Decomposition rate of cereal straw as affected by soil placement.
Post-harvest crop residues are an important resource for arable farmers because they add organic matter to the soil and recycle plant nutrients. Returning crop residues to the soil has long been advocated as a practice that helps sustain soil quality and fertility over the long-term (Campbell and Zentner 1993). Post-harvest residues may affect several factors that influence yield of the following crop, including plant establishment, nutrient availability, soil temperature, water infiltration and evaporation, and disease and pest populations. The challenge is to manage residues in a way that maximises the benefits while minimising problems associated with their retention. It is especially important that undecomposed residues do not hinder mechanical operations such as seed drilling. A thorough knowledge of the factors affecting decomposition is a pre-requisite to develop best management practices for crop residues.
Decomposition is a complex process that is influenced by the chemical and physical properties of the residue (e.g. lignin and N contents, particle size) and by soil factors that control microbial activity (Parr and Papendick 1978). Due to more favourable conditions for microbial activity, soil-incorporated residues normally decompose at a faster rate than residues on the soil surface (Parr and Papendick 1978; Christensen 1986; Douglas and Rickman 1992; Curtin et al. 1998; Wang et al. 2001). Interactions between residues and the soil create the environmental conditions under which decomposing microorganisms function. Most importantly, these interactions control the water potential of the straw, the rate of microbial colonisation, and N supply to the decomposing microorganisms (Parr and Papendick 1978; Wang et al. 2001). The disposition of residues within the soil may affect the rate of decomposition. Where residues are incorporated by ploughing, they become concentrated beneath the inverted plough layer and limited soil-residue contact may slow microbial colonisation and water absorption by the residues and restrict N availability to sites of decomposition. When residues are mixed through the soil, as occurs when the soil is cultivated using a disced implement, these constraints are alleviated because of intimate and extensive soil-residue contact.
Little work has been done to quantify the effects of within-soil placement on residue decomposition (Kanal 1995). From a laboratory study, Parr and Reuszer (1959) reported that early decomposition was slower where straw was localised rather than mixed through the soil, but total decomposition over the 6-week incubation differed little between the 2 placement methods. However, caution is required when extrapolating laboratory decomposition data to the field (Parr and Papendick 1978).
Residue decomposition may be determined either by measuring mass loss (e.g. Douglas and Rickman 1992; Beare et al. 2002) or by measuring C[O.sub.2]-C evolution from decaying residue (e.g. Bremer et al. 1991; Jensen et al. 1997; Berg and McClaugherty 2003). Because of operational simplicity, mass loss using a litter bag technique has been the preferred method in many decomposition studies (Heal et al. 1997). Crop residue, contained in fibreglass mesh bags, is placed in the soil at a depth of 0.15 or 0.20 m and undecomposed residue measured by recovering the litter bag containing the remaining material (Beare et al. 2002). The litter bag technique simulates the placement achieved when residue is incorporated by ploughing. However, the technique has been criticised on the grounds that the mesh could hinder entry of soil animals (earthworms and arthropods), which may be important decomposers in some systems. Possibly a more serious criticism of the litter bag method is that it can underestimate decomposition because of poor soil-residue contact (Cogle et al. 1987).
Although soil-mixed residues may give a more meaningful estimate of decomposition than the litter bag technique, recovery of unconfined residues is problematic because decomposition and fragmentation result in residue particles too small to be physically removed. The unrecovered residue could be quantified using [sup.14]C labelling techniques, but a high level of technology is needed to produce and analyse the labelled material (Cogle et al. 1987). Measurement of C[O.sub.2]-C output from decomposing residue provides an alternative method of evaluating the effects of within-soil residue placement. Although monitoring of C[O.sub.2] emissions during the decomposition period is time-consuming, it does provide information on the dynamics of the decomposition process at time scales that are not possible using the litter bag technique, which, as normally employed, measures mass loss over relatively long time steps (several weeks).
Data obtained using the litter bag technique have been used to develop simple models to predict the decomposition of incorporated cereal straw (Douglas and Rickman 1992). These models have been tested against litter bag data from a range of studies (Douglas and Rickman 1992), but we do not know if they realistically predict decomposition when residues are distributed through the soil. In this paper we report on a field study comparing decomposition of cereal straw measured using the litter bag technique (simulating the residue placement achieved by ploughing) with decomposition of unconfined residues mixed through the soil (simulating disc incorporation).
Materials and methods
The experiment was on a Wakanui silt loam (Immature Pallic Soil; Hewitt (1993)) at the Crop & Food Research farm near Lincoln, Canterbury, New Zealand (43[degrees]38'S, 172[degrees]30'E). The sand, silt, and clay contents were 180, 620, and 200 g/kg, respectively. At the start of the experiment, the soil (0-0.20 m) had total C and N contents of 23.2 and 1.7 g/kg, respectively, mineral N content of 47 mg/kg, microbial biomass C, determined by the fumigation-extraction method (Vance et al. 1987; Horwath and Paul 1994) of 207 mg/kg; and pH ~6. Bulk density values for the 0-0.075, 0.075-0.15, and 0.15-0.20m layers were 1.15, 1.16, and 1.31 Mg/[m.sup.3], respectively. The site had been fallow in the growing season before our experiment. This was preceded by a cereal crop (1999-2000) and another fallow season (1998-99). Soil moisture content at initiation of the experiment was 14.0% (gravimetric basis).
