Our results suggest that within the sagebrush steppe of the INL area, alpha, beta, and phylogenetic beta diversity among sites are much more strongly related to physical disturbance than to fire. That phylogenetically distinct groups of plants occupied different ends of the disturbance-stability gradient (Fig. 6; lower left panel) suggests that a plant's ability to tolerate a particular physical-disturbance regime is more likely inherited than rapidly adapted to, which is evidence of phylogenetic niche conservatism (e.g., Donoghue 2008). Our results also suggest that observations of higher level taxa such as Old versus New World Astragalus apply generally to the sagebrush flora of the INL regions and perhaps elsewhere. Higher-level taxa generally have a predilection to a particular disturbance regime (Fig. 6, lower left panel). Consequently, selection of species for restoration purposes should focus on lineages preadapted to specific disturbance regimes, in addition to gene pools (e.g., Jones and Monaco 2009) and competitive genotypes (Goergen et al. 2011).
Fire disturbance was not detected as strongly influencing patterns of alpha, beta, and phylogenetic beta diversity (compare Figs. 4, 5; Fig. 6, right two panels), which suggested that fire represents a disturbance regime to which many sagebrush plants can easily adapt, a finding also found for plant lineages inhabiting the fire-prone South American cerrado (Simon and Pennington 2012; Hughes et al. 2013). Our results agree with Ratzlaff and Anderson (1995), who suggested that postfire recovery of sagebrush steppe to a high diversity of native herb, grass, and shrub species does not require augmentation by reseeding programs because the physical disturbance involved may impede natural recovery. It may well be the case that alpha, beta, and phylogenetic beta diversity of postfire plant communities in the sagebrush steppe is more dependent on the prefire physical-disturbance regime than on fire itself (Fig. 6; compare two lower panels), a finding that agrees with that of Seefeldt and McCoy (2003).
Plant biodiversity in the sagebrush steppe
This study represents one of the few with a focus on plant diversity in the sagebrush steppe. Welch (2005) overviewed the impacts of sagebrush steppe fragmentation on the biodiversity of mainly animals. Anderson and Inouye (2001) analyzed long-term plant diversity data from the INL sagebrush and found an increase in both alpha and beta diversity over a 45 year period, which suggested that as sagebrush steppe recovers from physical disturbance (i.e., early 1900s cropping and grazing), it tends toward a dynamic biodiversity equilibrium and not a static climax community. They found that the plant diversity of the kind harbored by the INL sagebrush steppe was largely resistant to colonization of introduced species and dominance of Bromus tectorum (cheatgrass). Our results are in agreement, especially for sagebrush steppe away from the well-maintained roads and overgrazed rangelands.
We found alpha diversity of all species to be relatively unaffected by physical or fire disturbance (Fig. 3), which agrees with Seefeldt and McCoy (2003). Alpha diversity of native plant species, however, was most affected by physical disturbance (Fig. 4). Indeed, the general increase in alpha diversity detected for many plant functional groups and growth forms with respect to infrequent physical disturbance regimes (Fig. 4) is likely related to the increase in native plant diversity. This is suggested by plants groups with theoretically alternative traits, such as plant groups marked by airborne propagules versus gravity-dispersed propagules, which are both more diverse in settings with an infrequent physical disturbance regime.
In contrast to alpha diversity, patterns of beta and phylogenetic beta diversity (e.g., Fig. 7) are more informative about the profound effects of physical disturbance. The turnover of species and higher taxa is much greater among transects belonging to different physical-disturbance categories (Fig. 6; PCO axis 1 in both left-hand panels) than to different fire categories (Fig. 6; PCO axis 2 in both right-hand panels).
Figure 7. Map of sagebrush study sites (Lavin et al. 2013 and M. Lavin, unpubl. data) along a megatransect that bisects the sagebrush biome from the foothills of the Carson Range near Reno, Nevada, northeast to the Charles M. Russell National Wildlife Refuge, Phillips County, Montana. Climate diagrams are arranged from left to right: Reno (Thomas Creek, Nevada, 1672 m, −119.83020, 39.39801), Boise (Table Rock, Idaho, 983 m,−116.15380, 43.60524), Idaho National Laboratory (INL) (near Atomic City, Idaho, 1509 m, −112.85510, 43.47758), Monida (Monida Pass, Montana, 2071 m, −112.37410, 44.65242), CMR (Reynolds Road, Montana, 786 m, −107.67470, 47.75141). For the climate diagrams, bars = mean monthly precipitation (right-hand y axis), upper solid line = mean monthly temperature highs, and lower dashed line = mean monthly temperature lows (left-hand y axis). January through December are indicated numerically along the x axis. Bioclimatic variable 11, mean temperature during the coldest quarter, is contoured on the map. This contour and the climate diagrams suggest that the INL region is more climatically similar to sagebrush areas to the north and east than to the south and west.
