Yes, because of several other reasons. The translation isn’t always done by the same person or on the same end. This has shown itself several times in the GRIN catalog in other cultivars, not necessarily pyrus. Sometimes it’s the donator that translates. Sometimes it’s the receiver that translates. Hence, especially in the second case you end with minor variations.
I can speak Mandarin and several dialects, I can tell you the variation isn’t as great unless it’s not a mutually intelligible language like Hokkien or Wenzhou. The subdialects are equivalent to me visiting my friend in Boston and listening to a Bostonian accent. They may say the word slightly differently (and wrong btw 
) like pizzer but I still completely understand them with and without context. And even if they say things different, we all write the same way in English the same way as they write the same way in Chinese.
Combine this with the fact that we have misspellings of cultivars that were developed within the continental states because tags and labels are copies of copies that get passed along, and bam, you have another reason there are variations in spelling. Even in those cases, if we drop of vowel or consonant from like asmede kernl aepple, we still know this is a misspelling of ashmead kernel apple or at the very least we know it’s English not french or german, latin, or some rare off the wall cultivar from deep in the jungles discovered and saved by ancient native tribe and passed in their native tongue to colonists who approximated it into their language - spanish or whatever.
Having said all this, it is completely possible that tse li is meant to be separate and distinct from tsu li and not a misspelling. I would consider that possibility for any conflict where two cultivars share similar cultivar names. That said, 酥梨 is 100% a category of pear in China. This is a term that one would find both in nursery and grocery trade. Also, there as is often the case, there are multiple inaccuracies in the GRIN database, which either arises from human error in data entry or just being flat our wrong. There are cases where nectarines are labeled as peaches and vice versa.
China is a big place. I don’t know every location or am well versed in every province. Shandong is a fairly well populated province though, and as far as I am aware, there is no county or district called tse.
I wouldn’t exclude the local dialect as a possibility, but if like in the case of tse, if this was like 中原官话, the consonant use would be different. It’s really no different than if someone wrote out a Bostonian accent. You know that drop of the r exist. Same as in the case of Chinese dialects, there some key traits that would make it apparent it was a dialect.
If this was Wenzhou we might as well be talking Klingon, because there is no way any of you would be able to pronounce any thing correctly.
I always advocate for preserving the original spelling of the cultivar in the native language if possible, because it always helps to prevent these problems. Like recently, I discovered Fire Crystal Persimmon in the US is probably not even the Fire Crystal in China because the Chinese name they used as a syn is a completely a different cultivar.
