Effect of Arsenic Contamination on Rice Production and Human Health: Insights from Farmers' Perceptions*

M. Shahe Alam1, J. Kabir2, M.A. Islam3 and M.A.Salam3

 Paper submitted for presentation in the ASAE International conference on ’Meeting the Challenges Facing Asian Agriculture and Agricultural Economics toward a Sustainable Future ‘ to be held in Hanoi, Vietnam during 13-15 October,2011 . 1, Former CSO and Head, Agricultural Economics Division, Bangladesh Rice Research Institute, 2,3, Senior Scientific Officer and Scientific Officers respectively, Agricultural Economics Division, Bangladesh Rice Research Institute, Gazipur 1701, Bangladesh

Effect of Arsenic Contamination on Rice Production and Human Health: Insights from Farmers’ Perceptions

M. Shahe Alam, J. Kabir, M.A. Islam and M.A.Salam

Abstract

Much of the shallow groundwater in southeastern and southwestern parts of Bangladesh is naturally contaminated with arsenic exposing more than 40 million people to unsafe levels of arsenic in drinking-water and potentially threatening rice production and food security as well. A study was undertaken to assess the level of arsenic contamination in rice production, determine the level of changes in intake of rice in the arsenic-overwhelmed population and find out the possible effect of arsenic contamination on human health. Sample survey was carried out in three arsenic-prone districts. Analysis revealed that, the yield of modern rice in arsenic-contaminated plots was substantially low compared to that in less-contaminated plots. Most sample households used hand-operated tubewell-water for drinking and daily household purposes. The proportion of arsenic-affected patients was higher in Kachua (14%) than that in Bhanga area. About 37% and 30% of the household heads in Kachua and Bhanga respectively opined that consumption of contaminated rice was another cause of arsenicosis. Women were more exposed to arsenic contamination since the proportions of female patients in all locations were much higher compared to their male counterparts. No children aged below 5 years, were suffering from arsenic problems indicating that arsenicosis is expressed after long-term intake of arsenic through water and food. The findings indicate that the study subjects have inadequate access to arsenic-free water and consequently people had the sufferings. Due to the excessive use of arsenic-contaminated irrigation water and chemical fertilizers, soil health might have been deteriorated substantially. The Government should undertake various action programmes to make people aware of the arsenic problems in Bangladesh.

Keywords: Arsenic contamination, groundwater, modern rice, arsenicosis, health hazard, food safety; ______The authors are Chief Scientific Officer, Senior Scientific Officer and Scientific Officers respectively, Bangladesh Rice Research Institute, Gazipur 1701, Bangladesh.

1. Introduction In Bangladesh, the cultivation of irrigated winter (Boro) season rice has increased tremendously since 1970s and at present the area under Boro is about 4.71 million hectares which is about 42% of the total rice area contributing about 57% of the total rice production in the country (BBS, 2009). Available statistics indicate that, about 57% of the total arable land area is currently irrigated from shallow tube wells (BER, 2010). However, much of the shallow groundwater in south-eastern and south-western part of Bangladesh is naturally contaminated with arsenic (As) exposing more than 40 million people to unsafe levels of arsenic in drinking water and potentially threatening food security. Build up of arsenic in soil due to use of arsenic contaminated irrigation water has been shown to elevated levels of arsenic in paddy soils causing reduction in rice yield and eventually high arsenic concentration in rice grains (Islam et al., 2007; Panaullah et al. 2009; Khan et al., 2010). The arsenic content of rice grain is generally much higher than that of upland cereal crops because of the relatively high availability of soil arsenic under reduced conditions (William et al., 2007). This arsenic hazard is a great concern of our country since about 25% people in Bangladesh are affected by arsenic contamination due to drinking of arsenic contaminated water from tube-wells (WHO, 2001). Moreover, it is suspected that there will be possible reduction of crop production due to arsenic contamination if the issue remains unattended. The country cannot afford these adverse effects since it is already struggling to meet the demand for food emanating from population growth. High concentration of arsenic has been found in ground water from thousands of hand tube-wells under 60 out of 64 districts across the country. The groundwater in these districts has been reported to be contaminated at various degrees and about 85 million people may be at some health risk due to the ingestion of this contaminated water (Huq et al., 2001) and Ghani et al, 2004). However, adverse impact of arsenic contamination of ground water has been categorized as primary and secondary by Khuda (2001). The primary impact is on the health of individuals who are exposed to arsenic poisoning through drinking ground water lace with arsenic. After several years of low level arsenic exposure, various skin lesions appear. These are manifested by different symptoms i.e. dark spots, white spot, keratosis of hands and feet. The secondary impact is an out come of the primary impact and is reflected in the socioeconomic consequences like

inability to do economic/productive works, social exclusion, problems of getting married etc (Jaim et al, 2007). Study on different aspects of agriculture, such as: effects of arsenic contamination on crop, animal and human health is very limited in Bangladesh. Taking into consideration of the above background, this study was undertaken to evaluate the impact of arsenic on modern rice production and human health in the selected arsenic contaminated areas of Bangladesh.

1.1. Specific objectives The specific objectives of the present study were: 1. to assess the level of arsenic contamination in crop production and understand the magnitude of land degradation due to arsenic contamination, 2. to evaluate the productivity and profitability differences in modern rice production due to arsenic contamination, 3. to understand the level of change(s) in intake of rice and other food crops in the arsenic overwhelmed areas and assess the effect of arsenic on human health; and 4. to document farmers’ coping mechanism in overcoming the arsenic related problems.

