._\ . ‘ l’ ‘ s ’ . 4 , THESlS 9% 9,- 2'33" % 3;. '4_.'-.‘ _. ' 57- 15"“ m“ a W I/flllflfl// /W I LM‘IL“ ’u- be tho; 0 :2? 3291/g////00l i .ifi 1) f" ' -". m. ”5.1.5:.” 1 f: ‘3 @4741” §_, 6 E,, :f “- ‘cm‘: 9 “46%;; :fi" r, ' 4* it "1:. M (.3 fl 3 o 531.: This 18 to certify that the, __ _. . . dissertation entitled - - SOIL DENITRIFICATION: EFFECT OF OXYGEN AND MOISTURE AND MEASUREMENT IN THE FIELD presented by ALAN JOHN SEXSTONE has been accepted towards fulfillment of the requirements for Ph.D. . Microbiolo 8 d 9y wee "‘Tubi ic Health (AW 3% Major professorU / N ——----- “45.” Date Ava- )1! I 1983 0'7 639 s. ' ‘ “v" I fifl‘w . . l Wh _ 4; ‘N ' W ‘ \~ I I PLACE IN RETURN BOX to remove We checkout from your record. T0 AVOlD FINES return on or before date due. II I l MSU to An motive Action/Ewe! Opportunity lnetitwon W H/ w m1 SOIL DENITRIFICATION: EFFECT OF OXYGEN AND MOISTURE AND MEASUREMENT IN THE FIELD By Alan John Sexstone A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Microbiology and Public Health 1983 ABSTRACT SOIL DENITRIFICATION: CONTROL BY OXYGEN AND MOISTURE AND MEASUREMENT IN THE FIELD by ALAN JOHN SEXSTONE Denitrification is known to cause loss of combined nitrogen from agricultural soils, however methodological limitations have made realistic measurement of these losses under natural conditions very difficult. In this study a soil core method using the acetylene inhibition of nitrous oxide reduction was developed to measure field denitrification rates in agricultural soils. Recirculation of the pore space gases in a closed system allowed rapid distribution of acetylene to active denitrification sites and equilibrium of the N20 produced with the recirculating gaseous phase. Denitrification rates could be measured on a soil core within 2 hours which allowed many replicate samples to be evaluated and variability of the mean rate estimate described. Nitrogen losses were determined to be 11.6 kg-N-ha"1-mo‘1 for an aggregated clay loam soil and 6.1 kg--N°ha"1°mo"l for a largely unaggregated sandy loam soil. The response of soil denitrification to increased moisture was compared in the two soils. Increased denitrification rates were observed in both soils following water inputs of at least 1 cm following irrigation or rainfall. The denitrification rate in the sandy loam soil began to increase immediately after water addition and reached a maximum rate within 3 to 5 hours. Denitrification rates returned to preirrigation levels within 12 hours. A similar, but slower denitrification response occurred in the heavier textured clay loam soil; 8 to 12 hours elapsed before a maximum rate was observed, and 48 to 60 hours was required before the original rate was Alan John Sexstone restored. Peak field losses of 1.1 and 1.9 kg-N°ha"'1'day'1 occurred following water inputs of 7 and 4 cm, corresponding to air filled porosities of 37 and 30% in the sandy loam and clay loam soils, respectively. Nitrogen losses from the clay loam soil were double that of the sandy loam soil although the sandy loam soil received almost twice the water input, reflecting the difference in the temporal duration of the period of highest N-loss. In laboratory studies denitrification rates increased with increasing moisture content both in the presence and absence of oxygen.i The increased aerobic rate was attributed to increased anaerobic microsites due to decreased oxygen diffusion in the wetter cores. Increased anaerobic rates were attributed primarily to increased substrate availability, since no increase in denitrifying enzyme content could be detected. Cores collected after irrigation exhibited a greater response to rewetting than did dry cores wet to the same moisture content. Denitrification rates followed a hyperbolic relationship with pore space oxygen concentrations. This relationship depended on soil moisture content. Wetter cores exhibited a higher percentage of the potential anaerobic rate at a given oxygen concentration when compared with drier cores. A clay loam soil achieved 10% of the potential anaerobic rate at higher pore space oxygen concentrations when compared with a sandy loam soil at the same air filled porosity. Oxygen diffusion coefficients that provided the best fits to the experimental data were estimated from a model predicting soil anaerobiosis. An oxygen microelectrode was modified to measure oxygen concentrations in wet aggregates of a silt loam soil. The microelectrode had a sensing tip area of 3 pm, and oxygen measurements Alan John Sexstone could be made in as little as 0.1 mm increments to a depth of 12 mm. When aggregates were incubated in air, steep oxygen gradients usually occurred over very small distances from the aggregate surface. The smallest aggregate exhibiting a completely anaerobic center had a radius of 4 mm, although small aggregates (radius 5 6 mm) were generally oxic. Larger aggregates (radius 2.10 mm) often had measurable anaerobic centers, with the exception of those from a native prairie soil which exhibited irregular oxygen profiles, apparently due to oxygen intrusion caused by old root channels. Oxygen profiles obtained in 45 degree increments around an aggregate circumference were used to construct contour maps of oxygen concentrations within the aggregate. Oxygen gradients were often assymmetric, suggesting non-uniform oxygen consumption. An average oxygen diffusion coefficient of 8.5 x 10'“6 cm2°s’1 was determined for saturated aggregates. The aggregate anaerobic radii, calculated assuming radial diffusion, were similar to those measured directly with the electrode. Anaerobic centers were present in all aggregates that denitrified, but not all aggregates with anaerobic zones denitrified. The denitrification rate did not correlate with the size of the anaerobic zone, indicating that factors other than anaerobic volume alone determined the observed rates. To Julie and Sean ii ACKNOWLEDGEMENTS Dr. James Tiedje provided me with the opportunity to grow scientifically in an exciting and challanging environment. I thank him for all he has taught me, his continuing support, good advice, patience, and friendship. I have learned much from the many individuals who have also worked for Dr. Tiedje during this time. I acknowledge many long conversations with Joe Robinson, Dan Shelton, and Dave Myrold. I particularly thank Tim Parkin for getting me back out in the field, and for our many collaborations. A special thank you goes to Niels Peter Revsbech for the Opportunity to work with microelectrodes. The opportunity to interact with these individuals, and the other excellent students and post docs in Dr. Tiedje's lab have made the past five years an experience I will always remember. I thank the members of my guidance committee for their comments on and careful reading of my dissertation: Dr. Breznak, Dr. Klug, Dr. Shubert, and Dr. Ellis. Finally, I thank Julie for her love, for her willingness to stay while I completed something she knew was important to me, for her help in keeping this all in perspective, and of course for Sean. iii TABLE OF CONTENTS LIST OF TABLES O O O O O O O O O O 0 LIST OF FIGURES. . . . INTRO DUCT I ON 0 O O O O O O O O O O 0 CHAPTER I. CHAPTER II. CHAPTER III. APPENDIX A. APPENDIX 3. LITERATURE CITED . . . TEMPORAL RESPONSE OF SOIL DENITRIFICATION RATES TO RAINFALL AND IRRIGATION . . . . . . . . . . . MATERIALS AND METHODS. RESULTS. 0 O O O O O 0 DISCUSSION . . . . . . LITERATURE CITED . . . INTERACTIVE CONTROL OF BY OXYGEN AND MOISTURE MATERIALS AND METHODS. RESIILTS. C O O O O O 0 DISCUSSION . . . . . . SOIL DENITRIFICATION RATES LITERATURE CITED . . . . . . DIRECT MEASUREMENT OF OXYGEN DENITRIFICATION RATES IN MATERIALS AND METHODS. RESULTS. . . . . DISCUSSION . . . LITERATURE CITED PROFILES AND SOIL AGGREGATES THE EFFECT OF LONG TERM SOIL ACIDITY DENITRIFICATION RATES. MATERIALS AND METHODS. . . RESULTS AND DISCUSSION LITERATURE CITED . . . . . OXYGEN FLUX MEASURED WITH PLATINUM AS RELATED TO SOIL DENITRIFICATION LITERATURE CITED . . . iv ELECTRODES RATES . . . Page vi 11 16 18 29 46 50 52 54 S7 74 77 79 83 92 109 113 115 116 118 120 126 129 Table LIST OF TABLES CHAPTER I Page Denitrification rates and air filled porosity prior to and following irrigation. . . . . . . . . . . . . . . . 37 Comparison of cumulative nitrogen losses from two 30113eeeeeeeeeeeeeeeeeeeeeeeeeee44 CHAPTER II The effect of water addition on denitrification rates 1nac18y108n18011eeeeeeeeeeeeeeeeeeee58 Effect of water input on denitrification rates in a clay loam and sandy loam soil. . . . . . . . . . . . . . . 6O Respiration rates and estimated intra-aggregate diffusion coefficient for a clay loam and sandy loam 8011.00.00...0.0.0.00000000000073 CHAPTER III Comparison of denitrification rate, respiration rate, and anaerobiosis in aggregates from cultivated and prair1e80113eeeeeeeeeeeeeeeeeeeeeee108 APPENDIX B The effect of increasing soil moisture on oxygen flux and denitrification rates in a clay loam soil. . . . . . . 130 Figure LIST OF FIGURES CHAPTER I Relationship between air filled porosity and volume percent moisture for a clay loam (squares) and a sandy loam (triangles) soil. . . Gas flow system used to determine denitrification iDBOilcoreseeeeeeeeeeeeeeeeee Valve configuration used to inject gas samples onto the column of the gas chromatograph and to direct flow to the detector or vent. . . . . . . Mean denitrification rate of replicate soil cores packed with sieved clay loam soil. . . . . . Denitrification response of soil cores collected from the clay loam (squares) and sandy loam (triangles) soil irrigated in the laboratory withaZcmwaterinput............. Mean field denitrification rates of replicate soil cores collected from the clay loam (squares) and the sandy loam (triangles) soils at intervals following water inputs . . . . . . . . Mean daily denitrification rates from a clay loam soil receiving the indicated water inputs . Mean daily denitrification rates, including 952 confidence intervals of the mean, from a clay loam soil receiving the indicated rainfall . . . Mean daily denitrification rates from a sandy loam soil receiving the indicated water inputs . CHAPTER II Effect of oxygen additions on the anaerobic denitrification rate of a clay loam soil . . . . Aerobic denitrification rates in a clay loam soil at varying oxygen concentrations. . . . . . Percentage of the anaerobic rate at varying pore space oxygen concentrations in a sandy loam soil vi Page 20 23 26 31 33 35 39 41 43 64 66 68 Figure CHAPTER II Pore space oxygen concentration necessary to achieve 10% of the anaerobic rate for cores at varying air filled porosities. . . . . . . . . . . . Best fits of numerical simulations of Z anaerobic volume vs. pore space oxygen concentration (symbols) to experimentally determined relation- ship between X anaerobic rate and pore space oxygen concentration (solid lines) . . . . . . . . . CHAPTER III Diagram and dimensions of the oxygen microelectrode cathOde(A)andt1p(B)eeeeeeeeeeeeeee Flow through gas chamber for determining oxygen profiles in 8011 aggregates. e e e e e e e e e e e e Examples of oxygen profiles obtained in silt loam aggregates..................... Maps of oxygen concentrations within silt-loam aggregates..................... Oxygen diffusion in aggregates . . . . . . . . . . . Error function graph, constructed according to Duursma Values of C(x,t) obtained by numerically integrationg Fick's second law using different values Of Da. 0 O O O O O O O O O O O O O O O O O O O O O 0 APPENDIX A Linear correlation of denitrifying enzyme activity and soil pH from 204 samples of a sandy loam soil. . Influence of pH on denitrification enzyme activity in soil collected from areas of low and high pH. . . APPENDIX B Relationship between electrode current and applied effective voltage used to calculate current at EffeCtive VOltage 0f 0e5 e e e e e e e e e e e e e e Relationship between percent anaerobic rate, oxygen flux, and pore space oxygen concentration in a clay loam soil at 26 percent moisture. . . . . . . . vii Page . 