Continuity Optional: Geometry of Loewner Chains with Jumps ✛ ✎
Context
A hundred years ago, Charles Loewner devised an ingenious way to encode the evolution of planar domains into an ordinary differential equation. This made it possible to solve problems about conformal mappings, but even Loewner himself didn’t know whether the growth is always generated by a continuous curve. In the 21st century, interest in the theory exploded thanks to the random processes of SLE (Schramm–Loewner evolution), which turned out to be universal limits of many lattice models in statistical physics. Yet until now, fundamental questions about discontinuous driving functions remained wide open.
Methods
The study rests on a sophisticated analysis of growing hulls in the complex upper half-plane. The authors work carefully with capacity parametrization (hcap($K_t$)=2t) and construct explicit chains of crosscuts that separate newly born pieces. Building on Carathéodory’s theorem, they prove that for a discontinuous càdlàg function W(t), the Loewner equation ∂_t^+ $g_t$(z) = 2/($g_t$(z)-W(t)) uniquely determines the conformal map $g_t$. The connectivity of hulls is investigated through the topology of the added sets.
Results
The main result is a one-to-one correspondence between locally growing hulls and càdlàg functions (Theorem 1.2). At each time t, the hull gains either a “bubble” bubble—an open simply connected set of swallowed points—or a compact connected set of boundary points $P_t$ (Theorem 1.4). The hulls are generated by a function η(t) if and only if their union with the real line is path-connected (Theorem 1.6). If the radial limit η(t) = lim_{y→0+} $g_t^{-1}$(W(t-)+iy) exists, then η has right limits and forms a càglàd function (Theorem 1.7). Moreover, when η is left-continuous, the hulls are locally connected (Theorem 1.8).
Implications
These findings close a long‑standing gap in Loewner theory. They make it possible to work with processes whose driving functions have jumps—for instance, Lévy processes in random geometry. Classifying the added points is vital for a rigorous analysis of phase transitions in two-dimensional lattice models. In addition, the fresh perspective on generating functions helps distinguish curves from more exotic fractal structures.
Future development
The next step will likely be to carry these results over to the radial case (inside the unit disk) and to measure-driven chains. It is also interesting to study how hulls behave under even weaker conditions—for example, when local growth fails and the hull grows from many points simultaneously. This might shed light on the behavior of rare events in random media.
Impact
The work will influence probability theory, complex analysis, and mathematical physics. It provides tools for studying boundary effects in models where stochastic domain evolution arises, including hydrodynamics, growth theory, and polymer physics.
Next steps
The immediate challenge is to determine under what additional conditions the existence of a generating function guarantees its left-continuity without assuming that the hulls have empty interior. Explicit examples for all possible topological types listed in the paper also need to be constructed.
Key open problems
One unsolved problem is proving the existence of the SLE trace for κ > 8. The new boundary-behavior results may simplify this analysis. Another open question remains: Is a hull driven by a continuous function always locally connected? Advanced Markov methods could provide insight.
The knowledge machine's view
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
The paper examines the evolution of growing hulls in the upper half-plane under discontinuous driving functions W(t), using the capacity parametrization hcap(K_t)=2t and explicit slits. The authors establish a one-to-one correspondence between locally growing hulls and càdlàg functions, classify the added sets as bubbles or compact boundary sets, and provide conditions for the existence of a generating function η with various continuity properties.
Where its strength is
The main strength is the complete characterization of the relationship between the analytic properties of the driving function and the topology of the growing sets, including criteria for linear and local connectedness. This provides a rigorous tool for analyzing discontinuous processes that were previously excluded from classical theory.
Where it could go next
- It might be interesting to see how universal limit laws from a work on why different systems behave alike manifest in the context of random Loewner chains with jumps—for instance, when studying scaling limits of lattice models with nonstandard weights.based on: Why Different Systems Behave the Same arXiv:2607.09079
- A related step comes from a study on how simple rules generate complex structures: one could try to apply a similar perspective to the generation of discontinuous driving functions that produce fractal or self-similar hulls.based on: Life at the Pencil's Tip: How Simple Rules Give Rise to Cells arXiv:2607.27402
- A possible direction is to consider how ideas about phase transitions and freezing into patterns from a work on electrons transfer to the classification of added sets at critical parameters of discontinuous processes.based on: Electrons Freeze into a Pattern arXiv:2607.17027
Worth trying
- Check whether discontinuous driving functions with bubbles provide natural examples of universality in the spirit of a related work on the similarity of systems.
- Test the slit construction for explicitly building hulls generated by simple piecewise constant W with prescribed jumps.
- Compare the topological properties of hulls for Lévy processes with the pattern results from a work on electrons.
Why this may matter
Extending Loewner theory to discontinuous driving functions opens the way to rigorous analysis of stochastic processes with jumps in geometry and physics, including critical phenomena and fractal structures, and may clarify universality in irregular growth regimes.
Nearby works in our archive
- Life at the Pencil's Tip: How Simple Rules Give Rise to Cellssimilarity 0.58
- Why Different Systems Behave the Samesimilarity 0.58
- Electrons Freeze into a Patternsimilarity 0.57
- How a Neural Network Eavesdropped on the Noise of Lifesimilarity 0.57
- Round Trip: How a Neural Network Checks Itself Without Hintssimilarity 0.57
- One-Way Flow: The Secret of the Lymphatic Pumpsimilarity 0.56
The archive is sparse around this work: the closest review is only 0.58, against a usual 0.64. It looks like few have drilled nearby.
Though the Loewner equation looks like a harmless ODE, it can generate mind‑boggling geometric shapes. For instance, the spiral from Example 1.10 winds around a circle and, at the moment of closing, instantly swallows a whole disk, and this happens under an absolutely continuous driving function!
Key numbers
- critical Hölder exponent for the SLE trace: 4
- capacity growth rate: 2
- maximum number of left limits of η(t): a continuum
- Hausdorff dimension of SLE_κ (κ<8): 1 + κ/8