Natural analogues of degraded ecosystems enhance conservation and reconstruction in extreme environments.

Ecosystem rehabilitation strategies are grounded in the concept that coexisting species fit their environments as an outcome of natural selection operating over ecological and evolutionary timescales. From this perspective, re-creation of historical environmental filters on community assembly is a necessary first step to recovering biodiversity within degraded ecosystems; however, this approach is often not feasible in severely damaged environments where extensive physiochemical changes cannot be reversed. Under such circumstances management goals may shift from restoring historical conditions to reconstructing entirely new ecosystems or replicating natural ecosystems that may be locally novel but of regional conservation importance. This latter goal may be achieved by introducing to damaged sites species already adapted to filters maintaining the degraded state, through targeting assemblages from natural ecosystems biophysically analogous to the degraded state, here termed "degraded-state analogue" (DSA) ecosystems. This hypothesis predicts that, in high-stress sites where recruitment of previous inhabitants is strongly microsite-limited, DSA species will be primarily propagule-limited; furthermore, communities invaded by DSA species should shift in structure to reflect properties associated with high-value DSA target ecosystems. We tested these predictions by experimentally sowing long-abandoned limestone quarry floors with 18 perennial grass and forb species characteristic of rare natural limestone pavements called "alvars." Alvar species established successfully under a range of microsite conditions manipulated to alter suspected constraints on colonization, including nitrogen deficiency, excessive CaCO3, and competition with weeds. Alvar species performed equivalently to seeded weed species known to thrive on quarry floors. Resident communities doubled in species richness following alvar species addition, supporting 17-20 species/0.18 m2 (95% confidence interval) and providing refuge to regionally restricted or threatened species including Iris lacustris, Solidago ptarmicoides, and Liatris cylindracea. In contrast, maximum-diversity reference plots on a pristine alvar supported 20-23 species/0.18 m2. Strong propagule limitation but weak microsite constraints on quarry colonization by alvar species combined with establishment of species-rich communities comparable to natural alvar biodiversity hot spots confirms that targeting DSA assemblages in ecosystem reconstruction can promote both efficient site colonization and ex situ biodiversity conservation within difficult-to-restore anthropogenic wastelands.

Influences of microhabitat constraints and rock-climbing disturbance on cliff-face vegetation communities.

Many researchers report that rock climbing has significant negative effects on cliff biota. Most work on climbing disturbance, however has not controlled for variation in microsite characteristics when comparing areas with and without climbing presence. Additionally, some researchers do not identify the style or difficulty level of climbing routes sampled or select climbing routes that do not represent current trends in the sport. We solved these problems by sampling climbing areas used by advanced "sport" climbers and quantifying differences in microtopography between climbed and control cliffs. We determined whether differences in vegetation existed between pristine and sport-climbed cliff faces when microsite factors were not controlled. We then determined the relative influence of the presence of climbing, cliff-face microtopography, local physical factors, and regional geography on the richness, abundance, and community composition of cliff-face vascular plants, bryophytes, and lichens. When we did not control for microsite differences among cliffs, our results were consistent with the majority of prior work on impacts of climbing (i.e., sport-climbed cliff faces supported a lower mean richness of vascular plants and bryophytes and significantly different frequencies of individual species when compared with pristine cliff faces). When we investigated the relative influences of microtopography and climbing disturbance, however the differences in vegetation were not related to climbing disturbance but rather to the selection by sport climbers of cliff faces with microsite characteristics that support less vegetation. Climbed sites had not diverged toward a separate vegetation community; instead, they supported a subset of the species found on pristine cliff faces. Prior management recommendations to restrict development of new climbing routes should be reevaluated based on our results.

Experimental separation of resource quantity from temporal variability: seedling responses to water pulses.

We tested the hypothesis that higher temporal variability in water supply will promote higher species richness of germinating and surviving seedlings using assemblages of 70 species of herbaceous plants from limestone pavement habitats. In a two-factor greenhouse experiment, doubling the total volume of water added led to greater germination (measured as number of germinated seeds and species) and establishment (survival and biomass) but the effects of temporal variability depended on the response variable considered. Low pulse frequencies of water addition with total volume added held constant resulted in greater temporal variability in soil moisture concentration that in turn promoted higher density and richness of germinated seedlings. Low pulse frequencies caused an eight-fold greater mortality in the low total volume treatment and biomass production to decline by one-third in the high total volume treatment. The effects of increasing temporal variability in water supply during recruitment stages can thus be opposite on different components of plant fitness and may also depend on total resource quantity. While greater species richness in more temporally variable soil moisture conditions was attributable to sampling effects rather than species-specific responses to the water treatments, species relative abundances did vary significantly with temporal variability. Changes in the amplitude or frequency of resource fluctuations may alter recruitment patterns, and could have severe and relatively rapid effects on community structure in unproductive ecosystems.

THE EFFECT OF WINTER FIELD CONDITIONS ON THE DISTRIBUTION OF TWO SPECIES OF UMBILICARIA: I. CO2 EXCHANGE IN RECIPROCALLY-TRANSPLANTED THALLI.

Two lichen species of the genus Umbilicaria were transplanted into each other's habitat in order to determine the role of winter field conditions in regulating distribution patterns. U. vellea (L.) Ach., which normally grows in a steeply inclined, snow free habitat, was transplated into the level-ground, snow-covered habitat of U. deusta (L.) Baum. The opposite was done for U. deusta. Suitable controls to test the effect of transplanting were also run for both species. The reciprocal transplantation was done in each of two years, one of which had a normal snow load, the other which had almost no snow cover. The effect of reciprocal transplantation was to reduce significantly the carbon fixing ability in U. vellea in the year with the normal snow-load. Transplantation had no effect on U. deusta. The effects on CO2 exchange were only slight for U. vellea in the year with the lesser snow-load. It is therefore concluded that existing winter field conditions, and especially the nature of the snow cover, maintain the existing distribution pattern of the lichens in the field.