SSSAJ Grow Your Career with SSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 3 August 2006
Published in Soil Sci Soc Am J 70:1512-1521 (2006)
DOI: 10.2136/sssaj2005.0338
© 2006 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Curtin, D.
Right arrow Articles by McCallum, F. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Curtin, D.
Right arrow Articles by McCallum, F. M.
Agricola
Right arrow Articles by Curtin, D.
Right arrow Articles by McCallum, F. M.
Related Collections
Right arrow Nutrient Management
Right arrow Nitrogen
Right arrow Soil Analysis

Soil Biology & Biochemistry

Hot Water-Extractable Nitrogen as an Indicator of Soil Nitrogen Availability

D. Curtin*, C. E. Wright, M. H. Beare and F. M. McCallum

New Zealand Institute for Crop & Food Research, Private Bag 4704, Christchurch, New Zealand

* Corresponding author (curtind{at}crop.cri.nz)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
There is keen interest among soil scientists in identifying chemical assays that may be used as predictors of soil N mineralization potential. Our objective was to determine if hot water-extractable N (16-h extraction at 80°C) is a useful predictor of mineralizable N and plant N availability. In a group of 30 New Zealand soils, representing different management histories and parent materials, hot water extracted between 2.6 and 8.7% of total N. The extracted N consisted mainly (~80%) of organic N, with the remainder being NH4–N, generated by hydrolysis of heat-labile organic N. The C/N ratio of the extracted organic matter was relatively low (mean 8:1 vs. 11:1 for total organic matter), indicating that it included N-rich substrates (i.e., substrates likely to have high mineralization potential). However, about three-quarters of the extracted organic N was relatively recalcitrant, i.e., it did not hydrolyze to ninhydrin-reactive N (NH4–N, amino acid-N, amino sugar N) when treated with 1 M HCl for 6 h at 80°C. The contribution of mineralized N to plant N uptake was measured using a greenhouse-grown oat (Avena sativa L.) crop, which received no added N. Hot water-extractable N accounted for 50% of the variation in plant N derived from mineralization (PNDM), compared with 16% for total soil N, 32% for anaerobically mineralizable N (AMN), and 24% for NH4–N released by hot 2 M KCl. The best predictor of PNDM was N mineralized in a 28-d aerobic incubation at 20°C (79% of variability in PNDM explained).

Abbreviations: AMN, anaerobically mineralizable nitrogen • Aerobic N, nitrogen mineralized under aerobic conditions • HWC, hot water-extractable C • PNDM, plant nitrogen derived from mineralization


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
NITROGEN supplied by mineralization of soil organic matter can be an important N source for crops. Predicting the contribution of this N source is problematic, partly because quantification of mineralizable N requires a laborious incubation assay, which may take several weeks to complete (Campbell et al., 1993). Soil scientists are keenly interested in developing chemical assays that could be used as proxies for mineralizable N because such tests are likely to be cheaper and more rapid and precise than incubation assays (Gianello and Bremner, 1986; Jalil et al., 1996). In theory, mild extractants should be superior to harsh chemicals (e.g., acids), which inevitably cause hydrolysis of some non-labile N. Extraction with hot aqueous solutions has been used to measure labile organic C (Sparling et al., 1998; Ghani et al., 2003), carbohydrates (Puget et al., 1999; Haynes, 2000; Ghani et al., 2003), and plant-available N and S (Gianello and Bremner, 1986; Blair et al., 1991). The extractants used have included hot water (Ghani et al., 2003; Gregorich et al., 2003) and hot KCl solutions at concentrations ranging from 0.25 to 2 M (Gianello and Bremner, 1986; Blair et al., 1991).

Decomposition studies by Gregorich et al. (2003) provide evidence that the hot water-extractable fraction of soil C is highly labile. Their results suggest that organic C extracted by hot water is comprised of two kinetically discrete fractions. The most labile component (about one-third of hot water soluble C) had a turnover time of <1 d at an incubation temperature of 35°C. The second C fraction was also quite labile (turnover time ~80 d).

