<<

UC Berkeley UC Berkeley Previously Published Works

Title Pit and their impacts on quality: a systematic review.

Permalink https://escholarship.org/uc/item/7dn8t3qn

Journal Environmental health perspectives, 121(5)

ISSN 0091-6765

Authors Graham, Jay P Polizzotto, Matthew L

Publication Date 2013-05-01

DOI 10.1289/ehp.1206028

Peer reviewed

eScholarship.org Powered by the California Digital Library University of California Review

Pit Latrines and Their Impacts on Groundwater Quality: A Systematic Review Jay P. Graham1,2 and Matthew L. Polizzotto3 1Department of Environmental and Occupational Health, and 2Department of Global Health, George Washington University School of and Health Services, Washington, DC, USA; 3Department of Soil Science, North Carolina State University, Raleigh, North Carolina, USA

with groundwater­ contamination by pit Ba c k g r o u n d : Pit latrines are one of the most common human excreta disposal systems in latrines. In particular, we a) calculated global low-income countries, and their use is on the rise as countries aim to meet the -related pit coverage, b) systematically reviewed target of the Millennium Development Goals. There is concern, however, that discharges of chemi­cal empirical studies of the impacts of pit latrines and microbial contaminants from pit latrines to groundwater may negatively affect human health. on ground­water quality, c) evaluated latrine Ob j e c t i v e s : Our goals were to a) calculate global pit latrine coverage, b) systematically review siting standards, and d) identified knowl- empirical studies of the impacts of pit latrines on groundwater quality, c) evaluate latrine siting edge gaps regarding the potential for and standards, and d) identify knowledge gaps regarding the potential for and consequences of ground­ water contamination by latrines. ­consequences of groundwater­ ­contamination by latrines. Me t h o d s : We used existing survey and population data to calculate global pit latrine coverage. We reviewed the scientific literature on the occurrence of contaminants originating from pit latrines Methods and considered the factors affecting transport of these contaminants. Data were extracted from peer-reviewed articles, books, and reports identified using Web of ScienceSM, PubMed, Google, and Global pit latrine coverage. We used exist- document reference lists. ing survey data to estimate the percentages of people per country who a) use pit latrines for Di s c u s s i o n : We estimated that approximately 1.77 billion people use pit latrines as their primary means of sanitation. Studies of pit latrines and groundwater are limited and have generally focused sanitation,­ b) do not have any sanitation­ facili- on only a few indicator contaminants. Although groundwater contamination is frequently observed ties, and c) use groundwater­ sources for drink- downstream of latrines, contaminant transport distances, recommendations based on empirical ing water [see Supplemental Material, Table S1 studies, and siting guidelines are variable and not aligned with one another. (http://dx.doi.org/10.1289/ehp.1206028)]. Co n c l u s i o n s : In order to improve environmental and human health, future research should Data from the most recent reports for each examine a larger set of contextual variables, improve measurement approaches, and develop better country were obtained from Demographic criteria for siting pit latrines. and Health Surveys (USAID 2012), Multiple Ke y w o r d s : groundwater, latrine, privy, sanitation, siting standards, water quality. Environ Health Indicator Cluster Surveys (UNICEF 2012), and Perspect 121:521–530 (2013). http://dx.doi.org/10.1289/ehp.1206028 [Online 22 March 2013] China’s Economic, Population, Nutrition, and Health Survey (WHO/UNICEF 2012a, 2012b). We included improved latrines [flush An estimated 2.6 billion people lack access into the ground and covered with a con- and toilets that pour/flush to pit latrines to —defined as facilities crete slab or floor with a hole through which (water is poured by hand for flushing), venti- that hygienically separate human excreta from excreta falls. Unimproved pit latrines are lated improved latrines, and pit latrines with human contact [World Health Organization those without slabs or platforms. slabs] and unimproved latrines (traditional (WHO)/UNICEF 2010)]. Improved sanita- In concert with sanitation goals, the latrines, pit latrines without slabs, and shared tion includes water-based toilets that flush UN has also set explicit targets to increase latrines) when estimating pit latrine use (see into sewers, septic systems, or pit latrines; the proportion of the global population Supplemental Material, p. 2, for definitions of simple pit latrines; and ventilated improved using an improved drinking-water source types of sanitation). Composting toilets, con- pit latrines. There is strong evidence that (WHO/UNICEF 2012c). In the context of sidered improved facilities, were not included access to improved sanitation can reduce diar- low-income countries, water from improved in our analysis, nor were sanitation facilities rhea morbidity and mortality as well as soil- sources is frequently derived from ground­ for which final disposal of human excreta is transmitted helminths (Albonico et al. 2008; water via protected springs, protected dug unknown (e.g., hanging latrines and bucket Cairncross et al. 2010b). , tube wells, and boreholes (UN 2008). latrines). For estimates of the proportions The United Nations (UN), through the Thus, the use of ground­water (which typically of improved versus unimproved latrines, we Millennium Development Goals, has set a receives no subsequent treatment to improve assumed that unspecified latrines were split target of halving by 2015 the proportion of quality) for supplies is increas- evenly between improved and unimproved. the population without sustainable access ing dramatically (Rosa and Clasen 2010). Data for people without a sanitation facility to improved sanitation (WHO/UNICEF Because of the increasing uses of both pit include “no facility” and “open 2012c). To achieve this target, approximately latrines and ground­water resources in low- 1 billion people in urban areas and 900 mil- income countries, there is concern that pit Address correspondence to J.P. Graham, School of lion people in rural areas must gain access to latrines may cause human and ecological Public Health and Health Services, Department of improved sanitation by 2015 over the base- health impacts associated with microbiological Environmental and Occupational Health, George line year, 1990 (WHO/UNICEF 2012c). In and chemical contamination of groundwater.­ Washington University, Washington, DC 20037 USA. Telephone: (202) 994-2392. E-mail: jgraham@ low-income countries [with a gross national Pit latrines generally lack a physical barrier, gwu.edu income per capita of ≤ US$1,025 (World such as concrete, between stored excreta and Both authors contributed equally to this article. Bank 2013)], many households use improved soil and/or ground­water (van Ryneveld and Supplemental Material is available online (http:// or unimproved pit latrines because of their Fourie 1997). Accordingly, contaminants dx.doi.org/10.1289/ehp.1206028). low cost and availability (Cairncross et al. from pit-latrine excreta may potentially leach We thank T. Barto, D. Galan, R. Hinton, and three 2010a; Jain 2011). Improved pit latrines into groundwater,­ thereby threatening human anonymous reviewers for helpful input to the analysis. The authors declare they have no actual or potential are the most basic and inexpensive form of health through well-water contamination. In competing financial interests. improved sanitation. They typically consist of this study, we assessed the known and mea- Received 18 September 2012; accepted 11 March a pit—circular, rectangular, or square—dug sured environmental health impacts associated 2013.

