The Walker and Syers model of phosphorus (P) transformations during pedogenesis is widely accepted for the development of humid ecosystems, but long-term P dynamics of more arid ecosystems remain poorly understood. We tested the Walker and Syers model in semiarid piñon-juniper woodlands by measuring soil P fractions under tree canopies and in intercanopy spaces along a well-constrained, approximately 3000 ka (1 ka = 1000 years) volcanic substrate age gradient in northern Arizona, USA. The various pools of soil P behaved largely as predicted; total soil P and primary mineral P declined consistently with substrate age, labile inorganic P increased early in soil development and then declined at later stages, and organic phosphorus increased consistently across the chronosequence. Within each site, soils under tree canopies tended to have higher concentrations of labile and intermediately available P fractions compared to intercanopy soils. However, the degree of spatial heterogeneity conferred by tree islands was moderated by the stage of soil development. In contrast, tree islands had no influence on within-site distribution of more recalcitrant soil P pools, which appear to be controlled solely by the stage of pedogenesis. Coincident with declines in total P, primary mineral P, and labile inorganic P, we found that phosphatase enzyme activity increased with substrate age; a result consistent with greater ecosystem-level P demand on older, more highly weathered substrates. Our results suggest that, compared to humid climates, reduced inputs of water, energy, and acidity to semiarid ecosystems slow the rate of change in P fractions during pedogenesis, but the overall pattern remains consistent with the Walker and Syers model. Furthermore, our data imply that pedogenic change may be an important factor controlling the spatial distribution of labile P pools in semiarid ecosystems. Taken together, these data should both broaden and unify terrestrial ecosystem development theory.

The long‐term dynamics of carbon (C) and nitrogen (N) in semiarid ecosystems remain poorly understood. We measured pools and fluxes of surface soil C and N, as well as other soil properties, under tree canopies and in intercanopy spaces at four sites that form a volcanic substrate age gradient in semiarid piñon‐juniper woodlands of northern Arizona, United States. Clay content and soil water‐holding capacity increased consistently with substrate age, but both soil organic C and N increased only up to the 750,000 year site and then declined at the oldest (3,000,000 year) site. Measures of soil C and N flux displayed a similar pattern to total C and N pools. Pools and fluxes of C and N among the three canopy types became more homogeneous with substrate age up to the 750,000 year site, but disparity between tree and intercanopy microsites widened again at the oldest site. The δ15N of both tree leaves and surface soils became progressively more enriched across the substrate age gradient, consistent with a N cycle increasingly dominated by isotope fractionating losses. Our results point to consistencies in patterns of ecosystem development between semiarid and more humid ecosystems and suggest that pedogenic development may be an important factor controlling the spatial distribution of soil resources in semiarid ecosystems. These data should help both unify and broaden current theory of terrestrial ecosystem development.

Observational studies and experimental evidence agree that rising global temperatures have altered plant phenology-the timing of life events, such as flowering, germination, and leaf-out. Other large-scale global environmental changes, such as nitrogen deposition and altered precipitation regimes, have also been linked to changes in flowering times. Despite our increased understanding of how abiotic factors influence plant phenology, we know very little about how biotic interactions can affect flowering times, a significant knowledge gap given ongoing human-caused alteration of biodiversity and plant community structure at the global scale. We experimentally manipulated plant diversity in a California serpentine grassland and found that many plant species flowered earlier in response to reductions in diversity, with peak flowering date advancing an average of 0.6 days per species lost. These changes in phenology were mediated by the effects of plant diversity on soil surface temperature, available soil N, and soil moisture. Peak flowering dates were also more dispersed among species in high-diversity plots than expected based on monocultures. Our findings illustrate that shifts in plant species composition and diversity can alter the timing and distribution of flowering events, and that these changes to phenology are similar in magnitude to effects induced by climate change. Declining diversity could thus contribute to or exacerbate phenological changes attributed to rising global temperatures.

No abstract

No abstract