Download figure to PowerPoint
The Intermountain Flora (Cronquist et al. 1972; to Holmgren et al. 2012) centerpieces the high levels of plant diversity and endemism in an area of western North America that is dominated by short-statured sagebrush. In contrast, the sagebrush steppe has been reported to harbor only moderate levels of plant diversity (e.g., West and Young 2000). Perceptions of low to moderate plant diversity in the sagebrush steppe may be a product of most studies involving degraded sagebrush steppe or an analytical focus that ignores rare species (e.g., <1–5% cover; e.g., Davies et al. 2012; McCune and Grace 2002). Our results agree with The Intermountain Flora and underscore the high levels of plant diversity in high-native-cover sagebrush steppe (e.g., Fig. 3). Furthermore, such high levels of plant diversity can be fully utilized in a community phylogenetic analysis because the many rare or singleton species (often herbs) will have close relatives among the total sample of species, and thus will be informative of regional patterns of phylogenetic beta diversity. This is illustrated by our finding that more variation was captured along the first two PCO axes when MNT rather than Euclidean distances were used (Fig. 6).
Ecological stability of the sagebrush steppe
Native diversity was higher in more stable sites and lower in more physically disturbed sites. This may be a reflection of the sagebrush steppe representing one of the most ecologically stable forms of northern latitude vegetation under historic disturbance regimes and, therefore, not prone to immigration by colonizing species. Climate data extracted from WorldClim (Hijmans et al. 2005) revealed that sagebrush steppe is either in water deficit or frozen for much of the year (Fig. 7; climate diagrams). Long-lived perennial plant species are common to this type of sagebrush steppe (e.g., Wambolt and Watts 1996; West and Young 2000; Wambolt and Hoffman 2001; Cooper et al. 2011) and this axiomatically suggests that they have evolved to survive in an environment where the favorable season is short, erratic, and unpredictable. Regular or occasionally profound physical disturbances would lower the probability of sagebrush-steppe-adapted plants amortizing the cost of growing new leaves and other photosynthetic organs when climatic conditions do not allow for much of a growing season. Such a climatically harsh environment that includes a high rate of physical disturbance would favor a completely different suite of species (e.g., a diversity of annual Amaranthaceae, Brassicaceae, and Poaceae) than what predominates in high-native-cover sagebrush steppe.
In contrast, introduced plant species diversity was not as well explained by the physical disturbance gradient, with introduced species occupying both roadside and high-native-cover sagebrush areas. We found introduced species to be abundant and diverse in physically disturbed settings, but they were also diverse, though were less abundant, in the stable sagebrush steppe. Within high-native-cover sagebrush steppe, small patches of regular physical disturbances (e.g., localized small-mammal burrowing, trampling, grazing, etc.) harbor a diversity of introduced species (e.g., annual Brassicaceae), as well as natives plants (e.g., annual species of Cryptantha, Lappula, and Mentzelia). Regularly disturbed roadsides may well be serving as conduits for invasion into small disturbed patches within the stable sagebrush matrix. However, stable patches do not similarly exist in regularly disturbed roadsides and rangeland, so native species characteristic of stable conditions do not often invade (or reinvade) these disturbed ecologies, especially those well-traveled and well-grazed. This explains why native species diversity tracks physical disturbance much more strongly than introduced species diversity in the sagebrush steppe.
These findings have distinct implications concerning the choice of species used in restoration projects. Physical and fire disturbances continue to push sagebrush steppe across thresholds from which restoration to a predisturbance state is becoming increasingly difficult (e.g., Davies et al. 2011) and has prompted the consideration of primary, secondary, tertiary, and quaternary gene pools of potential restoration species (e.g., Jones and Monaco 2009). Primary gene pools come from locally adapted populations of the candidate restoration species whereas the quaternary restoration gene pool comes from a surrogate species of the same functional group when the original species is no longer adapted to the modified (degraded) environment. If physical disturbance in short-statured sagebrush steppe imposes an evolutionary impediment to adaptation as our results suggest, then species pools also should be considered when choosing plant material for restoration. In order for the expectations of restoration outcomes to be realistic, practitioners must consider the evolutionary limitations of plant lineages that disturbance regimes impose, and either choose species that can inhabit those conditions, or drastically alter the disturbance regime to favor a level of stability that is associated with the desired species that often represent the restoration targets.
