Evaluation of a pilot constructed wetland system as an appropriate technology for septage treatment in Cambodia

Chiew, Hannah

Masters of Applied Science- MASc

Civil Engineering

University of British Columbia

Septic tanks are one of the most commonly used forms of onsite sanitation systems in developing countries. However, lack of regulation on the design and installation and the lack of proper disposal for the septage generated may cause serious environmental degradation and pose serious public health hazards. This study presents the findings of a pilot-scale constructed wetland system in Cambodia for the treatment of septage for 8 months. After a 6-month Acclimatization Period, the system was operated at two hydraulic loading rates (HLRs) of 15.5 and 27.4 mm/day to evaluate the impact of HLR on the treatment efficacy. During the 8-month period, the system achieved mass removal efficiency of 96% - 98% for total suspended solids (TSS), 80% - 97% for turbidity, 61% - 78% for biological oxygen demand (BOD) and 61% - 90% for chemical oxygen demand (COD), but was less efficient in removing total phosphorus (TP) and total nitrogen (TN), which exhibited efficiencies of 52% - 58% and 57% - 66%, respectively. The system also achieved 1.56 – 2.25 log reduction for E. coli and 0.49 – 0.81 log reduction for total coliform (TC). Although the HLR was almost doubled during the 2-month period, the effluent quality did not degrade appreciably in most cases, and caused little or no impact on the treatment efficacy. It is possible that harvesting prior to the start of the HLR exercise had an adverse impact on the eco system and de-stabilized the system, causing the results with lower HLR (or higher HRT) to be worse than expected. Another possible reason for the lack of difference could be due to the high variability of influent quality. It is possible that there is no apparent benefit in decreasing the HLR in this case. The k-C* , or first order removal model, was used to model the longitudinal concentration profile and estimated the removal rate constant. There were no apparent relationships between HLR and the removal rate constants in the k-C* model. The effluent may be suitable for localized and restricted irrigation with careful crop selection, or re-used with dilution due to its high nitrogen content.

Surfactant modified zeolites to reduce microbial pathogen exposure and Understanding Pathogen Exposure via Fruits and Vegetables

Palmy Jesudhasan¹,
Ken Widmer²,
Andrew Shantz³
Suresh D. Pillai¹

¹Texas A&M University, College Station, Texas, USA
²Gwangju Institute of Science and Technology, Gwangju, S. Korea
³Resource Development International, Cambodia

Exposure to microbial pathogens is a serious problem with global significance. The problem of microbial pathogens is an especially serious in under-developed and developing countries such as in Cambodia, Vietnam, India, Thailand and many other countries in this region. Groundwater is susceptible to fecal contamination from a variety of sources including sewage dumps, leaking septic tanks, infiltration of surface water contaminated by human and animal wastes, and careless disposal of septic wastes. In addition to microbial contamination of drinking water it is essential that we identify all routes of exposure to microbial contaminants. This includes foods both raw and cooked.

The focus of the proposed project is utilize “surfactant-modified zeolites” (SMZ), a patented technology jointly developed at Texas A&M University to determine whether it could serve as an inexpensive approach to limit human exposure of microbial pathogens in drinking water in Cambodia. The outcome of this project would be deeper understanding of the utility of the SMZ technology as a low-cost drinking water filtration technology in Cambodia. The availability of low-cost filter media can make a significant impact on reducing the exposure to waterborne pathogens in the Mekong River Basin. The underlying hypothesis of the project was that SMZ-filters will function effectively to removal microorganisms in ground waters found in Cambodia. The activities focused on laboratory studies to evaluate the performance of SMZ as a filter media to remove selected bacterial contaminants such as fecal indicators such as E.coli and male-specific coliphages. The phage MS2 was used as the candidate male-specific coliphage. A 90% reduction in E.coliand MS2 phage was observed in the initial laboratory trials. Further studies are, however, still needed to optimize the removal/reduction as a function of SMZ filter media volume.

