Socioeconomic determinants of mangrove exploitation and management in the Pangani River Estuary, Tanzania

. Mangrove forests in Tanzania are gazetted as state forest reserves, implying that exploitation is controlled by the state. However, their continued degradation and loss is largely associated to the inadequate enforcement of management measures against uncontrolled extractive exploitation to support local livelihoods. Local management institutions are therefore advocated to enhance mangrove conservation. This study explored socioeconomic determinants of exploitation patterns and management of mangroves in the Pangani River Estuary, using two coastal communities (Bweni and Pangani Magharibi) as case study sites. Data were collected through focus group discussions, key informant interviews, household questionnaires, and field observations. Quantitative data were analyzed for both descriptive and statistical inferences, while qualitative information was subjected to content analysis. Residence time of household, household main occupation, household size, and cost of alternatives to mangroves as a source of domestic fuel were all factors positively associated with mangrove resource use. The two communities differed in their perceptions on the role of local institutions in management of mangroves. Over half of respondents (56%) in Bweni agreed that interventions by Beach Management Units (BMUs) enhanced mangrove conditions, whereas only about 16% of the respondents in Pangani Magharibi had similar perceptions. Overall, 55% of the respondents were not impressed with the performance of government institutions in implementing conservation measures for sustainable use of mangroves. Exploration and promotion of feasible alternative livelihood activities and improved stakeholders’ collaborative arrangements are recommended for sustainable exploitation and management of mangroves in the study area.


INTRODUCTION
Mangrove forests offer a range of important ecosystem services of both socioeconomic benefits and ecological merits (Tanner et al. 2019). Among the many local benefits, mangroves are traditionally exploited by coastal communities as a source of firewood, charcoal, and building materials (Mangora andShalli 2014, Nchimbi et al. 2020). Ecologically, they protect the coastline from erosion (Wagner and Sallema-Mtui 2016), store a significant amount of carbon (Alavaisha and Mangora 2016, Gullström et al. 2021, and serve as nursery ground for fish and invertebrates, i.e., supporting fisheries productivity (Nagelkerken et al. 2008, Mwandya et al. 2010, Kimirei et al. 2016. The nursery role of mangroves is supported by availability of food for young fish (i.e., directly from mangrove leaves and detritus or indirect as periphyton and phytoplankton in estuaries) and the structural complexity of mangrove sub-habitats (e.g., roots and tree trunks) providing shelter for fish, crustaceans, and mollusks from predation (Gajdzik et al. 2014).
Despite their economic and ecological values, the balance between mangrove resource exploitation and conservation is challenging in many regions where they occur (Friess et al. 2019). Globally, 336,300 ha of the mangrove area were lost between 2000 and 2016, with the Southeast Asian region suffering from about 80% of those losses that were associated with human activities (Goldberg et al. 2020). Africa has lost about 55,000 ha of mangroves over a 20-year period, from 1996 to 2016 (Spalding and Leal 2021), and nearly 3000 ha per year of mangroves have been lost in the Western Indian Ocean (WIO) region (Tanzania, Kenya, Mozambique, Madagascar) during the last 25 years (Bosire et al. 2016, as cited in Maina et al. 2021). The widespread mangrove loss is likely to continue and spread in many other nations if mangroves are not sustainably managed and monitored (UNEP 2014). Overexploitation and conversions of mangrove areas to farmland, salt pans, and aquaculture, as well as poor enforcement of management prescriptions are cited as the main human-induced drivers of mangrove degradation, especially in developing countries (Mangora 2011, Suman 2019. With regard to these anthropogenic drivers of mangrove degradation, various conservation measures, e.g., restoration programs and participatory management approaches, are advocated at international, national, and local levels to reverse the loss of mangroves, and to promote sustainable utilization and conservation (Romañach et al. 2018). However, the degree of reliance on mangroves and their associated ecosystem services is context-specific and depends on the socioeconomic characteristics of resource users (Brander et al. 2012), management policies, and initiatives instituted to conserve mangroves (Badola et al. 2012, Romañach et al. 2018. For example, Handavu et al. (2019) revealed that socioeconomic variables such as household income, household size, and gender, which may vary from one community to another, are among the most critical factors driving the use of various forest products. At the same time, lack of transparency and inadequate attention to community voices in conservation of natural resources destabilize management strategies (Bennett and Dearden 2014). Appropriate engagement shapes communities to recognize and appreciate the role played by mangroves in supporting local livelihoods, identify impacts of mangrove overuse and the rationale for rehabilitation of the degraded areas to achieve sustainability (Arumugam et al. 2021).
In Tanzania, mangroves are legally gazetted as state forest reserves, and occur along almost the entire coastline (Semesi 1992, Wang et al. 2003, Mangora et al. 2016. The largest fractions are found in estuarine-and river mouths spreading across the country from the northern to southern districts (Semesi 1992, Wang et al. 2003, Mangora et al. 2016. Coastal communities in the different parts of Tanzania have depended on mangrove forests for generations to support their livelihoods and well-being (Mainoya et al. 1986, Semesi 1992, 1998. In recognizing their values and importance, the government in collaboration with development partners developed the national mangrove management plan in 1991 (Semesi 1992), and implemented it through the Mangrove Management Project (MMP), which began in 1994 to accentuate sustainability (Wang et al. 2003) before it phased out in 2004 (Cooksey et al. 2006). Capitalizing on this management plan, different stakeholders attempted to design a number of other collaborative arrangements and restoration initiatives in different parts of the country to enhance mangrove management and conservation (Mangora 2011). Joint forest management (JFM) is one of the participatory forest management (PFM) efforts developed to promote sustainable utilization of forest resources, including mangroves following its inclusion in the National Forest Policy of Tanzania of 1998, and the Forest Act 2002 (Beymer-Farris and Bassett 2012, Treue et al. 2014, Mwansasu 2016. However, in practice JFM has not yet been fully executed to prevent mangroves from being degraded because of a range of factors. First, under JFM, the government owns the forest, and for state forest reserves, this means that all decisions about, e.g., human access and use rights are made by the government (Beymer-Farris and Bassett 2012, Mwansasu 2016). Second, it is difficult to achieve effective collaborative management for mangrove forests because of the absence of specific formal agreements on community rights to use mangrove resources and the lack of shared management responsibilities (Mshale et al. 2017). These issues have exacerbated conflicts between the government (forest owners) and local communities (users of resources; Ngomela 2007, Mangora 2011. Moreover, implementation of a management plan faces various challenges that jeopardize fully secured sustainable use of mangrove resources, mainly because of increasing pressures from human exploitation, inadequate engagement of local communities, and inconsistent management interventions and enforcement (Mangora 2011, Mshale et al. 2017. Furthermore, these management challenges exhibit significant spatial and temporal variations that are much influenced by the prevailing socioeconomic conditions of the communities around mangroves (Nyangoko et al. 2021). Nchimbi et al. (2020) also demonstrated that community interests in conservation can remain high if their opinions are appropriately included in the management planning and if there are alternative livelihood options that would meet the needs of local users, while otherwise the loss of this resource is likely to continue.
Mangroves are estimated to cover an area of 158,100 ha in mainland Tanzania (MNRT 2015), and nearly 19,000 ha of mangrove area were lost between 1990 and 2015 (FAO 2015, as cited in Njana 2020). Generally, Tanzania's mangroves are disappearing at a rate of around 0.7% per year (FAO 2007), with some local areas being more degraded than others (Wang et al. 2003, Alavaisha andMangora 2016). For example, the Rufiji Delta, which holds the largest mangrove area (45,519 ha) in the country, loses about 378 ha of its mangroves each year, corresponding to loss rate of 0.5% annually (Monga et al. 2018). Mangrove loss is also exemplified by the Wami River Estuary, where intensive harvesting of these resources for charcoal production and fuel-wood needs prior to its gazettement under management authority of Saadani National Park (protected area), has threatened not only ecosystem integrity but also local fisheries (McNally et al. 2011). This could match the case of the Pangani River Estuary, where the contribution of mangrove forests to local livelihoods and mangrove degradation has been reported over the last few decades (Turpie et al. 2005, Ngomela 2007), but the link between socioeconomic variables of dependent communities and mangrove resources extraction, as well as how local people perceive strategies for managing mangrove ecosystems is not well captured. In an effort to further contribute to the design of appropriate mangrove management interventions, which would promote common ground for collaborative arrangements with local communities, the present study explored the influence of socioeconomic drivers on mangrove resource use, and community perceptions on effectiveness of mangrove management of the Pangani River Estuary. Specifically, the study was set to answer the following questions: (1) What are the socioeconomic drivers that determine the exploitation pattern of mangrove resources of the Pangani River Estuary? (2) What are the community perceptions on effectiveness and enforcement of mangrove forest management interventions for the Pangani River Estuary?

