A company stores data in an on-premises Oracle database and needs the data available in Amazon Aurora PostgreSQL for analysis. The company connects on-premises to AWS using an AWS Site-to-Site VPN and must capture changes that occur in the source database during migration. Which solution meets these requirements?

This solution is correct because AWS SCT is ideal for converting the Oracle schema to a compatible Aurora PostgreSQL schema, addressing the database heterogeneity. AWS DMS is the appropriate service for migrating existing data and, crucially, replicating ongoing changes (Change Data Capture - CDC) from the on-premises Oracle database to Aurora PostgreSQL. This ensures data consistency during and after the migration, fulfilling the requirement to capture changes. The first incorrect option only mentions a "full-load migration task," which would not capture ongoing changes. The second and fourth incorrect options involve migrating data to S3 first, then importing it. While possible for bulk transfers, this method is less efficient for continuous change capture from an operational database and doesn't directly support CDC like DMS does. AWS DataSync is primarily for file/object storage, not relational databases, and Snowball is for large-scale offline data transfer, neither of which directly addresses real-time change capture for database migration.