The effects of altered precipitation patterns on ecosystem responses in a desert rangeland

TitleThe effects of altered precipitation patterns on ecosystem responses in a desert rangeland
Publication TypeConference Paper
Year of Publication2004
AuthorsSnyder K.A., Cable J.M., Huxman T.E., Maxwell CJ
Conference Name4th International Conference on Applications of Stable Isotope Techniques to Ecological Studies
Date PublishedApril 19, 2004
ARIS Log Number174875
Keywordsaltered, carbon exchange, desert rangeland, ecosystem responses, precipitation patterns, precipitation pulses, water use
AbstractMany desert regions are characterized by seasonally dominated rainfall regimes. Global climate change models predict regional changes in average seasonal precipitation, but few models address how changes in the delivery of precipitation within a season will affect ecosystem responses. In the southwestern US, summer rainfall is highly unpredictable and punctuated by intervening dry periods of variable lengths. These patterns of precipitation pulses and subsequent dry periods may differentially affect water use and net ecosystem carbon exchange, depending on the ability of various ecosystem components (e.g., plants, soil microbes, and biological soil crusts) to respond to increases in water availability and maintain function during soil dry-down. The objective of this study was to examine how a 60-mm increase in summer precipitation (46% increase relative to mean summer rainfall) applied in two contrasting treatments affected carbon exchange and water use. Eighteen 4x6-m plots containing both a mesquite shrub (Prosopis glandulosa) and scattered black grama grass (Bouteloua eriopoda) were assigned to one of three watering treatments (n=6): ambient precipitation (controls), ambient plus frequent, small (5-6 mm) rainfall events applied weekly, and ambient plus infrequent, large (20-24 mm) events applied monthly. Treatments were applied for 12 weeks during summer using deuterium-enriched water. Stable isotope ratios of hydrogen (dD) in plant xylem water and carbon (d13C) in respired CO2 were used to partition water source use and respiratory CO2 flux. The dD of mesquite xylem sap indicated that mesquite was able to use water from both small and large precipitation events, while control plants accessed water from deep within the soil profile. Changes in soil moisture, mesquite predawn water potentials, and mesquite stem elongation were highest in monthly watered plots, intermediate in weekly plots, and lowest in control plots. Although large events appeared to favor carbon gain of C3 shrubs, biological soil crust communities may have taken advantage of smaller more frequent events. Daytime measurements of net ecosystem carbon exchange on subplots containing soil, plant roots, and biological soil crusts indicated control and monthly plots were losing CO2 to the atmosphere, while weekly plots were acquiring CO2. After application of 2 mm of water to all plots, net uptake of CO2 was observed in all plots, but the magnitude of crust photosynthetic response was greatest in weekly watered plots. Keeling plot analyses of the d13C of respired CO2 during dark hours indicated that the contribution from different sources (roots, microbes, and soil crusts) varied with watering treatment. Small frequent rainfall appeared to promote the proportional contribution of microbial respiration or C4 roots to total respiratory flux, while infrequent large rainfall resulted in a greater proportion of CO2 originating from biological soil crusts or C3 root activity. Control plots that had received little ambient rainfall had more negative d13C values, indicating another potentially deeper C3 dominated source of respired CO2.