Götsche–Kool virtual elliptic genus conjecture for rank 2 sheaf moduli spaces

Let SS be a smooth projective surface with b1(S)=0b_1(S)=0, pg(S)>0p_g(S)>0, KS0K_S\ne0, and with only Seiberg–Witten basic classes 00 and KSK_S. Let H,c1,c2H,c_1,c_2 be chosen so that there are no rank 22 strictly Gieseker HH-semistable sheaves on SS with Chern classes c1,c2c_1,c_2, and write M:=MSH(2,c1,c2)M:=M_S^H(2,c_1,c_2). Let vd(M)\operatorname{vd}(M) be its virtual dimension, and let Ellvir(M)\operatorname{Ell}^{\mathrm{vir}}(M) denote its virtual elliptic genus. Götsche–Kool's virtual elliptic genus conjecture. Ellvir(M)\operatorname{Ell}^{\mathrm{vir}}(M) is given by the coefficient of pvd(M)p^{\operatorname{vd}(M)} in ψS(p,q,y)\psi_S(p,q,y), with

ψS(p,q,y)=8(12L2(ϕ0,1))χ(OS)(2L4(2ϕ0,12ϕ0,32)L(2ϕ0,12)L2(2ϕ0,12ev(q12,y)ϕ0,12(q2,y2)+2ϕ0,122))KS2,\psi_S(p,q,y)=8\left(\frac{1}{2\mathsf L_2(\phi_{0,1})}\right)^{\chi(\mathcal O_S)}\left(\frac{2\mathsf L_4(2\phi_{0,\frac12}\phi_{0,\frac32})\mathsf L(-2\phi_{0,\frac12})}{\mathsf L_2\left(-2\phi_{0,\frac12}^{\operatorname{ev}}|_{(q^{\frac12},y)}-\phi_{0,\frac12}|_{(q^2,y^2)}+2\phi_{0,\frac12}^2\right)}\right)^{K_S^2},

where

L2(ϕ0,1)=L2(Ell(K3))12=(χ10(p2,q,y)p2Δ(q)ϕ2,1(q,y))12.\mathsf L_2(\phi_{0,1})=\mathsf L_2(\operatorname{Ell}(K3))^{\frac12}=\left(\frac{\chi_{10}(p^2,q,y)}{p^2\Delta(q)\phi_{-2,1}(q,y)}\right)^{\frac12}.

Here ϕ0,12\phi_{0,\frac12} and ϕ0,32\phi_{0,\frac32} are the quasi-Jacobi forms introduced in the paper, and L\mathsf L, L2\mathsf L_2, and L4\mathsf L_4 are the corresponding lift operators. The formula is a rank 22 analogue of the Dijkgraaf–Moore–Verlinde–Verlinde formula and is supported by the universal structure and calculations discussed in the paper, but remains conjectural.

Sources & referencesView supporting material

Primary source

Lothar Göttsche and Martijn Kool, “A rank 2 Dijkgraaf-Moore-Verlinde-Verlinde formula”, arXiv:1801.01878 (2019).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.