102 . 104 106 . 123 . 125 . 131 . 134 INTRODUCTION Intensive nitrogen fertilization to maximize the yield of agronomic crops did not become prevelant in this country until after the second world war (Scarseth, 1942). Ten years later Allison (1955), reported the ”enigma” that nitrogen inputs could not be accounted for by measured nitrogen outputs from these systems. Based on this unaccounted for nitrogen, he suggested that between 10 to 302 of the nitrogen inputs were lost from soil due to denitrification. The following year, Hauck and Melstead (1956) discussed some of the methods available at that time to measure denitrification in the laboratory, and concluded that the rates measured under these artificial conditions were very high when compared with projected field losses. In the 30 years since these reports, progress has been made on the basic physiology and biochemistry of denitrifying bacteria, however, quantitative direct measurements of denitrification losses from soils have remained elusive. Lack of this information necessitates the currently practiced strategy of routine over-fertilization. This represents an economic loss to the farmer, and can cause increased nitrate pollution of groundwater (Keeney, 1982). Increased fertilizer use may also increase N20 emissions, which have been implicated in stratospheric ozone depletion, and gradual warming of the planet (McElroy et al., 1977; Wang et al., 1976). The early observations of Allison and Hauck pose a question that is central to this dissertation, i.e. how can denitrification rates be reliably measured to obtain good estimates of denitrification N-losses in the field, and what environmental factors control these losses? In this introduction I will not attempt to review all aspects of denitrification; indeed the many lines of current active research make this a prohibitive task. The interested reader is referred to several recent reviews that cover advances in this rapidly changing area of investigation: Payne, (1973); Delwiche and Bryan, (1976); Stouthamer, (1976); Focht and Verstraete, (1977); Rolston, (1981); Ryden, (1981); Payne, (1981); Knowles (1981,a,b; 1982); and Firestone(1982). I will briefly review literature pertaining to environmental factors observed to control denitrification rates in soil, as well as methodological approaches that have been used in attempts to measure field denitrification N-losses from soil systems. Denitrifying bacteria are facultatively anaerobic heterotrophs that in the absence of 02 can use N03" and N02" as alternate terminal electron acceptors, and stoichiometrically reduce these compounds to the gaseous products, N20 and N2. The latter is generally the dominant product from soil, although factors have been identified which increase the proportion of N20 produced (Firestone et al., 1980). When 02 is available these organisms preferentially live as aerobes. Oxygen is ‘known to both repress synthesis of denitrifying enzymes, and to inhibit the activity of these enzymes once synthesized, therefore denitrification must occur when the soil is anaerobic (Knowles, 1982). The macropores of soil, unless completely saturated with water are generally found to contain 02 at concentrations near those of the atmosphere (Parkin and Tiedje, 1983). Anaerobic zones in soil are thought to occur only as a result of active 02 consumption by bacteria, fungi, and plant roots. The rate of 02 consumption must exceed the rate of 02 supplied by diffusion for anaerobiosis to occur. The rate of 02 supply in turn depends on the thickness of the water film on soil particles, since oxygen diffuses at a rate 10,000 to 100,000 times slower through water than through air. The potential respiratory demand of soil heterotrOphs, as well as potential activity of soil denitrifiers, also depends on the availability of readily utilizable carbon, and in the latter case, the supply of NO3‘. Methods to measure natural denitrification rates must maintain the physical structure of the soil since this controls the extent of anaerobiosis and the substrate supply to potential active sites of denitrification. The presence of anaerobic microsites in soil is discussed further in Chapter III of this dissertation. Most previous studies of environmental factors controlling denitrification rates have been performed using long term incubations (weeks) with soils that often have been air dried and sieved such that all naturally occurring spatial arrangements within the soil are destroyed. Such artificial homogeneity make extrapolations to the field situation equivocal, but they have been useful in identifying important controlling parameters. It has been generally observed that increasing the soil moisture content, or decreasing the oxygen content of the incubation increases the denitrification rate observed. Various measures of the soil water and oxygen status have been used in an attempt to describe critical levels below which denitrification ceases to occur. Estimates of such critical moisture tensions include 33 kPa (Pilot and Patrick, 1972; Bremner and Shaw, 1958); 25 kPa (Ryden et al., 1979), and 10 kPa (Rolston et al., 1978). Other types of measurements that have been found useful include soil mositure content (Burford and Stefanson, 1973); air filled porosity (Pilot & Patrick, 1972); percent pore space oxygen (Wijler and Delwiche, 1954); percent water holding capacity (Nommik, 1956; Bremner and Shaw, 1958) and redox potential (Bailey and Beauchamp, 1973). As pointed out by Papendick and Campbell (1978), diffusion of a gas in soil is directly related to water content for all soil textural classes and not to the energy status of the water in the system; the latter is described by moisture tension. I have used water content or percent air filled porosity to describe the water status throughout this dissertation. Denitrification rates have been observed to increase with increasing nitrogen fertilization rates (Broadbent and Carlton, 1980). They have also been positively correlated with both ”readily available” and mineralizable carbon (Burford and Bremner, 1975); exogenous carbonaceous inputs such as manure (Rolston, 1978), or plant materials (Brar et al., 1978); and with association with the plant rhizosphere (Smith and Tiedje, 1979a). Soil nitrogen budgets provided the earliest estimates for and are still important indirect measures of field denitrification losses (Allison, 1966; Legg and Meisinger, 1982). Reported losses vary from O to 702 of the applied fertilizer with an average of 20 to 302. The general stratagy is to measure the nitrogen pools present in a soil and to assess inputs and exports over a period of time before constructing a balance. Nitrogen unaccounted for by difference is used as the denitrification estimate. Estimates of denitrification by this difference method reflect the cumulative errors of measuring all components of the balance. The transient nature of mineralization and immobilization reactions, difficulties with determining accurate leaching losses, and analytical difficulties in accurately measuring total N pools are a few of the problems encountered in constructing an accurate balance. The sensitivity of the balance approach can be increased by labelling a specific nitrogen pool with nitrogen enriched or depleted in 15N. Use of 15N for such studies has been reviewed by Hauck and Bremner (1976). As pointed out by Legg and Meisinger, use of 15N in budget studies is not equivalent to budgets where only unlabelled pools are measured. The former approach focuses on the total N-cyle of the system. The latter is a measure of how the label interacts with the system by tracing its fates through the various pools. It is possible to account for 15N fertilizer lost from a system, but it is difficult to access total denitrification by this method. Problems include heterogeneity of label distribution, the necessity of accurate time dependent measurement of 15N ratios of N pools, and the many assumptions necessary for the varying fates and recycling of N. It may be possible to obtain improved denitrification rates by this method, if analysis employing simultaneous estimation of major N-cycle rate processes is employed (Tiedje et al., 1981; Myrold and Tiedje, 1982). Denitrification losses estimated by difference contain much uncertainty, and require long term experiments to be accurate. Studies of this kind cannot elucidate short term temporal responses that determine when important nitrogen losses occur. It would be desirable therefore to directly measure the process. Substrate disappearance cannot be used in field situations, since nitrate has assimilatory and dissimilatory fates other than denitrification (Tiedje et al., 1981). Monitoring the appearance of the terminal product is complicated by the presence of an atmospheric background of 782 N2. One solution has been to follow 15N2 and 15N20 appearance from highly enriched 15NO3' by mass spectrometry. This approach was used by Rolston et al. (1976; 1978) who added large quatities of this isotOpe as fertilizer to replicated 1 m2 field plots, and monitored 15N flux by sampling gas accumulation under plexiglass covers placed on the soil surface for 1 to 3 hours. He observed increasing N-flux with increasing water content, soil temperature, fertilizer application rate, and manure addition. For plots near saturation with high manure additions, fluxes of 70 kg-'N-ha"'1°day'1 were reported, representing a 702 fertilizer loss over a 1 month period. The lowest loss observed at 230 c was 2.5 kg-N-ha’l-day"l from uncropped sites. These results contrast those of Mosier et al.