Treatments and trial establishment
Wheat (Triticum aestivum L.) and barley (Hordeum vulgare L.) straw used in the experiment was collected as standing stubble from plots on the Crop & Food Research farm in the month preceding the experiment. The initial composition of the straws is given in Table 1. The straws were air-dried and cut into 50-mm lengths before use in the experiment.
The trial was laid out as an incomplete Latin square with 5 replicates of 7 treatments. The treatments were: (1) control (no added straw); (2) barley straw (in 4-mm mesh bags) placed horizontally in the soil at a depth of 0.20 m (simulated plough treatment); (3) wheat straw (in 4-mm mesh bags) placed horizontally in the soil at a depth of 0.20 m; (4) barley straw mixed through the 0-0.20 m soil layer (simulated disc treatment); (5) wheat straw mixed through the top 0.20 m of soil; (6) barley straw placed on soil surface; and (7) wheat straw placed on soil surface. The straw application rate was equivalent to 7 t/ha, which is about the average rate of straw production in the Canterbury region.
The experiment was established on 6 April 2001 (i.e. early autumn). Soil was dug from microplots (0.30 by 0.30 by 0.20 m deep) by hand. The sides of the holes (but not the bases) were lined with polythene to prevent lateral diffusion of C[O.sub.2] from the microplots. In the simulated disc treatments, straw was thoroughly mixed through the excavated soil before repacking it into the hole. In the litter bag treatments, mesh bags (0.30 by 0.30 m; 4-mm mesh) containing straw were placed at the bottom (0.20 m depth) of the holes, and the excavated soil was well mixed before refilling the holes. The control plots and plots to be used for the surface straw treatments were also excavated and the soil mixed before returning it to the holes. The microplots were packed to approximately the original bulk density. To prevent removal by wind, surface-placed straw was covered with mesh (4 mm) that was anchored to the ground using metal pegs. Plots were kept plant-free during the experiment by hand removing seedlings as they emerged. Soil moisture content (0-0.20 m) was measured periodically during the experiment by time domain reflectrometry using 20 cm waveguides.
Measurement of C[O.sub.2] emissions
Emissions of C[O.sub.2] were monitored from the day following trial establishment and measurements were usually made every 2-3 days until emissions declined due to low soil temperatures in winter, when less frequent monitoring was deemed adequate. A plastic chamber (0.24 m diam.; height 0.25 m) with a sharpened base was pressed into the soil to a depth of 40 mm (the chamber covered 50% of the plot area). The C[O.sub.2] concentration in the chamber was measured using an infrared gas analyser (PP Systems Soil Respiration system, Stotfold, Hitchin, UK) for 120 s, and C[O.sub.2] flux calculated (in units of g C[O.sub.2]/[m.sup.2].h) from the rate of increase in C[O.sub.2]. Emissions of C[O.sub.2] were measured for all plots except those with surface straw, where the thick layer of straw made insertion of the chambers impractical. Measurements were usually made in the mid-morning to early afternoon period. To avoid compaction of soil in the vicinity of the plots during C[O.sub.2] measurements, wooden walkways were used to access the plots.
Carbon dioxide fluxes were monitored as described above between 7 April and 18 June 2001. Towards the end of that period, the increase in chamber C[O.sub.2] concentration during the 120-s measurement period was quite small (generally <10 [micro]L/L) and the system was approaching its detection limit. Subsequent monitoring was done using chambers containing alkali traps to absorb emitted C[O.sub.2]. These chambers, which enclosed an area of 225 c[m.sup.2], usually remained in place for 24 h, during which evolved C[O.sub.2] was absorbed in traps containing 20mL of 0.25 M NaOH. The absorbed C[O.sub.2] was determined by back titrating with 0.25 M HCl. Comparisons carried out on 29 May and 7 June showed that C[O.sub.2] fluxes measured by the 2 techniques were very similar. The final measurement using the alkali traps was on 30 August. Cumulative C[O.sub.2]-C emissions were estimated from initiation of the experiment to 18 June (period during which the infrared analyser was used) and from then until the end of the experiment (alkali trap period) by trapezoidal integration beneath the C[O.sub.2] flux v. time curve.
Recovery and analysis of undecomposed straw
The experiment was terminated after 158 days (10 September 2001). Straw remaining on the surface straw plots was carefully collected so as to minimise soil contamination. Soil was collected from each microplot in 50-mm increments (0-0.05, 0.05-0.10, 0.10-0.15, and 0.15-0.20 m depths). The litter bags were retrieved and placed in plastic bags for transport to the laboratory.
Straw was taken from the litter bags, shaken gently on a 1-mm sieve to remove adhering soil, and weighed. Samples were dried at 60[degrees]C to determine moisture content. Subsamples of dry, ground residue were heated in a muffle furnace at 550[degrees]C for 5 h to determine ash content. Further subsamples were taken to measure total C and N using a LECO C/N/S analyzer (LECO Corporation, St Louis, MI). Residue mass and residue C and N contents were corrected to an ash-free basis to account for variable amounts of contaminating soil. Surface straw was treated similarly to litter bag straw.
