Kähler Sigma Models and Supersymmetric Target Geometry

A four-dimensional rigid $\mathcal N=1$ theory of chiral multiplets with a two-derivative D-term is a nonlinear sigma model whose scalar target is Kähler. The superspace potential $K(\Phi,\Phi^\dagger)$ produces the metric $g_{i\bar j}=\partial_i\partial_{\bar j}K$; component covariance then forces the fermion connection and a curvature four-fermion interaction. The potential $K$ is only local and may change by a Kähler transformation between patches, while the metric and action are global. Stronger conclusions such as hyperkähler or special Kähler geometry require additional supersymmetry and a specified multiplet sector.

Required background. Supersymmetric Action Principles and Component Reduction supplies the chiral expansion and auxiliary workflow. Bundle Connections, Curvature, Gauge Transformations, and Bianchi Identities supplies coordinate-covariant differentiation.

Helpful background. Smooth Manifolds, Tangent and Cotangent Bundles, and Tensor Fields is a compact geometry repair route.

Chiral superfields are local complex coordinates

Let $z^i$ be the scalar components of chiral superfields $\Phi^i$. On a target patch $U\subset\mathcal M$, take

$$S=\int d^4x\left[ \int d^4\theta\,K(\Phi,\Phi^\dagger) +\left(\int d^2\theta\,W(\Phi)+\text{h.c.}\right) \right].$$

Reality of $K$ gives a Hermitian tensor

$$g_{i\bar j}=K_{i\bar j} =\frac{\partial^2K}{\partial z^i\partial\bar z^{\bar j}}.$$

Unitarity of the two-derivative theory requires $g_{i\bar j}$ to be positive definite on the field region being used. The associated two-form

$$\omega=i g_{i\bar j}\,dz^i\wedge d\bar z^{\bar j}$$

is closed because mixed derivatives commute: $d\omega=0$. Thus the target is Kähler on every nonsingular patch. This is the geometric content of the supersymmetric D-term, first emphasized in Zumino 1979, pp. 203–206.

The Kähler potential is not a scalar observable. On an overlap,

$$K_{(a)}(z,\bar z) =K_{(b)}(z,\bar z)+F_{ab}(z)+\bar F_{ab}(\bar z),$$

with $F_{ab}$ holomorphic. The metric is unchanged, and the extra full superspace integral vanishes up to a spacetime derivative. Global well-definedness therefore belongs to the patching data of $K$, not to a single preferred potential. A later supercurrent question is more restrictive: an improvement involving $K$ may fail to be global even though the sigma-model action is perfectly well-defined.

Component reduction reconstructs the connection and curvature

Define the Kähler connection

$$\Gamma^i{}_{jk}=g^{i\bar l}\partial_jg_{k\bar l}, \qquad D_\mu\psi^i =\partial_\mu\psi^i+\Gamma^i{}_{jk}\partial_\mu z^j\psi^k.$$

To keep the sign convention reproducible, define the four-index tensor used below by

$$\mathscr R_{i\bar j k\bar l} =\partial_i\partial_{\bar j}g_{k\bar l} -g^{m\bar n} (\partial_i g_{k\bar n}) (\partial_{\bar j}g_{m\bar l}).$$

Some geometry texts define the Riemann tensor with the opposite overall sign; the coefficient in the action must be translated together with that definition. After combining the noncovariant auxiliary terms into

$$\widehat F^i =F^i+\frac12\Gamma^i{}_{jk}\psi^j\psi^k,$$

the off-shell component action can be organized as

$$\begin{aligned} \mathcal L={}& g_{i\bar j}\left( \partial_\mu z^i\partial^\mu\bar z^{\bar j} +i\bar\psi^{\bar j}\bar\sigma^\mu D_\mu\psi^i +\widehat F^i\widehat{\bar F}^{\bar j} \right)\\ &+\frac14\mathscr R_{i\bar j k\bar l} \psi^i\psi^k\bar\psi^{\bar j}\bar\psi^{\bar l} +\widehat F^iW_i -\frac12\nabla_iW_j\,\psi^i\psi^j +\text{h.c.}, \end{aligned}$$

where

$$\nabla_iW_j=\partial_iW_j-\Gamma^k{}_{ij}W_k.$$

Terms can be redistributed between $\widehat F$ and the fermion bilinears by an auxiliary-field redefinition; the displayed covariant form fixes that choice. Component derivations in another curvature convention must reproduce the same target-coordinate scalar action and four-fermion amplitude. A full superspace reduction is given in Weinberg 2000, §26.8, pp. 102–106.

Eliminating the covariant auxiliary gives the bosonic potential

$$V_F=g^{i\bar j}W_i\bar W_{\bar j}.$$

The fermion mass/Yukawa tensor is the covariant Hessian $\nabla_iW_j$, not the coordinate-dependent second partial derivative alone. At a supersymmetric vacuum $W_i=0$, the connection term vanishes there, but it is needed away from the vacuum and under nonlinear field redefinitions.