(1982) who could detect no denitrification losses using this method from his sites. Problems with this technique include the high cost of highly enriched 15N and the high application rates necessary to detect sufficient quantities of N-gas for analysis by ratio mass spectrometry. This latter consideration makes this technique inappropriate in situations where the natural pool is normally low, as in most unfertilized sites. An alternative to ratio mass spectrometry is the use of gas chromatography coupled with quadripole mass spectrometry which has a much improved detection limit for 15N2 (Focht et al., 1980; Focht and Stolzy, 1978). Denitrified nitrogen is unavailable for assimilation, whether it remains in the soil macrOpores or has escaped to the atmosphere. A general criticism of cover techniques is that a flux as opposed to the actual rate of loss is measured. Flux measurements reflect only the rates of gaseous diffusion from the soil matrix to the atmosphere rather than actual denitrification rates. In wet soils, long lag periods can obscure short term temporal responses. As much as 70% of the gas produced can remain in the soil matrix under these conditions (Rice and Smith, 1982). A further underestimation occurs if downward diffusion is ignored. This later point has been considered in recent measurements by 'Verbruggen and Vlassak (1983), however measuring this component significantly increases the effort required to obtain a flux estimate, decreasing the number of replicate measurements that are possible. 4A alternative approach to addition of 15NO3- suggested recently is the use of a gas lysimeter to measure dilution by denitrification of added 15N2 (Limmer et al., 1982). This method is attractive since no disturbance of natural N pools is required. However the technique requires that the natural N2 background be reduced by flushing with an inert gas, and requires high production rates before sufficient dilutiOn occurs for detection. This gas lysymeter is a sophisticated apparatus, and does not easily lend itself to the replication necessary to obtain good field loss estimates. The radioactive isotOpe l3N provides for extremely sensitive N2 detection (Tiedje et al., 1979), however cannot practically be used in the field since the half life is 10 minutes and requires a cyclotron or Van de Graff generator to produce. However, this technique has attractive features for some laboratory studies, particularly those with natural samples with low pool N03” sizes. Promising initial rate measurements have recently been performed in this laboratory utilizing a direct injection technique (Jorgensen, 1978) with undisturbed cores of forest soils (G. P. Robertson, personal communication). Gas chromatography is another major technique used to measure rates of product formation from denitrification. N2 can be detected directly if background N2 is removed with an inert gas such as argon or helium, (Burford and Stevanson, 1973), however this approach is insensitive. The product N20 can be measured with much greater sensitivity (30 ppb detection limit) using 63N1 electron capture gas chromatography. The observation that low concentrations of acetylene (0.01 to 0.1 atm) inhibits the nitrous oxide reductase of denitrifying bacteria, causing stoichometric accumulation of N20 from N03" has greatly increased the applicability of this approach (Balderston at al., 1976; Yoshinari and Knowles, 1976). Acetylene inhibition of N20 reduction has been used in laboratory studies to measure denitrification in soil (Yoshinari et al., 1977; Klemedtsson et al., 1977; Smith et al., 1978), and the technique provides the basis of recent attempts to measure field denitrification rates. Use of this method requires that several cautions be observed: the blockage has been observed to fail in soils after long term incubations (160 hours) by Yeomans and Beauchamp (1978); sufficient concentrations of acetylene (>102) must be used in situations where the nitrate pool is less than 1 1.1g-N-g'l (Smith and Tiedje, 1979b); commercial acetylene contains contaminants such as acetone which, unless removed, can increase denitrification rates in incubations of longer than 3 to 5 days (Gross et al.,1982); and most importantly, acetylene has been observed to inhibit nitrifying bacteria and their activity in soil (Hynes and Knowles, 1980; Walter et al., 1979). These considerations necessitate that the acetylene inhibition technique be used in relatively short term incubations, and in situations were the indigenous nitrate pool is in sufficient supply. Applications of the acetylene inhibition technique in the field have used two main approaches; cover methods, and use with soil cores. With cover methods, acetylene is added the soil by diffusion and N20 is collected under a cover as it diffuses out. The most extensive studies reported to date are those of Ryden et al. (1979a,b) and Ryden and Lund (1980) who performed experiments on several irrigated California soils used for vegetable production. They allowed acetylene to diffuse into the sampling area surrounding injection tubes. A cover was then placed over the area and flux measurements determined by sweeping the headspace within the cover through a molecular sieve trap. N20 could be removed from the traps for analysis by gas chromatography by displacement with water. They observed rates between 7.9 and 19.5 kg-N'ha"'1'mo"l with fertilizer losses ranging from 14 to 522. Using a variation of this approach, Duxbury et al. (1982) observed losses of only 1.5 kg-N°ha"l over a 3 month period. The other major application of the acetylene inhibition technique involves removal of undisturbed cores from the field for incubation under controlled conditions in the presence of acetylene. Rice and Smith (1982) have described a method in which acetylene saturated water is added to the surface of soil cores prior to incubation to help facilitate acetylene distribution. Acetylene amended air is then passed over the tap of the core and the N20 concentration of the outflow is measured. With knowledge of a gas flow rate, a flux from the core surface can be calculated. This method has been used in a comparative study of soils subjected to minimum tillage techniques which exhibited greater N-losses than conventionally tilled soils. A second approach is to place the core sample within a second closed container into which acetylene is injected and allowed to passively diffuse into the sample. The denitrification rate is monitored by following N20 accumulation in the closed container over a period of 24 hours (Aulakh et al., 1982; Svensson et al., 1980). Using this method Aulakh et al. observed losses 10 of 0.5 to 1.5 kg-N-ha'1°mo"1 estimated from once weekly samples. Svensson et a1. measured high losses of 30 to 60 kg-N-ha‘1°mo’l from fertilized barley. These methods may suffer from a similar criticism to cover methods, i.e the rate measured is likely to reflect the physical rate of N20 diffusion rather than the biological rate of production. This is particularly true since the rate also reflects the speed and completeness with which acetylene can diffuse to active sites of denitrification. In this laboratory, we have deveIOped an acetylene soil core system that addresses these criticisms, and provides additional versatility for experimental evaluation of factors controlling the observed rate of denitrification (Parkin et al, 1983). Use of this method to determine denitrification N-losses from agricultural soils, as well as to describe short term temporal control of denitrification losses is described in Chapter 1. Chapter II further addresses the interactive control of denitrification rates by soil moisture content and oxygen concentration, and relates this control to an existing model which describes the extent of soil anaerobiosis. Finally, Chapter III reports direct measurement of anaerobic microsites in soil, and discusses these results with respect to observed denitrification rates. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. LITERATURE CITED Allison, F. E. 1955. The enigma of soil nitrogen balance sheets. Adv. Agron. 7:213-250. Allison, F. E. 1966. The fate of nitrogen applied to soil. Adv. Agron. 18:219-258. Aulakh, M. S., D. A. Rennie, and E. A. Paul. 1982. Gaseous nitrogen losses from crOpped and summer fallowed soils. Can. J. Soil Sci. 62:187-196. Bailey, L. D., and E. G. Beauchamp. 1973. Effect of moisture, added N03“ and macerated roots on N03” transformation and redox potential in surface and subsurface soils. Can. J. Soil. Sci. 53:219-230. Balderstron, W. L., B. Sherr, and W. J. Payne. 1976. Blockage by acetylene of nitrous oxide reduction in Pseudomonas perfectomarinus. Appl. Environ. Microbiol. 103:31-36. Brar, S. S., R. H. Miller, and T. J. Logan. 1978. Some factors affecting denitrification in soils irrigated with wastewater. J. Water. Pollut. Control Fed. 50:709-717. Bremner, J. M., and K. Shaw. 1958. Denitrification in soil. II. Factors affecting denitrification. J. Agric. Sci. 51:39-52. Broadbent, F. E., and A. B. Carlton. 1980. Methodology for field trials with Nitrogen-15 depleted Nitrogen. J. Environ. Qual. 9:236-242. Burford, J. R., and J. M. Bremner. 1975. Relationships between denitrification capacities of soils and total, water soluble, and readily decomposable soil organic matter. Soil Biol. Biochem. 7 : 389-3 94 e Burford, J.R., and R.C. Stefanson. 1973. Measurement of gaseous losses of nitrogen from soils. Soil Biol. Biochem. 5:133-141. Delwiche, C. D., and B. A. Bryan. 1976. Denitrification. Annu. Rev. Microbiol. 30:241-262. Duxbury, J. M., and P. K. McConnaughey. 1982. Direct measurement of denitrification in a corn field fertilized with urea or Ca(NO3)2. Agronomy Abstracts. p. 186. Firestone, M. K., R. B. Firestone, and J. M. Tiedje. 1980. Nitrous oxide from soil denitrification: Factors controlling its biological production. Science 208:749-751. Firestone, M. K. 1982. Biological denitrification. p. 289-326. In F. J. Stevenson (ed.) Nitrogen in agricultural soils. Agronomy 22: Am. Soc. Agron., Madison, Wis. 11 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 126. 27.. 12 Focht, D. D. and W. Verstraete. 1977. ”Biochemical ecology of nitrification and denitrification”. .Ig Advances in Microbial Ecology, v.1. M. Alexander (ed). Plenum Press. Focht, D. D., N. Valoras, and J. Letey. 1980. Use of interfaced gas chromatography-mass spectrometry for detection of concurrent mineralization and denitrification in soil. J. Environ. Qual. 93218-223e Focht, D. D., and L. H. Stolzy. 1978. Long-term denitrification studies in soils fertilized with (NH4)2804-N15. Soil Sci. SOCe Ame Je 42:894-898e Gross, P. J., J. M. Bremner, and A. M. Blackmer. 1982. A source of error in measurement of denitrification by the acetylene blockage method. Agronomy Abstracts. p. 188. Hauck, R. D., and S. W. Melstead. 1956. Some aspects of the problem of evaluating denitrification in soils. Soil Sci. Soc. AEe PrOCe 203361-3640 Hauck, R. D., and J. M. Bremner. 1976. Use of tracers for soil and fertilizer nitrogen research. Adv. Agron. 28:219-266. Hynes, R. R., and R. Knowles. 1978. Inhibition by acetylene of ammonia oxidation in Nitrosomonas eurOpaea. FEMS Microbiol. Latte 4:319-321e Jorgensen, B. B. 1978. A comparison of methods for the quantification of bacterial sulfate reduction in coastal marine sediments. I. Measurement with radiotracer techniques. Geomicrobiol. J. 1:11-27. Keeney, D. R. 1982. Nitrogen management for maximum efficiency and minimum pollution. p. 605-641. .ZE.F° J. Stevenson (ed.) Nitrogen in agricultural soils. Agronomy 22: Am. Soc. of Agron., Madison, Wisc. Klemedtsson, L., B. H. Svensson, T. Lindberg, and T. Rosswall. 