI
PhotobyK.Villholth/IWMI
PhotobyK.Villholth/IWMI
PhotobyY.Altchenko/IWMI
http://www.slideshare.net/IWMI_media