Plant taxa that represent reclamation and restoration targets of physically disturbed sagebrush steppe have been enumerated briefly Lavin et al. (2013). Suffice it to say that restoration benchmarks for sagebrush steppe should include a diversity of shrubs in the Amaranthaceae and Asteraceae, succulents in the Cactaceae and Crassulaceae, and Portulaceae, native bunchgrasses belonging to the tribes Stipeae (needlegrasses) and Triticeae (wheatgrasses), and herbaceous species including New World Astragalus, Orobanchaceae (especially Castilleja, Cordylanthus, and Orthocarpus), Eriogonum and Oxytheca (Polygonaceae), dry-adapted Apiaceae (e.g., Cymopterus, Lomatium, Perideridia, and Pteryxia), and perennial Cryptantha (Boraginaceae). Because of ecosystem stability requirements, these taxa may not be the best candidates for initial introduction into a degraded sagebrush steppe. More disturbance-tolerant taxa affiliated with open dry environments might serve this purpose better. Such taxa could include root-sprouting shrub species of the genera Artemisia, Chrysothamnus, Ericameria, and Tetradymia (e.g., Davies et al. 2011). Such potential candidates for reclamation would most likely respond to the immediate attentions of reclamation efforts.
The idea that communities with more species are more resistant to invasion is supported by our results. Phylogenetic beta diversity patterns suggest that high-native-cover sagebrush steppe is resistant to invasion by higher-level taxa such as Cleomaceae, Euphorbiaceae, introduced Amaranthaceae, and introduced and native species of the family Fabaceae and the genus Bromus. Bromus, according to our studies (e.g., Lavin et al. 2013), for example, is a higher-level taxon with a predilection for disturbance-prone settings. Bromus inermis and B. japonicus, for example, are found along the paved roads in the INL region but are rarely if ever found in the adjacent INL sagebrush steppe. Even Bromus tectorum, although occurring through much of the INL area, is often at low abundance within high-native-cover sagebrush steppe, a finding consistent with Anderson and Inouye (2001). We suggest that the mechanism of resistance to invasion is not that the sagebrush steppe is more diverse, as postulated by the diversity-invasibility hypothesis (e.g., Fridley 2011), but that it is more stable. The diversity-invasibility hypothesis suggests that new species are precluded because they cannot find resources to utilize because all the niches are filled. The results of our study suggest that colonizing species cannot successfully immigrate and replace the plant species adapted to the sagebrush steppe because they are not adapted to stable conditions and thus end up being precluded from the species pool. At higher taxonomic levels, most Fabaceae and all Cleomaceae and Euphorbiaceae, for example, will not thrive in the stable sagebrush steppe because such a setting does not cater to colonizing and disturbance-tolerant species that respond opportunistically to the immediate growing conditions.
The introduction of crested wheatgrass (Agropyron cristatum) is considered as an impediment to the restoration of sagebrush steppe. It can potentially resist grazing (e.g., Ganskopp et al. 1992) and dominate the seed bank (e.g., Marlette and Anderson 1986) in vegetation that otherwise would be characteristic of sagebrush steppe. Where the probability of restoration success is low and the threat of annual grass invasion high in Wyoming sagebrush steppe, crested wheatgrass has been suggested as a practical restoration alternative (Asay et al. 2001). Davies et al. (2011) go so far as to suggest that ecological processes in sagebrush plant communities may not be substantively changed when native perennial bunchgrasses are replaced with crested wheatgrass. Our results suggest that a slightly more nuanced view is required.
Our sampling of the INL green strips suggests that where livestock grazing is rare if at all present (as in the INL), crested-wheatgrass-dominated vegetation may provide the stability necessary for a diversity of Asteraceae and Amaranthaceae shrubs to assemble. Although green-strip transects were generally low in plant diversity and thus low in many plant functional types, they were notably diverse in shrub species (Fig. 4; panels labeled “shrub species” and “Artemisia sect. Tridentatae”). Overall, the plant composition and diversity of green-strip transects was low and very different from those of the high-native-cover sagebrush steppe with respect to Euclidean distances (Fig. 6; upper left, dark green open circles). With MNT distances, green-strips harboring a diversity of shrubby Asteraceae and Amaranthaceae were not as different from the sagebrush steppe transects relative to some of the paved road transects (Fig. 6; lower left, dark green open circles). These results suggest that if crested-wheatgrass-dominated vegetation is subjected to little if any livestock grazing, a diversity of semi-arid-adapted shrubs including those palatable Amaranthaceae can successfully immigrate and establish in this ecological setting, whereas exotic annuals including Bromus tectorum cannot. Perhaps stands of Agropyron cristatum, in cases of highly degraded sagebrush steppe, can give the required levels of stability needed to restore Wyoming big sagebrush steppe. From a management perspective, the duration of that stability would have to encompass many decades if not over a century especially for Wyoming big sagebrush steppe (e.g., Anderson and Inouye 2001; Monsen 2004; Cooper et al. 2011; Davies et al. 2011; Kay and Reid 2011) and presumably for other short-statured forms of this vegetation.