Environmental Health Perspectives • v o l u m e 121 | n u m b e r 5 | May 2013 521 Graham and Polizzotto in bush/field.” National survey data do not Results 2007). The largest chemical concerns from typically charac­terize shared facilities because Global pit latrine coverage. Globally, there is excreta disposed in on‑site sanitation systems they are considered unimproved sanitation. great variability in latrine coverage. We esti- are considered to be nitrate [British Geological Therefore, for shared sanitation, we applied mate that approximately 1.77 billion people Survey (BGS) 2002; Fourie and Vanryneveld the average proportion of facilities that were around the world use some form of pit latrine 1995; Pedley et al. 2006], phosphate (Fourie pit latrines (44%) based on seven national as their primary means of sanitation [Figure 1; and Vanryneveld 1995), and chloride (BGS surveys that provided more detailed informa- see also Supplemental Material, Table S1 2002) (see Supplemental Material, Table S3). tion (see Supplemental Material, Table S1). (http://dx.doi.org/10.1289/ehp.1206028)]. Microbiological contaminants associated Groundwater use comprised both improved In addition, we estimate that 48% of people with pit latrines. Concentrations of most fecal and unimproved modes of accessing ground­ using pit latrines use facilities charac­terized micro­organisms decline after excretion, but water, including tube wells and boreholes, as improved, whereas the remainder uses these microorganisms­ may still impair ground­ protected wells, protected springs, unprotected shared or unimproved facilities (e.g., tradi- water quality. Several approaches have been wells, and unprotected springs, but not cen- tional latrines or pit latrines without slabs). used to define the quantities and transport tralized water sources that may originate from The number of users per latrine varies by distances of latrine-derived microbial contami- groundwater.­ locale, but based on the excreta produc- nants. The majority of studies that assessed To calculate the global totals for pit latrine tion rates of Feacham et al. (1983), globally microbiological­ quality of groundwater­ in rela- use, we multiplied the country-wide percent- per day, as much as 2.1 billion kilograms of tion to pit latrines applied culture-based assays ages by the UN estimates of 2010 populations and 0.6 billion kilograms of are to measure fecal indicator bacteria (Table 1), (UN 2011) and summed all data presented deposited into latrines. In the countries where including total coliforms, fecal coliforms, and in Supplemental Material, Table S1 (http:// pit latrines are prevalent (see Supplemental Escherichia coli (previously known as Bacillus dx.doi.org/10.1289/ehp.1206028). We used Material, Table S1), > 2 billion people depend coli), which occur in high concentrations in our estimate of global latrine use in conjunc- on ground­water for their primary drinking the feces of healthy adults and have epide- tion with estimated excreta production rates of water supply. miological evidence to support their use as 1,200 g urine/person/day and 350 g wet feces/ These calculations are among the first indicators of water quality (Wade et al. 2003). person/day for rural settings estimates of the numbers of people using pit Caldwell conducted five experimental­ stud- (Feacham et al. 1983) to estimate daily quanti- latrines and groundwater­ in low-income coun- ies in the 1930s and included the colon ties of urine and feces deposited into latrines. tries. Because some national survey data are aerogenes group and anaerobic bacteria, in Review of studies on groundwater­ contami­ several years old, estimates have a fair degree addition to B. coli, in the analyses (Caldwell nation from pit latrines. To find relevant of uncertainty at the country level. However, 1937a, 1937b, 1938a, 1938b; Caldwell and documents describing ground­water contami- our estimate for the total number of people Parr 1937). Only one study analyzed viruses nation derived from pit latrines, we searched without any sanitation facility (1.11 billion) (adeno­virus and rota­virus) to charac­terize the Web of ScienceSM (http://webofknowl- is in agreement with the independently cal- ground­water quality in relation to pit latrines edge.com/), PubMed (http://www.ncbi.nlm. culated Joint Monitoring Program 2010 esti- (Verheyen et al. 2009). We found no studies nih.gov/pubmed), and Google (http://www. mate for (1.1 billion people) that assessed protozoa or helminths, which google.com/) using the following keywords: (WHO/UNICEF 2012c), which suggests that typically exhibit little movement in ground­ “pit latrine” AND “groundwater”;­ “privy” our approximations may be more robust at the water because of their size (Lewis et al. 1982). AND “groundwater”;­ “” AND “ground­ global level. In addition, our estimate of the The extent to which microbes from pit water”; “sanitation” AND “ground­water”; total 2010 population for countries included latrine wastes may be transported and con- “pit latrine” AND “aquifer”; “privy” AND in this analysis (5.22 billion) is consistent taminate groundwater­ largely depends on the “aquifer”; “toilet” AND “aquifer”; “sanitation” with the UN population estimate for “less environmental­ context of the area, particularly AND “aquifer”; “pit latrine” AND “ground developed regions” [or “all regions of Africa, hydrological and soil conditions. Nearly half water”; “privy” AND “ground water”; “toi- Asia (excluding Japan), Latin America and the of the studies assessing microbial contami- let” AND “ground water”; “sani­ta­tion” AND Caribbean plus Melanesia, Micronesia and nants used experi­mental approaches. These “ground water”; “pit latrine” AND “water Polynesia”] of 5.66 billion (UN 2011). studies included either the installation of test quality”; “privy” AND “water quality”; “toi- Studies on groundwater­ contamination wells to measure the quality of water sampled let” AND “water quality”; “pit latrine” AND from pit latrines. Twenty-four studies directly down­gradient of pit latrines, the collection of “well water”; “privy” AND “well water”; and assessed the transport of contaminants or soil samples, or both. Kligler (1921) sampled “toilet” AND “well water.” We also searched applied statistical methods to estimate a soil at varying distances from > 50 pit latrines the resulting reference lists and contacted measure of risk associated with the presence under wet and dry conditions. The maximum experts to identify additional articles. To pro- of pit latrines (Table 1); these studies assessed distance of bacterial contamination found was vide a critical review of the literature­ on the either chemical contaminants (4 studies), 5.5 m from latrines and occurred under wet occurrence of microbiological­ and chemical microbial contaminants (2 studies), or and sandy soil conditions. Kligler (1921) sug- contaminants originating from pit latrines, both (18 studies). Human excreta are the gested that a vertical distance of ≥ 3–4.5 m we more fully characterized the studies that main input to pit latrines, although other between the bottom of the pit and the water either directly assessed the fate and transport inputs may contribute significantly to pit table would maintain safe groundwater­ qual- of contaminants from pit latrines or studies contents depending on local practices [see ity. In several experimental­ studies on pit that applied statistical methods to estimate a Supplemental Material, Inputs to Pit Latrines, latrines and ground­water, Caldwell (1937a, measure of risk associated with the presence of p. 3, for additional details (http://dx.doi. 1937b, 1938a, 1938b) and Caldwell and pit latrines. By synthesizing existing results in org/10.1289/ehp.1206028)]. Parr (1937) found varying transport distances terms of siting guidelines for pit latrines and harbor a large number of microbes, including (ranging from 3 to 25 m) among B. coli (i.e., well installation, we identified research gaps bacteria, archaea, microbial eukarya, viruses, E. coli), colon aerogenes (i.e., total coliform that must be addressed in order to make better- and potentially protozoa and helminths (see bacteria), and anaerobes, depending on the informed decisions to protect water quality Supplemental Material, Table S2) (Feachem degree of soil saturation and the groundwater­ and safeguard human health. et al. 1983; Ley et al. 2006; Ramakrishna flow velocity. In a study of a latrine placed in

522 v o l u m e 121 | n u m b e r 5 | May 2013 • Environmental Health Perspectives The impact of pit latrines on groundwater quality an alkaline alluvium soil, Dyer (1941) reported quality in pre­existing wells and factors such source and at least 1 latrine within a radius that movement of total coliforms was limited as proximity of pit latrines to assess latrine of 50 m (Figure 2). These authors hypothe­ to < 7 m from the pit. A relatively short trans- impacts on groundwater. At a study site in sized that during the wet season, viruses were port distance was also found in , characterized by a shallow water table transported by ground­water flow in the upper where high fecal coliform counts [> 10 colony and fractured rock aquifer, high concentrations part of the soil, whereas viral transport in the forming units (cfu)/100 mL] were detected of fecal coliforms were found in domestic wells dry season was more likely a result of virus- only 1 m from a pit latrine (Still and Nash located near pit latrines and septic tanks (Pujari contaminated . 2002). Dzwairo et al. (2006) found fecal and et al. 2012). At a contrasting site, charac­terized Associations between ground­water con- total coliform­ contamination greatly reduced by alluvial formations, the authors detected tamination and factors related to sanita­tion > 5 m from pit latrines. no or low levels of fecal contamination (Pujari facilities are complicated by the co-occurrence In a study of 12 pour/flush latrines, Banerjee et al. 2012). In a geo­referenced spatial study of multiple contaminant sources, particu- (2011) found that transport of total and fecal of viral contamination, Verheyen et al. (2009) larly when information on groundwater­ flow coliforms increased during the monsoon period sampled 287 drinking-water sources (247 water patterns is not available. A study of ground­ and in sandy soils. The author noted that the wells, 25 pumps, and 15 surface water samples) water quality in an informal settlement of maximum travel distance of bacteria was 10 m proximate to 220 latrines. Adenoviral DNA found detectable total and fecal from pits (Figure 2). In contrast, in a study in was repeatedly detected in 26 water sources, coliforms in more than two-thirds of study Zimbabwe, Chidavaenzi et al. (1997) found and rotaviral RNA was detected in 1 source. boreholes and existing domestic wells (Zingoni that groundwater­ contamination was higher In multiple rounds of sampling, 40 of the et al. 2005). The abundance of pit latrines, in the dry season than in the wet season, with 287 drinking-water sources were positive for used in > 75% of the households, and the ­coliforms detected up to 20 m from a pit. viral contamination at least once. Verheyen presence of informal trading areas within the Nearly one-fourth of the studies analyzed et al. (2009) found a significant positive asso- settlement were likely sources of fecal pollu- associations between micro­biological water ciation between viral contamination of a water tion. The authors suggested that shallow wells

Population using pit latrines for sanitation 0–20% 21–40% 41–60% 61–80% 81–100% No data

Population using groundwater for drinking 0–20% 21–40% 41–60% 61–80% 81–100% No data

Figure 1. Percentage of low-income country populations using pit latrines as a primary sanitation facility (A) and groundwater as a primary drinking water source (B). Countries with no data presented were not included in the analysis.

