AMMONIA EMISSION FROM MINERAL

AMMONIA EMISSION FROM MINERAL FERTILIZERS AND FERTILIZED CROPS Sven G. Sommer,1 Jan K. Schjoerring2 and O.T. Denmead3 1Department of Agricultural Engineering, Danish Institute of Agricultural Sciences, Research Centre Bygholm, PO Box 536, DK-8700 Horsens, Denmark 2Plant Nutrition Laboratory, Department of Agricultural Sciences, The Royal Veterinary and Agricultural University, Thorvaldsensvej 40, DK-1871 Frederiksberg C, Denmark 3CSIRO Land and Water, GPO Box 1666, Canberra ACT 2601, Australia I. Introduction II. Mineral Fertilizer Consumption III. Ammonia Volatilization from Mineral Fertilizers A. Production and Transport of NH3 in the Soil–fertilizer– atmosphere Interface B. Temporal NH3 Loss Pattern C. Hydrolysis of Urea D. Soil Hþ E. Soil CEC F. Solid Phase Processes G. Climate and Infiltration H. Microbial Processes (nitrification/immobilization) IV. Ammonia Emission from Crop Foliage A. Transport of NH3 Between Leaves and the Atmosphere B. Magnitude of NH3 Losses C. Physiological Processes Involved in NH3 Emission from Crops V. Management Strategies A. Techniques for Reduction of NH3 Emission B. Fertilizer Composition C. Flooded Fields (rice paddies) D. Injection of Anhydrous Ammonia E. Crop Emissions as Affected by Fertilizer Application F. Ammonia Emission from Decomposing Plant Material G. Absorption by Crops VI. Measurement Techniques A. Tracer Techniques B. Enclosures C. Micrometerological Methods E. Gradient Diffusion Methods F. Eddy Correlation G. Relaxed Eddy Accumulation or Conditional Sampling H. Lagrangian Dispersion Models 557 Advances in Agronomy, Volume 82 Copyright q 2004 by Academic Press. All rights of reproduction in any form reserved 0065-2113/03 $35.00 ARTICLE IN PRESS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 VII. Conclusions and Perspectives Appendix A Acknowledgments References A thorough understanding of the physical and chemical processes involved in NH3 emission from inorganic N fertilizers and fertilized crops is required if reliable and operational NH3 emission factors and decision support systems for inorganic fertilizers are to be developed, taking into account the actual soil properties, climatic conditions and management factors. For this reason, the present review focuses on processes involved in NH3 volatilization from inorganic nitrogen fertilizers and the exchange of ammonia between crop foliage and the atmosphere. The proportion of nitrogen lost from N fertilizers due to NH3 volatilization may range from <0 to .50%, depending on fertilizer type, environmental conditions (temperature, wind speed, rain), and soil properties (calcium content, cation exchange capacity, acidity). The risk for high NH3 losses may be reduced by proper management strategies including, e.g., incorporation of the fertilizer into the soil, use of acidic fertilizers on calcareous soils, use of fertilizers with a high content of carbonate-precipitating cations, split applications to rice paddies or application to the soil surface beneath the crop canopy. The latter takes advantage of the relatively low wind speed within well-developed canopies, reducing the rate of vertical NH3 transport and increasing foliar NH3 absorption. Conversely, NH3 is emitted from the leaves when the internal NH3 concentration is higher than that in the ambient atmosphere as may often be the case, particularly during periods with rapid N absorption by the roots or during senescence induced N-remobilization from leaves. Between 1 and 4% of shoot N may be lost in this way. q 2004 Academic Press. I. INTRODUCTION Agriculture is recognized as a major source of atmospheric ammonia (NH3), contributing 50% of global NH3 emissions (Schlesinger and Hartley, 1992). Recent inventories have shown that mineral fertilizers and plants account for about 20% of the total emission of NH3 in Europe (ECETOC, 1994; Pain et al., 1998; Hutchings et al., 2001) and 23% of global emission of NH3 is derived from fertilizers and field-applied manure (Bouwman et al., 2002). Ammonia is a chemically active gas and readily combines with NO3 2 and SO4 22 in acid cloud droplets to form particulates (Asman et al., 1998). The NH4þ particles can be transported over long distances before being dry- or wet-deposited, while 558 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 gaseous NH3 usually is deposited much closer to the source (Asman and van Jaarsveld, 1991). The depositedNH3 may cause acidification and eutrophication of natural ecosystems (Schulze et al., 1989). Therefore, the United Nations has included NH3 in the Convention on Long-range Transboundary Air Pollution (CLRTP), and in addition, the EU Commission has set a limit—the NH3 ceiling—to the emission of NH3 from European countries (EEA, 1999). For farmers, the loss of fertilizer nitrogen (N) due to NH3 emission may significantly reduce N-fertilizer efficiency, contributing to a rather low overall efficiency of applied N, i.e., ,50% in the tropics and ,70% in temperate areas (Malhi et al., 2001). In order to prevent potential negative consequences of gaseous NH3 losses, farmers may also apply fertilizer-N in excess of crop requirements, which will increase the loss of N to the environment and production costs. Although the use of fertilizers increased dramatically during the 20th century (Fig. 1), the recent trend is for their stabilized or decreased use, as awareness of the economic consequences and environmental impacts of N losses is encouraging more efficient N utilization. In the 1980s, emission factors were introduced to calculate NH3 emission from European agriculture (Buijsman et al., 1987). Country-specific emission factors were introduced in inventories in the 1990s (Misselbrook et al., 2000; Hutchings et al., 2001). Major uncertainties are still associated with the use of NH3 emission factors for inorganic fertilizers, because they in many cases are highly empirical or have been derived from measurements carried out under conditions that deviate considerably from modern management practices associated with handling and applying fertilizers. As an example, the generally used emission factor for urea is 15% for Europe and 25% for the tropics (see e.g., Bouwman et al., 1997), which contrasts with the fact that NH3 emission from urea can be completely avoided if the fertilizer is incorporated into the upper soil layers Consumption, Tg N year–1 Developed world Developing world World Year 1960 1970 1980 1990 2000 Million tonnes per year 0 20 40 60 80 0 5 10 15 20 25 30 35 Other straight N NPK-N Ammonium nitrate Anhydrous-NH3 Nitrogen sollutions Ammonium phosphates Calcium Nitrate Ammonium Other NP-N Ammonium sulphate Urea Figure 1. Left: Global consumption of mineral fertilizers (IFA Statistics, 2002); Right: Distribution of the fertilizer consumed in late 1990 (Bouwman et al., 1997). AMMONIA EMISSION 559 ARTICLE IN PRESS 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 (Harrison and Webb, 2001), or reduced to well below 10% if applied to a wellestablished crop (Schjoerring and Mattsson, 2001). More reliable NH3 emission estimates could potentially be derived from mathematical models based on the physico-chemical processes controlling NH3 emission from fertilizers and their interactions with soil, canopy and atmospheric variables. Complex, mechanistic models of NH3 emission exist (Avnimelech and Laher, 1977; Fleisher et al., 1987; Kirk and Nye, 1991; Genermont and Cellier, 1997), but there are still difficulties in describing the controlling processes and their interactions. The substantial requirement for input data makes these models difficult to apply in decision support systems. The present review focuses on the processes underlying NH3 volatilization from inorganic N fertilizers. The influence of various soil and climatic parameters on the production of NH3/NH4þ during dissolution of the fertilizer is described and related to the temporal variation in NH3 losses and their accumulated values. The existing knowledge on the ability of growing crop plants to reduceNH3 losses from fertilizers lying on the soil surface beneath the canopy is outlined, as is the case for the magnitude and mechanisms of NH3 emission from the crop foliage itself. Finally, a description of measurement techniques for quantifyingNH3 losses under field conditions is included, since methodological aspects are important for assessing loss rates reported in the literature as well as for planning new experiments aiming at obtaining more and better knowledge onNH3 volatilization. Taken together, the information in the review provides an integrated picture of fertilizer-derived NH3 emissions, facilitating the development of decision support systems intended to limit the volatile loss of fertilizer-N from fertilizers and plants. II. MINERAL FERTILIZER CONSUMPTION Figure 1 shows the consumption of mineral fertilizers up to the end of the 20th century. In the 1950s, mineral fertilizers became cheap and the use of mineral-N increased, reducing dependence on leguminous N fixation and animal manure as sources of plant nutrients. In the developed world, the consumption of N fertilizers increased until the late 1980s, but has declined since then due to stricter environmental legislation and a growing number of set-aside policies in the European Union. The consumption of fertilizers in central and Eastern Europe as well as in the Commonwealth countries (EFMA, 1997) fell significantly during the 1990s due to the economic crisis that emerged after the fall of the Berlin Wall. Nitrogen fertilizer consumption has increased steadily in the developing world since 1960, mainly due to the increased use of mineral fertilizers in Asia (IFA Statistics, 2002). Urea constitutes about 38% of the total consumption of nitrogenous fertilizers, which is reported to be about 77 Tg N year21 (Bouwman et al., 1997). Urea is an 560 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 organic compound (amide) with a very high concentration of N (46%). After application to soil, the urea is hydrolyzed to ammonium; therefore, urea is included in the category of mineral fertilizers. Other straight N fertilizers than urea include ammonium bicarbonate (ABC) and ammonium chloride. Ammonium bicarbonate comprises 90% of the other straight N fertilizers, and is the second most used N-containing mineral fertilizer after urea (Fig. 1). It is widely used in China, where it is produced locally by small- and medium-sized manufacturers (Andreas Pacholsky, 2000; personal communication). Apart from urea and ABC, the usage of straight N fertilizers, in order of consumption, is ammonium nitrate (AN) . anhydrous ammonia (AA) . nitrogen solutions . calcium–ammonium–nitrate (CAN). Nitrogen solutions often consist of urea—ammonium–nitrate solutions (UAN), which contain 28–32% N (EFMA, 1997). Examples of multi-nutrient N-containing fertilizers are NPK, NP and ammonium sulfate. These fertilizers are either produced by chemical reactions (complex or compound fertilizers) or by mechanical blending of relevant minerals (EFMA, 1997). The nitrogen content of the fertilizers may be very variable (5–26%) and the content of ammonium and nitrate may vary between fertilizers. III. AMMONIA VOLATILIZATION FROM MINERAL FERTILIZERS A. PRODUCTION AND TRANSPORT OF NH3 IN THE SOIL–FERTILIZER–ATMOSPHERE INTERFACE Emission of NH3 from applied fertilizers follows transport of NH3 from the surface of an ammoniacal solution to the atmosphere. The solution can be on or within the soil surface or within plants. The rate of emission is determined by the concentration gradient and resistance to NH3 transport between the surface and the atmosphere as controlled by atmospheric transport processes, the chemical composition of the solution, and transformations of Total Ammoniacal Nitrogen (TAN) (NH3 2 N þ NH4þ 2 N) in the soil and plants. The air above the surface can be envisaged as a laminar or turbulent-free layer close to the surface and, above this, a turbulent layer. Ammonia gas at the liquid–air interface (x) is transported through the laminar layer by molecular diffusion and then through the turbulent layer to the free atmosphere by turbulent diffusion. The instantaneous rate of NH3 loss (Fv) may be given by the following equation, as shown by Rachhpal-Singh and Nye (1986a) and van der Molen et al. (1990): AMMONIA EMISSION 561 ARTICLE IN PRESS 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 NHþ4 $NH3 " þHþ ð1Þ Fv ¼ Kb £ ðx 2 NH3;aÞ ð2Þ where Kb is a bulk transfer coefficient, and x and NH3,a are, respectively, the partial pressure of NH3 in the air at the soil/plant–air interface and in the free atmosphere. The transfer coefficient is highly dependent on wind speed, but also depends on atmospheric stability. x is determined by the concentration of TAN and equilibrium processes in the solution (Freney et al., 1983; Sherlock and Goh, 1985): ½NH3;L ¼ TAN 1 þ 10ð0:09018þ2729:92=T 2 pHÞ ð3Þ x ¼ ½NH3;L £ 101477:7=T21:69 ð4Þ where NH3,L is the concentration of NH3 in the solution, and x is the equilibrium concentration of NH3 in the atmosphere in immediate contact with the soil solution. The change in NH3,L concentration of a solution of TAN and bicarbonate at different pH and temperatures is shown in Fig. 2. It is seen that below pH 7, the concentration of NH3,L is low. Therefore, the increase in NH3,L with temperature is most noticeable when pH is .7. Soil buffers and cations and anions added with fertilizers will affect soil pH and, hence, NH3 emissions. Furthermore, NH3 Figure 2. Effect of temperature and pH on NH3 concentrations in a liquid solution of TAN and TIC under conditions with no emission of NH3 and CO2 gases. It is assumed that 100 kg urea is mixed homogeneously in the top 1 cm soil layer and the soil has a water content of 30% (vol/vol); thus after hydrolysis, TAN concentration is ca. 0.1 mol l21 and TIC is 0.05 mol l21. 562 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 emissions will be reduced through reduction of TAN as a result of NH3 loss, infiltration of TAN or fertilizers below the soil surface by rainfall, reaction with the cation exchange complex, immobilization, and uptake by plants and oxidation by nitrifying organisms. Transport processes at the gaseous interface and in the air layers above the soil surface can be modeled by use of a bulk transfer coefficient (Eq. (2)). The transfer coefficient (Kb) is parameterized by a series of resistances that are additive and are related to the resistances to transport by diffusion and convection. The resistances comprise an aerodynamic resistance (ra) representing the resistance of the turbulent air layer between observation height and the aerodynamic roughness length of the surface (van der Molen et al., 1990; Genermont and Cellier, 1997), a laminar resistance (rb) representing the resistance to transport across the laminar boundary layer near the surface, which is dominated by molecular diffusion, and an interfacial resistance (rc) representing the resistance to transport within the soil or plant solution layer to the air interface (van der Molen et al., 1990). Kb ¼ 1 ra þ rb þ rc ð5Þ A similar set of resistances describes NH3 loss from water bodies, including rice paddies (Denmead and Freney, 1992). The resistance ra in the turbulent layer can be calculated from aerodynamic theory using developments by, for example, Paulson (1970) and Monteith and Unsworth (1990): ra ¼ ln½ðz 2 dÞ=z0 2cgðz 2 dÞ kup ð6Þ where z is the height of measurement within the boundary layer of the emitting field, d is the zero plane displacement (the height at which wind speed appears to go to zero), z0 is the roughness length of the surface, cg is a function that accounts for the effects of atmospheric stability on gas transport and is defined in Appendix 1, k is von Karman’s constant (< 0.4), and up is the friction velocity, a measure of the flux of momentum between the atmosphere and the ground. The friction velocity can be measured directly by eddy covariance or estimated from the profile of horizontal wind speed u: up ¼ ku ln½ðz 2 dÞ=z0 2cmðz 2 dÞ ð7Þ The stability function cm is given in Appendix 1. The roughness length and the zero plane displacement vary with surface characteristics. z0 and d are assumed to be linear functions of the canopy height (h) (Monteith and Unsworth, 1990): z0 < 0:1h: ð8Þ AMMONIA EMISSION 563 ARTICLE IN PRESS 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 d < 0:6h: ð9Þ The resistance of the laminar boundary layer rb above the crop canopy or soil surface can be estimated using the empirical relationship of Thom (1972): rb ¼ 6:2u20:67 p ð10Þ Van der Molen et al. (1990) estimate the resistance rc within the soil surface layer as rc ¼ 0:5Ll DaqKH þ Dg ð11Þ where the numerator represents the average distance over which the ammoniacal N has to be transported to reach the soil surface, and Daq and Dg are the soil– liquid and soil–gas diffusion coefficients for ammoniacal N. Sutton et al. (1998) and Nemitz et al. (2001) present formulations for the resistances to NH3 transport in plant leaves and canopies, and Denmead and Freney (1992) do the same for NH3 transport across the viscous sublayer below the air–water interface in water bodies. Soil surface roughness leads to an increase in turbulence, which leads to an increase in friction velocity (Brutsaert, 1982), and accordingly to an increase in the exchange between the soil surface and the atmosphere. Thereby, surface NH3 gas becomes better mixed and dispersed in the near atmosphere, and NH3 emission is increased. Consequently increasing wind speed and temperature will increase NH3 emission from applied fertilizers (Eqs. (1)–(10)), but on days with high global radiation, increasing wind speed will reduce soil and plant surface temperatures and thereby the emission potential of the solution. Thus, a number of studies of NH3 emission from manure applied to fields have shown that NH3 emission is not always related to wind speed (Bussink et al., 1994; Beauchamp et al., 1978; Sommer et al., 1997). The effect of the internal boundary layers may be illustrated by considering a field amended with ammoniacal fertilizers and surrounded with untreated fields. Downwind from the leading edge, the atmospheric NH3 concentration will gradually increase due to the input of NH3 volatilized from the soil solution, but the effect of the increase on emission rate will depend on the surface boundary conditions. The driving force for NH3 emission is the concentration difference between the interface of the soil or plant liquid containing TAN and the atmosphere (Eq. (2)). If NH3 is freely available at the surface, as in surface spreading of slurries, the surface concentration will remain more or less unchanged and the NH3 emission will gradually decline downwind. The dynamics of the emission will resemble those for a constant concentration boundary condition at the surface, or a radiation boundary condition if the wind speed is varying in time. If, however, NH3 production is controlled by soil processes, such as the hydrolysis of urea incorporated 564 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 into the soil, the emission rate will be virtually unaffected by plot size or fetch. Ammonia concentrations in the soil air will simply increase in response to the increase in the concentration of the air above and the dynamics will resemble those for a constant flux condition at the surface. In the case of slurry spreading, increasing the area to which fertilizer is applied at a constant application rate will be expected to reduce the emission rate of NH3 as a percentage of the amount of TAN applied (Genermont and Cellier, 1997). The initial NH3 emission rates after fertilizer application will tend to be larger from small plots than big fields, and to decline faster because TAN has been reduced due to the larger emission during the first few hours. The influence of field size on NH3 loss has been demonstrated by results of simulations for smaller plots (0–25 m) calculated by Vlek and Craswell (1981) for a flooded soil using a diffusion model (Bouwmeester and Vlek, 1981). Figure 3 shows the dependence of emission rate on plot size, calculated from Philip’s (1959) analysis for a constant concentration boundary condition at the surface. Figure 3 indicates that the average emission rate from a small treated plot with a fetch of only 1 m would be around 2.5 times that from a large field with a fetch of 1000 m, and even for a plot with a fetch of 100 m, the emission rate would still be about 30% higher than that from the large field. B. TEMPORAL NH3 LOSS PATTERN NH3 emission from a soil applied fertilizer containing TAN is a function of dissolved NH3 in equilibrium with the atmosphere either directly or through soil pores. According to the dissociation of NH4þ and Henry’s constant (Eqs. (3) and (4)) the emission is related to the concentration of TAN and Hþ(pH ¼ 2log(Hþ)). Length of field, m 0 200 400 600 800 1000 F/F1000 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Figure 3. Ammonia emission rate (F) as a fraction of the emission rate (F1000) at 1000 m calculated using the model developed by Philip (1959). AMMONIA EMISSION 565 ARTICLE IN PRESS 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 In consequence, doubling the TAN concentration will double NH3 emission, whereas doubling the Hþ concentration will halve the emission. Typical patterns of NH3 volatilization over time for urea and diammonium phosphate are shown in Fig. 4. Since urea fertilizer contains no TAN, no NH3 is emitted immediately after urea application to the soil. Subsequently, the urea in contact with the soil will absorb water, and urea will hydrolyze, producing TAN and bicarbonate (HCO3 2). The rate of hydrolysis is related to the amount of soil water absorbed and temperature; therefore, the initial lag phase with no NH3 volatilization may vary. The NH3 loss rate declines after 5–10 days due to a reduction in NH3 concentration as TAN is volatilized, dissolved in increasing volumes of soil water, leached by rain into the soil, absorbed to soil, or transformed to nitrate (Black et al., 1985; Haynes and Williams, 1992). The rate of NH3 volatilization from urea has been described by a logistic equation by Stevens et al. (1989), showing that maximum loss rates may occur within one to 10 days after application, depending on soils and environmental conditions. A sigmoidal model was used by Sommer and Ersbøll (1996) to relate cumulated loss of NH3 from urea to days from application, showing that for loamy soils half the total loss may be reached 2–7 days after urea application (Table I). The pattern of NH3 volatilization from urea is affected by whether urea is applied in pellets, in solution, or crushed to a fine powder, because dissolution of and diffusion of urea into the soil solution will be different. From urea applied as a powder and in solution, the emission will occur earlier than after application of prilled or granulated urea. This delay in emission from prilled or granular urea is related to hydrolysis of the applied urea (see Section III.C on hydrolysis of urea). Other ammoniacal fertilizers are composed of NH4þ salts of phosphate, sulfate, or nitrate; these salts are readily dissolved in the soil water after application of Days from application 0 2 4 6 8 10 12 Cumulative NH3 volatilization (% of N) 0 2 4 6 8 10 12 14 16 18 DAP Urea Figure 4. Ammonia volatilization from urea and diammonium phosphate (DAP) applied to a sandy loam measured with a wind tunnel (after Sommer and Jensen, 1994). 