On the \(K(1)\)-local homotopy of \(\mathrm{tmf}\wedge\mathrm{tmf}\) (Q2230806)

Adams spectral sequences based on \(H\mathbb{F}_p\) or \(MU\) are classic tools in computational algebraic topology. Their \(E_2\)-terms are, at least in principle, readily computable as certain Ext-groups. In contrast, when basing an Adams spectral sequence on a ring spectrum like connective \(K\)-theory \(ko\), there is no good formula for its \(E_2\)-term. Nevertheless, Mahowald has made good use of the \(ko\)-based Adams spectral sequence and used it in particular to prove the height \(1\) telescope conjecture at the prime \(2\). A research program, driven by Behrens and many coauthors, is to transfer the successes of the \(ko\)-Adams spectral sequence one height higher by using \(tmf\) instead, the spectrum of connective topological modular forms. It arises as the connective cover of a spectrum \(Tmf\) related to the moduli stack of generalized elliptic curves. The crucial step to work with the \(tmf\)-based Adams spectral sequence is to understand \(tmf \wedge tmf\). Chromatic homotopy theory suggests to study \(L_{K(i)}tmf \wedge tmf\) first. The localizations vanish for \(i>2\) and are quite well understood for \(i=0\) and \(i=2\). The paper under review computes \(L_{K(1)}tmf \wedge tmf\). The basic method is to use an \(E_{\infty}\)-presentation of \(L_{K(1)}tmf\), originally due to Hopkins and further studied by Laures, and the theory of \(\Theta\)-algebras. The result also computes a number-theoretic object, namely the moduli of pairs of generalized elliptic curves with an isomorphism betwen their formal groups. Unfortunately, chromatic assembly would only yield \(L_{E(2)}tmf \wedge tmf\) or \(Tmf\wedge Tmf\) and not \(tmf \wedge tmf\) itself. Thus additional information is needed beyond the chromatic approach based on \(K(i)\)-localizations to understand \(tmf \wedge tmf\). Nevertheless, the paper under review is a good step towards this goal. Moreover, it appears to be well written and can be recommended to anybody interested in \(K(1)\)-local \(tmf\).