Environmental Health Perspectives • v o l u m e 121 | n u m b e r 5 | May 2013 523 Graham and Polizzotto and boreholes in the study area, as well as the (2002) found fecal coliforms and streptococci­ liners as a way to reduce groundwater­ con- incomplete lining of most latrines, contributed in sediments 10 m below latrines. tamination from pit latrines, Nichols et al. to high levels of groundwater­ contamination Movement of bacteria from latrines is (1983) found fecal coliforms in soil samples (Zingoni et al. 2005). In a study conducted often limited by formation of a “scum mat,” taken adjacent to only one of five peat-lined in Moldova, Banks et al. (2002) concluded which develops around the latrine pit and pits, compared with three of three unlined that groundwater­ within villages was reduces the movement of fecal bacteria pits. The one peat-lined pit that showed con- likely caused by latrines, livestock and stored (BGS 2002; Caldwell 1937a). This mat (also tamination was located in shallow and rocky manure, solid-waste landfills, and leakage from referred to as a “biologically active layer,” soil and was under saturated conditions. wastewater pits. “biolayer,” or “clogged” zone) enhances bac- Chemical contaminants associated with Even in areas with a high density of pit teria removal through filtration and predation pit latrines. Nitrate. Because of high con- latrines, microbiologi­ cal­ groundwater­ contami­ by antagonistic organisms, but it may take centrations of nitrogen in human excreta, nation­ may not necessarily be detected. Three several months to develop around new latrines its adverse impacts to human health, and its studies found no strong positive association (Caldwell and Parr 1937). In addition, clog- use as an indicator of fecal contamination, between poor bacterio­logical water quality ging may result from blockage of soil pores by nitrate has been the most widely investi- and sanitary surveys or proximity to latrines solids that have been filtered out, swelling of gated chemical­ contaminant derived from (Ahmed et al. 2002; Howard et al. 2003; clay minerals, and precipitation of insoluble pit latrines. Consumption of high concentra- Tandia et al. 1999), although Ahmed et al. salts (Franceys et al. 1992). In a study testing tions of nitrate in drinking water is known to

Table 1. Summary of selected studies that assessed groundwater or soil contamination associated with pit latrines.a No. of latrines Experimental Subsurface Sampling Water quality Source Country in studyb design conditions time frame parametersc Conclusions Vinger et al. 2012 South Africa 15 Sampled existing No data June–July Ammonia, nitrate, nitrite Higher levels of contaminants wells observed at distances < 11 m from pit latrines Pujari et al. 2012 India 7 Sampled existing Fine loamy silt, Summer and Fecal coliforms, total No to low levels of nitrate and wells sandy loam, monsoon dissolved solids, nitrate fecal coliforms observed intermittent clay seasons Banerjee 2011 India 12 Installed test Saturated and Premonsoon Total coliforms, fecal Movement of chloride tracers and wells unsaturated soils and monsoon coliforms, chloride coliforms limited to < 10.2 m of gravel, sand, seasons solution used as tracer from pits silt, clay, and laterite Verheyen et al. Benin 220 Sampled existing No data Wet and dry Adenovirus, rotavirus Viral contamination of 2009 wells seasons, groundwater associated with 2003–2007 latrine proximity Dzwairo et al. Zimbabwe 3 Installed test Saturated and February–May Ammonia, nitrate, Fecal coliform movement greatly 2006 wells unsaturated sandy 2005 turbidity, pH, reduced > 5 m from pits; all soils conductivity, total nitrate levels and 99% of coliforms, fecal ammonia levels met WHO coliforms drinking water standards Zingoni et al. Zimbabwe Not specified Sampled existing No data No data Na, Zn, Cu, Co, Fe, Elevated levels of nitrate and 2005 wells and phosphate, nitrate, coliform bacteria in most parts installed test total coliforms, fecal of study area wells coliforms Mafa 2003 Botswana Not specified Sampled existing Fractured rock July and Broad set of Elevated levels of nitrate in wells overlain by August 2000 hydrochemical analyses several zones where pit latrines alluvial sediment, were common clay, sand, and weathered rock Banks et al. 2002 Kosova, Not specified Sampled existing No data 1996–2000 Chloride, sulfate, Elevated levels of nitrate likely Moldova, wells and potassium, nitrate from latrines Siberia springs Howard et al. Uganda Not specified Sampled Highly variable: clay Monthly, Fecal streptococci, fecal No significant relationship 2003 protected to sandy soils March 1998 coliforms, nitrate between microbiological springs through April contamination and pit latrine 1999 proximity Still and Nash South Africa 1 Installed test No data Bimonthly, Fecal coliforms, nitrate Low levels of nitrate (< 10 mg/L) 2002 wells 2000–2002 and fecal coliforms (10 cfu/100 mL) found > 1 m of latrine Ahmed et al. 2002 Bangladesh Not specified Sampled existing Two aquifer systems; 2- to 8-week Fecal streptococci, fecal Bacteriological water quality wells clay, silt, and fine intervals, coliforms, broad set of generally good (< 10 fecal to coarse sand 1998–1999 hydrochemical analyses coliforms/100 mL); water quality poorly correlated with sanitary surveys Chidavaenzi et al. Zimbabwe 2 Installed test Stratified fine-grain Wet and dry Nitrogen, coliforms Rapid reductions in coliform, 2000 wells sandy soils seasons sulfate, and nitrogen levels within 5 m from pits; contami­ nation present up to 20 m

Table continued

524 v o l u m e 121 | n u m b e r 5 | May 2013 • Environmental Health Perspectives The impact of pit latrines on groundwater quality cause methemoglobinemia, and associations ground­water nitrate concentrations near Vinger et al. 2012); and nitrate can be formed with cancer in humans have been observed, latrines were above local background lev- and lost through natural­ soil processes (Jacks although not consistently (Fewtrell 2004; els, even if they remained below or near the et al. 1999). Jacks et al. (1999) used mass- WHO 2011). The WHO-recommended WHO guideline (Baars 1957; Caldwell and balance calculations to estimate that 1–50% guideline for nitrate in drinking water is Parr 1937; Chidavaenzi et al. 2000; Jacks et al. of nitrogen leached to ground­water from 50 mg/L (WHO 2011). Concentrations of 1999; Zingoni et al. 2005). latrines in Botswana. Although significant nitrate in well water near latrines are highly High nitrate concentrations have been quantities of leached nitrate may have been variable. Although a number of studies that attributed to latrines through association and lost to denitrification in poorly drained soils, detected total or fecal coliforms did not assumptions based on general proximity, but the calculations suggested that nitrogen loss detect elevated nitrate concentrations in wells pinpointing­ the actual sources of nitrate in from latrines helped describe the high nitrate (Ahmed et al. 2002; Dzwairo et al. 2006; groundwater­ has proved challenging (WHO concentrations of ground­water (50 mg/L) in Howard et al. 2002; Padmasiri et al. 1992; 2006). Nitrate may be derived from numerous the area. The authors concluded that moving Still and Nash 2002), other studies have potential sources in urban and rural environ­ drinking wells outside of the habituated area reported nitrate concentrations > 100 mg/L ments, including latrines, plant debris, animal would help avoid nitrate contamination of (Banks et al. 2002; Girard and Hillaire-Marcel manure, garbage repositories, livestock pens, drinking water. 1997; Lewis et al. 1980; Mafa 2003; Pujari soil, and fertilizers (Girard and Hillaire-Marcel, Girard and Hillaire-Marcel (1997) used et al. 2012; Tandia et al. 1999). Frequently, 1997; Howard et al. 2002; Melian et al. 1999; nitrogen isotopes to determine the source of

