OR/15/009 Appendix

Lapworth D J, Carter R C, Pedley S and MacDonald A M. 2015. Threats to groundwater supplies from contamination in Sierra Leone, with special reference to Ebola care facilities. British Geological Survey Internal Report, OR/15/009.

Table A1 Groundwater quality surveys in representative regions in Sub-Saharan Africa (n=51), adapted from Lapworth et al (2015).

Area	Geology	Sample sites (n)	Results from selected water quality parameters*		Sampling time frame	Conclusion and sources of contamination	Reference
2Bombali, Sierra Leone	Granitic Basement	Wells (60)	FC 0-80, mean 16.6 SEC 38-554 NO3 25-280 Turb, and other majors, pH <6.5		Single study during the wet season May–June 2010	Wells contaminated with FC, 60% above. Who standards. Low pH concern for corrosion.	Ibemenuga and Avoaja (2014)
3Njala, Sierra Leone	Granitic Basement	Springs and wells (8)	FC 50–39k, mean 3.2k FS 5–2k		Monthly Wet and dry season sampling	Increased contamination during the onset of dry season and at the start of rainy season	Wright (1986)
3Moyamba, Sierra Leone	Granitic Basement	Springs and shallow wells (13)	FC 15–251k FS 12–63k, mean 501 SEC 7.6–206, mean 30 Turb, pH 5–6.5		Transition from dry to wet season, multiple sampling occasions	Increase risk during onset of wet season sustained risk during dry season for wells. No sanitation, open defecation practiced.	Wright (1982)
3Bo, Sierra Leone	Granitic Basement	Wells (33) lined and unlined	FC 0–75, mean 19.6 NO3 0.5–28, mean 7.7 PO4 0.01–11.5, mean 1.7 SEC 39–1281, mean 362		Wet season	Distance from field significant predictor of FC, not distance from toilet/PL	Jimmy et al. (2013)
3Conakry, Guinea	Volcanic rocks, fissured	Wells (69)	Mod.wells FC 370-1x105 FS 90-9k NO3 2-46 NH4 0.06-7 Cl 17-130 F 0-0.16 Turb. 1-70	Trad. wells FC 50-2 x105 FS 150-2 x104 NO37-51 NH4 0.01-8 Cl 8-284 F 0.0.38 Turb. 1-63	Dry season April-May 1994	Widespread contamination by nitrate and FC linked to poor sanitation and well construction	Gélinas et al. (1996)
2Various. Ivory coast	Basement	Boreholes (230)	NO3 mean 69		1981 and 1982	High nitrate (up to 200 mg/L) linked to domestic pollution and deforestation	Faillat (1990)
2Bolama City, Guinea Bissau,	Sandy soils and Cenozoic–Modern sediments	Wells (28)	SEC 27-326, mean 136 Turb. 1-26, mean 6.5 TC 0-23000, mean 2306 FC 0-5000, mean 410 Fecal Enterococci 0-850, mean 74 NO3 0.9-55.3, mean 16.6 NH4 0.01-1.37, mean 0.11 NO2 0.03-0.13, mean 0.04 Cu, Fe, Cr, As,		July 2006	80% of wells contaminated with FC linked to widespread use of PL	Bordalo and Savva-Bordalo (2007)
2Cotonou, Benin	Quaternary to mid Pleistocene sandstone	Dug wells in upper aquifer in densely populated area (379)	SEC 320-1045 Mn 0.06-0.19 NO3 10.4-118 PO4 <0.05-21.6 SO4 3.14-86.3		May 1991, August 1991 and April 1992	High P and K concentrations in upper aquifers linked to anthropogenic pollution	Boukari et al. (1996)
1Kumasi, Ghana	Precambrian Basement	Hand-dug wells (10)	TDS 6-230, mean 113 NO3 0-0.968, mean 0.16 PO4 0.67-15, mean 7.8 TH 8-103, mean 54 TC and EC <20		N/A	Water quality survey showed that water quality parameters were within WHO drinking water guideline values	Nkansah et al. (2010)
3Kumasi, Ghana	Precambrian Basement	Borehole and wells in peri-urban communities (9)	Fe 0.001-0.955 Mn 0.018-0.238 Pb 0.005-0.074 TC 3-16.8×106 FC 1.5-4.37×104 Enterococci 1.3-53.5		Monthly between Dec 2000 and Jan 2001	Poor quality overall, contamination linked to proximity to PL and refuse tips as well as livestock	Obiri-Danso et al. (2009)
