Soil respiration response to precipitation manipulation in piñon-juniper woodland

Background/Question/Methods

Soil respiration contributes > 75% of the annual ecosystem respiration in water-limited terrestrial ecosystems and is roughly equivalent to plant belowground carbon flux at an annual scale. As the frequency and intensity of drought increases in the southwestern US, changes in plant available water will likely change ecosystem carbon cycling and storage through changes in plant activity, tree mortality, and decomposition. We examined the link between precipitation and soil respiration in a large-scale precipitation manipulation experiment in piñon-juniper woodland at the Seviletta LTER. The experiment had four treatments repeated in three blocks: 1) ambient precipitation, 2) removal of 47% of the annual precipitation using troughs, 3) ambient precipitation with troughs installed upside down so that they removed no water, and 4) ambient plus supplemental irrigation from overhead sprinklers. Soil respiration was measured monthly with 30.5 cm diameter collars at nine locations per plot with three collars in each of the three vegetation cover types (piñon canopy, juniper canopy, and no tree canopy). We derived annual estimates for each plot and cover type by 1) normalizing measured values to flux rates at 15C based on autochamber data temperature response, 2) adjusting the normalized plot or type mean to daily mean soil temperature, and 3) interpolation between measurements using daily mean daily mean temperature within plot and cover type.

Results/Conclusions

The artificial drought and low ambient precipitation strongly reduced soil respiration, and respiration was higher under the tree crowns than for the between-tree openings. Annual soil respiration varied among years and treatments from a low of about 80 g C m^-2 yr^-1 for the 47% precipitation-removal treatment in a very dry year (140 mm annual precipitation) to a high of 630 g C m^-2 yr^-1 for all treatments in a wetter year (330 mm annual precipitation). Within a year, annual soil respiration varied with the amount of precipitation received by each treatment, with slopes of 0.23-0.77 g C m^-2 yr^-1 mm^-1 (R² of 0.23-0.69). Within treatments across years, the slope of the regression of soil respiration with precipitation was greater and more consistent among treatments (0.98-1.32 g C m^-2 yr^-1 mm^-1) and with less variability (R² of 0.66-0.97). Precipitation appears to be a good index of soil respiration in this ecosystem, but the data indicate that soil water balance, precipitation timing, and plant activity may also play important roles.