Soils from the soil-mixed straw treatment were oven-dried (30[degrees]C) for at least 16 h, after which large straw pieces were removed by hand. Further straw separation was achieved by passing the soil through a small, laboratory-based thresher fitted with a 5-mm sieve. Finally, smaller pieces of straw were recovered using a seed dresser fitted with 4- and 2-mm screens. Stones were removed from the separated residue by hand. Straw collected from the 4 sampled depths of each plot was combined. Residues were recovered from soil taken from the control plots using a combination of hand picking (larger pieces), and sieving using the thresher and seed dresser, as outlined above. The recovered residue was dried and analysed for ash, C, and N. Straw recovery from the soil-mixed straw treatment was estimated by subtracting the mean amount of ash-free residue measured in the 5 control plots.
Soil mineral N at end of experiment
Mineral N was extracted from subsamples of field-moist soil from each depth increment (0-0.05, 0.05-0.10, 0.10-0.15, 0.15-0.20m) by extraction with 2M KCl and measuring nitrate- and ammonium-N using standard colourimetric procedures (Keeney and Nelson 1982).
The data were examined with analysis of variance (ANOVA) using GENSTAT 8 (GenStat 2005). The C[O.sub.2] flux data for each measurement date were analysed separately. A probability level of 0.05 was used to determine significance.
Results and discussion
Weather and soil conditions
When straw was applied, the soil was moderately warm (mean daily temperature in the top 0.10 m ~14[degrees]C), but dry (moisture content of 14% w/w). Soil temperature remained relatively high (generally 12-14[degrees]C in top 0.10 m) during the following 40 days, after which there was a rapid decline with the onset of winter (Fig. 1). Soil temperature during the winter period of mid May to the end of August (days 41 to 147) was generally below about 6-7[degrees]C in the 0-0.10 m layer (mean 5.4[degrees]C). Total rainfall during the experiment was 181 mm, which is less than the long-term mean for the period (~300 mm). During the first 28 days, there was a total of only 0.6 mm of rain, and dry soil conditions persisted until 17 mm of rain fell between days 29 and 35.
Decomposition of low N (high C : N ratio) straws can be limited by N availability, particularly in the early phase of decomposition (Christensen 1986). The straws used in this study had C:N ratios that were not especially high (64 for barley and 89 for wheat; Table 1) and soil mineral N at initiation of the trial was relatively high (117 kg/ha in the 0-0.20 m layer). Thus, N availability is unlikely to have limited decomposition of incorporated straw, though it may be one of the factors limiting decomposition of surface straw (Wang et al. 2001).
[FIGURE 1 OMITTED]
Despite the relatively low soil moisture content, there was a large pulse of C[O.sub.2] production in the first 4 days where straw was mixed through the soil (Fig. 2). Litter bag straw resulted in smaller increases in C[O.sub.2] emissions during this initial period. Emissions of C[O.sub.2] from the soil-mixed straw plots decreased sharply following the initial pulse, though they remained above those measured in litter bag plots. There was a trend for C[O.sub.2] emissions from all treatments to decrease from about day 8 to ~day 24 (30 April). Measurements made on day 28 (4 May), before the 17 mm of rain between days 29 and 35, showed that the soil-mixed straw treatment continued to have a significantly (P < 0.05) higher output of C[O.sub.2] than the litter bag treatment (mean of 0.17 v. 0.12 g/[m.sup.2] x h). In response to rainfall on days 29-35, there was another pulse of C[O.sub.2] production with a peak at about day 33. During this period, values for soil-mixed straw were again substantially higher than for the litter bags. By day 40 (mid May), the soil was approaching field capacity (Table 2) and temperature was beginning to decline (Fig. 1). Low temperature was likely the major abiotic limitation to straw decomposition during the remainder of the experiment. From about day 55 (late May) through to the end of the experiment, C[O.sub.2] fluxes were generally higher for the straw-treated plots than from the control, but effects of straw placement were small or non-significant.
The high initial C[O.sub.2] output where straw was mixed through the soil was likely due to rapid decomposition of labile compounds in straw, as observed in previous work (Cogle et al. 1989). Even though the soil was relatively dry, soil-mixed straw apparently absorbed sufficient moisture through extensive soil-straw contacts to enable microorganisms to rapidly initiate decomposition of labile components such as the water-soluble material, which made up 7-9% of straw-C (Table 1). Fungi, which are more tolerant of low soil water potentials than are bacteria (Paul and Clark 1989), are likely to have played a dominant role during this early decomposition stage. Limited soil-straw contact in the case of the litter bags reduced the decomposition, probably by slowing absorption of water from the soil and delaying microbial colonisation of the straw.
In this field experiment, effects of within-soil straw placement persisted over a relatively long period (~2 months from time of incorporation). This contrasts with observations from laboratory studies where decomposition rate of localised straw was slower than that of soil-mixed straw during the initial 7 days only (Parr and Reuszer 1959). Total output of C[O.sub.2]-C from initiation of the experiment to 18 June (period during which the infrared gas analyser was used to measure C[O.sub.2] emissions), averaged 83 g/[m.sup.2] for soil-mixed straw plots, 61 g/[m.sup.2] for the litter bag plots, and 34g/[m.sup.2] in the no-straw control (Table 3). Amounts of C[O.sub.2]-C released during the remainder of the experiment were little affected by straw placement.