Table 1. Continued. No. of latrines in Experimental Subsurface Sampling Water quality Source Country studyb design conditions time frame parametersc Conclusions Jacks et al. 1999 Botswana 4 Sampled existing Well-drained and No data Phosphorous, nitrogen Variable nitrate leaching from pit wells poorly drained soils isotopic ratios, chloride latrines Tandia et al. 1999 Senegal Not specified Sampled existing Fine to coarse sand July and Broad set of Nitrate contamination in water wells November hydrochemical strongly correlated with latrine 1989 analyses, fecal proximity coliforms Nichols et al. USA 8 Installed test 3 latrines on clayey June and Nitrate, phosphorus, Latrines with peat liners reduced 1983 wells soil; 3 on shallow August fecal coliforms movement of phosphorus and loam; 2 on sand; all 1975–1979 fecal coliforms but not nitrate. soils well-drained Lewis et al. 1980 Botswana 30 pit latrines in Sampled existing Clayey soils and October 1977 Broad set of Contamination of wells near the study area wells and test fissured rock through hydrochemical latrine with E. coli and nitrate; wells February analyses, E. coli, rapid transport of chloride tracer 1978 chloride solution used as tracer Baars 1957 Netherlands Not specified Sampled soil and Unsaturated sandy September Ammonia, E. coli, nitrate Contamination in soil samples existing wells soils 1951 and limited to < 1.5 m from latrines January and March 1952 Dyer 1941 India 1 Installed test Saturated and December– Chloride, nitrate, total Movement of total coliforms wells unsaturated September coliforms limited to < 7 m from pit alkaline alluvium soils Caldwell 1938a USA 3 Installed test Fine gravel to clayey May– Bacillus aerogenes, B. coli movement limited to 3 m wells soils November anaerobes, odor, pH, from pits 1933 B. coli Caldwell 1938b USA 1 Installed test Fine gravel to clayey November Nitrate, dissolved Limited movement of B. coli to wells soils 1932– oxygen, chloride, 3 m from pit and chemicals to November nitrite, pH, odor, colon 24 m 1933 aerogenes group, B. coli, anaerobes Caldwell and Parr USA 8 bored hole Installed test Partially saturated May 1932– Nitrate, dissolved Movement of bacteria and 1937 latrines wells fine gravel to May 1933 oxygen, chloride, chemicals to within 10 m and clayey soils nitrite, pH, odor, colon 26 m of latrine, respectively aerogenes group, B. coli, anaerobes Caldwell 1937b USA 1 envelope pit Installed test Unsaturated fine May– Colon aerogenes group, Bacteria greatly reduced to within latrine wells gravel to clayey November pH, odor, B. coli, 2 m from pit soils 1933 anaerobes Caldwell 1937a USA 1 Installed test Saturated fine gravel August 1932– Colon aerogenes group, Movement of bacteria to within wells to clayey soils November pH, odor, B. coli, 25 m of latrine 1933 anaerobes Kligler 1921 USA 50 Sampled soil Saturated and Wet and dry B. coli, B. aerogenes Bacterial movement limited to at varying unsaturated sand, seasons, < 5.5 m from pit distances sandy clay, and clay 1918–1919 Abbreviations: Co, cobalt; Cu, copper; Fe, iron; Na, sodium; Zn, zinc. aOnly studies that either directly assessed the transport of contaminants from pit latrines or studies that applied statistical methods to estimate a measure of risk associated with the presence of pit latrines are included. bNo specific data were provided on the density or number of pit latrines in the study area. cCulture-based assays were used for all microbiological tests, except for Verheyen et al. (2009), who used genotyping methods.

Environmental Health Perspectives • v o l u m e 121 | n u m b e r 5 | May 2013 525 Graham and Polizzotto nitrate pollution in a fractured rock aquifer contamination of well water; an area with shal- downstream. In a small study, Padmasiri et al. of Niger. Due to fermentation of feces and low ground­water was more susceptible to pol- (1992) observed decreases in soil nitrate con- ammonia volatilization in latrines, isotopic lution from latrines than an area with a deeper centrations at 1.5 m from the latrine. Overall, enrichment of residual matter creates a nitrate water table. In eastern Botswana, buildup of although data are sparse, direct measure­ments source that is isotopically distinguishable from nitrogenous latrine effluent in soils and subse- and estimates of lateral transport distances nitrate of other sources. Nitrate concentrations quent downward leaching of nitrate appeared for high levels of pit latrine–derived nitrate— in wells reached 11.6 milliequivalents/L, which to promote dissolved nitrate concentrations where it has been detected—range from may have been a consequence of contamina- > 500 mg/L in ground­water (Lewis et al. approximately 1 to 25 m (Caldwell 1938b; tion by latrines and deforestation (Girard and 1980); the authors concluded that the fissured Caldwell and Parr 1937; Chidavaenzi et al. Hillaire-Marcel 1997). The authors cautioned bedrock aquifer allowed for rapid contami- 2000; Lewis et al. 1980; Still and Nash 2002; that, given annual population growth rates nant transport. Whereas soil type immediately Vinger et al. 2012) (Figure 2). and increased latrine densities, wells that had below the pit is likely to influence the degree Chloride. After nitrate, chloride has been safe nitrate concentrations at the time of the of nitrate transport (Caldwell and Parr 1937), the most commonly investigated chemical study might become polluted in the future. associations with soil type have not always been indicator of ground­water contamination from A more common approach in identify- observed (Nichols et al. 1983). In addition, latrines because of its high concentrations in ing nitrate sources has been to compare areas in an area with high nitrogen loading from excreta and its relative mobility in the sub- with similar environmental characteris­ tics­ but latrines but where ground­water was devoid of surface. Although there are no known health different population and latrine densities. By oxygen, nitrate concentrations were minimal, risks from chloride in drinking water, con- analyzing water samples from installed bore- presumably because of denitrification (Ahmed centrations > 250 mg/L may affect the taste holes in an informal settlement in Zimbabwe, et al. 2002). and acceptability of water (WHO 2011). In Zingoni et al. (2005) demonstrated that the Thus, both environmental conditions and a study from Botswana, Lewis et al. (1980) highest nitrate concentrations in ground­water human factors are major drivers of nitrate con- found the highest chloride concentrations in (20–30 mg/L) were associated with the high- tamination from latrines, and the highest con- soils closest to latrines. In Bangladesh, dis- est population and pit latrine densities of the centrations in well water are expected to be solved concentrations reached 400 mg/L settlement. In Siberia and Kosova, nitrate found downstream of areas with high latrine at shallow depths, but then decreased with concentrations were sometimes > 100 mg/L use (Chidavaenzi et al. 2000; Mafa 2003; depth and distance from latrines (Ahmed in groundwater­ of villages with high latrine Vinger et al. 2012). After nitrate is leached et al. 2002). Chloride is typically transported densities and minimal septic tanks, but con- from latrines, a number of factors may control with minimal retention during ground­water centrations were below hazardous levels in travel distance. Certain chemical contaminants flow, and concentrations frequently track with agricultural and unpopulated settings (Banks may be transported farther than microbial nitrate levels (Banks et al. 2002; Caldwell et al. 2002). Groundwater nitrate concentra- contaminants because they are not as inhib- 1938b; Caldwell and Parr 1937; Jacks et al. tions have also been correlated with proximity ited by the bio­layer that commonly forms 1999; Lewis et al. 1980; Tandia et al. 1999) to pollution sources, including pit latrines, in around latrines (Caldwell and Parr 1937). unless subsurface conditions promote nitrate Senegal and South Africa (Tandia et al. 1999; Similarly, peat-lined pits were associated with reduction (Ahmed et al. 2002). Variable distri- Vinger et al. 2012). reduced bacterial and phosphate transport butions of latrine contaminants resulting from Environmental factors also play a role in from latrines but appeared to be ineffective in pumping and seasonal fluctuations have been governing groundwater­ pollution from latrines. limiting nitrate (Nichols et al. 1983). In con- demonstrated by studies using chloride salts as Pujari et al. (2012) compared the impacts of trast, Chidavaenzi et al. (2000) estimated that tracers (Banerjee 2011; Lewis et al. 1980). on‑site sanitation in two Indian megacities­ and the nitrogen influence from latrines extended Ammonia. Ammonia, derived either concluded that hydrogeological­ conditions only 5 m from the latrine source, whereas directly from latrine waste or following were strong predictors of the threat of nitrate microbial contamination extended up to 20 m denitrification­ of nitrate released from latrines, has not been reported to accumulate apprecia- 60 bly in ground­water near latrines. In a study Bacteria of three pit latrines, Dzwairo et al. (2006) Viruses f WaterAid 2011 50 observed only one incidence of ammonium Chemicals + Latrine siting guidelines (NH4 ) > 1.5 mg/L in well water that was 40 microbiologically contaminated by latrines. In ground­water with latrine-derived nitrate con- Sphere Project 2011 30 centrations that exceeded 500 mg/L, Lewis g + a g et al. (1980) found NH4 at < 0.2 mg/L in c + 20 all wells but one, which had NH4 at 3 mg/L. h Lewis et al. 1982 and + Similarly, NH4 was below the South African h a e j Franceys et al. 1992 10 National Standard (2 mg/L) in all water b e i a a samples analyzed by Vinger et al. (2012).

