It is said that "if one/some of the nodes are unavailable for a configured amount of time (< 3 hours), Cassandra will automatically send the data to the node(s) when they become available again"
Suppose Node is unavailable for more than 24 hr and some write operations were done within these hours, then how Cassandra will synchronize all data to that node?
If the node was down more than specified period of time (3 hours by default, but configurable), then you will need to perform data repair using the nodetool repair (see documentation) - there is no automatic repair happening (except so-called read repair that may synchronize individual items when they are read with consitency levels higher than ONE/LOCAL_ONE)
