Ecosystem hydrologic and metabolic flashiness are shaped by plant community traits and precipitation

Abstract

Understanding the hydrologic and carbon cycling consequences of precipitation variability in dryland ecosystems requires improved appreciation and accounting of how above- and belowground biophysical processes differ in their response to rainfall. Our objective was to contrast the sensitivity of dryland ecosystem evapotranspiration (ET), gross ecosystem productivity (GEP), and ecosystem respiration (Re) in response to inter- and intra-annual precipitation variability in a nearby grassland, savanna, and shrubland ecosystems in southeastern Arizona. To do this, we modified the Richards-Baker index, which quantifies the flashiness of a stream’s hydrograph, to calculate analogous indices of ecosystem hydrologic and metabolic flashiness. In this way, ecosystem flashiness describes the frequency and rapidity of short-term fluctuations in H2O and CO2 exchange in response to precipitation while preserving the sequence of day-to-day variation in fluxes using tower-based time-series of daily averaged ET, GEP and Re. We calculated annual hydrologic, GEP, and Re flashiness (fET, fGEP and fRe respectively) using 6 years of daily-averaged fluxes estimated from eddy covariance. In contrast to our prediction, annual fGEP was consistently greater than annual fRe. Furthermore, we predicted that increasing rooting depth would correlate with a decline in annual fET and fGEP. In fact, annual fGEP was similar between the grassland, savanna, and shrubland. Whereas the response of annual fET and fGEP to annual precipitation was plant community dependent and generally declined with increasing rainfall, annual fRe did not vary in response to precipitation. The effect of late summer storms on fGEP was plant community dependent such that shrubland fGEP and fRe strongly declined in response to rainfall whereas grassland and savanna fGEP was relatively unresponsive. Conceptually similar to hydrologic flashiness, ecosystem flashiness may provide an additional lens through which to observe the influence of resource availability, shifts in community composition, and disturbance on ecosystem hydrologic and carbon cycling.