Dataset 781
Structure of small tropical islands freshwater fish and crustacean communities: a niche- or dispersal-based process?
Realm: Freshwater
Climate: Tropical
Biome: Small river ecosystems Central latitude: -17.609970
Central longitude: 50.929020
Duration: 14 years, from 2004 to 2017
Climate: Tropical
Biome: Small river ecosystems Central latitude: -17.609970
Central longitude: 50.929020
Duration: 14 years, from 2004 to 2017
3970 records
44 distinct species
Across the time series Sicyopterus lagocephalus is the most frequently occurring speciesMethods
A total of 86 and 61 sites were sampled from 2004 to 2017 along the downstreamupstream gradient of the main rivers of Mayotte (Figure 1b) and Reunion (Figure 1c), respectively. In Mayotte, sampling was repeated between one and seven years at each site (Figure 1b). As a result, 161 surveys were conducted with 12 to 23 surveys each year from 2008 to 2012, in 2014, 2015, and 2017, six in 2013 and 40 in 2016. In Reunion, sampling was performed from one to 12 different years at each site (Figure 1c) resulting in 366 surveys (i.e., 29 to 35 surveys each year from 2004 to 2014, four in 2015 and 22 in 2016). Fish and crustaceans were sampled using a portable electro-shocker (Deka, 3,000 or Hans Grassl, IG 200). The sampled river length was at least ten time the mean river width. The upstream and downstream limits of the sites were blocked by a net or a physical barrier (i.e., waterfall and weir) to limit fish and crustacean emigration and immigration. To ensure sampling effectiveness and representativeness, two sampling procedures were applied depending on the river width (Olivier et al., 2004). When the river width was less than five meters, the whole area of the river site was prospected by fishermen moving upstream (i.e., complete sampling). When the river width was greater than five meters, the sampling was stratified by hydromorphic units (i.e., cascade, rapid, riffle, run, shoal, and pool). In this case, the hydromorphic units were firstly identified based on mean water velocity and depth following Malavoi and Souchon (2002), and their surfaces were estimated using a laser meter (Leica DISTO D5). Then, several sampling units ranging from 10 to 25 m2 were distributed within each hydromorphic unit proportionally to its surface, ensuring that a minimum of 200 m2 was sampled across the whole site. Finally, each sampling unit was prospected by moving upstream using the electro-shocker. For both sampling procedures, fish and crustacean were captured by a minimum of three operators equipped with hand nets (width 50 cm, maximum mesh size four mm) and positioned immediately downstream to the operator manipulating the electro-shocker. The abundance of each taxon was estimated based on capture per effort units (CPUE, ind.m-2) whatever the fishing procedures. For the complete sampling procedure, the abundance of one taxon was the number of individuals divided by the fished area (m2). For the stratified sampling procedure, the abundance of each taxon was first calculated within each hydromorphic unit by averaging the number of individuals captured in each sampling unit divided by its area. Then, the abundance of one taxon in a sampling site was obtained by the mean of the abundance in all hydromorphic units weighted by their relative surface proportion in the river site.” [Extracted from Lagarde et al 2021] Additional: abundance units are number of individuals / 100 m2. Information in "grain size": total average fished area (254 m2) and total average fished river length (65 m) across sites. In Mayotte, the average fished area per site is 201 m2, and the average fished length is 63 m. In Reunion, the average fished area is 274 m2, and the average fished length is 66 m.Citation(s)
.
In (Eds.),
(p. ).
:
.
,
(),
.