Lateral travel distance from latrine (m) a a d h 0 Padmasiri et al. (1992) reported that soil + concentrations of NH4 decreased substan-

Kigler 1921 Dyer 1941 tially between 1 and 1.5 m from latrine pits. CaldwellCaldwell 1937a 1937bCaldwellCaldwell 1938a 1938b Banerjee 2011 Caldwell 1938b Banerjee 2011 Lewis et al. 1980 Vinger et al. 2012 Ammonia tends to accumulate and persist Dzwairo et al. 2006 erheyen et al. 2009 Still and Nash 2002 V Still and Nash 2002 Caldwell and Parr 1937 Chidavaenzi et al. 2000 Caldwell and Parr 1937Chidavaenzi et al. 2000 under anaerobic conditions, and high concen- trations are likely when the water table inter- Figure 2. Lateral travel distances of different contaminants emanating from pit latrines in relation to select latrine/water-point siting guidelines. Verheyen et al. (2009) and Vinger et al. (2012) used existing wells to sects the base of the latrine pit (Ahmed et al. approximate distances, whereas all other studies used test wells to measure distances. 2002; Baars 1957; Dzwairo et al. 2006). aB. coli; btotal coliforms; ccoliforms; dfecal coliforms; etotal and fecal coliforms; fadenovirus and rotavirus; gchemical Other chemicals derived from pit latrines. stream (nitrate, nitrite, and chloride); hnitrate; initrogen; jsalt tracer. Nitrite concentrations in well water from near

526 v o l u m e 121 | n u m b e r 5 | May 2013 • Environmental Health Perspectives The impact of pit latrines on groundwater quality latrines have typically been below drinking proximity, both environmental­ and anthropo- and should terminate no less than 1.5–2.0 m water standards (Baars 1957; Vinger et al. genic factors must be considered. above the water table. Banerjee (2011) con- 2012), although when present, it has been Among the studies we reviewed, specific cluded that, with the exception of fissured found in association with nitrate and chloride recom­mendations for minimizing latrine rock, the safe distance between a pit latrine (Caldwell 1938b; Caldwell and Parr 1937). effects on ground­water quality varied. Nichols and water source is 10 m. Vinger et al. (2012) Phosphate is fairly immobile, and when it et al. (1983) suggested that pit liners, such as suggested that wells are likely to be contami- was released from latrines, its penetration into peat liners, should not be used as a substitute nated if pit latrines are < 12 m away. soils was minimal (Padmasiri et al. 1992); for proper soil conditions, and recommended Countries and development agencies often peat liners further reduced potential transport that latrines not be built in thin, rocky soils. have siting standards for latrine construction. (Nichols et al. 1983). Accordingly, phosphate Dzwairo et al. (2006) highlighted the need to In Haiti, for example, latrines must be sited concentrations in well water have not been a) analyze critical parameters such as depth of at least 30 m from any surface water source detected at concentrations above water quality the infiltration layer and direction of ground­ or drinking water source, and the bottom of standards in association with pit latrines water flow; b) develop alternative sanitation the pit must be at least 1.5 m above the maxi- (Zingoni et al. 2005). options, such as raised or lined pit latrines, to mum height of the water table (Reed 2010). Elevated ground­water potassium concen- minimize ground­water impacts; and c) apply South Africa’s groundwater­ guidelines recom- trations may also be derived from latrines, and an integrated approach, involving geotech- mend that pit latrines are located at least 75 m concentrations have been shown to correlate nology and hydrogeology, to solve sanitation from water sources (Still and Nash 2002). The with those of nitrate and chloride (Banks et al. problems. Pujari et al. (2012) recom­mended WHO suggests minimal risk of ground­water 2002). The effect of latrines on sulfate con- that latrines be discouraged in rocky areas with pollution where > 2 m of relatively fine soil centrations remains unclear, perhaps because shallow water tables. They also suggested that exists between a pit and the groundwater­ table, of the prevalence of sulfate sources and the systematic lithological­ and hydrogeological­ assuming fill rates are < 50 L/m²/day (Franceys number of processes that may remove sulfate mapping be conducted and that parameters et al. 1992). Furthermore, 15 m is suggested as from solution in the subsurface. Although such as the depth of the water table, soil charac­ the safe lateral separation between pit latrines Banks et al. (2002) found no evidence that teristics, and rock strata be considered prior to and the groundwater­ supply; this distance can latrines influenced sulfate concentrations in installing latrines. Pujari et al. (2012) advised be reduced if the well is not directly down­ well water, Chidavaenzi et al. (2000) observed that ground­water sources in areas served by gradient of the pit (Franceys et al. 1992). increases in sulfate concentrations near on‑site sanitation systems should be monitored However, in a more recent and conservative latrines during the wet season. Latrines also by responsible agencies; monitoring should recommendation that seeks to account for a have been associated with increased well-water include nitrate, chloride, and fecal coliforms. wide variety of contexts, WaterAid (2011) turbidity (Dzwairo et al. 2006). Finally, Mafa To minimize the leaching of nitrate, Jacks et al. suggests that latrines and water sources should (2003) measured high concentrations of dis- (1999) suggested a) painting latrine ventila- be at least 50 m apart (WaterAid 2011). For solved organic carbon in wells down­gradient tion tubes black to increase daytime ventilation disaster response situations, the Sphere Project of latrines, which might contribute to reduc- rates; b) increasing the pH of latrine contents to (2011) has recommended 30 m as a mini- ing conditions and elevated dissolved iron increase ammonia volatilization; c) sealing pits mum standard for the lateral distance between concentrations (Zingoni et al. 2005). to prevent nitrate leaching and promote deni- on‑site sanitation systems and water sources, trification; and d) diverting urine for use as a although this value could be adjusted based on Discussion fertilizer for deep-rooted crops. Finally, a num- the nature of subsurface features. Pit latrine guidelines for mitigating ground­ ber of the studies suggested that pit latrines did Overall, threats to groundwater­ qual- water impacts. In relation to on‑site sanitation, not appear to pose a major threat to ground­ ity from on‑site sanitation can be mitigated the factors controlling transport of microbial water quality or public health (Caldwell 1938a, through technology design, risk assessment, and chemical contaminants in the subsurface­ 1938b; Chidavaenzi et al. 2000; Howard et al. development of protection zones, and moni­ have been the subject of several reviews (BGS 2003; Kligler 1921); this conclusion, which toring (Lawrence et al. 2001; Lewis et al. 2002; Dillon 1997; Gerba et al. 1975; Lewis runs counter to general consensus, may have 1982; Robins et al. 2007). For septic systems et al. 1982; WHO 2006), and there is exten- been influenced by the specific latrine siting, and more complex on‑site sanitation tech- sive literature that more broadly quantifies environmental conditions, and experimental nologies, manuals and siting guidelines are contaminant transport processes in ground­ designs of the studies. widely accessible (e.g., U.S. Environmental water (e.g., Schijven and Hassanizadeh 2000). Given the varying transport distances Protection Agency 2002), and technology Soil/rock type, natural and human-altered observed for microbiological and chemical choices generally depend on the available land groundwater­ flow rates and paths, and the contaminants originating from pit latrines area for drain fields and vertical separation biogeo­ chemical­ environment of the subsurface­ (Figure 2), researchers have identified a range to the water table. Step-by-step strategies for all govern contaminant travel distances and of latrine siting guidelines. In their compre- site-specific analyses of safe sanitation options rates. Tracking the movement of contami­ hensive review about the risks for groundwater­ appropriate for low-income countries have nants is further complicated by microbial contamination by on‑site sanitation sources, been outlined by the BGS (Lawrence et al. die-off and chemical transformations, which Lewis et al. (1982) noted the “traditional” 2001). The BGS guidelines provide a set of may occur hetero­geneously over space and guideline of 15 m as a safe distance between rules for determining the optimum horizontal­ time. The potential for widespread ground­ wells and sanitation units. On the basis of separation between sanitation facilities and water contamination from pit latrines is also statistical associations between latrines and drinking-water sources for a variety of hydro­ affected by social factors, such as latrine use, nitrate concentrations in water sources, Tandia geological environments. These guidelines latrine densities, maintenance, and ground­ et al. (1999) recommended­ distances of 20 m, have been tested in Bangladesh (Ahmed et al. water pumping. Latrine type, design, materi- 36 m, and 48 m for pits that are in use for 2002), Uganda (Howard et al. 2003), and als, and construction quality also influence < 1 decade, 1–2 decades, and > 2 decades, Argentina (Blarasin et al. 2002) and have been contaminant containment­ and leaching from respectively. Banks et al. (2002) suggested advocated as sensible practice for aquifers for pit latrines. Thus, to effectively evaluate the that pit latrines should be located no less than which data are limited and therefore do not safety of pit latrine and ground­water source 15–30 m from groundwater­ abstraction points otherwise lend themselves to conventional

