Skip to main content

Drivers of the Distribution of Fisher Effort at Lake Alaotra, Madagascar


Understanding how fishers make decisions is important for improving management of fisheries. There is debate about the extent to which small-scale fishers follow an ideal free distribution (IFD) – distributing their fishing effort efficiently according to resource availability rather than being influenced by social factors or personal preference. Using detailed data from 1800 fisher catches and from semi-structured interviews with over 700 fishers at Lake Alaotra, the largest inland fishery in Madagascar, we show that fishers generally conform to IFD. However, there were differences in catch: effort relationships between fishers using different gear types as well as other revealing deviations from the predictions of IFD. Fishers report routine as the primary determinant of their choice of fishing location, explaining why they do not quickly respond to changes in catch at a site. Understanding the influences on fishers’ spatial behaviour will allow better estimates of costs of fishing policies on resource users, and help predict their likely responses. This can inform management strategies to minimise the negative impacts of interventions, increasing local support for and compliance with rules.


Inland fisheries are widely recognised as significant sources of food and income for rural communities but most are fully exploited or overfished (FAO 2010; Welcomme 2011). Understanding fisher behaviour is critical to designing interventions that account for how fishers may respond to management in order to minimise the adverse impacts of interventions, and thereby increase the likelihood of fisher support and compliance (Wilen et al. 2002; Salas and Gaertner 2004; Cinner et al. 2008). To date, most fisheries literature focuses on commercial fishing fleets in developed countries whereas comparatively little is known about the complexities of subsistence or artisanal fisheries in developing countries (Welcomme et al. 2010). However, many of the factors that influence fisher decision-making and behaviour in commercial fisheries may be applicable in subsistence or artisanal contexts: economic factors such as risk strategy, access to gear or vessels, and fish prices (van Oostenbrugge et al. 2001; Cinner and McClanahan 2006; Tidd et al. 2011); biological factors such as fish densities and distribution (Gillis et al. 1993; Abernethy et al. 2007); environmental factors such as weather and human activity (Cinner and McClanahan 2006; Daw et al. 2011); social and institutional factors such as traditions and spatial or temporal restrictions (Pálsson and Durrenberger 1990; Bjarnason and Thorlindsson 1993; Dinmore et al. 2003); as well as fisher-specific factors such as gear or bait preferences, effort, and fishing experience and ability (Parker and Sutherland 1986; Christensen and Raakjær 2006; Smith and Zhang 2007).

Rational choices and utility maximisation are primary tenets of traditional economic theory (Morse 1997; Güth 2008), and models based on these tenets have frequently been applied within fisheries research to explain fisher decision-making and behaviour (Holland 2008; Daw et al. 2011). Such models assume that fishers have complete knowledge of fishery characteristics and use this information to make fishing decisions to maximise their personal utility, with profit often used as a proxy for utility (see Daw et al. 2011). However, the applicability of microeconomic theory has been challenged by anthropologists (e.g., Ryan and Bernard 2006; Miller et al. 2014) and behavioural economists (e.g., Gelcich et al. 2007; Hastie and Dawes 2010), based on empirical evidence for how decisions are actually made rather than how they should be made. Accordingly, it is now increasingly recognised that fishers’ strategies or choices can vary considerably among individuals and involve a range of compromises that drive their patterns of fishing behaviour (Abernethy et al. 2007; Daw 2008; Holland 2008).

The concept of ideal free distribution (IFD) was first developed as a model of how animals distribute themselves among several patches of resources, the number of individuals being proportional to the amount of resources available at each location (Fretwell and Lucas 1969). The term ‘ideal’ assumes that harvesters have accurate knowledge of the distribution of targeted resources (such as fish species) and the term ‘free’ assumes that resource users (fishers) are able to move between locations without constraint (Kacelnik et al. 1992; Gillis 2003). IFD has been applied to small-scale fishers but there is debate as to the extent to which it is a useful model to explain decision making of human harvesters (Abernethy et al. 2007), since other social or individual factors may be more important predictors of fishing behaviour.

For this study we examined fisher behaviour at Lake Alaotra, the largest lake in Madagascar and base for the nation’s most productive inland fishery (Andrianandrasana et al. 2005). We use information from in-depth semi-structured interviews and structured catch interviews with fishers from Anororo village to address four key questions:

  1. 1.

    What appear to be the main drivers of effort and choice of fishing location?

  2. 2.

    Do fishers follow an ideal free distribution in their effort?

  3. 3.

    What appear to be the drivers of deviations from IFD?

  4. 4.

    Under what circumstances do fishers change their fishing behaviour, particularly location?

We consider our results in the light of current and planned conservation interventions, which include no take zones and gear restrictions.


Study Site

The Alaotra wetland in northeast Madagascar is internationally recognised as an important area for biodiversity conservation. It was declared a Ramsar site in September 2003 (Ramanampamonjy et al. 2003) and gazetted as a new protected area by the government of Madagascar in 2007 (Andrianandrasana 2009). Lake Alaotra covers 200 km2 and has a seasonal maximum depth of 4 m (Moreau 1979; Vanden Bossche and Bernacsek 1991; Ferry et al. 2009). The marsh adjoining the lake covers 230 km2 (Andrianandrasana et al. 2005; Ferry et al. 2009). Both the lake and the marsh areas are used by fishers (Fig. 1). There are two main climatic seasons: the wet season occurs from December to April and is hot with heavy rain and rising water levels, while the dry season from May to November is cooler and water levels decrease. There is up to two metres difference in water level between the high in March and the low in November (Moreau 1979; Ferry et al. 2009).

Fig. 1
figure 1

Map of Lake Alaotra showing management zones within the lake and adjacent marsh, catch interview sites, and the centroids of fishing locations used by local fishers as recorded in the catch interview data

The human population in the Lake Alaotra area has increased from 109,000 in 1960 (Pidgeon 1996) to 550,000 in 2003 (Andrianandrasana et al. 2005). The number of fishers operating in Lake Alaotra increased from approximately 1000 in the 1960s to over 4000 by 1989 (Wilmé 1994) and continues to increase with over 7000 fishers currently operating in the lake (H. Andrianandrasana, DWCT, pers. comm.), adding further pressure on already heavily exploited and overexploited fish stocks (Vanden Bossche and Bernacsek 1991; Pidgeon 1996).

