1. Not yet, but a suffix array has lower space requirements than a suffix tree.

2. In Dan Gusfield's book 'Algorithms on Strings, Trees, and Sequences' several inexact matching solutions based on suffix trees are discussed (for example in chapter 12.2). If there are up to k errors without insertions and deletions, it is the k-mismatch problem (see chap. 9.4) with a solution running in O(km), where k is the number of errors and m is the length of the string. Otherwise if the errors shall include insertions/deletions it is called the k-difference problem.

Hope that helps.