Increased atmospheric CO2 concentrations caused by fossil fuel consumption are affecting global climate, increasing temperatures and precipitation variability in many regions. Such changes in climate can exceed the physiological tolerances of existing long-lived plant species or, alternatively, make conditions more favorable, increasing species productivity and abundance. One result is likely to be large scale vegetation change as some species die in areas where they can no longer survive while others increase in abundance and expand their distribution. The drought-induced mortality of piñon pine (Pinus edulis) over last six years in the southwestern US may be one example of this effect. Across this broad region, piñon mortality has ranged from 40-95% while co-occurring juniper (Juniperus spp.) has exhibited much more limited mortality (2-25%). This differential mortality has decreased woody plant cover, altered species distributions and is likely to affect key ecosystem functions like carbon uptake and sequestration, and water balance.
This project is a collaboration between the University of New Mexico (WT Pockman PI) and Los Alamos National Laboratory (NG McDowell PI) to test two hypotheses 1) plant mortality occurs when conditions exceed plant physiological limits or when conditions near these limits impair plant energy gain via photosynthesis, resulting in carbon starvation and a subsequent inability to defend against pathogens and 2) piñon pine, the species more susceptible to drought, outperforms juniper during periods of increased water availability. These hypotheses will be evaluated using ecosystem scale precipitation manipulation to study the mechanistic basis of responses of pinon-juniper woodland at the Sevilleta National Wildlife Refuge in central New Mexico . Using a rainfall exclusion treatment (excluding 50% of ambient precipitation), combined with a mechanistic model (Sperry et al 1998, 2002), we will evaluate the role of limitations on water uptake and transport in the response of piñon and juniper to severe drought. Water addition treatments (150% of mean annual precipitation) in adjacent plots will allow us to measure the effects of increased rainfall. The study will allow us to distinguish among a number of proposed explanations for the basis of the observed differential mortality during drought and the responses during periods of above-average rainfall that may actually pre-dispose the system to larger responses during drought.
The research site at the Sevilleta National Wildlife Refuge, an NSF-funded Long Term Ecological Research (LTER) site, consists of 12 large plots (40 m x 40 m) in the Los Pinos mountains. Three plots are assigned to each of the four treatments (untreated control, rainfall exclusion, rainout control, and water addition). Rainfall exclusion plots are equipped with acrylic channels covering 50% of the soil surface to exclude a similar amount of rainfall. Rainout control plots receive channels that do not divert water, allowing us to distinguish the effect of the channels on light and temperature from the effect of altered water availability in the rainfall exclusion plots. Water addition plots precipitation is supplemented using an overhead rainfall simulator. All plots are equipped with automated measurements of soil moisture and plant water uptake. To assess the changes that occur in each species during prolonged periods of high or low water availability we will use a water transport model to synthesize our field measurements and assess the behavior of each species with respect to its physiological limits. We will also use our modeling framework to estimate impacts of drought on carbon assimilation and water-use efficiency. Model synthesis will facilitate hypothesis testing and provides a tool for future predictions of piñon and juniper response to drought. The results of this study will provide important new information regarding the likely responses of piñon-juniper ecosystems to incipient changes in climate throughout the southwest deserts.