Fisheries management authority for Lake Alaotra is vested in the Service Régional de la Pêche et des Ressources Halieutiques (Fisheries Service). Local fishers are represented by a Federation of Fishers that is intended to work with the Fisheries Service to co-manage resources by developing, implementing, and monitoring fishing regulations. In practice, the Federation of Fishers is perceived by many fishers as an extension of the Fisheries Service and therefore distrusted (APCW pers. obs.). There is a general lack of enforcement and poor compliance with all regulations, which were drawn up with limited consideration or knowledge of impacts on local fishers (Wallace et al. 2015). Consequently, the Lake Alaotra Management Plan is largely viewed as a ‘paper park’ that is failing to meet conservation goals (Fig. 1).

Anororo village is a relatively large community of approximately 8000 people on the western edge of Lake Alaotra within the Alaotra-Mangoro region of Madagascar (PCD 2004), adjacent to marsh and lake habitat, and was selected for (i) its large population of fishers using a variety of habitats and fishing methods, (ii) proximity to current and planned fishery management interventions, and (iii) local dependence on fishery resources for subsistence and commercial activity. Current conservation interventions are based on the 2006 Lake Alaotra Management Plan, which aimed to improve the sustainability of the fishery, reduce pressures on the wetland, and conserve rare bird and mammal species, particularly the Critically Endangered Alaotran gentle lemur (Razanadrakoto and Rafaliarison 2005).

The subsistence and cash economy of Anororo centres on fishing and rice cultivation. Although some households engage in both activities, we targeted people who are primarily fishers with very few, if any, alternative sources of food and income. Anonoro-based fishers are male and almost exclusively (99 %) from the Sihanaka ethnic group. Those we interviewed were broadly representative of the landless population in Anororo, who have lower incomes and fewer assets than local land-owning rice farmers, in terms of ethnicity, age distribution, level of education, and number of household members supported (Le Courtois 2010).

Fishers operate independently from dugout canoes and use a broad range of gear types including traps, gill nets, cast nets, and line & hook. Traps and line & hook are used passively overnight with usually 24 h between fish collections. Gill nets were observed being used in three ways: (i) passively overnight, (ii) passively while waiting, and (iii) actively. Cast nets are used actively or passively with bait. Fishing occurs in lake, marsh, and lake-marsh edge habitats. The main market for fish is commercial buyers from Antananarivo, Madagascar’s capital; local market and bicycle collectors also operate in the village.

Data Collection

Data were collected in June and July 2009 and from October 2009 to December 2010, comprising two dry seasons (2009 and 2010) and one wet season (2010). We conducted three types of interviews with Anororo-based fishers: structured catch interviews (n = 1800); semi-structured background interviews (n = 405); and subsequent follow-up interviews (n = 221). Local research assistants conducted interviews in Malagasy and translated responses into French.Footnote 1 All interviews were conducted opportunistically with fishers who were willing to participate. Fishers were informed that participation was voluntary and that their identities and responses would not be shared with anyone.

Catch interviews were conducted at two sites (one on the lake and one in the marsh on the main thoroughfares to the village where buyers wait) prior to fishers selling their catch. Fishers do not return to the village with their catch if they intend to sell it, which was the case for most fishers. Background interviews were conducted by walking through all areas of the village and approaching people who had some involvement in fishing (identified with help from local guides and key informants). Follow-up interviews were conducted 11 to 12 months later, and only with fishers who had participated in background interviews. No fishers refused to be re-interviewed. A total of 784 fishers, approximately 85 % of fishers in the village, participated in one or more types of interview. A total of 110 fishers participated in all three types of interviews. Respondent codes were assigned to all participating fishers to preserve their anonymity (Bernard 2002) as well as to triangulate the data.

Catch Interviews

(5 to 10 min) comprised a series of questions about fishing activity that day, including the fishing location, gear type, and effort used. A mapping exercise conducted with senior and experienced fishers, followed up with visits to take GPS waypoints, provided locations of locally important fishing areas (Fig. 1; Wallace 2012). A total of 537 individual fishers participated in catch interviews; 248 fishers (46 %) were interviewed more than once. Catch interviews also included counting and measuring the fish caught (n = 27,064). Total catch weights were estimated using species-specific length-weight relationships (Wallace 2012).

Background Interviews

collected information on demographics, reliance on fishing for livelihood, and fishing behaviour including type of gear and fishing location(s) used. A total of 158 fishers participated in catch interviews as well as background interviews.

Subsequent Follow-Up Interviews

collected information on perceptions of the status of the fishery and management interventions, as well as how fishers’ spatial distribution has changed over time, particularly whether they added or dropped locations since the previous background interview and the reasons for changes.

Data Analysis

We first analysed background interviews to identify key demographic factors that best explain variation in effort across fishers. Second, we plotted data on catch and effort for fishing locations for comparison with the 1:1 prediction of the catch:effort relationship expected under IFD, and examined in detail those locations that deviated from IFD. Finally, we used background and follow-up interview data to examine drivers of distribution and changes in distribution of effort. Quantitative data, including frequencies and proportions, were analysed using R version 2.14.2 (R Development Core Team 2012).

Analysis of Fisher Effort

To identify different groups of fishers and determine the factors influencing fisher effort, measured as time spent fishing, we analysed data from background interviews with fishers (n = 405) using a negative binomial generalised linear model (GLM). Five socioeconomic characteristics of fishers were assessed as explanatory variables: (1) age category, (2) total number of people supported in the household, (3) level of education (up to primary school or secondary school and above), (4) whether the fisher’s household has an alternative livelihood or source of income, and (5) type of gear (e.g., traps, gill nets) used for fishing. Time spent fishing is a meaningful measure of fisher effort in both biological and economic terms, and allows for comparison between gear types as well as between fishing locations at varying distances from the village (Abernethy et al. 2007; Daw et al. 2011). Our measure of effort and analysis of IFD did not include travel time, due to potentially confounding factors such as other activities undertaken en-route to fishing location as well as individual differences in paddling speed related to age and/or physical fitness. Previous research within the Lake Alaotra system confirms time spent fishing influences catch weight but travel time does not (Wallace et al. 2015).