1977. The use of acetylene inhibition of nitrous oxide reductase in quantifying denitrification in soils. Swed. J. Agric. Res. 7:179-185. Knowles, R. 1981a. Denitrification, p. 323-369. .IE.E° A. Paul and J. Ladd (ed.), Soil biochemistry, vol. 5. Marcel Dekker Inc., New Yorke Knowles, R. 1981b. Denitrification, p. 315-329. .EE.F° E. Clark and T. Rosswall (ed.), Terrestrial nitrogen cycles. Ecol. Bull. (Stockholm) no. 33. Swedish Natural Science Research Council, Stockholm. Knowles, R. 1982. Denitrification. Microbiological Reviews. 46:43-70e 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. A40. iii. 42.. 13 Legg, J. 0., and J. J. Meisinger. 1982. 8011 nitrogen budgets. p. 503-566. .In F. J. Stevenson (ed.) Nitrogen in agricultural soils. Agronomy 22: Am. Soc. of Agron., Madison, Wis. Limmer, A. W., K. W. Steele, and A. T. Wilson. 1982. Direct field measurement of N2 and N20 evolution from soil. J. Soil Sci. 33:499-508. McElroy, M. B., S. C. Wofsy and Y. L. Yung. 1977. The nitrogen cycle: pertubations due to man and their impact on atmospheric N20 and 03. Phil. Trans. R. Soc. 2778:159-181. Mosier, A. R. 1982. Gaseous nitrogen emissions from northeastern Colorado cropped and native soils. Agronomy Abstracts p. 193. Myrold, D. D., and J. M. Tiedje. 1982. Simultaneous estimation of several rates of N-cycle processes. Agronomy Abstracts. p. 194. Nommik, H. 1956. Investigations on denitrification in soil. Acta Agr. Scand. 6:195-228. Papendick, R. I. and G. S. Campbell. 1978. Theory and measurement of water potential. p. 1-022. .£2.J° F. Parr, W. R. Gardner, and L. F. Elliot (ed.) water potential relations in soil microbiology. S.S.S.A. Special Publication 9: Soil Science Society of America, Madison, Wis. Payne, W. J. 1973. Reduction of nitrogenous oxides by microorganisms. Bact. Rev. 37(4):409-452. Payne, W. J. 1981. Denitrification. John Wiley & Sons. New York. 214pe Parkin, T. B., and J. M. Tiedje. 1983. Application of a soil core method to investigate the effect of oxygten concentration on denitrification. Soil Biol. Biochem. Accepted. Parkin, T. B., Kaspar, H. K., Sexstone, A. J., and Tiedje, J. M. 1983. A gas-flow soil core method for measuring field denitrification rates. Soil Biol. Biochem. Accepted. Pilot, L., and W. H. Patrick, Jr. 1972. Nitrate reduction in soils: effect of soil moisture tension. Soil Sci. 114:312-316. Rice, C. W. and M. W. Smith. 1982. Denitrification in no-till and plowed soils. Soil Sci. Soc. Am. J. 46:1168-1173. Rolston, D. E., M. Fried, and D. A. Goldhamer. 1976. Denitrification measured directly from nitrogen and nitrous oxide gas fluxes. Soil Sci. Soc. Am. J. 40:259-266. Rolston, D. E., D. L. Hoffman, and D. W. Toy. 1978. Field measurement of denitrification. I. Flux of N2 and N20. Soil Sci. Soc. Am. J. 42:863-869. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. .54. 515. 14 Rolston, D. E. 1981. Nitrous oxide and nitrogen gas production in fertilizer loss. p., 127-149. In C. C. Delwiche (ed.) Denitrification, nitrification, and—ztmospheric nitrous oxide. John Wiley and Sons. New York. Ryden, J. C., and L. J. Lund. 1980. Nature and extent of directly measured denitrification losses from some irrigated vegetable crap production units. Soil Sci. Soc. Am. J. 44:505-511. Ryden, J. C., L. J. Lund, and D. D. Focht. 1979a. Direct measurement of denitrification loss from soils. I. Laboratory evaluation of acetylene loss from soils. I. Laboratory evaluation of acetylene inhibition of nitrous oxide reduction. Soil Sci. Soc. Alle Je 43:104-1100 Ryden, J. C., L. J. Lund, J. Letey, and D. D. Focht. 1979b. Direct measurement of denitrification loss from soils. II. Development and application of field methods. Soil Sci. Soc. Am. J. 43:110-118. Ryden, J. C. 1981. Gaseous nitrogen losses. p. 277-312. .In I. K. Iskander (ed.) Modeling wastewater renovation. John Wiley and Sons. New York. Scarseth, G. D. 1942. Purdue University Agricultural Experiment Station Bulletin 482. Smith, M. S., M. K. Firestone, and J. M. Tiedje. 1978. The acetylene inhibition method of short-term measurement of soil denitrification and its evaluation using nitrogen-13. Soil Sci. SOCe Me Jo 42:611-615e Smith, M. S. and J. M. Tiedje. 1979a. The effect of roots on soil denitrification. Soil Sci. Soc. Am. J. 43:951-955. Smith, M. S. and J. M. Tiedje. 1979b. Phases of denitrification following oxygen depletion in soil. Soil Biol. Biochem. 11:261-267. Stouthamer, A. H. 1976. Biochemistry and genetics of nitrate reductase in bacteria. Adv. Microb. Physiol. 14:315-375. Svensson, B. H., L. Klemedtsson, and T. Rosswall. 1980. Denitrification. p. 84. Ig_Ecology of Arable Land. Sveriges Lantbruksuniversitet. Uppsala. Tiedje, J. M., R. B. Firestone, M. R. Firestone, M. R. Betlach, and M. S. Smith, and W. H. Caskey. 1979. Methods for the production and use of nitrogen-13 in studies of denitrification. Soil Sci. Soc. Am. J. 43:709-719. Tiedje, J. M., J. Sorensen, and Y. Y. Chang. 1981. Assimilatory and dissimilatory nitrate reduction: Prospectives and methodology for simultaneous measurements of several nitrogen cycle processes. p. 331-342. .Ig_F. E. Clark and T. Rosswall (ed.) Ecological Bulletin 33. Swedish Natural Science Research Council, Stockholm, Sweden. 56. 57. 58. 59. 60. 61. 62. 15 Verbruggen, J. and K. Vlassak. 1983. Improvements of the acetylene inhibition technique for measuring denitrification in the field. Transactions of International Society of Soil Science. p. 138. Walter, H. M., D. R. Keeney, and I. R. Fillery. 1979. Inhibition of nitrification by acetylene. Soil Sci. Soc. Am. J. 43:195-196. Wang, We Ce, Ye Le Yung, Ae Le Lac18, Te M0 and Jo Ee Hansone 1976e Greenhouse effects due to man-made perturbations of trace gases. Science. 1974:685-689. Wijler, J., and C. C. Delwiche. 1954. Investigations on the denitrifying process in soil. Plant 8011 5:155-169. Yoshinari, T. and R. Knowles. 1976. Acetylene inhibition of nitrous oxide reduction by denitrifying bacteria. Biochem. Biophys. Res. Commun. 69:705-710. Yoshinari, T., R. Hynes and R. Knowles. 1977. Acetylene inhibition of nitrous oxide reduction and measurement of denitrification and nitrogen fixation in soil. Soil Biol. Biochem. 9:177-183. Yeomans, J. C., and E. G. Beauchamp. 1978. Limited inhibition of nitrous oxide reduction in soil in the presence of acetylene. Soil Biochem. 10:517-519. CHAPTER I TEMPORAL RESPONSE OF SOIL DENITRIFICATION RATES T0 RAINFALL AND IRRIGATION The rate of nitrogen loss from soil due to denitrification is thought to be increased by factors that increase the extent of anaerobic sites in the soil (Firestone, 1982).\:Increasing soil moisture acts as one such factor by decreasing the rate of oxygen diffusion through the soil matrix, allowing the development of anaerobic microsites as oxygen consumption exceeds the rate of its diffusive supply. Several laboratory studies have demonstrated increased nitrogen loss with increasing moisture content from soils incubated in the presence of an aerobic soil atmosphere (Nommik, 1956; Pilot and Patrick, 1972; Baily and Beauchamp, 1973).? These studies have generally involved the use of disturbed soils and long term incubations of several weeks. (Smith and Tiedje (1979) suggested that it was important to understand the short term denitrification response to increasing soil moisture in order to better characterize denitrification N-losses from field soils following rainfall or irrigation. In limited laboratory studies,ithey demonstrated that the denitrification rate increased rapidly following a lag period of 6 hours when the soil moisture content of a sandy loam soil was increased from 512 of saturation to 942 of saturationgi McGarity and Rajaratnam (1973) recognized that the denitrification rate in field soils would be controlled by episodic factors such as increasing soil moisture, and suggested that techniques were necessary for accurate short term measurements that preserved the physical control of biological rates imposed by soil structure. 0f the techniques 16 17 subsequently deve10ped for measurement of denitrification in the field, variations of the acetylene inhibition of N20 reduction technique (Yoshinari et al., 1977) appear to be the most suitable for these short term field measurements. Artificially irrigated soil cores have been used to compare denitrification as affected by tillage practice on several field sites (Rice and Smith, 1983). No-till.soils showed a significantly greater nitrogen flux than conventionally tilled soils, which was attributed to the higher soil moisture content maintained in no-tilled soils. Aulakh et al (1982) made weekly measurements of various field sites using a soil core technique and observed that the denitrification rate was greatly elevated when rainfall decreased the soil air filled porosity (APP). Ryden et al. (1979), and Ryden and Lund (1980) used a soil cover technique to measure denitrification and reported that the N20 flux was greatest following irrigation when the ZAFP decreased below 222. In this paper I report the use of a soil core method coupled with acetylene inhibition of N20 reduction (Parkin et al., 1983a,b) to follow the short term response of the denitrification rate following rainfall or irrigation. The temporal dynamics and duration of this response was examined in both light and heavy textured soils. MATERIALS AND METHODS Field Sites Field sites were established on the Michigan State University farms, East Lansing, Michigan. During May-July and in September, 1981 the experimental areas were located on a Capac clay loam soil (Aeric Ochraqualf) of 342 clay, 302 sand, 36% silt, pH 6.8 (Sites I and II). The field site during May-July, 1982 was a Spinks sandy loam (Psammentic Hapludalf) of 132 clay, 712 sand, 16% silt, pH 6.5 (Site III). The sites were fallow during the experimental periods, and had been planted alternately to corn and soybeans in previous years. Soybeans were harvested from the fall clay loam site just prior to experimentation. Plots were rectangular in shape, approximately 20 m x 10 m. Each site was roto-tilled prior to the start of the experiment. Site I and Site II were fertilized at day 0 with 4.4 kg-N-ha-l KNO3- and 5 kg-N°ha'1 KNO3", respectively. Site III was fertilized with 14.2 kg-N°ha"1 NH4NO3". The sites were chosen for comparison of denitrification rates in fine and coarse textured soils. During the 44 day experiment at Site I, rainfalls of greater than 1 cm were recorded on days 18, 19, and 27. The site also received a 2 cm irrigation on day 41. Site II was sampled for 19 days and received 1 cm or greater rainfall on days 7, 12, 15, and 17. Site III was sprinkler irrigated at 2 cm or greater on day 5, 15, and 28; and received a 3 cm rainfall on day 16. Water input at each site was determined with a rain gague. Soil cores 7.6 cm in diameter and 7.6 cm long were collected from both soils for physical characterization. Ten cores of each soil type 18 19 Fig. 1 Relationship between air filled porosity and volume percent moisture for a clay loam (squares) and a sandy loam (triangles) soil. 20 a e a as a :8 so unammsomamnuuiv Figure 1 MOISTIRE [INTENT (Willi PERCENT) 21 *were saturated in the laboratory and a moisture desorption characteristic determined over a range of 0 to 0.5 MPa. Dry bulk density averaged 1.3 Mg-nf3 for the clay loam and 1.7 Mg- ‘3 for the sandy loam. The relationship of air filled porosity (AFP) to