If I was interested in deriving an equation for the conservation of momentum for a fluid, I could write down an expression for the change in momentum density of a fluid point using the Reynolds transport theorem:

$$\frac{\partial \rho \vec{v}}{\partial t} + \nabla \cdot (\rho \vec{v} \otimes \vec{v}) = \vec{f}$$

I could also try to write down Newton's law from a Lagrangian perspective for a parcel of fluid:

$$\frac{D(\rho \vec{v})}{Dt} = \vec{f}$$

Transforming this into a Eulerian perspective:

$$\frac{\partial \rho \vec{v}}{\partial t} + \nabla (\rho \vec{v})\cdot\vec{v} = \vec{f}$$

This expression differs from the expression derived from the Reynolds transport theorem by a $\rho \vec{v}(\nabla \cdot \vec{v})$ term—both expressions are the same for an incompressible flow field, but not when the flow is compressible.

Did I make a mistake in applying Newton's law, or is there a valid reason for the discrepancy?


If the left hand side is differentiated property using the product rule for differentiation, we obtain $$\frac{\partial \rho \vec{v}}{\partial t} + \nabla \cdot (\rho \vec{v} \otimes \vec{v}) =\rho\frac{\partial \vec{v}}{\partial t}+\vec{v}\frac{\partial \rho}{\partial t}+\vec{v}\cdot\nabla(\rho \vec{v})+\rho (\vec{v}\cdot \nabla) \vec{v}$$The middle two terms drop out because of the continuity equation, so we are left with $$\frac{\partial \rho \vec{v}}{\partial t} + \nabla \cdot (\rho \vec{v} \otimes \vec{v}) =\rho\frac{\partial \vec{v}}{\partial t}+\rho (\vec{v}\cdot \nabla) \vec{v}=\rho\frac{D\vec{v}}{dt}$$

  • $\begingroup$ Does that imply that the Lagrangian expression for Newton’s law I wrote above is incorrect? I would’ve expected it to include the density within the derivative, like the “traditional” analogue. $\endgroup$ – aghostinthefigures Oct 14 '20 at 21:58
  • $\begingroup$ Your 2nd and 3rd equations are incorrect. Just write down the starting equation in cartesian component form, and you will see that what I have done is correct. $\endgroup$ – Chet Miller Oct 14 '20 at 22:06
  • $\begingroup$ Oh don’t worry, I agree with you—I asked because I saw a derivation of Bernoulli’s law using that second equation, and was unsure if it was actually sensible to use it in cases where the flow isn’t incompressible. $\endgroup$ – aghostinthefigures Oct 14 '20 at 22:16
  • $\begingroup$ As a follow-up, I just realized what my mistake was: the correct way to write down a Lagrangian formulation of Newton's law requires the inclusion of a volume element! $$\frac{D(\rho \vec{v} dV)}{Dt}$$ with the usual rules of differentiating volume elements produces the right answer—the same for conservation of mass. $\endgroup$ – aghostinthefigures Oct 15 '20 at 22:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.