Could this word be its origins?
হাওড় hāoṛa n. an extensive marsh or quagmire or fen.
Thank you! This is certainly a much better phonetic match, and the semantics is probably good too, as there were marshlands all around that region. Still, there are a couple of issues with it:
1) I am not sure this word was ever used in South-West Bengal. I only know it as the name for shallow marshy lakes in Sylhet, like 400 Km away from Howrah.
2) If it was indeed used in SW Bengal, what was (or, maybe even 'is'?) its exact form? I have only ever known it as হাওর (i.e. with a final /r/) in the Sylhet context, but since the r-R phonetic distinction has been dead in Bangladesh for a while, that is not necessarily a deal-breaker for this proposal. In fact, the dictionaries seem to be split on the issue: Samsad has ড়/R, Haricharan and Bangladesh Bangla Academy's evolutionary dictionary (বিবর্তনমূলক বাংলা অভিধান; BBA) have র/r (Haricharan marks it as relating to Sylhet), and Sukumar Sen's "Etymological Dictionary of Bengali: c. 1000-1800" have both (with a quotation containing /r/ from Harivamsha by the 17th c. poet Bhavananda - also likely from Sylhet, as far as I could ascertain). Sen and BBA both give the etymon as Sanskrit সাগর/saagara.
If the etymology is correct (and it's a big 'if'), then the initial /h/ is possible only in the far Eastern Bengal, e.g. Sylhet, not around Howrah, where the actual form of this etymon is সায়র/sayôr. So, that brings us back to problem (1) - was this word ever used at the right place?