Study area
This study was conducted in two peri-urban communities; Bweni and Pangani Magharibi in Pangani Township of the Tanga region in northern Tanzania (05°24′00″-05°25′36″ S; 38°58′47″-38°59′ 60″ E (Fig. 1). The two communities are situated close to mangroves of the Pangani Estuary, which represents the terminal end of the Pangani River (Fig. 1). The Bweni site is more connected to rural settings and situated on the southern bank of the Pangani River. Pangani Magharibi is a more urbanized area on the northern bank of the Pangani River. Regardless of any differences in the main source of livelihoods between the two sites, these communities are generally characterized by livelihood occupations, such as subsistence rain dependent farming, artisanal fisheries, small businesses (e.g., food vending and kiosks), small scale livestock keeping and utilization of both mangrove and inland terrestrial forest resources.
The area covered by mangroves in the Pangani Estuary is about 753 ha (Turpie et al. 2005). The actual size of the degraded area is unknown (Ngomela 2007), but cutting of mangroves for timber has affected these ecosystems (Turpie et al. 2005). Thus, the reimposed ban on mangrove harvesting in 2016 across the country may have reduced reliance on mangroves and improved their conditions, but this needs to be further verified. Eight mangrove species occur in the estuary, which is dominated by Avicennia marina, Ceriops tagal, and Rhizophora mucronata. Other mangroves species found in the area are Bruguiera gymnorrhiza, Heritiera littoralis, Lumnitzera racemosa, Xylocarpus granatum, and Sonneratia alba (Lamtane et al. 2014). The area experiences hot and humid tropical conditions with average temperatures of https://www.ecologyandsociety.org/vol27/iss2/art32/ 20.5 °C and 30 °C during coolest and hottest months, respectively, and receives an average total annual rainfall of about 1240 ± 333 mm (Njana 2015). According to the national population census of 2012, Bweni and Pangani Magharibi had a total population of 1263 and 6262 people, which increased from 1187 and 4929 people recorded in 2002, respectively (URT 2013). Two local natural resource management institutions, the Beach Management Unit (BMU) and the Village Natural Resource Committee (VNRC), exist in the study area, but their functions differ. The BMU exists in each site and primarily focuses on management of fisheries resources and their associated habitats, including mangroves. Their responsibilities include patrolling, enforcing rules and regulations, and raising awareness about fisheries management and associated aquatic environments. The two sites share the same VNRC, which focused on protecting both inland terrestrial forests and mangrove resources, with the mandate to raise awareness and implement by-laws pertaining to overall environmental protection and pollution control throughout the studied communities.