Years of fishing experience was significantly and strongly positively correlated with fisher age (r = 0.81, p < 0.0001) and therefore not included as an explanatory variable. Almost all fishers interviewed (99 %) were from the Sihanaka ethnic group, and 98 % had lived in Anororo since birth. Most interviewees (86 %) stated that fishing was their primary livelihood, and many also had an alternative source of income during the calendar year. Lack of variation in these factors meant it was not possible to examine their influence on fisher effort and they were therefore not included in the analysis.

We used Akaike’s Information Criterion (AIC) model selection to rank models and quantify the magnitude of difference between them, and model averaging to determine model-averaged coefficients (Burnham and Anderson 2002; Bolker et al. 2009). The global model was run using the MASS package in R; the MuMIn package was used for model comparison and averaging. Following Burnham and Anderson’s (2002) rule of thumb, all models where AIC differences were less than four were included in the candidate set of models for model averaging. AIC differences of <4 was chosen because the weight or support for subsequent models decreased considerably at this point. No single model was clearly superior to others in the candidate set of models, suggesting that model averaging would provide a more robust understanding of the system and reduce model selection bias effects (Burnham and Anderson 2002).

Ideal Free Distribution

IFD states that harvesters will distribute themselves in relation to resource availability and predicts that the proportion of aggregate effort will be equal to the proportion of aggregate catch at each location (Fretwell and Lucas 1969). We analysed data from catch interviews for a total of 1757 fishing trips by 515 individual fishers; separate analyses were conducted for a sub-sample of 788 fishing trips by 151 individual fishers who had also participated in background interviews. Catch interviews where no fish were measured (n = 39) because fishers had sold their catch prior to interview, did not want to have their fish measured, or did not have time for their fish to be measured, were excluded from analyses. Interviews with fishers who had fished in the immediate vicinity of the village were infrequent (n = 4) and also excluded from analyses. Proportions of catch (measured as total weight caught) and effort (measured as total number of hours spent fishing) observed at fishing locations over the study period were calculated across all gear types and for each gear type. Deviation was calculated as proportion of catch divided by proportion of effort over the study period by gear type. A positive deviation (>1) occurs where proportion of catch exceeds proportion of effort; a negative deviation (<1) occurs where proportion of effort exceeds proportion of catch. We examined the attributes of particular sites that deviated from IFD to determine drivers of fisher behaviour.

Stated Drivers of Fisher Behaviour

Fisher responses to semi-structured interview questions were categorised into common themes for analysis. Response sample size varies according to whether background or follow-up interviews were conducted and because fishers sometimes gave vague or ambivalent responses that could not be categorised. Data are presented as the percentage of interviewees providing a particular response to questions regarding (a) reasons for choosing a fishing location, and (b) whether they changed location(s) chosen for fishing during the study period and why. We summarized differences among fisher groups and compared them using chi-square and Fisher’s exact tests. In particular we were interested in drivers of change in location, and whether fishers who changed did so because conditions had deteriorated (circumstances ‘pushed’ them out) or were perceived to have improved in the new location (‘pulled’ them to the new location). We hypothesized that if fishers truly conformed to IFD they would predominantly be ‘pulled’ into new locations for better catches in order to maximise returns per unit of effort.


Drivers of Fishing Effort

Results from the negative binomial GLM indicated that gear type and number of people supported were significant predictors of time spent fishing (Table 1). Fisher age category, level of education, and presence of an alternative livelihood in the household were not significant explanatory variables. These results confirmed that gear type could be used to categorise fishers at a broad scale.

Table 1 Results of the negative binomial generalised linear model of fisher profile variables explaining fisher effort measured as time spent fishing

The characteristics and fishing activity of fishers who had participated in catch interviews as well as background interviews (n = 151) were therefore grouped by gear type for comparison. The characteristics of fishing activity, in particular mean catch per trip in kilograms and mean effort (time spent fishing in hours) per trip, differed significantly across gear types (Table 2).

Table 2 Characteristics of fishers and their fishing activity by gear type

Spatial Distribution of Fisher Effort

Across all fishing trips, irrespective of gear type and fisher identity, fishers appear to conform to IFD; the proportion of effort (i.e., time spent fishing) allocated to fishing locations is directly proportional to the proportion of catch derived from those locations (Fig. 2a). There are however differences in deviation from IFD amongst fishers using different gear types (Fig. 2b to f). Notably, trap fishers appear to adhere more to IFD than gill net fishers.

Fig. 2
figure 2

Proportions of catch and effort observed at fishing locations in Lake Alaotra over the study period (n = 1757 catch interviews with 515 individual fishers), calculated across all gear types and for each gear type. Catch was measured as total weight caught and effort was measured as total number of hours spent fishing at the location over the period. Solid circles represent fishing locations within restricted areas; open circles represent locations within non-restricted areas (see Table 3 for characteristics of each labelled location). The dotted line represents the 1:1 prediction of IFD

Heteroscedasticity within the dataset is clear, and considerably more variation occurs at fishing locations where proportions of catch and effort are high (Fligner-Killeen test, χ 2 = 24.64, df = 6, p < 0.001). There were no explicit patterns or differences between years (2009 vs. 2010) to explain this deviation from IFD. Linear models to explore factors influencing deviation for each gear type were inconclusive. However, for each gear type, some general patterns can be drawn from the characteristics of locations that deviate particularly strongly from IFD (Table 3).