3Ilesha, Nigeria	Basement	Wells (86)	Mean results: NO3 35 Cl 34 SO4 2.8		Single survey	Evidence of anthropogenic impact on water quality degradation using PCA	Malomo et al. (1990)
1Benin City, Nigeria	Quaternary to mid Pleistocene sandstone	Boreholes and open wells (6)	Pb 0.03-0.25 Zn 0.98-7.19 Cr 0.02-1.1 Cd Nd-0.23 FC 4600-240000 FS 600-35000		Single survey	Elevated Pb, Cr, Cd and Zn attributed to indiscriminate waste disposal and FC occurrence linked to PL, soak- always and septic tanks	Erah and Akujieze (2002)
2Calabar, Nigeria	Tertiary to recent sands and gravels	Existing wells (20)	BOD 0.06-4.09, mean 1.72 N 0.09-3.5, mean 2.15 Cl 0.1-1, mean 0.45 FC 0.75-4.32, mean 1.86		N/A	FC, nitrate and Cl had a positive correlation with urbanisation	Eni et al. (2011)
1Ibadan, Nigeria	Basement, banded gneiss and schist	Existing wells (N/A)	TSS 159-186.6, mean 174 Cl 1.1-10, mean 5 TC 2300-9200, mean 5120		Dry season	Gross pollution of groundwater attributed to poor well construction, PL and waste management	Ochieng et al. (2011)
2Ibogun, Pakoto, Ifo, Ogun State, Nigeria	Cambrian basement geology and weathered regolith	Dug wells, communities of 5000-20,000 people (20)	TDS 100-2200 TH 6-246 NO3 0.8-88 TC 0-0.6 (cfu x105) FC 0-0.2 (cfu x105) FS 0-0.7 (cfu x105)		July–August 2009	Water quality standards for nitrate, FC, FS not met for significant proportion of wells	Adelekan (2010)
1Lagos, Nigeria	Alluvium over sedimentary	Urban wells (18)	TDS 79-1343, mean 514 TH 24-289, mean 110 Na 8-274, mean 79 NO3 0.05-1.51, mean 0.4 Pb 0-1.9, mean 1.6 Zn 0-4.2 mean 0.3		Survey August to October 2004	Sources of contamination included sanitation, textiles, pharmaceuticals, food, tanneries, motor industry	Yusuf (2007)
1Surulere, Lagos, Nigeria	Alluvium over sedimentary	Wells and boreholes in a middle class area (49)	Al 1-99 µg/L Cd 1-98 µg/L Pb 1-24 µg/L		July 2009	Pb and Cd above WHO drinking water standards in >30% of sites	Momodu and Anyakora (2010)
1Abeokuta, Nigeria	Basement igneous and metamorphic	Shallow wells including sanitary survey (40)	All bacterial count>20 Maximum 800 EC+PA+SAL		December 2005	Shallow groundwater is highly contaminated with bacteria. Sources include pit latrines, livestock and solid waste	Olabisi et al. (2008)
2Abeokuta, Nigeria, urban & peri-urban	Basement igneous and metamorphic	Shallow wells (76)	Urban (mean) TDS 402 TH 30.3 NO3 12.02 PO4 0.21 Pb 0.25 Zn 0.12 TC 10500	Peri-urban (mean) TDS 263 TH 31.7 NO3 10.7 PO4 0.03 Pb 0.19 Zn 0.09 TC 10000	Dry season	Mean values for Pb, nitrate EC and TC > WHO standards. Trading, textiles, transport, cottage industries, pit latrines Generally higher in dry season	Orebiyi et al. (2010)
1Peri-urban area, Abeokuta, Nigeria	Basement igneous and metamorphic	Hand-dug wells (25)	TDS 50-270, mean 163 NO3 2.97-40.7, mean 17.6 NH4 0-0.59, mean 0.11 PO4 12-86 µg/L , mean 46 TH 12-210 , mean 106		Rainy season 2008	Direct surface run off into wells is suggested as possible contamination source	Taiwo et al. (2011)
1Warri River plain, Delta, Nigeria	Alluvial Benin formation	Boreholes near WW treatment plant	TDS 16-81 COD 0.4-44.4 NO3 0.3-1.2 Fe 0.05-0.15		2 year sampling campaign	River infiltration, municipal wastewater, agriculture, oil industry	Ibe and Agbamu (1999)