Research design and data collection
The field work was carried out from late April to early May 2016, at both community and household levels, to explore the influence of socioeconomic drivers on mangrove resource use and community perceptions on mangrove forest management interventions in the communities of Bweni and Pangani Magharibi. Mixed social research methods, including focus group discussions (FGDs), key informant interviews (KIIs), and household questionnaire surveys (HQSs; Katikiro 2016) were used to gather data. Direct field observations were conducted to enrich the data. Prior to the actual data collection, a reconnaissance survey was carried out, during which consultative meetings were performed with community leaders, influential elders, local natural resources management committee leaders (BMUs and VNRC), and a few households, to introduce the study, seek consent, and collect important preliminary information that facilitated the design of actual survey tools.

Focus group discussions and key informant interviews
One FGD was conducted at the community level in each study site, which included a mixed group of mangrove resource users Ecology and Society 27(2): 32 https://www.ecologyandsociety.org/vol27/iss2/art32/ (fishers, mangrove cutters, food vendors, and farmers). Each group discussion involved a small active group of eight participants (two individuals from each occupation) of mixed gender and age for the efficient discussions, as suggested by Hennink (2014). Discussions were moderated using prepared checklists adopted from Ansong and Røskaft (2011) and Roy et al. (2013), and modified to meet the purposes of the study. During the discussions, participants identified all important socioeconomic activities in the study area, and local uses of mangrove resources. Participants were also asked to indicate how they perceived the conditions of mangroves over time and reasons for changes (based on time line analysis). In addition, a wealth ranking exercise (Endamana et al. 2016) was conducted to understand the relationship between the wealth status of community members and patterns of exploitation of mangrove resources. Accordingly, three wealth groups (poor, middle, and rich) were identified by participants to reflect households' assets in their community based on wealth decisive factors (type of houses, source of household energy, farm size, monthly income, number of meals taken per day, and livestock keeping). The participants were also asked to give opinions on (a) the performance of natural resource management institutions, (b) conservation programs and projects involved in mangrove management in the study area, (c) how they perceive those interventions, and (d) what should be done to conserve mangroves, while improving livelihood of dependent communities. The information generated from the FGDs were crosschecked by conducting in-depth interviews with 14 key informants, including (i) the Pangani District Forest Officer (DFO), (ii) the District Forest Manager (DFM) for Tanzania Forest Services (TFS), (iii) the village chairperson in each site, (iv) two influential people in each site, (v) two BMU leaders in each site, and (vi) two VNRC leaders in the study area (Appendix 1).
A key informant was regarded as an individual who had deep knowledge about the study site, socioeconomic patterns, and historical perspectives of management of natural resources (Orchard et al. 2016). The DFO and DFM were, however, selected as key informants based on their expertise and duties in managing mangrove forests regardless of how long they had lived in the study area.

Household survey
A semi-structured questionnaire (Appendix 2) was administered to 30 households randomly selected in each site. Household heads were targeted because they are responsible for decision making and obliged to provide basic needs in their families (Boafo et al. 2016). In their absence, any other adult representative member conversant with household deeds was asked to speak for the household. The questionnaire used was adopted from the study by Adhikari et al. (2004), and modified based on the information obtained from the FGDs and KIIs. The information collected included socioeconomic and demographic characteristics, resources collected from mangroves to support livelihoods, and perceptions on local and state management interventions. To determine whether households are dependent on mangrove resources for their livelihoods or not, respondents were asked to mention at least one key resource that they extract from mangroves to support their subsidence or commercial needs. Based on the description by Newton et al. (2016), those respondents who revealed that they extract resources from the mangrove forest in some way were classified as mangrove dependent, and vice versa as mangrove non-dependent. The perceptions on effectiveness and enforcement of management interventions were assessed based on predefined mangrove management and conservation statements explored at the community level, where respondents were asked to depict a statement and rank their responses as either strongly disagree, disagree, neutral, agree, or strongly agree per given statement.

Field observations
With the assistance from two local residents in each study site, a physical visit to the mangrove forests was carried out to observe mangrove conditions and resources collected from the forests, followed by observations on different socioeconomic activities that took place in the study area to verify the gathered information from the FGDs, KIIs, and HQSs.

Data analysis
Information collected from the household survey was analyzed statistically using SPSS v. 16. Binary logistic regression was used to deduce relationships between socioeconomic drivers and exploitation pattern of mangrove resources in the study area. Data from the studied sites (Bweni and Pangani Magharibi) were combined to increase the strength of associations between predictors and response variables that fit the logit model. The model was chosen because it can simultaneously analyze impacts of both continuous and categorical explanatory variables to reduce bias for their influence (Sperandei 2014). The variables included in this model are summarized in Table 1. Descriptive statistics (percentage distributions) were used to analyze respondents' perceptions on effectiveness and enforcement of mangrove forest management interventions between the two study sites. Content analysis (Yates and Leggett 2016) was used to analyze qualitative information collected from the FGDs and KIIs. With regards to this, all information noted were grouped together followed by establishment of major themes, which were further subdivided into sub-themes, and finally to the smallest meaningful units that represented issues described by respondents in relation to mangrove exploitation and management.

