The correct approach is C because performing an insert-only merge (using MERGE command) with a matching condition on a unique key ensures that new records are only inserted if they do not already exist in the Delta table. This effectively deduplicates against previously processed records.

• A is incorrect because schema enforcement checks data types and structure, not duplicates.
• B (VACUUM) manages file storage and does not address deduplication.
• D suggests a full outer join and overwrite, which is inefficient and risky compared to the atomic MERGE operation.