volumetric moisture content for the two soils is shown in Fig. 1. Total porosity was 522 for the clay loam and 402 for the sandy loam soil. Sample Collection {The field sites were sampled by collecting intact core samples for denitrification rate determinations. Some additional disturbed samples were collected in plastic bags and were refrigerated in the laboratory for later use. The soil cores were collected with a steel corer of 30 cm length and 6.8 cm diameter. A plexiglass or polyvinyl chloride (PVC) core liner (24 cm long) was placed inside the steel corer and held in place with a massive screw cap. The inner diameter of the metal cutting tip and of the core liner was 4.7 cm. The corer was then driven into the ground with a slide hammer. Compaction was usually minimal (O-SZ). Between 4 to 36 cores were collected at Site I nearly every day during the course of the experiment. During the experiments at Site II and III, the sampling regime was altered such that during relatively dry periods cores were collected at 3 to 4 day intervals. After rainfall or irrigation, sampling was increased to intervals of 6 to 12 hours. At each sampling during these experiments, between 12 to 56 cores were collected. After sampling , rubber stoppers were placed in the ends of the core liners which were returned immediately to the laboratory. Determination of Field Denitrification Rates Denitrification rates were determined by measuring N20 accumulation using a variation of the acetylene inhibition of nitrous oxide reduction technique (Yoshinari et al., 1977). This method is described in detail 22 Fig. 2 Gas flow system used to determine denitrification in soil cores. A, moisturizing flask; B, on-off valve; C, quick-fit connector; D, valving system in gas chromatograph; E, quick-fit connector. 24 by Parkin et al. (1983a). Acetylene enriched air was recirculated through the pore spaces of the soil cores in a closed system described in Fig. 2. Prior to rate measurements, cores were attached to the gas recirculation system and flushed with moistened air for 10 to 15 minutes. This was achieved using a manifold which divided the air stream into four portions, each leading to a soil core. With the on-off valve at point B open, and the quick connect at point C broken, the entire recirculation system could be flushed. Closing valve B, and reconnecting the gas line at C, isolated each core as a closed system. Acetylene was added to each core and the gas recirculated through the soil core and sample 100ps (D) on a gas chromatograph by means of a membrane pump. Typical flow rates were 300 ml°min"'1. samples from two soil cores could be injected alternately into a 63Ni electron capture detector for N20 determination using the gas sampling valving descibed in Fig. 3. Two gas chromatographs were used; a Perkin Elmer 910 and a Varian 3700, equipped with a total of four detectors allowing denitrification rate determination on eight cores simultaneously. Details of gas chromatograph operation and gas separation have been reported (Kasper and Tiedje, 1980). When one column leading to a detector was in the injection mode, the other column was backflushed at twice the carrier gas flow rate to remove acetylene, freon and water which have retention times longer than N20. Samples were injected in 5 min intervals, hence the sampling interval for an individual core was 10 min. N20 accumulation in the recirculating gas was measured in the presence of 20% (v/v) acetylene. This relatively high acetylene concentration was used to insure that blockage of further N20 reduction was complete. Linear rates of N20 production could be determined within 25 Fig. 3 Valve configuration used to inject gas samples onto the column of the gas chromatograph and to direct flow to the detector or vent. 26 CARRIER T° .22. Far SOIL SOIL CORE CORE ‘ TO CORE CORE DETECTOR SAMPLE SAMPLE LOOP LOOP ,VENT BACKFLUSH LJ . GAS COLUMN 1 COLUMN 2 B-PORT VALVE 1 B-PORT VALVE s-PORT VALVE 2 Figure 3 27 1 to 2 hours. Rates are reported on a per gram soil basis. Addition of acetylene to the soil cores resulted in a dilution of the 02 content of the recirculating gas from 212 to 182. The 02 concentration hithe recirculation gas could be adjusted to match that measured at the field sites with an in _§_i_t_u_ gas sampling technique (Parkin and Tiedje, 1983). For the cases reported here, soils always exhibited greater than 142 02 in their macropores. Mean daily denitrification rates were determined from the individual rates of soil cores collected at each sample time. We have determined that denitrification rates at these field sites follow a log normal distribution (Parkin et al., 1983). Statistics apprOpriate to this distribution were used to calculate mean daily rates and 95% confidence intervals from formulas according to Blais and Carlier (1968). Use of these expressions results in asymmetric confidence intervals. Total nitrogen losses from the entire sampling period were calculated by integrating the mean daily losses over the time course of each field experiment. Laboratory Denitrification Measurements Freshly collected field cores from Site I and Site III were used in laboratory incubations to determine the temporal response of denitrification following water input. For these experiments 2 cm of distilled water was added to the soil surface prior to incubation on the gas recirculation system in the presence of 187. 02 and 202 acetylene. Cores which received no water additions served as controls. Concentrations Of Nzo-N in the recirculating gas phase were determined at 30 to 60 min intervals for periods Of up to 24 hours. A second experiment was performed to monitor denitrification losses from a uniformly fertilized and wetted soil. Soil from Site I was 28 coarsely sieved (0.5 cm mesh), and adjusted to 222 moisture with a solution containing KNO3‘ to a final nitrate concentration of 30 ug:g'1. PVC cylinders were packed with 500 g of this soil to a bulk density of 1.5 Mg-m'3. The cylinders were covered with plastic wrap and incubated in the dark at 25° C. Denitrification rates were determined on 4 to 8 cores at each of the following times: 0, 0.5, l, 3, 5, 8, 12, 15, 19, and 22 days. RESULTS Sieved cores, which had uniformly distributed water and N03“, exhibited initially high denitrification rates which reached a maximum of 300 ng-N-g"l °day"1 at 24 hours (Fig. 4). Following this period the rates declined rapidly over a 4 day period. In laboratory incubations unamended soil cores Obtained from sandy loam and clay loam soil exhibited a constant rate Of denitrification in the presence of a recirculating atmosphere of 182 oxygen for up to 20 hours (Fig. 5, dark squares). When water equivalent to a 2 cm rainfall was added to these soils at zero time, the rate of denitrification increased according to a temporal pattern characteristic of that soil. The denitrification rate was initially constant for 1 to 3 hours and 8 to 12 hours in the sandy loam and clay loam soils, respectively, followed by a period of increasing N20 production. A second linear increase in N20, usually at least 10 times higher than the prewetted rate occurred within 5 hours in the sandy loam and within 14 hours in the clay loam soil. Field denitrification rates were determined using short term (1 to 2 hour) incubations of replicate cores removed from the field at various times following soil wetting by irrigation or rainfall (Fig. 6). A similar temporal pattern to the laboratory incubations was observed. Denitrification rates generally returned to prewetting levels within 12 hours at the site with sandy loam soils and within 60 hours at the site with clay loam soils. In general a increase in the denitrification rate could be Observed in both soils following at least a 1 cm water input. Shown in Table 1 are the mean denitrification rate and 952 29 30 Fig. 4 Mean denitrification rate and 952 CI of replicate soil cores packed with sieved clay loam soil which had been adjusted to 22 percent gravimetric moisture content at day 0. 31 12 10 6 DAYS Figure 4 § § E § § § § ° 1-111011) N‘ON Fig. 5 32 Denitrification response of soil cores collected from the clay loam (squares) and sandy loam (triangles) soils irrigated in the laboratory with 2 cm of water at zero time. Control cores , as illustrated by the clay loam soil (dark squares) , received no water addition. 33 Figure 5 34 Fig. 6 Mean field denitrification rates of replicate soil cores collected from the clay loam (squares) and the sandy loam (triangles) soils at intervals following water inputs. 35 o munwfim mg: e e e a e e a o... a a 4‘8 fl .— u a gm arm. 8m 1... .smN E 36 confidence intervals prior to and following a 7 cm and 2 cm water input to the sandy loam and clay loam soils, respectively. The mean denitrification rate increased by a factor of ten to 209 and 383 ng-N g'l day-1, respectively, following soil wetting when compared with prewetted levels. The increased denitrification rate following rainfall shown in Table 1 corresponded to air filled porosities of 37 to 382. The highest denitrification rate observed at Site III was 300 ng-N 3‘1 day'"1 and occurred following a 4 cm rainfall. During the initial 27 days of the first field trial (Fig. 7) between 4 to 12 cores were taken almost daily. The sampling schedule was then revised to better, measure the transient temporal denitrifcation response following rainfall and to obtain increased numbers of samples to obtain better estimates of the mean rateu Following the new schedule 12 to 56 cores were collected every 3 to 4 days during dry periods, with sampling increased to every 6 to 12 hours immediately following rainfall or irrigation. This approach is illustrated (Fig. 8) with data obtained at Site II. The mean rate declined slowly following fertilization as the soil dried then increased rapidly following a 4 cm rainfall at day 7. The denitrification rate was only slightly stimulated after a second heavy rainfall (8.6 cm) at day 13, and was not increased at all by two subsequent 2 cm rainfalls. The lack of response was apparently not due to depleted soil N03" since an average of 5 ppm was present in the tap 30 cm of this soils on day 19. A similar sampling schedule was also followed at the sandy loam site (Site III). The four irrigation inputs were all greater than 2 cm and caused increased denitrification rates in all cases (Fig. 9). The mean cumulative N-loss obtained by integrating the area under the daily Nhloss curves shown in Figs. 7 and 9 is summarized in Table 2. 37 Table 1. Denitrification rates and air filled porosity prior to and following irrigation.