Soil
"Fragile cryptobiotic crusts, themselves of significant biological interest, play a critical role throughout the Monument, stabilizing the highly erodible desert soils and providing nutrients for plants..."
Presidential Proclamation, September 1996

Cooperative NRCS-BLM Soil Survey
Prior to completion of the 2005 soil survey, soils across much of the Monument had never been formally surveyed. The survey includes descriptions for over 130 individual soil series that comprise approximately 200 map units. The survey area encompasses a wide variety of geologic, physiographic, and climatologic settings (Figure 1). Besides reflecting the ‘state-of-the-art’ in soil survey methods, a key feature of the survey is the correlation between soil map units and ecological sites. This data provides a framework for interdisciplinary research, such as developing and validating state-and-transition models for use in assessing, managing, and monitoring rangelands.

Predictive Modeling of Biological Soil Crusts as a Tool for Improved Range Management
Contact Information: Matt Bowker, Northern Arizona University, Soil Ecology Lab

Biological soil crusts (BSCs) comprise diverse soil surface communities, prevalent in semi-arid regions, which function as ecosystem engineers and perform numerous important ecosystem services. Because these crusts are in decline due to surface disturbance, and their loss is a component and accelerator of desertification, there is a clear need for tools that allow BSCs to be accounted for in range management.

To determine if the potential BSC cover and composition could be predicted for a large area based upon a parsimonious set of GIS data layers (soil functional types, precipitation, and elevation), we sampled low disturbance sites in GSENM. We used classification and regression trees to model cover of four crust types (light cyano-bacterial, dark cyanobacterial, moss, lichen), species richness, one cyanobacterial biomass proxy (chlorophyll a), and indices of N and C-fixation.

The correlation between predicted and observed values for the dark cyanobacterial, moss, lichen and species richness models were moderate to high (R2 = 0.49, 0.64, 0.55, and 0.60 respectively) (Figure 2). In addition, cover of late successional BSC elements (moss +lichen + dark cyanobacterial) was also successfully modeled (R2 = 0.64). Our models of annual N-fixation, maximal C-fixation, whole-site soil stability, surface roughness and cyanobacterial crust soil stability performed moderately well (R2 = 0.38, 0.50, 0.57, 0.59 and 0.40 respectively). We were less successful with light cyanobacterial models (R2 = 0.22), and the models for chlorophyll a had poor predictive power (R2 = 0.09). We believe that our difficulty modeling chlorophyll a concentration in soil is related to the occurrence of a severe drought during the course of the study. It is likely that chlorophyll a photodegraded during this dry time and that BSC cyanobacteria experienced mortality. Variables describing cover of BSC type appeared to be less prone to temporal changes.

Throughout the Colorado Plateau, BSCs relate strongly to each of the three key attributes (soil, hydrologic, biotic) considered in rangeland health assessments (Pellant et al. 2000). Our models allow BSC reference conditions to be quanitifed and thereby facilitate consistent comparisons between the actual condition of BSCs at a given site and their potential condition (Figure 3), so that sites experiencing desertification can be identified and management actions can be taken. Our results also provide a means for identifying sites that should be priorities for conservation, such as those with high species richness of those that provide ecosystem services at high rates.