A generic population model is formulated for radiation risk assessment of natural biota. The model demonstrates that effects of radiation on the population survival do not follow directly the effects on individual organisms. Dose rates resulting in population extinction can be analytically calculated. Besides individual radiosensitivity, two key parameters were found to determine the survival potential of a population under chronic radiation stress: the ratio “biomass losses/biomass synthesis,” and the lump amount of limiting resource in the environment. A benchmark scenario “Population response to chronic irradiation” developed within the IAEA Programme EMRAS II was calculated for generic populations of mice, hare/rabbit, wolf/wild dog, and deer/goat chronically exposed to different levels of ionizing radiation. In the conditions of the benchmark scenario, model populations survived normally (>90% of the control value) at dose rates below the following levels: 3 mGy day−1 for wolf/wild dog; 10 mGy day−1 for deer/goat; 14 mGy day−1 for hare/rabbit; and 20 mGy day−1 for mice. The model predictions showed a relatively high survival potential of short-lived and productive species such as mice. At the same time, populations of long-lived animals with slow and radiosensitive reproduction such as wolf/wild dog were candidates to extinction at chronic exposures above 5 mGy day−1. Recovery of short-lived and productive species took a much shorter time compared with long-lived and slow reproductive species.