+ Denitrification rate Time after Air filled Gravimetric Soil irrigation porosity moisture Mean 951 0.1. -(hr>- -- - (dC/dx) where P is the volume percent solids (0.7) and Da the average diffusion coefficient measured for the aggregates. An average oxygen concentration gradient, dC/dx, was measured directly with. the electrode from 8 to 10 concentrations measured at random at a 1 mm depth on the aggregate surface. RESULTS The oxygen microelectrode could be inserted into the soil aggregates in as little as 0.1 mm increments by a micromanipulator. Typical oxygen concentration profiles obtained in this way are shown in Fig. 3. Curve A shows the steep oxygen gradient that can occur over a small distance. Complete anaerobiosis occured after only 3.5 mm in this aggregate, which had a radius of 7 mm. Curve B also revealed anaerobiosis in another aggregate” however, the oxygen gradient was more gradual and less regular than the previous example. Curve C was obtained on an aggregate slightly larger than the ones for A or B, but a minimum oxygen concentration of 0.03 atm was measured at the aggregate center. The oxygen concentration increased as the electrode passed beyond the center. Curve D was obtained on.a native prairie aggregate and shows surface oxygen intrusion up to the 5 mm depth within the aggregate, which can likely be attributed to root channels observed within these aggregates. Oxygen profiles were obtained on.a total of 57 aggregates whose radii ranged from 3 to 12 mm. Of these, 12 aggregates exhibited at least one profile showing complete anaerobiosis. The smallest aggregate observed with an anaerobic center had a radius of 4 mm. Aggregates with radii greater than 10 mm generally exhibited at least one profile showing a zone of no oxygen, however, it was not uncommon for larger aggregates, particularly from undisturbed prairie, to have measurable oxygen concentrations throughout. 92 93 Figure 3. Examples of oxygen profiles obtained in silt loam aggregates. Aggregate radius (mm) was 7,8,9,9 for aggregates A,B,C,D, respectively. A-C were aggregates from a cultivated field; D was from an uncultivated native prairie. 94 m ousmfim 2.. e 25 was: as sass e ’4 HI- E ’4 0'3 m w .2 a n l. u u o u E m < a a .m . . .W 2 5 em. 4 4 5 b“ 4. r4 .2 ’4 . / 95 2 womafiucoo m muswfim A a a see; #552 as seas e a e N 9 L1 ‘) V, . o A“ n" A" n" “0 nu 0. .m . I) o .e b 0 n o a D, U II; , 9 0 r s o .2 W b I) G b/ nv Aw“ 9 bx >31 .mfi. y, 96 Profiles around a.circumference of selected aggregates were used to construct a contour map of intra-aggregate oxygen concentrations (Fig. 4). Aggregate A exhibited a small anaerobic zone which was skewed from the aggregate center. Aggregate B showed regular and steep oxygen gradients around the entire aggregate circumference which resulted in a large and regular anaerobic center. Prior to these measurements, this aggregate had been rolled to give the aggregate a more spherical shape. In {all aggregates treated in.this manner, the observed oxygen gradient increased after rolling although not all rolled aggregates exhibited anaerobic centers. Aggregate C exhibited oxygen concentrations as low as 0.01 atm but no strictly anaerobic sites were observed. Oxygen concentrations, determined at several depths within an aggregate 200 sec after the surrounding atmosphere was rapidly changed from water saturated air to 1002 oxygen, were used to determine the intra-aggregate diffusion coefficient (Fig. 5). 'These values were used to construct an error function plot (Fig. 6) from which a value of Da could be determined, in this case 1.1 x 10"5 cm2 8‘1. The closed symbols are from the experiment where the imposed gradient was 1002 N2; the same value for the diffusion coefficient was obtained. The mean oxygen diffusion coefficient determined from four aggregates was 8.5 x 10"6 cm2 8'1. In addition to oxygen profiles, the denitrification and respiration rates were determined for several aggregates (Table 1). Those aggregates with a measurable denitrification rate 97 Figure 4. Maps of oxygen concentrations within silt-loam aggregates. A and B were aggregates from a cultivated field; C was from an uncultivated prairie. c111 98 (D-j m— 3 l 0 3 distancdmm) Figure 4 Q— (0-1 99 CA4 6 5 0 distancdmm) 9'1 (0-1 Figure 4 continued 100 '- PA 1 g 21 7 15 *- 10 1 0 CD I I r 1 I 1 1 9 6 s 0 a 6 9 distancdmm) Figure 4 continued 101 Figure 5. Oxygen diffusion in aggregates. Shown are the recorder output of 02 concentration at 0.4, 0.8, 1.2 mm below the surface of the aggregate. The point of change in the atmosphere surrounding the aggregate from water saturated air to 1002 02 is indicated by the dotted line. 102 1 min 0.8 mm . . 1b 1.2 mm 1 3 2 1 7.02233 otocemoEum .3 NO Figure 5 Figure 6. 103 Error function graph, constructed according to Duursma and Hoede (1967), used to estimate the intra-aggregate 02 diffusion coefficient from the relationship: where C x,t) is the oxygen concentration at depth x(cm) and time t(200 sec) after rapidly changing the oxygen concentration to 1002 02 (O) or 1002 N2 (0) at the aggregate surface. The radiating lines are theoretical C(x t)/Co values for different values of the diffusion coefficient (Da) and are used to estimate the measured value of Da by interpolation. 104 0.5 .... 0.05 _ 0.02 - 0.01 - LllLl 1 1%31 4'1 1 \ fi-A\\\ 053 5 10 15 20 25 3035 40 Figure 6 00‘ Cm 105 always had a demonstrable anaerobic zone, however, some aggregates with anaerobic centers showed no denitrification. Aggregates obtained from a cultivated field (CA) had higher respiration than did aggregates obtained from native prairie (PA). The oxygen consumption rates calculated from the measured oxygen fluxes were generally less than those determined by C02 production, but agreed well with values reported for other soils at similar carbon contents (Greenwood, 1975). These values were used to calculate the anaerobic radius predicted by expressions derived by Smith (1980). The predicted anaerobic radius was similar to the average radius for the anaerobic zone measured by the oxygen microelectrode, and where different, the model tended to underestimate the radius. The magnitude of the denitrification rate did not correspond to aggregate size or to the anaerobic radius, indicating that factors other than anaerobic volume alone were determining the observed denitrification rates. Figure 7. 106 Values of C(x,t) obtained by numerically integrating Fick's second law using different values of Da. Solid line represents the oxygen concentrations experimen- tally measured as described in Figure 5. The symbols correspond to different values of Da (x 10’6cm2°g‘1) used in the numerical integration: were 11, 7, 5.9, and 3. 107 0 10 00120100 500 $12003 Figure 7 108 voawsumuov you 1 .e.a usooo vasonm madcap agnouommm on umcu mmumowvaw nowumasoamu 1 Awuma vuum>wuasuaa unmomfivm Scum no AHuHso m scum voawmuno u oumwmumwm some so mmafimoue o~ ou e no enhances weapon mama % o.m o.<~ o.m~ ~.~ ~.o n.~ Nauaso Bonn moumwouwwm cw mHmOfinouomam use .mumu aowumuwemou .mumu moaumowuwuufiaov mo nomaumeaoo .meom mwuwmum was .fi wanna DISCUSSION The oxygen sensing tip of the microelectrodes used in this study is much smaller than in any oxygen electrodes previously used in soil. The microelectrodes were made more sturdy for use in soil by increasing the wall thickness of the of the outer glass capillary. This could be achieved without effecting the inner diameter of the membrane or the size of the inner cathode. Although the electrode tips are flexible, they are still quite fragile because of their small size. In this initial study we chose to work with a soil containing little sand, since individual sand particles are much larger than the microelectrode and may damage the sensing tip. We worked only with wet aggregates to facilitate insertion of microelectrodes. Successful oxygen measurements were made in aggregates at moisture contents as low as 282. In future work it will be important to determine whether anaerobic sites exist in drier aggregates, since we know that denitrification, can occur in soils at lower moisture contents than were used in this study. The measurement of oxygen under drier soil conditions will probably be limited to one profile in each aggregate, as the air channel created when the electrode is withdrawn will not refill with water as it did in our wet aggregates. Greenwood & Goodman (1967) have provided the only other direct measurements of oxygen profiles in aggregates; they used 1 cm diameter, KCl saturated spheres. They concluded that their 109 110 electrode tip caused compaction when inserted into the aggregates causing steeper oxygen profiles. The tip of their electrode was 0.5 mm, 100 times larger area than the electrode used in this study. They measured complete anaerobiosis within 2 mm of the surface of their aggregate constructs. Their result is consistent with the observation that oxygen profiles in rolled or disturbed aggregates are steeper. The profiles shown in Fig. 3 illustrate the necessity of measuring oxygen concentrations with high spatial resolution within soil aggregates. Steep oxygen gradients occurred in most aggregates over very small distances from the aggregate surface. Greenwood predicted that the smallest aggregate containing an anaerobic center would have a radius of 9 mm, assuming a D8 of 1 x 10'5cm2°s'l and a respiration rate of 4.31 x 10’7ml°m1'1°s"l (Greenwood, 1975). We have measured anaerobic zones in aggregates with radii as small as 4 mm. The respiration rates determined for the aggregates in our study, were similar to those assumed by Greenwood, however, we measured a lower average oxygen diffusion coefficient of 8.5 x 10‘6cm2-s‘1. Our mean diffusion coefficient agrees well with the range of values predicted for water saturated aggregates (Smith, 1980). We used an integrated form of Fick's second law and the derivative form of Fick's first law describing one dimensional diffusion to obtain estimates of the intra-aggregate diffusion coefficient and the total aggregate oxygen flux, respectively. Although the aggregates were spherical, use of these simplified calculations was apprOpriate, since the radii of the aggregates 111 (Z 8 mm) were large relative to the very small sensing tip of an electrode (1 to 3 pm) and the shallow depth of electrode insertion (0.35 to 1.2 mm). On this scale, oxygen would arrive at the inserted electrode tip, diffusing from an apparently planar surface boundary. Previous workers have successfully described steady state oxygen profiles in water saturated aggregates using one dimensional diffusion (Greenwood and Goodman, 1967), but they also included an oxygen consumption term in their formulation. Our estimates of Da did not account for oxygen consumption in the aggregate since we assumed to be insignificant during the short duration (200 sec) of the experiment. This ommission could cause a slight underestimation of Da. the results of numerically integrating Fick's second law over a range of Da values to provide estimates of oxygen concentration with time at a fixed depth in an aggregate are shown in Fig. 7. Empirically comparing these results with oxygen concentration vs time of an actual experiment show that our experimental estimate of Da appears to provide a good fit with the data, and thus give us confidence that our assumptions were reasonable. The contour maps (Fig. 4) provide a two dimensional visualization of the oxygen profiles and anaerobic centers that occur in soil aggregates. Anaerobic zones were often assymmetric and did not always occur at the aggregate center. This suggests that sites of oxygen consumption within the aggregate are not uniformly distributed, presumabLy due to a non-uniform distribution of organic carbon within the aggregates. Because of 112 this anisotropy, a single oxygen profile or a two dimensional composite of oxygen profiles will not necessarily give a complete picture of aggregate anaerobiosis. As an example, the aggregate pictured in Fig. 2 B has no measurable zone of complete anaerobiosis in the two dimensional slice pictured, however, profiles obtained from at 90 degrees to the composite plane did contain an anaerobic zone, explaining the low but measurable denitrification rate obtained from this aggregate. A detailed description of anaerobiosis within a single aggregate will require many profiles obtained in all three dimensions. Presence of an anaerobic profile was necessary before a denitrification rate was measured on any aggregate, however, some aggregates with anaerobic centers exhibited no denitrification. An anaerobic region of soil alone is not sufficient for denitrification. Soil denitrifiers also need a source of carbon as well as the presence of sufficient nitrate. Furthermore they may not be present at all anaerobic sites. Current mathematical models rely on the measurement of the soil anaerobic volume to subsequently predict the denitrification rate (Leffelaar, 1979; Smith, 1980). ‘Measured denitrification rates in this study did not necessarily correlate with aggregate size or anaerobic radius, however, we have not yet made both measurements on a sufficient tunmber of aggregates to evaluate a reliable correlation. It seems likely, however, that a spatial description of carbon and NO3" distribution will also be necessary before denitrification within one aggregate can be sufficiently predicted. l. 8. 