Category: Pop. Genetics

Summary: Testing whether moderate dormancy durations maximize persistence for branching lineages in autocorrelated Markov environments.

Dormancy can protect a lineage during bad periods, but protection is only useful if dormant individuals return at the right time. This experiment asks whether there is a middle ground where dormancy lasts long enough to bridge persistent bad runs, yet not so long that lineages miss frequent recoveries in the environment.

The script runs deep branching-process simulations under Markov-switching conditions and compares persistence across different fixed dormancy durations. Rather than locating a spectral threshold, it measures long-run survival and growth outcomes over very long trajectories, with BOINC replication supplying many independent seeds.

The scientific interest is mechanistic: if interior peaks in persistence exist, finite dormancy duration may be one reason why. That would make dormancy not just a binary trait, but a quantitatively tunable response to environmental memory.

Method: Long single-seed branching-process simulations in Markov environments, aggregated across many BOINC seeds to compare dormancy durations.

What is measured: Persistence and growth outcomes across dormancy durations, condition-level performance, and support for an interior optimum.