Good correlations with microbial biomass C suggest that the C extracted in hot water is partly of microbial origin (Sparling et al., 1998). Based on chemical composition and pyrolysis-field ionization mass spectrometry studies, Leinweber et al. (1995) concluded that the organic matter extracted in hot water originates from soil microbial biomass, root exudates, and lysates.

There has been an upsurge in interest in dissolved and water-extractable organic matter in recent years, as exemplified by the convening of an international conference on the topic in 2001 (Kalbitz and Kaiser, 2003). While papers presented at that forum focused mainly on C, several authors identified a need for further work on the role of dissolved organic N in the cycling of N in the soil–crop system (Chantigny, 2003; Kalbitz and Kaiser, 2003). Given the evidence that it is readily decomposable, hot water-extractable organic matter deserves attention as a source of potentially mineralizable N.

Very few authors have measured hot water-extractable N and even fewer have attempted to determine its biodegradability. While it seems reasonable to assume that the behavior of hot water-extractable N would parallel that of C, there are also compelling reasons why this might not be the case. Carbon solubilized by hot water is measured as dissolved organic C but organic N may be partly hydrolyzed to inorganic N (NH4–N) in hot extractants (Gregorich et al., 2003). In fact, the NH4–N extracted in hot 2 M KCl has been proposed as a test for potentially mineralizable N (Gianello and Bremner, 1986; Jalil et al., 1996). There appears to be no published information on the relative amounts of NH4–N and organic N extracted by hot water. Some of the NH4–N released during hot water treatment will adsorb onto cation exchange sites on soil colloids. Partitioning of NH4–N between the solution and exchange phases will depend on factors such as cation selectivity of the exchange sites and the concentration of competing cations in solution. Recovery of the adsorbed NH4–N requires the additional step of extraction with a salt solution such as KCl.

The objectives of the study were to: (i) determine amounts and forms of N extracted in hot water from a diverse group of soils; and (ii) relate hot water-extractable N to labile and mineralizable pools of soil N and to plant N uptake.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soils
Thirty soils (0- to 15-cm depth), representing different management histories and a range of soil types, were collected at different locations throughout New Zealand (Table 1). Management history has a strong influence on N mineralization potential of New Zealand soils. Land that is under ryegrass (Lolium perenne L.)–clover (Trifolium repens L.) pasture for several years can accumulate large amounts of labile N whereas soils with a history of intensive arable cropping typically have a low mineralization potential (Goh, 1983). The sampled soils included nine that had been under long-term (8–10 yr) ryegrass-clover pasture and a further nine that had been under long-term (8–10 yr) arable cropping. Short-term (<3–4 yr) pasture and arable soils each had four representatives. The soils were derived from sedimentary parent materials except for four soils derived from volcanic ash from the Pukekohe area, south of Auckland. Further details of the soils may be obtained from Curtin and McCallum (2004).


View this table:
[in this window]
[in a new window]
  		Table 1. Management history and total and labile C and N in 30 soils used in the study.

 
Water-Extractable Carbon and Nitrogen
Water-extractable C and N were measured using samples (in duplicate) of air-dry soil (<2 mm) as described by Ghani et al. (2003). In a preliminary step, readily soluble organic matter was removed by extracting with deionized water at room temperature (hereafter referred to as cold water extraction). This involved shaking 4-g samples of soil with 30 mL of deionized water in 50-mL centrifuge tubes for 30 min. The soil-water suspension was then centrifuged (2600 x g), and the supernatant filtered through a filter paper (Whatman #42) that had been preleached with 40 mL of deionized water. The centrifuge tube plus wet soil was weighed to calculate the entrained water volume. Another 30-mL aliquot of water was added and, after mixing to resuspend the soil, the tubes were placed in a hot-water bath at 80°C for 16 h. The tubes were then centrifuged and the supernatant solution collected, as described above. Finally, the soils were extracted with 30 mL of 2 M KCl to remove adsorbed NH4–N.

