Study on sustainable livelihoods of fishers from Chalan beel, Bangladesh

An assessment was carried out to observe the impacts of fish sanctuary on the livelihoods from fishers of Ruhul beel and Bamonji beel under Chalan beel in Pabna district of Bangladesh. Data were collected from selected sanctuary sites with prepared and pretested questionnaire and analyzed through statistical tests like ANOVA (Analysis of Variance), regression, correlation, Chi-square were used to identify the relationships between variables and significant differences/association among them. From physical capital point of view, it was found that 91.3% of fishermen’s houses were katcha, 6.52% were semi-pacca, and only 2.17% were pucca in case of Ruhul beel and in case of Bamonji beel 90.54% of fishermen’s houses were katcha, 8.11% were semi-pacca, and only 1.35% were pacca. There was no significant variation (ANOVA; P>0.05, Chi Square test; non-significant at 0.05 level) among the residences by well-being. From human capital, average age of the household heads (HHH) was 37.95(±) 1.354 for Ruhul beel respectively, whereas for Bamonji beel the average age were 38.05(±) 1.064. The medium aged group fishermen were dominant in Ruhul beel than Bamonji beel. There were no significant difference (P>0.05) found between the age groups by well-being. From income capital it was observed highest part of income were come from fishing contributing 64.89% (653.72±128.82 US$), 43.58% (431.89±74.66 US$) for Ruhul beel and Bamonji beel respectively. As per as natural and social capital, average used land holding was 0.274±0.136 ranging 0.065 ha to 0.534 ha for Ruhul beel and 0.228±0.305 ha ranging 0.040 to 0.267 ha for Bamonji beel. The average land holdings did not vary significantly (P>0.05) between the sites.


Introduction
A livelihood comprises the capabilities, assets (including both material and social resources) and activities required for a means of living. A livelihood is sustainable when it can cope with and recover from stresses and shocks maintain or enhance its capabilities and assets, while not undermining the natural resource base. With some minor changes this is also the definition adopted by DFID (1999). Usually about all the fishers in Bangladesh are poor, illiterate, unorganized and also neglected by the society. Almost all the fishermen live below the poverty line in our country. Their access to resources is so limited because of interfere of elite, rich and musclemen. Maximum fishers have no fishing equipment of their own. Most of them are landless and don't even have homestead land but build their houses on rivers banks or in government has land or other's lands. They rarely have to access to bank or any formal credit because of lack of their resources for giving mortgage to get the loan. Consequently, they have to depend on private moneylenders who usually charge a very high rate of interest or with the condition of selling catches in a very low and fixed price. Most of them undertake fishing on a share basis with the boat-gear owner or moneylender or as daily laborer or on a monthly salary basis by which the fishers are deprived. Through this way most of them are exploited by the middleman, money lenders, aratder, dadonder, musclemen etc. because of their low social status, poor, landless and disorganized (Ali et al., 2003). Although Government has given the chance to fishers for access to resources and credit but in real they don't get the chance due to lack of organizational forum for their interests and their low socio-economic status. Very recently they are getting financial, social and technological support from NGOs and other government offices like DoF, BRDB, Co-operative office etc. Livelihood Resources-the basic material and social, tangible, and intangible assets that people use for constructing their livelihoods-are conceptualized as different types of 'capital' to stress their role as a resource base 'from which different productive streams are derived from which livelihoods are constructed' (Scoones, 1998).

Materials and Methods
The study was designed to assess the impacts of fish sanctuary on the livelihoods of the fishers in Ruhul beel (RB) and Bamonji beel (BB). The major issue was the comparative analysis of different fisher's households, adoption and adaptation of integrated farming systems and assesses the institutional context in relation to their impact on livelihoods. Both qualitative and quantitative data collection methods were used in the study in an interactive way (Sayed, 2010). Data were collected at different level from the household of the selected areas based on the livelihood approach (Carney, 1998). At the initial stage of data collection, findings of Participatory Community Appraisals (PCAs) and outcomes of the research were presented and cross checked/triangulated with different levels of fishers. The research process incorporated data relating to different aspects covering the availability of the assets; the vulnerability context; transforming structures and processes including development policies, development strategies, and other related issues like agriculture practices; common livelihood strategies; and the livelihood outcomes of local people. At the beginning of the study, secondary data were collected from different sources which included books, journals, scientific publications, dissertations and bibliographies that were relevant to the study were used as useful resource to enrich the information for understanding analysis and evaluation (Karim, 2006). Information on government policy particularly on integrated aquaculture systems, current trends of aquaculture, socio-economic, environmental and institutional factors that can be considered as important to the study were gleaned from different sources.

