The effects of simulated rainfall events during the summer fallow period on
extractable and microbial carbon (C), nitrogen (N), and phosphorus (P) in
soils under either continuous wheat or the second-year pasture phase of a
2-pasture-1-wheat rotation in the Western Australian wheatbelt were
investigated. A ‘single wetting’ treatment (45 mm rainfall on Day
0) was compared with a 55 mm ‘multiple wetting’ treatment (45 mm
at Day 0, 5 mm at Day 3, and 5 mm at Day 8). Soil samples from 0{10 cm
depth were taken prior to, and at regular intervals up to 14 days following,
the inital wetting event. Soil extracts were assayed for total soluble N
(TSN), total oxidisable C (TOC), Olsen-P, and ninhydrin-positive compounds
(NPC). Prior to the simulated rainfall events, extractable TSN and TOC in the
air-dry fallow soils were significantly higher (P <
0·01), and Olsen-P significantly lower, for the pasture land use
compared with the continuous wheat. However, subsequent to wetting there were
no significant differences between the 2 land uses, or single and multiple
wetting treatments, for extractable TSN, TOC, Olsen P, or NPC. Extractable
soluble organic N (SON), calculated by subtracting measured inorganic N from
TSN, decreased within 48 h of each wetting event to a minimal value but, after
the first 2 wetting events, subsequently increased to at least the prewet
value. Microbial C, N, and P were estimated from the difference in TOC, TSN,
and Olsen-P of extracts from fumigated and unfumigated soils (microbial
‘flush’) and microbial C and N were also estimated from the NPC
‘flush’. There was generally good agreement between the 2
estimates of microbial N (NPC and TSN,
R2 = 0·50), but less
so for the 2 estimates of microbial C (NPC and TOC,
R2 = 0 ·31). There
was no significant difference in microbial C, N, or P between the 2 land uses,
but there was a highly significant response of the microbial biomass to
wetting events and also significant differences in temporal patterns between
the single and multiple wetting treatments. Microbial C and N increased in the
period following initial wetting, more rapidly in the wheat than the pasture,
reaching a peak at Day 2 for wheat and Day 3 for pasture. Subsequently, for
the single wet treatment, there was a steady linear decline in microbial C and
N until Day 10, whereas over the same period (Days 4-10) in the multiple wet
treatment there were 2 highly significant quadratic responses to time,
manifest as a linear increase in microbial C and N following each re-wetting
event, to a peak value 24 h after the event, and a subsequent decline to the
pre-wet value after a further 24 h.
High-frequency monitoring reveals nutrient sources and transport processes in an agriculture-dominated lowland water system