1Warri River plain, Delta, Nigeria	Quaternary and older sedimentary sequences	Dug wells	Fe 0.32-2.75 Pb 0.058-0.443 Ni 0.008-0.188 V 0-4 Cr 0-9 Cd 0.75-8.5 Zn 0-1.8		N/A	Pb, Ni exceed WHO standards. Sources include Warri River, settlement, refinery. Highest values in village 3 km from refinery	Aremu et al. (2002)
1Masaka, Nigeria	Cretaceous sandstone and clay	Dug wells, high density (12)	TDS 528-935 NO3 44.5-92.5 Alk 67-179 Cl 41-118 Fe 0.085-0.199 Cr 0.005-0.0126 TC 25900-78400		Samples taken in wet season	WHO standards exceeded for a range of contaminants including nitrate, TDS, Cr, Cd and TC. High density settlement with shallow water table	Alhassan and Ujoh (2011)
2Yaounde, Cameroon	Basement	Springs and wells in high density area (> 40)	SEC 18.2-430, mean 87 FC 60% >100 FS 5%>100		One-off survey	Groundwater’s in high density zones show significant degradation (chemical and microbiological), linked to PL	Ewodo et al. (2009)
2Douala, Cameroon	Alluvium over Pliocene sand and gravel	Springs , wells and boreholes (72)	SEC 25-362 NO3 0.21-94.3 FC 0-2311		One-off survey	High levels of FS indicative of contamination from PL, related to age and density of settlement	Takem et al. (2010)
2Kinshasa, DR Congo	Alluvial and sedimentary sequences	Wells including sanitary survey	Dry season TDS 180-450 NO3 76-118 PO4 0.53-4.6 TH 110-149 Pb 0.04-0.09 Cd 0.13-0.20		Wet season TDS 200-710 NO3 97-198 PO4 3.6-14.6 TH 17-52.5	Latrines, metal works, solid waste dumps are main sources of contamination	Vala et al. (2011)
2Dakar, Senegal	Quaternary	Wells (56)	NO3 0-122		July-October 1997	Nitrate contamination from point-source seepage in urban areas	Cissé Faye et al. (2004)
2Mekelle, Ethiopia	Mesozoic sediments	Wells, springs and boreholes (100)	SEC 542-5300 TDS 330-3454 NH4 0.01-2.38 NO3 0.21-336 Cl 5.76-298 F 0-1.27, PO4 0.001-0.58		N/A	Highly variable water quality indicative of a range of redox zones and sources of contamination	Berhane and Walraevens (2013)
2Bahir Dar, Ethiopia	Weathered and fractured Alkaline Basalt	Dug wells and protected pumps in inner, middle and outer zones (8)	Middle and inner city TDS 20-600 NO3 0.18-57.2 NH4 0-12 Cl 46-270 FC 93% of sites Mean 1.5 log cfu EC 80% sites mean 1.4 log cfu	Outer city TDS 20-70 NO3 0.08-8.8 NH4 0-12 Cl 0-40	Sampling over a 5 month period 2006/2007	Groundwater contamination linked to population density and urbanisation. All dug wells and boreholes had microbiological contamination in excess of WHO/EU standards. Dug wells had significantly higher FC.	Vala et al. (2011)
1Addis Ababa, Ethiopia	Volcanics	Boreholes and springs (9)	Alk 8-41 NO3 0.72-35 NO2 <0.01 COD 6.8-41 Cl 6.8-28 PO4 <0.03-0.1 Pb 4.6-25 SEC 300-1200 TC 0-34000		Various	The authors made a link between the surface water quality and groundwater quality. Major sources of contamination inferred were domestic waste, and industrial pollution from textile industry and petrol stations	Abiye (2008)
1Addis Ababa, Ethiopia	Volcanics	Springs and boreholes (10)	Zn 0.87-146 Ni 0.31-0.98 Cu 0.44-1.82 Pb 4.3-56.2 Cd <0.1-0.2 Co <0.1-0.12		2002	Geogenic sources of heavy metals is the likely sources of groundwater contamination in this setting due to high heavy metal concentrations in soils and rocks	Alemayehu (2006) Goshu and Akoma (2011) Goshu et al. (2010)