Phases of the study
The study was performed in the following four phases; 2.1.1. Phase I Two villages were selected from two locations of more or less similar hydrographic characteristics (sanctuary site in Ruhul beel at Patulipara village and contriol site in Bamonji beel Hatgram village) initially from Bhangura upazilla under Pabna district where the first phase of the study carried out. Before incepting the study, a meeting with key informants with the villagers was organized by the help of Upazila Fisheries Officer and then by the help of the villagers a village map was drawn.

Phase II
After finalizing the site selection questionnaire was formulated and test survey was conducted to collect data/regarding livelihood asset-base, livelihood strategies, livelihood outcomes and vulnerability context of fishing households. As 3 facilitators/enumerators were involved in data collection, variation in consistency, interpretation, unit of measures and methods of presentation of data was possible. Finally, data collection was carried out from the field level using the finalized questionnaire (Haque, 2007).

Phase III
The third step was a yearly household monitoring survey using a structured pretested questionnaire. This was conducted with the same households sampled for the survey in Step-2. The purpose of this survey carried out with fishing households was to investigate the changing patterns of livelihood systems focusing on the impacts of fish sanctuary.

Phase IV In this
Step the impact of fish sanctuary was analyzed making a comparison with the control site (BB). The main objective of the intervention was to assess the effect of change through fish sanctuary on associated livelihoods. Fishers expected to assess the effectiveness of altering fishing systems on productivity and overall livelihoods (Karim, 2006). 2.2. Data processing and analysis Initially data were entered in Microsoft Excel and exported to SPSS 20. Errors were detected and necessary corrections were made accordingly after exporting. Primary analysis (descriptive, comparative mean, graphs, Pivot tables etc.) were carried out using Microsoft Excel. Finally, quantitative and qualitative data from the study were analyzed by using SPSS/windows version 20, which offered statistical tools applied to social sciences (Field, 2005). Statistical tests like ANOVA (Analysis of Variance), regression, correlation, Chi-square were used to identify the relationships between variables and significant differences/association among them. The tools and the programs used for different type of data analysis were summarized.

Quantitative and qualitative analysis
Descriptive statistics such as frequency distribution, mean, standard deviation (SD), Standard error (SE), maximum, minimum, percentage, crosstabs and case summary were used for preliminary analyses. Other statistical analysis especially regression analysis, correlation coefficient, Chi-square and other as required to examine and understand the association of variables and its direction and magnitude. The statistical significance of results was estimated by using ANOVA test and t-test. ANOVA is a powerful statistical test where two or more independent estimates of the variance for the dependent variables are compared (Gay, 1976).

ANOVA (Analysis of Variance)
Due to heterogeneity of different sites, wellbeing, season and groups, there were many variations within the sample which were independently associated with other variables (Karim, 2006). One-way analysis of variance with post hoc analysis was used in order to identify the intra and inter group variations between different wellbeing, season (years) and locations which influenced the livelihoods, adoption, resources, production and management systems of the beels. Location, socio-economic group, farming systems and season were included as independent fixed variables.

