Salinity-induced loss and damage to farming households in coastal Bangladesh

3.1 Study areas, population and socio-demographic settings

As mentioned above, the study was conducted in four villages (Jelekhali, Kultoli, Patarakhola and Tengrakhali) in Shyamnagar Upazilla in Satkhira district. Of the 360 participants in the study, 69% were men and 31% women. Participants were relatively young to middle aged, with about 50% (both male and female) between the ages of 20 and 40; about 24% were over 50 years of age. Nearly 24% of respondents had completed primary education (Class 1-V), 26% had completed junior education (Class VI-X) and 11% had had a university education. It was found that 26% of respondents know how to write their name only, while 13% were totally illiterate. The main occupation in the study villages was agriculture farming; 81% of respondents were involved with agriculture farming as their main occupation. The most significant sources of income for the study households were: rice cultivation (98%), livestock rearing (94%) and economic trees⁵ (88%). Other income was from fishing (56%), farm labour (39%), non-agricultural activities (65%), remittances (8%) and other sources (7%).

The study indicates that 28% of the total income in the study areas comes from the rice cultivation. The study also found that the annual average income per household among the study population is US$1,335. This makes the per capita income of the study population US$303 (1,335/4.4), which is 64% lower than the national per capita income. Thus, poverty remains one of the major factors pushing local communities towards greatest vulnerability to extreme events.

3.2 History and trend of rice production and salinity

Islam (2004) states that many coastal districts, including Satkhira, are facing increased levels of salinity in agricultural fields. A recent report shows that more than 1 million hectares of arable land in Bangladesh are affected by salinity intrusion caused by slow- and rapid-onset events (SRDI, 2010). It also points out that 71% of cultivated areas in Shyamnagar Upazilla are affected by high-level salinity (above 12 dS/m). According to the BBS, the net cultivated area in Satkhira decreased by about 7% from 1996 to 2008 (BBS, 2008). It was found that immediately after cyclone Aila, total rice production in Satkhira decreased from 0.53 million tons in 2008 to 0.46 million tons in 2009 (BBS, 2009, 2010). The report also indicates that production of the principal rice crop (aman rice) in Satkhira decreased substantially, from about 0.3 million tons in 2008 to 0.2 million tons in 2010.

The Ministry of Environment and Forests (2005) predicts that a sea level rise (SLR) of 14 cm, 32 cm and 88 cm will affect Bangladesh by 2030, 2050 and 2100 respectively. The predicted SLR is expected to push saline water further inland and affect not only rice production but also all other agricultural practices in the future. A recent estimate is that the country may lose 0.2 million tons of crops to saline intrusion in a moderate climate scenario, but that this might be more than double in a severe climate scenario (Huq and Ayers, 2008).

3.3 Community perception of salinity levels in rice fields

Focus group participants and questionnaire respondents claim that salinity intrusion has increased sharply over the last decade and particularly over the past five years (see Figure 3). Current high salinity in rice fields is caused to a large extent by cyclone Aila, which hit the area in 2009. In 2012, when the fieldwork took place, more than 80% of farming households experienced high salinity in rice fields, compared to 2% and 13% ten and five years ago respectively (Figure 3). The proportion of salinity-free farmland has gone down over the past 20 years, from more than 60% to nil. Almost all saline-free and low-salinity farmland has turned into medium- or high-salinity farmland, which has a severe impact on agricultural productivity.

A comparison of survey results from the four study villages shows that although the level of salinity in rice fields has varied from village to village over the last 20 years, the trends are the same: ever-increasing salinity (see Figures 3 and 4). Tengrakhali is currently the most adversely affected village, as noted by more than 90% of respondents; salinity level has increased from 36% to over 90% in this village in the last five years. The other three villages have been affected by high salinity mainly in last five years. Jelekhali village seems to be the least affected (Figure 4).

