There may be a more elegant solution offered to this later, but here is one way.
1. Use the "dfm eventtype" command line to lower the severity of the Volume Full event so that it is not forwarded to email or pagers as part of the Critical or Emergency eventtype severity levels. This will remove alarms from all volumes (bear with me).
2. Create a new group in OM to house all the volumes you want to be alarmed on. If all the volumes on a given storage appliance should receive alarms, add the entire appliance as a group member. The same is true for Aggregates. This is a cumbersome solution if the volumes you care about are widely scatttered. It is very easy to script a solution for this kind of problem if the volumes you want to be alarmed on have a consistent naming convention. You could also easily automate the addition of new volumes to this group in the future, based on a scheduled script that matched the naming convention and automatically performed the group addition.
3. Create a new alarm in OM which will catch the Volume Full event, but only for the newly created group and attach the appropriate paging and alarming to that new alarm. You can attach pagers and email notificaitons to specific events for specific groups regardless of what severity they are (see step one).
