Potential climate change induced modiications in mangrove ecosystems: a case study in Benin, West Africa

Abstract

Mangroves are one of the most threatened ecosystems globally. Likewise, they benefit many restoration efforts. However, these efforts have often disregarded potential effects of climate change on selected species and foreseen changes in physico-chemical conditions of mangrove ecosystems. This study aimed to model current and future climatic and physico-chemical conditions in occupied mangroves’ niche, in order to derive implications for successful long-term restorations. Presence records of mangroves’ indicator species and corresponding physico-chemical variables were collected. Bioclimatic data were obtained from Africlim. Kruskal–Wallis and Nemenyi pairwise tests for multiple comparisons were used to test the among sites (spatial) variations of climatic and physico-chemical variables within mangroves’ niche. Multiple linear regression (MLR) and artificial neural network (ANN) were used to build predictive models of salinity, dissolved oxygen, and conductivity; using most meaningful climatic variables to tropical mangroves—potential evapotranspiration, temperature seasonality, mean temperature of the warmest quarter, moister index of the moist quarter, and moister index of the driest quarter as predictors. Results showed that there are strong spatial variations of climatic and physico-chemical variables within mangroves’ niche. ANN outperformed MLR and was then used to predict trends in salinity, dissolved oxygen, and conductivity by year 2055. Based on foreseen trend in bioclimatic variables, conductivity, salinity, and dissolved oxygen will experiment significant changes under Representative Concentration Pathways 4.5 and 8.5, with most severe changes in case the later scenario occurs. Foreseen salinization of sites may be at the advantage of Avicennia germinans but to the prejudice of Rhizophora spp.