Refining the 'pulse-reserve' model for arid central Australia: Seasonal rainfall, soil moisture and plant productivity in sand ridge and stony plain habitats of the Simpson Desert

In dry ecosystems, water supply is highly variable in time and space, occurring either as short pulses of shallow soil water or long phases of deep soil water. There may be correspondence between hierarchical soil water pulses and hierarchical biological responses, but this has not been directly tested for arid Australia. We compared plant community responses to an isolated short rain pulse (about 130mm, November 2008) versus the start, mid and later stages of an extended high rainfall phase (about 540mm, February 2010 to March 2011) at a site on the western edge of the Simpson Desert. We sampled across contrasting habitats: tall longitudinal sand ridges and adjacent stony clay plains (gibber). Thus, we compared responses to spatially variable soil moisture availability (sand vs. clay soil; run-off/on relief) and temporally variable rainfall supply. Our results supported the hierarchical resource pulse-biological response concept for dry ecosystems, with the short and long rainfall phases modulating distinct responses among functional types. Shallow-rooted annual grasses showed an equivalent positive response to the isolated pulse as to the first pulse of the high rain phase; whereas short-lived forbs, longer-lived (2-3 years) grasses and persistent forbs had peak abundance during the high rainfall phase. Perennial grass and shrub productivity did not change over time, though there was a distinct trend towards heightened sexual reproduction at the mid-late stages of the high rain phase. We also found that species seasonal growth tendencies were pronounced, and that spatial aspects of moisture availability resulted in different composition patterns at the habitat scale with the gibber and sand ridge sites showing strong and weak correspondence with temporal rainfall patterning respectively. Thus, we confirm that plant productivity models for this region should include the effects of rainfall connectivity and soil texture as well as species seasonal growth constraints and rooting attributes.