The first problem addressed by this article is the enumeration of some families of patternavoiding inversion sequences. We solve some enumerative conjectures left open by the foundational work on the topics by Corteel et al., some of these being also solved independently by Lin, and Kim and Lin. The strength of our approach is its robustness: we enumerate four families F1 ⊂ F2 ⊂ F3 ⊂ F4 of pattern-avoiding inversion sequences ordered by inclusion using the same approach. More precisely, we provide a generating tree (with associated succession rule) for each family Fi which generalizes the one for the family Fi−1.The second topic of the paper is the enumeration of a fifth family F5 of pattern-avoiding inversion sequences (containing F4). This enumeration is also solved via a succession rule, which however does not generalize the one for F4. The associated enumeration sequence, which we call the powered Catalan numbers, is quite intriguing, and further investigated. We provide two different succession rules for it, denoted ΩpCat and Ω steady , and show that they define two types of families enumerated by powered Catalan numbers. Among such families, we introduce the steady paths, which are naturally associated with Ω steady . They allow us to bridge the gap between the two types of families enumerated by powered Catalan numbers: indeed, we provide a size-preserving bijection between steady paths and valley-marked Dyck paths (which are naturally associated with ΩpCat).Along the way, we provide several nice connections to families of permutations defined by the avoidance of vincular patterns, and some enumerative conjectures. arXiv:1808.04114v2 [math.CO] 14 Dec 2018 Proposition 2. Any inversion sequence e = (e 1 , . . . , e n ) is a Catalan inversion sequence if and only if for any i, with 1 ≤ i < n, if e i forms a weak descent, i.e. e i ≥ e i+1 , then e i < e j , for all j > i + 1.Proof. The forward direction is clear. The backwards direction can be proved by contrapositive. More precisely, suppose there are three indices i < j < k, such that e i ≥ e j , e k . Then, if e j = e i+1 , e i forms a weak descent and the fact that e i ≥ e k concludes the proof. Otherwise, since e i ≥ e j , there must be an index i , with i ≤ i < j, such that e i forms a weak descent and e i ≥ e k . This concludes the proof as well.