Site: Jornada Basin LTER
Explanations for unsuccessful attempts to limit shrub expansion using various methods, including livestock removal, have emerged from JRN studies. Current dominance by creosotebush of this exclosure (right panel) built in 1920 on a site dominated by black grama grasslands (left panel) is attributed to spatial processes of redistribution of water, nutrients, and seeds across scales that overwhelmed effects of release from grazing on grass-shrub competition.
USDA ARS, Jornada Experimental Range archive

Much of ecological research in terrestrial systems focuses on detailed understanding of local processes, such as competition for limiting resources, on fine-scale dynamics of individual plots that are extrapolated to broad-scale patterns of ecosystem types and biomes. Although this perspective can explain short-term dynamics at fine scales of plants and patches in drylands, landscape-scale dynamics over decadal time scales often cannot be extrapolated from finer scales of study. Thus, the largest errors in prediction will occur at the scales of most interest to land managers and decision makers.

A synthesis of long-term research at the JRN is providing new insights into these perplexing dynamics. Long-term studies of aboveground net primary production (ANPP) and biodiversity illustrate the importance of nonequilibrium dynamics in response to climate. Results show that a sequence of wet years initiates a sequence of events that results in higher production and biodiversity than expected based on rainfall. Plant-soil water feedbacks and patch scale redistribution of soil, organic matter, and nutrients are key processes governing dynamics at broader scales. These nonlinear increases in herbaceous plants on desertified shrublands are predicted to result in a state change reversal in the future under a directional increase in precipitation. These responses are also contingent on environmental conditions and the spatial context of a location. Mesquite shrublands on sandy soils with little topographic relief are more susceptible to reversal than creosotebush shrubland on sandy clay soils with well-established petrocalcic layers that affect water available to grasses. Distance to seed sources can also generate spatial heterogeneity, time lags, and nonlinear dynamics in the recovery of grasses. Researchers found that changing pattern-process relationships across scales can generate emergent behavior at broader scales that can explain historic patterns of change at the Jornada, and for a wide range of ecosystem types at other sites. This emergent behavior resulting from cross-scale interactions cannot be predicted from fine-scale patterns and processes.

Connections among plant, patch and landscape scales at the Jornada were expanded conceptually to regional and continental scales, and provides important linkages between LTER site-based science and continental-scale research being investigated by the National Ecological Observatory Network. Technologies based on these concepts are being applied to management of many of the hundreds of thousands of hectares of rangelands in the western US as well as in other arid and semiarid regions globally.

Graph for
Large variability in aboveground net primary production (ANPP) through time for grassland (upland) and shrubland (mesquite) ecosystem types can be explained by positive feedbacks at plant to patch scales that emerge following a sequence of wet years. Higher ANPP than expected based on rainfall in 2006-2008 (red points) results from herbaceous plants recruited in the wet summers of 2004-05 that captured water from patch-scale overland flow in each successive wet year to increase rain use efficiency. These interactions across scales are increasingly recognized as key to understanding variability in time and space in drylands at the Jornada and globally.
D.P.C. Peters, J. Yao, O.E. Sala, and J.P. Anderson Global Change Biology (submitted).