10. ll. 12. LITERATURE CITED Clark, L. C. Jr., R. Wolf, D. Granger, and Z. Taylor. 1953, Continuous recordining of blood oxygen tensions by polarography. J. Appl. Physiol. 6:89-193. Crank, J. 1956. ”The Mathematics of Diffusion”. Oxford Univ. Press, London and New York. Currie, J. A. 1961. Gaseous diffusion in porous media. III. Wet granular materials. British J. Appl. Phys. 123275-281 o Duursma, E. R., and C. Hoede. 1967. Theoretical, experimental, and field studies concerning molecular diffusion of radioisotOpes in sediments and suspended solid particles of the sea. Netherlands J. of Sea Research. 3:423-457. Enoch, B., and V. Falkenflug. 1968. An improved membrane system for oxygen probes. Soil Sci. Soc. Am. Proc. 32:445-446. Firestone, M. K. 1982. Biological denitrification. p. 289-326. $2.F' J. Stevenson (ed.) Nitrogen in agricultural soils. Agronomy 22. Am. Soc. of Agron., Madison, Wisc. Fluhler, H., M. S. Ardakan, T. E. Szuszkiewicz, and L. H. Stoley. 1976. Field measured nitrous oxide concentrations, redox potentials, oxygen diffusion rtes, and oxygen partial pressures in relation to denitrification. Soil Sci. 122:107-114. Greenwood, D. J. 1961. The effect of oxygen concentration on the decomposition of organic materials in soil. Plant Soil. 14:360-376. Greenwood, D. J., and D. Goodman. 1967. Direct measurement of the distribution of oxygen in soil aggregates and in columns of fine soil crumbs. J. Soil. Sci. 18:182-196. Greenwood, D. J. 1975. Measurement of soil aeration. p. 261-272. ‘13 soil physical conditions and crap production. M. N. Agric. Fish., Food, Tech., Bull. no 29, H.M.S.O. London. Harris, R. F., G. Chester, and O. N. Allen. 1966. Dynamics of soil aggregation. Adv. Agron. 18:107-160. Leffelaar, P. A. 1979. Simulation of partial anaerobiosis in a model soil in respect to denitrification. Soil Science. 113 13. 14. 15. 16. 17. l8. 19. 20. 21. 22. 23. 114 128:110-120. Lemon, E. R. and Erickson, A. E. 1952. The measurement of oxygen diffusion in soil with a platinum microelectrode. Soil Sci. Soc. Am. Proc. 16:160-163. McGill, W. B., H. W. Hunt, R. W. Woodmansee, and J. O. Reuss. 1981. Phoenix - A model of the dynamics of carbon and nitrogen in grassland soils. p. 49-115. In: Terrestrial nitrogen cycles: processes, ecosystem strategies, and management impacts. Ecol. Bull. 33. Stockholm, Sweden. McIntyre, D. S. 1970. The platinum microelectrode method for soil aeration measurement. Adv. Agron. 22:235-283. Naylor, P.F.D., and Evans, N.T.S. 1960. An electrode for measuring absolute oxygen tension in tissues. J. poldrogr. Soc. 2:22-24. Revsbech, N. P., J. Sorensen, T. H. Blackburn, and J. P. Comholt. 1980. Distribution of oxygen in marine sediments measured with microelectrodes. Limnol. and Oceanogr. 25:403-411. Revsbech, N. P., and D. M. Ward. 1983. Oxygen microelectrode that is insensitive to medium chemical composition: Use in an acid microbial mat dominated by Cyanidiumcaldarium. Appl. and Environ. Microbiol. 45:755-759. Sampson, R. J. 1978. Surface II Graphics System. ‘lg John C. Davis (ed.) Series on spatial analysis, number one. Kansas Geological Survey. Lawrence, Kansas. Sexstone, A. J., T. P. Parkin, J. M. Tiedje. 1983. Temporal response of soil denitrification to rainfall and irrigation. 8011 8C1. SOC. Am. J. (Bmeitted)e Smith, R. A. 1980. A model of the extent of anaerobic zones in aggregated soils, and its potential application to estimates of denitrification. J. Soil. Sci. 31:263-277. Wiley, C. R., and C. B. Tanner. 1963. Membrane covered electrode for measurement of oxygen concentration in soil. Soil Sci. Soc. Am. Proc. 27:511-515. Yoshinari, T., R. Hynes, and R. Knowles. 1977. Acetylene inhibition of nitrous oxide reduction and measurement of denitrification and nitrogen fixation in soil. Soil Biol. Biochem. 9:177-183. APPENDIX A THE EFFECT OF LONG TERM SOIL ACIDITY ON SOIL DENITRIFICATION RATES Decreasing pH is generally observed to decrease denitrification rates, both of known denitrifying bacteria, and in soil. Observed pH optima generally occur between pH 7.0 and 8.0 (Bollag et al., 1970; Valera and Alexander, 1961; Nommik, 1956). Klemedtsson et al. (1978) observed that nutrient additions to an acid peat soil had no effect on denitrification rates, however increasing the pH from 3.5 to 6.5 greatly stimulated denitrification. Other workers have measured significant denitrification rates in acidic soils (Van Cleemput and Patrick, 1974; Gilliam and Gambrell, 1978; Muller et al., 1980). It is not clear whether the denitrification rate measured in very acidic soils is the result Of small pOpulations of denitrifiers in protected microsites of neutral pH, of general pOpulations of denitrifiers functioning poorly in low pH environments, or of populations of denitrifiers with low pH optima (Firestone, 1982). Measurement of biological denitrification in soil under acidic conditions is complicated by chemical instability of N02” and the potential reactions of this compound with ammonia (Broadbent and Clark, 1965); amino acids and complex organic species (Soulides and Clark, 1958; Stevenson et al., 1970), and Fe2+ and Cu2+ ions in soil (Buresh and Moragham, 1976; Moragham and Buresh, 1977; Wullstein and Gilmour, 1966). These reactions are summarized by Knowles (1981). The gaseous products that have been reported from these reactions include NO, N20, N2, and N02. In this study we report the effect of long term soil acidity on the pH Option of denitrification activity in these soils eis reported. 115 MATERIALS AND METHODS The site sampled for this study was a‘ 15 m x 30 111 area of Spinks sandy loam soil located on the Michigan State University farms, East Lansing, Michigan. This site was established in 1959, to assess the effect of various nitrogen carriers on soil acidification (Wolcott, 1965). Fertilizer treatments were continued until 1977 and resulted in subplots Of p11 as low as 3.7. In 1965, one-half of the study site was limed, restoring the soil pH to rear pretreatment levels. The site was sampled during August 1982, as part of a larger study to evaluate spatial variablity of soil. denitrification rates (Parkin et al., 1983). Soil cores (4.5 cm diameter x 25 cm long) were collected at 60 cm intervals along four transects, which ran between areas of low and high pH. Additional bulk soil was also collected and refrigerated at 4°C for later analysis. Each core was coarsely sieved (0.5 cm) and used to prepare a soil slurry. Twenty-five grams of soil and 50 ml of distilled H20 was added to a 250 ml Erlenmeyer flask. Each flask was attached to the previously described gas recirculation system and flushed with argon for 15 to 30 minutes to achieve anaerobiosis. Acetylene was added (202 v/v) and the slurry stirred while a denitrification rate was determined over a period of 1 to 2 hours. The denitrification rate was measured by following increasing concentrations of N20 by gas chromatography. This assay is similar to that described by Smith and Tiedje (1979), and can be used as a measure of the activity of existing denitrifying enzymes in soil. Soil pH was determined in 2:1 slurries with 1N KCl, and N03- concentrations were determined in IN KCl extracts on a Technicon auto 116 117 analyzer. .A similar assay was performed using the bulk soil samples in which the pH Of the soil had been adjusted. Soil pH was increased variously by adding saturated CaCO3", 1 N NaOH or 25 mM MOPS buffer; and decreased using additions of 12 HCl or 112804 prior to the denitrification assay. Flasks were autoclaved and 1 mM N03” added prior to rate determinations. A second set of controls was run using 3 gm of low and high pH soil and 3 ml H20 in anaerobic culture tubes. Soils were autoclaved twice prior to gassing with argon and addition of acetylene. Replicates Of non-sterile soils were also run. NaNOz'Dwas added to a final concentration of 1, 0.1, 0.01, 0.001 mM. Headspace gases were sampled at 8 and 24 hours. RESULTS AND DISCUSSION Denitrifying enzyme activity for the site ranged from 100 to 20,000 ng-N g'l-day’l. Areas of low pH (< pH 5) averaged 1800 ng-NongI-day‘l, while areas of higher pH averaged 8500 ng-N -gm71'day‘1. The correlation of ln enzyme activity and pH was significant; r - 0.661 at P < .001 (Fig. 1). Other workers have observed similarly significant correlations (Muller et al., 1980; Bremner and Shaw, 1958; Dubey and Fox; 1974), while some workers have Observed no correlation of denitrifying activity and pH (Khan and Moore, 1968; Burford and Bremner, 1975). The establishment of long term low pH values at these sites may have selected for acid tolerant denitrifiers. Soils from areas Of low pH (3.9) exhibited highest denitrification rates at a pH of 3.8, while soils from higher pH sites (6.8) exhibited highest rates at pH 7 (Fig. 2). This relationship was observed using several treatments to adjust the soil pH, suggesting that the Observed Optima was not caused by a specific chemical addition. Since the soil was a largely unaggregated sandy loam and was slurried, most organisms should not have been protected from the pH of the solution suggesting that microsites of high pH were not responsible for the Observed rates. Previous workers have Observed chemical decomposition of NOZ" in acid soils. Natural nitrate pools prior to imposed anaerobiosis ranged between 5 and 30 ug-N ' g’l. Initially high rates of nitrate reduction in the slurry assay could cause transient accumulations of NOZ" which, through chemical decomposition, might account for the N-gas Observed in the low pH samples. The dominant products attributed to chemical 118 119 decomposition of N02" reduction are NO and N02, with little N20 found unless metals are involved (Moragham and Buresh, 1977). Sterile soils from the low pH subplot incubated in the presence of 1 mM to 1 11M N02" exhibited no detectable N20 over a 24 hour period. Non-sterile soils exhibited N20 production rates of 2000 ng-N-g-day from NOZ'at pH 3 indicating that biological reduction was the dominant fate in these soils. The low pH Optimum for denitrification in acid soil suggests that acidopholic denitrifiers have been selected for by the 24 year period of decreasing pH. l. 2. 