This report summarises research that was conducted during the period 1 January to 31 December 2000. Major achievements include the setting up of the environment laboratory at Dhaka University, establishment of field, soil, vegetable and ground water sampling procedures, appointment of new scientists in both Bangladesh and Australia, and establishment of the proposed field survey on dietary intake of As. Along with these activities, two research staff and a doctor from Bangladesh and two technical officers from Ballarat University, Victoria were trained at the Adelaide laboratories. Although most of the proposed subprojects have been established, much work needs to be done during 2001 to consolidate these activities.

Surface and subsurface soils were sampled from a number of contaminated sites from the least and most severely As affected districts in Bangladesh, railway contaminated sites in South Australia and old minesite contaminated sites currently used for cropping in Ballarat, Victoria. Data from the limited sampling in Bangladesh reveals considerable variability in total As content of soils, tubewell water and vegetable crops and cooked food. In general, total As content of surface and subsurface soils was less than 50 mg kg-1. However, appreciable concentrations of As were released in soil water extract indicating high bioavailability and potential for As uptake by crops. This was also reflected in the crop As content which ranged from <0.1 mg kg-1 to >2.2 mg kg-1. Analyses of food samples from a number of households suggests significant ingestion of As by the Bangladeshis living in the villages that were sampled for food, vegetables and water. It is, however, not clear whether the ingestion is primarily from the use of tubewell water for cooking or from the arsenic content of crops and animal products. Amongst the food samples, egg was the only food product that did not contain As. In contrast to soils from Bangladesh, the total As content of rail track contaminated soils from South Australia ranged from 50 < to > 500 mg kg-1. A similar range in As content was also recorded for Ballarat samples.

This report summarizes research that was conducted during the period 1 January to 31 December 2001. Major achievements include (a) extensive sampling of soils to assess the extent and severity of arsenic contamination, (b) assessment of arsenic content of crops grown at both historically arsenic contaminated sites and sites that have been subjected to irrigation with arsenic contaminated groundwater and (c) a major workshop on "Managing Arsenic in the Australasia Pacific Region". Along with these activities, the hydride generation equipment at Ballarat University was set up and calibrated for the analyses of arsenic in soil and plant digest extracts and water samples. However, work was constrained by the tragic road accident in Bangladesh that resulted in the death of two Bangladeshi team members and serious injury to three other team members causing much distress and disruption to our research. Loss of trained staff has caused considerable grief to the rest of the team members and much work needs to be done to rebuild the team capacity in BangaldeshBangladesh - this could lead to significant disruptions to Australian research as well, during 2002.

Following on from year 2000 study, surface and subsurface soils were sampled from a number of contaminated sites from the least and most severely As affected districts in Bangladesh, railway contaminated sites in South Australia and old mine-site contaminated sites currently used for cropping in Ballarat, Victoria. Data from these studies in Bangladesh revealed low level contamination of surface soils. In general, total As content of surface and subsurface soils was less than 50 mg kg-1. However, examination of As content of crops from these sites revealed appreciable bioaccumulation of As that ranged from <0.1 mg kg-1 to >20 mg kg-1. Analyses of food samples from a number of households, together with total daily intake of food crops, suggests that approximately 50% of As ingested by Bangladeshis from these sites is via food. Work is now in progress to assess whether ingestion is primarily due to As present in food crops or from the tube well water that is being used for cooking. In contrast to soils from Bangladesh, the total As content of rail track contaminated soils from South Australia ranged from < 50 mg kg-1 to > 500 mg kg-1. A similar range in As content was also recorded for Ballarat samples. Further detailed studies using the contaminated soils revealed significant bioavailabiltity of As in these soils. This was also reflected in both plant samples collected from the contaminated sites and glasshouse studies that show 38 mg kg-1 of As in pealed radish grown using the contaminated soils.