2. Methodology 2.1. Study area selection Under the IFAD collaboration, the soil science, irrigation & water management (IWM) and plant breeding divisions of BRRI have carried out field experiments on the prevalence of arsenic in the irrigation water and crops in three selected locations under three districts of Bangladesh. The areas were Kachua upazila of Chandpur, Faridpur sadar and Bhanga upazila of Faridpur district. Based on the reports of earlier experiments, sample locations for the present socio-economics study were identified. Two villages under each of the above Upazila were selected for carrying out in-depth socio-economics survey. 2.2. Sampling and data A comprehensive list of the rice producing farms in each village was prepared through taking help of concerned Sub-assistant Agriculture officer. For the present study, the farms those used shallow tube wells (STW) for crop production were the representative samples. Out of those, fifty farms from each village were selected following random

sampling technique. Data were collected through directly interviewing the selected farmers using pre-designed questionnaire during November, 2009 - May, 2010. Collected data were scrutinized, edited and compiled using appropriate computer software. Both descriptive and inferential statistics were employed in analyzing the data. To examine the mean differences for different items of input use, productivity and profitability between rice fields near the shallow tube well (more contaminated) and far end of the command area (less contaminated), mean test was done employing the following formula:

x1  x2 t  s 2 s 2 1  2 n n2 1

Where,

1 2 S 2  x  x 1 n 1 i 1

1 2 S 2  x  x 2 n 1 j 2

i, j= 1, 2, 3……n.

3. Results and Discussion 3.1. Socio economic profile of the sample households The socio-economic profile of the sample farm households are presented in Table 1. The average family size was slightly higher (6.0) in Kachua compared to those of other two locations. The overall family size was 5.57 which is higher than the national average of 4.9 (BBS, 2009). In terms of literacy, the proportion of people under primary education in all the three locations was more or less similar. About one-fourth of the household members in all locations were illiterate. It was also evident that, majority of the household members had either primary or secondary levels of education and the household members with secondary school certificate (SSC) and above were considerably low in all three study locations.

Table 1: Socio-economic characteristics of household members in the selected study areas.

Socio-economic characteristics Kachua Bhanga Faridpur All Sadar Locations Average family size: 6.0 5.6 5.1 5.57 Male: 2.9 2.8 2.8 2.83 Female: 3.2 2.8 2.3 2.77 Education of the family members (%) Illiterate: 20 16 28 21.33 Primary: 39 36 32 35.67 Secondary: 21 32 30 27.67 SSC & HSC passed: 18 13 8 13.00 Graduation & Masters: 2 3 2 2.33 Total: 100 100 100 100 Main sources of income (%) Agriculture 46 48 53 49.00 Business 23 13 20 18.66 Service 7 8 14 9.67 Wage 5 2 7 4.67 Remittance 16 26 10 15.33 Others 3 3 2 2.67 Total 100 100 100 100 Total annual income (Tk./HH): 131667 184659 151793 156,039

Household annual income was higher in Bhanga (Tk. 184,659) followed by Faridpur sadar (Tk. 151,793) and Kachua (Tk. 131,667). Agriculture was the main source of income of the households in all three locations. However, the share of agriculture income was higher (53%) for the sample farms under Faridpur sadar compared to that of Bhanga (48%) and Kachua (46%). In Faridpur sadar and Kachua, business was the second important income source. The share of remittance in household income was 26% for the households under Bhanga and it was a bit lower in case of the households of other two locations. The major proportion of households in all three study locations had ‘Kaccha’ house. Number of houses per household was 1.8, 1.43 and 1.28 in Kachua, Bhanga and Faridpur sadar, respectively.

3.2. Land holdings and tenancy The sample farms were categorized into four farm-size groups, i.e. marginal, small, medium & large, and their distribution are presented in Table 2. The bulk of the sample farms in all three study villages (above 50%) fell under the small farm size category. The proportion of

marginal farms was a bit higher in Kachua compared to those of Bhanga and Faridpur sadar. However, the proportion of large farms was remarkably low in all the study villages. Among the sample farms in the study villages, 50, 62 and 65% farms in Kachua, Bhanga and Faridpur sadar respectively, were the owner operators. The proportion of owner-cum tenant (i.e. part tenant) farms varied from 32 to 42% in different locations. It is important to note that, the level of tenant farms was much higher in Kachua compared to that of Bhanga and Faridpur sadar. Table 2: Land holding and tenancy status of the sample households in different locations

Items % of farms Kachua Bhanga Faridpur sadar Farm size: Marginal ( < 0.49 acre) 32 13 20 Small (0.50 – 2.49 acre) 54 67 52 Medium (2.50 – 7.5 acre) 12 18 25 Large ( > 7.50 acre) 2 2 3 Tenural status: Owner operator 50 62 65 Part tenant 42 36 32 Tenant 8 2 3

3.3. Effect of arsenic contamination on crops and lands Farmers at all three locations mentioned about the reduction of land quality due to arsenic contamination. About 83, 74 and 69% farmers of Bhanga, Kachua and Faridpur sadar pointed out that, Boro rice fields were contaminated with arsenic. More than 80% farmers both in Bhanga and Faridpur sadar expressed their opinion about the decrease in soil fertility because of arsenic contamination, while 63% farmers in Kachua also had the same notion. However, other arsenic related problems on rice fields were the reddish coloration of irrigation channel & rice fields, hardness of lands etc as reported by majority of the farm households in all three study sites (Table 3).

Table 3: Prevailing effects of arsenic contamination on crop fields

Items % farmers opined Kachua Bhanga Faridpur Sadar Crop fields contaminated with arsenic 74 83 69 Increases soil salinity 12 4 7 Decreases soil fertility 63 87 87 Irrigation canals and rice fields become red 92 89 80 Land become hard 91 87 69 Rice yield near irrigation channel/STW is low 81 82 46 No effect/No idea 3 8 9 Note: Multiple responses considered

3.4. Use of irrigation equipments

Both shallow tube wells (STWs) and deep tube wells (DTWs) were used by the farmers for irrigation purpose. However, the present study concentrated mainly on STW users in the study villages. The average depth of STWs in the study areas ranged from 80-90 feets (Table 4). In terms of average duration of installation, the highest was in Bhanga area (13 years), while the lowest duration was observed in Kachua (about 9 years).