Socioeconomic and demographic profiles of the households
The socioeconomic and demographic characteristics of respondents in the studied communities are summarized in Table  2. Almost 67% of the respondents were male and about 47% were in the age group of 36 to 55 years. Over 70% of the respondents were married. The average household size was five persons. In terms of education, 53% of the respondents in Bweni and 40% in Pangani Magharibi had completed the primary level of education. Fishing was the main livelihood occupation in Bweni, accounting for ~37% of the responses, while in Pangani Magharibi farming (33%) was the main livelihood occupation. The average household monthly income was estimated to be 187,500 Tanzanian shillings. In Bweni, 50% of the respondents reported to have stayed in the area for over 15 years, while only 33% had stayed for such a long period in Pangani Magharibi.

Socioeconomic drivers of mangrove resource exploitation
The common uses of different mangrove resources reported by respondents during the FGDs in both studied communities are summarized in Table 3. Building materials and firewood were the most commonly extracted products in the area. The dependence pattern on mangrove resources by occupation in the study area are summarized in Figure 2. In Bweni, 70% of the households were mangrove dependent (i.e., they extracted mangrove products for their livelihoods only), whereas 60% in Pangani Magharibi were mangrove non-dependent (i.e., they did not extract mangrove products for their livelihoods). In both sites, fishers and farmers were most dependent on mangrove resources compared to people with other occupations. Based on the binary logistic regression model testing the influence of selected socioeconomic factors on mangrove resource extraction (Table 4), only four predictors were significantly associated with the extraction of mangrove resources for livelihoods in the study area, including residence time of household, household main occupation, household size, and cost of alternative resources to substitute use of mangrove wood as cooking fuel. Because the coefficients of these independent variables were all positive, a unit increase in these factors would increase the number of people who were engaged in mangrove resource extraction. An analysis of households' wealth status based on wealth decisive factors (Table  5) indicated that poor households often exert more pressure on mangrove resources than rich households. Table 6 summarizes legal, institutional, and policy instruments related to mangrove management in the study area. Thirteen institutions and conservation projects/programs responsible for management of mangroves resources in the study area were identified during the FGDs and KIIs. Some of these conservation and management programs have been phased out, while others are still in operation. The past donor funded projects/programs like the Mangrove Management Project (MMP) and Tanga Coastal Zone Conservation and Development Program (TCZCDP) were reported to be effective during their implementation phases. This implies that the efforts from these projects/programs on, e.g., awareness raising, capacity building, and participatory monitoring have led to improved management and conservation of mangroves, which is a result of the collaborative work with local communities, who got empowered to take control of their own resources, which in turn has reduced

Fig. 2.
Household dependence patterns on mangrove resources by occupation in the study area (n = 30 per site). Extraction of poles, firewood, fish, and honey were mentioned as preferred resources by mangrove dependent occupants.
mangrove degradation. However, it was observed that many people in the area were unaware about the existence of a recent climate change adaptation project entitled "Developing core capacity to address adaptation to climate change in productive coastal zones of Tanzania," which implies little sensitization, participation, and engagement of the communities. Historical narratives and community perceptions on the conditions of mangroves and drivers of changes over time (Table 7) indicate that severe mangrove degradation occurred between 1965 and 1993, while from 1994 to 2004, the degree of such degradation was reduced to a moderate level, and from 2005 to present, there has been a resurgent degradation of mangrove forests. Generally, the two studied communities had different perspectives on the effectiveness of mangrove forest management interventions for sustainable mangrove utilization and associated conditions (Fig.  3). The majority of the respondents in Bweni (57%) either agreed or strongly agreed with the perception that BMU interventions have led to improved conditions for mangroves, whereas onlỹ 17% of respondents in Pangani Magharibi had similar opinion.
Ecology and Society 27(2): 32 https://www.ecologyandsociety.org/vol27/iss2/art32/ Overall, 55% of the respondents (i.e., agreed or strongly agreed responses in both study sites) indicated that state management agencies (the DFO and TFS) are weak in their enforcement of regulations for conservation and sustainable use of mangroves.