Environmental Health Perspectives • v o l u m e 121 | n u m b e r 5 | May 2013 527 Graham and Polizzotto vulnerability assessment (Ahmed et al. 2002; not been investigated proxi­mate to pit latrines, remains unclear whether these alternative sys- Blarasin et al. 2002; Howard et al. 2003). but they should be quantified and their poten- tems are affordable and culturally acceptable Moving forward. Pit latrine and ground­ tial for transport needs to be assessed. There to poor populations in low-income countries water usage are prevalent in a rapidly growing has also been little research on disposal of other (Mariwah and Drangert 2011). segment of the world population. Given that chemicals, such as lime, pesticides, and clean- Balancing risks. Despite the potential approximately 1.11 billion people currently ing agents, into latrines. Finally, it remains for ground­water contamination, pit latrines have no sanitation facility [see Supplemental unclear whether effects of latrine wastes on remain an important strategy for improving Material (http://dx.doi.org/10.1289/ the geochemical environment­ of groundwater­ human excreta disposal. These systems are the ehp.1206028)], pit latrine coverage is expected may increase downstream contamination. For most basic option for low-income countries to to increase as people attempt to move up the instance, excreta contains high quantities of decrease rates of open defecation and increase sanitation ladder from open defecation to basic organic carbon (Feacham et al. 1983), and access to improved sanitation. An intensive sanitation (WHO/UNICEF 2012c). Our plumes of carbon from latrines may promote effort is needed to develop more robust—yet analysis­ of existing literature­ reveals five key reducing conditions within groundwater­ (Mafa viable—approaches to siting pit latrines and knowledge gaps that could be addressed to 2003), leading to reductive release of trace ele- water sources. Proposed guidelines should improve our understanding and management­ of ments associated with native aquifer materials be tested empirically to ensure protection of groundwater­ contamina­ tion­ from pit latrines. (Harvey et al. 2002). ground­water quality after implementation Siting latrines in relation to wells. Global climate change. Global climate under local conditions. Groundwater flow paths are among the change is widely recognized as a threat to the most important factors controlling contami- safety and reliability of drinking water and Conclusions nant transport from latrines to water points. sanitation supplies, particularly in low-income We estimate that approximately 1.77 billion In many areas, the sub­surface flow pattern countries (WHO 2009; World Bank 2012). people around the world use pit latrines. This is unknown. Groundwater flow models are To date, no studies have specifically addressed number is expected to increase as popula- needed to better define the limits of chemi- these threats in relation to pit latrines and tions grow and countries strive to meet the cal transport and pathogen dispersion (Pedley groundwater­ quality. Many sprawling urban Millennium Development Goals. The use et al. 2006), particularly for complex ground­ slums, as well as poor rural communities, are of groundwater­ as a primary drinking-water water systems such as fractured rock aquifers. currently situated in coastal zones that are source is also increasing. Accordingly, there is It is often difficult to determine whether a flood prone or have high groundwater­ tables, a growing need to understand how pit latrines contamination source is a pit latrine or animal especially in East Asia (Djonoputro et al. may adversely impact ground­water quality waste and agricultural sources; better assess- 2010). Rising sea levels will increase the preva­ and human health. ment of ground­water flow conditions will lence of flooding and slowly raise groundwater­ Despite the widespread global reliance enable identification of dominant contaminant levels, limiting the ability for safe vertical sepa- on both pit latrines and ground­water, we sources. In locations where horizontal separa- ration between latrine pits and the saturated found a limited number of studies that have tion of latrines and water points is not pos- zone. Over shorter time periods, escalation of explicitly examined links between ground­ sible (e.g., routinely flooded regions), vertical storm intensities will increase the probability and contamination from pit separation has been promoted (Lawrence et al. that ground­water tables will rise above the latrines. Within these studies, the quality of 2001), but such siting guidelines are not well bottoms of pits at some point during the year; experimental techniques and chosen indicator defined. An improved understanding­ of con- thus, it is likely that contaminant transport contaminants varied greatly. In multiple studies taminants leaching from pit latrines and the from pit latrines to ground­water will increase. conducted near the same location, there were transport pathways involved is needed particu- Flooding will also likely under­mine efforts to substantial differences in transport distances larly for managing sanitation­ in densely popu- increase access to basic sanitation. Urban plan- of micro­biological and chemical contaminants lated areas, such as refugee­ camps and informal ning and housing develop­ment programs will (Caldwell 1937a, 1937b, 1938a, 1938b; settlements, as well as areas with rapidly grow- need better estimates of the potential effects Caldwell and Parr 1937). Nevertheless, based ing populations. Siting guidelines need to con- of climate change on on‑site sanitation, as well on available reports, researchers who looked for sider population pressures and the potential as additional information to determine appro- ground­water contamination from pit latrines for increased ground­water abstraction, which priate sanitation facility designs for different frequently detected it, and studies observed will alter transport distances and rates. target populations. travel distances of up to 25 m, 50 m, and 26 m Understudied and emerging contami- Improved sanitation technologies. for unsafe concentrations of bacteria, viruses, nants. To date there has been a focus on a Technological upgrades to pit latrines may and chemicals, respectively (Caldwell 1937b; limited number of contaminants that may substantially reduce microbiological and Caldwell and Parr 1937; Verheyen et al. 2009). be found in human excreta. Microbiological chemical threats to ground­water quality. Although these contaminant transport distances monitoring­ has primarily relied on fecal indica- Latrine liners can minimize seepage of pit could potentially be exceeded under certain tor bacteria, whereas nitrate has been the focus contents to ground­water, and raised latrines conditions (e.g., in fractured rock aquifers), of most chemical studies. In a recent study of may help mini­mize groundwater­ contami- most studies of pit latrine–derived contaminants ground­water in rural Bangladesh, Ferguson nation by increasing vertical separation and actually showed transport distances that were et al. (2012) noted that culture-dependent fecal promoting aerobic digestion of waste (Dillon less than half of the maximum values. Areas indicators were not always able to predict total 1997; Dzwairo et al. 2006; Nichols et al. with shallow ground­water and areas prone bacterial pathogens. Pit latrine additives are 1983). Urine-diverting toilets, painted ven- to flooding present the greatest risks, because used to reduce pit contents, odor, and insect tilation tubes, and chemical amendments to vertical separation is required between the base problems, but little research exists on their latrines can minimize nitrate formation and of latrine pits and the saturated zone. makeup or the prevalence of their use (Buckley release to ground­water (Jacks et al. 1999). The ability to make informed decisions et al. 2008). Organic chemical contaminants, Composting toilets and about water and sanitation options is largely including endocrine disruptors and pharma­ technologies may reduce microbial risks and inhibited by a scarcity of data, especially ceuticals, that may be excreted in urine and minimize chemical leaching from pit latrines regarding the influence of environmental con- feces and may persist in the environment have (Dillon 1997; Endale et al. 2012). However, it ditions on potential contamination. Guidelines

528 v o l u m e 121 | n u m b e r 5 | May 2013 • Environmental Health Perspectives The impact of pit latrines on groundwater quality

