Habitat is often characterized by more than one variable. For example, per section 4.1.1, little minnow need shallow water for spawning (no deeper than 3 ft), which is an expression of habitat suitability based on a single variable, depth. If little minnow also needed water moving at more than 0.5 ft per second, ideally between 1 and 2 ft per second, and no more than 3 ft per second for spawning, then habitat suitability would be based on two variables, depth and velocity. Habitat suitabilities are typically expressed as a number from 0 (wholly unsuitable) to 1 (perfectly suitable).
In HEC-EFM, “multivariate analyses” numerically combine suitabilities for different variables into a single measure of suitability, which is useful for habitat mapping and when tallying habitat provided. In EFM, multivariate analyses are applied only for compound flow regimes (2-dimensional input).
Computations for multivariate analyses are done last during an EFM run because input to multivariate analyses are generated while computing results for individual flow regime-relationship pairings. Involving more than one flow regime-relationship pairing makes multivariate analyses quite different than other EFM analyses, which are almost always done independently for each flow regime-relationship pairing.
Using 2-dimensional data and multiple pairings can make multivariate analyses complex. This chapter details creation of multivariate analyses and related numerical options, validations to ensure viable analyses, and output.