The incorporated straw accounted for a large part of the emitted C[O.sub.2]-C, particularly where straw was mixed through the soil (Table 3). Soil organic matter had been depleted due to a history of intensive cropping, and as a result, background soil respiration was relatively low. The C[O.sub.2]-C attributable to straw (estimated by subtracting C[O.sub.2]-C emitted from no-straw control from that released from the straw treated plots) was almost twice as high for the soil-mixed straw v. litter bag straw (averaged over straw type) until 18 June (first 73 days). Over the entire experiment, C[O.sub.2]-C generated from soil-mixed straw was, on average, 1.66 times that from litter bag straw (Table 3). These results demonstrate that within-soil placement can have a strong and persistent influence on straw decomposition. Results obtained using the litter bag technique may seriously underestimate decomposition rate of straw when it is incorporated using an implement that distributes it through the soil matrix.
The importance of straw placement/distribution is likely to be greatest when the soil and straw are dry at the time of incorporation. Under moist conditions, localised straw would wet up relatively rapidly and the decomposition process may be less affected by within-soil straw placement. However, in many grain-producing areas, soil can be depleted of available moisture by harvest, and under these conditions, mixing straw through the soil may be a practical way to speed up decomposition prior to establishing the next crop.
Straw type did not have a strong influence on C[O.sub.2] emissions. Estimated straw-derived C[O.sub.2]-C (averaged over placement methods) was 55 g/[m.sup.2] for barley straw v. 47 g/[m.sup.2] for wheat straw (difference significant, P = 0.03). There was no evidence of an interaction between straw type and incorporation method. There is evidence in the literature that barley straw decomposes slightly more rapidly than wheat straw (Summerell and Burgess 1989), although in a litter bag experiment, Christensen (1985) found no difference between wheat and barley straws. The slightly greater decomposition of barley straw may be a reflection of its higher N and water-soluble C contents (Table 1).
[FIGURE 2 OMITTED]
Straw recovered from the soil at the end of the experiment was substantially less when it had been soil-mixed than when it was enclosed in litter bags (Table 4). Mass loss of straw was 31-33% for the litter bags, whereas 61-71% of the soil-mixed straw was not recovered by careful hand picking and mechanical sieving. As expected, decomposition of surface straw was much less than that of incorporated straw; mass loss of surface straw was < 15% for both barley and wheat straw (Table 4).
Residue mass loss data for the litter bags are consistent with other local measurements. Beare et al. (2002) reported mass loss of barley straw from litter bags of about 35% in the period 5 May-15 September, similar to our values of 31-33% between 6 April and 10 September. Our mass loss values for litter bag straw also compare well with predictions of the Douglas and Rickman (1992) model. That model predicts straw mass loss based on air temperature (degree-days above a base of 0[degrees]C), placement (incorporated or surface-placed straw), and initial straw N content. An assumption implicit in the model is that incorporated residues decompose at the same rate regardless of whether they are localised or mixed through the soil. Predicted mass loss by the Douglas-Rickman model for incorporated straw was 42% for barley and 39% for wheat. Predicted loss for surface straw was 12-13%. These results suggest that that the Douglas-Rickman model is useful in predicting decomposition of surface-placed residues (no-tillage scenario) and residues that are localised within the soil. However, the model may systematically underpredict decomposition of residues that are well distributed through the soil.
It is likely that mass loss of soil-mixed straw was overestimated to some extent because of the difficulty of recovering all of the undecomposed residue. Tests of recovery efficiency showed that, after initial mixing of straw into microplots, essentially all of the straw was recovered using our extraction method (Francis et al. 1999). However, the efficiency of recovery may decrease as decomposition progresses, because straw becomes brittle and more fragmented, making it difficult to extract. From a practical perspective, straw fragments not recovered by our extraction technique would pose no difficulties for the farmer as they would not interfere with mechanical operations such as seedbed preparation and seed drilling.
As would be expected, straw C recovery values were very similar to those for straw mass (ash-free basis) recovery (Table 4). There was evidence of a straw type x placement interaction on straw mass and C recovery; recovery values for barley were significantly less than those of wheat straw in the soil-mixed treatment. Straw type had a strong influence on straw N recovery. On average, recovery of wheat straw N was 70%, compared with 57% for barley straw. The barley contained a significant amount of nitrate-N (640 mg N/kg v. 16 mg N/kg for wheat straw: Table 1). Leaching of N[O.sub.3]-N, which made up almost 10% of total N in barley straw, would largely explain the lower recovery of N for barley straw. The surface straw lost relatively more N than C. While leaching of N[O.sub.3]-N may partly account for the N loss from barley straw, leaching of soluble organic N must also have occurred to account for N losses from surface straw. Low N recovery from soil-mixed straw is likely to be partly due to failure to extract all of the undecomposed straw fragments.