Extensive research was conducted in Australia and Bangladesh during the period 1 January to 30 November 2002. Sampling and analyses of soils, plants and water continued in both Australia and Bangladesh with key towns and villages in both countries having been identified as key sites requiring further in-depth sampling. Along with these activities research was conducted at CSIRO Land & Water, Adelaide to further investigate the behavior of arsenic in soils under laboratoryconditions, the impact of microbial activity on arsenic speciation in alkaline soils as well as determining the human bioaccessibility of arsenic in selected contaminated sites for risk assessment predictions. Further studies were also conducted under glasshouse conditions to investigate the potential phytotoxicity of arsenic to commonly grown garden vegetables in several major South Australian soil groups. Research in highly contaminated soils revealed that arsenite is the dominant arsenic species present in highly alkaline soils and that this process may be controlled by abiotic rather than biotic processes. Studies of approximately 42 soils collected from various arsenic contaminated sites in South Australia and Victoria revealed that the bioaccessibility of arsenic contaminated soil is generally less than 50% of the total arsenic content of the soil. This has important implications when quantifying the exposure pathways and the risk associated with arsenic contaminated sites.
In Ballarat, the University of Ballarat continued to build an extensive database on a further two highly contaminated areas of central Victoria which are undergoing rapid changes in land use. This database will later be transferred onto a GIS mapping package to be used by regulators, farmers and the community. The major findings from this extensive survey were that there is considerable off-site migration of arsenic downstream of the contaminated site which is continuing today. This has led to a significant increase in the area of land outside the contamination source which contains elevated concentrations of arsenic and may be being utilized for other agricultural purposes.
In Bangladesh, three villages were identified for further extensive sampling; these being in the Munshigang, Narayanganj and Comilla districts. These three districts have a well documented history of patients presenting with symptoms of arsenocosis at the Dhaka Community Hospital. At each village soil, plant and groundwater samples were collected and analysed for arsenic content. Speciation of groundwater samples revealed arsenic present as arsenite and concentrations ranging between 5 and 800 mg L-1. The current Australian drinking water standard for arsenic is 7 mg L-1. Analyses of soil samples collected from agricultural areas around each village found that soil concentrations of As were generally less than 20 mg kg-1. However, examination of As content of crops from these sites revealed appreciable bioaccumulation of As that ranged from less than 0.1 to greater than 20 mg kg-1 indicating that irrigation groundwater may be a significant source of arsenic for the crops. Glasshouse studies were conducted for a select number of crops which have shown to accumulate arsenic. Studies revealed that arum consistently accumulated higher concentrations of arsenic than other crops studied irrespective of the environmental conditions the crops were grown in. The addition of organic manures were shown to have little or no effect on the accumulation of arsenic but the addition of low levels or phosphate fertiliser (20 kg P ha-1) increased the amount of arsenic accumulated while elevated additions of phosphate (40 kg P ha-1) decreased the accumulation of arsenic by the crops studied.

This project is being conducted in Bangladesh with the collaboration of Dhaka University. This four year study commenced in 1999 and focused on the pathways of arsenic transfer from arsenic contaminated groundwater to humans. Field surveys have collected groundwater, soil, plant and human health information from three villages in the districts of Munshiganj, Narayanganj and Comilla. These three districts have a well documented history of patients presenting with symptoms of arsenocosis at the Dhaka Community Hospital. Complimenting the field surveys, daily dietary information from Bangladeshi farming communities was also collected from Sirajdikhan village in Munshiganj, Chandina village in Comilla and Sonargaon village in Narayanganj. Collection of the daily dietary information is the first step in evaluating the risk to Bangladeshi villagers associated with ingesting arsenic from various sources. The major potential sources of arsenic exposure identified during the preliminary arsenic exposure pathway assessment conducted during a meeting held in Dhaka during November between Australian and Bangladeshi scientists. These major exposure pathways included the consumption of water and food, particularly vegetables, contaminated with arsenic. Of particular interest is the amount, frequency and arsenic concentration in rice consumed by the Bangladeshi villagers. To supplement the arsenic database for rice further rice grain and straw samples have been supplied by the Bangladesh Agricultural Research Institute. These rice samples have been collected from various districts throughout Bangladesh and represent dry season rice (Boro rice) which has been irrigated with arsenic contaminated groundwater. Analysis of these rice samples indicate elevated concentrations of arsenic in both the rice grain and straw. Arsenic concentrations in the rice grain ranged from 0.04 to 0.76 mg kg-1 dry weight with a mean arsenic concentration of 0.310.25 mg kg-1 dry weight. Rice straw concentrations were markedly higher ranging from 0.48 to 10.48 mg kg-1 dry weight and with a mean arsenic concentration of 3.893.5 mg kg-1 dry weight.

