Hi Michelle,
If the journal platform keeps a log of web requests, you should be able to filter these for just the OAI-PMH requests, and then analyse that data.
Hopefully the logs will contain the User-Agent and the originating IP address of the requests.
Some harvesters send a useful user agent e.g.
"IRUS_metadata_harvesting_bot"
"Summon/1.0"
Others just the software being used to harvest e.g.
"Java/11.0.4"
"OAIHarvester/2.0"
NB The user-agent information may not be reliable in all cases. The harvester can claim to be something it's not. How much trust you place in this data is up to you.
The IP address may provide some useful information if it can be resolved to an organisation. In a lot of cases it will resolve to a cloud-computing platform, which probably won't tell you anything useful.
Cheers,
John