Desert ecosystems are increasingly affected by the human-driven
environmental pressures currently changing global landscapes and climates. Plant species are especially susceptible due to their limited ability to move and a frequent reproductive dependence on pollinators that could themselves be displaced. Rising temperatures, aridity and development have already led to decreasing vegetation cover and increasing isolation of desert plant populations (1). It is essential that we identify the limitations climate and landscape already exert on species in these zones, as well as the ability of damaged habitats to recover from disturbance. Using this information we can inform future efforts to protect these fragile ecosystems as damage continues. In my dissertation I address these issues through three parts: First, I document the recovery of an area ravaged by prolonged desertification in the Sahara to determine if recent greening represents true reestablishment of lost habitat, or simply growth within refugial patches that were never lost during drought. Second, I examine reproductive limits in a representative shrub species in the Mojave, Acacia greggii A Gray. In A. greggii I determine if variation in pollinator availability or climate currently influences seed production across The Mojave National Preserve. Third, I identify movement of this species to determine if current climate or landscape features limit dispersal. I have shown that decreased environmental pressure in the Sahara has not guaranteed recovery for damaged areas. In the Mojave I have shown climate and landscape restrictions on reproduction and movement of Acacia greggii. These results indicate that changing environmental conditions will further restrict species in these areas and emphasize the need for swift action to halt possibly irreversible damage.

Detecting landscape change in a desert environment:

The last 100 years have seen drastic environmental changes in the Sahara and Sahel where a prolonged drought beginning in the 1970s led to pronounced vegetation loss. Previous work with coarse scale remote sensing imagery has shown dramatic expansion of the Sahara during the drought (1). However, more recent work shows a greening trend since the mid-1990s (1). I sought to determine if this indicates actual recovery of lost vegetative area, or greening of refugial patches that were never lost during the drought. To do this, I examined a 32,000 km2 area in The Islamic Republic of Mauritania reported to have a recent increase in vegetation cover. I used high-resolution satellite images collected every 15 days from 1990 to 2010. Like previous work, I found vegetation recovery in this region, but significant patterns of continued loss are ongoing in areas away from refugia.

Reproductive limitations of Acacia greggii:

Desert ecosystems are patchy landscapes where pollinator availability is critical to the reproductive success of many flowering plant species. Previous work has shown disruption of synchrony in plant-pollinator interactions, and plant population decline in response to climate change and habitat fragmentation (2, 3). To identify the current constraints that variable climate and pollinator availability have on species in the Mojave, I examined the reproduction of A. greggii. This species may be unaffected by existing variation in climate and pollinators, indicating little impact if either were to significantly change. If, however, a relationship does exist, it would demonstrate susceptibility to increased environmental pressure. I measured the average number of seeds produced per inflorescence on 105 trees at 21 sites in The Mojave National Preserve. At all study sites I concurrently surveyed pollinator abundance. Study sites varied in elevation, which is sharply correlated with climate, and can thus represent climatic variance. Both climate and pollinator abundance significantly influenced reproductive success. These findings illustrate a negative relationship with lower elevations on reproduction. As temperature and aridity increase in this area distribution of this species will become more restricted to higher elevation.

Movement of a Mojave Desert shrub at multiple spatial scales:
Identifying past and present movement patterns is essential to understanding the influence of a changing landscape and climate on the viability of a species. The rate of movement determines the ability of a species to reestablish a damaged population after disturbance, or maintain genetic variation when populations are separated. Acacia greggii occurs across a broad elevation gradient, and has populations separated or defined by distinct geographic features. It is possible, that this species has succeeded due to its ability to disperse widely in spite of these factors. If, however, its movement is affected by them, it will indicate susceptibility to future change. Using molecular markers I developed in the lab, I addressed these alternative hypotheses for A. greggii within the Mojave National Preserve. To identify broad and fine-scale influences Isampled individuals at two spatial scales. I genotyped 460 trees from 23 regional sites (6,200km2), and 200 trees along three parallel dry washes (8km2).

Additionally, to estimate the contribution of pollen movement to the adult pattern, I genotyped 240 seeds collected from 20 mothers within two of these washes.For these genotypes I related the degree to which distance, climate, and dominant landscape features related to genetic variability. I found little genetic differentiation between the 23 regional sites sampled, indicating a history of frequent migration events even at the greatest distances. The genetic variation in these individuals, as well as those samples within the three washes, was highly correlated with elevation, again used as a proxy variable for climate. This indicates a strong climatic influence either directing movement or exerting selective pressure on A. greggii. The model to explain genetic variance of trees in the washes was improved by incorporating the wash within which the tree was placed as an explanatory variable. This indicates a restriction to movement between parallel washes where A. greggii does not occur. We observed this same pattern in the pollen flow data which shows a strong directional movement pattern along washes. This suggests that for A. greggii pollen, dry-washes are serving as corridors for movement. Development in the Mojave Desert has already shifted drainage patterns, thus altering avenues of movement for this and similar species. Additionally, increasing aridity in this zone will likely push movement patterns into higher elevation further restricting range.

1) Anyamba, A. and C. J. Tucker (2005) Journal of Arid Environments 63(3): 596-614.
2) Agular, R. et al. (2006) Ecology Letters 9: 968-980.
3) McCarthy, J. (2002) Conservation Biology 15 (2): 320-331.