A secondary objective of the project was to screen a selected number of produce items that are normally consumed raw or with minimal processing to identify the occurrence of fecal contamination indicators. Twenty four different commodities were purchased from the local market and analyzed for the presence of coliforms, E.coli, somatic coliphages and male-specific coliphages. Standard methods were used for these analyses. Out of 24 samples, 100% of the samples were positive for coliforms. This result was not surprising since these samples are handled with bare hands and the presence of coliforms on fresh fruits and vegetables in a market is not surprising. However, 14 out of 24 samples were positive for E.coli which is indicative that they were probably exposed to fecal contamination. However, the presence of E.coli in tropical environments is not necessarily indicative of fecal contamination solely because E.coli can multiply in moist environments. Three out of the 24 samples (12.5%) were positive for male-specific coliphages, which is a strong indication of the presence of fecal contamination in these samples. The samples that tested positive for male-specific coliphages were salad, chirneangvorng, andchirrona. Six out of the 24 samples (25%) were positive for somatic coliphages. Somatic coliphages can be considered an indicator of fecal contamination. However, they are not as strongly correlated with fecal contamination as compared to male-specific coliphages. What was indeed surprising was the large concentration of male-specific coliphages in the samples that tested positive. In the three samples that tested positive (salad, chirneangvorng, andchirrona), the concentration of male-specific coliphages were 459 PFU (plaque forming units)/gram, 600 PFU/gm, and 293 PFU/gm. These results indicate that exposure to microbial pathogens is a very strong possibility through such commodities.

Overall, these results suggest that similar to other countries, a multi-pronged approach to reduce exposure to microbial pathogens is needed in Cambodia. Ground water and fresh fruits and vegetables need to be adequately treated or disinfected prior to human use. Reducing exposure to microbial pathogens will have significant improvement on the disease burden in the community.

Quantifying the Risk: Arsenic Contamination with Particular Focus on Rice

Marc O'Connor
School of the Built Environment
Northumbria University


Arsenic contamination within groundwater has been reported throughout Cambodia, with elevated levels in the Kandal province, however no peer‐reviewed study has focused on arsenic contamination from rice. This study is unique in that it contains the levels of arsenic in rice and directly relates this to total exposure and health effects. Rice samples were obtained from three markets (n=50) sourcing from all over Cambodia and were analysed for total arsenic using an Inductively Coupled Plasma Mass Spectrometry (ICPMS), a range of 4µg/kg to 1,166 µg/kg with an observed mean of 137µg/kg contributing 56% of the WHO Maximum Tolerable Daily Intake (MTDI) of 2µg of arsenic per kg bodyweight. Black sticky rice was observed to contain the highest concentration of arsenic with a mean value of 664µg/kg which alone contributed 277% of the MTDI. Secondary data of arsenic concentrations from groundwater used as drinking water in numerous provinces were added to the to the equation, were the daily intake of arsenic was above the Cambodian Maximum Contaminant Level (MCL) in each province. The cumulative exposures to arsenic contaminated areas such as, Kampong Chhnang, Kampong Thom and Takeo, which are not vulnerable to elevated arsenic levels in groundwater, are exposed to an increase in arsenic from riceof 311%, 314% and 381% of the total arsenic intake. The hazard quotient (HQ) was used to calculate the risk of populations exposed to arsenic from both groundwater and rice, taking exposure time into consideration for each province. When HQ > 1 then adverse health effects are considered occurring, with just groundwater the Kandal province was the only province above the threshold, however inclusive of both groundwater and rice Kratie and Prey Veng also passed the threshold; Signifying the importance of arsenic contamination from rice.

