Abstract
6 7 Random environmental fluctuations pose major threats to wild populations. As patterns of 8 environmental noise are themselves altered by global change, there is growing need to identify 9 general mechanisms underlying their effects on population dynamics. This notably requires 10 understanding and predicting population responses to the color of environmental noise, i.e. its 11 temporal autocorrelation pattern. Here, we show experimentally that environmental 12 autocorrelation has a large influence on population dynamics and extinction rates, which can be 13 predicted accurately provided that a memory of past environment is accounted for. We exposed 14 near to 1000 lines of the microalgae Dunaliella salina to randomly fluctuating salinity, with 15 autocorrelation ranging from negative to highly positive. We found lower population growth, 16 and twice as many extinctions, under lower autocorrelation. These responses closely matched 17 predictions based on a tolerance curve with environmental memory, showing that non-genetic 18 inheritance can be a major driver of population dynamics in randomly fluctuating 19 environments. 20 21