Corrections to denitrification measurements in Ochs and Milburn (2003), with a revised view of the importance of denitrification to N-loss from agricultural soils of the Mississippi Delta.

In our evaluation of the effects of wintertime flooding on nitrogen and phosphorus cycling in soils from the Mississippi Delta (Ochs and Milburn 2003), we report data for denitrification rates in flooded and non-flooded agricultural soil cores. The results for denitrification shown in FIGURE 4a and 4b were reported in units of mg [N.sub.2]O-N [ha-.sup.-1] [day.sup.-1]. This was an error. The correct units for denitrification in this FIGURE should be grams [N.sub.2]O-N [ha.sup.-1] [day.sup.-1] (see FIGURE 4a and 4b revised).

With this correction, we re-evaluate the importance of denitrification in this study. In Table 2 of the original study (Ochs and Milburn 2003) we provided estimates for the percent losses of nitrogen, relative to ambient total N pools, in cotton and soybean fields over the course of a 90-day winter period. In Table 2-revised, we correct these data. Applying the corrected measurements, at the maximum values of denitrification measured prior to soil drainage (12-21 g N [ha.sup.-1] [day.sup.-1], depending on crop type and flooding regime), percentage losses of soil N, over a 90-day period, would be between 0.09-0.16%.

Our corrected values for denitrification help, but do not fully, explain the observed reductions in N[O.sub.3]-N in both flooded and non-flooded soybean soils, as shown in Fig. 2b of Ochs and Milburn (2003). Losses of N[O.sub.3]-N from flooded soils over the course of our 57-day experiment were 6.22 mg kg [soil.sup.-1] from cotton field soils, and 3.9 mg kg [soil.sup.-1] in soybean soils. Applying our corrected average values of denitrification (6 g N [ha.sup.-1] [day.sup.-1] in cotton field soils, 7-12 g N [ha.sup.-1] [day.sup.-1] in soybean soils depending on flooding regime), losses of 0.34 kg N [ha.sup.-1] and 0.40-0.68 kg N [ha.sup.-1] can be attributed to denitrification in cotton and soybean field soils, respectively, during this period. For a soil depth of 10 cm (average soil dry weight = 1.3 X [10.sup.6] kg [ha.sup.-1]), these denitrification rates account for 4% of the observed decline of N[O.sub.3]-N in cotton soils and 8-13% of the observed decline in the soybean field soils. Clearly, although denitrification can't explain the magnitude of the decline in NO3-N in these soils, it is not an insignificant process.

Perhaps most importantly, is the role that denitrification may play in removing N[O.sub.3]-N that might otherwise be lost from the fields in runoff. This point was not fully considered in Ochs and Milburn (2003). At the average daily rates of denitrification we observed (see above), when extended over a 90-day winter period, losses of N by denitrification from these soils would exceed 0.5 kg N [ha.sup.-1] for cotton field soils, and between 0.6-1.1 kg N [ha.sup.-1] for soybean field soils. Extrapolated over thousands of hectares, these results indicate soil N removal by denitrification on the order of hundreds or thousands of kilograms, even during winter. Actual rates of loss would be tightly correlated with temperature, and labile C content (Ochs and Milburn 2003).

Denitrification in removal of N from agricultural soils in the Mississippi Delta can be appreciable (c.f. Zwart et al.), but it remains unclear how important it is as a mechanism of N-loss compared to erosion of soil particles, or runoff of soluble N. McDowell et al. (1989), for instance, report losses of N in solution and in soil particles from cotton soils near Clarksdale, MS exceeding 42 kg [ha.sup.-1] [year.sup.-1]. The primary purpose of our research (Ochs and Milburn 2003) was to investigate extended seasonal flooding as a mechanism to reduce N losses both by reducing soil erosion and by promoting denitrification in flooded, temporarily anoxic soils. Although we did not discern differences in denitrification between continually flooded and non-flooded (but occasionally wetted) soils, it is evident from our results, and from Green (1998), that controlled seasonal flooding can greatly reduce erosional losses of soil particles and associated nutrients. We encourage this practice to be studied further.

References

McDowell, L.L., G.H. Willis and C.E. Murphree. 1989. Nitrogen and phosphorus yields in run-off from the silty soils in the Mississippi Delta, U.S.A. Agriculture, ecosystems and environment 25: 119-137.

Ochs, C.A. and S.A. Milburn. 2003. Effects of simulated wintertime flooding to control erosion on selected chemical and microbial properties of agricultural soils in the Mississippi Delta. J. Mississippi Academy of Sciences 48: 102-114.

Zwart, K.B., G.L. Velthof, J-W. van Groenigen, M. Heinen, B. van der Graft, and L. Boumans. 2004. Denitrification in agricultural soils in the Netherlands. Workshop on advanced approaches to quantify denitrification. Woods Hole, MA. pp. 1-13.

[FIGURE 4 OMITTED]

Clifford A. Ochs

Department of Biology

University of Mississippi 

Table 2 -- revised. Percent losses of nitrogen and carbon by
denitrification and respiration relative to initial concentrations of
total nitrogen and organic carbon in cotton (CT) and soybean (SOY) soils

        TN (a,b)        C (a,b)         Denitrification (c)
Soil    kg [ha.sup.-1]  kg [ha.sup.-1]  kg [ha.sup.-1] 90 [d.sup.-1]

CT-NF   1220            15300           1.17
CT-F    1220            14800           1.47
SOY-NF  1220            15400           1.93
SOY-F   1250            14900           1.56

        Respiration (c)
Soil    kg [ha.sup.-1] 90 [d.sup.-1]  %Denitrif (d)  %Resp (d)

CT-NF   17.4                          0.09           0.11
CT-F    27.0                          0.12           0.18
SOY-NF  20.4                          0.15           0.13
SOY-F   43.8                          0.12           0.29

(a) Pre-flooding means of total N and organic C concentrations in soils.
(b) Soil weight (dry) is 1.3 X [10.sup.6] kg [ha.sup.-1] for a soil
depth of 10 cm.
(c) Maximum mean values for the flooded period (all are from day 57).
(d) Percent values are relative to pre-flooding concentrations in soils.