Sustainable Application of Bayoxide E33 Iron-Based Arsenic Adsorbent for Cambodia Drinking Water

Christopher O. Cope
David. A. Sabatini
University of Oklahoma WaTER Center

Millions of people in the world are exposed to dissolved concentrations of arsenic (As) exceeding 50mg/L. The World Health Organization’s guideline is 10 mg/L. Arsenic is naturally dissolved in many groundwaters of the world, including parts of Bangladesh, India, and Cambodia, which have large populations of exposed persons. Extended consumption of dissolved arsenic can result in arsenicosis. Symptoms of this chronic disease include skin lesions, black foot disease, diabetes, and nervous, hepatic, haematological, and renal damage (Hughes, 2002). The symptoms of arsenic’s carcinogenicity include tumors of the skin, lungs, urinary bladder, liver, prostate, and kidneys, as well as other organs (Obinaju, 2009).

The US Environmental Protection Agency’s (USEPA) Arsenic Treatment Technologies Demonstration Program, targeted at small and rural communities, has found Bayoxide E33 (E33) adsorptive media to be among the most efficient and cost effective technologies at removing naturally occurring arsenic from municipal groundwater treatment and distribution systems. The defining characteristic of the E33 adsorptive media is its relatively high surface area. At a specific surface area of approximately 130 m²g^-1, E33’s high internal surface area, combined with its ferric hydroxide matrix, provides numerous adsorptive sites for arsenic compounds to come out of solution and attach onto the surface of the media.

As one of the most successful and overall cost-effective treatment technologies in the USEPA’s study, questions arise as to whether or not this media would be cost-effective in community, wellhead, and a point-of-use (POU) treatment systems in the developing world. This study aims to characterize the adsorptive behavior and economic sustainability of E33 applications in developing world countries such as Cambodia and Bangladesh.

Rapid small-scale column studies (RSSCts) indicate similar breakthrough curves and media adsorptive capacities as full-scale pilot tests for a fraction of the time and money. This study will characterize the arsenic adsorptive behaviors of E33 using RSSCTs, which will synthesize full-scale pilot treatment systems, in the lab and with two Cambodian wells of differing water qualities. These same well tests will also indicate the impact of varying water quality characteristics (e.g. pH, DO, alkalinity, and competing ions) in Cambodian groundwaters. To confirm the effectiveness of the media as predicted by the RSSCTs, full-scale wellhead and POU pilot systems will be developed and tested to treat the same two Cambodian wells.

Once the arsenic adsorptive efficiency of the media has been determined, economic analysis will predict the affordability and economic sustainability of E33 for centralized, wellhead, and POU treatment systems. This will help determine the advantages and disadvantages of these approaches and enable different NGO, government, and private sector entities to assess the various implementation methods.

Continuous Monitoring Approaches for Conventional Water Quality Parameters in the Mekong River-Tonle Sap System

Kim Irvine, Buffalo State, State University of New York

Jeff Richey, Gordon Holtgrieve, University of Washington

Juha Sarkkula, Finnish Environment Institute

In assessing the potential impacts of development, such as hydropower construction, urban expansion, deforestation, agricultural runoff, and climate change on the Mekong-Tonle Sap system, it is essential to collect basic water quantity and quality data. These data can be used in support of the application of decision-making tools, such as mathematical models, to help inform management of the basin. Good policy must be founded on good data. The Mekong River Commission (MRC) oversees the collection and maintains the most readily available and extensive water quantity and quality dataset for the basin. The MRC water quality database represents monthly sample results for 99 sites across the basin, which can provide reasonable benchmarking, but for modeling purposes the MRC data must be supplemented with data having better temporal and spatial resolution. We monitored turbidity, dissolved oxygen, conductivity, temperature, pH, and fluorescence at seven different locations on the Tonle Sap-Mekong-Bassac system between 2004 and 2010 at 15 to 30 minute intervals. Monitoring was done using either Hydrolab Datasonde 4’s or YSI Datasonde 6600’s and 6920’s. All site measurements represent a fixed depth of between 0.5 and 2.0 m, although we also have profile data at three of the sites.