Residues with high C:N ratios can cause temporary immobilisation of soil N after incorporation (Parr and Papendick 1978). Soil mineral N levels (0-0.20 m) at the end of the experiment were little affected by the straw treatments (Table 2). The final mineral N levels were much lower than at the beginning of the experiment (~20 v. 117 kg/ha), indicating that there was significant loss of N during the winter period, presumably by leaching and denitrification. Any straw-induced N immobilisation would not be detectable, as leaching/denitrification had an over-riding influence on mineral N changes during the experiment.
Evolved C[O.sub.2]-C v. C recovery
Straw recovery data are consistent with the C[O.sub.2]-C output values, in that both approaches suggest that decomposition was less in litter bags than when straw was mixed into the soil. However, the estimates of straw decomposition based on C[O.sub.2]-C output were substantially less than those obtained by mass loss. For example, mass loss of C from the litter bags averaged 96g/[m.sup.2], compared with 38g/[m.sup.2] of straw-C respired as C[O.sub.2]. Explanations might include either underestimation of C[O.sub.2]-C output or incomplete recovery of undecomposed straw. Fluxes of C[O.sub.2]-C were monitored frequently during the high emission period following straw addition, with less frequent measurements during the cool winter period (Fig. 2); it is unlikely that any high C[O.sub.2] emission episodes were missed using this sampling protocol. Measurements using the infrared gas analyser (first 73 days) were usually made in mid-morning/early afternoon when temperature was close to the diurnal maximum. As no adjustment was made for temperature-related diurnal variation in C[O.sub.2] emissions (Akinremi et al. 1999), it is likely that C[O.sub.2]-C output was overestimated rather than underestimated. Measurements of C[O.sub.2] emissions ceased about 10 days before straw recovery; however, estimates obtained by extrapolating the data in Fig. 2 suggest that straw-C respired as C[O.sub.2] during that 10-day period would be quite small (<2 g C/[m.sup.2]). Cumulative C[O.sub.2]-C evolution represented an average of 21% of added C where straw was soil-mixed. This does not appear unrealistically low compared with some reports. For example, in a year-long study, Jensen et al. (1997) found that only 23-31% of residue C (residue incorporated by rotovating) was respired as C[O.sub.2], whereas mass loss would be expected to be ~70%. In a laboratory experiment in which canola (Brassica napus) residues were mixed into soil, 33-35% of residue-C was respired in a 6-month incubation at 20[degrees]C (Bhupinderpal-Singh et al. 2006).
The litter bag technique is a well-established method and there is no reason to believe that there was any significant loss of undecomposed straw during recovery of the bags from the plots, transport to the laboratory, or in subsequent processing. There may have been some leaching of water-soluble organic matter from the litter bags during the winter, but leaching probably would not make a large contribution to straw mass loss (Berg and McClaugherty 2003). Losses from the litter bags may have occurred due to fragmentation or removal by earthworms and arthropods, but the extent of such losses is difficult to quantify. Very few earthworms were noted during set-up of the experiment, suggesting that straw removal by earthworms would be minimal.
Only part of the decomposed straw C is likely to have been respired as C[O.sub.2]; some will also have been assimilated into microbial biomass and microbial metabolites (Paul and Clark 1989; Witter and Dahlin 1995; Devevre and Horwath 2000). The relative amounts of straw C respired as C[O.sub.2]-C v. synthesised into microbial tissue depends on microbial growth efficiency, which is calculated from the equation:
[C.sub.s] = [C.sub.m](1 + y/(100 - y)) (1)
where [C.sub.s] is the total C decomposed, [C.sub.m] is C respired as C[O.sub.2], and y is the efficiency of C use for biosynthesis of new biomass (Paul and Clark 1989). Utilisation efficiency values used in soil C turnover models are: 60% for the easily decomposable straw fraction (~15% of straw mass); 40% for the slowly decomposable straw fraction; and 10% for lignin (Paul and Clark 1989). These microbial efficiency values are consistent with recent measurements that indicated y was in the range 40-60% for rice straw during a 160-day decomposition period (Devevre and Horwath 2000). Substituting mean values for litter bag straw (i.e. [C.sub.s] = 96 g C/[m.sup.2] and [C.sub.m] =38 g C/[m.sup.2]) into Eqn 1 gives a value of 60% for y, which is at the high end of the range observed by Devevre and Horwath (2000) and similar to the model value for the easily decomposable straw (Paul and Clark 1989). In the case of litter bag straw, where mass loss was only ~30%, easily decomposable material could account for a large part of the mass loss, and a growth efficiency value of 60% may not be unrealistic.
Decomposing straw is known to support a large microbial population (Beare et al. 2002) but that microbial C would be measured as part of the recovered straw C. For microbial C synthesis to explain the difference between C loss from litter bags and C[O.sub.2]-C output, the microbial C would have to be located outside the litter bag so as not to be measured as part of the recovered straw C. It has been demonstrated that microbial biomass in soil above and below buried litter bags may contain substantial quantities of residue-derived C (Fliessbach et al. 1995). Recently, evidence has been presented that filamentous fungi can translocate straw C into the soil while transferring N in the opposite direction (Frey et al. 2003). The latter workers studied fungally mediated C translocation from surface straw; however, this mechanism is likely to operate to an even greater extent with buried bag straw because the soil-straw interface is larger (above and below the straw layer). Frey et al. (2003) observed that, where [sup.13]C-labelled wheat straw (~2 t/ha) was placed on the soil surface for 35 days, significant straw-derived C (120-150 mg C/kg soil) was measured in the underlying 25 mm soil layer as a result of fungal translocation. Extrapolation of these findings to our study suggests that C translocation by fungi to the soil above and below the litter bags could amount to 20-30 g C/[m.sup.2], assuming that translocated C increases in proportion to the straw application rate. Unfortunately, no measurements were made to quantify changes in soil microbial biomass. Although fungal C translocation may offer a partial explanation for the difference between mass loss of C and C[O.sub.2]-C evolved from the litter bags, further work is needed to examine the role of fungal C translocation as a mechanism of C loss from buried straw.
