Yes, the epoch_ticks is seconds in the epoch since 1970.
The zulu_timestamp is computed from the epoch_ticks. And for portions of the 2013 corpus, the two are off by an hour because of a timezone locales error. This caused the stream_id, which starts with epoch_ticks, to disagree with the date_hour directories --- in some cases.
jrf