A major component of this project is the communication of information developed by the project with Government, NGOs and the farming community. To this end an education pamphlet has been developed as part of a simple education strategy for the education of Bangladeshi villages impacted by the contamination of arsenic. The development of this pamphlet has gained the support of UNICEF (Bangladesh) and the preliminary design and outline in English has been completed by researchers. The pamphlet will be translated into Bangladesh by collaborators at Dhaka University before distribution to villagers already surveyed by Dhaka University collaborators.

It is now well recognised that the ingestion of arsenic (As) contaminated groundwater is the major cause of As poisoning in Bangladesh. Although scientific research in the last 5 years has focused on ingestion of As through the ingestion of contaminated groundwater, the observation that poisoning among the Bangladeshi population is not consistent with the level of water intake has raised questions on potential pathways of As ingestion. Inconsistencies in the clinical symptoms of As poisoning recorded in Bangladesh raise questions on the pathways of As transfer from groundwater to the Bangladeshi population. While the food habit, nature and amount of food intake play some role in the As dilemma, recent research under ACIAR funding (LWR1 1998/003) reveals significant potential for uptake of As via ingestion of backyard vegetables and rice irrigated with As contaminated groundwater. A dietary survey of over 60 households from 3 districts recorded over 100 food types ingested by the Bangladeshi's in rural villagers. The contributions of various components to the daily As intake are contributed by drinking water, but cooking water is also important. Arsenic intake via food was dominated by the consumption of rice, with arum and other components contributing only a moderate amount of As to the rural Bangladeshi daily As intake in comparison. The modeling of the daily As intake by Bangladeshi's from the three villages surveyed found that 42 % of the villagers consume a diet in which the daily As intake exceeds 220 g As day-1. The current US EPA guidelines recommend that the oral human exposure reference dose (RfD) for As does not exceed 0.0003 mg kg-1 day-1 without causing an adverse non-cancer effect. However, considerable variation in the dietary As intake between the 3 villages was modeled. Assessing the contribution of the food components only to the daily dietary As intake indicates that only 5 % of village A population received an As intake that exceeded the 220 g As day-1 RfD guideline, which is considerably lower than the 42% of the village B population who exceed the same guideline. One of the major issues that needed to be addressed was the potential bioavailability of As ingested in the Bangladeshi diet. It is generally considered that As in groundwater is 100 % bioavailable. However, the human bioavailability of As in plant material was unknown. Paddy rice was cultivated over a period of 6 months in glasshouse conditions to examine As bioavailability in cooked As contaminated rice using the swine animal model. Two scenarios were investigated; cooked As contaminated rice, and uncontaminated rice cooked in As contaminated water. In both scenarios As was potentially bioavailable, although the bioavailability of As in cooked contaminated rice was only 45 % (mean) compared to 85 % (mean) for the uncontaminated rice cooked in contaminated water.

A major component of this project is the communication of information developed by the project with Government, NGOs and the farming community. To this end an education pamphlet has been developed as part of a simple education strategy for the education of Bangladeshi villages impacted by the contamination of arsenic. The development of this pamphlet has gained the support of UNICEF (Bangladesh) and the preliminary design and outline in English has been completed by researchers. The pamphlet will be translated into Bangladesh by collaborators at Dhaka University before distribution to villagers already surveyed by Dhaka University collaborators.