are available for site-specific assessment, and Cairncross S, Bartram J, Cumming O, Brocklehurst C. 2010a. and Sanitation Program. Available: http://www.wsp.org/ general procedures for siting latrines with Hygiene, sanitation, and water: what needs to be done? sites/wsp.org/files/publications/wsp-rwanda-sanitation- Plos Med 7(11):e1000365; doi:10.1371/journal.pmed.1000365 lessons.pdf [accessed 27 March 2013]. respect to water points are common (Lawrence [Online 16 November 2010]. Kligler IJ. 1921. Investigation on Soil Pollution and the Relation et al. 2001). However, recommenda­ tions­ for Cairncross S, Hunt C, Boisson S, Bostoen K, Curtis V, Fung ICH, on the Various Types of Privies to the Spread of Intestinal mitigating groundwater­ impacts can be both et al. 2010b. Water, sanitation and hygiene for the preven- Infections. New York:Rockefeller Institute for Medical tion of diarrhoea. Int J Epidemiol 39(suppl 1):I193–I205. Research. qualitatively and quantitatively variable, and it Caldwell EL. 1937a. Pollution flow from pit latrines when an Lawrence AR, Macdonald DMJ, Howard AG, Barrett MH, remains unclear whether these guidelines can impervious stratum closely underlies the flow. J Infect Dis Pedley S, Ahmed KM, et al. 2001. Guidelines for Assessing be implemented under local conditions. Many 61(3):270–288. the Risk to Groundwater from On-Site Sanitation. Keyworth, Caldwell EL. 1937b. Study of an envelope pit privy. J Infect Dis Nottingham:British Geological Survey. Available: http:// countries already face serious developmental 61(3):264–269. www-esd.worldbank.org/esd/ard/groundwater/pdfreports/ challenges and may not have the resources— Caldwell EL. 1938a. Pollution flow from a pit latrine when perme- Guidelines_assessing_risk_to_Gdwtr.PDF [accessed human and economic—to implement guide- able soils of considerable depth exist below the pit. J Infect 28 March 2013]. Dis 62(3):225–258. Lewis WJ, Farr JL, Foster SSD. 1980. The pollution hazard to lines (Robins et al. 2007). Caldwell EL. 1938b. Studies of subsoil pollution in relation to village water supplies in eastern Botswana. ICE Proc In general, siting guidelines vary greatly, possible contamination of the ground water from human 69(2):281–293. and it is often unknown which (if any) empiri­ excreta deposited in experimental latrines. J Infect Dis Lewis WJ, Foster SSD, Drasar BS. 1982. The Risk of Ground­water cal studies were used to derive the guidelines. 62(3):272–292. Pollution by On-Site Sanitation in Developing Countries. Caldwell EL, Parr LW. 1937. Ground water pollution and the Duebendorf, Switzerland:International Reference Centre for Thus, there is a need to empirically test the bored hole latrine. J Infect Dis 61(2):148–183. Waste Disposal. effectiveness of specific guidelines under a vari- Chidavaenzi M, Bradley M, Jere M, Nhandara C. 2000. Pit latrine Ley RE, Peterson DA, Gordon JI. 2006. Ecological and evolu- ety of conditions in order to better merge pit effluent infiltration into groundwater: the Epworth case tionary forces shaping microbial diversity in the human study. Schriftenr Ver Wasser Boden Lufthyg 105:171–177. intestine. Cell 124(4):837–848. latrine siting guidelines with realistic ground­ Chidavaenzi M, Jere M, Bradley M. 1997. Pit latrine effluent Mafa B. 2003. Environmental Hydrogeology of Francistown: water threats. Given the scale of pit latrine infiltration into groundwater. In: Proceedings of the 23rd Effects of Mining and Urban Expansion on Groundwater use, future studies must examine additional WEDC Conference, 1–5 September 1997, Durban, South Quality. Lobatse, Botswana:Botswana Department of Africa. Available: http://wedc.lboro.ac.uk/resources/ Geological Survey and Federal Institute for Geosciences contaminants beyond standard indicators, conference/23/Chidavae.pdf [accessed 29 March 2013]. and Natural Resources. Available: http://www.limpoporak. monitor temporal changes in water quality Dillon P. 1997. by Sanitation on Tropical org/_system/DMSStorage/3471en/EnvHydrogeology%20 parameters, and evaluate alternative technolo- Islands. IHP-V Project 6–1. Paris:UNESCO (United Nations FTown_sec.pdf [accessed 28 March 2013]. Educational, Scientific and Cultural Organization). Available: Mariwah S, Drangert JO. 2011. Community perceptions of gies. In addition, efforts are needed to better http://unesdoc.unesco.org/images/0010/001065/106522eo. human excreta as fertilizer in peri-urban agriculture in understand the effects of population pressures pdf [accessed 29 March 2013]. Ghana. Waste Manage Res 29(8):815–822. and climate change in order to make more Djonoputro ER, Blackett I, Rosenboom JW, Weitz A. 2010. Melian R, Myrlian N, Gouriev A, Moraru C, Radstake F. 1999. Understanding sanitation options in challenging environ- Groundwater quality and rural drinking-water supplies in informed decisions that optimize latrine and ments. Waterlines 29(3):186–203. the Republic of Moldova. Hydrogeol J 7:188–196. ground­water use and improve environmental Dyer BR. 1941. Studies of ground water pollution in an alkaline Nichols DS, Prettyman D, Gross M. 1983. Movement of bacteria and human health. alluvium soil. Ind J Med Res 29(4):867–877. and nutrients from pit latrines in the Boundary Waters Dzwairo B, Hoko Z, Love D, Guzha E. 2006. Assessment of the Canoe Area Wilderness. Water Air Soil Poll 20(2):171–180. impacts of pit latrines on groundwater quality in rural Padmasiri JP, Jayatilake GM, Kotuwegedara JPK. 1992. Re f e r e n c e s areas: a case study from Marondera district, Zimbabwe. Groundwater pollution from double pit latrines. In: Water, Phys Chem Earth 31(15–16):779–788. Environment and Management: Proceedings of the Ahmed KM, Khandkar ZZ, Lawrence AR, Macdonald DMJ, Endale YT, Yirsaw BD, Asfaw SL. 2012. Pathogen reduction 18th WEDC Conference, 30 August–3 September 1992, Islam MS. 2002. Appendix A: an investigation of the efficiency of on-site treatment processes in eco-sanitation Kathmandu, Nepal. 239–241. Loughborough, UK:Water, impact of on-site sanitation on the quality of groundwater system. Waste Manage Res 30(7):750–754. Engineering and Development Centre, Loughborough supplies in two peri-urban areas of Dhaka, Bangladesh. In: Feachem RG, Bradley DJ, Garelick H, Mara DD. 1983. Sanitation University of Technology. Assessing Risk to Groundwater from On-site Sanitation: and Disease: Health Aspects of Excreta and Wastewater Pedley S, Yates M, Schijven JF, West J, Howard G, Barrett M. Scientific Review and Case Studies. Keyworth, UK:British Management. New York:John Wiley & Sons Ltd. 2006. Pathogens: health relevance, transport and attenuation. Geological Survey, 37–67. Available: http://r4d.dfid.gov.uk/ Ferguson AS, Layton AC, Mailloux BJ, Culligan PJ, Williams DE, In: Protecting Groundwater for Health: Managing the Quality pdf/outputs/r68692.pdf [accessed 26 March 2013]. Smartt AE, et al. 2012. Comparison of fecal indicators with of Drinking-Water Sources (Schmoll O, Howard G, Chilton J, Albonico M, Allen H, Chitsulo L, Engels D, Gabrielli AF, Savioli L. pathogenic bacteria and rotavirus in groundwater. Sci Total Chorus I, eds). Geneva:World Health Organization, 49–80. 2008. Controlling soil-transmitted in pre-school- Environ 431:314–322. Available: http://www.who.int/water_sanitation_health/ age children through preventive chemotherapy. PLoS Negl Fewtrell L. 2004. Drinking-water nitrate, methemoglobinemia, publications/PGWsection1.pdf [accessed 29 March 2013]. Trop Dis 2(3):e126; doi:10.1371/journal.pntd.0000126 [Online and global burden of disease: a discussion. Environ Health Pujari PR, Padmakar C, Labhasetwar PK, Mahore P, Ganguly AK. 26 March 2008]. Perspect 112:1371–1374. 2012. Assessment of the impact of on-site sanitation sys- Baars JK. 1957. Travel of pollution, and purification en route, in Fourie AB, Vanryneveld MB. 1995. The fate in the subsurface of tems on groundwater pollution in two diverse geological sandy soils. Bull World Health Organ 16:727–747. contaminants associated with on-site sanitation–a review. settings—a case study from India. Environ Monit Assess Banerjee G. 2011. Underground pollution travel from leach pits Water SA 21(2):101–111. 184(1):251–263. of on-site sanitation facilities: a case study. Clean Technol Franceys R, Pickford J, Reed R. 1992. A Guide to the Development Ramakrishna BS. 2007. The normal bacterial flora of the human Environ Policy 13(3):489–497. of on-Site Sanitation. Geneva:World Health Organization. intestine and its regulation. J Clin Gastroenterol 41(5):S2–S6. Banks D, Karnachuk OV, Parnachev VP, Holden W, Frengstad B. Available: http://www.who.int/water_sanitation_health/ Reed B. 2010. Emergency Excreta Disposal Standards and Options 2002. Groundwater contamination from rural pit latrines: hygiene/envsan/onsitesan.pdf [accessed 29 March 2013]. for Haiti. Leicestershire, UK:DINEPA (Direction Nationale de examples from Siberia and Kosova. J Chartered Inst Water Gerba CP, Wallis C, Melnick JL. 1975. Fate of wastewater bac- l’Eau Potable et de l’Assainissement) and Global WASH Environ Manage 16(2):147–152. teria and viruses in soil. J Irrig Drain Div 101(3):157–175. Cluster. Available: http://wedc.lboro.ac.uk/resources/pubs/ BGS (British Geological Survey). 2002. Assessing Risk to Girard P, Hillaire-Marcel C. 1997. Determining the source of Emergency_EDS_and_options_for_Haiti.pdf [accessed Groundwater from On-site Sanitation: Scientific Review and nitrate pollution in the Niger discontinuous aquifers using 28 March 2013]. Case Studies. Keyworth, UK:BGS. Available:http://r4d.dfid. the natural 15N/14N ratios. J Hydrol 199:239–251. Robins NS, Chilton PJ, Cobbing JE. 2007. Adapting existing gov.uk/pdf/outputs/r68692.pdf [accessed 28 March 2013]. Harvey CF, Swartz CH, Badruzzaman AB, Keon-Blute N, Yu W, experience with aquifer vulnerability and groundwater Blarasin M, Cabrera A, Frigerio C, Bettera S. 2002. Appendix C: Risk Ali MA, et al. 2002. Arsenic mobility and groundwater protection for Africa. J Afr Earth Sci 47(1):30–38. of ground­water pollution by on-site sanitation, geochemical extraction in Bangladesh. Science 298(5598):1602–1606. Rosa G, Clasen T. 2010. Estimating the scope of household models and geoindicators, Rio Cuarto City, Argentina. In: Howard G, Pedley S, Barrett M, Nalubega M, Johal K. 2003. water treatment in low- and medium-income countries. Assessing Risk to Groundwater from On-site Sanitation: Risk factors contributing to microbiological contamination Am J Trop Med Hyg 82(2):289–300. Scientific Review and Case Studies. Keyworth, UK:British of shallow groundwater in Kampala, Uganda. Water Res Schijven JF, Hassanizadeh SM. 2000. Removal of viruses by soil Geological Survey, 94–105. Available: http://r4d.dfid.gov.uk/ 37(14):3421–3429. passage: overview of modeling, processes and param- pdf/outputs/r68692.pdf [accessed 28 March 2013]. Howard G, Teuton J, Luyima P, Odongo R. 2002. Water usage eters. Crit Rev Environ Sci Technol 30(1):49–127. Buckley CA, Foxon KM, Brouckaert CJ, Rodda N, Nwaneri C, patterns in low-income urban communities in Uganda: Sphere Project. 2011. Minimum standards in water supply, Balboni E, et al. 2008. Scientific Support for the Design and implications for water supply surveillance. Int J Environ sanitation and hygiene promotion. In: Humanitarian Operation of Ventilated Improved Pit Latrines (VIPS) and the Health Res 12(1):63–73. Charter and Minimum Standards in Disaster Response. Efficacy of Pit Latrine Additives. WRC Report No. TT 357/08. Jacks G, Sefe F, Carling M, Hammar M, Letsamao P. 1999. Geneva:The Sphere Project. Available: http://www.ifrc. Gezina, South Africa:Water Research Commission. Available: Tentative nitrogen budget for pit latrines–eastern Botswana. org/PageFiles/95530/The-Sphere-Project-Handbook-20111. http://www.wrc.org.za/Knowledge%20Hub%20Documents/ Environ Geol 38(3):199–203. pdf [accessed 28 March 2013]. Research%20Reports/TT357-08%20-Developing%20 Jain N. 2011. Getting Africa to Meet the Sanitation MDG: Still DA, Nash SR. 2002. Groundwater contamination due to communities.pdf [accessed 29 March 2013]. Lessons from . Washington, DC:World Bank Water pit latrines located in a sandy aquifer: a case study from