Organic C in the cold and hot water extracts was determined using a Shimadzu TOC Total Organic Carbon Analyzer (Shimadzu Corp., Kyoto, Japan). Ammonium- and NO3–N in the water and KCl extracts were determined using standard colorimetric procedures (Keeney and Nelson, 1982). Total N in the cold and hot water extracts was determined using persulfate to oxidize organic N and NH4–N to NO3–N, as described by Cabrera and Beare (1993). Organic N was estimated by subtracting mineral N from total N. In the case of the KCl extracts, only mineral N was measured. Corrections were applied for carry-over of N from the previous extract based on the volume of solution carried over and its N concentration.

Ninhydrin-reactive N (i.e., free amino acid N, amino sugar N, and NH4–N) in the hot water extracts was determined as described by Joergensen and Brookes (1990), using the amino acid leucine as the calibration standard. Preliminary tests showed that other forms of ninhydrin-reactive N (NH4–N; glucosamine-N) gave very similar calibration curves to leucine-N.

The rate of release of N (and C) to hot water was examined using three selected soils. After the initial cold water extraction step, samples were extracted with hot water for periods ranging from 1 to 16 h (water was preheated to 80°C before adding it to the soil). Finally, the soils were extracted with 2 M KCl to remove adsorbed NH4–N. Carbon and N in these extracts were determined as described above.

Soil Organic Matter Fractions
Total soil C and N were determined using a LECO CNS-200 analyzer (LECO Corp, St. Joseph, MI). Methods used to measure mineralizable N have been described in detail by Curtin and McCallum (2004). Briefly, AMN was determined as the amount of N generated during a 7-d anaerobic incubation at 40°C (Keeney and Bremner, 1966). Nitrogen mineralization under aerobic conditions was measured by incubating soils at 20°C and optimum moisture content (95% of field capacity; Campbell et al., 1994) for 28 d. Mineralized N was estimated by subtracting mineral N in the soils before incubation from the amounts determined after the incubation.

Ammonium N release in hot 2 M KCl was determined by heating a soil-KCl suspension (3 g soil and 20 mL of 2 M KCl) in a glycol bath at 95°C for 16 h (Gianello and Bremner, 1986). Ammonium N released by hot KCl was calculated by subtracting native NH4–N (extracted with 2 M KCl at room temperature) from the total NH4–N in the hot KCl extract. Microbial biomass was determined by the chloroform fumigation-extraction method (48-h fumigation, 1-h extraction with 0.5 M K2SO4; Horwath and Paul, 1994). Carbon in the K2SO4 extracts was measured using a Shimadzu TOC-5000A Total Carbon Analyzer. Microbial biomass C was estimated assuming an extraction efficiency factor of 0.38 (Vance et al., 1987).

Total C and N, AMN, and hot KCl N were measured on air dry soil (<2 mm); microbial biomass and N mineralization under aerobic conditions were determined using field-moist soil (<4 mm). All C and N measurements were made in duplicate.

Plant Nitrogen Availability
Plant N uptake was measured in a pot experiment using oats (Avena sativa L., cultivar Croa 50) as the test plant. Seeds were sown in plastic pots (30 seeds pot–1) containing the equivalent of 3 kg of dry soil and grown to maturity in a greenhouse. There were four replicate pots of each soil, arranged in a randomized block design. The plants were grown during the autumn-winter period under natural daylight conditions. Greenhouse temperature was set at 18°C during daytime and 12°C at night. No N was added, so the plants were totally dependent on soil N (mineral N plus mineralized N). To ensure that other nutrients (P, K, S) were not limiting, a solution containing K2SO4 and KH2PO4 was applied after seedling emergence to supply 40 mg P kg–1, 100 mg K kg–1, and 20 mg S kg–1. The pots were watered regularly to prevent plant moisture stress. Plants were harvested at maturity (140 d after sowing) by cutting stems about 1 cm above the soil surface. These were separated into straw and grain, dried (60°C) and analyzed for N (LECO CNS-200 analyzer). An estimate of plant N derived from mineralization was obtained by subtracting mineral N in the soil at sowing from total N recovered in aboveground biomass (soil mineral N at harvest was very low; average 4 mg kg–1).