Socioeconomic drivers influencing on mangrove resource use
Dependence on forest resources, especially by rural and suburban communities, is shaped by socioeconomic factors operating at the household and community levels (Handavu et al. 2019). The results of this study demonstrate this viewpoint, albeit the level of influence of these factors varied, and were associated to different reasons and implications. The observed similarity between gender of household heads in terms of mangrove resource use implies that many people regardless of their gender extract resources from mangroves. However, some respondents during the FGDs and KIIs mentioned that men tend to exploit more resources (e.g., harvesting poles for making boats and fish traps throughout the year) than women (e.g., occasional collection of firewood and crabs) because men are traditionally responsible for household welfare and are obliged to ensure security of household basic needs. As such, men were mostly associated with illegal exploitation of mangroves on the grounds of sustaining household welfare. Examples of these illegal activities are cutting of mangrove trees for charcoal making and digging in mangrove areas for fishing bait. Our results conform to the findings of Feka et al. (2011) and Eneji et al. (2015), who also reported that exploitation of forest resources is carried out by both men and women, and that men tend to extort a multitude of resources from forests, while women only exploit a few resources for subsistence use. The age of household heads also had no significant influence on mangrove extraction, as both young and old people indicated that irrespective of their age, they would extract resources from mangroves as long as they had the physical ability to visit the mangrove forest. Our findings are similar with the study by Garekae et al. (2017), who also found that the elderly tend to reduce forest extraction activities because of a decline in physical strength. Household size was positively related with mangrove resource use. This implies that large households, especially in rural and suburban areas, tend to exploit more natural resources than small households because of an increased demand by the household members. As poverty prevails in poor coastal societies, large families may not be able to afford basic necessities and hence exert pressure on the nearest resources for products like firewood, mangrove crabs, prawns, and medicines. This finding is in agreement with the observation that households with large families, especially those with limited income opportunities, are highly reliant on forest resources to meet their basic needs (Handavu et al. 2019). https://www.ecologyandsociety.org/vol27/iss2/art32/ The level of education did not significantly influence the mangrove exploitation, which matches the results of the FGDs and KIIs that many people regardless of their educational status extract resources from mangroves. This was mainly attributed to limited employment opportunities in the studied communities, where a completion of secondary education alone was no guarantee of getting formal government or private employment. This in turn leads to less reliable sources of income and increased poverty, and thus a similar pattern of resource use among those with informal, primary, and secondary education to provide for subsistence needs and income generation. Contrary, Inoni (2009) reported that higher levels of education make harvesting of forests unattractive to the local elites. This contradiction stems from the fact that, unlike the Pangani community, the studied society by Inoni was characterized by more livelihood options like largescale dairy farming, which provided employment for rural dwellers who had some form of formal education, and thus were less dependent on mangrove resources as an income source.
According to Giliba et al. (2011), an increase in the number of years of residence around a forest in rural areas significantly influences the use of forest products, which is in agreement with the findings of this study. People with long residence time know their environment better than people with shorter residence time, and thus, they are more likely to go into forests to extract mangrove resources to secure their livelihoods. In addition, because the study area is dominated by fishers (Table 2), who have stayed there for a long time (over 15 years), they are often expected to be more reliant on mangrove resources compared to other types of occupations. This is due to the nature of their livelihood activities, as many fishermen in the area own boats and canoes, which are made of wood from mangrove trees. Fishers could also use their fishing boats to harvest and carry building poles, firewood, and other mangrove products, which is difficult for other occupants who do not own a boat or canoe. With regards to this, household main occupation was found to be positively related to mangrove resource use. This implies that an increase in number of fishers in the studied communities would increase the number of people who are engaged in mangrove resource extraction, which is consistent with the findings of Adhikari et al. (2010), who indicated that fishing communities rely on mangroves to collect fisheries products. Some respondents in the study area mentioned that the use of small fiberglass boats would replace wooden boats made from mangrove trees, and would facilitate offshore fishing, and thus reduce the pressure on mangroves. Because some of the respondents were engaged in mud crab fishing, training and development of crab fattening, using small bamboo cages in mangrove areas, could supplement the income for marginalized fishers without degrading the mangroves.
The costs of alternative energy sources for cooking significantly influenced the extraction of mangroves for firewood. This was associated with increased costs of purchasing gas, kerosene, or electric stoves in the studied communities. For instance, the price for refilling an Oryx gas cylinder of 5 kg as an alternative energy Ecology and Society 27(2): 32 https://www.ecologyandsociety.org/vol27/iss2/art32/ Table 3. Common local uses of mangrove resources in the study area, based on focus group discussions and key informant interviews.

Uses
Community narrative on the uses Firewood Species C. tagal and R. mucronata are preferred for firewood because of their good burning quality and they produce little smoke. The A. marina is also used as an alternative because Ceriops and Rhizophora species have become scarce in the study sites. Dead and/or live trees are collected or cut for firewood use by households in both sites. However, the demand of fuelwood was higher in Bweni than Pangani Magharibi. Primarily women collected firewood for household use. Few men collect firewood to sell. The commercial use of these resources required authorization from village authorities and then applying for a permit from the District Forest Manager for Tanzania Forest Services.
There is a danger of forest degradation in the area because firewood collection is carried out almost daily. Nowadays, firewood collection for selling is banned by the government. Despite the ban, signs of illegal cutting are still evident.

Building
The poles from C. tagal and B. gymnorhiza species are preferred for house building, making local furniture, supporting peg in fences and vegetable gardens because the wood is strong, resistant to insects, and lasts longer. Timber from S. alba is used for making furniture and boats. Poles are commonly harvested products in the area, and primarily collected by men. Poles are demanded for both commercial and domestic uses. However, compared to firewood, demand for poles for household use is not on a daily basis. Recently, permit or licences for pole harvesting has been suspended to reduce mangrove degradation.

Traditional medicines
The species X. granatum and A. marina are exploited for traditional health care by a few members of the community, but presently their use is very low compared to the past. Xylocarpus granatum leaves are boiled for treatment of stomach-ache and skin rashes. Fruits of A. marina are grinded for treatment of wounds.

Charcoal making
The species R. mucronata and C. tagal are harvested for charcoal making. Charcoal from mangroves produces high calorific value and less ash. Although charcoal making is forbidden, it is done in secret by a few men.
Boat making The species H. littoralis and S. alba are harvested for making fishing boats/canoes. Each species has its own role but sometime their roles are substituted/interchanged depending on availability of materials. Timber/poles from S. alba are preferred to make the boat hull (main body) and boat keel (bottom) because of their large size and good bending capability. H. littoralis poles are used to make boat boom (pole for direction) because of their strength and hardness.

Fisheries
Mangrove areas serve as nursery, feeding, and breeding sites for a variety of fish, and thus support fisheries. Crab collections, shrimps, and finfish fishing were the most cited mangrove-fishing activities, and were dominated by men.