566 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 Table I Models for Predicting NH3 Volatilization from Surface-Applied Ammoniacal Fertilizer Fertilizer Equation Definition of variables References Urea F ¼ a=ð1 þ be2ctÞ 2 að1 þ bÞ Nmax ¼ ab=ð1 þ bÞ tmax ¼ lnðbÞ=c for b . 0 tmax ¼ 0 for b # 0 F is the NH3 volatilization (% of urea-N or % of TAN), t is the time in days from application of urea and a, b, and c are time-related parameters Nmax is total cumulated NH3 volatilization (% of urea-N) tmax is the time in days where the loss rate is at maximum Al-Kanani et al. (1995) Urea F ¼ Nmaxð1 2 e2btÞc Asman and van Jaarsveld (1991) NH4þ mineral fertilizer F ¼ Nmaxð1 2 e2btÞ Asman and van Jaarsveld (1991) Urea F ¼ 3:67 þ 22:28ðDMÞ DM ¼ dry matter per unit surface (25 £ 65 cm2) Asman et al. (1998) Urea F ¼ 12:1 2 0:97CEC þ 0:02CEC2 2 0:011SMC þ 0:049T CEC ¼ cation exchange capacity (cmol kg21), SMC ¼ soil moisture content (g g21) and T ¼ temperature (8C) Atwell et al. (2002) Urea F ¼ 20:84 2 0:0452TA TA is total acidity (mmol kg21) Avnimelech and Laher (1977) Data were taken from Stevens et al. (1989), Sommer and Ersbøll (1996), Hoult and McGarity (1987), McGarry et al. (1987), Stevens et al. (1989). In the study of McGarry et al. (1987), a closed chamber was used and emission was measured by absorbing NH3 in a solution of acid, in the other studies emission was measured with dynamic chambers in the laboratory. AMMONIA EMISSION 567 ARTICLE IN PRESS 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 the fertilizer, and NH3 volatilization will start within a short time after their application. The NH3 volatilization rate will decline due to the same processes as mentioned previously for the decline in volatilization rate after application of urea. In consequence, the NH3 volatilization pattern shows no lag phase after application and the cumulated loss may therefore be described by a simple exponential equation (Table I). C. HYDROLYSIS OF UREA Urea is the diamide of carbonic acid and as such an organic fertilizer. Like other amides, urea in solution is hydrolyzed to NH3 and bicarbonate (HCO3 2). No NH3 is lost from urea that has not been transformed and emission of NH3 from urea applied in the field is, therefore, closely related to hydrolysis of urea by the enzyme urease: COðNH3Þ2 þ 2H2O$NH3 þ NHþ4 þ HCO23 ð13Þ This reaction produces a mixture of NH3, ammonium (NH4þ) and bicarbonate (HCO3 2). The rate of hydrolysis of urea is related to urease activity, the availability of water (Eq. (13)), pH, and temperature (Bremner and Douglas, 1971; Fig. 5). Very dry soils with water content below permanent wilting point and a dry air environment will delay or inhibit urea hydrolysis; thus at soil water potential ,21.5 MPa, hydrolysis has been shown to be insignificant (Al-Kanani et al., 1991). Increasing water content of the soil will increase the rate of hydrolysis (Vlek and Carter, 1983; McInnes et al., 1986a; Reynolds and Wolf, 1987). Air humidity influences hydrolysis of urea because prilled and granular urea are pH 7 8 9 10 0 2 4 6 8 10 12 Activity, micro mol g–1 soil h –1 Urea, mol N l–1 (soil solution) 0 2 4 6 8 10 0 10 20 30 Idasoil, pH 7.5 40 Marshallsoil, pH 6.8 Edinasoil, pH 6.1 Soil pH 4.9 Soil pH 7.8 Soil pH 6.4 Figure 5. Rate of urea hydrolysis related to soil pH (left: Tabatabai and Bremner, 1972) and concentration of urea (right: Nye, 1992). 568 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 hygroscopic and will absorb water at high air humidity. In consequence, NH3 emission may be significant from urea applied to a dry soil if air humidity is high (Bouwmeester et al., 1985; Reynolds and Wolf, 1987; Black et al., 1987a). Hydrolysis of urea may be delayed in soils low in pH or after the addition of acidifying anions mixed with the urea (Ouyang et al., 1998); optimum pH for soil urease has been reported to range from pH 8 to 9 (Fig. 5). Figure 5 shows that urease activity varies between soils and that for some soils the rate of hydrolysis is not much affected by pH. This may explain why Rachhpal-Singh and Nye (1986a) found that the rate of hydrolysis only varied insignificantly when soil pH was in the range from 6.6 to 8.6. In soils where water content is not limiting, urease activity increases with temperature (Vlek and Carter, 1983). Urease activity also increases with increasing urea concentration (Tabatabai and Bremner, 1972). The hydrolysis of urea applied in pellets is slow compared with that applied in solution or as a fine powder, due to the slow diffusion of urea into the soil where urease is abundant (Vlek and Carter, 1983). Therefore, hydrolysis may reach a maximum within 3–5 days for pellets and 24–48 h for urea in solution or as a fine powder (Lyster et al., 1980; Black et al., 1987a; Thomas et al., 1988; Whitehead and Raistrick, 1990). Recous et al. (1988) showed that urea applied in the field was hydrolyzed with a half-life of 1.9 days at 3.58C; on the other hand, urea well mixed with the soil was hydrolyzed with a half life of 22, 15 and 6 h when incubated in the laboratory at 4, 10 and 208C. This indicates that hydrolysis rate increases significantly with temperature, and that urea mixed with the soil is hydrolyzed rapidly. At urea concentrations below 1 mol N l21 (soil) the rate of hydrolysis initially increases linearly with concentration, then reaches an optimum value, eventually decreasing at higher urea concentrations (Fig. 5). This may be explained as substrate inhibition, which can be described by a Michaelis–Menten equation (Nye, 1992). For the purpose of predicting hydrolysis in soils low in urea and with pH in the range from 6.5 to 8.4, urease activity related to pH has been described using the Michaelis–Menten equation at substrate concentrations up to 0.2 mol urea-N per liter of soil followed by a straight line at urea concentrations above 0.2 mol N per liter (Rachhpal-Singh and Nye, 1986b). It is important to realize that after application of urea fertilizer in the field, the maximum concentration of urea may be 10 mol N l21 (soil) near the fertilizer granules (Nye, 1992), and that urea hydrolysis has been observed to occur at concentrations below 1 mol N l21(soil) in most of the studies cited; therefore, the algorithms given cannot predict hydrolysis near the applied granules. Hydrolysis of urea applied to flooded paddy soils for rice production may be affected by redox potential as well as pH. Thus at low redox potential (anoxic), half-time of the hydrolysis (t1/2) has been shown to be 11.2–24.8 h compared with 7.4–13.8 h for oxidized suspensions of acid soils (Lindau et al., 1989). In a suspension of calcareous soils, hydrolysis rate is not affected significantly by AMMONIA EMISSION 569 ARTICLE IN PRESS 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 differences in redox potential (Lindau et al., 1989); however, the rate of hydrolysis in calcareous suspension was greater (t1/2 of 1.7–1.9 h) than in the suspension of acid soil. The effect of organic and inorganic compounds added to urea with the function of delaying hydrolysis of urea and thereby reducing NH3 emission is discussed in Section V on management strategies. D. SOIL H1 One of the most important rate controlling factors of NH3 emission from applied fertilizers is the Hþ(pH ¼ 2log(Hþ)) concentration. At pH values higher than 7, the concentration of NH3,L (see Eqs. (3) and (4)) is at a level that may produce significant emissions from the applied fertilizer (Fig. 2). Therefore, the fertilizers may be grouped as acidic and alkaline with, respectively, a low and high potential for NH3 emission (Table II). Soil alkalinity will also interact with fertilizers and affect the concentration of Hþ. Furthermore, the fertilizer application will create “hot spots,” and soil Hþ concentration will typically show a trend from the fertilizer granule microsite to surrounding unaffected soil (Rachhpal-Singh and Nye 1986a,b). After application of ammonium salts (phosphate, nitrate, sulfate, etc.), the emission of NH3 will produce acids of the anions (Eq. (14)), as shown for ammonium sulphate (AS) (AS ¼ NH4HSO4) in the following example: NH4HSO4 þ H2O$NH3 " þHþ þ HSO24 ð14Þ Each mole of emitted NH3 will increase the concentration of Hþ by one mole. Therefore in acidic soils or poorly buffered soils, the pH will decline due to the NH3 emission from applied NH4HSO4, and the process of NH3 emission will not continue to the right, because little NH3,L will be available for volatilization and the total loss of TAN will be low. In calcareous soils, emission will be related to the solubility of the anion (Fenn and Hossner, 1985; Du Preez and Burger, 1988). The solubility of nitrate (NO3 2), Table II Fertilizers Grouped in Relation to the Acidity or Alkalinity They Produce When Dissolved in Soil Water Acid Moderate acid Alkaline (NH4)2SO4 (NH4)H2PO4 (NH4)HCO3 NH4NO3 (NH4)2CO3 NH4Cl NH3 (NH4)2HPO4 (NH3)2CO 570 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 for example, is high compared to SO4 22 and hydrogenphosphate (HPO4 22), which may precipitate with Ca2þ (Fenn and Hossner, 1985). Due to precipitation of the anions HPO4 22 and SO4 22, carbonate (CO32) will dissolve and this will increase the pH of the soil solution (Larsen and Gunary, 1962; Fenn et al., 1981). The pH in calcareous soils will be between 7 and 8 and most of the total inorganic carbonate (Total inorganic carbon (TIC) ¼ HCO3 2 þ CO3 22 þ CO2) will be in the form of HCO3 2; thus the reaction in the soil will be as follows: ðNH4Þ2X þ CaCO3 $NH3 þ NHþ4 " þHCO23 þ CaX# ð15Þ ðNHþ4 Þ þ HCO23 $NH3 " þCO2 " þH2O ð16Þ ðNH4Þ2X þ CaCO3 $2NH3 " þCO2 " þCaX # þH2O ð17Þ As mentioned above, HCO3 2 is an alkaline anion and the precipitation of the acid anions (X ¼ HPO4 22 or SO4 22) and dissolution of CaCO3 will increase pH, thereby increasing the potential for NH3 emission. It is, therefore, a general rule that emission is lower after application of fertilizers containing soluble anions than from fertilizers containing anions that may precipitate. Hydrolysis of urea will produce a mixture of NH3, NH4þ, HCO3 2 and CO3 2 and this may increase pH, because NH3 and CO3 22 are bases (pKa ¼ 9.48 for NH3/NH4þ and pKa ¼ 10.4 for HCO3 2/CO3 22). Application of the fertilizers ammonium bicarbonate (ABC ¼ (NH4)HCO3) or ammonium carbonate (AC ¼ (NH4)2CO3)) may also increase soil pH due to the alkaline properties of NH3 and CO3 22. After hydrolysis of urea or in soil amended with ABC, the pH in the microsites affected by the fertilizers will be .8; at this high pH a high proportion of TAN will be of the NH3 form and most of the TIC will be in the form of HCO3 21. In consequence, each mole of emitted NH3 will increase the concentration of Hþ by one mole (Eq. (1)), and emission of CO2 will reduce the Hþ concentration by one mole (Eqs. (18), (19), Fig. 6)). CO22 3 þ H3Oþ $HCO23 þ H2O ð18Þ HCO23 þ H3Oþ $CO2 " þH2O ð19Þ Due to the combined NH3 and CO2 emission from these fertilizers, the NH3 emission may be very high after application of urea, ABC, and AB to soils (Roelcke et al., 2002). The alkaline fertilizers such as ABC, urea, and diammoniumphosphate (DAP) generally cause a much higher NH3 emission than neutral or acidic fertilizers such as AS, CAN, or monoammonium phosphate (MAP), because pH is a dominant factor controlling NH3 emission. Inorganic ammonium compounds containing anions that are not producing precipitating calcium solids (NO3 2, Cl2) will reduce pH and the emission will be significantly lower than the emission from fertilizers precipitating with Ca ions (Fenn et al., 1981). The addition of, for example, calcium chloride, potassium chloride, sulfate, or triple superphosphate AMMONIA EMISSION 571 ARTICLE IN PRESS 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 reduces pH and in consequence will reduce NH3 emissions (Sloan and Anderson, 1995; Christianson et al., 1995; Fan et al., 1996; Ouyang et al., 1998; Goos et al., 1999). The typical pattern of accumulated NH3 emission from fertilizers will generally show NH3 emission rates in the following order ABC . urea . DAP . AS . CAN ¼ MAP due to the alkaline properties of the fertilizers and interaction of precipitation of anions (Du Preez and Burger, 1988; Whitehead and Raistrick, 1990; Sommer and Jensen, 1994; Ouyang et al., 1998; He et al., 1999; Zia et al., 1999; Roelcke et al., 2002). From soils with limited buffer capacity, the NH3 volatilization will increase at increasing application rates of ABC and urea, because pH will be high for a long period (Black et al., 1985; Fenn et al., 1987). However, emission of NH3 from acid fertilizers (SO4 22, PO4 32, Cl2 etc.) will decrease at increasing concentrations of TAN in fertilized soils with limited buffer capacity (Avnimelech and Laher, 1977), because NH3 emission will immediately cause a reduction in pH locally (Eq. (1)). From soils high in NH3, mol ltr.–1 CO2: NH3 =1:1 CO2: NH3 =2:1 CO2: NH3 =0:1 pH of soil solution 0.00 0.01 0.02 0.03 0.000 0.005 0.010 0.015 0.020 0.025 5 6 7 8 9 10 NH3 emission, mol l –1 Figure 6. Effect of emission of NH3 and CO2 from a liquid solution of TAN (NH3 þ NH4þ), TIC (CO2 þ HCO3 2 þ CO3 22), and Cl2. The initial concentrations are TAN 0.1 mol l21, TIC 0.05 mol l21, and Cl2 0.06 mol l21, and pH is 6.85. It is assumed that either equal amounts (mol) of TIC and TAN (CO2:NH3 ¼ 1:1), 2 units of TIC to 1 unit of TAN (CO2:NH3 ¼ 2:1) or 1 unit of TAN (CO2:NH3 ¼ 0:1) is emitted. 572 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 buffer capacity, soil pH is of importance irrespective of the concentration of TAN (Avnimelech and Laher, 1977). NH3 volatilization from granulated fertilizers and AA has been correlated with the total acidity and to titratable acidity of soils (Izaurralde et al., 1987; Stevens et al., 1989;Sommer and Ersbøll, 1996). On non-calcareous soils, the accumulated NH3 losses (Nmax) from urea and CAN are positively related to soil pH and inversely to total acidity (Fig. 7), i.e.,NH3 losses are closely related to both soil pH and Hþ buffering capacity of the soil (Avnimelech and Laher, 1977; Ferguson et al., 1984; Stevens et al., 1989). Therefore, for the purpose of predicting losses 0 100 200 300 400 500 0 5 10 15 20 25 30 pH(H2O) 3 4 5 6 7 8 9 TA, mmol kg–1 Cumulated NH3 volatilization (% of N) Cumulated NH3 emission (% of N) 0 10 20 30 40 F=6.12pH-19.7, R2=0.70 F= –0.05TA+22.4 R2=0.52 Figure 7. Cumulative NH3 volatilization related to soil pH and total acidity (TA) of the soil (adapted from Stevens et al., 1989; Sommer and Ersbøll, 1996; Whitehead and Raistrick, 1990). AMMONIA EMISSION 573 ARTICLE IN PRESS 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 from urea applied to the field, the relation between Nmax and soil pH may be used (Table I), as shown in Fig. 7; this relationship explains about 70% of the variation in three different studies. The volatilization from ammoniacal mineral fertilizers may also be predicted by the relation between NH3 volatilization and pHmax (Fig. 8), but the pHmax of the soil surface after fertilizer application should be measured or estimated. More complicated mechanistic models presented by Avnimelech and Laher (1977); Rachhpal-Singh and Nye (1986a,b); Fleisher et al. (1987); and Kirk and Nye (1991) may be used for predicting the risk of volatilization when applying different types of fertilizers to the field; the problem with these models is that the input data for the models may often be missing. E. SOIL CEC TAN in soil will be partitioned between the three soil phases—liquid, solid, and gas phase. The NH4þ component in solution will be in equilibrium withNH4þ on the solid phase exchange site (Fleisher et al., 1987). At pH , 8, more than 95% of the TAN (Fig. 2) will be of NH4þ form and can be exchanged with exchangeable cations (ex-C). Most agricultural soils contain the divalent cations Ca2þ and Mg2þ with a high affinity for adsorption; the exchange of NH4þ on exchange sites therefore has to be defined with the activity ratio law (Russel, 1977): NHþ4 þ ex 2 C$ex 2 NH4 þ Cþ ð20Þ NHþ4 pffiCffiffiaffi2ffiþffiffiþffiffiffiMffiffiffigffiffi2ffiþffi ¼ K Ex 2 NH4 Ex 2 Ca;Mg ð21Þ 3 4 5 6 7 8 9 10 Cumulated NH3 volatiliztion (% of N) pH max 0 10 20 30 40 50 60 F=0.245exp(0.5858 pHmax)R2 = 0.607 Figure 8. Cumulative NH3 volatilization from urea and mineral fertilizers related to maximum pH after soil amendment (adapted from Whitehead and Raistrick, 1990; Sommer and Ersbøll, 1996). 574 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 It can be seen from Eqs. (20) and (21) that increase in cation exchange capacity (CEC) will increase NH4þ adsorbed to the soil (Avnimelech and Laher, 1977; McGarry et al., 1987; Whitehead and Raistrick, 1993). At low cation exchange capacity, adsorption of NH4þ will not affect NH3 emission; at CEC ca. 250 mmol kg21 the effect of absorption tends to be insignificant (O’Toole et al., 1985), whereas when CEC is above this level, NH3 volatilization is reduced significantly. Further, NH4þ does not exchange easily with Ca2þ, thus high Ca2þ concentrations in the soil may reduce the effect of a high soil CEC. This will be the case in calcareous soils, where high concentrations of exchangeable Ca2þ will reduce the fraction of NH4þ absorbed. The consequence of the ratio law is that increasing soil water content due to rain will change the equilibrium and the divalent cations in solution will be exchanged with NH4þ (Chung and Zasoski, 1994). Conversely, if the solution is concentrated by water being removed due to drying, NH4þ will exchange with divalent cations on the CEC. Thus, during a drying event after fertilizer application, the concentration of NH4þ in solution will not increase linearly with evaporation of water, and in consequence the exchange process will reduce the rate of NH3 emission expressed relative to initial TAN content per time unit and increase the duration of the period with significant emission rates; this effect is most pronounced in soil high in CEC, as can be deduced from the work of Fleisher et al. (1987). This retention to CEC during drying events may, in addition to the solid phase theory (see below), contribute to an explanation of why NH3 emission may be low from a soil that has been dried (Fenn and Kissel, 1976). Furthermore, Fenn and Kissel (1976) have shown that in very dry soils there is a physical adsorption of gaseous NH3 to soil. F. SOLID PHASE PROCESSES Ammonia losses from DAP-amended soil tend to be lowest from limed soils as compared to AS (Sommer and Ersbøll, 1996); the lower NH3 emission from DAP in Ca-rich soil has been ascribed to precipitation of calcium ammonium phosphate or magnesium ammonium phosphate (struvite) (Larsen and Gunary, 1962; Whitehead and Raistrick, 1990). The conditional stability constants for the solid phases of struvite may be calculated according to principles described in detail by Ringbom (1963) and Stumm and Morgan (1981); the reaction is as follows: MG2þ þ NHþ4 þ PO32 4 $MGNH4PO4ðsÞ ð22Þ Thus, in the microsites with fertilizers, precipitation of struvite is mainly controlled by Mg2þ, phosphate, and pH, as NH4þ is present in large amounts. AMMONIA EMISSION 575 ARTICLE IN PRESS 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 Struvite precipitates and is stabile at pH .6, and will consequently be present as stable solid phases above this pH level. Thus, struvite can be formed after addition of urea and ABC to most soils and in some soils after addition of acid fertilizers; the amount of struvite precipitation will, as mentioned above, be related to Mg2þ, phosphate, and pH. The precipitation of solid phases of ammonium will increase during a drying event, due to increasing concentrations of the ions in solution. Salts with hydrated water may in absence of “free” water, i.e., in a dry soil, donate a proton to NH3, transforming it into NH4þ (Evangelou, 1990). The loss of a proton produces a negative charge on the salt crystal, which facilitates the absorption of NH4þ. Thus besides the precipitation reactions proposed by Fenn et al. (1981), this mechanism may explain the reduction in NH3 emission from surface-applied ammoniacal fertilizers applied with Ca or Mg salts. G. CLIMATE AND INFILTRATION The effect of climate on NH3 volatilization from urea differs significantly from mineral fertilizers, because urea has to be hydrolyzed before volatilization will start, and hydrolysis is related to water availability and temperature. Thus after application of urea to a dry soil there will be no hydrolysis of the urea and no NH3 volatilization. A rain event immediately after surface application will dissolve the urea and, as urea movement lags only slightly behind the waterfront, the urea will infiltrate the soil (Fenn and Miyamoto, 1979). In the soil, urea is transformed