The dynamics shaping Lake Victoria (LV) today encapsulate many of the key features of critical socio-ecological systems (SES) under stress in the early Anthropocene. They are the current background against which Phase I of MultiTip achieved several successes for a better understanding of tipping points and resilience of this globally relevant SES.

A new tipping point concept for Lake Victoria. We have in hand a mathematically rigorous, general, analytical, but also operationalized tipping point concept for the LV resource system. This concept, the subcritical bifurcation of system dynamics, was validated in the single-species Nile perch context, constituting a first capstone of the project (Marciniak-Czochra, 2021). Behind this deliverable stands an important shift that MultiTip has set into motion in modeling the LV resource system, moving from computational box models to a more analytical approach that emphasizes system characteristics like stable equilibria, regime shifts and tipping points over merely computed trajectories.

New methods of data-model integration. MultiTip has succeeded in a critical task, namely the integration of socio-ecological survey data and new analytical models of the resource system, using the spectrum slope of the fish size distribution as a management-relevant integration point. Using the model to structure four sets of survey data (Catch Assessment Survey, Frame Survey, Hydroacoustic Survey, Bottom-Trawl Survey) and using the survey data to calibrate the analytical model for the Nile perch with optimization and Bayesian Inference (Edwards et al., 2017; Miles, 2019), the project team is now able to adjudicate between conflicting survey evidence. This constitutes a breakthrough in the analysis and interpretation of size-based fish population data at LV and its rigorous integration into a numerical state-of-the-art stock model. It also has - in the spectrum slope - a validated handle on a key management indicator of population health and reproductive potential (Blanchard et al., 2014; Diekert et al., 2010), a critical factor for system stability (see Figure 1). Stock size distribution

Figure 1. Left Panel: Size distribution of the Nile perch population in Uganda, Tanzania, and Kenya from the hydroacoustic survey 2020. Right Panel: We identified the spectrum slope (top) as a key management indicator of the resilience of the fish stock. From the spectrum, we simulated how fishing selectivity (middle) translates to catches (bottom).

New models of size selectivity. MultiTip deepens into the outsized impact at LV of gear choice on the resilience of the population through modeled size selectivity of the gear (see Figure) (Gómez-Cardona, Kammerer, and Mrosso, 2022). Gear here means dagaa seine nets, Nile perch gillnets, longlines, bait, light attractors, chemicals, explosives, and all other relevant inputs used for catching fish. Gear and slot-size regulations are a characteristic feature of the LV fishery, and policy-makers and experts have been engaging in a continuous debate about their specific merits for some time now (Peter and van Zwieten, 2018). MultiTip succeeded in estimating the size selectivity of the LV Nile Perch fishery and therefore its impact on yield and population health (Gómez-Cardona, Kammerer, and Mrosso, 2022).

New evidence on the impacts of gear regulations. Policy-makers in Uganda, Kenya, and Tanzania have recognized the importance of protecting the juvenile fish stock for some time. The result is regulations about legal catch size (50 – 84 cm), about minimum legal longline hook sizes, and minimum gillnets’ mesh sizes in the Nile perch fishery (LVFO, 2016). MultiTip has been generating novel insights on how gear regulations impact on the resource system: Our gear selectivity analysis provides evidence that the gillnet fishing fleet is capturing increasingly larger fish and that this trend correlates with increased enforcement of the mesh size regulation (Gómez-Cardona, Kammerer and Mrosso, 2022). The hook fishing fleet, by contrast, is not experiencing any changes in the average size of the fish capture (idem) despite an increasing use of undersized hooks (Aloo et al, 2017). This evidence suggests that the mesh size regulation is effective in protecting juvenile fish stock while the hook size regulation appears to miss its intended effect.

Compliance subsidies as novel regulatory instruments. The current governance of LV’s resource system relies on a coercive regime favoring direct enforcement. In theory, this is supposed to deter illegal gear use as fishers consider the probability of detection and punishment. In reality, the system leads to sporadic, but harsh enforcement that generates limited deterrence at high cost, destroys assets (boats and gear, some of it legal), and regularly leads to fatalities among both fishers and enforcers (Kolding et al., 2014; Obiero et al., 2015; Cepic and Nunan, 2017). A field intervention has tested regulatory alternatives to the current detect-and-burn approach: This intervention at 20 landing sites in Tanzania showed that subsidies on legal gear have the potential to drive out illegal gear at acceptable cost (see Figure 2), but also evidence that soft measures that appeal to ‘good fishing practices’ are ineffective (Diekert et al. 2022). The field evidence provides a first proof-of-principle regarding the feasibility of compliance subsidies on legal gear as a novel tool. This success serves as a capstone result of Phase 1. At the same time, it is the basis for a scaled-up intervention in WP2 of Phase 2 to validate and generalize the results and to examine whether the governance can undergo a shift to a more cooperative regime.


