I am looking at Bernoulli's equation formulated in two dimensions, which reads $$\frac{v^2}{2}+c_p\theta_0\pi+gz \equiv \mathrm{const.}$$ along a given streamline $\gamma$, where $v=|\mathrm{v}|$ for the flow field $\mathrm{v}(x,z)$ which is in steady state (i.e. $\frac{\partial\mathrm{v}}{\partial t}\equiv 0$) and $\pi=\left(\frac{p}{p_0}\right)^\frac{R}{c_p}$ is the Exner function. The potential temperature $\theta$ is assumed to be constant equal to $\theta_0$ along $\gamma$. Now if we assume $\pi$ to be constant along $\gamma$ as well, we end up with an upper bound to the maximum upward displacement an air parcel with initial velocity $v_0$ and initial height $z_0$ can attain, namely $$z_{crit}= z_0+\frac{v_0^2}{2g},$$ since $$v^2=0 \text{ if and only if } z-z_0=\frac{v_0^2}{2g}.$$ But is the assumption that $\pi$ is constant along $\gamma$ a physically useful one? I cannot find a discussion concerning this in any textbook on meteorology I currently posses, neither in internet resources. Any advice/resource would be appreciated.


1 Answer 1


If you assume both hydrostatic and pure geostrophic balance, that is a valid assumption.

In Einstein notation, $$u_i=-\frac{1}{f \rho}\frac{\partial P}{\partial x_j}\epsilon_{ij3}$$

If we look at the equation for the streamline:$$u_i=-\frac{\partial \psi}{\partial x_j}\epsilon_{ij3}$$, then we can see that

$$-\frac{\partial \psi}{\partial x_j}\epsilon_{ij3}=-\frac{1}{f \rho}\frac{\partial P}{\partial x_j}\epsilon_{ij3}$$

which simplifies down to $$\frac{\partial \psi}{\partial x_j}=\frac{1}{f \rho}\frac{\partial P}{\partial x_j}$$. Thus, for isochoric (incompressible) geostrophic flow on an f-plane, the streamfunction only depends solely on pressure.

  • $\begingroup$ Thanks for another answer. I have a few remarks: The geostrophic flow equation should have the Coriolis parameter $f$ in the demoninator, right? I do not see how you arrive at this formulation of the streamline equation... this reads to me as $u_1=\frac{\partial \psi}{\partial x_2}, u_2=-\frac{\partial \psi}{\partial x_1}$, and $u_3=0$, with $\epsilon_{ijk}$ the Levi-Civita symbol? $\endgroup$
    – chriss
    Apr 15, 2020 at 7:59
  • $\begingroup$ Yes, I will edit that. My bad. Streamlines are isopleths of the streamfunction (glossary.ametsoc.org/wiki/Streamfunction). I realize now that I have the order of the Levi-Civita symbol backwards (see link in comment). I'll edit my answer to reflect those corrections. $\endgroup$ Apr 15, 2020 at 14:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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