Statistics
Relationships between water-extractable N and other measures of labile and plant available N were evaluated using linear regression analysis and correlations.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
The 30 soils represented a broad range of organic matter contents, with total organic C ranging from 17 to 59 g kg–1 (Table 1). Pasture soils tended to have more organic matter than soils under arable crops. Mean content of total organic C in the nine long-term pasture soils included in the study was 39 g kg–1 compared with 26 g kg–1 for the nine long-term arable soils. These results are in keeping with previous findings in New Zealand that intensively cropped soils typically have about one-third less organic matter than similar soils under permanent pasture (Haynes, 1991; Saggar et al., 2001). The labile fractions (microbial biomass, mineralizable N) were usually more strongly affected by management than was total organic matter. On average, microbial biomass was about twice as high in long-term pasture soils compared with long-term arable soils.

Cold Water-Extractable Nitrogen and Carbon
Cold water soluble C (Table 2) ranged from 164 to 634 mg kg–1 (mean of 308 mg kg–1), similar to values reported for maize-cropped soils in Ontario, Canada (range 280 to 570 mg kg–1; Gregorich et al., 2003). Measurements of cold water soluble C were made on air-dry soil and the values are likely to be higher than those for field-moist soils. Rolston and Liss (1989) showed that cold water soluble C values for air-dry soils were more than twice those of field-moist samples.


View this table:
[in this window]
[in a new window]
  		Table 2. Amounts of C and N extracted from 30 soils by cold water followed by hot water and in a subsequent extraction with 2 M KCl.

 
Cold water-extractable organic N showed a significant, but not very close (R2 = 0.69), correlation with cold water-extractable C. The few reported studies that have included measurements of both N and C suggest that the ratio of C to organic N in cold water extracts can be quite variable. In a recent review, McDowell (2003) cited evidence that dissolved organic N and C may not be closely coupled because the factors governing production of soluble C and N can be different. For example, dissolved organic N may double as a result of N fertilization without a corresponding increase in dissolved C (McDowell, 2003). In our study, the C/N ratio of cold water extractable organic matter (mean 13.8) was generally wider than that of whole soil organic matter (mean 11.0), possibly indicating the importance of recent plant debris with high C/N ratio as a source of the extracted organic matter (Ghani et al., 2003).

Concentrations of cold water-extractable C and N were significantly higher for pastures than for arable soils (Table 3). On average, organic N extracted by cold water was 0.9% of total N in long-term pastures compared with 0.75% in long-term arable soils (difference not significant at P = 0.05). Short-term pasture and short-term arable cropping were each represented by only four soils, precluding meaningful comparison with the long-term systems.


View this table:
[in this window]
[in a new window]
  		Table 3. Mean levels of total and labile N and C in soils under long-term pasture and long-term cropping (each land-use class was represented by nine paddocks).

 
Hot Water-Extractable Nitrogen and Carbon
Hot water-extractable C (HWC) was a much larger C pool than cold water C. On average, HWC represented 81% of total water-soluble C (cold plus hot water extractable). The proportion of total soil organic C extracted in hot water was quite variable (2–7.5%) and, as a result, the relationship between HWC and total C was relatively weak (R2 = 0.44). Pastoral soils had more HWC than long-term arable soils (Table 3) and the proportion of total C that was extractable in hot water was significantly greater under pasture (Table 3). This supports suggestions that HWC is more sensitive to changes in land use than is total C (Ghani et al., 2003).

Most (average 80%) of the N liberated by hot water was in an organic form. However, significant quantities of NH4–N were found in the hot water extract itself and in the subsequent KCl extract (Table 2). Native soil NH4–N, which was present in trace amounts only (average 1.2 mg kg–1), was largely removed by cold water extraction. Thus, the NH4–N released by hot water can be attributed to hydrolysis of heat-labile organic N compounds, likely to include amino sugars (Gianello and Bremner, 1986; Gregorich et al., 2003).

Ammonium N extracted by KCl after the hot water treatment represented 53 to 71% of the total NH4–N liberated by hot water. Overall, there was a good linear relationship between NH4–N in hot water (x) and NH4–N subsequently extracted by KCl (y):

Formula 1[1]
The regression equation had a significant positive intercept, which probably signifies that the soils had some exchange sites with high affinity for NH4 (i.e., sites capable of adsorbing NH4 when its concentration in solution is very low). The intercept value (7 mg NH4–N kg–1) translates to 0.05 cmolc kg–1, which is <1% of the average cation exchange capacity of the soils (6.1 cmolc kg–1). The regression equation suggests that, once sites with strong affinity for NH4 were filled, NH4 was partitioned on an approximately 1:1 basis between the solution and exchange phases. This equation could allow realistic estimates of adsorbed NH4 based on the concentration of NH4 in the hot water extract, thus avoiding the need for the additional KCl extraction step.