Table 3 Characteristics of locations with greatest deviation from the ideal free distribution by gear type

An important factor for location selection is convenience, i.e., distance travelled. For trap fishers, for example, although Andratsilanina (ADT; positive deviation) is distant from Anororo, it is very suitable for setting traps at the lake-marsh edge and perceived to have high fish abundance. In comparison, Amparihy (AMP; negative deviation) is characterised by extensive human traffic and degraded habitat due to close proximity to Anororo. Despite the relatively high level of habitat degradation, fishing, and other activities at Amparihy, fishers continued to fish in this location due to its proximity to the village, indicating that they value convenience more highly than potentially higher catches elsewhere. For gill net fishers Ankororo (ANK) and Sahabe (SHB), both deviating positively from IFD, are low-traffic areas lacking vegetation in which nets can be damaged, entangled, or dislodged, in contrast to Ambavasaha (ABV; negative deviation) which is closer to Anororo. Disproportionately high fishing activity, given the returns, at Ambavasaha indicates that gill-net fishers are risk averse when it comes to personal safety, often stating that fishing near the lake edge was safer than further offshore despite smaller catches. For all gear types, returns were generally better at sites more suitable for the gear and/or within a protected area. Persistence at sites with lower returns was mostly attributable to convenient proximity to Anororo.

Proportions of catch and effort were also compared to IFD for the sub-sample of fishers (n = 151) who also participated in background interviews (Fig. S1). The sub-sample conformed to IFD in a similar pattern to that for the larger sample of all fishers participating in catch interviews (n = 515). The total sample of fishers participating in background interviews (n = 405) was therefore used to further explore the drivers of fisher behaviour.

Factors Influencing Fishers’ Spatial Behaviour

Although fishers participating in catch interviews used a single location per fishing trip, fishing locations were used adaptively according to changing conditions over the year; fishers use a median of two (range 1 to 6) locations over the calendar year (Wallace 2012). Almost all fishers participating in background interviews (98 % of 405) stated they had continued to fish at the same location(s) over the last 5 years. Routine was the most frequently cited reason to use a location, followed by perception of good catches at these sites (Table 4). Being familiar with or having good knowledge of a fishing location was also stated as important; however, only seven of these fishers cited a long family history of fishing at a location as a reason for using it. Typical responses by fishers when asked why they use their location(s) include: “It is my habit and I don’t know other locations;” “I am used to getting mostly good catches at this location;” “My location puts me at ease for fishing;” and “I know the route from habit and do not have to fear getting lost.

Table 4 Reasons provided by fishers (n=403) for choosing fishing locations. The proportion of fishers stating each reason is grouped by gear type

The vast majority (94 %) of fishers interviewed use traps or gill nets; therefore we focussed on responses from these two groups, for whom the reasons for selecting fishing locations differed significantly (Fisher’s exact test, p = 0.024). Specifically, a greater proportion of trap fishers cited familiarity and travel time as reasons for selecting a location (chi-square tests: familiarity χ 2 = 11.79, df = 1, p < 0.001; travel time χ 2 = 6.53, df = 1, p = 0.011). Reasons for selecting a fishing location did not differ between age categories (Fisher’s exact test, p = 0.597).

Despite the high degree of consistency in relation to location choice during background interviews (Table 4; routine), 81 % of fishers subsequently participating in follow-up interviews (n = 221) stated they were using a different set of locations since the background interview due to unusual seasonal changes. Fishers stated that low rainfall and extended cooler temperatures early in 2010 limited fish movement and growth, resulting in reduced stock and ultimately reduced catch sizes. Consequently, the most frequently cited reasons for fishers being pushed out of or pulled into different fishing locations were related to catch size, travel, and water level (Table 5). In most cases, factors that pushed fishers out of their preferred fishing location(s) were more powerful, indicating their reluctance to move otherwise, and is consistent with initial responses specifying routine as the main driver of location choice. Only travel-related factors pulled rather than pushed a higher proportion of fishers to a new location, which was always closer to Anororo. Reasons for switching locations did not differ among gear types or among age categories (Fisher’s exact tests, p = 0.742 and p = 0.420, respectively).

Table 5 Reasons provided by fishers (n = 178) for being pushed out of or pulled into other fishing locations. The number and proportion of fishers stating each reason are grouped for pushed and pulled

A fisher’s choice of location is influenced, and often constrained, by type of gear used (e.g., traps cannot be used in open lake habitat because they need to be fixed to supporting materials such as reeds to hold them in place). Only 4 % of fishers participating in background interviews (n = 405) had changed their gear type in the previous five years. Most fishers (80 %) specified routine and/or competence as the primary reasons for continuing to use their preferred type of gear. Representative statements include: “It is my habit, I have used nets for a long time;” “I am competent with traps and do not know how to use other methods;” “I know how to make nets;” and “Traps are easy to use and are my tradition.


Examining drivers of fisher effort and behaviour, this study shows that Anororo-based fishers within Lake Alaotra’s small-scale fishery generally conform to IFD, and they choose locations throughout the lake and marsh they believe will provide good or consistent (but not necessarily maximal) catches. Departures from IFD are primarily tied to convenience and/or the distinctive environmental or anthropogenic characteristics of a fishing site, which often assume importance for gear-specific reasons. Our finding that trap fishers adhere more to IFD than gill net fishers is counter-intuitive – trap fishers can move location less easily because their traps are left overnight and not readily moved large distances, whereas gill net fishers can in theory choose their location on a daily basis.

Type of gear used is the major determinant of fishing effort and choice of fishing location. Gear-based variation in fisher effort and spatial distribution often characterises subsistence or artisanal fisheries, particularly in developing tropical countries (Abernethy et al. 2007; Daw et al. 2011; Hamerlynck et al. 2011). Environmental and habitat factors may constrain the type(s) of gear a fisher can use, or render some methods more suitable than others (see Welcomme 2001). Whereas variation in gear use often occurs to target different species (Gillis et al. 1993; Abernethy et al. 2007), for Anororo-based fishers this variation is mediated by differences in habitat within their fishing arena. For example, gill nets require relatively large open areas clear of obstructions, which are typically further from the village and involve greater travel time, while fishers using hand methods require shallow water and/or marsh habitat, which occur closer to the village. These gear-specific influences of environmental factors affect fishers’ choices of fishing location (and therefore spatial distribution), demonstrating that gear type is central to understanding fisher behaviour within multi-habitat fisheries.