Evolved C[O.sub.2]-C accounted for an average of about 40% of straw C lost from the litter bags, whereas about 30% of the unrecovered C in the soil-mixed straw treatments could be attributed to C loss as C[O.sub.2] (Table 4). This would imply that straw C recovery was not greatly underestimated in the soil-mixed treatments. Based on the ratio of C[O.sub.2]-C evolved to C mass loss observed for the litter bags, loss of straw C in the straw-mixed treatments would average 160 g/[m.sup.2], compared with the measured value of 209 g/[m.sup.2]. The difference may be assumed to represent C in straw fragments that were too small to be physically recovered from the soil. These small, partially decomposed straw fragments could legitimately be considered part of the soil organic matter, most probably the light fraction or particulate organic matter. Increases in light fraction organic matter (material recoverable by flotation on a dense liquid) following incorporation of straw have been demonstrated by Magid et al. (1997b).
A strict definition of decomposition is difficult and residue decomposition has, by default, been defined by the method used to measure it (Berg and McClaugherty 2003). Residue mass loss comprises C respired as C[O.sub.2], C partitioned to the microbial biomass, and C in residue fragments that become part of soil C pools such as light fraction and particulate organic matter. The latter may represent a relatively large part of the mass loss where unconfined straw is used. Models capable of predicting decomposition of unconfined straw could assist greatly in developing practical management practices for post-harvest residues. For this purpose, straw that has physically fragmented beyond a certain point (e.g. passes through a 2-mm sieve) could effectively be regarded as decomposed, although that criterion may not fit with some definitions of decomposition. Where these residue fragments need to be specifically quantified, this may be achieved using density separation techniques (Magid et al. 1997a).
Summary and conclusions
This study has confirmed that straw mixed through the soil decomposes more rapidly than when it is localised, as occurs when plough-incorporated. Effects of within-soil placement on decomposition persisted for about 2 months following incorporation. Over the course of the experiment, cumulative C[O.sub.2] evolution from soil-mixed straw was 1.7 times that from litter bag straw. Straw mass recovered at the end of the experiment averaged 68% for litter bag straw compared with 34% for soil-mixed straw and 86% for surface-placed straw. Carbon respired as C[O.sub.2]-C accounted for 29-46% of mass loss of incorporated straw-C. The unrecovered straw-C presumably accumulated in soil C pools such as microbial biomass and particulate organic matter, with the latter likely to be an important repository where straw was mixed through the soil. Straw type had only small effects C[O.sub.2]-C emissions and mass loss, although there was evidence that barley straw decomposed more rapidly than wheat straw, possibly because it was higher in water-soluble C and had a higher N content than wheat straw. The litter bag technique may seriously underestimate decomposition rate of straw that is mixed through the soil, and decomposition models need to be revised to take within-soil placement into account.
Funding for this research was provided by the Foundation for Research, Science and Technology. We thank Sarah Glasson and Charles Wright for technical assistance, and Esther Meenken and Ruth Butler for statistical advice.
Manuscript received 8 June 2007, accepted 8 January 2008
Akinremi OO, McGinn SM, McLean HDJ (1999) Effects of soil temperature on soil respiration in barley and fallow plots. Canadian Journal of Soil Science 79, 5-13.
Beare MH, Wilson PE, Fraser PM, Butler RC (2002) Management effects on barley straw decomposition, nitrogen release, and crop production. Soil Science Society of America Journal 66, 848-856.
Berg B, McClaugherty C (2003) 'Plant litter: decomposition, humus formation, carbon sequestration.' (Springer-Verlag: Berlin)
Bhupinderpal-Singh, Rengel Z, Bowden JW (2006) Carbon, nitrogen and sulphur cycling following incorporation of canola residue of different sizes into a nutrient-poor soil. Soil Biology & Biochemistry 38, 32-42. doi: 10.1016/j.soilbio.2005.03.025
Bremer E, van Houtum W, van Kessel C (1991) Carbon dioxide evolution from wheat and lentil residues as affected by grinding, added nitrogen, and the absence of soil. Biology and Fertility of Soils 11, 221-227. doi: 10.1007/BF00335771
Campbell CA, Zentner RP (1993) Soil organic matter as influenced by crop rotations and fertilization. Soil Science Society of America Journal 57, 1034-1040.