to TAN, which will be absorbed by soil colloids; therefore, TAN is not as easily transported as urea (Black et al., 1987a). The effect of rain may therefore be variable and related to soil humidity and air dryness prior to the rain and also to the amount of water added during the rain event. More NH3 volatilizes after urea application on a wet soil than on a dry soil, because the humidity will initiate urea hydrolysis (Fenn and Miyamoto, 1979). An example of the interaction between application of urea and climate is shown in Fig. 9; in late March 1993 with no rain after application, ca. 9% of the N was lost, while only about 2% was lost in 1994 with a rain event of 13 mm immediately after application (Schjoerring and Mattsson, 2001). The 13 mm of rain in 1994 transported urea into the soil. The second round of urea application in late April resulted in a loss of 7–8% of the applied N in both years (Fig. 9). However, due to very dry conditions in 1993, the NH3 emission was delayed for more than 2 weeks until a light shower was received mid-May, accelerating the dissolution of the fertilizer granules and the hydrolysis of the urea. The total amount of NH3-N lost was 13 and 9 kg N ha21 in 1993 and 1994, respectively. The effect of convective and diffusive transport of urea in the soil is important when predicting NH3 emission. Diffusion may reflect changes in TAN 576 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 concentration and pH around the fertilizer granules and convection the gross transport. Complex models predicting diffusive and convective transport of incorporated urea have been developed by Rachhpal-Singh and Nye (1986a,b) and by Kirk and Nye (1991). These models indicate that convective transport into the soil as affected by rain and transport to the soil surface during drying conditions are important for the prediction of the magnitude of NH3 volatilization. Less than 10 mm of rain has little effect on the magnitude of volatilization (McInnes et al., 1986a; Ryden et al., 1987), and at rainfall greater than 10–16 mm, losses are reduced if urea remains in a non-hydrolyzed form (Black et al., 1987b) and no loss is determined after 20—25 mm of rain (Fenn and Miyamoto, 1979; Bouwmeester et al., 1985). For ammoniacal fertilizers, one may assume that about 20 mm of rain is sufficient to reduce NH3 volatilization significantly. Furthermore, transport of TAN into the soil may also be affected by soil humidity. Model simulations of Nye (1992) showed that emissions tended to increase when the moisture level was at about field capacity and also when soil was drier. At high soil water content, diffusion of TAN in the liquid phase increases with increasing soil water, and in dry soil diffusion of NH3 in the vapor phase increases with reduction in water content compensating for the reduced TAN diffusion in the liquid phase. The rate of NH3 volatilization from urea will be affected by temperature because both hydrolysis rate and NH3 transfer from the liquid to the atmosphere increase with increasing temperature (Black et al., 1985; McGarry et al., 1987). NH3 emission, kg NH3-Nha–1 season–1 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 August April May June July Winter wheat applied urea March 1994 1993 Figure 9. Cumulated NH3 emissions ^ SE from urea applied to growing winter wheat crops. Vertical arrows denote time of urea application. The reason for the low loss in 1994 was that 13 mm of rain fell immediately after the urea was applied so that the urea was transported into the soil. Very dry conditions in 1993 delayed the NH3 emission for more than two weeks until a light shower was received mid-May accelerating the dissolution of the fertiliser granules and the hydrolysis of the urea. AMMONIA EMISSION 577 ARTICLE IN PRESS 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 Thus, the lag phase will be shorter and the initial loss rates higher at high soil temperatures. Total losses may not be affected by changes in temperature, because volatilization will continue for a longer period at low than at high temperature (Fenn and Kissel, 1974; Harper et al., 1983); consequently total loss may be related to the potential of losses given by soil and other variables (McGarry et al., 1987). A low temperature in combination with rain may reduce losses significantly, especially because hydrolyses is slow and urea may infiltrate the soil during showers (Fenn and Miyamoto, 1979). Thus, Harper et al. (1983) concluded that the amount and distribution of rain after urea application appeared to control the total NH3 volatilization from urea application. Ammonium phosphate or sulfate, which may precipitate with calcium, has a different relation to temperature than AN, which will not precipitate (Fenn and Kissel, 1973). The precipitate-forming fertilizers showed no differences in volatilization due to change in temperature at high (.275 kg NH4-N ha21) and low (,66 kg NH4-N ha21) application rates. Volatilization rates increased from 26 to 45% of TAN with a temperature increase from 12 to 328C, when the precipitate-forming fertilizers were applied to a calcareous soil at application rates of 110 kg NH4-N ha21. Volatilization from AN increased from 14 to 45% of applied TAN with a temperature increase from 12 to 328C and was not affected by changes in application rates (Fenn and Kissel, 1974). These patterns may be related to interaction of soil buffer capacity and concentrations of TAN as mentioned above; i.e., in soils at high buffer capacities and low NH4þ concentrations, the soil buffer will control emission, and at high NH4þ concentration, emission is related to NH4þ, which equilibrium with NH3 is affected by the temperature. Wind will affect losses of NH3 from fertilizers; if humidity is high enough to keep the salts in solution, volatilization is expected to increase with wind (Eqs. (1)–(10)). In the few field studies using non-interfering micrometeorological techniques it has, however, been shown that wind cannot be identified as the most important variable and wind has only been shown to affect the volatilization from urea significantly during short periods (Harper et al. (1983), or the effect of wind speed has not been significant during the study (McInnes et al., 1986a). These experiments may back up the assumption that the interaction between wind and soil temperature may confound the effect of wind. H. MICROBIAL PROCESSES (NITRIFICATION/IMMOBILIZATION) Emission of NH3 from TAN supplied to the soil in form of urea or ammonium salts will compete with depletion of the TAN by microorganisms through immobilization to soil organic N or nitrification to NO3 2 (Malhi and McGill, 1982; Recous et al., 1992). The rate of TAN transformation by microorganisms 578 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 depends on a range of factors, including the population density of microorganisms, soil temperature, water and oxygen concentrations of the soil and inhibition due to high NH3 concentrations. This means that it is difficult to generalize the contribution of microbial activity to the emission pattern of NH3, but the following general assessment is based on the initial period of 0–14 days after application of ammoniacal fertilizers, i.e., in the period with significant NH3 emission from the fertilizer-treated soil. Microbial transformation of TAN to organic N (assimilation/immobilization) takes place immediately after TAN is dissolved in soil solution. Rate of assimilation is related to availability of carbon in the soil rhizosphere, TAN concentration, soil water content and temperature (Recous et al., 1988; 1999). Few studies have quantified the relationship between environment and immobilization; however, it is assumed that immobilization is less responsive to temperature than, for example, mineralization (Murphy et al., 2003). Immobilization has been reported to account for 5% of the TAN applied within the first 2–3 days after spreading and 8–10% after 14 days at an average soil temperature of 3.58C (Recous et al., 1988) and 3.5% during the initial 7 days at 238C (Sørensen, 2001). In humid sandy tropical soils, the rate of microbial immobilization was 12% of TAN applied during 50 days of incubation of urea or (NH4)2SO4 (Atwell et al., 2002), confirming that the response to temperature is not very significant and the assimilation by soil microorganisms is relatively low both in temperate and tropical soils. Furthermore, soil type seems not to affect immobilization, as similar immobilization rates have been reported for different soils in rotation after a wheat crop, probably because immobilization is limited by readily assimilable carbon available for the soil microflora (Recous et al., 1992). Nitrification of TAN mixed with soil is largely affected by soil pH, being negligible at soil pH values lower than ca. 4 and increasing linearly with pH increasing from 4 to 6.2 in a sandy soil (Winter and Eiland, 1996). Increasing TAN (substrate) concentrations increases nitrification until either salt effects or NH3 toxicity reduces nitrification (Malhi and McGill, 1982). Thus, at a level higher than 250–300 mg NH4þ-N g21 (soil), nitrification has been shown to be inhibited after mixing (NH4)2SO4 in soils having a pH of 5.8–7 (Malhi and McGill, 1982; Flowers and O’Callaghan, 1983); the inhibition was primarily a salt effect, as the NH3 fraction of TAN is low in soils at pH , 7. At high pH, the concentration of the NH3 component of TAN will increase and reach levels toxic to the microflora (Malhi and McGill, 1982). Nitrification can be expected at “permanent wilting point” and increases with increasing water content (Malhi and McGill, 1982; Flowers and Callaghan, 1983). Near water saturation, nitrification is absent due to a shortage of oxygen (Malhi and McGill, 1982). The nitrification rate increases linearly with soil water content (zero order process; Flowers and O’Callaghan (1983) and is ca. 2.8 times higher at 0.033 MPa than at 21.5 MPa, respectively, at field capacity and permanent wilting point (Malhi and McGill, 1982). AMMONIA EMISSION 579 ARTICLE IN PRESS 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 Temperature dependence of nitrification is different in warm or tropical soils and temperate soils (Fig. 10), however 15N pool dilution studies have shown that nitrification may be very rapid in most climatic regions (Watson et al., 2000; Watson et al., 2002 quoted in Murphy et al., 2002). Daily nitrification rates of 2.2–7.9% for an acid sandy soil and a sandy loam at 238C have been reported by Sørensen (2001) and 4–6% at 58C by Flowers and O’Callaghan (1983), at substrate concentrations of 50–100 mg NH4þ-N g21 (soil). These rates correspond to the nitrification rates calculated by Malhi and McGill (1982) using a simple first-order model with input of substrate concentration, temperature, and water content of the soil. A generally applicable model will need inclusion of the effects of pH and changes in the population of nitrifiers (Gilmour, 1984; Grant, 1994). Immobilization may cause reduction in NH3 emission, but the effect is considered to be negligible because the process is relatively slow compared with volatilization from applied mineral fertilizers, i.e., daily immobilization rates is 0.5–1.3% of the TAN or urea applied compared to NH3 emission rates of 5–10% immediately after hydrolysis of applied urea or dissolution of ammonium salts (see Fig. 4). The studies also indicate that immobilization of NO3 2 is very slow (i.e., about 5% of the immobilization rate of TAN; Mary et al. (1998)) and may therefore not be accounted for in NH3 emission studies. Contrary to immobilization, nitrification should be accounted for when interpreting NH3 emission experiments or when developing models to predict NH3 emission. Temperature, °C Nitrification rate, micog–1g(soil) d–1 0 10 20 30 40 50 60 0 1 2 3 4 5 2.5°C 25°C Concentration of TAN, microg TAN-N g–1(soil) 0 100 200 300 Soil pH 5.0 5.5 6.0 6.5 7.0 7.5 Soil I Soil II Soil III Figure 10. Left: Measurements of nitrification in Alberta soils from an area with a mean annual air temperature of 2.58C (Malhi and McGill, 1982) and North Australian soil with a mean annual air temperature of 258C (Myers, 1975). The figure is inspired by Malhi and McGill (1982). Right: Change in soil pH after completion of nitrification of (NH4)2SO2 mixed with the soil and no NH3 emission. Soil I: gray luvisol; Soil II: dark gray chernozemic, and Soil III: black chernozemic (Malhi and McGill, 1982). 580 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 Nitrification may affect NH3 emission through reduction in TAN and by reducing pH, as nitrification of 1 mol NH4þ produces 2 mol Hþ, according to the following equation: NHþ4 þ 2H2O$NO21 3 þ 2H3Oþ ð23Þ The study of Flowers and O’Callaghan (1983) showed that nitrification of 100 mg NH4þ-N g21 (soil) resulted in a reduction of 0.5 pH units and nitrification of 250 mg NH4þ-N g21 (soil) resulted in a reduction of 1 pH unit. Nitrification, therefore, reduces NH3 emission due to reduction in both concentration of TAN in soil solution and a reduction in the NH3 component of TAN. Figure 10 depicts the change in soil pH measured in the study of Malhi and McGill (1982), showing that the change in pH is related to amount of added TAN and differences in soil buffering capacity. The addition of fertilizers may cause a local change in soil pH, the magnitude of the change being related to the concentration of fertilizer, soil pH–buffer capacity, and fertilizer type. In consequence of the change in pH due to application of the fertilizers, the nitrification rates was in the following order urea . DAP . AS in the study of McInnes and Fillery (1989). IV. AMMONIA EMISSION FROM CROP FOLIAGE A. TRANSPORT OF NH3 BETWEEN LEAVES AND THE ATMOSPHERE In plants, the major source of NH3 is TAN dissolved in the water film in the mesophyll cell walls of leaves (the so-called apoplastic solution; Husted and Schjoerring, 1995). The concentration of TAN and Hþ is affected by uptake of N, translocation, and transformation of N, which varies with plant developmental stage, climate, and fertilization. The NH3 flux, FNH3 , between a single plant leaf and the atmosphere can be described as: FNH3 ¼ gleaf ðx 2 NH3;aÞ ð12Þ where gleaf is the conductance to diffusion of NH3 between the atmosphere and the interior of the leaf, and x is the NH3 concentration of the air in the substomatal cavities and intercellular air spaces within the leaf. Whether a leaf will act as a sink for or a source of atmospheric NH3 depends on the difference in internal and external NH3 concentration. If NH3,a exceeds x, NH3 will be absorbed, while in the opposite case emission will occur. When NH3,a equals x, no net NH3 flux occurs between the leaf and the atmosphere. The internal NH3 concentration at which FNH3 is zero is called the stomatal compensation point for NH3 AMMONIA EMISSION 581 ARTICLE IN PRESS 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 (Eq. (12); Farquhar et al., 1980; Husted et al., 1996). As evident from Eq. (12), the rate and direction of NH3 fluxes between plant leaves and the atmosphere at a given atmospheric NH3 concentration are controlled by the conductance to NH3 transfer and the internal NH3 concentration. The leaf resistance (inverse of conductance) to NH3 transfer includes a stomatal and cuticular resistance in parallel and usually also includes a boundary layer resistance in series with the two other terms. The latter is a function of the aerodynamic properties of the leaf, wind speed, and canopy turbulence, and will generally be an order of magnitude smaller than the maximum stomatal resistance. The cuticular resistance (Rcut) is extremely high for NH3, probably in the range of 2000–40,000 s m21 (van Hove et al., 1989). This means that hardly any NH3 will pass through the cuticle. However, NH3 can readily be deposited on the cuticular surface due to the presence of a surface water film. Since NH3 is highly soluble in water, moist leaf surfaces can act as a storage compartment for atmospheric NH3, making the leaf surface a temporary sink for NH3. B. MAGNITUDE OF NH3 LOSSES In general, NH3 is emitted from intensive agricultural ecosystems (Fig. 11), while semi-natural ecosystems act as NH3 sinks (Sutton et al., 1993a; Schjoerring et al., 1998). The wild understory plant, Luzula, and three native grass species showed low NH3 compensation points of 0.5–2 nmol mol21 (Hanstein et al., 1999; Hill et al., 2001). Crop plants such as oilseed rape or barley usually show 0 2 4 6 8 10 12 14 16 UK UK DK,1989-1990 DK, 1993-1994 Barley Wheat Oilseed rape Ammonia emission,kg NH3-Nha–1 DK, 1994 USA, 1985 DK,1994 DK,1993 DK,1993 Figure 11. Seasonal NH3 emissions from agricultural crops. Data from Harper et al. (1987); Mattsson and Schjoerring (2001); Sutton et al. (1993b); Schjoerring et al. (1993); Yamulki et al. (1996). 582 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 NH3 compensation points between 2 and 6 nmol mol21 (Husted and Schjoerring, 1995, 1996; Husted et al., 1996; Mattsson et al., 1997). In barley, the NH3 compensation point changed in relation to the developmental stage although the plants were grown under constant N limitation (Husted et al., 1996). In a field experiment in the Netherlands, intensively managed ryegrass showed NH3 compensation points varying over the season between 1 and 7 nmol mol21 (van Hove et al., 2002). For the same crop species grown in Scotland, Loubet et al. (2002) monitored NH3 compensation points ranging from 0.02 mg NH3 m23 in periods between fertilizations up to 10 mg NH3 m23 just after fertilizations. Under laboratory conditions, both ryegrass and Bromus erectus showed very high NH3 compensation points up to 18 nmol mol21 particularly when supplied with high levels of NH4þ to the growth medium (Mattsson and Schjoerring, 2002). Quantification of the NH3 exchange between the atmosphere and the canopy of barley, wheat and oilseed rape over two growing seasons show that the crop foliage is a net source of NH3 to the atmosphere, with emissions ranging between 1 and 5 kg NH3-N ha21 year21 (Fig. 11). Very high NH3 emissions of up to 15 kg NH3-N ha21 per season were reported for winter wheat in the United States by Harper et al. (1987). Harper et al. (1996) and Plantaz (1998) also measured high daily NH3 emissions from grassland in the Netherlands during spring and summer and from these emissions, high NH3 compensation points were derived. In contrast, based on measurements of stomatal NH3 compensation points by apoplastic bioassay, van Hove et al. (2002) concluded that plants in intensively managed grasslands in the Netherlands would not contribute to atmospheric NH3 loadings. For wheat, oilseed rape, and barley, the accumulated NH3 loss over a growing season constituted between 1 and 4% of the applied N, or between 1 and 4% of the total shoot N (Schjoerring and Mattsson, 2001). The loss increased under conditions with a high N concentration in the foliage and was positively correlated with the above-ground crop N content at anthesis, but not with that at final maturity. There were no indications that NH3 emissions were larger under conditions unfavorable for nitrogen remobilization from vegetative plant parts (low N harvest index). Nevertheless, a distinct peak in NH3 emission occurred during senescence. NH3 emissions from plant stands, measured under simulated environmental conditions in wind tunnels, ranged between 0.8 and 1.4% of the N content of the shoot, equivalent to 1.1–2.9 kg NH3-N ha21 (Mannheim et al., 1997). The highest emissions were observed in faba beans, whereas the emissions in winter wheat, spring rape, and white mustard were lower. The total NH3 emissions were not affected by removing a part of the ears (sink reduction), but emissions occurred earlier, as did the plant senescence. This suggests that the NH3 emissions are closely related to senescence (Schjoerring et al., 1993; Mannheim et al., 1997). Emission of NH3 has also been suggested to contribute to the decline in shoot nitrogen content that is often observed in agricultural crops AMMONIA EMISSION 583 ARTICLE IN PRESS 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 during the generative growth stage (Wetselaar and Farquhar, 1980; Schjoerring et al., 1989; Francis et al., 1993). In this context, it is important to emphasize that assessments of NH3 losses based on measurements of changes in 15N-labeled or total above-ground N are indirect and other pathways of nitrogen loss may influence the results, resulting in overestimation of losses (see below in Section VI on Measurement techniques). It can be concluded that plant communities on agricultural cropland represent a net source of NH3 to the atmosphere. Net emissions range from below 1 up to 5 kgNH3-N ha21 per season, depending on plant species, crop nitrogen economy status, and climatic conditions. Crops will in many areas represent a significant input of NH3 to the atmosphere and NH3 losses may become large enough to significantly affect crop N budgets. C. PHYSIOLOGICAL PROCESSES INVOLVED IN NH3 EMISSION FROM CROPS In addition to being taken up directly from the soil, NH3/NH4þ is produced in plant tissues by a number of different metabolic processes. For example, NH4þ is produced via nitrate reduction, through the fixation of atmospheric nitrogen by root nodules, by photorespiration in leaves, and through the phenyl propanoid pathway (Hirel and Lea, 2001). Ammonium may also be released during reassimilation of nitrogen transport compounds (e.g., asparagine, glutamine, arginine, and ureides) and through breakdown of other N compounds during senescence and remobilization. These processes take place in different cell organelles and in different tissues leading to both spatial and temporal variation in tissue NH4þ concentrations. It is therefore difficult to predict the proportion of the NH4þ present in plant leaves that at any stage will be contributing to the emission of NH3. A further complicating factor is that environmental conditions, particularly nitrogen supply, also affect tissue NH4þ levels (see below). 