A data engineering team uses a pipeline to ingest Kafka source data into a Multiplex bronze table, partitioned by topic and month. A new engineer notices the 'user_activity' topic contains Personal Identifiable Information (PII) that must be deleted every two months per the company’s SLA. How can table partitioning assist in meeting this requirement?