3.4 Impacts and evidence

Farmers in the Shyamnagar study villages mainly cultivate amanrice varieties, which grow between April and August. In the past 20 years, the pattern of amanproduction in Ramjanagar Union has been quite irregular. The data show that total aman production was more than 8,500 tons between 1993 and 1997, but was less than 8,000 tons from 1998 to 2006, except in 2002 when it was 8,460 tons. In 2008, the production of aman was about 6,800 tons. Since then, it has decreased substantially by 31% and 15% in 2009 and 2010 respectively in the study union (UAO, 2012). According to UAO data, a similar pattern was observed in the other study union, Munshiganj – although in group discussions and community consultations farmers mentioned that all four study villages in both the abovementioned unions had ‘zero production’ of rice in 2009, immediately after cyclone Aila hit. More specifically, the yield of the aman crop has shifted from an average of 2.9 tons per hectare in 2008 to 0 tons per hectare in 2009 in all the study villages due to high salinity in the rice fields, as reported by local farmers. In the year after Aila (2010), some farmers in the study villages tried aman cultivation again, despite high salinity levels, but their yields were less than 1 ton per hectare. In 2011, farmers in all four villages were able to slightly increase their amanyields to an average of 1.3 ton/ha. However, the yield was less than 50% of that of 2008 (the year before cyclone Aila). It may be worth mentioning that rice production data provided by study area farmers may be more reliable. In addition, most impacted villages were selected to explore interactions between salinity intrusion and rice productions.

There were various responses regarding the changes in rice production over the last 20 years. About 76% of farmers in the study villages believed rice production has decreased over the years. More than half of the respondents (55%) mentioned that rice production had ‘decreased a lot’ over the last 20 years: the highest (82%) was in Patarakhola and the lowest (41%) in Kultoli. This significant decrease in Kultoli and Patarakhola is fairly consistent with high levels of salinity in rice fields in both villages, as mentioned above. On the other hand, at least 11% of farmers claimed that rice production increased a lot over the years: the highest (24%) was in Kultoli and the lowest (2%) in Patarakhola.

Changes in rice production cause food crises in the study villages. A majority (63%) of the study households were found to be facing a food crisis at some time of the year. Figure 6 shows that the study population faces a food crisis that varies from month to month, with peaks in September and October. These are the later months of the monsoon, just before the harvesting of aman rice. During the winter months (December to February) study households experience less food shortages. The study finds that households in Ramjannagar Union (Tengrakhali and Patarakhola villages) experience slightly more food shortages compared to households in Munshiganj Union (Jelekhali and Kultoli villages). On average, 69% of households in Ramjannagar Union face a food crisis in October, while in Munshiganj Union the figure is 67%. In group discussions and consultation meetings, it was explained that the food crisis in the study villages of Ramjannagar Union is much worse than in Munshiganj, although the survey finding does not adequately reflect that.

In group discussions and interviews, participants explained that the impacts of salinity intrusion affect men and women differently, mainly because of the level of exposure and different roles men and women play in the household. For example, one male respondent said, “I am losing rice and vegetable production due to salinity intrusion in the agricultural fields, which is directly affecting my income”, while a woman respondent reported, “I have to walk at least 1.5 km every day, especially during pre-monsoon (March-April-May) and winter (December-January-February) season, to collect water for drinking from the nearest pond/PSF (pond sand filter) in which the salinity tastes low.”

3.5 Salinity intrusion is the main cause of declining rice production in the study areas

Figure 5 showed that 75% of respondents reported declining rice production. Figure 7 indicates the main reasons given for declining rice production. Most respondents (96%) identified salinity intrusion as the main cause of declining rice production, followed by lack of rainfall (73%). Other factors mentioned included excessive rainfall in short periods of time (called ‘sky floods’ locally), pest attack, not having fertiliser at the right time, water logging, high cost of cultivation, and lack of irrigation water. The group of 23% who reported increasing rice production consisted mainly of relatively rich farmers who were able to invest in improved technology, fertilisers, improved seeds and others external inputs. Other farmers who were able to increase rice production had land with low salinity, sufficient fresh water for irrigation, and access to saline-tolerant rice varieties. All the farmers strongly believed that salinity intrusion in soil and water is the main challenge to rice farming in the study areas.

3.6 Existing adaptation measures

To deal with salinity intrusion and associated impacts, farmers in the study area have adopted a number of non-field and field-based adaptation measures. Field-based measures are mainly aimed at reducing the impact of salinity intrusion on agricultural production. Non-field-based measures aim to make livelihoods less dependent on farming. Field-based measures⁶ mainly include using saline-resistant cultivars, reducing salinity by washing rice fields, adjusting irrigation practices, and repairing or reconstructing irrigation channels. Non-field measures⁷ include temporary migration for work (either within the home district or to another nearby district), changing eating habits and lifestyles, switching to non-agricultural income-generation activities, etc.