Pearson's chi-square test
The chi-square test was used to find out relationship between two categorical variables (Snedecor and Cochran, 1989). Chi-square method was used to test whether two (or more) variables are: (1) independent or (2) homogeneous. The chi-square test for independence examined whether knowing the value of one variable helps to estimate the value of another variable. The chi-square test for homogeneity examines whether two populations have the same proportion of observations with a common characteristic. Though the formula is the same for both tests, the underlying logic and sampling procedures vary. Following formula used for this test where: ∑ Oi = an observed frequency Ei = an expected (theoretical) frequency, asserted by the null hypothesis

Correlation and regression analysis
The correlation coefficient, sometimes also called the cross-correlation coefficient, is a quantity that gives the quality of a least squares fitting to the original data. The correlation coefficient is also known as the productmoment coefficient of correlation or Pearson's correlation. The main result of a correlation is called the correlation coefficient (or "r"). It ranges from -1.0 to +1.0. The closer r is to +1 or -1, the more closely the two variables are related. Regression analysis is used to identify the linear association between independent variables used collectively to predict the dependent variables (Miles and Shevlin, 2001).

Triangulation and validation
The findings of the survey were shared and validated in each spot. At the end of the survey key findings were shared with participants and nonparticipants in each of two spot. At the end of presenting the results they were asked to comment and suggest if any findings were been contradictory in real. 3. Results

Physical capital
Physical Capital is a category of livelihood assets. It comprises the basic infrastructure and physical goods that support livelihoods. The findings on physical capitals were recorded during the present study was as follows;

Residence
All the fishers lived in own house other than only one fisherman who lived in other's house in RB in both years. But in case of BB no fisherman lives in other's house. By well-being more than 50% house owner were marginal (Table 1). There was no significant variation (ANOVA; P>0.05, Chi-Square test; non-significant at 0.05 level) in residences by well-being.

Housing status
Four types of houses were found among the fishermen in the study areas which were; 1) Katcha that were made of tin, bamboo, jute stick and trees leaves with mud floor, 2) Semi-pacca that were made of brick in one part either floor or wall but the roof was made of tin, and 3) Pucca that were made by brick eigther the wall or floor or roof. All together 91.3% of fishermen's houses were katcha, 6.52% were semi-pacca, and only 2.17% were pucca in case of RB and in case of BB 90.54% of fishermen's houses were katcha, 8.11% were semi-pacca, and only 1.35% were pacca. There was no significant variation (ANOVA; P>0.05, Chi Square test; non-significant at 0.05 level) among the residences by well-being (Table 1).

Use of electricity
Overall 60.87% households had no electricity connection in RB and in case of BB 77.03% house hold had electricity connection. The largest number of HH with electricity availability was marginal by well-being. It was found that electricity facilities were comparatively lower in RB than BB. A significant difference was found (ANOVA; P<0.001) between locations (but not between years) for electricity availability (F=39.19) and sanitation status (F=354.83) ( Table 1).

Drinking water
All of the fishermen had the access of clean and safe drinking water. Although, 7% and 11% fishermen from RB and BB respectively, did not have own tube-well but they used tube-well water from others (Table 1).

Cooking fuel
All the HH use straw, branches of tree etc. (khori) for cooking as fuel in both the beels (Table 1).

Human capital
Human Capital is a category of livelihood assets. It represents the skills, knowledge, capacity to work and good health that together enable people to pursue different livelihood strategies and achieve their livelihood outcomes.

Age group
The age of the household heads (HHH) was ranged between 20 years and 63 years with an average of 37.46(±)1.354 years and 38.46(±)1.354 years for RB-1 and RB-2, respectively whereas for BB-1 and BB-2 the average age were 38.00(±)1.064 and 39.00(±)1.064, respectively (Table 2). The second highest group was young aged group (≥30<50) which was ranged from 16.22% to 25.68% in BB-2 and BB-1, respectively. The old aged (>50 years) fishers HHH were the lowest in numbers. The medium aged group fishermen were dominant in RB than BB. There were no significant difference (P>0.05) found between the age groups by well-being.

Family size
The overall mean household size of the fishers' population surveyed for both beel were from 4.70±1.24 to 4.76±1.12 for RB and 4.53±1.10 to 4.55±1.05 for BB ranging from 2 to 7 for both the beels (Table 3). The largest family size was found in poor for both beel and the comparatively smaller sized family was found in marginal for both the beels (Figure 1). Family size was not affected by well-being, location and years (P>0.05).

