Canonical Correspondence Analysis (CCA) is a multivariate method designed to elucidate the relationships between biological assemblages of species and their environment. It extracts synthetic environmental gradients from ecological data sets, which are used to describe and visualize the differential habitat preferences (niches) of taxa through ordination diagrams. Unlike linear multivariate methods such as Partial Least Squares (PLS2), canonical correlation analysis, and redundancy analysis, CCA is better suited for unimodal functions of habitat variables. The paper discusses the key assumptions underlying CCA and focuses on interpreting CCA ordination diagrams. It also illustrates advanced uses, such as ranking environmental variables and statistical testing of effects, using a macroinvertebrate data set. The paper concludes with comparisons to other methods like correspondence analysis, discriminant analysis, PLS2, and co-inertia analysis. Additionally, a new method called CCA-PLS, which combines the strengths of CCA and PLS2, is proposed in an appendix. CCA has been widely used in aquatic sciences, particularly in limnology, marine biology, and palaeolimnology, for identifying environmental gradients and understanding how species respond to external factors.Canonical Correspondence Analysis (CCA) is a multivariate method designed to elucidate the relationships between biological assemblages of species and their environment. It extracts synthetic environmental gradients from ecological data sets, which are used to describe and visualize the differential habitat preferences (niches) of taxa through ordination diagrams. Unlike linear multivariate methods such as Partial Least Squares (PLS2), canonical correlation analysis, and redundancy analysis, CCA is better suited for unimodal functions of habitat variables. The paper discusses the key assumptions underlying CCA and focuses on interpreting CCA ordination diagrams. It also illustrates advanced uses, such as ranking environmental variables and statistical testing of effects, using a macroinvertebrate data set. The paper concludes with comparisons to other methods like correspondence analysis, discriminant analysis, PLS2, and co-inertia analysis. Additionally, a new method called CCA-PLS, which combines the strengths of CCA and PLS2, is proposed in an appendix. CCA has been widely used in aquatic sciences, particularly in limnology, marine biology, and palaeolimnology, for identifying environmental gradients and understanding how species respond to external factors.