Honey production
Traditional beekeepers set their bee hives in the mangrove forest to be colonized by wild bees for production of honey. This activity is dominated by men.
source for cooking, which could replace the use of mangroves as wood fuel, was 25,000 Tanzanian shillings (TZS; in April 2016) in the study sites. This is too expensive when compared with an average monthly income (187,500 TZS in 2016) of households in the studied communities. Under such circumstances, poor communities living around mangrove areas, who are in the majority, will continue to rely on forest resources as a cheap source of fuel. This finding concurs with the results of Kilahama (2008) and Doggart et al. (2020), who reported that a small increase in the kerosene price and electricity tariffs adds a burden to users, and thus exerts a demand for cheap charcoal from forests and woodlands. Based on the discussions in the FGDs and KIIs, community members in the study area perceived that improving the bad conditions of the road between Tanga and Pangani with tarmac would be an option that not only would reduce the costs for transporting affordable alternative energy sources, but could also lead to multiple benefits such as investments in the tourism industry and transportation of agricultural and fisheries products to the market areas (i.e., Tanga and Moshi cities). Promoting the planting of fast-growing genetically improved coconut palms was also mentioned by participants as an important initiative to provide husks and trees that tentatively could reduce the demand for firewood and building materials. Other interventions mentioned by key informants included introduction of biogas technology and low-cost solar stoves to substitute traditional biomass-based fuels from mangrove trees. To accomplish this, the government should engage in an open dialogue with the communities and provide them with the knowledge and skills required for biogas installation and usage as well as on how to cook with a solar stove.
In this study, mangrove exploitation was not significantly influenced by household monthly income, implying that many people regardless of their income (low-income, middle, and highincome), partly relied on mangrove resources (i.e., honey, fish, crabs, and poles) as a source of income to meet their livelihoods. This was associated with limited income-generating opportunities and a low level of education among the majority of respondents, which limit their ability to pursue formal employment in the area and neighboring cities (e.g., Tanga), a finding that is in conformity with what has been reported elsewhere (Okello et al. 2019). The promotion of alternative livelihood activities is seen by local communities in the study area as a viable way to improve rural livelihoods and reduce dependence on the forests. With regard to this, participants of the FGDs in the studied communities mentioned that with a growing demand for meat products and eggs in Pangani Township and Tanga City, the provision of lowinterest loans by the government would aid in the establishment of poultry farming (broilers and layers), which would provide food for household consumption, while also contributing to household income. There was also an extensive interest in modern beekeeping among community members, where many interested groups showed up and sought technical support from the Pangani District.
Ecology and Society 27(2): 32 https://www.ecologyandsociety.org/vol27/iss2/art32/ of the coefficient (β) indicates that a unit increase in a specific variable could increase the number of people involved in mangrove resource extraction by a factor of the observed odds ratio (Exp β). Mangrove extraction (dependent variable) was assessed by yes or no responses, where respondents were asked if they had extracted any resources from mangroves for their livelihoods or not.

