Does the following second covariant (in terms of Kähler geometry) derivative of Kähler potential vanish? \begin{equation} K_{ij}\equiv\nabla_i\nabla_j K=0, \end{equation} \begin{equation} K_{i^*j^*}\equiv\nabla_{i^*}\nabla_{j^*} K=0? \end{equation} Indices represent complex coordinates ($A_i$ and $\bar{A}_i$) on Kähler manifold.

up vote 0 down vote accepted

The answer is in general No. Take e.g. the Fubini-Study Kaehler potential

$$K~=~\ln D, \qquad D~=~1+ Q,\qquad Q~=~\sum_{k=1}^n z^k \bar{z}^k, \tag{1}$$

with Hermitian metric

$$g_{\imath\bar{\jmath}} ~=~ \partial_{\imath} \bar{\partial}_{\bar{\jmath}}K~=~ \frac{\delta_{\imath\bar{\jmath}}}{D}-\frac{\bar{z}^{\imath}z^{\bar{\jmath}}}{D^2}~=~ D^{-1}\left(\delta_{\imath\bar{\jmath}}-\frac{\bar{z}^{\imath}z^{\bar{\jmath}}}{D}\right), \tag{2} $$

and inverse metric

$$ g^{\bar{\imath}\jmath}~=~D(\delta^{\bar{\imath}\jmath}+\bar{z}^{\bar{\imath}}z^{\jmath} ) , \tag{3}$$

and Hermitian Christoffel symbols

$$ \Gamma_{\imath\jmath}^{\ell} ~=~ \partial_{\imath}g_{\jmath\bar{k}}~g^{\bar{k}\ell}~=~-\frac{\bar{z}^{\imath}\delta_{\jmath}^{\ell}}{D}-\frac{\bar{z}^{\jmath}\delta_{\imath}^{\ell}}{D}. \tag{4}$$

The covariant derivative of the Kaehler potential is

$$ \nabla_{\ell}K~=~\partial_{\ell}K ~=~\frac{\bar{z}^{\ell}}{D}.\tag{5} $$

Now calculate the sought-for quantity

$$ \nabla_{\imath}\nabla_{\jmath}K ~=~ \nabla_{\imath}\partial_{\jmath}K ~=~\partial_{\imath}\partial_{\jmath}K -\Gamma_{\imath\jmath}^{\ell}~\partial_{\ell}K ~=~\frac{\bar{z}^{\imath}\bar{z}^{\jmath}}{D^2} ~\neq~0, \tag{6}$$ which does not vanish.

  • Then, as I asked in physics section, in application to supersymmetric theories, $\Box z$ and $\partial z\partial\bar{z}$ terms cannot be used interchangeably (box and partials WRT ordinary spacetime). And I don't understand how to get the latter term, without introducing the $\partial z\partial z$ and $\partial\bar{z}\partial\bar{z}$ terms. – Kosm Jun 19 '16 at 21:21
  • Why don't your i's and j's have dots? Are they sick? – Harry Wilson Jul 21 '16 at 9:58
  • Purely for aesthetic reasons to avoid having bars and dots stacked on top of each other. – Qmechanic Jul 21 '16 at 10:54

Your Answer


By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

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