4. 5. 7. 8. 9. 10. ll. 12. 13. 14, LITERATURE CITED Bollag, J. -M., M. L. Orcutt, and B. Bollag. 1970. Denitrification by isolated soil bacteria under various environmental conditions. Soil Sci. Soc. Am. Proc. 34:875-879. Broadbent, F. E., and Clark, F. E. 1965. Denitrification. .95 Soil Nitrogen W. V. Bartholomew and F. E. Clark (eds). American Society of Agronomy, Madison, Wisc. pp. 344-359. Buresh, R. J., and Moraghan, J. T. 1976. Chemical reduction of nitrate by ferrous iron. J. Env. Qual. 5:320-325. Burford, J. R., and Bremner, J. M. 1975. Relationships between denitrification capacities of soils and total, water-soluble, and readily decomposable soil organic matter. Soil Biol. Biochem. 13389-394. Dubey, H. D., and R. H. Fox. 1974. Denitrification losses from humid tropical soils of Puerto Rico. Soil Sci. Soc. Sm. Proc. 38:917-920. Firestone, M. K. 1982. Biological denitrification. p.289-326. $2.F° J. Stevneson (ed). Nitrogen in agricultural soils. Agronomy 22: Am. Soc. Agron., Madison, Wisc. Gilliam, J. W. and R. P. Gambrell. 1978. Temperature and pH as limiting factors in loss of nitrate from saturated Atlantic Coastal Plain soils. J. Environ. Qual. 7:526-532. Khan, M. F. A., and A. W. Moore. 1968. Denitrifying capacity of some Alberta soils. Can. J. Soil Sci. 48:89-91. Klemedtsson, L., B. H. Svensson, T. Lindberg, and T. Rosswall. 1978. The use of acetylene inhibition of nitrous oxide reductase in quantifying denitrification in soils. Swed. J. Agric. Res. 7:179-185. Knowles, R. 1981a. Denitrification, p. 323-369. .12.E° A. Paul and J. Ladd (ed.), Soil biochemistry, vol. 5. Marcel Dekker Inc., New York. Knowles, R. 1982. Denitrification. Microbiological Reviews. 46:43-70. MOraghan, J. T., and Buresh, R. J. 1977. Chemical reduction of nitrite and nitrous oxide by ferrous iron. Soil Sci. Soc. Amer. J. 41:47-50. Muller, M. M., V. Sundram, and J. Skujins. 1980. Denitrification in low pH spodosols and pests determined with the acetylene inhibition method. Applied Environ. Microbiol. 40:235-239. Nommik, H. 1956. Investigations on denitrification in soil. Acta Agric. Scand. 6:195-228. 120 15. 16. 17. 18. 19. 20. 21. 22. 121 Parkin, T. B., A. J. Sexstone, J. A. Robinson, and J. M. Tiedje. 1983. A geostatistical analysis Of denitrification rates, denitrification enzyme activity, moisture, nitrate, and pH. Soil Sci. Soc. Am. J. (prepared). Smith, M. S. and J. M. Tiedje. 1979. Phases of denitrification following oxygen depletion in soil. Soil Biol. Biochem. 11:261-267. Soulides, D. A. and Clark, F. E. 1958. Nitrification in grassland soils. Soil Sci. Soc. Amer. Proc. 22:308-311. Stevenson, F. J., Harrison, R. M., Wetselaar, R., and Leeper, R. A. 1970. Nitrosation of soil organic matter, III: Nature Of gases produced by reaction of nitrite with lignins, humic substances, and phenolic constituents under neutral and slightly acidic conditions. Soil Sci. Soc. Amer. Proc. 34:430-435. Valera, C. L., and M. Alexander. 1961. Nutrition and physiology of denitrifying bacteria. Plant Soil 15:268-280. Van Cleemput, O. and Patrick, W. H. 1974. Nitrate and nitrite reduction in flooded gamma-irradiated soil under controlled pH and redox potential conditions. Soil Biol. Biochem. 6:85-88. Wolcott, A. R., H. D. Foth, J. F. Davis, and J. C. Shickluna. 1965. Nitrogen carriers: soil effects. Soil Sci. Soc. Am. Proc. 29:405-410. Wullstein, L. H. and Gilmour, C. M. 1966. Non-enzymatic formation of nitrogen gas. Nature 210:1150-1151. 122 Figure 1. Linear correlation of denitrifying enzyme activity and soil pH from 204 samples of a sandy loam soil. 123 .9 111011111 311121121 N"| 124 Figure 2. Influence of pH on denitrification enzyme activity in soil collected from areas of low and high pH. ID = pH 3.9; 0 = pH 6.8. 125 (O)L_uL-_5-N—6u N muswwm CON e 094 T 000 . com . A 1 mm on on cop mNF coop cm? (0),; P-N-fiu APPENDIX B OXYGEN FLUX MEASURED WITH PLATINUM ELECTRODES AS RELATED TO SOIL DENITRIFICATION RATES Lemon and Erickson (1952) described construction of bare platinum electrodes for measuring oxygen diffusion rate (ODR) in soil. The theory of Operation of these electrodes is simlar to that described in Chapter III. The platinum surface is rendered negative at 0.5-0.8 volts relative to an external Ag/AgCl reference electrode. The resultant potential causes 02 to be electrochemically reduced on the platinum surface. The surface area Of platinum is relatively large (0.06 cm diameter x:0.4 cm long), and consumes oxygen from the surrounding media when a potential is applied. The electric current recorded by these electrodes is proportional to the rate of 02 arrival at the electrode surface. This oxygen flux is related to the current by the following expression (Stolzy et al., 1981): i x 10"6 - n FAJ where i is the current in microamperes, n is the number of electrons required to reduce one molecule of 02, F is the Faraday constant, A is the electrode surface area, and J is the flux of 02 (umoles-Oz °cm"2°sec"1). The oxygen flux J, will vary with soil oxygen concentration and moisture content and can be used as an indication Of soil aeration status (Phene et al., 1976). Fluhler et al. (1976) has suggested that ODR measurements provide a statistical approach for estimating the effect of anaerobic microsites on the soil denitrification rate. Limitations of the method have been discussed in detail by McIntyre (1970). The current observed is affected by the voltage applied (Va). It is necessary to calculate an effective applied voltage 127 (Ve) by taking into account the resistance (R) between the platinum electrode and the Ag/AgCl reference: Ve - Va - R It is also necessary to physically separate the reference electrode from the measuring electrode by a distance of 5-10 cm to obtain reliable readings. The electrode surface can be poisoned by soil constituents and must be frequently cleaned. Finally, the method is appropriate only in relatively wet soils, since a water film must cover the electrode surface and be continuous with the reference electrode. Electrodes were constructed according to Van Dam and Erickson (1966) to evaluate relationships between ODR values and observed denitrification rates. Soil cores for denitrification measurements in Chapter I were modified with side ports such that 10 electrodes could be inserted through gas tight rubber septa. The Ag/AgCl reference was placed in contact with the soil surface through a rubber stopper at the top of the core. The soil core could be attached to the 833 a recirculation system and ODR measurements and denitrification rate coarse}? #99 sieved clay loam soil packed in the cores to a bulk density of 1.5 “3 simultaneously determined. Experiments were performed on a and wetted to moisture contents ranging from 22 to 302. 17 ODR values were determined using a Jensen Instruments Model 13 0 a? ‘19 rate meter. Resistance between each electrode and the referetl“e I 8) determined using a conductivity bridge (Bernstein and McGuirk. 197 119 / eat at VE- Relationships were constructed between resultant current 3nd app cathode voltage (Fig. l) and were used to calculate the curl? 0.5 volts. ‘3 ined 8 F1Sure 2 shows ODR values and denitrification rates determ e 1 e 18 ch varying pore space oxygen concentrations. The reported ODR va u ¥_———_‘__ 128 mean flux determined from 10 replicate electrodes. The ODR rate decreased rapidly as the 2AR begun to increase at low pore space oxygen concentrations. In an earlier study, Brandt et al. (1964) also observed denitrification rates in a clay loam soil at ODR values Of less than 0.2 g-Oz-cm220min'1. Differences in ODR values measured in soils of varying moisture contents were observable at pore space oxygen concentrations of 182 (Table 1) suggesting that ODR values, in addition to soil moisture content might be used to predict soil denitrification rates. Differences were not discernable at lower 02 concentrations. Measured differences in the mean ODR values at the various moisture contents were small (0.2 to 0.5 umoles-02°cm'2-min"l, with sufficient variability among electrode measurements (CV - 10 to 302) to Obscure differences. This variability may result from the large physical size of the electrode, which is in contact with many potentially different aeration sites. ODR values provide an additional indirect measure relating soil aeration to denitrification rates. The measured flux cannot, however, be used to quantitatively determine oxygen diffusion coefficients, which as discussed in Chapter II and III is the important parameter necessary to describe soil anaerobiosis and its effect on denitrification. 1. LITERATURE CITED Bornstein, J. and M. McGuirk, 1978. Modifications to a soil oxygen diffusion ratemeter. Soil Science. 126:280-284. Brandt, G. R., A. R. Wolcott, and A. E. Erickson. 1964. Nitrogen transformations in soil as related to structure, moisture, and oxygen diffusion rate. Soil Sci. Soc. Am. Proc. 28:71-75. Fluhler, E., L. H. Stolzy, and M. S. Ardakani. 1976. A statistical approach to define soil aeration in respect to denitrification. Soil Science 122:115-123. Lemon, E. R., and A. E. Erickson. 1952. The measurement of oxygen diffusion in soil with a platinum microelectrode. Soil Sci. Soc. Am. Proc. 16:160-1630 McIntyre, D.S. 1970. The platinum microelectrode method for soil aeration measurement. Adv. Agron. 22:235-283. Phene, Ce J., R. B. camprII, and Co W. DOtYO 1976. Characterization of soil aeration 12 situ with automated oxygen diffusion measurements. Soil Science. 122:271-281. Stolzy, L. H., D. D. Focht, and H. Fluhler. 1981. Indicators of soil aeration status. Flora. 171:236-265. Van Dorn, D. M., and A. E. Erickson. 1966. Factors affecting the platinum microelectrode method for measuring the rate Of oxygen diffusion through soil solution. Soil Science. 102:1-28. 129 130 Table 1. The effect of increasing soil moisture on oxygen flux and denitrification rates in a clay loam soil. moisture content oxygen flux 2 anaerobic rate (z) 22 24 26 30 0.47 i 0.35 |+ 0.31 I+ 0.28 |+ (umoles-OZ-cmfz'min'l) .06 .06 .04 .03 0.7 0.9 2.1 1.9 131 Figure 1. Relationship between electrode current and applied effective voltage used to calculate current at effective voltage of 0.5. Examples from 3 electrodes in a clay loam soil at 26 percent moisture. MICRUAMPS 132 15 18* 0 .2 f4 .0 EFFECTIVE VOLTAGE Figure 1 133 Figure 2. Relationship between percent anaerobic rate, oxygen flux, and pore space oxygen concentration in a clay loam soil at 26 percent moisture. 134 1s 12 a s e e - 1031mm PERCENT OXYGEN Figure 2 nICHIGnN smTE UNIV. LIBRARIES WI“WW”NWNWWNWWI‘IIHHIHHI 31293008364741