Riemann surface Study Guide

📖 Core Concepts Riemann surface – a connected one‑dimensional complex manifold; locally looks like an open set of ℂ. Complex vs. conformal structure – a complex structure gives holomorphic transition maps; pulling back the Euclidean metric yields a conformal (angle‑preserving) structure. Underlying real manifold – every Riemann surface is a 2‑dimensional real manifold equipped with a complex structure. Genus ( g ) – topological invariant counting “holes”; sphere (g = 0), torus (g = 1), higher‑genus surfaces (g ≥ 2). Uniformization theorem – any simply connected Riemann surface is conformally equivalent to exactly one of: The Riemann sphere ℙ¹(ℂ) (curvature +1) The complex plane ℂ (curvature 0) The unit disc 𝔻 (or upper half‑plane ℍ) (curvature ‑1) Compact vs. non‑compact – compact ⇒ every holomorphic ℂ‑valued function is constant (Maximum Principle); non‑compact ⇒ there exist non‑constant holomorphic functions (Stein property). Algebraic realization – by Chow’s theorem, every compact Riemann surface is a projective algebraic curve; its function field is a finite extension of ℂ(t). Key theorems – Identity theorem, Riemann–Roch, Riemann–Hurwitz, Hurwitz automorphism bound (|Aut| ≤ 84(g‑1) for g ≥ 2). --- 📌 Must Remember Definition – $X$ is a connected Hausdorff space with an atlas $\{(U\alpha,\phi\alpha)\}$ to the unit disk, transition maps $\phi\beta\circ\phi\alpha^{-1}$ holomorphic. Compact surface → constant holomorphic functions; non‑compact → admits non‑constant holomorphic functions. Uniformization types – sphere (elliptic), plane/cylinder/torus (parabolic), unit disc quotients (hyperbolic). Genus‑curve relation – hyperelliptic curve $y^{2}=Q(x)$ with $\deg Q = 2g+1$ or $2g+2$ gives genus $g$. Riemann–Hurwitz formula – for a non‑constant holomorphic map $f: X\to Y$ of degree $d$: $$\chi(X)=d\,\chi(Y)-\sum{p\in X}(ep-1),$$ where $\chi$ is Euler characteristic and $ep$ ramification index. Hurwitz automorphism bound – for $g\ge 2$, $|{\rm Aut}(X)|\le 84(g-1)$. Liouville’s theorem on maps – any bounded entire map $\mathbb{C}\to\mathbb{C}$ is constant; in particular, any holomorphic map $\mathbb{C}\to\mathbb{D}$ is constant. --- 🔄 Key Processes Identify the genus Write the defining equation (e.g., $y^{2}=Q(x)$). Compute $\deg Q$: if $\deg Q = 2g+1$ or $2g+2$, the genus is $g$. Classify the surface via uniformization Check if the surface is simply connected. Determine universal cover: sphere → elliptic, plane → parabolic, disc → hyperbolic. Apply Riemann–Hurwitz List ramification points and their indices $ep$. Plug into $\chi(X)=d\,\chi(Y)-\sum(ep-1)$ to solve for unknowns (e.g., degree $d$ or possible $g$). Use Chow’s theorem (compact case) Conclude the surface can be embedded as a projective algebraic curve; then work with polynomial equations. --- 🔍 Key Comparisons Elliptic (sphere) vs. Parabolic (plane/cylinder/torus) vs. Hyperbolic (disc quotients) Curvature: +1 vs 0 vs ‑1. Universal cover: ℙ¹(ℂ) vs ℂ vs 𝔻. Automorphism group: infinite Möbius group vs translations/rotations vs finite (for $g\ge2$). Compact vs. Non‑compact Holomorphic functions: constant vs. non‑constant exist. Function field: finite extension of ℂ(t) vs. often transcendental. Holomorphic map vs. Meromorphic map Holomorphic: no poles, locally $\mathbb{C}\to\mathbb{C}$ analytic. Meromorphic: allowed poles; on compact surfaces they always exist non‑trivially. --- ⚠️ Common Misunderstandings “Every compact Riemann surface has non‑constant holomorphic functions.” False – only constant ones exist. Confusing boundedness with non‑triviality – Liouville says a bounded entire map $\mathbb{C}\to\mathbb{C}$ is constant; it does not apply to maps $\mathbb{C}\to\mathbb{D}$ unless boundedness is explicit. Assuming all genus‑1 surfaces are the same – torus can have many complex structures (different $\tau$). Thinking the automorphism group is always infinite – for $g\ge2$ it is finite, bounded by $84(g-1)$. --- 🧠 Mental Models / Intuition “Patchwork quilt” – picture a Riemann surface as a quilt of tiny ℂ‑patches sewn together by holomorphic overlaps. Uniformization as a “DNA code” – the universal cover (sphere, plane, disc) uniquely determines the surface’s geometric “type”. Genus as “number of handles” – each handle adds a “hole”, raising curvature from +1 toward ‑1. --- 🚩 Exceptions & Edge Cases Genus 0: the only simply connected compact surface; uniformizes to the sphere, not the disc. Genus 1: can be parabolic (torus) or a cylinder (non‑compact). Liouville’s theorem does not forbid non‑constant holomorphic maps $\mathbb{C}\to\mathbb{D}$ that are unbounded (e.g., the exponential map composed with a Möbius transformation). Hurwitz bound is sharp only for very special “Hurwitz surfaces” (e.g., the Klein quartic). --- 📍 When to Use Which Uniformization → when you need the conformal type of a simply connected surface. Riemann–Hurwitz → when analyzing a non‑constant holomorphic map between compact surfaces (degree, ramification, genus relations). Riemann–Roch → to compute dimensions of spaces of meromorphic sections (e.g., number of independent holomorphic differentials). Chow’s theorem → whenever you must replace a compact Riemann surface by a projective algebraic curve (e.g., to use algebraic geometry tools). Hurwitz automorphism bound → to estimate or rule out large symmetry groups for $g\ge2$. --- 👀 Patterns to Recognize Polynomial degree ↔ genus – hyperelliptic equation $y^{2}=Q(x)$ with $\deg Q = 2g+1$ or $2g+2$. Ramification sum in Riemann–Hurwitz often appears as an even integer; check $ \sum (ep-1)$ for parity errors. Curvature sign matches the uniformization type: +1 (sphere), 0 (plane/cylinder/torus), –1 (disc quotients). Automorphism finiteness shows up for $g\ge2$; any answer implying infinite symmetries for such a surface is a red flag. --- 🗂️ Exam Traps Distractor: “A compact Riemann surface always admits a non‑constant holomorphic map to ℂ.” – Wrong; only constant maps exist by the Maximum Principle. Distractor: “Any holomorphic map from ℂ to the unit disc must be constant because of Liouville.” – Liouville applies to bounded maps into ℂ, not into the disc. Distractor: “All genus‑1 surfaces are conformally equivalent to the standard torus $\mathbb{C}/\mathbb{Z}$.” – False; the modulus $\tau$ varies (any $\tau$ with $\operatorname{Im}\tau>0$). Distractor: “The automorphism group of a genus‑2 surface can be arbitrarily large.” – Violates Hurwitz bound ($\le84(g-1)=84$). Distractor: “If a surface is hyperbolic, its universal cover must be the upper half‑plane and the unit disc simultaneously.” – They are conformally equivalent, but the classification uses one model; stating both as distinct is misleading. ---