Doctoral Dissertation Research: Effects of Climate Variation on Tree Mortality Across Spatially Heterogeneous Subalpine Forest Landscapes in the Front Range of Colorado

气候是森林生态系统动态的主要驱动因素，并在多个时空尺度上影响森林群落的树木死亡率。长期和/或严重干旱等气候事件可引发死亡率大幅上升。气候对树木死亡率的影响更为复杂，因为对气候敏感的干扰，如树皮甲虫的爆发，会促进树木的死亡。本博士研究的目标是通过整合来自永久样地的死亡率记录和树木死亡日期的树木年轮重建，以及与干旱和树皮甲虫爆发相关的死亡率模式的空间分析，研究气候变化对科罗拉多州前山脉亚高山森林树木死亡率的影响。重新测量了1980年代初最初安装的10个亚高山永久性森林样地，每个样地树木超过400棵，30个较小的样地平均50棵树，以测定径向生长和树木死亡率。树种包括恩泽曼云杉（Picea engelmannii）、亚高山冷杉（Abies lasiocarpa）、黑松（Pinus contorta）和柔韧松（Pinus flexilis）。从样地内的树木观测到的树木死亡年和树木年轮重建的树木死亡年将与当地和区域气候数据进行统计比较，以辨别与气候事件（例如极端干旱）的时间关联。还将根据树冠间隙的相对位置、树木的大小和生长率以及不同大小和生长率的树木的接近程度来检查样地内死亡率的空间格局。在更广泛的空间尺度上，将研究在前山岭北部亚高山分布的松木死亡率与自然环境空间变异性的关系。对30个地点进行密集采样，将深入了解气候直接导致的斑块大小和死亡率程度，以及山松甲虫（Dendroctonus ponderosae）引起的死亡率。将分析交叉年代的树木死亡年是否同步，并与当地和区域气候数据进行定量比较，以检验与气候事件的时间关联。利用地形信息、土层信息、影像解译的物种边界信息和历史病虫害信息，建立预测区域尺度上松木死亡率的空间逻辑回归模型。鉴于森林形成的长时间尺度，这项研究利用了从永久样地进行的长期观测，这对了解森林人口统计学至关重要。此外，在当前全球变暖趋势的背景下，分析气候变化和与气候有关的昆虫爆发作为树木死亡的因素。在美国科罗拉多州，高山松甲虫对高海拔松木的影响文献记载很少。这就提出了一种可能性，即在过去的大约100年里，目前由甲虫引起的松木死亡率的高水平是前所未有的，这可能是与科罗拉多落基山脉当前变暖趋势相关的模式的先兆。对野生动物来说，林松是一种特别重要的物种，它的种群目前受到干旱、甲虫攻击和水疱锈病感染的威胁。因此，土地管理者需要更好地了解这一物种过去和未来可能的死亡模式，以便为缓解决策提供信息。这项研究的数据将被纳入美国地质勘探局的一项合作计划，该计划旨在调查全球变化对美国西部山区的影响。

Climate is a major driver of forest ecosystem dynamics and operates on several spatial and temporal scales to affect tree mortality in forest communities. Large pulses of mortality can be triggered by climatic events such as prolonged and/or intense drought. Climatic influences on tree mortality are further complicated because tree mortality can be facilitated by climatically-sensitive disturbances such as bark beetle outbreaks. The goal of this doctoral research is to examine effects of climate variation on tree mortality in subalpine forests of Colorado's Front Range by integrating mortality records from permanent plots and tree-ring reconstructions of tree death dates with spatial analysis of mortality patterns related to drought and bark beetle outbreak. Ten permanent subalpine forest plots with greater than 400 trees each and thirty smaller plots with an average of 50 trees initially installed in the early 1980s were re-measured for radial growth and tree mortality. Tree species include Engelmann spruce (Picea engelmannii), subalpine fir (Abies lasiocarpa), lodgepole pine (Pinus contorta) and limber pine (Pinus flexilis). Both observed and tree-ring reconstructed tree death years from trees within the plots will be statistically compared to local and regional climate data to discern temporal associations with climate events (e.g. extreme drought). Spatial patterns of mortality within the plots will also be examined with regard to the relative location of canopy gaps, tree size and growth rate, and proximity of trees of varying sizes and growth rates. At a broader spatial scale, mortality of limber pine will be examined in relation to spatial variability of the physical environment throughout its subalpine distribution in the northern Front Range. Intensive sampling of 30 sites will provide insight into the patch size and extent of mortality directly driven by climate as well as mortality meditated by mountain pine beetle (Dendroctonus ponderosae). Cross dated tree death years will be analyzed for synchrony and quantitatively compared with local and regional climate data to test for temporal associations with climate events. A spatial logistic regression model that predicts limber pine mortality at a regional scale will be developed from topographic information, soil layers, photo-interpreted species boundaries, and historic pest information. This study capitalizes on long-term observations from permanent plots which are crucial to an understanding of forest demography given the long timescales over which forests are shaped. In addition, analysis of climate variation and climate-related insect outbreak as agents of tree mortality situates the study within the context of current global warming trends. There are few documentary records of mountain pine beetle affecting high elevation limber pine in Colorado. This raises the possibility that the current high-level of beetle-induced mortality in limber pine is unprecedented over approximately the past 100 years and may be a harbinger of patterns associated with the current warming trend in the Colorado Rockies. Limber pine is a particularly important species for wildlife, and its populations are currently threatened by drought, beetle attack, and infection by blister rust. Thus, land managers need a better understanding of past and probable future patterns of mortality in this species to inform mitigation decisions. Data from the study will be included in a collaborative USGS initiative aimed at investigating the effects of global change on mountain regions in the western United States.