1. NH3 Compensation Point The major route for NH3 exchange between plants and the atmosphere is through the leaf stomates. In order for NH3 to be emitted through the stomates, the concentration inside the leaf (i.e., compensation point) has to be higher than the ambient concentration (see Section IV.A). The leaf apoplast constitutes the interface between the atmosphere and the living leaf tissue, and the NH4þ concentration in the liquid phase of the apoplast (the water contained within the cell walls) is therefore a critical parameter in determining the gaseous NH3 concentration inside the leaf. The stomatal NH3 compensation point can be 584 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 calculated from measurements of the pH and NH4þ concentration in the apoplastic solution (Eqs. (3) and (4); Husted and Schjoerring, 1995; Mattsson and Schjoerring, 2002). Apoplastic NH4þ concentrations normally range between 0.01 and 2 mM but have been shown to be highly dynamic and closely coupled to the plant N metabolism (Mattsson et al., 1998; Nielsen and Schjoerring, 1998; Mattsson and Schjoerring, 2002). 2. N Uptake and Translocation There are different pathways for NH4þ to reach the apoplastic solution; either through xylem transport of NH4þ from the root or through NH3 efflux from the mesophyll cells. Xylem transport of NH4þ can be quite substantial, particularly if the plants are grown on high levels of nitrogen. The concentration of NH4þ in xylem sap increases with increasing supply of both NH4þ (Mattsson et al., 1998) and NO3 2 (Husted et al., 2000a). These increasing xylem sap NH4þ concentrations are usually reflected in the apoplastic NH4þ concentration (Mattsson et al., 1998; Finnemann and Schjoerring, 1999; Husted et al., 2000a). Inside the leaf cells, NH4þ can either be assimilated by cytosolic glutamine synthetase (GS1) or taken up into the chloroplasts and assimilated by the chloroplastic form of the enzyme (GS2). Depending on the capacity of these enzymes to assimilate NH4þ into glutamine and then by the enzyme glutamate synthase (GOGAT) to convert glutamine into glutamate, NH4þ may accumulate in the leaves. In general, high NH4þ concentrations in xylem sap and apoplast also result in a high leaf tissue extract NH4þ concentration (Mattsson et al., 1998), but there are also examples where leaf tissue NH4þ did not rise despite increasing xylem and apoplast concentrations (Husted et al., 2000a). In grasses grown with either NO3 2 or NH4þ, significant correlations between leaf tissue NH4þ concentration and both NH3 emission and apoplastic NH4þ concentration were observed (Mattsson and Schjoerring, 2002). Decreasing the activity of GS, either by using an inhibitor such as MSX or by using mutants or transgenic plants with lower assimilation capacity, usually results in increasing concentrations of NH4þ in various plant compartments. In consequence, NH3 emission to the atmosphere increases within a few hours after adding MSX to the growth medium in both barley and oilseed rape, because GS activity is decreased and tissue TAN concentrations increase (Husted and Schjoerring, 1995; Mattsson and Schjoerring, 1996). In barley mutants with reduced activity of chloroplastic GS, higher leaf tissue and apoplastic NH4þ concentrations have resulted in higher NH3 emission compared with wild-type plants (Mattsson et al., 1997). Emission of NH3 also seems to increase more with increasing temperatures in the mutants than in the wild-type plants, suggesting a higher sensitivity to photorespiration in GS mutants. A massive release of AMMONIA EMISSION 585 ARTICLE IN PRESS 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 NH3/NH4þ takes place during photorespiration, particularly at high temperatures (Leegood et al., 1995), making it extremely important for the plant to have an efficient reassimilation of this NH3. The influence of photorespiration on apoplastic and leaf tissue NH4þ concentrations was investigated in antisense GS2 oilseed rape plants, i.e., plants with reduced activity of the key enzyme responsible for re-assimilation of photorespiratory NH4þ wild-type plants (Husted et al., 2002). Despite a 50–75% lower in vitro leaf GS activity in the antisense plants, there was no tendency for these plants to have higher tissue NH4þ concentrations than wild-type plants. Antisense plants exposed to leaf temperatures increasing from 14 to 278C or to a three-fold increase in the O2/CO2 ratio did not show any change in steady state leaf tissue NH4þ concentration or in NH3 emission to the atmosphere. These results show that oilseed rape has a large surplus of GS in the leaves, which makes the plants less sensitive to increasing photorespiration than is the case for barley (Mattsson et al., 1997). The influence of N2 fixation on the NH3 emission potential has been investigated in white clover (Herrmann et al., 2002). Symbiotic N2 fixation and mineral N acquisition were shown to be well balanced with respect to the apoplast and plant tissue NH4þ concentrations, leading to equal NH3 compensation points in plants grown with or without an external NO3 2 supply. Field measurements over pea crops similarly indicate NH3 emissions comparable to those for barley and wheat (Schjoerring and Mattsson, 2001). Senescing plant material undergoes a sequence of biochemical and physiological events including protein degradation, which eventually leads to cell death. The massive release of NH3 in senescing material is thought to be a consequence of amino acid deamination and catabolism of nucleic acids (Brouquisse et al., 2001). Dark-induced senescence of barley plants and detached leaves of oilseed rape both showed increased NH3 emission which was synchronized with chlorophyll degradation and liberation of NH4þ in the leaf tissue (Schjoerring et al., 1998). V. MANAGEMENT STRATEGIES Fertilizers are spread mainly as granules or prills. The fertilizers are granulated and prilled because this reduces inhomogeneity and because these fertilizers can be spread evenly. Furthermore, additives mixed with the fertilizers can produce slow-release fertilizers or reduce transformations after application as, for example, with urea. Fertilizers are either broadcast onto the soil surface or injected into the soil. AA is usually injected or can be dissolved in irrigation water, and urea is sometimes incorporated for the purpose of reducing NH3 emission. In Asia, farmers broadcast nitrogen fertilizer (mainly urea) into the water in rice paddy fields. 586 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 A. TECHNIQUES FOR REDUCTION OF NH3 EMISSION Top dressing and partly covering urea may reduce emissions significantly (Fig. 12). The effect of placing urea is related to the depth and soil characteristics (Fenn and Kissel, 1976). Placing urea in a soil low in CEC and with low pH buffering capacity may create a zone of high concentrations of dissolved TAN and a high pH due to hydrolysis. In consequence, the concentration of NH3 may be very high in the zone affected by urea. In these soils, shallow placement of urea may have little effect on NH3 emission (Fig. 12) because the NH3 will be transported by diffusion to the surface and be lost (Blaise et al., 1996). Thus there is a high correlation between amount of urea applied and placement depth and emission, so at high application rates urea should be placed at greater depths than at low application rates. To eliminate NH3 emission from urea applied to calcareous soils, the fertilizer may need to be placed as deep as 5–7.5 cm (Fenn and Miyamoto, 1981; Ismail et al., 1991). Harrowing stubble before urea application may halve NH3 volatilization, because cultivation forms cracks and small hollows where urea prills will be protected from volatilization, and rain events or irrigation may leach urea into the soil (Bacon et al., 1986; Bacon and Freney 1989). Mixing ammoniacal fertilizers with the soil may be a less efficient reduction measure than injection to the same depth because a part of the mixed-in fertilizer will be close to the surface and TAN will be transported either by diffusion or convection upwards and be lost (Nye, 1992). Increasing the application rate may reduce the relative emission of NH3 from urea and ammonium fertilizers applied to calcareous soils (Du Preez and Burger, 1988). On acidic soils the proportion of urea-N lost due to NH3 emission has been Ammonia emission, NH3 pct. of applied N 0 5 10 15 20 25 30 35 Wind tunnel Dynamic chamber Broadcast Broadcast Placed 2.5 cm Placed 1 cm Placed 2 cm Mixed 2 cm Mixed 3 cm Figure 12. Ammonia emission from urea broadcast or placed at different depth to a bare soil (Bouwmeester et al., 1985; Nye, 1992). AMMONIA EMISSION 587 ARTICLE IN PRESS 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 shown to increase by increasing the application rate (Black et al., 1987; Watson and Kilpatrick, 1991). This discrepancy is due to interaction of fertilizer alkalinity and soil acidity. In acid soils, the pH buffer may prevent high increases in pH after addition of urea at low rates, whereas pH may increase in acid soils amended with urea at high rates because the amount of alkaline hydrolysis products will be higher than microsite acidity of the soil. Similarly, surfaceapplication of urea in bands may increase NH3 emission compared to broadspreading when applied to acid soils. Although, absorption of NH4þ may affect transport of TAN in the soil, model simulations of Nye (1992) have shown that soil pH buffer capacity is more important in influencing NH3 emission. Application of pelleted fertilizers or fertilizers in solution may contribute to local “hot spots” with high salt concentrations and pH different from the surrounding soil. In a hot spot with urea, the pH will be high compared with the pH of the surrounding soil, and in a hot spot with “acid” ammonium salts pH will be low. Immediately after band application the high urea concentrations may reduce urease activity due to substrate inhibition (Fig. 5) and as the uncharged urea may infiltrate more easily to greater depths than NH4þ this may result in reduced emission of NH3 (Fenn and Miyamoto, 1979, Bouwmeester et al., 1985). Straw has a much higher pH buffer capacity and pH than soil and a 20-times higher urease activity than the surface 10 mm of soil; consequently broadcasting urea to straw spread on the soil may contribute to a high potential of NH3 emission (McInnes et al., 1986b). Similar, applying urea to trash left from harvesting of sugarcane may cause significant losses of NH3 (Fig. 13) because the urea is not in Ammonia emisson, NH3 pct. of applied N 0 10 20 30 40 UAN Urea Urea Fine silt loam Fine silt loam Coarse silt Calcareous Disced-scarified Trash - heavy rain Trash - dry Trash W. wheat seeding, Z 05 W. wheat, tillering, Z 14.22 W. wheat, ear initiation, Z 30 Banana plantation AS Urea Urea Figure 13. Ammonia emission from UAN (urea ammonium nitrate), urea and AS (ammonium sulphate) applied to bare soil, sugarcane trash-covered soil, to winter wheat, and banana plantation. (Adapted from McInnes et al., 1986a,b; Bouwmeester et al., 1985; Bacon and Freney, 1989; Freney et al., 1992; Prasertsak et al., 2001.) 588 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 contact with soil and the trash has a high pH. Moving the litter aside on no-till soil and applying urea to the soil surface may reduce losses significantly (Touchton and Hargrove, 1982). B. FERTILIZER COMPOSITION Ammonia emission from fertilized soils has been reported to vary between negligible and 40% when measured with the micrometeorological mass balance technique (Fig. 14; Freney et al., 1992; McInnes et al., 1986a). The emission from a specific fertilizer will vary with climate, soil pH–buffer capacity, and CEC; therefore the average values shown in Figs. 12–14 cover considerable variation, as indicated by the relatively high standard variation. Ammonia emission is affected by the choice of fertilizer and rate of application. Fertilizers applied to calcareous or limed soils should preferably be acidic fertilizers with anions not forming calcium precipitates, e.g., AN or NH4Cl. Soils low in Ca may be fertilized with MAP, AS, or AN, which will reduce pH upon dissolution in soil water. Surface application of urea to acidic soils with a high pH–buffer capacity or a high CEC may not cause high losses. Urea should not be surface-applied to calcareous soils because TAN is at risk of being lost due to NH3 emission; therefore, incorporation would be recommended. Urea 0 10 20 30 40 DAP CAN AS Ammonia emission, NH3-N pct. of applied N Figure 14. Ammonia emission from ammoniacal fertilizers, i.e., urea, diammoniaum phosphate (DAP), calcium ammonium nitrate (CAN) and ammonium sulfate (AS) broadcast to crops, measured with wind tunnels (Sommer and Jensen, 1994; Velthof et al., 1990; van der Weerden and Jarvis, 1997). AMMONIA EMISSION 589 ARTICLE IN PRESS 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 An example of adapting fertilizer application to soil and crop is banana plantations in a tropical area in Queensland (Australia), where 17% of the applied urea-N may be lost due to emission of NH3 (Prasertsak et al., 2001). The fertilizers cannot be incorporated because this will damage the roots; therefore, Australian banana producers are recommended to switch from urea to AN, with a loss potential lower than that of urea (Prasertsak et al., 2001). On grassland, incorporation of solid fertilizers may also be detrimental to the plants, and on acid soils application of liquid urea ammonium nitrate instead of granular urea may reduce losses because the extent of microsites with high TAN concentrations is reduced. Lightner et al. (1990) showed that applying urea in liquid solution may reduce emission, with 35% compared with granular-applied urea. Application of inorganic salts with fertilizers may reduce NH3 emission significantly. The salts used should be soluble, thus CaSO4 is not efficient for reducing NH3 emission. Adding CaCl2 with urea will contribute to reduction in NH3 emission, because CO3 22 will precipitate as CaCO3 and Cl2 and NH4þ will reduce pH significantly as NH3 is lost. KNO3 or KCl may be used with urea to reduce NH3 emission, the Cl2 may contribute to produce an acid environment, reducing the NH3 emission potential. In addition, Kþ cation may exchange with absorbed Ca2þ, which will precipitate the CO3 22 produced during hydrolysis (Fenn et al., 1982). Sulfur applied with urea will hydrolyze to SO4 22 and reduce NH3 emission, but the effect may not be very significant in all environments, i.e., NH3 emission from sulfur-coated urea was equal to or somewhat lower than from untreated urea (10.1% as compared with 12.6%) in the study of Black et al. (1985), and natural finely ground sediments of pyrite (FeS) have been shown to reduce emission from urea by 54% (Blaise et al., 1996). Thiosulfate at 10% reduces hydrolysis rates and, under drying conditions, CaCl2 may also reduce hydrolysis (Black et al., 1985); the reduction is due to the low pH of the Cl2 solution (Fig. 5). In dry conditions, Cl concentration will be high due to slow diffusion and the pH reduction will be significant (Sloan and Anderson, 1995). Addition of H2SO4 or H3PO4 with urea to calcareous soils is questionable due to precipitation of the anions with calcium. In calcareous soils, adding acids that precipitate should, therefore, be avoided and either HCl or HNO3 may be used with urea to reduce losses of NH3 (Fenn and Richards, 1986). However, Fenn and Hossner (1985) do not recommend the use of these acids because urea nitric acid is a cold explosive in solid form and HCl may be noxious and injurious to plants. Furthermore, adding pyrite and H2SO4 may contribute an increase in the soil’s demand for lime, which will be an additional cost for the farmer. Urease inhibitors delaying hydrolysis of urea to TAN may be an option for reducing NH3 emission. Slowing down the hydrolysis allows time for urea to infiltrate into the soil by diffusion or convection after a rain event, because the uncharged urea is more readily transported in soil than the charged NH4þ (Fenn and Miyamoto, 1981). Of the inhibitors tested, the phosphoryl di- and 590 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 triamides, (N-(n-butyl)thiophosphorictriamide (NBPT) or phenylphosphorodiamidate (PPD) were found to be the most useful, and have proven effective to varying degrees depending on the environment and management (Byrnes and Freney, 1995). The efficiency is related to climate and soil conditions; i.e., applying the urea with urease inhibitor to a humid soil during a period with little rain may not reduce NH3 emission, whereas application followed by a heavy rainfall few days after application may prove very efficient. To become efficient, NBPT must be converted to the oxygen analogue (N-(n-butyl)phosphorictriamide, NBPTO); on the other hand, PPD is efficient immediately but is decomposed (departs from neutrality) rapidly (Bremner, 1995). In waterlogged soil or in floodwater in rice fields, the oxidation of NBPT applied with urea will be retarded and the inhibition of urea hydrolysis will be inefficient (Bremner, 1995). Therefore, Phongpan et al. (1995) applied urea with PPD and NBPT to rice fields. It appeared that initially PPD inhibited urease activity, and during this time at least part of the NBPT was converted to NBPTO; then as the activity of PPD declined, NBPTO inhibited the hydrolysis of urea. This combined urease inhibitor treatment reduced NH3 emission from 15 to 3% of the applied N. Hydroquinone is a less effective urease inhibitor than the phosphoramides but may have beneficial physiological effects on the plant (Bremner, 1995). Due its low cost, hydroquinone has been tested and is formulated with urea for use in China. The concentration of a chemical inhibitor required to suppress hydrolysis decreases with increasing granule size of the fertilizer, as seen with nitrification inhibitors (Singh et al., 1994). Thus, increasing granule size of the urea fertilizer may delay hydrolysis significantly because the contact to soil is reduced. Black et al. (1987a) reported that urea concentration was significant for 5 days when applying 8 mm granules, compared with 2 days after application of powder or ,4 mm granules. The pH in the granules increased to values between 8 and 9 within 2 days in both the small and large granules. Consequently, urea was hydrolyzed into a microsite with high pH and high TAN concentration for a longer period in the large granule than in small granules, causing the cumulated emission to be large. This was confirmed by the study of Watson and Kilpatrick (1991), who found little difference in NH3 emissions between different sized prills of applied urea. C. FLOODED FIELDS (RICE PADDIES) The fertilizer application to rice paddies is highlighted, because rice is the world’s most important food crop and 75% of the world’s rice is grown in several inches of water held in small dikes, i.e., paddies (Maclean, 1997). Management of the rice paddies may briefly be characterized by flooding of the dry paddy, puddling, transplanting of rice, vegetative and reproductive phases of rice, and AMMONIA EMISSION 591 ARTICLE IN PRESS 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 harvesting. Nitrogen uptake efficiencies range from 20 to 60% of applied N (Maclean, 1997). This low efficiency reflects poor management and the use of urea-N, leading to high NH3 emissions. Thus, reducing NH3 emission may act to optimize N-use efficiency. The water in paddies is buffered by bicarbonate (HCO3 2), which is the main contributor to the alkalinity of the water. Therefore, consumption of CO2 during daytime and respiration of CO2 during the night will cause large fluctuations in the pH of the water in rice paddies with a high photosynthetic biomass, especially algae (Fillery et al., 1986; Bowmer and Muirhead, 1987). The uptake of CO2 will increase pH during the daytime and respiration of CO2 will reduce pH at night. The diurnal variation in NH3 emission clearly reflects the variation in pH (Fillery et al., 1986). Alkalinity may be significantly higher in paddies irrigated with floodwater or alkaline well water compared with paddies irrigated with rainwater (Vlek and Crasswell, 1981; Fillery et al., 1984). The alkalinity and pH of the water will be affected by the applied fertilizers; thus (NH4)2SO4 decreased soil water pH fluctuation initially, but after 2 days pH fluctuations were similar to variations in pH of floodwater fertilized with urea in the study by Fillery et al. (1984). This indicates that after a few days, the HCO3 2 of the water, including CO2 from respiration, had buffered the pH to the initial values. In contrast, emission of NH3 from (NH4)2SO4 applied to water with low alkalinity would have resulted in a reduction in pH, which eventually would have reduced emission significantly (Vlek and Crasswell, 1981). Hydrolyzed urea will contribute to the alkalinity and may increase pH to 10 in weakly buffered water (Mikkelsen et al., 1978), and the NH3 emission may be significant. Removing the algae biomass or reducing the photosynthetic activity will reduce the diurnal variation in pH of the water, and the addition of a suitable photosynthetic inhibitor (terbutryne) with fertilizer may reduce the daytime pH of rice floodwater for up to 6 days and the potential NH3 emission by 43% (Bowmer and Muirhead, 1987). Adding urease inhibitors will delay hydrolysis of urea and reduce the concentration of TAN in the water, thereby reducing emission of NH3 (Fillery et al., 1986). The effect of urease inhibitors is variable and related to water chemistry. Mixing of the water due to wind and changes in buoyancy due to heating and cooling of surface water will vary during the day. Stratification of subsurface water may reduce emission in periods with heating of surface water and little mixing caused by wind. Radiation during the daytime will heat the surface water, which will enhance stratification, and cooling in the afternoon will lead to enhanced mixing (Leuning et al., 1984). In consequence, mixing may contribute to the high NH3 emission rates in the afternoon. Application