If you strictly enforce a 1:1 relationship between each Service and Repository (and each Repository maps to one database table) what you will end up with is database-driven design, where the structure of your code is determined not by the domain but by the schema of the persistence layer.
Given that persistence is nothing more than an implementation detail, this structure is less than ideal. You should be able to modify your schema, for example by normalizing, denormalizing, switching between SQL and NoSQL, etc., without modifying code outside the persistence layer.
To phrase it another way, a Service shouldn't care whether the data it retrieves comes from one table or ten tables or a flat file system, or a network call, or an in-memory cache, or anywhere else. To a Service, data is data and its origin is irrelevant.
With all that being said, it sounds like maybe your domain includes the concept of something like a TimestampService which needs to be implemented.
Bob Martin describes this concept of Domain-Driven Design thusly.
For a very long time there was an idea that the domain model was the database model. They were the same. You'd come up with an entity-relationship diagram that would describe the schema of the database, and that was your domain model.
The problem with that viewpoint is that it has no behavior, whereas the goal of the business is to implement behaviors. Now, you might need data structures to do that; but we want to know what the behaviors are. So, in the Clean Architecture, in Ports & Adapters, in all of these architectures, the business objects contain behavior... They might use data structures, but the business objects represent behavior.
...
Yes, you need to be able to sub-divide the data. You've got to be able to have a good data model; but the data model is not the behavior model. The data model is not the business model. The data model is not the domain.