RainMan: A Global Change Experiment Integrating Proximal Remote Sensing with Direct Measures of Semiarid Grassland Structure and Function
Abstract
Many drylands are experiencing temporal repackaging of precipitation: larger rainfalls separated by longer drought. Relatively little is known about coupled mechanistic responses of plants, soils, and biogeochemical cycling. Furthermore, dryland remote sensing is challenged by decoupling of greenness from physiology. Here we present a two-year rainfall manipulation experiment in 60 plots of a semiarid grassland in Arizona, US. Constant growing season precipitation was repackaged into four treatments with dry intervals of 3.5 - 21 days and inversely varying magnitude of 5 - 60 mm. We ask how fewer, larger rainfalls impact ecosystem structure and function above- and below-ground. We evaluate visible, hyperspectral and thermal measurements to extend the observability of our experiment and test assumptions about dryland remote sensing. Aboveground measurements include ET and CO2 exchanges, cover, composition, structural and functional traits, and ANPP. Belowground measures include soil biogeochemistry, depth profiles of automated water content and matric potential, and images quantifying root allocation with depth. Plot-level remote sensing includes RGB and thermal techniques.
Fewer, larger rainfalls enhanced moisture ≥25 cm belowground, resulting in ~400% greater aboveground biomass of deeply rooted perennial grasses. Water stress in the top 10 cm killed annual forbs and grasses, reducing cover and increasing surface temperature ~2.0 °C. Depth of maximal root production increased from 20 to 45 cm. Peak GPP and greenness were of similar magnitude but shifted up to 30 days later. Soil CO2 emissions were reduced, consistent with an infrequently wet surface layer, and ongoing analyses are assessing multi-year carbon budget. Automated RGB cameras tracked GPP and ET well early in the growing season but showed less skill during late season. Ongoing analyses are developing unique greenness time series for each dominant plant functional type, which promises to help disentangle the complex relationship between phenology and function in dryland remote sensing. Thermal imaging quantified the prolonged access to water by deep-rooted perennials as compared to annuals.- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2022
- Bibcode:
- 2022AGUFM.B13C..02B