Facing the need to diversify income sources due to declining productivity in cocoa plantations, cocoa farmers in Côte d’Ivoire are increasingly adopting aquaculture as an additional revenue stream, establishing fishponds in low-lying areas adjacent to or within their farms. Given the intensive use of chemical agricultural inputs in cocoa cultivation, this study aims to evaluate the ecological quality of water in fishponds integrated into cocoa farming systems using biomarkers analyses. Aquatic macroinvertebrates and microalgal flora were used as biological indicators to assess water quality in 6 fishponds integrated into cocoa plantations across two regions in Côte d’Ivoire: Méagui and Sinfra.
Water parameters were measured with a multiparameter probe. Ammonia (NH3) and phosphate (PO43-) levels were analysed with a photometer, while transparency was assessed via Secchi disk. Phytoplankton and macroinvertebrates were sampled using nets and sediment grabs. Water quality was evaluated using biotic indices, including Shannon diversity, Pielou’s equitability, biosedimentary index (IBS), planktonic index (IP), and macroinvertebrate families sensitivity to pesticides.
Results from physicochemical analyses indicate acceptable water characteristic for aquaculture. While, Shannon diversity and Pielou equitability indices from both algae and macroinvertebrates communities and also Biosedimentary Index (IBS) (4,6-5,46) classified water quality as marginal to poor.
Moreover, the Plancton Index (38-108) indicates average to poor ecological status of water. Macroinvertebrates families sensitive to pesticides were scarce (0–20%) in the whole ponds.
The integration of fishponds into cocoa agroecosystems presents economic opportunities but raises ecological concerns due to pesticide runoff. The prevalence of pollution-tolerant species and low proportions of sensitive macroinvertebrates underscore the cumulative impacts of agrochemical use. To mitigate risks, farmers should adopt integrated pest management strategies and explore organic alternatives to synthetic inputs.
Acknowledgement: The authors would like to thank the IRN ASACHA-IRD for sustain participation to WAS and D. Caruso from IRD for facilitation and advice.