Environmental Health Perspectives • v o l u m e 121 | n u m b e r 5 | May 2013 529 Graham and Polizzotto

Maputaland. In: Water Institute of Southern Africa Biennial van Ryneveld MB, Fourie AB. 1997. A strategy for evaluating http://whqlibdoc.who.int/publications/2011/9789241548151_ Conference. Durban, South Africa:Water Institute of the environmental impact of on-site sanitation systems. eng.pdf [accessed 1 April 2013]. Southern Africa, 1–6. Available: http://www.sswm.info/sites/ Water SA 23(4):279–291. WHO (World Health Organization)/UNICEF. 2010. Progress on default/files/reference_attachments/STILL%20and%20 Verheyen J, Timmen-Wego M, Laudien R, Boussaad I, Sanitation and Drinking-water–2010 Update. Available: NASH%202002%20Groundwater%20contamination%20 Sen S, Koc A, et al. 2009. Detection of adenoviruses http://whqlibdoc.who.int/publications/2010/9789241563956_ due%20to%20pit%20latrines%20located%20in%20a%20 and rotaviruses in drinking water sources used in rural eng_full_text.pdf [accessed 5 December 2012]. sandy%20aquifer.pdf [accessed 29 March 2013]. areas of Benin, West Africa. Appl Environ Microbiol WHO (World Health Organization)/UNICEF. 2012a. Estimates for Tandia AA, Diop ES, Gaye CB. 1999. Nitrate groundwater pol- 75(9):2798–2801. the Use of Improved Sanitation Facilities: China. Available: lution in suburban areas: example of groundwater from Vinger B, Hlophe M, Selvaratnam M. 2012. Relationship http://www.wssinfo.org/fileadmin/user_upload/resources/ Yeumbeul, Senegal. J Afr Earth Sci 29(4):809–822. between nitrogenous pollution of borehole waters and CHN_san.pdf [accessed 1 April 2013]. UN (United Nations). 2008. The Millennium Development distances separating them from pit latrines and fertilized WHO (World Health Organization)/UNICEF. 2012b. Estimates Goals Report 2008. New York:UN. Available: http://www. fields. Life Sci J 9(1):402–407. for the Use of Improved Drinking-Water Sources: China. un.org/millenniumgoals/pdf/The%20Millennium%20 Wade TJ, Pai N, Eisenberg JNS, Colford JM Jr. 2003. Do U.S. Available: http://www.wssinfo.org/fileadmin/user_upload/ Development%20Goals%20Report%202008.pdf [accessed Environmental Protection Agency water quality guidelines resources/CHN_wat.pdf [accessed 1 April 2013]. 27 March 2013]. for recreational waters prevent gastrointestinal illness? WHO (World Health Organization)/UNICEF. 2012c. Progress on UN (United Nations). 2011. World Population Prospects: The A systematic review and meta-analysis. Environ Health Drinking Water and Sanitation: 2012 Update. Available: http:// 2010 Revision, CD-ROM Edition. Data download available: Perspect 111:1102–1109. www.unicef.org/media/files/JMPreport2012.pdf [accessed http://esa.un.org/unpd/wpp/Excel-Data/population.htm WaterAid. 2011. Technology Notes. Available: http://www. 5 December 2012]. [accessed 1 April 2013]. wateraid.org/documents/plugin_documents/technology_ World Bank. 2012. Turn Down the Heat: Why a 4oC Warmer UNICEF. 2012. Multiple Indicator Cluster Surveys. Available notes_2011.pdf [accessed 5 December 2012]. World Must Be Avoided. Washington, DC:The World Reports/Datasets. Available: http://www.childinfo.org/ WHO (World Health Organization). 2006. Protecting Groundwater Bank. Available: http://climatechange.worldbank.org/sites/ mics_available.html [accessed 25 October 2012]. for Health: Managing the Quality of Drinking-water Sources. default/files/Turn_Down_the_heat_Why_a_4_degree_ USAID. 2012. Measure DHS: Available Datasets. Available: Geneva:WHO. Available: http://www.who.int/water_ centrigrade_warmer_world_must_be_avoided.pdf http://www.measuredhs.com/data/available-datasets.cfm sanitation_health/publications/protecting_groundwater/en/ [accessed 27 March 2013]. [accessed 25 October 2012]. [accessed 26 March 2013]. World Bank. 2013. How We Classify Countries. Available: U.S. Environmental Protection Agency. 2002. Onsite WHO (World Health Organization). 2009. Protecting Health http://data.worldbank.org/about/country-classifications Wastewater Treatment Systems Manual. EPA/625/R- from Climate Change: Connecting Science, Policy and [accessed 3 April 2013]. 00/008. Washington DC:U.S. Environmental Protection People. Geneva:WHO. Available: http://whqlibdoc.who. Zingoni E, Love D, Magadza C, Moyce W, Musiwa K. 2005. Agency. Available: http://water.epa.gov/type/groundwater/ int/publications/2009/9789241598880_eng.pdf [accessed Effects of a semi-formal urban settlement on groundwater uic/class5/upload/2007_12_12_uic_class5_techguide_ 29 March 2013]. quality Epworth (Zimbabwe): case study and groundwater uic-class5_2002_onsite_wwt_sys_man.pdf [accessed WHO (World Health Organization). 2011. Guidelines for quality zoning. Phys Chem Earth 30(11–16):680–688. 26 March 2013]. Drinking-water Quality. 4th ed. Geneva:WHO. Available:

530 v o l u m e 121 | n u m b e r 5 | May 2013 • Environmental Health Perspectives