Holomorphic curves in low dimensions. From symplectic ruled surfaces to planar contact manifolds (Q1753192)

This book presents ``modern'' proofs of the main results in McDuff's paper on the classification of rational and ruled symplectic 4-manifolds, using techniques that had not been fully developed when \textit{D. McDuff} wrote her paper [J. Am. Math. Soc. 3, No. 3, 679--712 (1990; Zbl 0723.53019)]. The book has several well written chapters summarizing the background material on \(J\)-holomorphic curves required for the proofs of the theorems, with an emphasis on Lefschetz pencils and fibrations. Much of the analytic background material is stated without proof. However, intuitive justifications for the necessary background results are included, while full proofs can be found in [\textit{D. McDuff} and \textit{D. Salamon}, \(J\)-holomorphic curves and symplectic topology. 2nd ed. Providence, RI: American Mathematical Society (AMS) (2012; Zbl 1272.53002)]. Chapter 7 of the book discusses Gromov-Witten invariants and a related theorem of McDuff's which says that a symplectic 4-manifold is symplectically uniruled if and only if it is a blowup of a rational or ruled surface. The main technical part of the proof is contained in [\textit{D. McDuff}, Ann. Inst. Fourier 42, No. 1--2, 369--392 (1992; Zbl 0756.53021)], but according to the author, ``I am not aware of any previous source in the literature that both contains the statement and explains why it is true.'' The last two chapters in the book discuss contact topology, and in particular, planar contact manifolds as a 3-dimensional analogue to rational and ruled surfaces in dimension 4, with an emphasis on the Weinstein conjecture and the classification of symplectic fillings. The main theorem of McDuff's proved in the first 6 chapters of the book is as follows, where \(\omega_{\mathrm{FS}}\) denotes the standard symplectic form on \(\mathbb{C}P^2\) and \([S] \cdot [S]\) denotes the homology intersection number. \textbf{Theorem A.} Suppose \((M,\omega)\) is a closed and connected symplectic 4-manifold containing a symplectically embedded 2-sphere \(S \subset M\) with \[ [S] \cdot [S] \geq 0. \] Then \((M,\omega)\) is either \((\mathbb{C}P^2,c\omega_{\mathrm{FS}})\) for some constant \(c > 0\) or it is a blown-up symplectic ruled surface. For the minimal case, i.e. when \(M\) does not contain any exceptional spheres, the author proves the following. \textbf{Theorem D} Suppose \((M,\omega)\) is a closed, connected and minimal symplectic 4-manifold that contains a symplectically embedded 2-sphere \(S \subset (M,\omega)\) with \([S] \cdot [S] \geq 0\). One then has the following possibilities: \begin{itemize} \item[(1)] If \([S] \cdot [S] = 0\) then \((M,\omega)\) admits a symplectomorphism to a symplectic ruled surface such that \(S\) is identified with a fiber. \item[(2)] If \([S] \cdot [S] = 1\) then \((M,\omega)\) admits a symplectomorphism to \((\mathbb{C}P^2,c\omega_{\mathrm{FS}})\) for some constant \(c > 0\), such that \(S\) is identified with the sphere at infinity \(\mathbb{C}P^1 \subset \mathbb{C}P^2\). \item[(3)] If \([S] \cdot [S] > 1\) then \((M,\omega)\) is symplectomorphic to one of the following: \begin{itemize} \item[(a)] \((\mathbb{C}P^2,c\omega_{\mathrm{FS}})\) for some constant \(c > 0\), \item[(b)] \((S^2 \times S^2, \sigma_1 \oplus \sigma_2)\) for some pair of area forms \(\sigma_1\), \(\sigma_2\) on \(S^2\). \end{itemize} \end{itemize} The author's approach to proving Theorem A and Theorem D yields the following generalization to Theorem A. \textbf{Theorem F.} Suppose \((M,\omega)\) is a closed and connected symplectic 4-manifold that contains a symplectically embedded 2-sphere \(S \subset (M,\omega)\) with \[ m := [S]\cdot [S] \geq 0. \] Then for any choice of pairwise distinct points \(p_1,\ldots , p_m \in S\), \((M,\omega)\) admits a symplectic Lefschetz pencil with base points \(p_1,\ldots, p_m\) (or a symplectic Lefschetz fibration if \(m=0\)), in which \(S\) is a smooth fiber and no singular fiber contains more than one critical point. Moreover, the set of singular fibers of this pencil (or fibration) is empty if and only if \(m \in \{0,1\}\) and \((M\backslash S,\omega)\) is minimal. These theorems are proved in Chapter 6 of the book. Chapter 2 of the book contains the necessary background on moduli spaces of pseudoholomorphic curves in symplectic manifolds, including discussions of Fredholm regularity, bubbling, gluing, nodal points, Gromov compactness, and orientations via the determinant line bundle. Subsection 2.2 focuses on dimension 4, where the author discusses ``positivity of intersections'' and sketches a proof of an important automatic transversality result unique to dimension 4. Chapter 3 discusses complex and symplectic blowups and Lefschetz pencils and fibrations. Chapter 4 is devoted to a discussion of compactness for moduli spaces of pseudoholomorphic curves in symplectic 4-manifolds, including complete proofs of two theorems concerning the convergence of sequences of pseudoholomorphic curves satisfying uniform energy bounds (Theorems 4.5 and 4.6). Chapter 5 contains a discussion of exceptional spheres, i.e. symplectically embedded spheres \(S\) that satisfy \[ [S] \cdot [S] = -1, \] and proofs of Theorems B and C in the book, which concern collections of 1-parameter families of pairwise disjoint exceptional spheres and maximal collections of pairwise disjoint exceptional spheres. Chapter 7 expands the discussion in the earlier chapters to uniruled symplectic 4-manifolds. Most of Chapter 7 is devoted to proving the following results. \textbf{Theorem 7.3.