BACKGROUND In fragmented landscapes, natural and anthropogenic disturbances coupled with successional processes result in the destruction and creation of habitat patches. Disturbances are expected to reduce metapopulation occupancy for species associated with stable habitats, but they may benefit species adapted to transitory habitats by maintaining a dynamic mosaic of successional stages. However, while early-successional species may be favoured by very frequent disturbances resetting successional dynamics, metapopulation occupancy may be highest at intermediate disturbance levels for species with mid-successional habitat preferences, though this may be conditional on species traits and patch network characteristics. Here we test this 'intermediate disturbance hypothesis' applied to metapopulations (MIDH), using stochastic patch occupancy simulation modelling to assess when does intermediate disturbance favour metapopulation occupancy. We focused on 54 virtual species varying in their habitat preferences, dispersal abilities and local extinction and colonization rates. Long-term metapopulation dynamics was estimated in landscapes with different habitat amounts and patch turnover rates (i.e. disturbance frequency). RESULTS Equilibrium metapopulation occupancy by late-successional species strongly declined with increasing disturbance frequency, while occupancy by early-successional species increased with disturbance frequency at low disturbance levels and tended to level-off thereafter. Occupancy by mid-successional species tended to increase along with disturbance frequency at low disturbance levels and declining thereafter. Irrespective of habitat preferences, occupancy increased with the amount of habitat, and with species dispersal ability and colonisation efficiency. CONCLUSIONS Our study suggests that MIDH is verified only for species associated with mid-successional habitats. These species may be particularly sensitive to land use changes causing either increases or decreases in disturbance frequency. This may be the case, for instance, of species associated with traditional agricultural and pastoral mosaic landscapes, where many species disappear either through intensification or abandonment processes that change disturbance frequency.

中文翻译：

---

物种特征，斑块更新和演替动态：中间干扰何时有利于种群居住？

背景技术在零散的景观中，自然和人为干扰以及演替过程导致栖息地斑块的破坏和产生。预计干扰会减少与稳定栖息地相关的物种的种群分布占有率，但它们可能会通过维持演替阶段的动态拼接而使适应临时栖息地的物种受益。然而，虽然早期成功物种可能会受到频繁发生的重置连续动态的干扰的影响，但对于中中期生境偏好的物种，中等干扰水平的种群迁移占有率可能最高，尽管这可能取决于物种特征和斑块网络特征。在这里，我们测试了适用于后代（MIDH）的“中间干扰假设”，使用随机斑块占有率模拟模型来评估中间干扰何时有利于人口居住。我们关注了54种虚拟物种，这些物种的生境偏好，扩散能力以及局部灭绝和定居率各不相同。在具有不同栖息地数量和斑块周转率（即干扰频率）的景观中，估计了长期的种群动态。结果随着干扰频率的增加，后期成功物种的均衡种群占有率显着下降，而在干扰水平较低时，早期成功物种的均衡种群占有率则随着干扰频率的增加而增加，此后趋于平稳。在低干扰水平下，中等成功物种的占有率随着干扰频率的增加而增加，此后下降。无论栖息地的偏好如何，其占有率都随着栖息地数量的增加以及物种的扩散能力和定殖效率的增加而增加。结论我们的研究表明MIDH仅针对与中继生境有关的物种进行了验证。这些物种可能对土地利用的变化特别敏感，因为土地利用的变化会导致干扰频率的增加或减少。例如，与传统的农业和田园马赛克景观有关的物种可能就是这种情况，其中许多物种通过改变干扰频率的强化或放弃过程而消失。结论我们的研究表明MIDH仅针对与中继生境有关的物种进行了验证。这些物种可能对土地利用的变化特别敏感，因为土地利用的变化会导致干扰频率的增加或减少。例如，与传统的农业和田园马赛克景观有关的物种可能就是这种情况，其中许多物种通过改变干扰频率的强化或放弃过程而消失。结论我们的研究表明MIDH仅针对与中继生境有关的物种进行了验证。这些物种可能对土地利用的变化特别敏感，因为土地利用的变化会导致干扰频率的增加或减少。例如，与传统的农业和田园马赛克景观有关的物种可能就是这种情况，其中许多物种通过改变干扰频率的强化或放弃过程而消失。