Christensen BT (1985) Wheat and barley straw decomposition under field conditions: effect of soil type and plant cover on weight loss, nitrogen and potassium content. Soil Biology & Biochemistry 17, 691-697. doi: 10.1016/0038-0717(85)90047-1
Christensen BT (1986) Barley straw decomposition under field conditions: effect of placement and initial nitrogen content on weight loss and nitrogen dynamics. Soil Biology & Biochemistry 18, 523-529. doi: 10.1016/0038-0717(86)90010-6
Cogle AL, Saffigna PG, Strong WM (1989) Carbon transformations during wheat straw decomposition. Soil Biology & Biochemistry 21, 367-372. doi: 10.1016/0038-0717(89)90145-4
Cogle AL, Saffigna PG, Strong WM, Ladd JN, Amato M (1987) Wheat straw decomposition in subtropical Australia. I. A comparison of [sup.14]C labelling and two weight-loss methods for measuring decomposition. Australian Journal of Soil Research 25, 473-479. doi: 10.1071/SR9870473
Curtin D, Selles F, Wang H, Campbell CA, Biederbeck VO (1998) Carbon dioxide emissions and transformation of soil carbon and nitrogen during wheat straw decomposition. Soil Science Society of America Journal 62, 1035-1041.
Devevre OC, Horwath WR (2000) Decomposition of rice straw and microbial carbon use efficiency under different soil temperatures and moistures. Soil Biology & Biochemistry 32, 1773-1785. doi: 10.1016/ S0038-0717(00)00096-1
Douglas CLJ, Rickman RW (1992) Estimating crop residue decomposition from air temperature, initial nitrogen content, and residue placement. Soil Science Society of America Journal 56, 272-278.
Fliessbach A, Sarig S, Walenzik G, Steinberger Y, Martens R (1995) Mineralisation and assimilation processes of [sup.14]C-labelled shoots of Stipa capensis in a Negev desert soil. Applied Soil Ecology 2, 155-164. doi: 10.1016/0929-1393(95)00058-S
Francis GS, Curtin D, Fraser P, Allen A, Tabley FJ, Butler R (1999) Postharvest residue management for cereals. Crop & Food Research Confidential Report 596, Christchurch, NZ.
Frey SD, Six J, Elliott ET (2003) Reciprocal transfer of carbon and nitrogen by decomposer fungi at the soil-litter interface. Soil Biology & Biochemistry 35, 1001-1004. doi: 10.1016/S0038-0717(03)00155-X
GenStat (2005) 'The guide to GenStat release 8--Parts 1-3.' (VSN International: Oxford)
Heal OW, Anderson JM, Swift MJ (1997) Plant litter quality and decomposition: an historical overview. In 'Driven by nature: plant litter quality and decomposition'. (Eds G Cadisch, KE Giller) pp. 3-30. (CAB International: Wallingford, UK)
Hewitt AE (1993) 'New Zealand Soil Classification.' (Manaaki-Whenua Press: Lincoln, NZ)
Horwath WR, Paul EA (1994) Microbial biomass. In 'Methods of soil analysis. Part 2. Microbiological and biochemical properties'. (Ed. RW Weaver) pp. 753-773. (Soil Science Society of America: Madison, WI)
Jensen LS, Mueller T, Magid J, Nielsen NE (1997) Temporal variation of C and N mineralization, microbial biomass, and extractable organic pools in soil after oilseed rape straw incorporation in the field. Soil Biology & Biochemistry 29, 1043-1055. doi: 10.1016/S0038-0717(97)00014-X
Kanal A (1995) Effect of incorporation depth and soil climate on straw decomposition rate in a loamy Podzoluvisol. Biology and Fertility of Soils 20, 190-196. doi: 10.1007/BF00336557
Keeney RR, Nelson DW (1982) Nitrogen--inorganic forms. In 'Methods of soil analysis. Part 2. Chemical and microbiological properties'. (Eds AL Page et al.) pp. 643-698. (ASA and SSSA: Madison, WI)
Magid J, Jensen LS, Mueller T, Nielsen NE (1997a) Size-density fractionation for in situ measurements of rape straw decomposition--an alternative to the litterbag approach? Soil Biology & Biochemistry 29, 1125-1133. doi: 10.1016/S0038-0717(96)00306-9
Magid J, Mueller T, Jensen LS, Nielsen NE (1997) Modeling the measurable: interpretation of field-scale C[O.sub.2] and N-mineralization, soil microbial biomass and light fractions as indicators of oilseed rape, maize and barley straw decomposition. In 'Driven by nature: plant litter quality and decomposition'. (Eds G Cadish, KE Giller) pp. 349-362. (CAB International: Wallingford, UK)
Parr JF, Papendick RI (1978) Factors affecting the decomposition of crop residues by microorganisms. In 'Crop residue management systems'. (Ed. WR Oschwald) pp. 101-129. (American Society of Agronomy: Madison, WI)
Parr JF, Reuszer HW (1959) Organic matter decomposition as influenced by oxygen level and method of application to soil. Soil Science Society of America Proceedings 24, 213-216.
Paul EA, Clark FE (1989) 'Soil microbiology and biochemistry.' (Academic Press: San Diego, CA)
Summerell BA, Burgess LW (1989) Decomposition and chemical composition of cereal straw. Soil Biology & Biochemistry 21, 551-559. doi: 10.1016/0038-0717(89)90129-6
Vance ED, Brookes PC, Jenkinson DS (1987) An extraction method for measuring microbial biomass C. Soil Biology & Biochemistry 19, 703-707. doi: 10.1016/0038-0717(87)90052-6
Wang H, Curtin D, Jame YW, McConkey BG, Zhou HF (2001) Simulation of soil carbon dioxide flux during plant residue decomposition. Soil Science Society of America Journal 66, 1304-1310.