of acidic fertilizers will reduce NH3 emission. Fig. 15 shows that emission of NH3 from urea may be higher than emission from (NH4)2SO4 when applied at transplanting. In Fig. 15, data from different studies have been used and comparing results from different studies may bias conclusions due to the effect of 592 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 different environmental conditions (e.g., effect of alkalinity, wind, temperature, etc.), but the experiments were carried out on similar soils and wind fluctuations were similar in the studies of Freney et al. (1981) and De Datta et al. (1991). Paddy rice in the tropics is fertilized with urea, which may be applied to the rice field before transplanting rice, to the rice at transplanting, to the floodwater ca. 4 weeks after sowing and at panicle initiation or booting (De Datta et al., 1989, 1991; Son and Buresh, 1994). In the study of De Datta et al. (1991), 46–54% of the urea-N applied immediately after flooding was emitted as NH3 due to the low buffer capacity of floodwater and a high pH after hydrolysis of the urea. Incorporating the first application of urea into the dry soil before flooding may reduce NH3 emission to about 10% compared with the emission of 46–54% of urea-N after broadspreading of the urea to the rice field after flooding (Fig. 15). This soil had a pH of 4.6 and a CEC of 250 mmol kg21 and because TAN was absorbed on soil colloids as NH4þ, the potential of NH3 emission from the incorporated urea was low. Incorporation of urea in flooded soils has little effect on NH3 emission from urea but significantly reduces the emission from (NH4)2SO4; probably urea incorporated in flooded soils is not adequately in contact with the soil to reduce emission (Fig. 15). In addition to incorporating urea before flooding rice fields, NH3 emission from paddy fields may also be reduced by matching the application to crop demand by split applications of urea to the floodwater. Urea application to rice may be delayed 14–16 days after transplanting in the period of vigorous growth and assimilation of the nitrogen applied (Son and Buresh, 1994). The plant uptake of N will thereby reduce NH3 emission and contribute to a higher fertilizer At transplanting, IRRI At transplanting, IRRI Incorporation in water, IRRI NH3 emission, % of applied N Incorporation in water, IRRI Incorporation dry soil, IRRI At transplanting, NWS Panicle, IRRI Urea 0 10 20 30 40 50 60 (NH4)2SO4 Panicle, IRRI Figure 15. Ammonia emission from rice paddies, measured with the micrometeorological mass balance technique at either IRRI (Philippines) or in New South Wales (Australia). (From Leuning et al., 1984; De Datta et al., 1989, 1991; Fillery and De Datta,1986; Fillery et al., 1986; Freney et al., 1981.) AMMONIA EMISSION 593 ARTICLE IN PRESS 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 efficiency of the applied urea-N. Little NH3 is emitted from urea applied to the rice field at panicle initiation or booting, because rice is rapidly assimilating the nitrogen, and shading of the water reduces temperature and wind speed (Freney et al., 1981; De Datta et al., 1989). Furthermore, shading of the water reduces pH increase due to respiration of algae. D. INJECTION OF ANHYDROUS AMMONIA The emission of NH3 from injected AA will be related to application rate, depth of injection, knife spacing, and soil buffering capacity (Izaurralde et al., 1990). Penetration depth of NH3 increases with decreasing soil water content (Blue and Eno, 1954;McDowell and Smith, 1958). Thus, fromAAinjected into a dry soil, the NH3 emission was 20% because NH3 retention capacity was low and a part of the injected NH3 could move through the air-filled pore space to the soil surface (Sommer and Christensen, 1992). On the other hand, the crevice left after injection into a wet soil was left open and the distance of NH3 penetrated into the soil to the open crevice was short and emission was as high as50% (Sommer and Christensen, 1992). Hopkins et al. (1963) showed that injection ofAAmay push water in the soil ahead of the gas and thereby reduce gas movement in water-filled soils, which will result in high TAN concentrations in soil water near the site of injection. Anhydrous ammonia is usually injected into the soil after winter or a rainy season. The injection will take place when soils are moist, as driving on wet soils is either impossible or will compact the soil. Under these conditions, little NH3 will be emitted from AA injected to depths below 10 cm (Baker et al., 1959; Denmead et al., 1977; Sommer and Christensen, 1992), because the furrow will close and soil water will absorb the NH3. Injection may enhance production of nitrous oxide (N2O) and NO, and reduction of NH3 emission from AA (280 kg N ha21) to 0.1% by injection to 25 cm may increase emission of N2O and NO to, respectively, 3.9 and 11.3% of the applied N (Jambert et al., 1997). Injection depth of AA where little or no NH3 emission will take place may be predicted using TAN-transport models with input of amount of applied AA, soil pH buffer capacity, diffusion as affected by soil water, and CEC (Izaurralde et al., 1990). The model calculations have shown, for a fine sandy loam and a silt loam soil, that emission is low at 15 cm injection and significantly higher (8%) at 5 cm injection depth. E. CROP NH3 EMISSIONS AS AFFECTED BY FERTILIZER APPLICATION Volatile NH3 losses from crops depend on seasonal variations in climatic conditions affecting the growth and nitrogen economy of the crops 594 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 (Schjoerring and Mattsson, 2001). In general, losses are expected to increase with the N concentration of the foliage and to the extent that this is controlled by other growth factors than fertilizer N availability; there may be no connection between NH3 emission and fertilizer application. As an example from a field study by Schjoerring and Mattsson (2001), lower NH3 emissions were measured from wheat plants fertilized according to optimum N-recommendations as compared to plants applied a reduced amount of N-fertilizer (75% of optimum), because the latter plants had a higher shoot N concentration during part of the growing season. In the same study, the accumulated NH3 loss over a growing season was positively correlated with the above-ground crop N content at anthesis, but not with that at final maturity, and there were no indications that NH3 emissions were larger under conditions unfavorable for nitrogen remobilization from vegetative plant parts (low N harvest index). Enhanced NH3 emission from leaves under conditions with excessive N absorption by roots has been observed in several laboratory experiments (see e.g., Mattsson and Schjoerring, 2002). A similar relationship seems also to be valid under field conditions; thus, as shown in Fig. 16, application of the acid fertilizers DAP and AS to a grass ley or young wheat plants in April resulted in a peak of NH3 emission between 5 and 12 days after fertilizer application. A similar peak was not observed when the same fertilizers were applied in early spring (March), i.e., before significant plant uptake of N, where the NH3 emission continuously 0 5 10 15 20 DAP, April DAP, March AS, April AS, March 0.5 0.4 0.3 0.2 0.1 0.0 Days from application of fertilizers Ammonia emission, g NH3 m–2 day–1 Figure 16. Wind tunnel measurements of NH3 emission from DAP and AS applied to grass (10– 15 cm) on 9 March 1992 and to winter wheat (5 cm) on 1 April 1992. In the March experiment, air temperature was 4.38C, soil temperature 5.78C and wind speed in the wind tunnel 4.2 m s21. In the April experiment, air temperature was 5.88C, soil temperature 7.48C, global radiation 10.9 MJ m22 and wind speed in the wind tunnel 4.2 m s21. In the April experiment, an atypical peak in emission was measured from 5 to 13 days after fertilizer application; this peak is attributed to NH3 emission from the weak seedlings not having capacity to transform NH4þ to amide/amines due to low global radiation. AMMONIA EMISSION 595 ARTICLE IN PRESS 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 declined following high initial loss rates. The extra peak in April is assumed to represent NH3 emission from the seedlings, following rapid nitrogen uptake in a period with low global radiation and, thus, a low photosynthetic activity. F. AMMONIA EMISSION FROM DECOMPOSING PLANT MATERIAL Decomposing crop residues may constitute a significant source of atmospheric NH3. In measurements by Mannheim et al. (1997), NH3 emissions from decomposing sugar beet leaves, potato tops, and field-bean straw ranged from 0.9 to 3.7% of the N content. The highest emissions, reaching from 8.6 up to 12.6 kg N ha21, occurred from sugar beet leaves and potato shoots with high water content, whereas the emissions from field-bean straw with high dry matter and N content were relatively low (3.1 kg N ha21 or 0.9% of the N content). The NH3 emission from sugar beet leaves was reduced 81% by plowing and 63% by mulching (Mannheim et al., 1997). A marked influence of moisture on the NH3 volatilization during decomposition was also observed for ryegrass herbage, where 20–47% of the herbage N was lost over a 70 day period if kept moist, while volatilization was less than 1% of herbage N during drying (Whitehead et al., 1988). The quantity of NH3 lost increased with herbage N concentration and temperature during decomposition (Whitehead et al., 1988). As an example, herbage containing 3% N on a dry matter basis lost 10% of its N through NH3 volatilization over a period of 28 days, whereas no volatilization was detected from herbage containing 0.9% N. The profound increase in NH3 volatilization with moisture and nitrogen content of plant residues, as well as with temperature, is consistent with the stimulating influence of these parameters on protein degradation and liberation of NH3/NH4þ in senescing leaves. An example of the influence of nitrogen status is shown in Fig. 17, where bulk leaf tissue NH4þ concentrations increased more during senescence of high-N leaves compared with low-N leaves, particularly when plants had been growing at relatively low light intensity, resulting in reduced C/N ratio relative to high-light grown plants (Fig. 17). The ratio between [NH4þ] and [Hþ], which reflects the NH3 emission potential from leaf litter, also increased dramatically during senescence (Fig. 17). Significant amounts of N may be lost from the plant by senescent leaves falling off and decaying on the ground. Husted et al. (2000b), and Nemitz et al. (2000) showed that in oilseed rape decomposing plant residues on the soil surface made a large contribution to NH3 emission. Some of the NH3 emitted from these leaves might be reabsorbed by leaves still attached to the stem (Nemitz et al., 2000). The soil has also been suggested as an NH3 emission source. However, only a small amount of NH3 emission from the soil was found and it was proposed that the soil acts mostly as a sink for NH3 (Neftel et al., 1998). 596 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 10 20 30 40 A 0 10000 20000 30000 40000 50000 60000 70000 0 B High light 0N High light 3N High light 6N Low light 0N Low light 3N Low light 6N day 0 day 2 day 4 Days of senescence Bulk tissue[NH4 +]/[H+]ratio Bulk tissueNH4 +,μmolg–1 tissuewater Figure 17. (A) Bulk tissue extract NH4þ concentrations and (B) NH3 emission potential as expressed by the [NH4þ]/[Hþ] ratio in leaves of Lolium perenne grown for 4 weeks in 0, 3, or 6 mM NO3 2 under high and low light conditions on the day of leaf excision and after 2 and 4 days of senescence in darkness. C/N values in low-light grown plants were 52, 12, and 10 for plants grown in 0, 3, or 6 mM NO3 2, respectively. The corresponding values for high-light grown plants were 63, 24, and 23. Values are means ^ SE for three replicates. (M. Mattsson and J. K. Schjoerring, unpublished results.) AMMONIA EMISSION 597 ARTICLE IN PRESS 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 G. ABSORPTION BY CROPS Plants are capable of absorbing atmospheric NH3 when the NH3 concentration in the atmosphere exceeds that in the substomatal cavities. The absorption responds linearly to ambient NH3 concentration over a very broad range even up to around 500 nmol mol21 (van Hove et al., 1987; Whitehead and Lockyer, 1987). From urea applied to a grass pasture, the NH3 volatilization decreased with increasing canopy density (Hoult and McGarity, 1987; Ping et al., 2000). The crop may have changed the microclimate and the canopy may have absorbed NH3 volatilized from the urea, thereby reducing the flux of NH3 from the soil– plant system as shown in Fig. 13 for urea applied to winter wheat at increasing physiological age. The maize canopy reduced NH3 emission from NH3 dissolved in irrigation water (Denmead et al., 1982), due to a reduction in wind speed, reduction of temperature and uptake of NH3 by the maize canopy. In oilseed rape, significant amounts of N may be lost before flowering in dropped leaves (Schjoerring et al., 1995), followed by significant NH3 emissions from the decaying leaves (Sutton et al., 2000). However, a significant part of the emitted NH3 is absorbed by leaves still attached to the plants, resulting in a very high cycling of NH3 within the canopy, much higher than the net exchange with the atmosphere above the canopy (Nemitz et al., 2000). VI. MEASUREMENT TECHNIQUES The following section gives an overview of the most widely used techniques for measuring NH3 emission from fertilizer applied to the soil. McGinn and Janzen (1998) and Harper (2003) have comprehensively reviewed techniques for measuring NH3 fluxes to and from the soil. A. TRACER TECHNIQUES Ammonia emission has been estimated by crop response to fertilization or by soil mass balance after application of fertilizer that in some studies have been enriched with N15 (see Denmead et al., 1977; Farquhar et al., 1980; No¨mmik, 1966; Moal et al., 1995; Morvan et al., 1997). Crop response has been shown to be a very unreliable estimator due to variation in total N uptake and uptake of soil-N, other loss pathways such as leaching or nitrification/denitrification, and differences in the release patterns of N14 and N15 (Schjoerring et al., 1989). Isotope techniques for estimating N losses have also been questioned, since they seem to overestimate the loss, because NH3 can be both absorbed and released; a plant grown with 15N-enriched fertilizer will tend to lose 15NH3 and gain 14NH3, 598 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 even if the net NH3 flux is zero (Francis et al., 1997). A significant transfer of 15N between labeled and unlabeled plants has been shown to occur via the atmosphere in controlled environment studies (Janzen and Gilbertson, 1994). This will result in an overestimation of losses estimated by 15N analyses. Soil mass balances are unreliable due to loss pathways other than NH3 emission and also because of variations in concentration of TAN or organic N in the soil, i.e., for comparison, recovery of bromide (Br2) added to soil in columns was between 78 and 116% (de Jonge et al., 2003), showing the low precision of this technique. Recently, Vandre´ and Kaupenjohann (1998) have described a method whereby the transfer factor of NH3 from a source to a passive sampler on experimental plots is determined by releasing NH3 at a known rate via a cylinder and tubing on standard comparison plots. The transfer factor is then applied to passive sampler measurements of concentration from manure-treated plots to determine NH3 release rate (i.e., flux) from treated plots. A similar approach has been used by Warland and Thurtell (2000) to infer rates of nitrous oxide evolution from soil. Sherlock et al. (1995) showed for bare soils that the emission of NH3 from applied mineral fertilizers can be calculated as the product of wind speed, NH3 gas in equilibrium with NH3 dissolved in the surface soil layers, and a transfer coefficient. The equilibrium concentration technique (JTI method; Svensson, 1994) is a micrometeorological method suitable for measuring NH3 emissions from small plots. It involves sampling close to the soil surface to measure the driving force for volatilization and the aerodynamic resistance to flux. The method has recently been verified against the IHF method for applications of urea fertilizer and manure to large plots (Misselbrook and Hansen, 2001). The above-mentioned techniques have not been used extensively. The most commonly used methods have been enclosures and the micrometeorological methods described below. B. ENCLOSURES Enclosures are much used in both field and laboratory experiments. The enclosures can be chambers placed on the soil surface with no air flow through the head-space, i.e., static chambers, or they may be dynamic chambers with lids through which the air is exchanged by means of ventilators or pumps. These methods are useful when emission measurements are required over well-defined areas (e.g., small plots) and for comparing treatments under identical environmental conditions (Livingston and Hutchinson, 1995). Chambers are popular because of their portability, versatility, relative simplicity, and high sensitivity. They permit process studies and experiments withmany treatments in numbers that could not be contemplated with conventional AMMONIA EMISSION 599 ARTICLE IN PRESS 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 micrometeorological approaches because of the large land areas that the latter require. Moreover, the large increase in gas concentration that occurs in the headspace means that chambers can detect fluxes that are 100 times smaller than can be detected by micrometeorological techniques (Denmead, 1994). 1. Static Chambers The flux (F) is calculated from the rate of increase in gas concentration in the enclosure just after the system has been closed. F ¼ ðV=AÞdx=dt ð24Þ where V is the volume of the head-space, A is the area of soil surface enclosed by the chamber, x is gas concentration and t is time. The increase in gas concentration is often measured by absorbing the NH3 in an acid solution and storing in an open container or absorbing it on to a filter. However, the gas concentration gradient from the emitting surface to the air beneath the enclosure decreases as concentration in the air increases. Hence, the size of the enclosure and the measurement period must be carefully selected to avoid negative feedback on the rate of diffusion of the gas. Thus, the measured emission may be lower than when using methods which pass air over the soil and fertilizer (Volk, 1959; Denmead, 1979; McGarry et al., 1987). Such chambers can be used when gas emission rate is controlled by soil processes, as is the case for N2O and CH4, but not in situations where air exchange has a large impact on the emission rate, as may be the case for NH3. 2. Dynamic Chambers Closed dynamic chambers have been used in the laboratory and in the field. Air is drawn through the chamber, and the rate of NH3 emission is determined from the NH3 enrichment of the air stream, using Eq. (25). In laboratory studies, the surface of the soil in the chambers is generally about 0.01–0.04 m2 and the chambers are closed continuously (Kissel et al., 1977; Bacon et al., 1986; Sommer and Ersbøll, 1996), whereas chambers with lids that automatically close during short intervals of NH3 emission measurements have been used in field studies (Kissel et al., 1977). The wind tunnel system described by Lockyer (1984) is an example of a large dynamic, open chamber covering a surface area of about 1 m2. It employs a fan to draw air over the treated area. Emission (F) from the area is calculated from: F ¼ ðNH3;o 2 NH3;iÞv ð25Þ where NH3,o and NH3,i are the NH3 concentrations in the outlet and inlet air, respectively, and v is the volume of air flowing through the tunnel over 600 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 the sampling period. Average NH3 recovery of between 74 and 90% has been found for wind tunnel systems (Sommer et al., 1991; van der Weerden et al., 1996). It is suggested that the NH3 trapping efficiency of wind tunnel systems should be checked on a regular basis to avoid errors in measurement. As discussed already, emissions from the small plots covered by chambers might be higher than those from a field to which fertilizers have been applied. Ryden and Lockyer (1985) showed that NH3 emission measured with a wind tunnel adjusted to the wind speed at 10 cm height in the open was 5% higher than the emission measured with the micrometeorological mass balance technique. Usually wind speed in small laboratory chambers is made deliberately high and emissions correspondingly represent maximum losses (Kissel et al., 1977; Bacon et al., 1986). C. MICROMETEOROLOGICAL METHODS The concept of this approach is to measure NH3 emissions from large, open experimental areas, which can be plots or entire fields. Micrometeorological methods have the great advantage of not being intrusive, and of integrating across heterogeneities in the experimental area. They include mass balance methods, gradient diffusion approaches, eddy correlation, and relaxed eddy accumulation techniques, and methods based on Lagrangian dispersion. 1. Mass Balance Methods These are probably the most widely used techniques for measuring NH3 emissions from larger plots manured with mineral fertilizers. This technology does not affect emissions from the plots to which fertilizer has been applied, but it should be remembered that emission from a small plot might be higher than from a large field when scaling up emissions from the former. As discussed earlier in the chapter, whether or not plot size is important in this respect depends on the boundary conditions at the surface. Intercalibration studies have shown that different mass balance techniques give similar results (Schjoerring et al., 1992; Sommer et al., 1995; Wood et al., 2000; Sherlock et al., 2002). 