Table 4: Status of irrigation equipments and level of irrigation. Items Upazila Kachua Bhanga Faridpur sadar Average depth of STW (feet) 80 90 82 Average frequency of irrigation for rice 16 28 22 Av. years of installation 9.37 13 9.5

3.5. Effects of arsenic contamination on rice Major problems encountered by the sample farmers in rice production due to arsenic contamination are furnished in Table 5. More than 70% farm households in Kachua reported that they were familiar with various arsenic related problems in rice production which were: less tillering of rice plants, plants become shorter in height, plants do not flower timely or uneven flowering etc. Similarly, about 60% farmers further observed that grains do not mature timely and more grains remain unfilled. However, those problems eventually resulted to reduction of rice yield as reported by 80% farmers in all three study areas. It is important to note that, comparatively higher proportion of farms under Bhanga were facing most of the yield retarding factors in growing MV Boro rice than those of both Kachua and Faridpur sadar.

Table 5: Farmers’ observation on the effects of arsenic contamination in rice

Items of visual effect % farmers opined Kachua Bhanga Faridpur Sadar Less tillering 77 91 87 Plants become shorter in height 75 92 82 Plants become red or yellowish 77 92 85 Plant growth not uniform 72 92 77 Plants do not flower uniformly 71 91 75 More unfilled grains 66 89 73 Decrease rice yield 84 89 82 High insect and disease infestation 41 27 14 Note: Multiple responses considered

Fig 1: Arsenic contaminated rice field with ununiform flowering

3.6. Steps taken by the farmers for improving the crop condition

Most of the sample farms under all the study locations took different types of measures for improving the condition of the affected crops. In general, farmers had the tendency of applying more fertilizers in arsenic contaminated plots. Nearly 70% farmers in Kachua applied more urea in the arsenic affected plots, while 64% farmers in Bhanga applied more urea to overcome the aforesaid problem (Table 6). Table 6: Farmers’ indigenous practices in reducing the effect of arsenic on land and crop productivity

% farmers followed Steps taken Kachua Bhanga Faridpur Sadar Apply less water 21 17 12 Drain out the water and allow the field to dry 21 29 20 Apply more urea (N) fertilizer 70 62 32 Apply more TSP (P) fertilizer 12 17 11 Apply more MP (K) fertilizer 25 36 10 Apply more gypsum fertilizer 82 79 59 Apply more zinc fertilizer 55 52 63 Apply plant growth regulator 69 59 30 Did mulching 80 61 32 Apply more cow dung 44 33 25 Apply Ash 23 16 16 Note: Multiple responses considered

In Kachua, almost 80% farms applied more gypsum and did mulching to the affected plots, while 60% farms in Bhanga and 30% farms in Faridpur sadar also did mulching. Usually, farms under the study areas apply less water or allow drying up the affected plot before mulching and this was practiced by 20% farms in all the sample areas. It was also observed that, more than 50% farms under all three study areas applied zinc sulphate to overcome the arsenic problem. Apart from these, 69% farms in Kachua applied growth hormone to accelerate the growth of affected crops while in Bhanga and Faridpur, growth regulator was also applied by 59 and 30 % farms respectively (Table 6).

3.7. Changes in cropping patterns over time Data on the major cropping patterns followed by the sample farms are presented in tables 7, 8 and 9. In the sample villages under all three study locations, previously farmers used to follow few cropping patterns. In the village under Faridpur sadar, farmers’ main pattern was Boro- fallow –T. Aman. Previously some of the farms used to grow pulses before growing Boro

followed by T. Aman. However, now a day, farmers are found to practice some new cropping patterns in addition to the earlier ones. Out of the new cropping practices, wheat followed by jute or wheat followed by Aus rice followed by Aman are the favorite patterns. It is important to note that, few farmers who faced problem of arsenic contamination in particular plots, they tended to change the cropping practices for those plots. Farmers adopted some new patterns with the understanding that, through changing the cropping practices they could overcome the arsenic problem. In the sample village under Bhanga upazila, almost similar changes in cropping patterns were observed (Table 8).

Table 7: Changes in Boro rice based cropping patterns due to arsenic contamination in Faridpur sadar

Previous patterns Present patterns Boro-Fallow-T.Aman Boro-Fallow-T.Aman Boro-vegetables-T.Aman Boro-vegetables-T.Aman Pulses-Boro- -T.Aman Pulses-Boro- -T.Aman Wheat-Jute-T.Aman Wheat-Aus-T.Aman Wheat-Jute-Fallow Wheat-Fallow-T.Aman Vegetables-Fallow-T.Aman

Table 8: Changes in cropping pattern due to arsenic contamination in Bhanga

[[ Previous patterns Present patterns Boro-Fallow-T.Aman Boro-Fallow-T.Aman Boro-Aus T.Aman Boro-Aus T.Aman Boro-Fallow-B.Aman Boro-Fallow-B.Aman Mustard -Boro- -Fallow Mustard -Boro- -Fallow Boro-Fallow-Fallow Boro-Fallow-Fallow Wheat-Jute-T.Aman Wheat/pulses-Jute-B.Aman Wheat/Chili-Jute-B.Aman Wheat-Jute-Fallow

