Folklore conjecture on nodal volume of harmonic functions

For every integer n≥3n\ge 3, there exists a constant Cn>0C_n>0 such that, for every nonzero real-valued function uu satisfying Δu=0\Delta u=0 in B4⊂RnB_4\subset\mathbb{R}^n and u(0)=0u(0)=0, one has Hn−1({x∈B2:u(x)=0})≥CnN(u)\mathcal{H}^{n-1}\bigl(\{x\in B_2:u(x)=0\}\bigr)\ge C_n\mathcal{N}(u), where N(u)=log⁡2 ⁣(sup⁡B1∣u∣sup⁡B1/2∣u∣)\mathcal{N}(u)=\log_2\!\left(\frac{\sup_{B_1}|u|}{\sup_{B_{1/2}}|u|}\right). The dependence on N(u)\mathcal{N}(u) is conjectured to be optimal.

References

Primary source

arXiv

Additional references

Progress summary

Refreshed
Claimed solved

A September 2026 preprint claims to establish the conjectured lower bound, but the result has not yet been independently verified.

The conjecture predicts a sharp dimension-dependent lower bound for the nodal hypersurface measure of a harmonic function in a Euclidean ball. The target is the linear dependence on the function's oscillation or frequency quantity.

Known results

  • In March 2023, an almost-sharp result proved a bound with exponent 1−ε1-\varepsilon for n≥3n\ge 3, while explicitly leaving the linear estimate as Conjecture 1.3.

September 10, 2026 claimed proof

Zhehui Wang's preprint A sharp lower bound for the nodal volume of harmonic functions claims the conjectured dimension-dependent bound in the stated Euclidean-ball setting. The proof is available only as an unrefereed preprint, so this is a claimed resolution rather than a verified theorem.

Current status (as of September 2026): The conjecture has a claimed proof in an unrefereed preprint, but independent verification is not recorded.

Sources

Solutions 0

No solutions have been posted yet.