The study found that 61% of households adopted some field-based (rice fields) adaptation measures; of these, 39% planted saline-resistant cultivars (BRRI 23, BRRI 47, BINA-8) (Figure 8). One would expect these measures to be most widespread in Tengrakhali, where salinity was highest. In reality, only 40% of households in Tengrakhali village implemented saline-resistant cultivars, whereas about 57% of households in Kultoli village had done so. It was clear from discussions with study households that awareness of the impact of salinity, access to saline-resistant cultivars, poverty and local knowledge about measures to combat salinity all play a vital role in whether adaptation measures are implemented. This might account for the fact that Jelekhali and Kultoli villages (Munshiganj Union) are more successful than Patarakhola and Tengrakhali villages (Ramjannagar Union) in terms of field-based adaptation. Although many households (27%) washed rice fields at different stages of cultivation to reduce soil salinity, planting saline-resistant cultivars remains the most common adaptation measure to increasing salinity in the study areas.

Of non-field adaptation measures, the study indicates that more than 70% households take out loans, reduce household expenses and change eating habits to cope with the impact of salinity on rice production. Table 1 shows that 55% of households try to earn more to overcome the crisis situation, while about 30% migrate to find work elsewhere. Many male members of the households studied migrate temporarily to work, either to the nearest district or to Dhaka, the capital city. Some households have migrated permanently, as confirmed during focus group discussions and interviews. In Tengrakhali village, non-field adaptation measures are very prominent compared to practices in other study villages. The study population in Tengrakhali were found to be the most likely to adopt most of the non-field adaptation measures – including taking out loans, reducing household expenditure, changing food habits and switching to non-farm activities, e.g., rickshaw/van pulling, selling dry food in the nearest market, etc (see Table 1). It should be noted that Tengrakhali is a very remote village, with higher poverty and lower education levels compared to Jelekhali and Kultoli villages. The remote villages traditionally lack infrastructure and face challenges in accessing up-to-date information, technology, etc.

3.7 Perception of the effectiveness of adaptation measures in the study areas

Field-based adaptation measures (e.g., saline-resistant cultivars, washing rice fields, raising seed beds, adjustments to irrigation, etc) were not adequate to address salinity risks. Most (64%) of the households who adopted the abovementioned measures reported that they did not adequately cope with the impacts of salinity intrusion; only 15% said the measures were effective. On the other hand, 39% of the study households did not implement any adaptive measures for rice cultivation. Most of these households cited lack of knowledge about salinity effects, lack of access to saline-resistant cultivars, poverty, etc as reasons for not implementing adaptation measures.

3.8 Estimated loss and damage caused by salinity intrusion

Current coping mechanisms are not sufficient to cope with increased levels of salinity caused particularly by the extreme weather events of 2009. Poverty, low-level resilience, and lack of alternative livelihoods, together with climate-induced hazards, are responsible for huge losses, not only for the villages studied but also for communities along the whole coast.

The study estimated amanrice production loss using 2008 as the base year. Figure 9 shows the cost of loss of rice production due to salinity intrusion caused by cyclone Aila for three consecutive years in the study villages. It appears that the total cost of loss was US$1.9 million for three years in the study villages. This estimate was based on data on aman rice production and farm gate prices for 2008 provided by the farmers in group discussions in the presence of farmers and local agriculture officers. This estimate was further scrutinised and validated in the presence of representatives of at least 20% of surveyed households from each of the study villages and local government agriculture officers. As mentioned at the beginning of Section 3, none of the study villages had had a rice crop in 2009. Therefore, the highest loss (US$0.76 m) was incurred in 2009, followed by US$0.72 m in 2010 (loss of 94%, compared to base year) and US$0.42 m in 2011.

The survey found that almost all the households in the study area incurred losses in rice production due to salinity intrusion, especially caused by cyclone Aila in 2009. The average loss per household wasUS$2,021 in 2009 for the complete loss of aman rice crop. In addition, on average, each of the sampled households had to spend US$34 for reconstructing or repairing irrigation systems, US$32 for washing rice fields to reduce salinity, US$41 for repairing seed beds and US$7 for buying drinking water in the same year. Moreover, on average, each affected source of water (e.g., ponds, tube-wells, etc) incurred costs of US$220 (on average) for cleaning or repairing.

The study also indicated that loss and damage associated with salinity intrusion in rice production affects income groups (extreme poor, poor and non-poor) differently. The income of extremely poor households from rice cultivation was affected most. In 2009, the loss incurred by the extreme poor was about 74% of their last annual household income, whereas other households lost 43-45% of their last annual income. In addition, from 2009 to 2011, all the sampled households were gradually adapting in some way to reduce loss in rice production. But the poor and extreme poor households were recovering their situation at a slower rate than non-poor households.