Marrital status
All the household heads were married except 2% from marginal fishermen of RB-1 and 4% from better-off of BB-1 however, 1% unmarried was found from marginal of BB-1 and BB-2 ( Figure 2). There was no significant differences (ANOVA; P>0.05, F= 0.722) between marrital status of fishers among location, seasons and wellbeing.

Educational status
Education is a basic right of all population, along with food, cloth and medication. Educational qualification of the fishers (HHH) of the study spot was highest up to JSC. The majority of fishers were able to sign only ( Figure 3). Overall 1% to 7% HHHs was literate up to PSC from different well-being categories. There was no JSC passed fishermen in RB and BB in better-off and poor category, respectively. The highest numbers of HHH were able to sign only and the capabilities of sign in better-off family were increased (8% to 26%) after establishing the sanctuary. There was no significant differences in (ANOVA; P=.077, F=2.589) education by site, year and well-being. Chi-square test shows that there was a significant variation in education (χ2 = 17.14a, P<0.01 for RB-1 and RB-2) by well-being categories. However, there was no significant variation (χ2= 2.488 b and 4.365 c for BB-1 and BB-2, P>0.05) within family size, site and year (Table 4) both for BB-1 and BB-2.

School going and dropout children
The result of school going and dropout children has shown in the Figure 4 and Table 5. The total children were 124±0.18, 128±0.16, 187±0.12 and 189±0.13 in RB-1, RB-2, BB-1 and BB-2, respectively. The percent of school going children were 78.23%, 92.97%, 68.56% and 66.67% in RB-1, RB-2, BB-1 and BB-2, respectively. The school going rate was increased 22.68% in RB but decreased 1.56% in BB. Percentage of dropout children was 17.84%, 8.40%, 38.28% and 48.41% in RB-1, RB-2, BB-1 and BB-2, respectively (Table 5). On the other hand, dropout percent was decreased 41.18% in RB but increased 24.49% in BB (Figure 4). There was a significant difference in school going (ANOVA; P<0.001, F=41.13) and dropout children (P<0.001, F=11.48) in different sites but there was no significant difference in total children (P>0.05, F=1.833) among different sites (Table 5a).   Table 6. Overall 100% fisher's main occupation was fishing in case of RB in both the years whereas in case of BB fishing was the main occupation of 38% in BB-1 and 27% in BB-2. Business as secondary occupation was added 2.17% in RB. In BB fishing as main occupation was decreased 28.95%. Agriculture and daily laborer as main occupation was increased 116.67% and 71.43%, respectively in BB.

Primary occupation by location, season and well-being
The result of the main occupation is shown in the Figure 5. The 100% fishermen's main occupation was fishing for all well-being categories of RB, however in case of BB agriculture, business and daily labor was the main occupation in some of poor, marginal and better-off category. The main occupation as business was increased in poor (4% to 11%), marginal (9% to 12%) but decreased in better-off (28% to 11%). On the other hand, the main occupation agriculture of the fishers was increased in marginal (1% to 3%) and better-off (11% to 15%) categories for BB that is the trends to migrate towards agriculture from fishing as main occupation were observed in BB of marginal and better-off categories. Fishing was the main occupation of 63% fishers in poor category which was decreased to 9% after establishing the sanctuary which indicates the upliftment of wellbeing. Fishing was increased as main occupation in marginal (15% to 48%) and better-off (14% to 37%) category of RB whereas decreased in BB as shown in Figure 5. There was a highly significant differences (ANOVA; P<0.001, F=7.169) in main occupation among all well-being categories (Table 6a).

Secondary occupation by location, season and well-being
The result of the secondary occupation has shown in the Figure 6. Daily labour was the dominant secondary occupation in RB by all well-being categories; however, fishing was the dominant secondary occupation in BB by marginal and better-off categories. The secondary occupation as fishing was increased in BB by 4% to 8% in poor, 14% to 22% in marginal and 31% to 34% in better-off category ( Figure 6) whereas agriculture as secondary occupation was increased for RB in marginal well-being (4 % to 8%). In case of RB in better-off category business was increased (upto 2%). The migrating trend in fishing was observed in BB of marginal and better-off categories towards secondary occupation. There was a significant difference (ANOVA; P<0.001, F=15.04) in secondary occupations among the well-being categories (Table 6a).