In most soils, NH4–N (soluble plus adsorbed) represented between 15 and 20% of the total N liberated by hot water, but the proportion of NH4–N tended to be higher in the volcanic ash-derived soils (Pukekohe soils), where it accounted for up to 30% of the released N (Table 2). The volcanic ash soils had high contents of Fe/Al oxides compared with all other soils in the study. Adsorption to oxide surfaces is an important factor governing the concentration of dissolved organic matter (Kaiser and Guggenberger, 2000) and it is possible that in the volcanic ash soils some organic matter was re-adsorbed before separation of the hot water extracts. This would account for the higher ratios of NH4–N to organic N found in these soils. Overall, total NH4–N was reasonably well correlated with organic N (R2 = 0.72) and C (R2 = 0.68) extracted in hot water, but the relationship was considerably improved (R2 = 0.93 for organic N and R2 = 0.90 for C) by exclusion of the volcanic ash soils.

The relationship between C and N liberated by hot water was excellent both when NH4–N in the subsequent KCl extract was included and excluded (Fig. 1 ). With the KCl extracted NH4–N excluded, the mean C/N ratio of hot water extractable organic matter was 9.1, which is less than that of the bulk organic matter (11.0). The C/N ratio decreased to an average of 8.1 when KCl extracted NH4–N was included. The amount of adsorbed NH4–N is sufficiently large that its exclusion would result in significant overestimation of the C/N ratio of hot water-soluble organic matter. The lower C/N ratio of hot water soluble vs. whole soil organic matter is consistent with suggestions (Sparling et al., 1998) that N-rich microbial biomass is an important source of organic matter extracted in hot water. Fractions of soil organic matter with high N mineralization potential are likely to be enriched in N (low C to N ratio) and the hot water extractable fraction meets this criterion. The organic N extracted by hot water was only weakly related to total soil N (R2 = 0.38) but the relationship between hot water-released NH4–N (soluble plus adsorbed) and total soil N was much better (R2 = 0.78), possibly suggesting that a wider range of organic matter fractions contributed to the NH4–N than to the organic N extracted by hot water.


Figure 1
View larger version (18K):
[in this window]
[in a new window]
  		Fig. 1. Relationship between hot water-extractable C and N when NH4–N recovered in the subsequent extraction with 2 M KCl was (A) included and (B) excluded.

 
Hereafter, unless specifically stated to the contrary, hot water-extractable N will refer to all of the N liberated by hot water (i.e., organic N and NH4–N in the hot water extract itself plus NH4–N in the subsequent KCl extract).

Ninhydrin Reactivity
Ninhydrin reactivity provides a means of characterizing organic N in soil extracts. Ninhydrin-reactive N includes amino acid N and amino sugar N as well as NH4–N. It is commonly used as a measure of microbial N in chloroform-fumigated soils (Amato and Ladd, 1988; Joergensen and Brookes, 1990). Estimates of the amounts of organic N in hot water extracts that reacted with ninhydrin were obtained by subtracting NH4–N from the total ninhydrin-reactive N. On average, about half of the ninhydrin-reactive N in hot water extracts of the 30 soils could be attributed to NH4–N (Table 4). The proportion of the organic N that reacted with ninhydrin was relatively small (average of 9.6% of the organic N; Table 4) indicating that free amino acids and amino sugars comprised a small fraction of the organic N in hot water extracts.


View this table:
[in this window]
[in a new window]
  		Table 4. Total ninhydrin N (TNN), NH4–N and estimated amounts of organic N that were ninhydrin-reactive (OrgNN) for soil hot water extracts before and after acid hydrolysis (6 h treatment with 1 M HCl at 80°C).