Table 9: Changes in cropping pattern due to arsenic contamination in Kachua

Previous patterns Present patterns Boro-Aus-T.Aman Boro-Aus-T.Aman Boro-Fallow-T.Aman Boro-Fallow-T.Aman Wheat-Aus-T.Aman Wheat- Vegetable -T.Aman Pulses-Chili -T.Aman

3.8. Input use for more arsenic contaminated & less contaminated fields

The comparative input use level and corresponding cost for MV Boro production in more arsenic contaminated fields (near the STW) and less contaminated fields (periphery of the command area) under different study locations can be viewed in Table 10. In general the cost of fertilizers for boro production was much higher for the more arsenic contaminated plots than the less contaminated plots. In terms of fertilizer by type, the use of gypsum and zinc sulphate in arsenic contaminated plots was almost three times higher than that used for less arsenic contaminated plots. The additional application of some of the chemical fertilizers and for doing mulching for the more arsenic affected plots, the level of human labour use was also much higher. The total cost of boro production for the arsenic contaminated plots was also much higher compared to that of less contaminated plots irrespective of study locations.

Table 10: Comparative costs of MV Boro rice cultivation in more arsenic contaminated and less arsenic contaminated plots during 2010

Items of inputs/cost Kachua, Chandpur Faridpur sadar Bhanga, Faridpur More Less More Less More Less arsenic arsenic arsenic arsenic arsenic arsenic affected affected affected affected affected affected plots plots plots plots plots plots Human labour (Tk./ha) 22937 21597 21108 19399 19138 17382 PT/DP cost (Tk/ha) 3428 3299 2904 3147 3544 3635 Seed (Tk/ha) 1149 1229 1469 1622 1275 1171 Manure (Tk/ha) 2712 2812 1599 1048 1337 474 Chemical fertilizer (Tk/ha) 7458 6651 10081 8409 9018 8599 Urea (Tk/ha) 2973 2859 3922 3536 3840 3773 TSP (Tk/ha) 2224 2185 2656 2658 2439 2547 MP (Tk/ha) 1304 1302 1475 1321 1465 1610 Gypsum (Tk/ha) 813 270 1664 675 1034 578 Zinc sulphate (Tk/ha) 144 35 364 219 240 91 Insecticide/Pesticide cost 1753 1434 1161 1025 731 708 (Tk/ha) Theovit (Tk/ha) 458 52 141 97 33 19 Herbicide cost (Tk/ha) 143 19 Irrigation cost (Tk/ha) 8904 8993 12336 12672 12155 12318 Total variable cost (Tk./ha) 48799 46067 50942 47438 47217 44320 Interest on operating capital @ 915 864 955 889 885 831 9% Rental value (Tk/ha) 14500 14500 13500 13500 12500 12500 Total cost (Tk./ha) 64214 61431 65397 61827 60602 57651

3.9. Productivity and Profitability

The level of grain yield obtained by sample farms and other pertinent economic analysis for different locations are presented in Table 11. The grain yields of MV Boro rice in the more arsenic contaminated plots in all three locations were substantially low compared to that of less contaminated plots. Consequently, the gross return from the more arsenic affected plots was also very low. In particular, the per hectare gross return for arsenic affected plots in Kachua and Faridpur sadar were Tk. 55708 and Tk. 59385 respectively, while gross return for less affected plots of those two areas were Tk. 80,188 and Tk. 91,449 respectively. The net return was negative for the arsenic affected plots for those two sites. The return from investment (BCR) was only 0.87 and 0.91 for the arsenic affected plots of Kachua and Faridpur implying that for each taka investment in Boro production for the arsenic affected plots, the concerned farmers earned less than a taka. However, in case of Bhanga, a different picture appeared. Boro producing farms in Bhanga had not to incur loss in growing Boro even for the arsenic affected plots. The rate of returns for the farms under Bhanga were 1.10 and 1.75 for the more arsenic affected and less affected plots respectively. This was possible may be because of the fact that, in Bhanga area Boro is grown as a single crop and because of better soil fertility, Boro producing farms usually obtain a bit better yield that helped the sample farms to achieve higher gross return & thus could avoid loss in Boro production.

Table 11. Productivity and Profitability of MV Boro rice cultivation in more arsenic contaminated and less arsenic contaminated plots during 2010

Items Kachua, Chandpur Faridpur sadar Bhanga, Faridpur More Less More Less More Less arsenic arsenic arsenic arsenic arsenic arsenic affected affected affected affected affected affected plots plots plots plots plots plots Yield: Grain (kg/ha) 3220 4635 3483 5364 3896 5934 Straw (kg/ha) 2576 3708 2786 4291 3117 4747 Gross Return (Tk./ha) 55708 80188 59385 91449 66425 101174 Grain (Tk/ha) 53131 76480 56599 87158 63309 96427 Straw (Tk/ha) 2576 3708 2786 4291 3117 4747 Gross Margin (Tk./ha) 6909 34121 8443 44011 19208 56854 Net Return (Tk/ha) -8506 18757 -6012 29622 5823 43523 BCR (Undiscounted) 0.87 1.31 0.91 1.48 1.10 1.75

3.10. Significance of the differences in input use and cost and return Results of ‘t’ test for different items of production and cost and returns for different locations are presented in Tables 12, 13 and 14. It was found that, per hectare human labour cost for the more arsenic contaminated plots was higher by Tk.1756, Tk.1706 and Tk.1340 in Bhanga, Faridpur sadar and Kachua respectively compared to the less arsenic contaminated plots. These differences were found highly significant. This suggests that there was a significant difference in the use of human labour for producing Boro rice in arsenic contaminated fields and less contaminated plots. It might have happened due to additional use of labor for mulching and fertilizer application to the arsenic affected plots. There were significant differences in the use of gypsum and zinc sulphate between the arsenic contaminated and less contaminated plots in all the locations. This seems quite reasonable, because additional gypsum and zinc sulphate were applied to the arsenic contaminated plots when crops became yellowish/red. There were significant difference in fertilizer cost at Faridpur sadar and Kachua. Differences in the cost of producing Boro rice between more arsenic contaminated plots and less arsenic contaminated plots were Tk. 2898/ha, Tk. 3504/ha and Tk. 1565/ha, respectively. This differences were also statistically significant implying that, the difference in total variable cost for producing Boro rice in more arsenic contaminated and less contaminated plots were really substantial. Table 12: Operation-wise mean differences of cost and return for producing Boro rice in more arsenic affected and less affected plots at Bhanga

