Precipitation pulses and carbon fluxes in semiarid and arid ecosystems

Abstract

In the arid and semiarid regions of North America, discrete precipitation pulses are important triggers for biological activity. The timing and magnitude of these pulses may differentially affect the activity of plants and microbes, combining to influence the C balance of desert ecosystems. Here, we evaluate how a “pulse” of water influences physiological activity in plants, soils and ecosystems, and how characteristics, such as precipitation pulse size and frequency are important controllers of biological and physical processes in arid land ecosystems. We show that pulse size regulates C balance by determining the temporal duration of activity for different components of the biota. Microbial respiration responds to very small events, but the relationship between pulse size and duration of activity likely saturates at moderate event sizes. Photosynthetic activity of vascular plants generally increases following relatively larger pulses or a series of small pulses. In this case, the duration of physiological activity is an increasing function of pulse size up to events that are infrequent in these hydroclimatological regions. This differential responsiveness of photosynthesis and respiration results in arid ecosystems acting as immediate C sources to the atmosphere following rainfall, with subsequent periods of C accumulation should pulse size be sufficient to initiate vascular plant activity. Using the average pulse size distributions in the North American deserts, a simple modeling exercise shows that net ecosystem exchange of CO₂ is sensitive to changes in the event size distribution representative of wet and dry years. An important regulator of the pulse response is initial soil and canopy conditions and the physical structuring of bare soil and beneath canopy patches on the landscape. Initial condition influences responses to pulses of varying magnitude, while bare soil/beneath canopy patches interact to introduce nonlinearity in the relationship between pulse size and soil water response. Building on this conceptual framework and developing a greater understanding of the complexities of these eco-hydrologic systems may enhance our ability to describe the ecology of desert ecosystems and their sensitivity to global change.

Appendix

We compiled flux duration curves [analogous to the stream flow duration curves (Searcy 1959)], to illustrate the differences in ecosystem CO₂ exchange characteristics for a pulsed ecosystem [a desert grassland (Jornada Experimental Range; Mielnick et al., in press)] and an ecosystem that experiences a relatively steady-state decline in soil water availability in time [a coniferous forest (Niwot Ridge AMERIFLUX site; Monson et al. 2002)]. We used 30- and 20-min averaged (Niwot Ridge and Jornada, respectively) peak growing season (June–August) NEE values observed over 4 years (1999–2002 and 1997–2000, Niwot Ridge and Jornada, respectively). Briefly, NEE data for the period of interest at each site were assigned a rank (r) in order of descending magnitude, positive to negative. A probability of exceedance (F) was calculated for each ranked NEE value (r) according to the formula:

$$F=[r/(n+1)]\times 100$$

where n is the number of ranked NEE values for the period of interest. Like flow duration analysis in hydrology (Searcy 1959; Vogel and Fennessy 1995; Potts and Williams 2004), flux duration analysis provides a convenient and repeatable standard for comparing patterns of ecosystem exchange between sites and between years at the same site. By ranking and assigning a frequency to ecosystem exchange values, flux duration analysis incorporates episodic high activity periods, such as those associated with precipitation pulses, and sustained low level fluxes during interpulse periods into a single calculation. As additional ecosystem scale flux data sets become available, it may be possible to broadly classify ecosystem flux duration curves as “pulsed-dominated” and “steady-state” similarly to the way hydrograph-derived flow duration curves can be described and classified by the physical, biotic and anthropogenic factors controlling stream flow (e.g., Vogel and Fennessey 1995; Smakhtin 2001).