Our finding that Anororo-based fishers generally conform to IFD contrasts with Abernethy et al. (2007) for artisanal Anguillan reef fishers, where the relatively substantial departures from IFD were linked to fisher age, experience, and target species, as well as type of gear. In our study, species were not explicitly targeted, fisher age or experience were not drivers of catch (see Wallace et al. 2015), and catch influenced fishers when selecting or changing fishing location(s). This suggests that fishers pursue rational strategies to the extent their knowledge permitted when distributing effort spatially, which is usually more characteristic of larger-scale commercial fisheries than artisanal systems (Branch et al. 2006; Powers and Abeare 2009). Despite this, and in contrast to the central assumption of IFD that all individuals aim to optimise profits (Gillis 2003; Abernethy et al. 2007), fishers do not fish in ways to maximise returns. Rather, in line with equivalent findings by Béné and Tewfik (2001), Cabrera and Defeo (2001), Salas and Gaertner (2004), and Daw (2008), fishers’ decisions on effort distribution are mediated by multiple trade-offs including convenience, routine, gear usability and maintenance, or predictability of catch. Commonly preferred strategies are (a) spend more time fishing closer to home rather than invest that time travelling to distant locations or (b) continue fishing in a familiar location where catches are more predictable but sometimes small (see Swain and Wade (2003) for similar strategies for fishers of snow crab in the Gulf of St. Lawrence, Daw (2008) for lobster fishers in Nicaragua, and Teh et al. (2012) for small-scale fishers in Sabah, Malaysia, indicating that a broad range of fishers compromise or satisfice instead of optimising utility (Simon 1955; Foxon 2006)).

We took a rational choice approach to understanding fishers’ decisions, by assessing the extent to which fishers’ decisions about spatial distribution conform to IFD. However, the range of influences and trade-offs reported by fishers suggest further research using alternative approaches to understanding decision-making may provide additional insights (Gintis 2007). Similarly, it would be informative to examine fishers’ behaviour over time and under changing conditions (including the implementation of new management) to determine the degree to which drivers and patterns of behaviour are fixed or adaptive and the degree to which they can be generalised to other fisheries.


The predominance of routine as a driving factor for Anororo-based fishers extends from choice of fishing location through to selection of gear and persistence with that type of gear. Specifically, fishers’ decisions about location and gear intertwine; once choices are made, persistence is bound by (i) familiarity with relevant site characteristics, (ii) competence with type of gear used, and (iii) perceived costs of changing location or gear. Provided catches remain adequate for the amount of effort invested, which depends on trade-offs with fishers’ interests other than catch, there is considerable inertia within the fishery and reluctance to change. Routine, habit, and/or familiarity with location or gear are increasingly identified as key factors in fisher decisions, such as in New England trawl fisheries (Holland and Sutinen 2000), urchin divers’ location choices in California (Smith 2005), preferred resource spaces by small-scale fishers in Malaysia (Teh et al. 2012), and gear use by Swedish fishers (Eggert and Tveteras 2004).

Catch size and travel costs (which interlink with water level and access to fishing locations) are most likely to motivate fishers to change their spatial behaviour. Poor catches and/or high travel costs frequently push changes in fishers’ spatial behaviour. The prospect of better catches and/or lower costs per se has less effect because of the constraining influences of routine and familiarity. These patterns show that although Anororo-based fishers conform generally to IFD and make rational spatial decisions, they are risk averse and sub-maximal catches suffice under conditions of uncertainty. A key implication of this is that these and similar fishers could be less responsive to purely economic incentives to modify their behaviour than commonly expected (see Holland 2008). By demonstrating that fishers’ behaviour is typically mediated by tradeoffs and fishers do not always maximise returns, our study makes an important contribution to our understanding of fisheries dynamics by evaluating and explaining fisher spatial behaviour at a scale relevant to conservation planning. This understanding could be used to advantage in fisheries management to ensure plans and actions accommodate resource-users’ behaviour; it could also inform the development of meaningful and realistic incentives to support interventions and comply with regulations.

From a fisheries management perspective it is pragmatic to understand and account for fisher behaviour collectively (Béné and Tewfik 2001; Holland 2008; Cinner et al. 2010). Our study suggests that trade-offs and variation in spatial behaviour within multi-gear artisanal fisheries may be best understood by grouping fishers according to their type of gear, and that IFD is a useful null hypothesis against which to examine differences between fishers in their distribution among locations. Our findings are likely to also be applicable for some commercial fisheries, particularly where artisanal and commercial fisheries overlap. Misunderstandings about diversity of interests and motivations influencing fishers’ spatial behaviour have often led to inappropriate and ineffective management interventions, or compromised compliance with regulations, by imposing greater costs on fishers than they are able to bear (Peterson and Stead 2011). Data from this study may be used with spatial planning tools and scenario analysis to inform development of reserve designs that account effectively for fisher behaviour, costs to fishers, and biodiversity goals.

Management actions are likely to have substantial impact in subsistence, artisanal, and developing-country settings because fishers will typically be socioeconomically highly invested in and dependent on fishing. These fishers will also probably be relatively poor (and hence vulnerable and less resilient to shocks), lack buffers to offset seasonal variation in catches and income, and have limited livelihood options (Hill 2011; Teh et al. 2012). Greater research attention should be afforded to understanding the relationships among fishers’ motivations, perceptions of the costs of management actions, and responses to those actions in order to recommend management actions and conservation interventions with minimal negative impacts for fishers who depend on the fishery for livelihood. A better understanding of these relationships could in turn increase fisher compliance and hence the effectiveness of such actions and interventions.


  1. APCW is fluent in French and over the course of the study, acquired working knowledge of Malagasy. JPGJ is fluent in conversational Malagasy.


  • Abernethy, K. E., Allison, E. H., Molloy, P. P., and Côté, I. M. (2007). Why do Fishers Fish Where They Fish? Using the Ideal Free Distribution to Understand the Behaviour of Artisanal Reef Fishers. Canadian Journal of Fisheries and Aquatic Sciences 64(11): 1595–1604.