2. Integrated Horizontal Flux (IHF) Methods This method equates the loss of NH3 from the surface of a treated plot with the difference between the amount of NH3 carried off the plot by the wind and the amount carried on to it (Denmead et al., 1977; Denmead, 1983; Wilson et al., 1983). It calculates the average surface flux density of NH3 in the treated area, AMMONIA EMISSION 601 ARTICLE IN PRESS 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 E, from the difference in the horizontal fluxes of NH3 across downwind and upwind boundaries: E ¼ 1 X ðzp z0 uðxd 2xuÞdz ð26Þ where X is the fetch (the distance traveled by the wind across the plot), u is horizontal wind speed, and xd and xu are the downwind and upwind atmospheric NH3 concentrations. The integration limit zp is the height at which the NH3 concentration is at background level. It should be noted that the integral in Eq. (26) is in terms of instantaneous values of u and the NH3 concentrations. The integral has been evaluated with mean wind speeds and mean concentrations. This neglects turbulent terms implicit in the transport equation, Eq. (26), and results in an overestimation of the true flux by 5–15% depending on the geometry (Raupach and Legg, 1984; Leuning et al., 1985; Wilson and Shum 1992; R.L. Desjardins et al., unpublished, 2003). In most studies, circular plots and making the “downwind” measurements at the plot center are used. The wind will always blow towards the center regardless of wind direction and the fetch will always be the same, viz., the plot radius. Radii from 3.5 to 40 m have been employed (Beauchamp et al., 1978; Gordon et al., 1988). A large simplification in technique for circular plots follows from developments by Wilson et al. (1982) and Denmead (1983), showing that for a given surface roughness and plot radius, there exists one particular height (ZINST) of measurement where the horizontal flux density is in a fixed ratio to the vertical flux density, regardless of atmospheric stability. The emission can thus be inferred from measurements of wind speed and atmospheric NH3 concentration (x) at a single height above the ground (Wilson et al., 1983). The flux F is calculated from: F ¼ u £ x Z ð27Þ where u¯ and x¯ are the mean wind speed and mean NH3 concentration, respectively, measured at ZINST. The term Z is the normalized horizontal flux (u¯x¯/F0), which is given in the form of nomograms for different surface roughnesses and plot radii by Wilson et al. (1982). The method offers considerable savings in labor and equipment, with results comparable to those obtained with the IHF method outlined above. This approach has been made even simpler by the development of passive samplers that measure the horizontal flux directly (Leuning et al., 1985; Sherlock et al., 1995). These require no power, no pumps, no anemometers, no data-loggers. A single sampler mounted at ZINST can return the mean NH3 flux from the treated plot over periods from 1 day to several weeks (for details see Freney et al., 1992). 602 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 3. Theoretical Profile Shape—Philip’s Solution (PTPS) McInnes et al. (1985) have used Philip’s (1959) analysis for a constant flux boundary condition to quantify the relationship between the surface flux and horizontal fluxes at any height. Circular geometry is used. Philip’s analysis predicts the concentration C1 generated at a particular height by a unit surface flux density F1 for a given wind speed. The true emission (F) is calculated from the concentration C measured at that height using the equation: F ¼ C C1 £ F1 ð28Þ Calculation of the fluxes at various heights requires measurements of air temperature and wind speed at two heights, as well as soil temperature and atmospheric stability. An advantage of the method is that it allows fluxes to be calculated from measurements at a height where the error to magnitude of measurement is smallest, but it is based on a constant flux surface condition, which may not always be appropriate. 4. Perimeter Profile Method The perimeter profile method is another mass balance method that employs four masts placed perpendicular to each other around the perimeter of an experimental area (Schjoerring et al., 1992). Arrays of flux samplers (Ferm, 1991) are mounted in pairs on masts around the boundary of a circular experimental area. The horizontal fluxes of the inward and outward pointing tubes are determined separately for each of several heights on each mast. The vertical flux of NH3 is then determined by stepwise summation of the difference between the inward and outward facing horizontal fluxes. This technique is laborious but has the advantage that there is little demand for a homogeneous environment around the plot to which fertilizer is applied. Denmead et al. (1998) describe a somewhat similar technique in which air is sampled at several heights along the full length of each boundary. It is designed particularly for situations in which there are scattered point sources, such as grazed pastures where NH3 is emitted from scattered dung and urine patches. D. GRADIENT DIFFUSION METHODS The vertical transport of gases in the surface layer is described by: Fg ¼ 2Kg›x=›z ð29Þ where Fg is flux of NH3, x¯ is its mean concentration over a sampling period long enough to encompass all the significant transporting eddies, and Kg is an eddy or AMMONIA EMISSION 603 ARTICLE IN PRESS 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 turbulent diffusivity for gases. Kg is determined by wind speed, height, aerodynamic roughness and atmospheric stability. Equation (29) leads to practical methods for calculating gas fluxes from knowledge of the flux of a tracer gas above the surface and the corresponding gradients or differences in mean atmospheric concentrations of the tracer and the gas. Aerodynamic methods use momentum as the tracer flux and horizontal wind speed as the tracer “concentration”. Energy balance methods use available energy as the tracer flux and a linear function of temperature and humidity as the “concentration”. Both methods are appropriate for large surface sources with fetches of hundreds of meters, and have been widely used for measuring NH3 flux to and from crops. 1. Aerodynamic Methods The basic equations are given in Appendix 1. If measurements are made at only two heights, the gas flux is calculated from: Fg ¼ k2ðu2 2u1Þðx1 2x2Þ {ln½ðz2 2dÞ=ðz1 2dÞ2½c1ðz22dÞ2c1ðz12dÞ}{ln½ðz22dÞ=ðz12dÞ2½c2ðz2 2dÞ2c2ðz12dÞ} ð30Þ In Eq. (30), the subscripts denote the two measuring heights, the overbars denote means over a suitable measuring period such as 20 or 30 min, and c1 and c2 are corrections for stability effects, given explicitly in Appendix 1. Aerodynamic methods have been much used for measuring NH3 fluxes from crops, soils, and water bodies (e.g., Denmead et al., 1978; Denmead, 1983; Harper et al., 1983, 2000; Sutton et al., 1993a,b, 2000; Genermont et al., 1998; Griffith and Galle, 2000). 2. Energy Balance Methods These methods are also discussed in Appendix 1. For concentrations measured at two heights, the gas flux is calculated from: Fg ¼ ðRn 2 G0 2 SÞðx1 2x2Þ rcp½ð T1 2 T2Þ þ ðe1 2 e2Þ=g ð31Þ where Rn is the net radiation receipt at the surface (incoming short- and longwave radiation minus reflected and re-emitted radiation), G0 is the flux density of heat into the soil at its surface, S is a storage term, T and e are air temperature and vapor pressure, and g is the psychrometric constant. Advantages of the method over the aerodynamic approach are that it does not require calculation of z0 or d and is applicable in all stability conditions. As well, the basic measurements can provide a measurement of evaporation rate, which 604 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 often has an important influence on NH3 production and emission. Its disadvantage is that it can give erroneous fluxes at night because both Rn and G are then small and difficult to measure, and net radiometers may not function effectively since their domes can become covered with dew. The same problems can arise in rainy periods. Aerodynamic methods are usually more reliable at night. Like the aerodynamic approach, the energy balance method is much used for measuring NH3 fluxes on a field basis. Some example applications are provided by Denmead et al. (1982, 2003) and Freney et al. (1992). It should be noted that sensors mounted at different heights above the surface have different fetches, so that problems arise in the use of gradient diffusion methods if the surface fluxes are spatially heterogeneous. E. EDDY CORRELATION This is the preferred micrometeorological method for measuring scalar fluxes, because it is a direct measurement requiring no simplifying assumptions (about similarity between the Ks) and no stability corrections, and it gives the vertical flux at the point of measurement. There are no problems with different footprints for different measurement levels as there are for gradient-diffusion methods. The basic method requires simultaneous measurements of the vertical wind speed w (usually by a sonic anemometer) and the gas concentration x at a sampling frequency rapid enough to catch all the significant eddies, say 10 Hz. The instantaneous vertical gas flux density ¼ wx, and the mean flux density is the average of wx over a period long enough to encompass all the effective transporting eddy sizes. There are problems in using the method with some trace gases because it requires fast gas measurement (10 Hz sampling), which is not yet possible for many trace gases, and density effects caused by simultaneous fluxes of sensible and latent heat may necessitate large corrections to the apparent flux (Webb et al., 1980). Recently, however, a fast-response gas analyzer employing a tunable diode laser has been developed for measuring atmospheric NH3 concentrations. Further, calculations by Wesely et al. (1989) indicate that for NH3, the corrections for density effects will usually be negligible, so that eddy correlation measurements of NH3 fluxes are likely to become more common. F. RELAXED EDDY ACCUMULATION OR CONDITIONAL SAMPLING The relaxed eddy accumulation method retains many of the advantages of eddy correlation inasmuch as it is a point measurement and requires no stability corrections. In addition, it does away with the need for rapid gas measurement. It still requires rapid measurement of the vertical wind speed w, but substitutes fast AMMONIA EMISSION 605 ARTICLE IN PRESS 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 solenoid valves for a fast-response gas analyzer. The anemometer controls a simple valving system. Air is sampled at a point at a constant rate and is diverted via the valves into “up” and “down” bins depending on the direction of the vertical wind. The contents of the bins are then analyzed for the mean gas concentrations of the ascending and descending eddies, but the analyzer can be of slow response and does not have to be mounted in the field or denuders may be used to trap the NH3 in the air streams (Zhu et al. 2000). The flux of NH3 is calculated from (Businger and Oncley, 1990): F ¼ bsððNHþ3 Þþ 2 ðNHþ3 Þ2Þ ð32Þ where b is a coefficient close to 0.6, sw is the standard deviation of w, and (NH3þ)þ and (NH3þ)2 are the respective average NH3 concentrations of the ascending and descending eddies (Businger and Oncley, 1990). G. LAGRANGIAN DISPERSION MODELS Lagrangian dispersion analyses adopt a coordinate system that travels with the dispersing entity. This is in contrast to the Eulerian analyses that we have described so far in this chapter, with the exception of the trajectory-simulation model of Wilson et al. (1982). Eulerian dispersion uses a fixed coordinate system and considers the passage of scalars at a point fixed in space. 1. Backward Lagrangian Stochastic Dispersion Model This model allows predictions of the strength of any surface source from onepoint measurements of wind speed and concentration at any location downstream (Flesch et al., 1995). The plot geometry and the location of the measuring point relative to the plot can be quite arbitrary, but must be known. The model calculates trajectories of air parcels backward in time from the sensor location to the source. It employs computer-simulated gas releases (about 10,000) to relate the surface flux density F0 to mean concentrations developed at specified heights and distances downwind. Required input information is z0, u¯ and gas concentration xg,z in excess of background at one particular height and distance downwind, plus atmospheric stability. Solutions have the form: F0 ¼ nuzxg;z ð33Þ where n is a coefficient calculated for the particular situation by the model. This method differs from the ZINST approach because it caters for any geometry and measurements can be made at any height downwind, but the stability needs to be known. Examples of its use are given by McGinn and Janzen (1998). 606 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2. Inverse Lagrangian Analysis Soils and plants can be both sources and sinks for NH3, so that measurements of the net NH3 exchange by a plant community will usually be insufficient for full understanding of ecosystem functioning. Additional information will be needed on the strengths and locations of the canopy sources and sinks. Recent research, notably by Raupach (1989a,b,c), has led to the development of a micrometeorological tool known as Inverse Lagrangian Analysis that allows the identification of the sites of gas exchange in plant canopies in a non-disturbing, continuous way from relatively simple observations of concentrations and turbulence parameters within and above plant canopies. It is beyond the scope of the present review to examine the analysis in any detail, but examples of its application to NH3 exchange in crops of corn, oilseed rape, and sugarcane can be found in Harper et al. (2000), Nemitz et al. (2000), and Denmead et al. (2003). VII. CONCLUSIONS AND PERSPECTIVES The emission of NH3 from inorganic N fertilizers varies considerably, depending on soil properties, fertilizer type and atmospheric conditions. Particularly the pH and pH buffer capacity of the soil and dissolving fertilizer salts may have a large impact on the NH3 emission. A simple classification tool that can be useful in simple decision support systems may be to classify soils as acid soils, calcareous soils, neutral soils, etc. and fertilizers as, e.g., alkaline, acid fertilizers, and fertilizers at risk of causing precipitates with Ca. The effect of temperature, wind and rain may interact and these interactions should be considered when estimating NH3 emission potentials. Thus, a high wind speed on days with high radiation may on the one hand reduce soil surface temperature, while on the other hand increase atmospheric NH3 transport, resulting in no net effect on NH3 emission. Temperature may significantly affect NH3 emission, but the total amount lost will not always increase, because the emission may be higher for a longer period at low than at high temperatures. Consequently, simple models relating emission to air temperature and wind speed may be erroneous, and more complex models will have to be developed, including calculation of surface temperature as affected by global radiation and wind. The effect of rain on NH3 emission from urea depends on timing and intensity—a light shower immediately after application may increase emission and heavy rain will reduce emission. Ammonia emission directly from the foliage of crops may on some, but not all, occasions be related to N fertilizer application. There are indications that AMMONIA EMISSION 607 ARTICLE IN PRESS 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 seedlings rapidly absorbing N via the roots may emit NH3 because internal N assimilation is too slow relative to N uptake. At later growth stages, the leaves may absorb NH3 emitted from fertilizer on the soil surface beneath the canopy, while during grain filling NH3 may again be emitted from the foliage as N is remobilised from senescing leaves. Biological activity in the soil associated with nitrification and N-immobilization may affect fertilizer NH3 emissions via effects on pH and NH4þ concentration. In models of ammonia losses from fertilized fields, both canopy–atmosphere NH3 exchange and nitrification should be included. Incorporation of fertilizers is among the most efficient techniques for reducing NH3 emission from dry soil. Splitting fertilizer applications so that the canopy more efficiently will be able to capture NH3 may also contribute to reduced emissions. Additives may influence loss of NH3 from applied fertilizers but their efficiency may be variable, e.g., depending on rain transporting urea into the soil. Furthermore, additives inhibiting urea hydrolysis are costly and additives reducing pH may either be a safety risk to farmers or may increase the demand for lime. The trend towards no-till crop production will increase the area of land covered with trash on which fertilizers are subsequently applied. Ammonia emission will be high from these fields and techniques should be developed to reduce NH3 emission without increasing soil tillage. Dynamic modeling of NH3 emission in relation to soil properties is challenging due to the many interacting processes. The physico-chemical and biological processes (pH, hydrolysis, cation adsorption, precipitation, nitrification etc.) affecting the gradient of ammonium and pH between the fertilizer granules and the soil may vary spatially and be difficult to predict. Thus, mixed models using empirical algorithms and mechanistic submodels may prove useful for the purpose of developing reliable emission-models with an acceptable and realistic need for input data. Atmospheric turbulence equations may predict transport of NH3 from a bare soil surface to the atmosphere, but may fail to describe NH3 emission from fertilizers on the soil beneath a plant canopy, i.e., the models may not precisely take into account how the canopy will change the vertical wind speed profiles and provide shade. Moreover, crops may absorb NH3, with the assimilation being related to NH3 concentration. Data are needed for validating models of ammonia emission and for the purpose of establishing relationship between emission and the most important emission factors. In the processing of data, systematic biases due to effect of measurement technique and plot size must be taken into account, i.e., NH3 emission from small plots may be too high compared with emissions measured on a large field scale. Therefore, results from small dynamic chamber studies should only be used qualitatively when assessing the effects of different soil, fertilizer or climatic factors on emission patterns. This bias should be corrected when scaling up the results to field scale or when developing decision support systems and calculating national inventories. 608 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 ACKNOWLEDGMENTS Hydro Agri supported this review financially. APPENDIX A. DERIVATION OF OPERATING FORMULAE FOR MEASURING GAS FLUXES BY AERODYNAMIC AND ENERGY BALANCE METHODS A.1 ATMOSPHERIC STABILITY This Appendix is prefaced with a short explanation about atmospheric stability. Stability is concerned with the relative effects of buoyancy (positive or negative) and wind shear on transport and mixing in the atmosphere. Atmospheres are described as being neutral, stable, or unstable depending on buoyancy. In neutral conditions, which occur for short periods around sunrise and sunset, there is essentially no temperature gradient and hence no buoyancy. Parcels of air displaced upwards are at the same temperature as their surroundings and are neither accelerated nor decelerated. Transport is subject only to mechanical turbulence. In stable conditions, which usually occur at night, the temperature gradient is positive (air is cooler closer to the ground) and parcels displaced upwards are heavier than their surroundings, so their motion is decelerated. Transport is inhibited. In unstable conditions, the usual daytime condition, the temperature gradient is negative (air is warmer closer to the ground) and parcels displaced upwards are lighter than their surroundings and so their motion is accelerated. Transport is enhanced. A common index of stability is the Monin– Obukhov length, L, which can be calculated from measurements of the temperature and wind speed gradients or from direct measurements of the heat and momentum fluxes (Paulson, 1970). A.2 AERODYNAMIC METHOD The methodology has been described in a number of reviews (e.g., Thom, 1972; Monteith and Unsworth, 1990; Denmead, 1994), and only summary results are given here. The general formula for calculating the gas flux from wind speed and concentration measurements is Fg ¼ k2z2 ›u ›z ›x ›z wmwg ¼ k2 ›u ›lnz ›x ›lnz wmwg ðA1:1Þ where the symbols are as used previously and wm and wg are functions to account for the effects of atmospheric stability on momentum and gas transport, respectively, and are given explicitly in the references above. AMMONIA EMISSION 609 ARTICLE IN PRESS 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 If wind speeds and concentrations have been measured at a number of points in the surface boundary layer, their gradients can be calculated and Eq. (A1.1) evaluated directly. Usually, however, wind speeds and concentrations are measured at only two levels in the boundary layer. Then, Eq. (A1.1) becomes: Fg ¼ k2ðu22u1Þðx1 2x2Þ {ln½ðz2 2dÞ=ðz1 2dÞ2½cmðz2 2dÞ2cmðz1 2dÞ}{ln½ðz2 2dÞ=ðz1 2dÞ2½cgðz22dÞ2cgðz12dÞ} ðA1:2Þ where cm and cg are integrated forms of the stability functions, wm and wg. In neutral conditions: cm ¼cg ¼0 ðA1:3Þ in stable conditions, cm ¼cg ¼25ðz2dÞ=L ðA1:4Þ and in unstable conditions, cm ¼2ln½ð1þxÞ=2þln½ð1þx2Þ=222tan21xþp=2 ðA1:5Þ and cg ¼2ln½ð1þx2Þ=2 ðA1:6Þ where x¼ð1216z=LÞ1=4 ðA1:7Þ If the friction velocity up is available from eddy correlation measurements, an alternative aerodynamic formulation is: Fg ¼ ku*ðx12x2Þ ln½ðz22dÞ=ðz12dÞ2½cgðz22dÞ2cgðz12dÞ ðA1:8Þ A.3 ENERGY BALANCE METHOD The energy balance at the surface is Rn ¼ H þlE þ G0 þ S; ðA1:9Þ where Rn is the net radiation receipt at the surface (incoming short- and longwave radiation less reflected and re-emitted radiation), H is the flux density of sensible heat from surface to atmosphere, lE is the flux density of latent heat, l being the latent heat of vaporization of water and E the surface evaporation rate, G0 is the flux density of heat into the soil at its surface, and S is a storage term, e.g., the change in heat stored in water bodies or the air or biomass of a plant community, or the solar energy fixed in photosynthesis. Rearranging Eq. (A1.9) 610 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 leads to H þlE ¼ Rn 2 G0 2 S ðA1:10Þ Following Eq. (17), H ¼ 2Khrcp›T=›z ðA1:11Þ lE ¼ 2Kgðrcp=gÞ›e=›z ðA1:12Þ and Fg ¼ 2Kg›x=›z ðA1:13Þ where most of the symbols are as defined previously, Kh and Kg are the eddy diffusivities for heat and gases or vapors respectively, e is vapor pressure, and g is the psychrometric constant (<66 Pa K21). Much micrometeorological research has shown that in most field situations, Kh ¼ Kg. On that assumption, Fg ¼ ðRn 2 G0 2 SÞ›x=›z rcp½›T=›z þ ð›e=›zÞ=g ðA1:14Þ or, for measurements at two levels, Fg ¼ ðRn 2 G0 2 SÞðx1 2x2Þ rcp½ð T1 2 T2Þ þ ðe1 2 e2Þ=g ðA1:15Þ Thus, H, lE and Fg can be determined from measurements of the energy fluxes at the surface and differences in temperature, humidity, and gas concentration between two heights above the surface. All the terms in Eq. (A1.15) can be measured directly with appropriate instrumentation. Advantages of the method over the aerodynamic approach are that it does not require calculation of z0 and d and is applicable in all stability conditions. REFERENCES Al-Kanani, T., MacKenzie, A. F., and Barthakur, N. N. (1991). Soil water and ammonia volatilization relationships with surface-applied nitrogen fertilizer solutions. Soil Sci. Soc. Am. J. 55, 1761–1766. Asman, W. A. H., and van Jaarsveld, H. A. (1991). A variable resolution transport model applied for NHx in Europe. Atmos Environ. 