Net panels

Figure 2: Left Panel: average legal sized net panel demand by price among Dagga fishers; Right Panel: average legal sized net panel demand by discount among Dagga fishers;

The surprising spatiality of the LV resource system. MultiTip initially adopted the literature’s common shortcut of characterizing the socio-ecological situation of the Lake in terms of averages. An unexpected outcome has been the recognition that the Lake’s spatial dimensions have some surprising socio-ecological consequences. Our spatial model of the Nile perch fishery shows that LV’s sheer size, combined with its shallowness, creates fish habitats protected from humans (see white patches in Figure 3). The reason is that these habitats are so distant from the nearest shore (some more than 70km) that artisanal fishing there is economically unprofitable given fishing technology, fuel cost, and fish prices (Gómez-Cardona, 2022). This model can explain the spatial pattern of most of the fleet fishing close to the shore, despite localized competition for fish (Peter, van Zwieten, 2018). It has demonstrated that the protected habitats can act as potential reserves, increasing the resilience of the LV resource system to both continued harvesting pressure and external shocks (Gómez-Cardona, 2022).

A universal tool for mental model elicitation. A key success has been the development of a universal tool for mental model elicitation. The M-Tool (van den Broek et al., 2021a) is an accessible offline application for tablets and an online-tool for other devices that allows participants to express their views and perceptions about causal relationships in their environment with minimal literacy and numeracy requirements. The tool has already been downloaded more than 100 times.

Disparate, yet convergent mental models among fishers. Van den Broek (2018) found a disparity in the perceptions among stakeholder about the causes of the decline in Nile perch catch, with important management implications. These were identified by stakeholders, field partners and the research team as a core area of interest. A follow-up study uncovered structural differences in perceptions of what drives the decline in Nile perch stock across stakeholders in the three riparian countries (Klein et al., 2021). An application of M-Tool to a sample of Tanzanian fishers supported its validity (van den Broek et al., 2021b), but also showed that fishers converge in their perception that fishers’ activities, in particular non-compliance with regulations, contributes to stock decline in a major way (van den Broek et al., under review).

M Tool

Figure 4: Left: Screenshot of the mapping screen of the M-Tool. The letters are placeholders for the system concepts the researchers would like to use for their study. Participants choose the concepts from the left panel and link these in the middle panel with the arrows in the right panel to demonstrate their perceptions of the interactions between these concepts. Right: Aggregate interview mental model in van den Broek et al., (2021b)


Asymmetric positive feedback effects in risk-taking among fishers. Risk is an essential part of fishing decisions and therefore determines anthropogenic pressure on the resource system. MultiTip investigated fishers’ risk behavior and found that there are positive feedback effects among fishers: When fishers learn that their peers are engaging in high-risk behavior, they increase their own risk-taking (see Figure 5). The feedback effect is asymmetric, however: Learning that their peers engage in low-risk behavior does not decrease risk-taking (Dannenberg et al. 2022). These results come from a lab-in-the-field experiment in which fishers took risky investment decisions under controlled conditions varying experience of good or bad luck and information about peer behavior.


Behavioral impacts of attribution to anthropogenic causes.  LV is a resource system that could be tipped into an adverse state by both natural (climate extreme events) and anthropogenic causes (overfishing; destructive fishing). We were able to demonstrate that it makes a measurable difference for user behavior whether they can attribute adverse system events to an anthropogenic cause (see Figure 6). If attribution to an anthropogenic cause is possible, then users respond to such events by reducing their anthropogenic stress on the systems, at a cost to themselves (Diekert et al. 2021). This result is the first rigorously derived evidence that attribution of adverse events to anthropogenic causes is behaviorally impactful. The evidence comes from a carefully designed experiment with over 3,000 participants in four treatment conditions. This breakthrough opens up the possibility of using information about causation strategically in the governance of the LV resource system.



Seitenbearbeiter: E-Mail
Letzte Änderung: 22.03.2022
zum Seitenanfang/up