Perceptions of mangrove forest management interventions
It is becoming increasingly clear that for collaborative management to be successful, it must be people-centered and include social-economic and political issues of communities (Saunders et al. 2010, Datta et al. 2012. Such inclusion allows communities to take part in management (Wells et al. 2010) and reduces the conservation costs in the face of limited human and financial resources (Mshale et al. 2017). In this study, we assessed the effectiveness and enforcement of mangrove forest management interventions in the Pangani Estuary. Institutionally, the Fisheries Act of 2003 provided frameworks for Beach Management Units (BMUs) to monitor fisheries activities and related ecosystems such as mangroves. In Bweni, community perceptions on the improved status of mangroves as a result of the work by the BMU were positive, indicating the relative success of the BMU interventions in meeting its objectives. Results of the FGDs and KIIs revealed that awareness raising about the importance of mangroves, patrolling and mangrove planting were the main strategies used by the BMU for conservation of mangrove-associated resources. The success of this work was partly due to good collaboration between BMU members and the local community, whereby people become willing to report any illegal activity to the BMUs. A similar result was reported by Okoth (2015), who found that patrolling by BMUs had reduced cutting of mangroves and contributed to some improvement of the mangrove condition in Bagamoyo on the coast of Tanzania.
Although BMUs are not a part of the Joint Forest Management (JFM) from a forest policy standpoint, the observations from Bweni still demonstrate that an interest in improving mangrove forest conservation for collaborative management can be achieved only if the local community and different institutions are actively involved in management processes. Mikoko Pamoja in Gazi Bay, Kenya, is a community-based project, where forests are comanaged in partnership with government and non-government organizations, and exemplifies the importance of investing in collaborative mangrove management (Wylie et al. 2016). Sales of carbon-credits from this multibeneficial, community-led initiative has reduced mangrove degradation and improved mangrove conservation via rehabilitation, monitoring, and surveillance, and also created employment opportunities for local residents, and contributed to the planting of fast-growing inland trees to replace wood harvested from mangroves for building (Herr et al. 2019). Furthermore, The Tanga Coastal Zone Conservation and Development Program (TCZCDP) in Tanzania (run from 1994 to 2007) also provided a prime case study of collaborative management that actively engaged communities in natural resource management. The program empowered coastal communities in the Tanga region to control their own resources, resulting in increased mangrove cover and reduced degradation during its operation (Wells et al. 2010). This success was accomplished through good consultation with local communities, capacity building, patrols, and the establishment of nonconsumptive uses of forest products such as beekeeping and seaweed farming. Moreover, the experience drawn from the Kinondoni Coastal Zone Management Program in Dar es Salaam, which was initiated in 2001 with the support from the Swedish government, also portrayed a well-established comanagement approach, where communities secured the coastal areas and its resources. The program provided education, seminars, and established beekeeping and vegetable gardening as alternative livelihoods, which enabled mangrove reliant villagers to reduce a lot of mangrove cutting (Himberg 2016).
The participants of the FGDs also mentioned that in the past, the Village Natural Resources Committee (VNRC) was successful in providing awareness and enforcing its bylaws (Table 6). This success was due to the existence of donor funded initiatives (i.e., TCZCDP and the Marine and Coastal Environment Management Project, which provided patrolling tools and financial motivation to the members of the VNRC. Phasing out of these donor-funded programs/projects jeopardized the capability of the VNRC to achieve its purpose, leaving it inactive and thus likely contributing to resurgent degradation of terrestrial and mangrove-based resources. This calls for a need for long-term government commitments, with the priority in participatory forest management (PFM) by people, rather than relying on shortterm international donor agencies. The majority of respondents (~47%) in Bweni were also positive about the capability of the village council to collaborate with some members of local communities to resolve disputes associated with mangrove utilization. The good relations were built on regular consultative meetings between the village council and the community, where local issues were discussed. The situation was different in Pangani Magharibi because of poor cooperation between the BMU members and both the village council and the local communities. This poor relationship was attributed to political affiliations, whereby most of the respondents claimed that a majority of the BMU members were inclined to political fractions rather than technical conservation, prompting continued mismanagement and irresponsible use of mangroves. The mismanagement of mangroves due to political influence as revealed in this study seems to be a national challenge; for instance, in the Rufiji Delta, politicians, particularly at the local levels, have issued statements advocating for overharvesting of mangroves for political gain (Mshale et al. 2017).
Despite the improved status of mangroves close to Bweni from 2005 to 2016, as reported by some community members during the FGDs, most of the respondents in both study sites admitted that formulated bylaws and their enforcement had not been fully executed. This was due to the existence of social ties and lack of working facilities. Social affection often hinders punishments related to illegal mangrove exploitation because BMU members fear creating a bad social relationship with members of their communities. The present findings concur with Kanyange et al. (2014), who also reported that close friendship between BMU members and culprits of resource degradation are among the issues leading to poor resource management along the coastline of Kenya and Tanzania. Another challenge mentioned by local elders in the studied communities includes friction between BMUs and VNRCs that represent an institutional conflict that may render less enforcement of management measures. Conflicts between the BMU, VNRC, and communities are also driven by the fact that not all local residents could fully understand all bylaws implemented by these local institutions, and thus some community members regard BMUs and VNRC as "puppet committees" that are working for the interest of the government to hinder the way local people benefit from mangroves.
We speculate that mismanagement of mangroves in the study area and across the country could also be associated with marginalization of women in local management institutions, with women often lacking decision power and hence their perceptions and interests on conservation are sometimes ignored or excepted. Active involvement of both genders, particularly women, in decision making has the potential to improve mangrove resource conservation because women have more conservationist preferences than men (Barrero-Amórtegui and Maldonado 2021). Although the context of the Tanzania Forest Policy framework advocates PFM for managing forest resources, evidence from the studied communities and across the country shows that existing policies and laws (e.g. National Forest Policy of 1998, Forest Act No. 14 of 2002, Fisheries policy of 1997, Fisheries Act No. 22 of 2003, and Fisheries Regulations of 2009 that emphasize co-management to ensure sustainable use of mangrove resources have failed to prevent mangroves from being degraded. This is because the PFM in Tanzania has not been exercised as intended, and the mechanisms for shared management responsibilities and partnerships between government agencies and local communities are not explicitly stipulated in the legislation. Moreover, for state forest reserves, all decisions including human access and use rights are made by the government rather than the people who live within the forest areas, which in turn creates a sense of dis-ownership of the resources (Mangora 2011) and impedes cooperation between government agencies and local communities (Brenner et al. 2018). Inadequate devolution of management rights and decision making to communities demonstrates the existence of a top-down management approach, in which local people have little control https://www.ecologyandsociety.org/vol27/iss2/art32/ Funded by NORAD and aimed at promoting sustainable utilization and conservation of mangroves through community sensitization and awareness to build capacity in management and restoration of degraded areas of mangroves. The project developed the National Mangrove Management Plan, which is still implemented, albeit outdated. The MMP was part of large program called Management of Natural Resources Program (MNRP), which aimed at improving livelihoods of rural communities through sustainable natural resource management. The MMP was implemented in three phases: phase I focused on capacity building (1994)(1995)(1996)(1997)(1998), phase II emphasized collaboration among stakeholders (1998)(1999)(2000)(2001)(2002), and phase III focused on improvement of livelihoods to communities living adjacent to forest (2002)(2003)(2004). It lessened mangrove degradation, but after it was phased out, resurgent degradation occurred. Tanga  1999 to present This local institution operates under local government authority framework and aims at managing all environmental resources at the community level and acts as a link between local communities and government officers like the District Forest Officer (DFO). The VNRC was active in past years, but it is now inactive after phasing out of short-term international donor agencies. VNRC has its bylaws that were signed by village councils to operate within the study communities. Forest Act No. 14 2002 to present This legal framework advocates forest management by improving efficiency in the use of forest products in a sustainable manner through encouraging local communities and private sector involvement. over the mangrove resources (Ngomela 2007), and thus promotion of the joint forest management (JFM), as emphasized by the Tanzania National forest policy, remains to be a theory on paper (Mangora 2011). The poor people who deeply depend on these resources for their livelihoods are perceived by policy makers in the context of enforcement of legal provisions as culprits of degradation, instead of being partners of conservation (Nyangoko et al. 2021). Government restrictions, including banning measures enacted in 2016, had also enraged a large number of mangrove-reliant households and deprived them of their livelihoods (Nyangoko et al. 2021). This has resulted in friction between the government and mangrove-dependent communities, a situation that has reduced the willingness of local people to participate in rehabilitation and conservation initiatives. The results of the FGDs exposed that forest officers rarely visit local communities and when they do, it is often for different missions (e.g., attending field work or escorting researchers to villages) rather than for community consultations. Accordingly, government agencies (TFS and DFO) are perceived weak in enforcing measures and regulations regarding conservation and sustainable use of mangroves. This result agrees with the note that inadequate government budget allocation, few forest extension staff, and lack of field equipment are challenges that affect effectiveness of collaborative arrangements. Specifically, the DFO   Highly degraded Rapid population growth, emergence and proliferation of commercial exploitation of mangrove resources, and poor management by government agencies.

