We are using EC2 plugin to distribute our builds on EC2 instances. We need clarity or an improvement on the understanding that how does Jenkins (or EC2 plugin) decides to TERMINATE nodes (or EC2 instances). How it decides what nodes need to get TERMINATES and when?

Let's say: We have one node (Minimum number of instances: 1) always available. But, we are facing some EBS space related issue if the instance stays for long time. Can we adjust or fine-tune the plugin so that every day at a specific time-period old node gets TERMINATED and a new node would take its place? Right now, we are not able to see an such setting that helps us to specify such functionality.
Kindly help us or navigate us to understand how actually it does work behind the scenes.
If possible kindly mention the same in the plugin documentation.