Assume that $J$ is an almost complex structure on torus $\mathbb{T}^2$. Let $X$ be a non vanishing vector field on the torus. Let $g$ be a Riemannian metric with corresponding $LC$ connection $\nabla$.

We define the following differential operator on $\chi^{\infty}(\mathbb{T}^2)$, the space of smooth vector fields on tori.

$$ D(Y)=( \nabla_X .\nabla_X +\nabla _{JX}. \nabla_{JX})(Y)$$

Is $D$ an elliptic operator? Does its index depend on $J,X$ and $g$?

When $\nabla, g, J$ are the standard structure on tori, is $D$ the standard Laplacian operator on $\Omega^1(\mathbb{T}^2)\simeq \chi^{\infty}(\mathbb{T}^2)$?(The later equivalency is done via metric $g$).


$D$ is an elliptic operator, as its symbol is given by $$\sigma(D)(p,\xi)=(\xi(X_p))^2+(\xi(JX_p))^2$$ which is non-vanishing for $0\neq\xi\in T_pT^2.$ The index is always 0, as your operator can be deformed continuously (via a family of second order elliptic operators) to the standard Laplacian (acting on pairs of functions). To prove this claim, take a smooth deformation of $J$ and $g$ to the standard $J_0$ and $g_0$ on $T^2$, as well as a scaling of $X$ to a vectorfield $X_0$ of constant length $1$ with respect to $g_0$. At that point, your operator has the same symbol as the standard Laplacian, and you can further deform by scaling the lower order terms to 0 to obtain the standard Laplacian.


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.