Agriculture Reference
In-Depth Information
of mineralized N likely kept soil NH
4
+
and NO
3
−
concentrations low, even
while the supply rate may have been high, given the activity of these en-
zymes. High rates of both gross mineralization and microbial immobiliza-
tion have been observed in an organic tomato system on similar soil in
the same landscape (Burger and Jackson, 2003). An alternative hypothesis
is that increased C availability increased denitrifi cation and subsequently
lowered soil NO
3
−
; however, N
2
O emissions over two growing seasons
were negligible in a separate case study of an organic Roma tomato fi eld
managed by one of the growers involved in this research (Smukler et al.,
2010) as well as in other organic tomato systems in this area (Burger et al.,
2005). Thus, denitrifi cation was probably low.
The lack of association between P availability, as indicated by Olsen
P, and the potential activity of P-cycling enzymes, phosphodiesterase and
alkaline phosphomonoesterase, is in contrast with previous work that
demonstrates a negative relationship between phosphatase activity and P
availability in non-agricultural systems (Olander and Vitousek, 2000 and
Allison et al., 2007). Across the 13 fi elds, these enzyme activities appear
related to microbial biomass; for instance, phosphodiesterase activity had
the strongest positive relationship with MBC of any enzyme (p < 0.001 R
2
= 0.727), suggesting that soil microbial biomass was more important than
P availability in regulating investments in phosphatases across these fi elds.
The relative importance of microbial community composition vs. envi-
ronmental factors in regulating enzyme expression remains unclear (Sin-
sabaugh et al., 2005, Allison et al., 2007, Frossard et al., 2012 and Reed
and Martiny, 2013). In this study, microbial community composition ex-
plained little unique variation in potential enzyme activities relative to soil
physicochemical properties. The plasticity of the resident microbial com-
munity to respond to environmental conditions may be high, as suggested
by the relatively large fraction of variation in potential enzyme activities
explained by soil physicochemical characteristics (37.7%). Moreover, a
large fraction of variation was also explained jointly by soil factors and
FAMEs (~27% of the total variation in the canonical variance partition-
ing analysis), indicating that microbial communities did infl uence activity
under specifi c environmental conditions, despite the low variation in com-
munity composition across this landscape (see below).
