Flather Curtis H, Bevers Michael
U.S. Department of Agriculture Forest Service, Rocky Mountain Research Station, Fort Collins, Colorado 80526, USA.
Am Nat. 2002 Jan;159(1):40-56. doi: 10.1086/324120.
A discrete reaction-diffusion model was used to estimate long-term equilibrium populations of a hypothetical species inhabiting patchy landscapes to examine the relative importance of habitat amount and arrangement in explaining population size. When examined over a broad range of habitat amounts and arrangements, population size was largely determined by a pure amount effect (proportion of habitat in the landscape accounted for >96% of the total variation compared to <1% for the arrangement main effect). However, population response deviated from a pure amount effect as coverage was reduced below 30%-50%. That deviation coincided with a persistence threshold as indicated by a rapid decline in the probability of landscapes supporting viable populations. When we partitioned experimental landscapes into sets of "above" and "below" persistence threshold, habitat arrangement became an important factor in explaining population size below threshold conditions. Regression analysis on below-threshold landscapes using explicit measures of landscape structure (after removing the covariation with habitat amount) indicated that arrangement variables accounted for 33%-39% of the variation in population size, compared to 27%-49% for habitat amount. Thus, habitat arrangement effects became important when species persistence became uncertain due to dispersal mortality.
一个离散反应扩散模型被用于估计栖息在斑块状景观中的假设物种的长期平衡种群数量,以检验栖息地数量和布局在解释种群规模方面的相对重要性。当在广泛的栖息地数量和布局范围内进行考察时,种群规模在很大程度上由纯粹的数量效应决定(景观中栖息地的比例占总变异的96%以上,而布局主效应占比不到1%)。然而,当覆盖率降至30%-50%以下时,种群反应偏离了纯粹的数量效应。这种偏离与一个持续阈值相吻合,表现为支持可行种群的景观概率迅速下降。当我们将实验景观划分为“高于”和“低于”持续阈值的集合时,栖息地布局在解释低于阈值条件下的种群规模时成为一个重要因素。对低于阈值的景观进行回归分析,使用明确的景观结构测量方法(去除与栖息地数量的协变量后)表明,布局变量占种群规模变异的33%-39%,而栖息地数量占比为27%-49%。因此,当由于扩散死亡率导致物种持续性变得不确定时,栖息地布局效应变得重要起来。