Witter E, Dahlin S (1995) Microbial utilization of [sup.14]C-labelled straw and [sup.13]C-labelled glucose in soils of contrasting pH and metal status. Soil Biology & Biochemistry 27, 1507-1516. doi: 10.1016/0038-0717(95) 00107-P
D. Curtin (A,B), G. S. Francis (A), and F. M. McCallum (A)
(A) New Zealand Institute for Crop & Food Research Limited, Private Bag 4704, Christchurch, New Zealand.
(B) Corresponding author. Email: email@example.com
Table 1. Composition of wheat and barley straws used in the experiment Water-soluble Total Total C (%) (A) C N Nitrate N C:N Cold Hot Ash Straw (%) (mg/kg) ratio water water (%) Wheat 44.4 0.50 16 89 3.3 4.2 6.5 Barley 43.7 0.68 640 64 3.9 4.7 6.3 (A) Cold water-soluble C was extracted in deionised water at room temperature and hot water-soluble C at 95[degrees]C (both 1-h extractions at a 10:1 water-to-straw ratio); dissolved C was measured using a Shimadzu Total Organic C Analyzer. Table 2. Soil volumetric moisture content, measured on 8 occasions, and mineral N at termination of the experiment (0-0.20 m depth) Volumetric moisture content (%) Treatment Day 6 Day 25 Day 31 Day 35 Day 40 Control 12.4 11.9 18.6 29.6 24.8 Barley, litter bag 11.6 11.1 15.1 27.2 23.1 Wheat, litter bag 11.4 11.4 15.3 27.0 23.3 Barley, soil mixed 12.2 11.5 17.6 30.0 25.4 Wheat, soil mixed 11.9 11.9 18.3 30.8 26.1 l.s.d. (P = 0.05) 1.2 Volumetric moisture content (%) Mineral N Treatment Day 47 Day 66 Day 125 (kg/ha) Control 25.0 28.2 25.2 16 Barley, litter bag 24.9 29.1 23.5 16 Wheat, litter bag 24.4 28.7 23.5 14 Barley, soil mixed 25.4 28.8 26.9 22 Wheat, soil mixed 25.8 29.5 26.5 18 l.s.d. (P = 0.05) 4 Table 3. Amounts of C[O.sub.2]-C evolved (g C/[m.sup.2]) during specified time periods when straw was confined in litter bags or mixed through the soil Cumulative C[O.sub.2]-C evolved Total Treatment Days 0-73 Days 73-146 (Days 0-146) Control 34.3 39.3 73.7 Barley, litter bag 61.7 52.1 113.8 Wheat, litter bag 59.4 50.7 110.1 Barley, soil mixed 86.2 56.7 142.9 Wheat, soil mixed 79.7 51.9 131.4 l.s.d. (P = 0.05) 7.4 3.8 8.7 Increase in C[O.sub.2]-C relative to control Total Treatment Days 0-73 Days 73-146 (Days 0-146) Control -- -- -- Barley, litter bag 27.2 12.9 40.1 Wheat, litter bag 25.1 11.4 36.4 Barley, soil mixed 51.9 17.3 69.2 Wheat, soil mixed 45.2 12.6 57.8 l.s.d. (P = 0.05) 7.6 4.4 9.6 Table 4. Effect of straw placement (in litter bags at 0.20 m, mixed through 0-0.20 m layer, placed on soil surface) on mass recovery and C and N contents of recovered wheat and barley straw Recovered (%) Treatment Straw mass Straw N Straw C Barley, litter bag 69.2 65.3 71.5 Wheat, litter bag 66.7 77.3 66.4 Barley, soil mixed 29.4 38.6 28.8 Wheat, soil mixed 38.7 51.2 36.0 Barley, surface 86.7 66.4 86.8 Wheat, surface 87.8 82.1 88.7 l.s.d. (P = 0.05) 3.8 6.1 5.9 Straw C C[O.sub.2]-C decomposed (A) evolved/ Treatment (g/[m.sup.2]) decomposed C (%) Barley, litter bag 87 46 Wheat, litter bag 104 35 Barley, soil mixed 218 32 Wheat, soil mixed 199 29 Barley, surface 40 n.d. Wheat, surface 35 n.d. l.s.d. (P = 0.05) 18 10 n.d., Not determined. (A) Straw C decomposed = straw C added minus straw C recovered.
|Printer friendly Cite/link Email Feedback|
|Author:||Curtin, D.; Francis, G.S.; McCallum, F.M.|
|Publication:||Australian Journal of Soil Research|
|Date:||Mar 1, 2008|
|Previous Article:||Microbial biomass carbon and some soil properties as influenced by long-term sodic-water irrigation, gypsum, and organic amendments.|
|Next Article:||Loss of phosphorus and nitrogen in runoff and subsurface drainage from high and low input pastures grazed by sheep in southern Australia.|