26A, 445–464. Asman, W. A. H., Sutton, M. A., and Schjørring, J. K. (1998). Ammonia: emission, atmospheric transport and deposition. New Phytol. 139, 27–48. Atwell, B. J., Fillery, I. R. P., McInnes, K. J., and Smucker, A. J. M. (2002). The fate of carbon and fertilizer nitrogen when dryland wheat is grown in monoliths of duplex soil. Plant Soil 241, 259–269. Avnimelech, Y., and Laher, M. (1977). Ammonia volatilization from soils: equilibrium considerations. Soil Sci Soc Am J 41, 1080–1084. AMMONIA EMISSION 611 ARTICLE IN PRESS 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 Bacon, P. E., and Freney, J. R. (1989). Nitrogen loss from different tillage systems and the effect on cereal grain yield. Fert Res. 20, 59–66. Bacon, P. E., Hoult, E. H., and McGarity, J. W. (1986). Ammonia volatilization from fertilizers applied to irrigated wheat soils. Fert. Res. 10, 27–42. Baker, J. H., Peech, M., and Musgrave, R. B. (1959). Determination of application losses of anhydrous ammonia. Agron. J. 51, 361–362. Beauchamp, E. G., Kidd, G. E., and Thurtell, G. (1978). Ammonia volatilisation from sewage sludge applied in the field. J. Environ. Qual. 7, 141–146. Black, A. S., Sherlock, R. R., Smith, N. P., Cameron, K. C., and Goh, K. M. (1985). Effects of form of nitrogen, season, and urea application rate on ammonia volatilization from pastures. N. Z. J. Agric. Res. 28, 469–474. Black, A. S., Sherlock, R. R., and Smith, N. P. (1987a). Effect of urea granule size on ammonia volatilization from surface-applied urea. Fert. Res. 11, 87–96. Black, A. S., Sherlock, R. R., and Smith, N. P. (1987b). Effect of timing of simulated rainfall on ammonia volatilization from urea, applied to soil of varying moisture content. J. Soil Sci. 38, 679–687. Blaise, D., Tyagi, P. C., and Khola, O. P. S. (1996). Ammonia volatilisation from urea as affected by the addition of iron pyrites and method of application. Nutr. Cycl. Agroecosyst. 46, 97–101. Blue, W. G., and Eno, C. F. (1954). Distribution and retention of anhydrous ammonia in sandy soils. Soil Sci. Soc. Am. Proc. 18, 420–424. Bouwman, A. F., Lee, D. S., Asman, W. A. H., Dentener, F. J., Van Der Hoek, K. W., and Olivier, J. G. J. (1997). A global high-resolution emission inventory for ammonia. Global Biogeochem. Cycl. 11, 561–587. Bouwman, A. F., Boumans, L. J. M., and Batjes, N. H. (2002). Estimation of global NH3 volatilization loss from synthetic fertilizers and animal manure applied to arable lands and grasslands. Global Biogeochem. Cycl. 16(2) (DOI 10.1029/2000GB001389). Bouwmeester, J. B., and Vlek, P. L. G. (1981). Wind-tunnel simulation and assessment of ammonia volatilization from ponded water. Agron. J. 73, 546–552. Bouwmeester, R. J. B., Vlek, P. L. G., and Stumpe, J. M. (1985). Effect of environmental factors on ammonia volatilization from a urea-fertilized soil. Soil Sci. Soc. Am. J. 49, 376–381. Bowmer, K. H., and Muirhead, W. A. (1987). Inhibition of algal photosynthesis to control pH and reduce ammonia volatilization from rice floodwater. Fert. Res. 13, 13–29. Bremner, J. M. (1995). Recent research on problems in the use of urea as a nitrogen fertilizer. Fert. Res. 42, 321–329. Bremner, J. M., and Douglas, L. A. (1971). Decomposition of urea phosphate in soils. Soil Sci. Soc. Am. J. 55, 1761–1766. Brouquisse, R., Evrard, A., Rolin, D., Raymond, P., and Roby, C. (2001). Regulation of protein degradation and protease expression by mannose in maize root tips: Pi sequestration by mannose may hinder the study of its signaling properties. Plant Physiol. 125, 1485–1498. Brutsaert, W. (1982). Evaporation into the Atmosphere. Reidel, Dordrecht, The Netherlands. Buijsman, E., Maas, J. F., and Asman, W. A. H. (1987). Anthropogenic NH3 emission in Europe. Atmos. Environ. 21, 1009–1022. Businger, J. A., and Oncley, S. P. (1990). Flux measurement with conditional sampling. J. Atmos. Ocean Technol. 7, 349–352. Bussink, D. W., Huijsmans, J. F. M., and Ketelaars, J. J. M. H. (1994). Ammonia volatilisation from nitric-acid-treated cattle slurry surface applied to grassland. Neth. J. Agric. Sci. 42, 293–309. Byrnes, B. H., and Freney, J. R. (1995). Recent developments on the use of urease inhibitors in the tropics. Fert. Res. 42, 251–259. Christianson, C. B., Carmona, G., Klein, M. O., and Howard, R. G. (1995). Impact on ammonia volatilisation losses of mixing KCl of high pH with urea. Fert. Res. 40, 89–92. 612 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 Chung, J.-B., and Zasoski, R. J. (1994). Ammonium–potassium and ammonium–calcium exchange equilibria in bulk and rhizosphere soil. Soil Sci. Soc. Am. J. 58, 1368–1375. De Datta, S. K., Trevitt, A. C. F., Freney, J. R., Obcemea, W. N., Real, J. G., and Simpson, J. R. (1989). Measuring nitrogen losses from lowland rice using bulk aerodynamic and nitrogen-15 balance methods. Soil Sci. Soc. Am. J. 53, 1275–1281. De Datta, S. K., Buresh, R. J., Sampson, M. I., Obcemea, W. N., and Real, J. G. (1991). Direct measurement of ammonia and denitrification fluxes from urea applied to rice. Soil Sci. Soc. Am. J. 55, 543–548. de Jonge, L. W., Sommer, S. G., and Jacobsen, O. H. (2003). Infiltration of and ammonia volatilization from pig and cattle slurry applied to harrowed and stubble soils. J. Environ. Qual. (In preparation). Denmead, O. T. (1979). Chamber systems for measuring nitrous oxide emission from soils in the field. Soil Sci. Soc. Am. J. 43, 89–95. Denmead, O. T. (1983). Micrometeorological methods for measuring gaseous losses of nitrogen in the field. In “Gaseous Loss of Nitrogen from Plant–Soil Systems” (J. R. Freney and J. R. Simpson, Eds.), pp. 133–158, Martinus Nijhoff/Dr W Junk, The Hague. Denmead, O. T. (1994). Measuring fluxes of CH4 and N2O between agricultural systems and the atmosphere. In “CH4 and N2O: Global Emissions and Controls from Rice Fields and other Agricultural and Industrial Sources” (K. Minami, A. Mosier, and R. Sass, Eds.), pp. 209–234. National Institute of Agro-Environmental Sciences, Tsukuba, Japan. Denmead, O. T., and Freney, J. R. (1992). Transfer coefficients for water–air exchange of ammonia, carbon dioxide and methane. Ecol. Bull. 42, 31–41. Denmead, O. T., Simpson, J. R., and Freney, J. R. (1977). A direct field measurement of ammonia emission after injection of anhydrous ammonia. Soil Sci. Soc. Am. J. 41, 1001–1004. Denmead, O. T., Freney, J. R., and Simpson, J. R. (1982). Dynamics of ammonia volatilisation during furrow irrigation of maize. Soil Sci. Soc. Am. J. 46, 149–155. Denmead, O. T., Harper, L. A., Freney, J. R., Griffith, D. W. T., Leuning, R., and Sharpe, R. R. (1998). A mass balance method for non-intrusive measurements of surface–air gas exchange. Atmos. Environ. 32, 357–369. Denmead, O. T., Raupach, M. R., Leuning, R., Dunin, F. X., and Freney, J. R. (2003). Inverse Lagrangian analysis of heat, vapor and gas exchange in plant canopies. In “Micrometeorological Measurements in Agricultural Systems” (E. J. Sadler, J. M. Ham, B. Tanner, J. M. Baker, and J. L. Hatfield, Eds.), American Society of Agronomy, Madison, (in press). Du Preez, C. C., and Burger, R. du T. (1988). Ammonia losses from ammonium-containing and—forming fertilizers after surface application at different rates on alkaline soils. Fert. Res. 15, 71–78. ECETOC. (1994). Ammonia Emissions to Air in Western Europe, Technical Report No. 62. European Centre for Ecotoxicology and Toxicology of Chemicals, Brussels, Belgium. EEA. (1999). Overview of National Programmes to Reduce Greenhouse Gas Emissions. European Environmental Agency, Final version, April 1999. EFMA. (1997). The Fertilizer Industry of the European Union: The Issues of Today, the Outlook for Tomorrow. European Fertilizer Manufacturers Association, Brussels, Belgium. Evangelou, V. P. (1990). Regulation mechanisms and field implications of ammonia bonding with various crystalline salts. Soil Sci. Soc. Am. J. 54, 394–398. Fan, M. X., MacKenzie, A. F., and Blenkhorn, H. D. (1996). A modified urea based NP fertilizer: Urea—TSP–MAP combinations. Fert. Res. 45, 217–220. Farquhar, G. D., Firth, P. M., Wetselaar, R., and Weir, B. (1980). On the gaseous exchange of ammonia between leaves and the environment: determination of the ammonia compensation point. Plant Physiol. 66, 710–714. Fenn, L. B., and Hossner, L. R. (1985). Ammonia volatilization from ammonium or ammoniumforming nitrogen fertilizers. Adv. Soil Sci. 1, 124–169. AMMONIA EMISSION 613 ARTICLE IN PRESS 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 Fenn, L. B., and Kissel, D. E. (1973). Ammonia volatilization from surface applications of ammonium compounds on calcareous soils. I. General theory. Soil Sci. Soc. Am. Proc. 37, 855–859. Fenn, L. B., and Kissel, D. E. (1974). Ammonia volatilization from surface applications of ammonium compounds on calcareous soils. I. Effects of temperature and rate of ammonium nitrogen application. Soil Sci. Soc. Am. Proc. 38, 606–610. Fenn, L. B., and Kissel, D. E. (1976). The influence of cation exchange capacity and depth of incorporation on ammonia volatilization from ammonium compounds applied to calcareous soils. Soil Sci. Soc. Am. Proc. 40, 394–398. Fenn, L. B., and Miyamoto, S. (1981). Ammonia loss and associated reactions of urea in calcareous soils. J. Soil Sci. Soc. Am. 45, 537–540. Fenn, L. B., and Richards, J. (1986). Ammonia loss from surface applied urea-acid products. Fert. Res. 9, 265–275. Fenn, L. B., Taylor, R. M., and Matocha, J. E. (1981). Ammonia losses from surface applied nitrogen fertilizer as controlled by soluble calcium and magnesium: general theory. Soil Sci. Soc. Am. J. 45, 777–781. Fenn, L. B., Matocha, J. E., and Wu, E. (1982). Substitution of ammonium and potassium for added in reduction of ammonia loss from surface-applied urea. Soil Sci. Soc. Am. J. 46, 771–776. Fenn, L. B., Malstrom, H. L., and Wu, E. (1987). Ammonia losses from surface-applied urea as related to urea application rates, plant residue and calcium chloride addition. Fert. Res. 12, 219–227. Ferguson, R. B., Kissel, D. E., Koelliker, J. K., and Basel, W. (1984). Ammonia volatilization from surface applied urea: effect of hydrogen ion buffering capacity. Soil Sci. Soc. Am. J. 48, 578–582. Ferm, M. (1991). A sensitive diffusional sampler. Report B-1020, Swedish Environmental Research Institute, Gothenburg. Fillery, I. R. P., and De Datta, S. K. (1986). Ammonia volatilization from nitrogen-sources applied to rice fields. 1. Methodology, ammonia fluxes, and N-15 loss. Soil Sci. Soc. Am. J. 50, 80–86. Fillery, I. R. P., Simpson, J. R., and De Datta, S. K. (1984). Influence of field environment and fertilizer management of ammonia loss from flooded rice. Soil Sci. Soc. Am. J. 48, 915–920. Fillery, I. R. P., Roger, P. A., and De Datta, S. K. (1986). Ammonia volatilization from nitrogen sources applied to rice fields. II. Floodwater properties and submerged photosynthetic biomass. Soil Sci. Soc. Am. J. 50, 86–91. Finnemann, J., and Schjoerring, J. K. (1999). Translocation of NH4þ in oilseed rape plants in relation to glutamine synthetase isogene expression and activity. Physiol. Plant 105, 469–477. Fleisher, Z., Kenig, A., Ravina, I., and Hagin, J. (1987). Model of ammonia volatilization from calcareous soils. Plant Soil 103, 205–212. Flesch, T., Wilson, J. D., and Yee, E. (1995). Backward time Lagrangian stochastic dispersion models, and their application to estimate gaseous emissions. J. Appl. Meteorol. 34, 1320–1332. Flowers, T. H., and O’Callaghan, J. R. (1983). Nitrification in soils incubated with pig slurry or ammonium sulphate. Soil Biol. Biochem. 15, 337–342. Francis, D. D., Schepers, J. S., and Vigil, M. F. (1993). Post-anthesis nitrogen loss from corn. Agron. J. 85, 659–663. Francis, D. D., Schepers, J. S., and Sims, A. L. (1997). Ammonia exchange from corn foliage during reproductive growth. Agron. J. 89, 941–946. Freney, J. R., Denmead, O. T., Watanabe, I., and Craswell, E. T. (1981). Ammonia and nitrous oxide losses following applications of ammonium sulphate to flooded rice. Aust. J. Agric. Res. 32, 37–45. Freney, J. R., Simpson, J. R., and Denmead, O. T. (1983). Volatilization of ammonia. In “Gaseous Loss of Nitrogen from Plant–Soil Systems” (J. R. Freney and J. R. Simpson, Eds.), pp. 1–32, Martinus Nijhoff/Dr. W. Junk, The Hague. Freney, J. R., Denmead, O. T., Sod, A. W., Saffigna, P. G., Chapman, L. S., Ham, G. J., Hurney, A. P., and Stewart, R. L. (1992). Factors controlling ammonia loss from trash covered sugarcane fields fertilized with urea. Fert. Res. 31, 341–349. 614 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 Genermont, S., and Cellier, P. (1997). A mechanistic model for estimating ammonia volatilisation from slurry applied to bare soil. Agric. Forest. Meteorol. 88, 145–167. Genermont, S., Cellier, P., Flura, D., Morvan, T., and Laville, P. (1998). Measuring ammonia fluxes after slurry spreading under actual field conditions. Atmos. Environ. 32, 279–284. Gilmour, J. T. (1984). The effects of soil properties on nitrification and nitrification inhibition. Soil Sci. Soc. Am. J. 48, 1262–1266. Goos, R. J., and Cruz, A. D. (1999). Effect of ammonium sulfate pretreatment on ammonia volatilization after urea fertilization. Commun. Soil Sci. Plant Anal. 30, 1325–1336. Gordon, R., Leclerc, M. Y., Schuepp, P., and Brunke, R. (1988). Field estimates of ammonia volatilisation from swine manure by a simple micrometeorological technique. Can. J. Soil Sci. 68, 369–380. Grant, R. F. (1994). Simulation of ecological controls on nitrification. Soil Biol. Biochem. 26, 305–315. Griffith, D. W. T., and Galle, B. (2000). Flux measurements of NH3, N2O and CO2 using dual beam FTIR spectroscopy and the flux-gradient technique. Atmos. Environ. 34, 1087–1098. Hanstein, S., Mattsson, M., Jaeger, H.-J., and Schjoerring, J. K. (1999). Uptake and utilization of atmospheric ammonia in three native Poaceae species: Leaf conductances, composition of apoplastic solution and interactions with root nitrogen supply. New Phytol. 141, 71–83. Harper, L. A. (2003). Ammonia: measurement issues. In “Micrometeorological Measurements in Agricultural Systems” (E. J. Sadler, J. M. Ham, B. Tanner, J. M. Baker, and J. L. Hatfield, Eds.), American Society of Agronomy, Madison, (in press). Harper, L. A., Catchpoole, V. R., Davis, R., and Weir, K. L. (1983). Ammonia volatilization: soil, plant and microclimate effects on diurnal and seasonal fluctuations. Agron. J. 75, 212–218. Harper, L. A., Sharpe, R. R., Langdale, G. W., and Giddens, J. E. (1987). Nitrogen cycling in a wheat crop: soil plant and aerial nitrogen transport. Agron. J. 79, 965–973. Harper, L. A., Bussink, D. W., van der Meer, H. G., and Corre, W. J. (1996). Ammonia transport in a temperate grassland. 1. Seasonal transport in relation to soil fertility and crop management. Agron. J. 88, 614–621. Harper, L. A., Denmead, O. T., and Sharpe, R. R. (2000). Identifying sources and sinks of scalars in a corn canopy with inverse Lagrangian dispersion analysis. II. Ammonia. Agric. For. Meteorol. 104, 75–83. Harrison, R., and Webb, J. (2001). A review of the effect of N fertilizer type on gaseous emission. Adv. Agron. 73, 65–108. Haynes, R. J., and Williams, P. H. (1992). Changes in soil solution composition and pH in urineaffected areas of pasture. J. Soil Sci. 43, 323–334. He, Z. L., Alva, A. K., Calvert, D. V., and Banks, D. J. (1999). Ammonia volatilization from different fertilizer sources and effects of temperature and soil pH. Soil Sci. 164, 751–758. Herrmann, B., Mattsson, M., Fuhrer, J., and Schjoerring, J. K. (2002). Leaf–atmosphere NH3 exchange of white clover in relation to mineral N nutrition and symbiotic N2 fixation. J. Exp. Bot. 53, 139–146. Hill, P. W., Raven, J. A., Loubet, B., Fowler, D., and Sutton, M. A. (2001). Comparison of gas exchange and bioassay determinations of the ammonia compensation point in Luzula sylvatica (Huds.) Gaud. Plant. Physiol. 125, 476–487. Hirel, B., and Lea, P. J. (2001). Ammonia assimilation. In “Plant Nitrogen” (P. J. Lea and J.-F. Morot-Gaudry, Eds.), pp. 79–99. Springer-Verlag, Berlin. Hopkins, D. F., Wells, G. L., and Butler, B. J. (1963). High-pressure injection of anhydrous ammonia, Trans. ASAE, Paper no. 62–645 presented at the winter meeting of the American Society of Agricultural Engineers in Chicago, December 1962, pp. 343–352. Hoult, E. H., and McGarity, J. W. (1987). The influence of sward mass, defoliation and watering on ammonia volatilization losses from an Italian ryegrass sward topdressed with urea. Fert. Res. 13, 199–207. AMMONIA EMISSION 615 ARTICLE IN PRESS 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 Husted, S., and Schjoerring, J. K. (1995). Apoplastic pH and ammonium concentration in leaves of Brassica napus L. Plant Physiol. 109, 1453–1460. Husted, S., and Schjoerring, J. K. (1996). Ammonia fluxes between Brassica napus plants and the atmosphere in response to changes in leaf temperature, light intensity and air humidity: interactions with stomatal conductance and apoplasticNH4þ andHþ concentrations. Plant Physiol. 112, 67–74. Husted, S., Mattsson, M., and Schjoerring, J. K. (1996). Ammonia compensation points in two cultivars of Hordeum vulgare L. during vegetative and generative growth. Plant Cell Environ. 19, 1299–1306. Husted, S., Hebbern, C., Mattsson, M., and Schjoerring, J. K. (2000a). A critical experimental evaluation of methods used for analysing ammonium in plant tissue, xylem sap and apoplastic fluid. Physiol. Plant. 109, 167–179. Husted, S., Schjoerring, J. K., Nielsen, K. H., Nemitz, E., and Sutton, M. A. (2000b). Stomatal compensation points for ammonia in oilseed rape plants under field conditions. Agric. For. Meteorol. 105, 371–383. Husted, S., Mattsson, M., Mo¨llers, C., Wallbraun, M., and Schjoerring, J. K. (2002). Photorespiratory NH4þ production in leaves of wildtype and GS2-antisense oilseed rape. Plant Physiol. 130, 989–998. Hutchings, N. J., Sommer, S. G., Andersen, J. M., and Asman, W. A. H. (2001). A detailed ammonia emission inventory for Denmark. Atmos. Environ. 35, 1959–1968. IFA Statistics (2002). International Fertilizer Industry Association: http://www.fertilizer.org/ifa/ statistics/STATSIND/tablen.asp. Ismail, K. M., Wheaton, F. W., Douglass, L. W., and Potts, W. (1991). Modeling ammonia volatilization from loamy sand soil treated with liquid urea. Trans. ASAE 34, 756–763. Izaurralde, R. C., Kissel, D. E., and Cabrera, M. L. (1987). Titratable acidity to estimate ammonia retention. Soil Sci. Soc. Am. J. 51, 1050–1054. Izaurralde, R. C., Kissel, D. E., and Cabrera, M. L. (1990). Simulation model of banded ammonia in soils. Soil Sci. Soc. Am. J. 54, 917–922. Jambert, C., Serca, D., and Delmas, R. (1997). Quantification of N-losses as NH3, NO and N2O and N2 from fertilized maize fields in south-western France. Nutr. Cycl. Agroecosyst. 48, 91–104. Janzen, H. H., and Gilbertson, C. (1994). Exchange of N-15 among plants in controlled environment studies. Can J. Soil Sci. 74, 109–110. Kirk, G. J. D., and Nye, P. H. (1991). A model of ammonia volatilization from applied urea. V. The effects of steady-state drainage and evaporation. J. Soil Sci. 42, 103–113. Kissel, D. E., Brewer, H. L., and Arkin, G. F. (1977). Design and test of a field sampler for ammonia volatilization. Soil Sci. Soc. Am. J. 41, 1133–1138. Larsen, S., and Gunary, D. (1962). Ammonia loss from ammoniacal fertilisers applied to calcareous soils. J. Sci. Food Agric. 11, 566–572. Leegood, R. C., Lea, P. J., Adcock, M. D., and Ha¨usler, R. E. (1995). The regulation and control of photorespiration. J. Exp. Bot. 46, 1397–1414. Lenschow, D. H. (1995). Micrometeorological techniques for measuring biosphere–atmosphere trace gas exchange. In “Biogenic Trace Gases: Measuring Emissions from Soil and Water” (P. A. Matson and R. C. Harriss, Eds.), Blackwell Science, Oxford. Leuning, R., Denmead, O. T., Simpson, J. R., and Freney, J. R. (1984). Processes of ammonia loss from shallow floodwater. Atmos. Environ. 18, 1583–1592. Leuning, R., Freney, J. R., Denmead, O. T., and Simpson, J. R. (1985). A sampler for measuring atmospheric ammonia flux. Atmos. Environ. 19, 1117–1124. Lightner, J. W., Mengel, D. B., and Rhykerd, C. L. (1990). Ammonia volatilization from nitrogen fertilizer surface applied to orchardgrass sod. Soil Sci. Soc. Am. J. 54, 1478–1482. Lindau, C. W., Reddy, K. R., Lu, W. F., Khind, C. S., Pardue, J. H., and Patrick, W. H. Jr. (1989). Effects of redox potential on urea hydrolysis and nitrification in soil suspensions. Soil Sci. 148, 184–190. 