Freshwater fisheries conservation can increase biodiversity

doubt, by answering questions about the species (colour, typical size, etc). This data allows us to calculate species richness, a count of the number of species caught in the last 12 months, considered to be a good indicator of fisheries’ health [19]. This approach is similar to the one used in [34] to monitor a smaller number of threatened Mekong megafauna, also in Lao PDR. There are several potential shortcomings to this approach, which are likely to be less impor- tant in the context of evaluating the impact of a conservation program. The first is that we do not measure species that are not caught or consumed, either because they are not part of the local diet or because they are protected. It is however unlikely that we would find large differ- ences in food habits between treatment and control villages in this dimension given that we work in a relatively small area of the country. The second is that differences in fishing practices (eg, restrictions on the use of nets with smaller mesh) may change the diversity of species caught. Existing studies that attempt to mea- sure the effect of sampling technique on biodiversity seem to suggest greater diversity in fish- ing gear (ie, sampling techniques) leads to higher estimates of species richness (see [35, 36]). Although this is likely a topic that would benefit from more empirical studies, it is unlikely to be a major driver of any difference between treated and control villages: as shown in the next section (Table 1), ComFishIII is not associated with a much higher incidence of restrictions on fishing gear.

One additional question is that recall data is potentially fraught with problems, given that memory is limited. However, the importance of fishing for the respondents (which is balanced across both types of villages) suggests that there is no reason to find a systematic difference in recall between treated and control units. Finally, the fact that the research team was not Table 1. Control vs ComFishIII villages. Variable Control ComFishIII Control—ComFishIII mean SD mean SD mean s.e. p-value A: village characteristics N households 103.1 55.79 112.5 62.89 -9.4 21.02 0.657 N households fishing 77.69 35.54 93.50 50.01 -15.81 15.33 0.312 N households fishing: main income 2.44 7.71 0.63 1.62 -1.81 1.97 0.372 N households farm 85.7 39.34 98.2 49.84 -12.5 15.87 0.438 N households with land 82.75 37.04 96.57 49.95 -13.81 7.75 0.382 N ethnic groups >5% 1.813 0.981 1.875 1.258 -0.062 0.40 0.877 N ethnic groups >10% 1.750 1.000 1.625 0.885 0.125 0.33 0.711 Share households main ethnicity 80.17 22.94 80.69 22.32 -0.52 8.00 0.950 Share households 2nd main ethnicity 10.94 12.46 12.48 14.25 -1.54 4.73 0.751 District/ Province HQ 0.00 0.00 0.06 0.25 -0.06 0.06 0.331 Resettled village 0.188 0.403 0.188 0.403 0.000 0.14 1.000 Village to be resettled 0.125 0.342 0.125 0.342 0.000 0.12 1.000 B: management rules FCZ 0.50 0.52 1.00 0.00 - 0.50 0.13 0.001 Restrictions - gear 0.88 0.34 1.00 0.00 -0.12 0.09 0.15 Restrictions - species 0.56 0.51 0.88 0.34 -0.32 0.15 0.05

Restrictions - quantity 0.25 0.45 0.44 0.51 -0.19 0.17 0.28 Monitoring 0.50 0.52 0.94 0.25 -0.44 0.14 0.005 Progressive penalties 0.75 0.45 1.00 0.00 -0.25 0.11 0.033 p-value for two sided t-test of difference in means (n = 32 villages, of which 16 participate in the Community Fisheries (ComFishIII) program. https://doi.org/10.1371/journal.pone.0233775.t001 PLOS ONE FCZ and biodiversity PLOS ONE | https://doi.org/10.1371/journal.pone.0233775 May 27, 2020 7 / 14