 
Hot water extracts were acidified with HCl (1 M) and heated at 80°C for 6 h to determine the susceptibility of the organic N to hydrolysis. This acid treatment caused hydrolysis of some organic N to NH4–N, which approximately doubled in concentration (Table 4). In addition, part of the organic N was hydrolyzed to ninhydrin-reactive organic N, which increased from an average of 14 mg N kg–1 before acid treatment to 25 mg N kg–1 after acid hydrolysis. Generally, between 20 and 30% (average 24%) of the organic N was hydrolyzed by acid treatment to ninhydrin-reactive forms (NH4–N, amino acids, amino sugars). Treatment with 1 M HCl at 80°C for 6 h is a moderately severe hydrolysis. Therefore, the fact that only about one-quarter of the organic N was hydrolyzed suggests that a significant portion of it was relatively recalcitrant. Tests on selected soils showed that extending the acid treatment from 6 to 16 h resulted in relatively small increases in the proportion of the organic N that reacted with ninhydrin (results not shown).

The hot water extracts were yellow to dark brown in color, presumably reflecting the presence of significant quantities of stable humic and fulvic acids. Although the N extracted by hot water has not been characterized, it is known that the organic N found in naturally occurring soil solution is quite heterogeneous, comprising compounds ranging from simple amino acids and amino sugars to polyphenol-bound compounds of high molecular weight, with the latter making up the bulk of the soluble organic N (Jones et al., 2004). Differentiation of the labile from the inactive forms of N would presumably improve the sensitivity of hot water N as an index of labile soil N, but further work is needed to determine how this separation can be best achieved. Our finding that a large proportion of the organic N extracted by hot water was resistant to acid hydrolysis appears to be at variance with reports that hot water soluble organic matter is easily biodegraded (Gregorich et al., 2003). Whether this variance is due to differences in the methodology used (i.e., chemical vs. biological techniques), differences in the chemical characteristics of organic matter in the two studies or perhaps other factors remains unknown and deserving of further investigation.

Time Course of Nitrogen and Carbon Release to Hot Water
In studies using hot water to measure labile organic matter, an extraction period of 16 to 18 h has been preferred (Sparling et al., 1998; Ghani et al., 2003; Gregorich et al., 2003), though the time course of C and N release does not appear to have been established. Since the C, organic N, and NH4–N released by hot water may originate from a range of organic components (carbohydrates, amino acids and proteins, amino sugars, and humic materials), the rate of their release is not necessarily the same. Results for three selected soils (long-term pasture and arable soils derived from sedimentary parent material and a volcanic ash soil) show that the rate of N and C release to hot water was rapid in the first hour and slowed markedly thereafter (Fig. 2 ). Carbon released in the first hour amounted to 37 to 44% of that liberated in the longest extraction period (16 h). Hot water-extractable N followed a very similar release pattern to C (Fig. 2); the C/N ratio of the released organic matter showed no consistent trend with extraction time (data not shown). There was a trend for NH4–N to increase as a proportion of total extracted N as time increased, though the increase was relatively small (Fig. 3 ).


Figure 2
View larger version (23K):
[in this window]
[in a new window]
  		Fig. 2. Cumulative amounts of C and N extracted (A, B), and rates of C and N release (C, D) from three selected soils as a function of extraction time. Error bars indicate standard deviations (error bars less than symbol size not shown). The two Lincoln soils were from property "D" (Table 1).

 

Figure 3
View larger version (19K):
[in this window]
[in a new window]
  		Fig. 3. Ammonium N, as a percentage of total N released by hot water, as a function of extraction time for three selected soils. The two Lincoln soils were from property "D" (Table 1).

 
The results show that there is a fraction of organic matter that is readily extracted in hot water and, once it is removed, relatively little additional C and N is released by extending the extraction period. The pattern of C and N release with time was similar in the three soils that were examined, suggesting that a ranking of soils in terms of their hot water-extractable organic matter levels would not change if the extraction time is altered. While this study did not reveal major changes in the composition of the dissolved organic matter when extraction time was changed, it would be interesting to know if the readily soluble organic matter (i.e., that released in the first hour or so) is more labile than that released over longer extraction periods.

