Ozsváth-Szabó and Rasmussen invariants of doubled knots (Q2492045)

Let \(\nu\) be an integer-valued knot invariant such that (1) \(\nu(K_{1}\sharp K_{2})=\nu(K_{1})+\nu(K_{2})\), (2) \(| \nu(K)| \leq g_{4}(K)\), and (3) \(\nu(T_{p,q})=(p-1)(q-1)/2\), where \(K_{1}\sharp K_{2}\) is the connected sum of two knots \(K_{1}\) and \(K_{2}\), \(g_{4}(K)\) is the \(4\)-ball genus of a knot \(K\), and \(T_{p,q}\) is the torus knot of type \((p,q)\) with \(p>0\) and \(q>0\). There are at least two such invariants; the Ozsváth-Szabó invariant defined by using knot Floer homology [\textit{P. Ozsváth} and \textit{Z. Szabó}, Geom. Topol. 7, 615--639 (2003; Zbl 1037.57027)], and the Rasmussen invariant defined by using Khovanov homology [\textit{J. A. Rasmussen}, Khovanov homology and the slice genus, arXiv:math.GT/0402131]. (These two invariants are indeed different, cf. [\textit{M. Hedden} and \textit{P. Ording}, The Ozsváth-Szabó and Rasmussen concordance invariants are not equal, arXiv:math.GT/0512348].) In the paper under review the authors prove the following results using only the three properties above, i.e., they do not use knot Floer homology or Khovanov homology: Let \(D_{\pm}(K,t)\) be the positive or negative \(t\)-twisted double of a knot \(K\). Then: (i) If \(\nu(D_{+}(K,t))=\pm1\), then \(\nu(D_{-}(K,t))=0\), and (ii) The sequence \(\nu(D_{+}(K,t))\) is non-increasing with respect to \(t\), and so there exists an integer \(t_{K}\) such that \(\nu(D_{+}(K,t))=1\) if \(t\leq t_{K}\) and that \(\nu(D_{+}(K,t))=0\) if \(t>t_{K}\). Moreover \(t_{K}\) satisfies the inequalities \(TB(K)\leq t_{K}<-TB(-K)\), where \(TB(K)\) is the Thurston-Bennequin number, see [\textit{D. Bennequin}, Astérisque 107/108, 87--161 (1983; Zbl 0573.58022)]. They also show: (iii) Given a \(2g\times 2g\) Seifert matrix \(A\), there exists a knot \(K_{i}\) with \(\nu(K_{i})=i\) bounding a Seifert surface of genus \(g\) with \(A\) as its associated Seifert matrix for every \(i\) with \(-g\leq i\leq g\).