Our most recent detailed sampling effort (17 August – 22 August, 2010) was conducted in the Tonle Sap Lake near Siem Reap. We sampled at three sites on a transect running from the edge of the flooded forest fringe to the open lake, 2-3.5 km out from the main shoreline. Samples were collected at each site throughout a 24 hour period in order to address system metabolism questions. Two YSI datasonde 6920’s were used to measure water temperature, conductivity, dissolved oxygen, pH, turbidity, and fluorescence. Because the YSI 6920’s can only house one optical sensor at a time, one of the 6920’s was fitted with a turbidity sensor and one was fitted with a fluorescence sensor. The YSI 6920 with the fluorescence sensor was programmed to record all parameters at 15 minute time intervals and was attached to the side of the boat so that measurements were done at a depth of approximately 0.5 m. The YSI 6920 with the turbidity sensor was used for water column profiling. Profiling was done at 0.25 m increments from 0.1 m below the surface to the lake bed. Water samples (0.5 m depth) were collected on an hourly basis at each site for oxygen isotope analysis (e.g. to determine ^16:18O₂ ratios) and alkalinity; four times per day for dissolved inorganic carbon analysis; and once per day for nutrients, chlorophyll a, and zooplankton analysis. Measurements of photosynthetically active radiation (PAR), wind velocity, temperature, and relatively humidity also were made each hour.

This most recent work in the Tonle Sap was part of a larger project led by the University of Washington that is seeking to examine regional-scale landscape dynamics in river basins in Southeast Asia relative to their connectivity to the South China Sea, with an emphasis on the Mekong River. The basic premise of the larger project is that the understanding of regional scale processes requires the higher resolution now possible with satellite data, process-based models, and field measurements. By focusing on how transient forcing of the atmosphere combines with land-use change at multiple space and time scales to mobilize water and carbon to the sea, we are examining the critical and poorly understood interfaces between the atmosphere, land surface and sea function. We are in the process of summarizing the data and developing a series of articles based on the results. An earlier summary of some of the data can be found in Irvine et al. (2007): http://www.buffalostate.edu/geography/documents/publication6.pdf. This field effort has benefited from the support of numerous RDI staff, Buffalo State, University of Washington, and Royal University of Phnom Penh graduate and undergraduate students since 2004.

Vulnerability of Arsenic-Free Wells to Future Contamination

Matthew L. Polizzotto, North Carolina State University

Rebecca B. Neumann, University of Washington

In South and Southeast Asia, contamination of groundwater with naturally-occurring arsenic affects the health of millions. Current efforts to mitigate arsenic exposure in the region include installing new wells that access arsenic-free groundwater. However, the vulnerability of clean groundwater to future arsenic contamination is unknown. Our goal is to assess the potential for arsenic contamination in groundwater that is currently arsenic-free. To accomplish this goal, we explicitly and systematically link field, laboratory, spectroscopic, and modeling approaches that comparatively quantify a multitude of biogeochemical and physical processes controlling arsenic in the environment.

In collaboration with RDIC, we have established a study site in the Kandal Province of Cambodia, between the Mekong and Bassac Rivers. Although the province is known for high (>100 ug/L) dissolved arsenic concentrations in domestic tubewells, at our site, dissolved arsenic concentrations are below the World Health Organization recommended limit (10 ug/L). However, the area is undergoing a period of relatively rapid development, resulting in land-surface changes and growth in groundwater withdrawals. Thus, the site, with its low arsenic concentrations and changing environmental conditions, is ideal for studying the vulnerability of wells to future arsenic contamination.

In our work, we will identify the processes that may stimulate future arsenic contamination and the time scales on which these processes may occur. Our findings will advance basic understanding of groundwater arsenic contamination, and will provide concrete, useful information about the sustainability of currently ‘safe’ wells – information that policy makers, development organizations, and individuals can use both locally to aid in decisions about specific water use options, and broadly to inform larger policy initiatives.