Seiberg-Witten-Floer homology and gluing formulae (Q1396014)

Writing a paper on Seiberg-Witten-Floer (SWF) homology is a tricky proposition for a couple of reasons. First, one needs to make several choices when defining the invariant. Different choices lead to different versions of the invariants. After deciding what route to take defining the invariants, one must decide what to present and what not to present. There are several different versions of SWF homology [\textit{M. Marcolli} and \textit{B. L. Wang}, Commun. Anal. Geom. 9, 451--639 (2001; Zbl 1026.57025)], [\textit{W. Chen}, Turk. J. Math. 21, 61--81 (1997; Zbl 0891.57021)]. In addition, Peter Ozváth and Zoltán Szabó have defined a Heegaard-Floer homology of 3-manifolds that is conjecturally equivalent to a SWF homology [\textit{P. Kronheimer}, \textit{T. Mrowka}, \textit{P. Ozsváth} and \textit{Z. Szabó}. Monopoles and lens space surgeries, http://www.arxiv.org/abs/math.GT/0310164], [\textit{P. Ozsváth} and \textit{Z. Szabó}, Holomorphic disks and topological invariants for closed three-manifolds, to appear in Ann. Math., http://www.arxiv.org/abs/math.SG/0101206], [\textit{P. Ozsváth} and \textit{Z. Szabó}, Holomorphic disks and three-manifold invariants: properties and applications, to appear in Ann. Math., http://www.arxiv.org/abs/math.SG/0105202]. Ultimately the different versions of SWK homology will be judged on their usefulness. One important role of SWF theory is to establish glueing formulae for computing the 4-dimensional version of the Seiberg-Witten invariant. The reason one must worry about what to present and what not to present in a paper on SWF theory is that there are many details that must be established (Is the square of the boundary zero? Is the invariant independent of the metric?, etc). The proofs of these important detils tend to be long and technical. However, the general outline of the arguments is familiar from the Donaldson-Floer and symplectic Floer homology theories. In this paper, the authors define two variants of SWF homology, a relative Seiberg-Witten invariant for 4-manifolds with boundary and a glueing theorem. In addition, they establish the existence of a \(\mathbb Z[u]\text{-module}\) structure on their first version of SWF homology. This same module structure is found on the Ozváth-Szabó Heegaard Floer homology. The glueing theorem in this paper generalizes an earlier glueing theorem of Morgan, Szabó and Taubes [\textit{J. W. Morgan}, \textit{Z. Szabó} and \textit{C. H. Taubes}, J. Differ. Geom. 44, 706--788 (1996; Zbl 0974.53063)].