1Addis Ababa, Ethiopia	Volcanics	Springs and wells (63)	Ni 2-152 µg/L Pb <1 Co 0.5-165 As <3 Zn <20-2100 Cu 1.5-164 Cd 0.3-12.3 Cr 18.2-214		Februrary-March 2004, July to September 2005	Urban area, leaching from polluted soils.	Demlie and Wohnlich (2006)
3Kisumu, Kenya (urban)	Sedimentary	Existing wells (191)	TTC 0->100k mean 894 NO3 0.06-45 mean 15 Cl 0-225 mean 796 F 3-29.6 mean 6.2		1998 and 2004	Density of PL within a 100 m radius was significantly correlated with nitrate and Cl but not FC (PC)	Wright et al. (2013)
2Lichinga, Mozambique and Timbuktu, Mali	Quaternary/ Basement gneiss-granite complex	Hand dug wells: Timbuktu(31), Lichinga (159)	Timbuktu SEC 221-2010 NO3-N 35 med Cl 500	Lichinga SEC 220 med NO3 5.6 med Cl 13.5	Timbuktu September 2002 to May 2003 Lichinga, April 2002–August 2004	Contamination of groundwater sources from on site sanitation traced using N:Cl	Cronin et al. (2007)
3Lichinga, Mozambique	Mudstone	Lichinga (25)	TTC, EF (Enterococi)		Monthly for 1 year	Higher risk at onset of the wet season and end of the dry season. Predominant source was from animal faeces rather than PL or septic tanks. (LR)	Godfrey et al. (2006)
2Kampala, Uganda	Weathered Basement	Wells and springs	High density NO3 mean 67 Cl mean 59 TC mean 14	Low density NO3 mean 22 Cl mean 21 TC mean 544	Contrasting hydrological conditions	Significantly higher contamination in high density regions compared to low density	Barrett et al. (1998)
3Kampala, Uganda	Weathered Basement	Springs (25)	TtC (FC) FS BLD-23000		Monthly between September 1998-March 1999	Evidence of rapid recharge to springs following rainfall. Local environment hygiene and improved sanitary completion shown to be more important than on-site sanitation for spring protection (LR)	Howard et al. (2003)
3Kampala, Uganda	Weathered Basement	Monitoring wells (16)	Dry season SEC 272-345 P BDL-0.11 N BDL-5.5 NO3 24-144 Cl 31-50.5 TC 0-131 FC 0-35	Wet Season SEC 280-372 P BDL0.04 N BDL-263 NO3 24-692 Cl 28-192 TC 29-10000 FC 6-8300	2003: weekly March-May and September in dry season, and June to August, wet season.	High population density with pit latrines and livestock sources identified. Microbiological water quality deterioration after heavy rainfall	Barrett et al. (1998)
1Kampala, Uganda	Weathered Basement	Boreholes and wells (28)	Limited inorganic and organic suit, no microbiology		September and October 2011	Nitrate concentrations suggest poor sanitation and diffuse contamination.	Nachiyunde, Kabunga et al. (2013)
3Uganda, Kampala (urban)	Weathered basement	Piezometers (10)	1.5 m down gradient of pit latrines NO3 5-90 Cl 50-1100 PO40.1-2 NH4 5-40		March-August 2010 biweekly sampling	PL found to be a significant source of nutrients (N) compared to waste dump. NH4 removal by nitrification	Nyenje et al. (2013)
1Lusaka, Zambia	Dolomite	Wells and streams in intensely urbanised area (9)	SEC 200-710 NO3 <0.1-43 NH4 <0.25-3.5, Cl 4.6-36 PO4 <0.1-4, B <1-10, As <0.2-0.49 Pb 0.14-0.67, Hg <0.4-13		July 2001	Values for nitrate and Hg were in excess of WHO standards on some occasions. Poor sanitation and solid waste disposal implicated.	Cidu et al. (2003)
2Lusaka, Zambia	Dolomite	Boreholes (7)	FC 0-45 TC 0-58 SEC 401-1060		Single survey	Evidence for contamination in health centre boreholes by FC, poor waste management implicated	Nkhuwa (2008)
