Data from: Edge influence on vegetation at natural and anthropogenic edges of boreal forests in Canada and Fennoscandia

1. Although anthropogenic edges are an important consequence of timber harvesting, edges due to natural disturbances or landscape heterogeneity are also common. Forest edges have been well-studied in temperate and tropical forests, but less so in less productive, disturbance-adapted boreal forests. 2. We synthesized data on forest vegetation at edges of boreal forests and compared edge influence among edge types (fire, cut, lake/wetland; old vs. young), forest types (broadleaf vs. coniferous) and geographic regions. Our objectives were to quantify vegetation responses at edges of all types and to compare the strength and extent of edge influence among different types of edges and forests. 3. Research was conducted using the same general sampling design in Alberta, Ontario and Quebec in Canada, and in Sweden and Finland. We conducted a meta-analysis for a variety of response variables including forest structure, deadwood abundance, regeneration, understorey abundance and diversity, and nonvascular plant cover. We also determined the magnitude and distance of edge influence using randomization tests. 4. Some edge responses (lower tree basal area, tree canopy and bryophyte cover; more logs; higher regeneration) were significant overall across studies. Edge influence on ground vegetation in boreal forests was generally weak, not very extensive (distance of edge influence usually < 20 m) and decreased with time. We found more extensive edge influence at natural edges, at younger edges and in broadleaf forests. The comparison among regions revealed weaker edge influence in Fennoscandian forests. 5. Synthesis. Edges created by forest harvesting do not appear to have as strong, extensive or persistent influence on vegetation in boreal as in tropical or temperate forested ecosystems. We attribute this apparent resistance to shorter canopy heights, inherent heterogeneity in boreal forests and their adaptation to frequent natural disturbance. Nevertheless, notable differences between forest structure responses to natural (fire) and anthropogenic (cut) edges raise concerns about biodiversity implications of extensive creation of anthropogenic edges. By highlighting universal responses to edge influence in boreal forests that are significant irrespective of edge or forest type, and those which vary by edge type, we provide a context for the conservation of boreal forests.

1. 尽管人为林缘是木材采伐的重要生态后果，但由自然干扰或景观异质性形成的林缘也同样普遍。温带与热带森林的林缘已得到充分研究，但在生产力较低、适应频繁干扰的北方针叶林（boreal forest）中，相关研究仍较为匮乏。
2. 本研究整合了北方针叶林林缘的植被数据，对比了不同林缘类型（火干扰边缘、采伐边缘、湖泊/湿地边缘；老龄林缘与幼龄林缘）、森林类型（阔叶林与针叶林）以及地理区域间的林缘效应差异。本研究的目标为量化各类林缘的植被响应，并比较不同类型林缘与森林的林缘效应强度与影响范围。
3. 研究在加拿大阿尔伯塔省、安大略省、魁北克省以及瑞典、芬兰采用统一的通用采样方案开展。针对一系列响应变量（包括森林结构、枯木丰度、植被更新、林下植被丰度与多样性，以及非维管植物盖度），我们开展了荟萃分析（meta-analysis）。此外，通过随机化检验确定了林缘效应的幅度与影响距离。
4. 部分林缘响应（如较低的林木断面积、林冠盖度与苔藓植物盖度，更多的倒木，以及更高的植被更新率）在所有纳入研究中整体显著。北方针叶林林缘对地面植被的影响通常较弱且范围有限（林缘影响距离通常小于20米），且随时间推移逐渐减弱。我们发现，自然边缘、幼龄林缘以及阔叶林内的林缘影响范围更广。区域对比结果显示，芬诺斯堪的亚地区的森林林缘效应相对更弱。
5. 综合分析表明，与热带或温带森林生态系统相比，森林采伐形成的人工林缘对北方针叶林植被的影响并未达到同等强度、广度与持续性。我们将这一相对抗性归因于北方针叶林较矮的林冠高度、自身固有的景观异质性，以及其对频繁自然干扰的适应性。尽管如此，自然干扰（火）与人为活动（采伐）形成的林缘所引发的森林结构响应存在显著差异，这引发了人们对大规模营造人工林缘的生物多样性影响的担忧。本研究明确了北方针叶林林缘效应中不受林缘与森林类型影响的普遍性响应，以及随林缘类型变化的特异性响应，可为北方针叶林的保护工作提供理论依据。