Abstract Tasmaphena lamproides is a rare snail found in northwest Tasmania. The species is eliminated by logging but re-establishes a population in ∼20-year-old native forest regeneration and builds up to pre-logging levels by ∼60 years. Major plantation development is planned within the range of the species. It is unlikely that T. lamproides will reinvade areas converted to plantation. To aid the conservation of T. lamproides the managing authority planned to retain a “biodiversity spine” (i.e. a string of contiguous coupes (logging units) that would be regenerated to native forest rather than converted to plantation) within areas earmarked for major plantation development. A PVA was used to assess the comparative impacts of different forest management scenarios. The management scenarios modelled involved differing levels of reservation, differing levels of native forest regeneration or conversion to plantation, different rotation lengths for native forest regeneration and different temporal patterns of logging of native forest. In a forest block with a major reserve simulations indicated that the population would decline to around 50% of the original population and thereafter remain fairly stable. For a forest block earmarked for plantation development where no major reserves occurred, simulations indicated the population would undergo a steep decline to around 20% of the original population. The extent of the recovery of the population before re-harvesting of the native forest coupes depended on the extent of the biodiversity spine. The probability of reaching low absolute population levels (<one or two thousand individuals) varied with the degree of plantation development but not linearly. Increasing the length of the rotation of native forest coupes lowered the probability of reaching low absolute numbers as did increasing the spread of logging of coupes over the length of the rotation rather than logging all of the coupes over a short time span. The model allows the managing authority to design a management scenario that meets a specific quantitative goal, such as a less than 10% probability of numbers falling below 1000 individuals. It also allows them to choose between a mix of different forest management strategies that could all potentially provide the same level of conservation benefit.