Jungle fishing adventures in French Guiana

Hot work fishing in the jungle

by Bill Beaumont,
GWCT Senior Fisheries Scientist

French Guiana (FG) is an overseas department and region of France located on the northeast Atlantic coast of South America. It borders Brazil to the south, and Suriname to the west.

The country has a very low population density of only 3 inhabitants per km2, with half of the population living in the metropolitan area of Cayenne, its capital. The country is the largest (by area) overseas department of France. As an overseas region, it is inside the European Union, and its official currency is the euro.

As the country is an overseas department of France the authorities are examining the status of the rivers and fish with regard to the Water Framework Directive that applies to Europe and calls for all rivers to achieve good or excellent status. Many of the rivers in FG are very low conductivity (that is they do not conduct electricity very well) this makes sampling with electric fishing difficult.

Dense jungle and amazing trees

Therefore fish sampling in FG has historically been carried out by using Rotenone to poison the fish in small creeks. Electric fishing has been tried but the method has not been very successful (Allard et al. 2014). However, the use of Rotenone is now subject to review by the French authorities and is likely to be banned due to the lethal effects.
 
Due to the links between GWCT and the French INRA scientists arising from the MORFISH project (who are also involved with fish research in FG) I was asked to review the methods used by Allard et al. (2014).

I found that whilst their equipment was moderately successful at 50 uS.cm-1, at lower conductivity fishing efficiency declined. I carried out some theoretical calculations to determine the power settings that would be needed to improve catch rate at the lower conductivities (Power Transfer Theory calculations) and found that in order to keep the same (50 S.cm-1 water) capture efficiency in 20 uS.cm-1 water, the output voltage would need to be increased to 1061 V DC.

These settings are beyond the capabilities of the equipment used by the FG researchers. However, I have had success in fishing low conductivity streams in UK using pulsed DC at voltages between 500 and 600 volts and have access to equipment that will fish up to 990 volts.
 
In order to demonstrate fishing using pulsed DC electric fishing methods I joined a small team of  five researchers from both Hydreco-Guyane and INRA in FG between 3rd to 9th June. We were based in the Hydreco-Guyane field research station at Petit Saut, where there is a large hydro-electric dam producing 116 MW of power. Creating the dam inundated 3,500,000 Hectares of forest and the research station has been set up to monitor the on-going impacts of the dam on the ecology of the area.



Home for four days

Accommodation was basic (tin roof and no walls, but did have a kitchen area, showers and electricity). Sleeping arrangements were hammocks slung in the single main room. The good thing about this arrangement was hearing the howler monkey chorus on the first morning, the bad was that the rain on the tin roof was deafening for much of the visit. Another downside was wearing chest waders, thick rubber gloves and a 15 Kg backpack electric fishing machine in 30 degree heat and 95% humidity!

Suitable sites in the forest were identified and enclosed with stop nets. We then carried out a 3-catch depletion population estimate of the reaches using electric fishing, followed by using rotenone to capture any remaining fish.

Most of the fish within the reach were small (<10 cm) and this often made seeing them in the slightly turbid water difficult. At some of the sites we were also working on the limits of the equipment (using 990 volts, 100 Hz pulsed DC). Estimates of the population by electric fishing were not very accurate (about 50 % of the ‘true’ population).

Exotic - but small fish

At another, smaller and slightly more conductive, site the total population (electric fishing population estimate + additional fish sampled by Rotenone) was within the estimate from the electric fishing data. The number of different species caught by electric fishing (which is the metric used to assess site status) was very good when compared with the total species caught by both methods.

At the largest site we fished, electric fishing caught 20 species and the Rotenone added two more (all this in just 25m of river!). At another site the electric fishing caught more species than the Rotenone including a species that is rarely caught by Rotenone.

Conclusions of using electric fishing methods were that it showed promise as a non-lethal method of assessing fish species composition in the low conductivity waters present in the area. In addition, in smaller streams above 20-25 uS.cm-1 it could provide population estimates of fish numbers.

For the future, I am going to explore ways in which the method can be refined for the difficult conditions in the streams and hope to return to further test the method. If the method does prove effective it will have implications for fish sampling in other low conductivity areas of South America (and elsewhere) where electric fishing is currently considered not feasible.

References: Allard et al. (2014). Fisheries Management and Ecology, 21, 234–243