    Article  Google Scholar 

  • Andrianandrasana, H. T. (2009). Développement des Systèmes de Télédétection en Vue d’améliorer le Suivi de l’Habitat et des Feux de Marais au Lac Alaotra. Mémoire de Diplôme d’Etudes Approfondies en Foresterie Développement et Environnement. Université d’Antananarivo, Antananarivo. 120 pp.

    Google Scholar 

  • Andrianandrasana, H. T., Randriamahefasoa, J., Durbin, J., Lewis, R. E., and Ratsimbazafy, J. H. (2005). Participatory Ecological Monitoring of the Alaotra Wetlands in Madagascar. Biodiversity and Conservation 14: 2757–2774.

    Article  Google Scholar 

  • Béné, C., and Tewfik, A. (2001). Fishing Effort Allocation and Fishermen’s Decision Making Process in a Multi-species Small-Scale Fishery: Analysis of the Conch and Lobster Fishery in Turks and Caicos Islands. Human Ecology 29(2): 157–186.

    Article  Google Scholar 

  • Bernard, H. R. (2002). Research Methods in Anthropology: Qualitative and Quantitative Methods, 3rd ed. AltaMira Press, Walnut Creek. 753 pp.

    Google Scholar 

  • Bjarnason, T., and Thorlindsson, T. (1993). In Defense of a Folk Model: The “Skipper Effect” in the Icelandic Cod Fishery. American Anthropologist 95(2): 371–394.

    Article  Google Scholar 

  • Bolker, B. M., Brooks, M. E., Clark, C. J., Geange, S. W., Poulsen, J. R., Stevens, M. H. H., and White, J.-S. S. (2009). Generalized Linear Mixed Models: A Practical Guide for Ecology and Evolution. Trends in Ecology & Evolution (Personal edition) 24(3): 127–135.

    Article  Google Scholar 

  • Branch, T. A., Hilborn, R., Haynie, A. C., Fay, G., Flynn, L., Griffiths, J., Marshall, K. N., Randall, J. K., Scheuerell, J. M., Ward, E. J., and Young, M. (2006). Fleet Dynamics and Fishermen Behavior: Lessons for Fisheries Managers. Canadian Journal of Fisheries and Aquatic Sciences 63(7): 1647–1668.

    Article  Google Scholar 

  • Burnham, K. P., and Anderson, D. R. (2002). Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach, 2nd ed. Springer, New York. 488 pp.

    Google Scholar 

  • Cabrera, J. L., and Defeo, O. (2001). Daily Bioeconomic Analysis in a Multispecific Artisanal Fishery in Yucatan, Mexico. Aquatic Living Resources 14(1): 19–28.

    Article  Google Scholar 

  • Christensen, A.-S., and Raakjær, J. (2006). Fishermen's Tactical and Strategic Decisions: A Case Study of Danish Demersal Fisheries. Fisheries Research 81(2–3): 258–267.

    Article  Google Scholar 

  • Cinner, J. E., Daw, T., and McClanahan, T. R. (2008). Socioeconomic Factors That Affect Artisanal Fishers’ Readiness to Exit a Declining Fishery. Conservation Biology 23(1): 124–130.

    Article  Google Scholar 

  • Cinner, J. E., and McClanahan, T. R. (2006). Socioeconomic Factors that Lead to Overfishing in Small-Scale Coral Reef Fisheries of Papua New Guinea. Environmental Conservation 33(1): 73–80.

    Article  Google Scholar 

  • Cinner, J. E., McClanahan, T. R., and Wamukota, A. (2010). Differences in Livelihoods, Socioeconomic Characteristics, and Knowledge About the Sea Between Fishers and Non-fishers Living Near and Far from Marine Parks on the Kenyan Coast. Marine Policy 34(1): 22–28.

    Article  Google Scholar 

  • Daw, T., Maina, J., Cinner, J., Robinson, J., and Wamukota, A. (2011). The spatial behaviour of artisanal fishers: Implications for fisheries management and development (Fishers in Space). Western Indian Ocean Marine Science Association (WIOMSA), Marine Science for Management (MASMA). 79 pp.

  • Daw, T. M. (2008). Spatial Distribution of Effort by Artisanal Fishers: Exploring Economic Factors Affecting the Lobster Fisheries of the Corn Islands, Nicaragua. Fisheries Research 90(1–3): 17–25.

    Article  Google Scholar 

  • Dinmore, T. A., Duplisea, D. E., Rackham, B. D., Maxwell, D. L., and Jennings, S. (2003). Impact of a Large-Scale Area Closure on Patterns of Fishing Disturbance and the Consequences for Benthic Communities. ICES Journal of Marine Science 60(2): 371–380.

    Article  Google Scholar 

  • Eggert, H., and Tveteras, R. (2004). Stochastic Production and Heterogeneous Risk Preferences: Commercial Fishers’ Gear Choices. American Journal of Agricultural Economics 86(1): 199–212.

    Article  Google Scholar 

  • FAO (2010). The State of World Fisheries and Aquaculture. FAO Fisheries and Aquaculture Department, Food and Agriculture Organization of the United Nations, Rome. 218 pp.

    Google Scholar 

  • Ferry, L., Mietton, M., Robison, L., and Erismann, J. (2009). Le lac Alaotra à Madagascar - Passé, Présent et Futur. Zeitschrift für Geomorphologie 53(3): 299–318.

    Article  Google Scholar 

  • Foxon, T. (2006). Bounded Rationality and Hierarchical Complexity: Two Paths from Simon to Ecological and Evolutionary Economics. Ecological Complexity 3(4): 361–368.

    Article  Google Scholar 

  • Fretwell, S. D., and Lucas, H. L. (1969). On Territorial Behavior and Other Factors Influencing Habitat Distribution in Birds. Acta Biotheoretica 19(1): 16–36.

    Article  Google Scholar 

  • Gelcich, S., Edwards-Jones, G., and Kaiser, M. (2007). Heterogeneity in Fishers’ Harvest Decisions Under a Marine Territorial User Rights Policy. Ecological Economics 61: 246–254.