Mean differences between more Std. t value df Cost and returns arsenic contaminated & less Deviation arsenic contaminated plots (Tk/ha) Human labour 1756 632 23.6*** 71 Manure 863 2270 3.2** 71 Urea 67 1835 0.3 71 Gypsum 456 968 4.0*** 71 Zinc sulphate 150 416 3.1** 71 Total fertilizer 419 3924 0.9 ns 71 Insecticide 23 2341 0.1 ns 71 Total variable cost 2897 7495 3.3** 71 Gross Return -34749 28013 -10.5*** 71 Gross Margin -37646 27937 -11.4*** 71 ***, **, * indicate significant at 1%, 5% and 10% level.

However, farmers obtained much lower gross return and gross margin for the arsenic contaminated plots than that of less contaminated plots in all three study locations. The

differences in gross return were Tk. 34749/ha, 32064/ha and Tk 24480/ha for Bhanga, Faridpur sadar and Kachua, respectively. Table 13: Operation-wise mean differences of cost and return for producing Boro rice in more arsenic affected and less affected plots at Faridpur sadar

Mean differences in Std. Deviation t value df Cost and returns cost and return (Tk/ha) Human labour 1709 1269 11.4*** 71 Manure 551 4200 1.1 ns 71 Urea 386 1353 2.4** 71 Gypsum 989 2123 4.0*** 71 Zinc sulphate 145 488 2.5** 71 Fertilizer 1672 3765 3.9*** 71 Insecticide 137 2101 0.6 ns 71 Total variable cost 3504 9960 3.0*** 71 Gross Return -32064 35251 -7.7*** 71 Gross Margin -35568 34893 -8.6*** 71 ***, **, * indicate significant at 1%, 5% and 10% level.

It could be inferred from this result that, farmers although apply significantly higher amount of inputs in more arsenic contaminated plots than that for less contaminated plots: but due to arsenic contamination, the crop condition remains poor and eventually they were able to reap less output and thus harness less benefit.

Table 14: Operation-wise mean differences of cost and return for producing Boro rice in more arsenic affected and less affected plots at Kachua.

Mean differences in Std. Deviation t value df Cost and returns cost and return (Tk/ha) Human labour 1340 1547 7.1*** 67 Manure -100 3687 -0.2 ns 67 Urea 114 845 1.1 ns 67 Gypsum 543 370 12.1*** 67 Zinc sulphate 109 291 3.1*** 67 Total fertilizers 807 2003 5.0*** 67 Insecticide 319 1476 1.8* 67 Total variable cost 2732 4185 3.1*** 67 Gross Return -24480 37979 -5.3*** 67 Gross Margin -27212 37576 -5.7*** 67 ***, **, * indicate significant at 1%, 5% and 10% level.

3.11. Level of food intake by the sample farms Data on weekly food intake under all three study areas are given in Table 15. It was evident that, weekly consumption of rice, wheat, fish, vegetables, potato by the households in Kachua is comparatively lower than those of Bhanga and Faridpur sadar. Table 15. Level of food consumption by the sample households in different study locations

Food items Kachua Bhanga Faridpur Sadar gm /head /week gm /head /week gm /head /week Rice 2608 3236 3217 Puffed rice 235 136 165 Wheat 299 373 432 Fish 309 436 453 Meat 112 102 84 Chicken 134 144 253 Milk 168 395 148 Egg 1 1.3 0.8 Vegetables 547 699 627 Potato 328 503 572 Pulses 91 125 96 Weekly consumption of puffed rice was higher in Kachua compared to those of Bhanga and Faridpur sadar. As the level of income per house hold in Kachua is low compared to those of Bhanga and Faridpur the level of overall nutrition intake by the family members in Kachua is also low compared to those of Bhanga and Faridpur (Table 15). This could be one of the causal factors behind the high intensity of arsenic affected patients in Kachua compared to that of Bhanga.

3.12. Sources and uses of drinking water Majority of sample households (almost 100%) under the study locations used tube well water for drinking and daily household purposes. It also appeared that, STW water was also used for drinking purpose by the households at varying degrees. However, the proportion of households using STW water was 28% in Kachua while it was less than 10% both in Bhanga and Faridpur sadar implying that the villagers both in Bhanga and Faridpur sadar had more access to drinking arsenic free water from sources other than STWs. It was also observed that, more than 20% households in Kachua and Bhanga harvested rain water for drinking and cooking purposes (Table 16).

Table16. Farmers’ sources of drinking water in different study locations

Sources of drinking % of total households Kachua Bhanga Faridpur Sadar Tube-well 100 100 100 STW 28 5 8 Rain water 23 22 13 Note: Multiple responses considered

3.13. Sources of arsenic contamination The causes of arsenic contamination in human health are described in Table 17. Both in Bhanga and Faridpur sadar, all respondents reported that drinking arsenic contaminated tube-well water was the main cause of arsenic poisoning in human health. Farmers had the perceptions on drinking water: one of the possible reasons behind it could be that, different agencies like GO and NGOs through their continued efforts had been able to increase awareness of the people about arsenic contamination in human health. In Kachua, about 37% household heads opined that, consumption of contaminated rice was another cause of arsenicosis disease. Similarly, about 29 and 21% respondents of Bhanga and Faridpur sadar respectively, reported the similar notion. However, sample households under all three locations opined that, consumption of contaminated vegetables may also be the cause of arsenic contamination in human health.