} If \((M,\omega)\) is a closed and connected symplectic 4-manifold, the following are equivalent: \begin{itemize} \item[(1)] \((M,\omega)\) is a symplectic rational surface or blown-up ruled surface. \item[(2)] \((M,\omega)\) admits a positively symplectically immersed sphere \(S \looparrowright M\) with \(c_1([S]) \geq 2\). \item[(3)] For some \(J \in \mathcal{J}_\tau(M,\omega)\), there exists a somewhere injective Fredholm regular \(J\)-holomorphic sphere \(u\) with ind\((u) \geq 2\). \item[(4)] For some \(A \in H_2(M)\) satisfying \(-2 + 2c_1(A) \geq 2\) and a dense subset \(\mathcal{J}^{\mathrm{reg}} \subset \mathcal{J}_\tau(M,\omega)\), there exists a somewhere injective Fredholm regular \(J\)-holomorphic sphere \(u\) with \([u] = A\) for every \(J \in \mathcal{J}^{\mathrm{reg}}\). \item[(5)] \((M,\omega)\) is symplectically uniruled. \end{itemize} \textbf{Corollary 7.4.} Each of the properties listed in Theorem 7.3 for symplectic 4-manifolds is invariant under symplectic deformation equivalence and symplectic blowup or blowdown. The results discussed in the first six chapters involve symplectically embedded spheres, whereas the results in Chapter 7 allow symplectically immersed spheres. In fact, one central issue addressed in Chapter 7 is the fact that a symplectically embedded sphere can become immersed under the blowdown operation. The equivalence of some of the conditions listed in Theorem 7.3 follow from results in the earlier chapters, and some of the implications involve Gromov-Witten invariants. Section 7.2 contains a heuristic discussion of general Gromov-Witten invariants and then a rigorous discussion of a simplified version of the invariants for symplectic 4-manifolds, while proving \((1) \Rightarrow (5) \Rightarrow (2)\). The most difficult part of Chapter 7 is the proof that \((2) \Rightarrow (1)\) in Theorem 7.3, which follows from a result in McDuff's paper on immersed spheres in symplectic 4-manifolds. \textbf{Theorem 7.34} (McDuff [Zbl 0756.53021]) Suppose \((M,\omega)\) is a closed symplectic 4-manifold admitting a positively symplectically immersed sphere \(S \looparrowright (M,\omega)\) with \(c_1([S]) \geq 2\). Then it also contains a symplectically embedded sphere with nonnegative self-intersection number. McDuff's original proof of this theorem only used pseudoholomorphic curves and algebraic topology, but the theorem can also be proved using Seiberg-Witten theory. The proof of Theorem 7.34 presented in Chapter 7 is a variation on McDuff's original proof and is independent of Seiberg-Witten theory. According to the author, the proof ``contains a few simplifications in comparison with the original, most of which are also due to McDuff but have not appeared in the literature before.'' The last two chapters in the book contain a fairly extensive survey of techniques and results in symplectic and contact geometry that involve pseudoholomorphic curves. Section 8.1 discusses the Arnol'd conjecture concerning a lower bound on the number of 1-periodic orbits of a Hamiltonian system on a symplectic manifold (Conjecture 8.1), Floer homology, and the Weinstein conjecture concerning the existence of a periodic orbit of a Reeb vector field on a closed contact manifold (Conjecture 8.11). Section 8.2 discusses symplectic cobordisms and symplectic fillings of contact manifolds, and Section 8.3 discusses punctured holomorphic curves, Morse-Bott contact forms, gluing, compactness, orientations, transversality, and intersection products in symplectic field theory (SFT). Chapter 9 surveys many interesting results concerning contact 3-manifolds that can be proved using the techniques discussed in Chapter 8. In Section 9.1 the author even provides an extensive sketch of the proof of the following result due to Gromov and Eliashberg. \textbf{Theorem 9.4} ([\textit{M. Gromov}, Invent. Math. 82, 307--347 (1985; Zbl 0592.53025)] and [\textit{Y. Eliashberg} [Lond. Math. Soc. Lect. Note Ser. 151, 45--72 (1990; Zbl 0731.53036)]) Weak symplectic fillings of \((S^3,\xi_{\mathrm{st}})\) are unique up to symplectic deformation equivalence and blowup. Moreover, every exact symplectic filling of \((S^3,\xi_{\mathrm{st}})\) is symplectomorphic to a starshaped domain in \((\mathbb{R}^4,\omega_{\mathrm{st}})\). Section 9.2 discusses fillings of the 3-torus and Giroux torsion, and Section 9.3 concludes the main part of the book with a discussion of open book decompositions and planar contact manifolds. Appendix A of the book sketches a proof of a folk theorem that says that ``a \(J\)-holomorphic foliation with an isolated transverse nodal singularity can be identified locally with the neighborhood of a critical point in a Lefschetz fibration.'' The book is well-written, well-referenced, and contains a large number of well-drawn multi-colored diagrams (at least in the pdf version). Anyone interested in McDuff's characterization of rational and ruled symplectic 4-manifolds or her theorem that says ``uniruled \(\Rightarrow\) rational/ruled'' should find this book quite useful. Not only does the book provide complete and detailed proofs of those theorems, but it also puts them in context and explains how they are used in Seiberg-Witten theory. Moreover, it also explains how the techniques used to prove the theorems relate to many other results in symplectic and contact geometry.