    Article  Google Scholar 

  • Gillis, D. M. (2003). Ideal Free Distributions in Fleet Dynamics: A Behavioral Perspective on Vessel Movement in Fisheries Analysis. Canadian Journal of Zoology 81(2): 177–187.

    Article  Google Scholar 

  • Gillis, D. M., Peterman, R. M., and Tyler, A. V. (1993). Movement Dynamics in a Fishery: Application of the Ideal Free Distribution to Spatial Allocation of Effort. Canadian Journal of Fisheries and Aquatic Sciences 50(2): 323–333.

    Article  Google Scholar 

  • Gintis, H. (2007). A Framework for the Unification of the Behavioral Sciences. Behavioral and Brain Sciences 30(1): 1–16.

    Google Scholar 

  • Güth, W. (2008). (Non)Behavioral Economics: A Programmatic Assessment. Zeitschrift für Psychologie / Journal of Psychology 216(4): 244–253.

    Article  Google Scholar 

  • Hamerlynck, O., Duvail, S., Vandepitte, L., Kindinda, K., Nyingi, D. W., Paul, J.-L., Yanda, P. Z., Mwakalinga, A. B., Mgaya, Y. D., and Snoeks, J. (2011). To Connect or not to Connect - Floods, Fisheries and Livelihoods in the Lower Rufiji Floodplain Lakes, Tanzania. Hydrological Sciences Journal 56(8): 1436–1451.

    Article  Google Scholar 

  • Hastie, R., and Dawes, R. M. (2010). Rational Choice in an Uncertain World: The Psychology of Judgement and Decision Making, 2nd ed. Sage Publications, Thousand Oaks.

    Google Scholar 

  • Hill, N. A. O. (2011). Livelihood diversification for conservation: Interactions between seaweed farming and fishing in Danajon Bank, central Philippines. PhD Thesis. Imperial College London & Institute of Zoology, London. 215 pp.

  • Holland, D. S. (2008). Are Fishermen Rational? A Fishing Expedition. Marine Resource Economics 23: 325–344.

    Article  Google Scholar 

  • Holland, D. S., and Sutinen, J. G. (2000). Location Choice in New England Trawl Fisheries: Old Habits Die Hard. Land Economics 76(1): 133–149.

    Article  Google Scholar 

  • Kacelnik, A., Krebs, J. R., and Bernstein, C. (1992). The Ideal Free Distribution and Predator–Prey Populations. Trends in Ecology & Evolution 7(2): 50–55.

    Article  Google Scholar 

  • Le Courtois, S. (2010). Household choices in rice cultivation in a social-ecological system and impacts on productivity: Lessons from Anororo, Lac Alaotra, Madagascar. Master of Science Thesis. Imperial College London, London. 79 pp.

  • Miller, B., Leslie, P., and McCabe, J. T. (2014). Coping with Natural Hazards in a Conservation Context: Resource-Use Decisions of Maasai Households During Recent and Historical Droughts. Human Ecology 42(5): 753–768.

    Article  Google Scholar 

  • Moreau, J. (1979). Essai d’application au Lac Alaotra (Madagascar) d’un Modèle d’étude des Pêcheries pour les Plaines d’inondation Intertropicales. Cahiers ORSTOM series Hydrobiologie 13(1–2): 83–91.

    Google Scholar 

  • Morse, J. R. (1997). Who is Rational Economic Man? Social Philosophy and Policy 14(1): 179–206.

    Article  Google Scholar 

  • Pálsson, G., and Durrenberger, E. P. (1990). Systems of Production and Social Discourse: The Skipper Effect Revisited. American Anthropologist 92(1): 130–141.

    Article  Google Scholar 

  • Parker, G. A., and Sutherland, W. J. (1986). Ideal Free Distributions When Individuals Differ in Competitive Ability: Phenotype-Limited Ideal Free Models. Animal Behaviour 34(4): 1222–1242.

    Article  Google Scholar 

  • PCD (2004). Plan communale de developpement CR Anororo - FID. Unpublished report, Toamasina, Madagascar. 43 pp.

  • Peterson, A. M., and Stead, S. M. (2011). Rule Breaking and Livelihood Options in Marine Protected Areas. Environmental Conservation 38(3): 342–352.

    Article  Google Scholar 

  • Pidgeon, M. (1996). An Ecological Survey of Lake Alaotra and Selected Wetlands of Central and Eastern Madagascar in Analysing the Demise of the Madagascar Pochard Aythya Innotata. World Wide Fund for Nature, Antananarivo. 146 pp.

    Google Scholar 

  • Powers, J. E., and Abeare, S. M. (2009). Fishing Effort Redistribution in Response to Area Closures. Fisheries Research 99(3): 216–225.

    Article  Google Scholar 

  • R Development Core Team (2012). R: A language and environment for statistical computing, R version 2.14.2. R Foundation for Statistical Computing, Vienna, Austria.

  • Ramanampamonjy, J. R., Rasoamampionona Raminosoa, N., Andrianandrasana, H. T., Wilmé, L., Durbin, J., Lewis, R. E., Razafindrajao, F., Razafindramahatra, L., Randriamanampisoa, H., Randriamahefasoa, J., and Rakotoniaina, L. J. (2003). Information Sheet on Ramsar Wetlands: Lake Alaotra Wetlands and Catchment Basin. Durrell Wildlife Conservation Trust Madagascar, Antananarivo. 12 pp.

    Google Scholar 

  • Razanadrakoto, D. R., and Rafaliarison, J. (2005). Délimitation des Zones de Frai Dans le lac Alaotra. Durrell Wildlife Conservation Trust, Antananarivo. 45 pp.

    Google Scholar 

  • Ryan, G. W., and Bernard, H. R. (2006). Testing an Ethnographic Decision Tree Model on a National Sample: Recycling Beverage Cans. Human Organization 65(1): 103–114.

    Article  Google Scholar 

  • Salas, S., and Gaertner, D. (2004). The Behavioural Dynamics of Fishers: Management Implications. Fish and Fisheries 5: 153–167.