Relationship between Hot Water-Extractable Nitrogen and Labile Carbon and Nitrogen Pools
There was a reasonably good correlation between hot water-extractable N and microbial biomass (Fig. 4 ), consistent with previous suggestions (Sparling et al., 1998) that the extracted material is partly of microbial origin. Hot water-extractable N was also correlated with organic N extracted in cold water, though there were several outliers. The quantities of N extracted by hot water were considerably larger than those measured in the two N mineralization assays. Hot water-extractable N was, on average, about five times that mineralized during the 28-d aerobic incubation at 20°C (mean of 33 vs. 167 mg N kg–1) and about twice that mineralized in the anaerobic incubation (mean of 73 vs. 167 mg kg–1). Relationships between mineralizable N and hot water-extractable N were not very close, particularly in the case of aerobically mineralizable N (Aerobic N) (Fig. 4). When two outlying observations were excluded, the relationship with Aerobic N improved considerably (R2 increased from 0.30 to 0.53). Because HWC and hot water-extractable N were closely related (Fig. 1), HWC exhibited similar relationships with mineralizable N to those observed for the hot water-extractable N (not shown).


Figure 4
View larger version (22K):
[in this window]
[in a new window]
  		Fig. 4. Relationship of hot water-extractable N (N in hot water plus NH4–N in subsequent KCl extract) with: (A) microbial biomass C; (B) organic N extracted in cold water; (C) anaerobically mineralizable N (AMN); and (D) N mineralized in 28-d aerobic incubation.

 
When the four volcanic ash soils were excluded, hot water-extractable N was closely related to hot KCl N (Fig. 5 ). The volcanic ash soils had especially high ratios of hot KCl NH4–N to hot water-extractable N. As noted earlier, these soils were unusual in that NH4–N made up a relatively large fraction of the N released by hot water. The relationship between NH4–N liberated by hot water (i.e., NH4–N in the hot water-extract plus NH4–N in the subsequent KCl extract) and hot KCl N was particularly close (Fig. 5), though somewhat more NH4–N was extracted in hot KCl than by hot water. This is probably mainly because of the higher KCl extraction temperature (95°C for hot KCl vs. 80°C for hot water). Also, the pretreatment step with cold water would have removed some hydrolyzable N that would otherwise be measured in the hot water extraction. Our results strongly suggest that hot water and hot 2 M KCl extract the same fraction of organic matter, though in the case of the hot KCl assay only the NH4–N component is measured. It may be argued on the basis of these results that the utility of the hot KCl assay as an indicator of mineralizable N is because of its relationship with hot water-extractable organic matter, rather than as a unique pool of labile N.


Figure 5
View larger version (13K):
[in this window]
[in a new window]
  		Fig. 5. Relationship of hot KCl N with total hot water-extractable N (A) and NH4–N released by hot water treatment (B). Note: data points marked with arrows in (A) are for volcanic ash (Pukekohe) soils.

 
Relationship between Hot Water-Extractable Nitrogen and Plant Nitrogen Uptake
The total amount of N recovered in grain and straw by the greenhouse-grown oat crop was equivalent to between 44 and 242 mg N kg–1 of soil (Curtin and McCallum, 2004). Plant N derived from mineralization was estimated by subtracting mineral N present in the soil at sowing (Table 1) from crop N uptake (mineral N at harvest was negligible). Most (average of 83%) of the N recovered in the crop (grain and straw) was derived from mineralization (Curtin and McCallum, 2004). The PNDM, which ranged from 25 to 134 mg kg–1 of soil (mean 66 mg kg–1), was considerably less than hot water-extractable N (mean 167 mg kg–1).

Hot water-extractable N was significantly correlated with PNDM, though only 50% of the variability in PNDM was explained (Fig. 6 ). Among the other indices of labile N tested, Aerobic N showed the best relationship with PNDN (~80% of the variability in PNDN explained; Fig. 6). Hot water-extractable N was more closely related to PNDM than were AMN (R2 = 0.32), hot KCl N (R2 = 0.24), or total soil N (R2 = 0.16).


Figure 6
View larger version (14K):
[in this window]
[in a new window]
  		Fig. 6. Relationship of aerobically mineralizable N (A) and hot water-extractable N (B) with estimated amounts of plant N derived from mineralization.