Table 17. Stated possible reasons of arsenic contamination in human health

Sources of arsenic as farmers’ idea % of respondents opined Kachua Bhanga Faridpur Sadar

Drinking contaminated water 98 100 100 Consumption of contaminated rice 37 29 21 Consumption of contaminated Vegetables 14 18 16 Note: Multiple responses considered

3.14. Prevalence of arsenic related diseases according to gender differences Distribution of arsenic affected patients by sex depicted that, the proportion of arsenic affected patient were higher (13.5%) in Kachua than that of Bhanga (Table 18). On the other hand, arsenic affected patient in Faridpur sadar was much lower (1.6%) compared to those of other two locations. Arsenic affected female patients were higher (14%) in Kachua followed by 9% in

Bhanga and about 2% female members in Faridpur sadar. However, the proportion of female patients in all three locations was a bit higher compared to their male counterpart. This result further implies that, women are more exposed to the arsenic contamination through different sources possibly due to under taking various household activities.

Table 18. Distribution of arsenic affected patients in the study areas by sex

Sex Kachua Bhanga Faridpur Sadar No. of % of No. of % of No. of % of patients total patients total patients total Male 35 13 23 7 6 1.30 Female 36 14 23 9 7 1.97 All 71 13.5 46 8.5 13 1.60

Distribution of arsenic affected people by age and locations could be viewed in Table 19. It appears that, among the sample households under three study locations, there was no case of children below 5 years of age suffering from arsenic problems. This may be due to the fact that arsenicosis is a dilatory process and its symptoms on health could be visible after long term exposure to arsenic due to taking arsenic contaminated water or from other sources. Family members of all age groups above 5 years were affected by arsenic contamination both in Kachua and Bhanga, while in Faridpur sadar, family members belonging to the age group of 6-14 years were not affected by arsenic. However, young generations belonging to the age group of 15-30 and 31- 45 years are comparatively more vulnerable to arsenic contamination in all the three study areas. In Kachua, the proportion of household members under the category of 31-45 years was 31% while it was more than 20 % both in Bhanga and Faridpur sadar. In Faridpur sadar, household members under the age ranging 46-60 years are also significantly affected (38%) by arsenic contamination. It was also evident that, the proportion of household members above 60 years of age with arsenic problems was 21 and 22 % in Kachua and Bhanga. But in Faridpur sadar, the proportion of household members under this category was quite low (8%).

Table 19. Distribution of arsenic affected patients ( by age) in the study areas

Age groups Kachua Bhanga Faridpur Sadar (years) No. of % of No. of % of No. of % of patients total patients total patients total 6-14 6 9 4 9 0 0 15-30 19 27 9 20 4 31 31-45 22 31 13 28 5 38 46-60 9 13 10 22 3 23 Above 60 15 21 10 22 1 8 Total 71 100 46 100 13 100

3.15. Visible symptoms of arsenic on human health

With regards to the questions about different symptoms on arsenic contamination, the sample inhabitants of Kachua reported that 70% of the victims had black spot in their body while 90% of them observed black spot on palm (Table 20). Table 20. Visible symptoms of arsenic contamination on human health in the study areas

Visible Symptoms Kachua Bhanga Faridpur Sadar (N = 71) (N = 46) (N = 13) No. of % of No. of % of No. of % of patients total patients total patients total Melanosis (Black 50 70 35 76 7 54 pigmentation on body) Melanosis (Black 64 90 32 70 8 62 pigmentation on palm) White spot on the 42 59 24 52 7 54 body Skin rash (Tiny red 36 51 40 87 3 23 spots on the skin) Allergy 68 96 37 80 12 92 Keratosis (rough, dry 57 80 28 61 8 62 skin lesion) Weakness 67 94 37 80 13 100 Gout ( Pain in Knees, 57 80 31 67 10 76 Fingers & toes/feet) Limb (Leg, Arm and 45 63 31 67 6 46 Wing) pains Pain in waist 28 39 23 50 7 54 Hair dropped 19 27 6 13 5 38 Note: Multiple responses considered

However, among other symptoms the most alarming ones were allergy, keratosis and weakness which were mentioned by 96, 80 and 94% of the sample respondents in Kachua. Gout, limb pain and skin rash were also reported by 80, 63 and 51 % respondents respectively in Kachua. In case of Bhanga, among different symptoms, the most prominent ones were skin rash, allergy and black spot on the body which were mentioned by about 80% of the sample respondents. Gout, allergy and pain in waist were the prominent symptoms of arsenicosis in Faridpur sadar as reported by more than 90% of the sample respondents. Black spot on palm, keratosis, hair dropped, white spot on body etc were also the remarkable symptoms as reported by about 62% sample respondents (Table 20).

3.16. Arsenic related illness and consequences Majority of the arsenic victims in all study locations had been suffering from arsenic related illness for 3 to 6 years. In sample villages of Kachua about 35% of the arsenic affected patients belonged to this group, while both in Bhanga and Faridpur sadar, about 46% patients were suffering for 3-6 years. However, there were patients who had been suffering for a bit lesser duration (1-3 years). The proportion of long term victims of arsenic contamination was higher in Kachua than those of Bhanga and Faridpur sadar. The proportion of arsenic victims under this duration (9-15 years) was 21% in Kachua while in Bhanga and Faridpur sadar only about 10% patients fell in this duration. Among all three study areas, proportion of arsenic contaminated victims and duration of sufferings were more in Kachua compared to those of Bhanga and Faridpur sadar (Table 21).

Table 21. Arsenic related illness and level of peoples’ sufferings in the sample villages

Years Kachua Bhanga Faridpur Sadar

No. of % of No. of % of No. of % of patients total patients total patients total 1.1-2.9 22 31 14 30 5 38 3.0-6.0 25 35 21 46 6 46 6.1-9.0 9 13 6 13 1 8 9.1-15.0 15 21 5 11 1 8 Total 71 100 46 100 13 100

3.17. Awareness and necessary measures taken Data on necessary measures taken by the sample farmers to cope with the arsenicosis problems are furnished in Table 22.

Table 22. Measures taken by the sample farmers for avoiding the effect of arsenic on health Items % farmers reported Kachua Bhanga Faridpur Sadar Drink arsenic free water 76 77 80 Cook with arsenic free water 81 77 75 Wash cookeries with arsenic free water 70 70 61 Eat more vegetables 40 33 30 Take vitamin C tablet 23 12 13 Take advice from the MBBS Doctor 31 29 34 Take advice from village Doctor 39 5 10 Take advice from ‘Kabiraj’ (village quack) 15 2 12

The people under the study areas felt that drinking arsenic free water would be one of the best safety measures against arsenic contamination on human health and as such 76, 77 and 80% farms in Kachua, Bhanga and Faridpur sadar, respectively adopted this measure. Likewise, about 70% respondents in Kachua used arsenic free water/rain water for cooking and washing purposes. Similarly, about 70 and 61% farms in Bhanga and Faridpur sadar respectively, followed the same practice for cooking and washing purposes. The sample farms under all three locations also have adopted some other measures like taking vitamin C, taking advice from doctors and advice from ‘Kabiraj’ (i.e., village quack).

3.18. Farmers’ perceptions and steps taken Due to growing awareness among the farmers about the effect of arsenic day by day they become confident on certain steps towards overcoming the problems. About 90% respondents in all three locations believed that, the very first precautionary measure for getting rid of arsenic problems is to drink arsenic free water. About 60% farms in Kachua perceived that having nutritious food and more vegetables could also be the safety measures in overcoming the problem of arsenic poisoning. Similarly, about 17 and 24% respondents in Bhanga and Faridpur sadar respectively, had the same impression. More than 20% respondents in Bhanga and Faridpur sadar further suggested that awareness about the harmfulness of arsenic should be built up among the villagers under all affected locations through awareness program, imparting training and through different media e.g. TV, radio, bill board, leaflet, posters etc (Table 23).

Table 23. Farmers' perceptions on overcoming the problem of arsenic in human health

Respondents perception % farmers opined Kachua Bhanga Faridpur Sadar Drink arsenic free water 96 94 98 Eat nutritious food 60 17 24 Eat more vegetables 52 44 37 Increase awareness 23 27 22 Eat arsenic free food 21 23 36 Be neat and clean 11 20 24 Note: Multiple responses considered

3.19. Economic and social impacts

It was evident that, the victims under all three study locations had been suffering from various social and economic problems. In Kachua, about 67% respondents mentioned that arsenic affected people lost their working capability while in Bhanga and Faridpur sadar 62% and 45 % farmers reported about this. The eventual impacts of the problems are really worth mentioning. Nearly 55% farms in Kachua and 55% farms in Bhanga mentioned that, the aforementioned problems ultimately resulted in reducing the level of income of the affected families. The other social problems arose for the arsenic affected females were really painstaking. Getting married appeared to be the main problem for women patients as reported by 34% respondents in both Kachua and Bhanga. However, to some of the victims, arsenic problem was believed to be the curse from God (Table 24). Table 24. Level of eventual effect of arsenic poisoning on farmers' socio-economic condition

Items % farmers reported Kachua Bhanga Faridpur Sadar Work efficiency reduced 67 62 45 Income reduced 52 54 43 Socially isolated 29 14 15 Marriage problem for women 34 34 21 Consider as 'Curse from God' 30 5 10 Note: Multiple responses considered

4. Conclusions and policy implications From the present study, the following conclusions could be drawn: More than 70% of the sample rice farmers were acquainted with various arsenic related problems in growing rice. Categorically the problems were: less tillering ability, plants become shorter, become red/yellowish, un-uniform flowering, unfilled grains etc, resulting to reduction of rice yield. Rice farmers, however, adopted few indigenous way of overcoming the problem. Majority of the farms applied more gypsum and did mulching to the rice fields. Due to the application of additional fertilizer and labour, the cost of boro production for the more arsenic affected plots was much higher compared to that of less affected plots in all the locations. On average the yield of MV Boro in more arsenic contaminated plots was 40-50% lower compared to that of less contaminated plots. It can be inferred from farmers' opinion that, soil fertility has been decreasing resulting in lower soil productivity. Although farmers were using higher doses of fertilizers in rice fields, yield did not increase indicating that land degradation was happening due to arsenic contamination through irrigation water. Per capita rice consumption by the households in Kachua were comparatively lower than those of Bhanga and Faridpur. Similarly, consumption of other food items (e.g., Fish, vegetable, potato, pulses etc.) was also substantially low in Kachua. Majority of households in the study locations used tube well water for drinking and other daily chores. STW water was also used for drinking purpose at varying degrees. Beside tube well water, consumption of contaminated rice was another cause of arsenicosis disease. The proportion of arsenic affected patient was higher in Kachua than that of Bhanga. Women were found to be more exposed to the arsenic contamination. The inhabitants belonging to the age of 20-45 years are comparatively more vulnerable to arsenic contamination. Majority of the victims in study areas suffered from arsenic related illness for 3 to 6 years. It is important to note that, both the proportion of arsenic victims and duration of sufferings were more prevalent in Kachua than Bhanga and Faridpur sadar. In Kachua, about 67% respondents mentioned that people with arsenicosis problem lost their working capability while in Bhanga and Faridpur sadar 62 and 45 % farms respectively, reported this. However, the eventual impacts of the problems are really worth mentioning. Nearly 55% farms in Kachua and 55% farms in Bhanga very confidently mentioned that the

aforementioned problems ultimately resulted to the reduction of their income generating activities (IGA) that led to have low household income. The other social problems arose for the arsenic affected females were really painstaking. Getting married appeared to be the main problem for women patients as reported by 34% respondents in both Kachua and Bhanga, while in Faridpur sadar 21% respondents expressed the same notion.