Financial capital
Financial Capital is a category of livelihood assets. Within the SL framework, it is defined as the financial resources that people use to achieve their livelihood objectives.

Income
The results of total income have been presented in the Table 7. Traditionally although fishing is the major and in some cases only source of income for fishers, individuals occasionally undertaken a variety of supplementary activities (business, agriculture, livestock, labour and others sources) which constitute a substantial part of their annual income. In the present study, the mean total income from all the sources together were 887.  RB (15.2%) whereas in case of BB the second highest income was added to total income from agriculture (23.17%). The income from all the sources was increased by 13.58% in case of RB however, income decreased by 4.77% in BB. There was no differences in yearly mean income by well-being between years (P>0.0) sites (P>0.05) (Table 7a). Comparison between total income and national is shown in Table 8.

Food and Nutrition 3.4.1. Monthly frequency of HH by food intake
Total 46 HH from RB and 74 HH from BB were observed for food intake survey. The fish intake was found to increase in marginal and better-off household and to decrease in poor HH for RB. Fish intake among the wellbeing categories, location and year was varied significantly (χ2= 9.56, 16.11, 12.85, 15.20; and P<0.01, 0.001, 0.05, 0.01 for RB-1, RB-2, BB-1 and BB-2, respectively). The meat, egg and milk consumption frequency in poor households was decreased however increased in marginal and better-off category. Consumption frequency of meat (day/HH/month) was not significantly different among by well-being, locatin and year (P>0.05). The egg intake frequency in all the fishers from both beels were <10 days per month other than only one household in marginal of RB-2 (Table 9). The egg intake was significantly (P<0.001) affected in RB by well-being, loction and year however not in BB (P>0.05) (Table 9a). Milk intake was not significantly (P>0.05) affected by the variation in total income, sites and years for both the beels. Table 9. Monthly food intake (HH/month) by well-being and rank (Very low = ≤10, low = >10≤15, medium = >15≤20, high = >20).

Contribution of income from fishing in total income
In all well-being categories for BB, contributions of income from fishing were decreased. But in marginal and better-off categories the income from fishing were increased and decreased in poor (Figure 7).

.3. Per capita fish intake by well-being groups
The results of per capita fish intake have been presented in the Table 10. The per capita fish intake was ranged between 46.10 and 55.94 gm/capita/day for RB whereas it was ranged between 40.96 and 52.6 gm/capita/day for BB during the whole study period. Overall fish intake was increased 11.61% in RB and decreased 9.77% in BB. In the poor fish intake/capita/day was increased 4.11% in RB and decreased 8.19% in BB whereas in case of marginal and better-off well-being category the fish intake/capita/day was increased 4.77% and 14.40% in RB, respectively. However, fish intake/capita/day was decreased 6.59% and 11.14% in BB, respectively. There was a significant differences in per capita fish consumption between well-beings (P<0.01) and location (P<0.05) but there was no significant differences among years (Table 10a).

Natural and social capital
Natural Capital is a category of livelihood assets. It is the term used for the natural resource stocks such as trees, land, clean air, coastal resources upon which people rely for their livelihood.
In the present study, it was found that some household heads have no cultivable land and depended on a small piece of homestead land for housing, food and income. The overall average used land holding was 0.274±0.136 ranging 0.065 ha to 0.534 ha for RB and 0.228±0.305 ha ranging 0.040 to 0.267 ha for BB (Table 11). The average land holdings did not vary significantly (P>0.05) between the sites.

Discussion
The nature of residence and housing conditions indicates the social status of the people. During the study it was found that more than 50% households were marginal by well-being. All the fishers of both beels lived in own house except one from RB. There was no significant variation (P>0.05) in residences by well-being. All