3Lusaka,Zambia	Dolomite	Private and public boreholes (N/A)	Alk 124-564, NO3 0.03-39, NO2 0.002-42, NH4 0.08-60 Cl 42-102, TC 1-TNTC FC 21-TNTC, BOD 2-69 COD 9-320		Various: 1995-2000	Hydrochem, microbiology and incidence of cholera outbreaks compiled to show the rapid deterioration of GW sources associated with poor sanitation	Nkhuwa (2003)
2Ndola, Zambia	Dolomite and basement lithologies	Wells (123) and boreholes (60) surface waters (41)	Wells (median) TC 7 Zn 11.4	Boreholes (med) TC 0 Zn 139	April–June 2013	Geological control on trace metal contamination. TC for wells>boreholes but no FC data collected.	Liddle et al (2015)
3Kabwe, Zambia	Dolomite and basement	Private (13) and public (12) boreholes, private wells (57)	Dry season Wells NO3 0.1-187 (18) FC 10-6800 (180) Boreholes NO3 0.1-38 (6) FC <2-28 (<2)	Wet season Wells NO3 0.15-174(22) FC 2-27600 (570) Boreholes NO3 0.1-41 (6) FC <2-760 (<2)	Dry and wet season 2013–2014	Widespread NO3 and FC contamination in shallow wells in both wet and dry seasons, wet>>dry. Generally good quality in peri-urban boreholes but evidence of contamination in some urban boreholes	Lapworth et al (2015)
3South Lunzu, Blantyre, Malawi	Weathered basement	Borehole, springs and dug well (9)	Dry season SEC 210-330 Cl 21-35 Fe 0.1-0.8 FC 0-5200 FS 0-640	Wet season SEC 306-383 Cl 14-29 Fe 0.4-0.7 FC 0-11,000 FS 0-7000	Wet and dry season on two occasions	Groundwaters highly contaminated due to poor sanitation and domestic waste disposal. 58% of residence use traditional PL	Palamuleni (2002)
3Southern Malawi	Weathered basement	Shallow wells (26)	Dry season NO3 0-2.6 NH4 detectable most samples FC 0-9k TC 0-17k As, F also	Wet season NO3 0-4.4 TC 0-77k FC 0-9k	Wet and dry season	Overall contamination levels higher during wet season for two districts and lower for one district and significantly higher in unprotected sources.	Pritchard et al. (2008)
2Tamatave and Foulpointe, Madagascar	Weathered basement and unconsolidate d sediments	Boreholes (53)	FC 73%>0, 55% 0-10, 54%>10 NO3 4.4-35, mean 23 Pb 1-215, mean ca. 5		One-off survey	Widespread drinking water contaminated with FC and concerns over Pb from pump materials	MacCarthy et al. (2013)
3Epworth and Harare, Zimbabwe	Granite	Wells andboreholes, transect of formal and informal zones (18)	NO3 0-30, mean 11 PO4 0-27.2, mean 3.03 FC 0-2, mean 0.75 (cfu x104)		Survey carried out withduplicate sampling	Pit latrines, faecal coliforms in older and informal trading areas, urban agriculture, home industries and commercial areas	Zingoni et al. (2005)

SEC-specific electrical conductivity, PCA=Principal component analysis, LR= logstic regression, TDS= total dissolved solids, TH=total hardness, , BOD-biological oxygen demand, COD=chemical oxygen demand, FC=faecal coliforms, EC= E. Coli, TC=total coliforms, FS=faecal streptococcus. Microbiological units as cfc/100 mL unless stated otherwise, TNCT=too numerous to count, BDL=below detection limit. Notation: ¹Case-studies presenting data from a limited number of sites (n<20), limited temporal resolution as a single survey or use only basic chemical indicators and limited analysis of the results; 2 Case studies which either draw from larger data sets or include both chemical and microbiological indicators but have limited data analysis regarding sanitary risk factors; 3 Case studies with greater temporal resolution or are accompanied by a more thorough analysis of the data, for example using statistical techniques to understand the significance different risk factors on water quality observations.

