a bit more clarifying comments about energy_total_modified

src/equations/bbm_1d.jl

@@ -18,7 +18,8 @@ The BBM equation is first described in Benjamin, Bona, and Mahony (1972).
 The semidiscretization implemented here is developed in Ranocha, Mitsotakis, and Ketcheson (2020).
 It conserves
 - the total water mass (integral of ``h``) as a linear invariant
-- the total modified energy
+- a quadratic invariant (integral of ``\eta^2 + \eta_x^2``), which is called here [`energy_total_modified`](@ref)
+  (and [`entropy_modified`](@ref)) because it contains derivatives of the solution
 
 for periodic boundary conditions.
 

src/equations/equations.jl

@@ -314,7 +314,11 @@ contain additional terms compared to the usual [`energy_total`](@ref).
 For example, for the [`SvaerdKalischEquations1D`](@ref) and the
 [`SerreGreenNaghdiEquations1D`](@ref), it contains additional terms
 depending on the derivative of the velocity ``v_x`` modeling non-hydrostatic
-contributions.
+contributions. For equations, which are not `AbstractShallowWaterEquations`,
+the modified total energy does not have to be an extension of the usual
+[`energy_total`](@ref) and does not have to be related to a physical energy.
+However, it is still a conserved quantity and contains derivatives of the
+solution.
 
 `q_global` is a vector of the primitive variables at ALL nodes.
 `cache` needs to hold the SBP operators used by the `solver` if non-hydrostatic