I don't think there is any way you can set a direct threshold.
You could use a cron job / Job Scheduler task to check the values on a regular basis.
Some of the values do put warning messages in the errorlog when scavenging starts to happen and these can be picked up by an errorlog monitor.