Table A2 Conceptual framework for hazard sources and pathways for groundwater and surface waters.

	Surface water	Traditional wells a	Springs	Improved wells	Boreholes b
Major hazard sources	Surface sources: These include open defecation by humans and animals, surface soil amendments, sewers, shallow drains and surface application of waste water	Surface sources: Same as for surface waters, materials used to draw water from collector contaminated with soil microbes and sanitary sources from hands Subsurface sources: These include all buried sources of solid and liquid waste (e.g. pit latrine, soak away, waste dump, and cemetery).	Surface sources: Same as for surface waters, materials used to draw water from collector contaminated with soil microbes and sanitary sources from hands Subsurface sources: These include all buried sources of solid and liquid waste.	Surface sources: materials used to draw water from collector contaminated with soil microbes and sanitary sources from hands Subsurface sources: These include all buried sources of solid and liquid waste.	Subsurface hazards: These include buried sources of solid and liquid waste (e.g. pit latrine, soak away, waste dump, and cemetery).
Major hazard pathways	Surface runoff, open sewer systems	Surface runoff directly into well, bypass pathway from use of contaminated materials (e.g. rope or bucket). Vertical and horizontal soil flow from buried hazard sources.	Surface runoff directly into spring collector, bypass pathway from use of contaminated materials (e.g. bucket). Vertical and horizontal soil flow from buried shallow hazard sources.	Vertical and horizontal soil and groundwater flow to well. Crack in sanitary seal, well lining. Bypass pathway from use of contaminated materials to draw water.	Horizontal groundwater flow in saturated zone to borehole intake.
Hazard susceptibility under high groundwater table conditions	High at all times	Very high due to lack of barrier to horizontal soil and shallow groundwater flow to well.	High due to limited soil attenuation and potential activation of shallow rapid horizontal pathways to spring	Moderate due to some protection from shallow horizontal soil and groundwater flow by casing. Some attenuation in saturated zone	Low due to narrow diameter of casing and generally deeper casing, and high attenuation capacity in saturated zone
Hazard susceptibility to extreme rainfall conditions	High due to strong link to runoff sources of contamination and limited attenuation potential	Very high due to strong link to runoff sources of contamination.	High due to strong link to surface runoff sources of contamination, difficulty in protecting spring catchment from encroachment by animals	Moderate due to reduced lateral pathways in soil and shallow groundwater. Erosion or bypass of sanitary/annular seal possible, large diameter means this is more likely	Low due to limited rapid pathways from surface or buried sources of hazards
Possible interventions for safer supply	Not suitable for drinking without treatment at household level*.	May be best to stop using unless there is no alternative source of water. Install well casing, sanitary seal, cover and use of alternative water lifting device such as hand pump. Generally not suitable for drinking without household treatment*.	Improved citing of springs and ensure better spring protection in surface capture zone, very difficult to manage in rural areas, this is not realistic in urban/peri-urban areas. Generally not suitable for drinking without treatment*.	Improved citing of wells in relation to sources of hazards. Stop main pathway from surface through use of rope and bucket, e.g. cap and install hand-pump. Deepen casing and improve sanitary seals. Often not suitable for drinking without treatment*.	Improved citing of borehole in relation to sources of hazards. Maintenance: replace cracked casing, ensure adequate sanitary seals are maintained. Often suitable for drinking without treatment if well maintained/cited.

^a Hand dug wells with no surface protection, b Assuming that the initial installation of a borehole is of a high standard, *Regular household treatment is not realistic in Sierra Leone or many other countries in SSA, if there is high turbidity (likely for surface waters) this may render treatment using chlorination only partially effective.