 
Results of this study suggest that, while hot water may not select specifically for potentially mineralizable N, the extracted N does represent a relatively labile part of the total soil organic N. Thus, it does have merit as an indicator of soil N supplying power. Because chemical assays, including the hot water assay, measure specific forms of N, they are likely to be more successful as indicators of gross N mineralization than of net N mineralization (gross N mineralization minus N immobilized). This is because chemical tests cannot make allowances for the N immobilized during the incubation. Differences between the soils in N immobilization may partly explain why net N mineralization in the aerobic incubation (Fig. 4) was not closely related with hot water-extractable N. Rates of N immobilization can be especially high in grassland soils as a consequence of raised availability of C to microorganisms (Ledgard et al., 1998). Curtin and McCallum (2004) presented evidence of greater N immobilization in the long-term pasture vs. arable soils used in this study. For example, the ratio of C/N mineralized during aerobic incubation was significantly higher for pasture than for arable soils (average 14:1 for pasture soils vs. 8:1 for arable soils). Hot water-extractable N performed better as a predictor of plant N uptake (Fig. 6) than of Aerobic N (Fig. 4). One possible reason is that, as plant roots compete with microorganisms for mineralized N, the extent of microbial N immobilization was reduced compared with the plant-free incubation system used to measure mineralizable N.

Hot water-extractable N was more closely related to PNDM (R2 = 0.50) than was hot KCl N (R2 = 0.24). Since the hot water assay includes both organic N and NH4–N, whereas hot KCl N is comprised solely of NH4–N, the superiority of hot water-extractable N may be attributed to the fact that all of the released N was included. Ammonium N generated in hot aqueous extractants (hot water or hot KCl) was strongly correlated with total soil N (R2 = 0.92 for hot KCl N and R2 = 0.78 for NH4–N released by hot water). This evidence that NH4–N released by hot aqueous extractants is determined to a large extent by the total N content of the soil raises doubts as to whether it does, in fact, originate mainly from labile or active N pools. Further work is needed to identify the source of the NH4–N hydrolyzed by hot KCl and hot water. In the meantime, we suggest that hot water-extractable N will provide a more reliable predictor of potentially available N. The ninhydrin-reactive fraction was no better as a predictor of PNDM than was total hot water-extractable N (results not shown).


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Hot water is a mild extractant that removed an average of 6% of total N. Nitrogen released by hot water was mainly in organic form (80% on average) but significant amounts of NH4–N were also released due to hydrolysis of heat-labile organic N. Although part of the extracted organic N was relatively recalcitrant (not susceptible to acid hydrolysis), the hot water-extractable N appears to represent a relatively labile part of the total soil organic N. As a predictor of N availability to the greenhouse-grown oat crop, hot water-extractable N was superior to total N, AMN, and hot KCl N but inferior to Aerobic N. The superiority of hot water-extractable N over the commonly advocated hot KCl test is, presumably, because all or the extracted N (organic N and NH4–N) was measured, whereas the hot KCl test includes only NH4–N. The latter comprises a relatively small and possibly not the most labile part of the N released by hot aqueous solutions.

While it does have merit as a soil N test, measurement of hot water-extractable N is analytically challenging (determination of the organic N may involve tedious wet chemistry and quantitative recovery of the NH4–N released by hot water requires a subsequent extraction with KCl). Fortunately, because there is a close relationship between N and C extracted by hot water, indirect estimates of N can be obtained by measuring dissolved C, which is easily determined using a total organic C analyzer. The rate of N release to hot water was especially rapid in the first hour or two of extraction and further work is needed to determine if this rapidly released N would provide a better indication of potentially available N than N measured following the usual overnight extraction.

Received for publication October 6, 2005.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Curtin, D.
Right arrow Articles by McCallum, F. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Curtin, D.
Right arrow Articles by McCallum, F. M.
Agricola
Right arrow Articles by Curtin, D.
Right arrow Articles by McCallum, F. M.
Related Collections
Right arrow Nutrient Management
Right arrow Nitrogen
Right arrow Soil Analysis

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome