- We are at risk of losing the sagebrush steppe in the floristic Great Basin to the invasion of Bromus tectorum L., cheatgrass. The floristic Great Basin includes the Central Basin and Range, the Northern Basin and Range, and the Snake River Plain. The Great Basin receives most of its precipitation as winter snow and experiences hot and dry summers. Early accounts of invasion by cheatgrass associated it with farming and grazing practices. The non-farmed areas in the region are still actively grazed and referred to as rangelands. On invaded sites, cheatgrass changes the flammability of fuels on invaded landscapes, across the Great Basin, from coarser fuels that are widely spaced to fine fuels that are continuous, filling interspaces between perennial plants. The fuel load created by cheatgrass regenerates annually. This has resulted in a change in the fire regime of the Great Basin from infrequent, small fires to more frequent large fires.
In arid lands globally, soil interspaces between perennial plants are typically filled by biological soil crusts (biocrusts). This is also true for ecoregions in and surrounding the Great Basin. Biocrusts are known to influence many ecosystem processes that cheatgrass influences, specifically nutrient cycling and availability of soil moisture.
However, little work has been done on biocrusts of the Great Basin and to my knowledge, no one had restored biocrusts within the Great Basin. I attempt to fill some of this knowledge "interspace" by relating biocrust presence to disturbances and cheatgrass invasion and to demonstrate the potential for biocrust restoration within this region.
Previous work in eastern Oregon demonstrated relationships between declines in biocrusts and increases in cheatgrass with increasing grazing intensity, soil temperature, and decreasing soil moisture. Grazing intensity influences the cover of biocrusts as well as the abundance and composition of native bunchgrasses. Native bunchgrasses influence the interspace gap size between perennial herbaceous vegetation which is directly associated with the cover of cheatgrass. In a region where grazing records may be incomplete and may exist in various forms of data, having a simple indicator of grazing impacts would be useful. It is also crucial that we have an understanding of what leads to loss of site resistance to cheatgrass. This previous work suggested that cover of biocrusts, in addition to bunchgrass composition, were associated with increased site resistance to cheatgrass.
In Chapter 2, I used current grazing records from a range of suspected grazing intensities, to examine the ability of both biocrusts and perennial vegetation to maintain site resistance to cheatgrass after fire. I examined the ability of mosses and lichens to maintain site resistance separately given that these are two very different kinds of organisms. Mosses are non-vascular plants and early colonizers of sites in primary succession. Lichens have a symbiotic relationship between a fungus and a photosynthesizing partner, a cyanobacteria, an algae or both. Using structural equation models, I corroborated that perennial vegetation and lichens are associated with increased site resistance to cheatgrass and that mosses are associated with and may facilitate both lichens and perennial herbaceous vegetation. Also in Chapter 2, I identified that burned sites were associated with increased grazing pressure by
livestock as shown by increases in cow dung density and increases in gap size between perennial herbaceous vegetation.
The Great Basin is managed for cover of perennial vegetation but it could also be managed for morphogroups of biocrusts. Considering morphogroups of biocrusts, which were shown in the Chapter 2 to be important for site resilience and resistance, I wanted to determine if there were site characteristics associated with biocrust distribution and recovery from disturbance, across the Great Basin. Outside of the Great Basin on the Columbia Plateau, others had found that mosses were still present on disturbed sites whereas lichens were often lost. In addition, biocrust species were more associated with soil properties than with grazing by livestock. Given that grazing by livestock and fire are common disturbances across the region, I wanted to know if the same relationships between biocrusts, soil properties and disturbance were true in the Great Basin. I found that cover of the lichen component of biocrusts was higher on sites that were both ungrazed and unburned. Factors related to disturbance characteristics were correlated with the recovery of biocrusts, even after accounting for time since fire. Factors related to disturbance, a composite of grazing and fire, were more important for structuring the cover and composition of morphogroups as opposed to environmental conditions. Lichens were the most sensitive morphogroup, compared to tall mosses, followed by short mosses which were favored by some disturbance but reduced in cover immediately after fire. Perennial grasses were also favored by some disturbance and perennial forbs did not show an obvious relationship with a disturbance gradient. Chapter 3 highlights that grazing by livestock and fire are common disturbances across the region so much so that the effects of one on the abundances of morphogroups could not be separated from the other.
Given the observed contributions of biocrusts to site resilience and resistance, I wanted to know if we could restore biocrusts in the field. Others have grown mosses in a lab setting but this was the first study to restore mosses in the Great Basin. I tested the influence of factors that are commonly used in the field of restoration for facilitating plant establishment. I tested the influence of season of inoculation (fall versus spring), the addition of organic matter (in the form of jute net), irrigation (in the spring season) and the climatic setting of moss the collection sites (for moss propagation), in comparison to the experiment site (warm, dry versus cool, moist) on moss growth. I used two moss species: a ruderal (Bryum argenteum) and a later successional species (Syntrichia ruralis). Moss cover increased when the climatic setting of the collection site matched the experiment site. Mosses were facilitated by the addition of the organic jute netting, putting on most of their growth in winter. Although there is still a great deal of work to be done developing moss material for restoration and working out inoculation rates of moss fragments, similar to seeding rates, land managers have another tool to consider when rehabilitating sites after disturbance.
Managing the Great Basin for biocrusts in the presence of grazing and fire will not only increase site resistance to cheatgrass but it will add to the conservation of ecosystem functions related to nutrient cycling, hydrologic cycling and soil erosion. Site resistance will be improved with increased periods of rest from grazing following fire. The lichen component of biocrusts is a more sensitive indicator of disturbance when compared with mosses or perennial vegetation but we are currently actively managing for perennial vegetation and not biocrusts. The moss component of biocrusts can be successfully restored in the Great Basin, without irrigation. This dissertation shows that land managers should consider a suite of organisms, in addition to perennial plants to achieve management goals and maintain site resistance to cheatgrass.