1994-2005
Improved with moderate degradation Implementation of conservation initiatives including MMP (1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004), TCZCDP (1994TCZCDP ( -2007 claimed that "We cooperate with local communities but not too much because we are few in numbers and we do not have enough money or equipment to perform regular patrols and provide training to local communities." Inadequate staffing and funding to effectively enforce laws and regulations related to mangrove resources remains to be a general challenge throughout the country. This is evidenced by Mshale et al. (2017), who explained that shortage of forest staff and inadequate facilities are among the major issues that undermine the management and conservation of mangrove forests in the Rufiji Delta. There is also a lack of adequate coordination among government institutions across the country to harmonize different actor interests involved in managing mangroves resources (Nyangoko et al. 2021). For example, although the forestry department designates all mangrove forests as reserves, the fisheries department issues permits for prawn farm development, and the Ministry of Minerals issues permits for salt extraction in the same designated mangrove forest reserve areas (Mangora 2011). Thus, in the absence of well-designed management measures, different actors will engage in a range of activities that threaten mangrove ecosystems (Maina et al. 2021).
Overall, an understanding of how socioeconomic drivers influence mangrove resource exploitation, and how local people perceive effectiveness of mangrove management, such as those shown in this study, enrich policy knowledge on interaction between people, mangroves, and conservation actors. Poor or lack of collaboration tends to exacerbate the problems of mangrove overuse and non-compliance, contributing to degradation of natural capital and negative perceptions on mangrove management efforts.

CONCLUSION
This study explored socioeconomic determinants of exploitation and management of mangroves in the Pangani River Estuary.
Residence time of household, household main occupation, household size, and cost of alternative resources to substitute mangroves as a source of domestic fuel, were positively associated with mangrove exploitation. Most of the respondents in Bweni appreciated the role of local institutions in enhancing mangrove conservation, while only a few in Pangani Magharibi had such similar opinion, albeit not congruent with the National Forest Policy that categorically places mangroves under the joint forest management strategy. Moreover, over half of the respondents indicated that state management agencies are weak in implementing conservation measures for sustainable use of mangroves. Considering these socioeconomic and management challenges, especially those related to resource extraction and legal capacities of local natural resource management institutions, provides a good opportunity to improve management interventions. Collaboration among all stakeholders, awareness raising, and enforcement mechanisms should be strengthened to address the need of both local people and ecosystem protection. To achieve this, the government through the forest services agency needs to reverse the existing forest policy and define clear legal mechanisms for engaging communities around mangroves. This may facilitate active participation of local people in the management and allow for the inclusion of a sense of ownership of resources through formally shared management rights that define agreements on areas of use and access rights as a bottom-up approach. To effectively support these communities, a call is made for government through responsible state agencies and departments to promote alternative income generating activities, recognize and appreciate the traditional and cultural reliance of communities on the mangrove resources, so that the best collaborative arrangements can be put in place for a win-win situation. More importantly, with help of village leaders and elders through open dialogue at the community level, marginalized local people, often the most poor and natural capital dependent, should be identified, empowered through measures such as soft loans, monitored and trained to begin working on small livelihoods projects that are vested in local contexts.
Responses to this article can be read online at: https://www.ecologyandsociety.org/issues/responses. php/13227 Ecology and Society 27(2): 32 https://www.ecologyandsociety.org/vol27/iss2/art32/ Author Contributions: B.P.N. was the lead author and responsible for coordinating all contributions from co-authors. The conception and design of the study was carried out by B.P.N., M.S.S., and M.M.M., analysis was conducted by B.P.N. and M.S.S., interpretation of data was done by B.P.N., M.S.S., M.M.M., M.G., and H.B., writing -review and editing was done by B.P.N., M.S.S., M.M.M., M.G., and H.B. All authors give final approval of submitted version of the paper.

Acknowledgments:
We express our gratitude to local communities of Bweni and Pangani Magharibi for their time and providing valuable information used in this study, and three anonymous reviewers for useful comments. This research was funded by the Swedish International Development Cooperation Agency (Sida) through the bilateral marine science program between Sweden and Tanzania.

Data Availability:
The data that support the findings of this study are available on request from the corresponding author, [B.P.N]. The data are not publicly available to respect the privacy of the responding communities.