Coordinate changes provide an independent check

Under a holomorphic change $z^i\mapsto z^{\prime a}(z)$,

$$\psi^{\prime a} =\frac{\partial z^{\prime a}}{\partial z^i}\psi^i.$$

$\partial_\mu\psi^i$ is not a target vector; the connection term in $D_\mu\psi^i$ cancels the second derivative of the coordinate map. Similarly, $\partial_iW_j$ is not a tensor, whereas $\nabla_iW_j$ is. The curvature term is then forced by supersymmetry and coordinate covariance. These transformation tests are independent of the original Grassmann expansion and catch many component errors.

The bosonic energy gives another check. For a static configuration with $W=0$,

$$E=\int d^3x\,g_{i\bar j}\, \boldsymbol\nabla z^i\mathbin{\cdot} \boldsymbol\nabla\bar z^{\bar j}\geq0$$

when the target metric is positive. A coordinate patch in which the matrix $g_{i\bar j}$ degenerates does not automatically signal a physical singularity; one must test whether another patch is regular and whether extra low-energy degrees of freedom enter.

The projective line illustrates patching

On a patch of $\mathbb{CP}^1$, take the Fubini–Study potential

$$K_{(z)}=f^2\log\left(1+\frac{|z|^2}{f^2}\right), \qquad g_{z\bar z}=\left(1+\frac{|z|^2}{f^2}\right)^{-2}.$$

On the patch $w=f^2/z$,

$$K_{(w)} =f^2\log\left(1+\frac{|w|^2}{f^2}\right),$$

and on the overlap the two potentials differ by a holomorphic plus antiholomorphic logarithm after the coordinate substitution. The metric and $\omega$ patch smoothly, while neither $K_{(z)}$ nor $K_{(w)}$ is global. Consequently a component calculation performed in one chart is trustworthy only while the field image stays in that chart, or after the patch transition has been included.

In four spacetime dimensions the coupling scale $f$ makes this nonlinear model an effective field theory: expanding $K$ generates interactions with negative-dimension coefficients. Curvature controls interaction strength, and the derivative expansion fails when momenta or field excursions probe the target’s cutoff or omitted degrees of freedom. Kähler geometry is a symmetry constraint, not a claim of ultraviolet completeness.

More supersymmetry imposes more geometry—but only in a named setting

The safe dimension-sensitive statements are:

• rigid four-dimensional $\mathcal N=1$ chiral sigma models have Kähler targets;
• rigid four-dimensional $\mathcal N=2$ hypermultiplet sigma models have hyperkähler targets under the standard two-derivative assumptions;
• the Coulomb-branch vector-multiplet geometry of four-dimensional $\mathcal N=2$ theories is rigid special Kähler, a different sector from a hypermultiplet target; and
• two-dimensional $(2,2)$ models with chiral multiplets give Kähler targets, while twisted or semichiral multiplets and torsion lead to generalized Kähler structures.

One must not infer “hyperkähler” merely from the word supersymmetric. The number of supercharges, spacetime dimension, multiplet type, torsion, and off-shell assumptions are part of the theorem. The classification logic and its scope are reviewed in Álvarez-Gaumé and Freedman 1981, pp. 443–451.

Common pitfalls

Treating K as global. The action may be global even when every Kähler potential is patchwise. This distinction becomes a real obstruction for some supercurrent improvements.

Using $W_{ij}$ as a tensor. Away from normal coordinates or a supersymmetric critical point, the covariant Hessian is $\nabla_iW_j$.

Equating a coordinate singularity with a new particle. First test the metric in another patch and invariant curvature. Only then ask whether a genuine singular locus requires additional low-energy fields.

Exercises

1. Verify the Kähler transformation. Show that $K\mapsto K+F(z)+\bar F(\bar z)$ leaves $g_{i\bar j}$ unchanged.

Solution

Because $F$ is holomorphic, $\partial_i\partial_{\bar j}F=0$, and similarly for $\bar F$. Hence $g'_{i\bar j}=\partial_i\partial_{\bar j}K'=g_{i\bar j}$.

2. Expand the $\mathbb{CP}^1$ metric. Find the first interaction in $g_{z\bar z}$ near $z=0$.

Solution

Using $(1+u)^{-2}=1-2u+O(u^2)$,

$$g_{z\bar z}=1-\frac{2|z|^2}{f^2}+O(|z|^4/f^4).$$

The scalar kinetic term therefore contains $-2|z|^2\partial z\partial\bar z/f^2$. Its coefficient has dimension $-2$, making the four-dimensional EFT character explicit.

Where the geometry continues

Gauge–Matter Systems adds a Hamiltonian group action and moment maps. F- and D-Flatness and Gauge Quotients constructs vacuum spaces, while Moduli-Space Metrics and Quantum Corrections separates protected complex structure from generally unprotected metrics.

References

• Álvarez-Gaumé, Luis, and Daniel Z. Freedman. “Geometrical Structure and Ultraviolet Finiteness in the Supersymmetric Sigma Model.” Communications in Mathematical Physics 80, no. 3 (1981): 443–451. DOI.
• Weinberg, Steven. The Quantum Theory of Fields, Volume III: Supersymmetry. Cambridge: Cambridge University Press, 2000, §26.8, pp. 102–106. DOI.
• Zumino, Bruno. “Supersymmetry and Kähler Manifolds.” Physics Letters B 87, no. 3 (1979): 203–206. DOI.