Limits to upward movement of subalpine forests in a warming climate.

Rapid changes to the earth’s climate are well underway, already influencing biological systems across the globe. Among the key impacts are potential shifts in the geographic range of many species or ecosystems, often predicted to move either poleward or upward in response to warmer conditions (1, 2). In mountain biomes, as upper elevations become more climatically hospitable to tree establishment, upward shifts in tree cover could cause alpine ecosystems (nonforest vegetation above treeline) to become both reduced in extent and increasingly isolated into smaller “sky islands.” However, currently we know relatively little about the actual nature of such shifts in treeline, or to what extent a warming climate alone will generate them. In PNAS, Macias-Fauria and Johnson (3) make an important step in quantifying the relative influence of temperature regime and geomorphic factors in determining current and future treeline across a large Rocky Mountain landscape. The authors’ data suggest that upward treeline shifts in a warming climate may be heavily constrained by geologic factors that influence the availability of growing substrates at high elevations, leading to much less, or at least much slower, tree colonization into alpine areas than predicted by climate alone. Macias-Fauria and Johnson (3) show that large extents of the alpine zone can be dominated by extremely steep slopes, cliffs, avalanche chutes, and exposed bedrock: that is, places where trees rarely can grow (Fig. 1). Prior efforts to model treeline shifts have focused primarily on the envelope of temperature and moisture conditions in relation to the physiological requirements of subalpine trees, with little consideration given to the underlying geomorphic template (topography, soils) upon which trees must ultimately grow. Although warming temperatures have been linked to tree expansion into alpine areas with relatively gentle topography and suitable soils (4, 5), such sites can be relatively limited in many high-mountain landscapes. As such, movement of woody vegetation into some alpine areas may occur over millennial rather than decadal time scales (i.e., periods relevant to the weathering and erosion of mountain surfaces rather than the reproduction of trees). This intuitive concept has been recognized by many researchers (6, 7), but there have been few studies quantifying the degree to which factors other than climate might drive spatial variability in treeline migration over large landscapes. The quantitative nichemodeling approach applied byMacias-Fauria and Johnson (3), incorporating both geomorphology and climate, moves us toward more realistic and sophisticated predictions of potential shifts in mountain treelines. More broadly, such efforts fit into a larger context of understanding the diverse set of factors that influence biogeographic shifts under a changing climate, many of which are not directly related to the climatic “envelope” in which species grow. Other subfields of ecology have long recognized these other factors, perhaps most notably in the conservation biology realm. For example, the ability of threatened plant and animal species to migrate latitudinally in response to changing conditions may be limited by the low connectivity of many human-modified landscapes (8), prompting a long-running and spirited debate as to whether we should engage in assisted relocation efforts (9, 10). Fundamental to this debate is an awareness that not only climate, but also the landscape template, will affect species’ movements. Like most responses to changing climate, treeline dynamics emerge from a series of complex and interacting drivers, each of which may have its own response to a changing environmental context. In addition to the critical factors of climate and geomorphology (3), altered disturbance regimes and