Policy implications

Although some farmers in the study areas are concerned about arsenic contamination and its further impact; others are still ignorant about this. The inhabitants who are aware, are already taking steps in using arsenic free water as far as possible; but availability of arsenic free water is not adequate and consequently some people are suffering from arsenicosis. The government should undertake action programs to make the people aware about the arsenic problem. Similarly, safety net programs in this regard are to be strengthened. 2) In the study areas, farmers from own perception applied more chemical fertilizers in rice fields to overcome the arsenic problem. Due to the excess use of arsenic contaminated irrigation water and chemical fertilizers, soil health might have been deteriorated. Researchers should come forward and carry out needed research to develop appropriate technology, so that, applying the technologies the soil could be saved and thus congenial environment could be created for crop production. 3) For growing Boro rice in the study areas, farmers used mainly Shallow Tube-well (STWs) for irrigation. Research findings indicated that STW water is the main source of arsenic contamination. In this regard government efforts should be strengthened in order to supply more DTWs replacing the STWs for irrigating Boro rice. 4) Research findings showed that, Boro rice production in the severely arsenic affected plots were loosing concern due to low yield, and because of this rice farmers felt discouraged in cultivating rice. This would eventually cause low rice production and create food shortage in the areas. The relevant research Institutes should take necessary step in carrying out needed research to solve the arsenic problems in crop production.

References:

Bangladesh Bureau of Statistics (BBS). 2009. Statistical Year book of Bangladesh. Statistics Division, Ministry of Planning. Government of the People’s Republic of Bangladesh. Bangladesh Economic Review (BER). 2010. Finance Division, Ministry of Finance, Government of the People’s Republic of Bangladesh, Dhaka. Ghani, M.A, M.A. Rashid and A.L. Shah. 2004. Arsenic contamination and its impact on food production and environment, Paper Presented in Third South Asia Water Forum (SA WAF-III), in Bangladesh. Bangladesh Water Partnership (BWP), LGED, Shere-e- Banglanagar, Dhaka. 13-15 July. Ghani, M.A. 2001. Studies on effect of using arsenic contaminated water on crop production and its carried over effect on food chain. Paper Presented in the Introductory Workshop Organized by the Ministry of Agriculture, GOB in Collaboration with UNDP and FAO, held at BIAM, New Eskaton Road, Dhaka, 15 May. Huq, S.M.I; Q.A.J. Ara, K. Islam, A. Jahid and R. Naidu, 2001. The possible contamination from arsenic through food chain, Proceedings of the KTH. Dhaka University Seminar, DU, Bangladesh. Islam, M.R., Jahiruddin, M and Islam, S. 2007. Arsenic linkage in the irrigation water-soil-rice plant systems. Pakistan Journal of Scientific and Industrial Research. 50(2):85-90. Jaim W.M.H and M. Hossain. 2007. Arsenic Contamination and Human Health; A Socioeconomic Study in Some Selected Areas of Bangladesh. Bangladesh Journal of Agricultural Economics Vol. XXX(2). P:81-91. Khan, M. Z. H. 2007. Managing the Arsenic Disaster in Water Supply: Risk Management, costs of Illness and Policy Choices for Bangladesh, South Asian Network for Development and Environmental Economics (SANDEE) working paper, no. 27, 2007, Kathmandu, Nepal. Khan, M.A., Islam, M.R. Panaullah, G.M., Duxbury, J.M., Jahiruddin, M. and Loeppert R.H. 2010. Accumulation of arsenic in soil and rice under wetland condition in Bangladesh. Plant and Soil. DOI:101007/s11104-010-0340-3.

Khuda, Z.R.M.M. 2001. Environmental Degradation: Challenges of the 21st Century. Environmental Society and Research Unit, Dhaka, Bangladesh.

Meharg, A.A and M. M Rahman 2003. Arsenic Contamination to Bangladesh Paddy field Soils: Implication for rice Contribution to Arsenic Consumption. Environ Sc Technol. 37(2): 224-234. Panaullah, G.M., Alam, T., Hossain, M.B., Loeppert, R.H., Lauren, J.G., Meisner, C.A.,

Ahmed, Z.U., Duxbury, J.M., 2009. Arsenic toxicity to rice (Oryza sativa L.) in

Bangladesh. Plant and Soil 317, 31–39.

Roy, Joyashree.2007. Estimating the Economic Benefits of Arsenic Removal in India: A case study from West Bengal, South Asian Network for Development and Environmental Economics (SANDEE) working paper, no. 21,2007, Kathmandu, Nepal. World Health Organization (WHO), 2001. Arsenic in Drinking Water. World Health Organization. [http://www.who.int/water- sanitation_health/dwq/arsenic3/en/]. Williams, P.N., Villada, A., Deacon, C., Raab, A., Figuerola, J., Green, A.J., Feldmann, J.,

Meharg, A.A., 2007. Greatly enhanced arsenic shoot assimilation in rice leads to elevated

grain levels compared to wheat and barley. Environmental Science & Technology 41,

6854-6859.