    Article  Google Scholar 

  • Simon, H. A. (1955). A Behavioral Model of Rational Choice. The Quarterly Journal of Economics 69(1): 99–118.

    Article  Google Scholar 

  • Smith, M. D. (2005). State Dependence and Heterogeneity in Fishing Location Choice. Journal of Environmental Economics and Management 50(2): 319–340.

    Article  Google Scholar 

  • Smith, M. D., and Zhang, J., (2007). Sorting models in discrete choice fisheries analysis. American Agricultural Economics Association Annual Meeting, 29 July - 1 August, 2007. Portland, Oregon. p. 31.

  • Swain, D. P., and Wade, E. J. (2003). Spatial Distribution of Catch and Effort in a Fishery for Snow Crab (Chionoecetes opilio): Tests of Predictions of the Ideal Free Distribution. Canadian Journal of Fisheries and Aquatic Sciences 60(8): 897–909.

    Article  Google Scholar 

  • Teh, L., Teh, L., and Meitner, M. (2012). Preferred Resource Spaces and Fisher Flexibility: Implications for Spatial Management of Small-Scale Fisheries. Human Ecology 40(2): 213–226.

    Article  Google Scholar 

  • Tidd, A. N., Hutton, T., Kell, L. T., and Padda, G. (2011). Exit and Entry of Fishing Vessels: An Evaluation of Factors Affecting Investment Decisions in the North Sea English Beam Trawl Fleet. ICES Journal of Marine Science 68(5): 961–971.

    Article  Google Scholar 

  • van Oostenbrugge, J. A. E., van Densen, W. L. T., and Machiels, M. A. M. (2001). Risk Aversion in Allocating Fishing Effort in a Highly Uncertain Coastal Fishery for Pelagic Fish, Moluccas, Indonesia. Canadian Journal of Fisheries and Aquatic Sciences 58(8): 1683–1691.

    Article  Google Scholar 

  • Vanden Bossche, J.-P., and Bernacsek, G. M. (1991). Source book for the inland fishery resources of Africa: Vol. 1. CIFA Technical Paper. No. 18.1, FAO, Rome. 240 pp.

  • Wallace, A. P. C. (2012). Understanding fishers’ spatial behaviour to estimate social costs in local conservation planning. PhD Thesis. Imperial College London, London. 333 pp.

  • Wallace, A. P. C., Milner-Gulland, E. J., Jones, J. P. G., Bunnefeld, N., Young, R., and Nicholson, E. (2015). Quantifying the Short-Term Costs of Conservation Interventions for Fishers at Lake Alaotra, Madagascar. PLoS ONE. doi:10.1371/journal.pone.0129440.

  • Welcomme, R. L. (2001). Inland Fisheries: Ecology and Management. Fishing News Books, Blackwell Science, UK. 358 pp.

    Book  Google Scholar 

  • Welcomme, R. L. (2011). Review of the state of world fishery resources: Inland fisheries. FAO Fisheries and Aquaculture Circular. No. 942, Rev. 2, FAO Inland Water Resources and Aquaculture Service, Fishery Resources Division, Rome. 110 pp.

  • Welcomme, R. L., Cowx, I. G., Coates, D., Béné, C., Funge-Smith, S., Halls, A., and Lorenzen, K. (2010). Inland Capture Fisheries. Philosophical Transactions of the Royal Society, B: Biological Sciences 365: 2881–2896.

    Article  Google Scholar 

  • Wilen, J. E., Smith, M. D., Lockwood, D., and Botsford, L. W. (2002). Avoiding Surprises: Incorporating Fisherman Behavior into Management Models. Bulletin of Marine Science 70(2): 553–575.

    Google Scholar 

  • Wilmé, A. L. (1994). Status, Distribution and Conservation of two Madagascar Bird Species Endemic to Lake Alaotra: Delacour’s Grebe Tachybaptus rufolavatus and Madagascar Pochard Aythya innotata. Biological Conservation 69(1): 15–21.

    Article  Google Scholar 

Download references


We thank all participants in this study, particularly the fishers of Anororo and field assistants Joachin Randriarilala, Solofoniaina Esperant Rakotonisainana, Luhanaud Andriamiarivola, Rado Zilia Randriamihamina, and Mr Rabemanisa. Funding was provided by the UK Economic and Social Research Council and Durrell Wildlife Conservation Trust (DWCT). Invaluable in-country logistical support was provided by the DWCT Madagascar team in Antananarivo and Ambatondrazaka. We also thank the Service Régional de la Pêche et des Ressources Halieutiques, the Ministère des Eaux et Forêts, and the Federation of Fishers in Ambatondrazaka for their support, as well as the Ministry of Environment and Forest for research permits. Ethics approval for the research was granted by Imperial College Research Ethics Committee. EN acknowledges the support of a Marie Curie Fellowship. This paper is a contribution to Imperial College’s Grand Challenges in Ecosystems and the Environment initiative. The data underlying this paper are available on request from the Lead Author.

Author information

Authors and Affiliations


Corresponding authors

Correspondence to Andrea P. C. Wallace or E. J. Milner-Gulland.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Fig. S1
figure 3

Proportions of catch and effort observed across all gear types at fishing locations in Lake Alaotra, calculated for all fishers who also participated in background interviews (n = 788 catch interviews with 151 individual fishers). The pattern shown is similar to Fig. 2a and so justifying the use of the background interview dataset for all future analysis. Catch was measured as total weight caught and effort was measured as total number of hours spent fishing at the location. Solid circles represent fishing locations within restricted areas; open circles represent locations within non-restricted areas. The dotted line represents the 1:1 prediction of IFD. (JPEG 17 kb)

High Resolution Image (TIF 673 kb)

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Wallace, A.P.C., Jones, J.P.G., Milner-Gulland, E.J. et al. Drivers of the Distribution of Fisher Effort at Lake Alaotra, Madagascar. Hum Ecol 44, 105–117 (2016).

Download citation

  • Published:

  • Issue Date:

  • DOI:


  • IFD
  • Fisher spatial behaviour
  • Multi-habitat fishery
  • Adaptation
  • Risk
  • Madagascar