616 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 Livingston, G. P., and Hutchinson, G. L. (1995). Enclosure-based measurements of trace gas exchange: application and sources of error. In “Biogenic Trace Gases: Measuring Emissions from Soil and Water” (P. A. Matson and R. C. Harriss, Eds.), pp. 14–51. Blackwell Science, Oxford. Lockyer, D. R. (1984). A system for the measurement in the field of losses of ammonia through volatilization. J. Sci. Food Agric. 35, 837–848. Loubet, B., Milford, C., Hill, P. W., Tang, Y. S., Cellier, P., and Sutton, M. A. (2002). Seasonal variability of apoplastic NH4þ and pH in an intensively managed grassland. Plant Soil 238, 97–110. Lyster, S., Morgan, M. A., and O’Toole, P. (1980). Ammonia volatilization from soils fertilized with urea and ammonium nitrate. J. Life Sci. R. Dublin Soc. 1, 167–175. Maclean, J. (1997). “Rice Almanac”, 2nd edn. IRRI, Los Ban˜os, Philippines. Malhi, S. S., and McGill, K. S. (1982). Nitrification in three Alberta soils: effect of temperature, moisture and substrate concentration. Soil Biol. Biochem. 14, 393–399. Malhi, S. S., Grant, C. A., Johnston, A. M., and Gill, K. S. (2001). Nitrogen fertilization management for no-till cereal production in the Canadian Great Plains: a review. Soil Tillage Res. 60, 101–122. Mannheim, T., Braschkat, J., and Marschner, H. (1997). Ammonia emissions from senescing plants and during decomposition of crop residues. Z. Pflanzenernaehr. Bodenkd. 160, 125–132. Mary, B., Recous, S., and Robin, D. (1998). A model for calculating nitrogen fluxes in soil using 15N tracing. Soil Biol. Biochem. 30, 1963–1979. Mattsson, M., and Schjoerring, J. K. (1996). Ammonia emission from young barley plants: influence of N source, light/dark cycles, and inhibition of glutamine synthetase. J. Exp. Bot. 47, 477–484. Mattsson, M., and Schjoerring, J. K. (2002). Dynamic and steady state responses of inorganic nitrogen pools and NH3 exchange in leaves of Lolium perenne and Bromus erectus to changes in root N supply. Plant Physiol. 128, 742–750. Mattsson, M., Ha¨usler, R. E., Leegood, R. C., Lea, P., and Schjoerring, J. K. (1997). Leaf–atmosphere ammonia exchange in barley mutants with reduced activities of glutamine synthetase. Plant Physiol. 114, 1307–1312. Mattsson, M., Husted, S., and Schjoerring, J. K. (1998). Influence of nitrogen nutrition and metabolism on ammonia emission from plant leaves. Nutr. Cycl. Agroecosyst. 51, 35–40. McDowell, L. L., and Smith, G. E. (1958). The retention and reaction of anhydrous ammonia on different soil types. Soil Sci. Soc. Am. Proc. 22, 38–42. McGarry, S. J., O’Toole, P., and Morgan, M. A. (1987). Effects of soil temperature and moisture content on ammonia volatilization from urea-treated pasture and tillage soils. Ir. J. Agric. Res. 26, 173–182. McGinn, S. M., and Janzen, H. H. (1998). Ammonia sources in agriculture and their measurement. Can. J. Soil Sci. 78, 139–148. McInnes, K. J., and Fillery, I. R. P. (1989). Modeling and field measurements of the effect of nitrogen source on nitrification. Soil. Sci. Soc. Am. J. 53, 1264–1269. McInnes, K. J., Kissel, D. E., and Kanemasu, E. T. (1985). Estimating ammonia flux: a comparison between the integrated horizontal flux method and theoretical solutions of the diffusion profile. Agron. J. 77, 884–889. McInnes, K. J., Ferguson, R. B., Kissel, D. S., and Kanemasu, E. T. (1986a). Field measurements of ammonia loss from surface applications of urea solution to bare soil. Agron. J. 78, 192–196. McInnes, K. J., Ferguson, R. B., Kissel, D. S., and Kanemasu, E. T. (1986b). Ammonia loss from applications of urea-ammonium nitrate solution to straw residue. Soil Sci. Soc. Am. J. 50, 969–974. Mikkelsen, D. S., De Datta, S. K., and Obcemea, W. N. (1978). Ammonia volatilization losses from flooded rice soils. Soil Sci. Soc. Am. J. 42, 752–730. Misselbrook, T. H., and Hansen, M. N. (2001). Field evaluation of the equilibrium concentration technique (JTI method) for measuring ammonia emission from land spread manure or fertiliser. Atmos. Environ. 35, 3761–3768. AMMONIA EMISSION 617 ARTICLE IN PRESS 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 Misselbrook, T. H., van der Weerden, T. J., Pain, B. F., Jarvis, S. C., Chambers, B. J., Smith, K. A., Phillips, V. R., and Demmers, T. G. M. (2000). Ammonia emission factors for UK agriculture. Atmos. Environ. 34, 871–880. Moal, J.-F., Martinez, J., Guiziou, F., and Coste, C.-M. (1995). Ammonia volatilization following surface-applied pig and cattle slurry in France. J. Agric. Sci. 125, 245–252. van der Molen, J., van Faassen, H. G., Leclerc, M. Y., Vriesema, R., and Chardon, W. J. (1990). Ammonia volatilization from arable land after application of cattle slurry: derivation of a transfer model. Neth. J. Agric. Sci. 38, 239–254. Monteith, J. L., and Unsworth, M. H. (1990). “Principles of Environmental Physics”. Edward Arnold, London. Morvan, T., Leterme, Ph., Arsene, G. G., and Mary, B. (1997). Nitrogen transformations after the spreading of a pig slurry on bare soil and ryegrass using 15N-labelled ammonium. Eur. J. Agron. 7, 181–188. Murphy, D. V., Recous, S., Stockdale, E. A., Fillery, I. R. P., Jensen, L. S., Hatch, D. J., and Goulding, K. W. T. (2003). Gross nitrogen fluxes in soil: theory, measurement and application of 15N pool dilution techniques. Adv. Agron. 79, 000–000. Myers, R. J. K. (1975). Temperature effects on ammonification and nitrification in a tropical soil. Soil Biol. Biochem. 7, 83–87. Neftel, A., Blatter, A., Gut, A., Hogger, D., Meixner, F., Ammann, C., and Nathaus, F. J. (1998). NH3 soil and soil surface gas measurements in a triticale wheat field. Atmos. Environ. 32, 499–505. Nemitz, E., Sutton, M. A., Gut, A., San Jose, R., Husted, S., and Schjoerring, J. K. (2000). Sources and sinks of ammonia within an oilseed rape canopy. Agr. For. Meteorol. 105, 385–404. Nemitz, E., Milford, C., and Sutton, M. A. (2001). A two-layer canopy compensation point model for describing bi-directional biosphere–atmosphere exchange of ammonia. Q. J. Roy. Meteor. Soc. 127, 815–833. Nielsen, K. H., and Schjoerring, J. K. (1998). Regulation of apoplastic ammonium concentration in leaves of oilseed rape. Plant Physiol. 118, 1361–1368. No¨mmik, H. (1966). Use of microplot technique for studying gaseous loss of ammonia from added nitrogen materials under field conditions. Acta Agric. Scand. 16, 147–154. Nye, P. H. (1992). Towards the quantitative control of crop production and quality. II. The scientific basis for guiding fertilizer and management practice, particularly in poorer countries. J. Plant Nutr. 15, 1151–1173. O’Toole, P., McGarry, S. J., and Morgan, M. A. (1985). Ammonia volatilization from urea-treated pasture and tillage soils: effects of soil properties. J. Soil Sci. 36, 613–620. Ouyang, D. S., Mackenzie, A. F., and Fan, M. X. (1998). Ammonia volatilization from urea amended with triple superphosphate and potassium chloride. Soil Sci. Soc. Am. J. 62, 1443–1447. Pain, B. F., van der Weerden, T. J., Chambers, B. J., Phillips, V. R., and Jarvis, S. C. (1998). A new inventory for ammonia emissions from UK agriculture. Atmos. Environ. 32, 309–313. Paulson, C. A. (1970). The mathematical representation of wind speed and temperature profiles in the unstable atmospheric surface layer. J. Appl. Meteorol. 9, 857–861. Philip, J. R. (1959). The theory of local advection. Int. J. Meteorol. 16, 535–547. Phongpan, S., Freney, J. R., Keerthisinghe, D. G., and Chaiwanakupt, P. (1995). Use of phenylphosphorodiamidate and N-(n-butyl)thiophosphorictriamide to reduce ammonia loss and increase grain-yield following application of urea to flooded rice. Fert. Res 41, 59–66. Ping, J., Bremer, E., and Janzen, H. H. (2000). Foliar uptake of volatilized ammonia from surfaceapplied urea by spring wheat. Commun. Soil Sci. Plant Anal. 31, 165–172. Plantaz, M. A. G. H. (1998). “Surface/atmosphere exchange of ammonia over grazed pasture”. Dissertation No. 2468. University of Wageningen, The Netherlands. Prasertsak, P., Freney, J. R., Saffigna, P. G., Denmead, O. T., and Prove, B. G. (2001). Fate of urea nitrogen applied to a banana crop in the wet tropics of Queensland. Nutr. Cycl. Agroecosyst. 59, 65–73. 618 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 Rachhpal-Singh, and Nye, P. H. (1986a). A model of ammonia volatilization from applied urea. I. Development of the model. J. Soil Sci., 37, 9–20. Rachhpal-Singh, and Nye, P. H. (1986b). A model of ammonia volatilization from applied urea. II. Experimental testing. J. Soil Sci. 37, 37, 9–29. Raupach, M. R. (1989a). A practical Lagrangian method for relating scalar concentrations to source distributions in vegetation canopies. Q.J. R. Meteorol. Soc. 115, 609–632. Raupach, M. R. (1989b). Stand overstorey processes. Phil. Trans. R. Soc. Lond. B 324, 175–190. Raupach, M. R. (1989c). Applying Lagrangian fluid mechanics to infer scalar source distributions from concentration profiles in plant canopies. Agric. For. Meteorol. 47, 85–108. Raupach, M. R., and Legg, B. J. (1984). The uses and limitations of flux–gradient relationships in micrometeorology. Agric. Water Manage. 8, 119–131. Recous, S., Fresneau, C., Faurie, G., and Mary, B. (1988). The fate of labelled 15N urea and ammonium nitrate applied to a winter wheat crop. Plant Soil 112, 205–214. Recous, S., Machet, J. M., and Mary, B. (1992). The partitioning of fertilizer-N between soil and crop: comparison of ammonium and nitrate applications. Plant Soil 144, 101–111. Recous, S., Aita, C., Mary, B., and Faurie, B. (1999). In situ changes in gross N transformations in bare soil after addition of straw. Soil Biol. Biochem. 31, 119–133. Reynolds, C. M., and Wolf, D. C. (1987). Effect of soil moisture and air relative humidity on ammonia volatilization from surface-applied urea. Soil Sci. 143, 144–152. Ringbom, A. (1963). “Complexation in Analytical Chemistry”. Interscience Publishers, Chichester. Roelcke, M., Li, S. X., Tian, X. H., Gao, Y. J., and Richter, J. (2002). In situ comparisons of ammonia volatilisation from N fertilizers in Chinese loess soils. Nutr. Cycl. Agroecosyst. 62, 73–88. Russel, E. W. (1977). “Soil Conditions and Plant Growth”, 10th edn. Longman, London, New York. Ryden, J. C., and Lockyer, D. R. (1985). Evaluation of a system of wind tunnels for field studies of ammonia loss from grassland through volatilisation. J. Sci. Food. Agric. 36, 781–788. Ryden, J. C., Whitehead, D. C., Lockyer, D. R., Thompsen, R. B., Skinner, J. H., and Garwood, E. A. (1987). NH3 emission from grassland and livestock production systems in the UK. Environ. Pollut. 48, 173–184. Schjoerring, J. K., and Mattsson, M. (2001). Quantification of ammonia exchange between agricultural cropland and the atmosphere: measurements over two complete growth cycles of wheat, oilseed rape, barley and pea. Plant Soil 228, 105–115. Schjoerring, J. K., Nielsen, N. E., Jensen, H. E., and Gottschau, A. (1989). Nitrogen losses from fieldgrown spring barley plants as affected by rate of nitrogen application. Plant Soil 116, 167–175. Schjoerring, J. K., Sommer, S. G., and Ferm, M. (1992). A simple passive sampler for measuring ammonia emission in the field. Water Air Soil Pollut. 62, 13–24. Schjoerring, J. K., Kyllingsbaek, A., Mortensen, J. V., and Byskov-Nielsen, S. (1993). Field investigations of ammonia exchange between barley plants and the atmosphere. I. Concentration profiles and flux densities of ammonia. Plant Cell Environ. 16, 161–167. Schjoerring, J. K., Bock, J. H. G., Gammelvind, L., Jensen, C. R., and Mogensen, V. O. (1995). Nitrogen incorporation and remobilization in different shoot components of field-grown winter oil-seed rape as affected by rate of nitrogen application and irrigation. Plant Soil 177, 255–264. Schjoerring, J. K., Husted, S., and Poulsen, M. (1998). Soil–plant–atmosphere NH3 exchange associated with Calluna vulgaris and Deschampsia flexuosa. Atmos. Environ. 32, 507–512. Schlesinger,W. H., and Hartley, A. E. (1992). A global budget for atmospheric NH3. Biogeochemistry 15, 191–211. Schulze, E.-D., De Vries, W., Hauhs, M., Rose´n, K., Rasmussen, L., Tamm, C.-O., and Nilsson, J. (1989). Critical loads for nitrogen deposition on forest ecosystems. Water Air Soil Pollut. 48, 451–456. Sherlock, R. R., and Goh, K. M. (1985). Dynamics of ammonia volatilization from simulated urine patches and aqueous urea applied to pasture. I. Field experiments. Fert. Res. 5, 181–195. AMMONIA EMISSION 619 ARTICLE IN PRESS 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 Sherlock, R. R., Freney, J. R., Bacon, P. E., and van der Weerden, T. J. (1995). Estimating ammonia volatilization from unsaturated urea fertilized and urine affected soils by an indirect method. Fert. Res. 40, 197–205. Sherlock, R. R., Sommer, S. G., Khan, R. Z., Wesley Wood, C., Guertal, E. A., Freney, J. R., Dawson, C. O., and Cameron, K. C. (2002). Ammonia, methane, and nitrous oxide emission from pig slurry applied to a pasture in New Zealand. J. Environ. Qual. 31, 1491–1501. Singh, Y., Malhi, S. S., Nyborg, M., and Beauchamp, E. G. (1994). Large granules, nests or bands: methods of increasing efficiency of fall-applied urea for small cereal-grains in North America. Fert. Res. 38, 61–87. Sloan, J. J., and Anderson, W. B. (1995). Calcium chloride and ammonium thiosulfate as ammonia volatilisation inhibitors for urea fertilizers. Commun. Soil Sci. Plant Anal. 26, 2425–2447. Sommer, S. G., and Christensen, B. T. (1992). Ammonia volatilization after injection of anhydrous ammonia into arable soils of different moisture levels. Plant Soil 142, 143–146. Sommer, S. G., and Ersbøll, S. K. (1996). Effect of air flow rate, lime amendments and chemical soil properties on the volatilization of ammonia from fertilizers applied to sandy soils. Biol. Fert. Soils 21, 53–60. Sommer, S. G., and Jensen, C. (1994). Ammonia volatilization from urea and ammoniacal fertilizers surface-applied to winter wheat and grassland. Fert. Res. 37, 85–92. Sommer, S. G., Olesen, J. E., and Christensen, B. T. (1991). Effects of temperature, wind speed and air humidity on ammonia volatilization from surface applied cattle slurry. J. Agric. Sci. 117, 91–100. Sommer, S. G., Mikkelsen, H., and Mellqvist, J. (1995). Evaluation of meteorological techniques for measurements of ammonia loss from pig slurry. Agric. For. Meteorol. 74, 169–179. Sommer, S. G., Friis, E., Bach, A., and Schjørring, J. K. (1997). Ammonia volatilization from pig slurry applied with trail hoses or broadspread to winter wheat: effects of crop developmental stage, microclimate and leaf ammonia absorption. J. Environ. Qual. 26, 1153–1160. Son, T. T., and Buresh, R. J. (1994). Urea management for lowland transplanted rice as affected by application of farmyard manure. Fert. Res. 39, 199–203. Sørensen, P. (2001). Short-term nitrogen transformations in soil amended with animal manure. Soil Biol. Biochem. 33, 1211–1216. Stevens, R. J., Laughlin, R. J., and Kilpatrick, D. J. (1989). Soil properties related to the dynamics of ammonia volatilization from urea applied to the surface of acidic soils. Fert. Res. 20, 1–9. Stumm, W., and Morgan, J. J. (1981). Aquatic Chemistry: an Introduction Emphasizing Chemical Equilibria in Natural Waters. John Wiley and Sons, New York. Sutton, M. A., Fowler, D., and Moncrieff, J. B. (1993a). The exchange of atmospheric ammonia with vegetated surfaces. I. Unfertilized vegetation. Q. J. R. Meteorol. Soc. 119, 1023–1045. Sutton, M. A., Fowler, D., and Moncrieff, J. B. (1993b). The exchange of atmospheric ammonia with vegetated surfaces. II. Fertilized vegetation. Q. J. R. Meteorol. Soc. 119, 1047–1070. Sutton, M. A., Burkhardt, J. K., Guerin, D., Nemitz, E., and Fowler, D. (1998). Development of resistance models to describe measurements of bi-directional ammonia surface-atmosphere exchange. Atmos. Environ. 32, 473–480. Sutton, M. A., Nemitz, E., Milford, C., Fowler, D., Moreno, J., San Jose, R., Wyers, G. P., Otjes, R. P., Harrison, R., Husted, S., and Schjoerring, J. K. (2000). Micrometeorological measurements of net ammonia fluxes over oilseed rape during two vegetation periods. Agr. Forest Meteorol. 105, 351–369. Svensson, L. (1994). A new dynamic chamber technique for measuring ammonia emissions from land-spread manure and fertilizers. Acta Agric. Scand. B 44, 35–46. Tabatabai, M. A., and Bremner, J. M. (1972). Assay of urease activity in soils. Soil Biol. Biochem. 4, 479–487. Thom, A. S. (1972). Momentum, mass and heat exchange of vegetation. Q. J. R. Meteorol. Soc. 98, 124–134. 620 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 Thomas, R. J., Logan, K. A. B., Ironside, A. D., and Bolton, G. R. (1988). Transformation and fate of sheep urine-N applied to an upland U.K. pasture at different times during the growing season. Plant Soil 107, 173–181. Touchton, J. T., and Hargrove, W. L. (1982). Nitrogen-source and methods of application for notillage corn production. Agron. J. 74, 823–826. Van Hove, L. W. A., Adema, E. H., Vredenberg, W. J., and Pieters, G. A. (1989). A study of the adsorption of NH3 and SO2 on leaf surfaces. Atmos. Environ. 23, 1479–1486. Van Hove, L. W. A., Heeres, P., and Bossen, M. E. (2002). The annual variation in stomatal ammonia compensation point of rye grass (Lolium perenne L.) leaves in an intensively managed grassland. Atmos. Environ. 36, 2965–2977. Van Hove, L. W. A., Koops, A. J., Adema, E. H., Vredenberg, W. J., and Pieters, G. A. (1987). Analysis of the uptake of atmospheric ammonia by leaves of Phaseolus vulgaris L. Atmos. Environ. 21, 1759–1763. Vandre´, R., and Kaupenjohann, M. (1998). In situ measurement of ammonia emissions from organic fertilizers in plot experiments. Soil Sci. Soci. Am. J. 62, 467–473. Velthof, G. L., Oenema, O., Postmus, J., and Prins, W. H. (1990). In situ field measurements of ammonia volatilization from urea and calcium ammonium nitrate applied to grassland. Meststoffen 1/2, 41–45. Vlek, P. L. G., and Carter, M. F. (1983). The effect of soil environment and fertilizer modifications on the rate of urea hydrolysis. Soil Sci. 136, 56–63. Vlek, P. L. G., and Crasswell, E. T. (1981). Ammonia volatilization from flooded soils. Fert. Res. 2, 227–245. Volk, G. M. (1959). Volatile loss of ammonia following surface application of urea to turf or bare soils. Agron. J. 51, 746–749. Warland, J. S., and Thurtell, G. W. (2000). A micrometeorological method for in situ denitrification measurements using acetylene inhibition. Agric. For. Meteorol. 103, 387–391. Watson, C. J., and Kilpatrick, D. J. (1991). The effect of urea pellet size and rate of application on ammonia volatilization and soil nitrogen dynamics. Fert. Res. 28, 163–172. Watson, C. J., Travers, G., Kilpatrick, D. J., Laidlaw, A. S., and O’Riordan, E. (2000). Overestimation of gross N transformation rates in grassland soils due to non-uniform exploitation of applied and native pools. Soil Biol. Biochem. 32, 2019–2030. Webb, E. K., Pearman, G. I., and Leuning, R. (1980). Corrections to flux measurements for density effects due to heat and water vapour transfer. Q. J. R. Meteorol. Soc. 106, 85–100. van der Weerden, T. J., and Jarvis, S. C. (1997). Ammonia emission factors for N fertilizers applied to two contrasting grassland soils. Environ. Pollut. 95, 205–211. van der Weerden, T. J., van der Moal, J. F., Martinez, J., Pain, B. F., and Guiziou, F. (1996). Evaluation of the wind-tunnel method for measurement of ammonia volatilization from land. J. Agric. Eng. Res. 64, 11–13. Wesely, M. L., Lenschow, D. H., and Denmead, O. T. (1989). Flux measurement techniques. In “Global Tropospheric Chemistry: Chemical Fluxes in the Global Atmosphere” (D. H. Lenschow and B. B. Hicks, Eds.), pp. 31–46. National Center for Atmospheric Research, Boulder, CO. Wetselaar, R., and Farquhar, G. D. (1980). Nitrogen losses from tops of plants. Adv. Agron. 33, 263–302. Whitehead, D. C., and Lockyer, D. R. (1987). The influence of the concentration of gaseous ammonia on its uptake by the leaves of Italian ryegrass, with and without an adequate supply of nitrogen to the roots. J. Exp. Bot. 38, 818–827. Whitehead, D. C., Lockyer, D. R., and Raistrick, N. (1988). The volatilization of ammonia from perennial ryegrass during decomposition, drying and induced senescence. Annals Bot. 61, 567–571. AMMONIA EMISSION 621 ARTICLE IN PRESS 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 Whitehead, D. C., and Raistrick, N. (1990). Ammonia volatilization from five nitrogen compounds used as fertilizers following surface application to soils. J. Soil Sci. 41, 387–394. Whitehead, D. C., and Raistrick, N. (1993). The volatilization of ammonia from cattle urine applied to soils as influenced by soil properties. Plant Soil 148, 43–51. Wilson, J. D., and Shum, W. K. N. (1992). A re-examination of the integrated horizontal flux method for estimating volatilization from circular plots. Agric. For. Meteorol. 57, 281–295. Wilson, J. D., Thurtell, G. W., Kidd, G. E., and Beauchamp, E. G. (1982). Estimation of the rate of gaseous mass-transfer from a surface source plot to the atmosphere. Atmos. Environ. 16, 1861–1867. Wilson, J. D., Catchpoole, V. R., Denmead, O. T., and Thurtell, G. W. (1983). Verification of a simple micrometeorological method for estimating the rate of gaseous mass transfer from the ground to the atmosphere. Agric. Meteorol. 29, 183–189. Winter, F. P., and Eiland, F. (1996). The effect of soil pH on nitrification in coarse sandy soil. In “Progress in Nitrogen Cycling Studies” (O. van Cleemput, Ed.), pp. 655–658. Kluwer Academic Publishers, The Netherlands. Wood, C. W., Marshall, S. B., and Cabrera, M. L. (2000). Improved method for field scale measurement of ammonia volatilization. Commun. Soil Sci. Plant Anal. 31, 581–590. Yamulki, S., Harrison, R. M., and Goulding, K. W. T. (1996). Ammonia surface-exchnage above an agricultural field in Southeast England. Atmos. Environ. 30, 109–118. Zhu, T., Pattey, E., and Desjardins, R. L. (2000). Relaxed eddy-accumulation for measuring ammonia volatilization. Environ. Sci. Technol. 34, 199–203. Zia, M. S., Aslam, M., Rahmatullah, Arshad, M., and Ahmed, T. (1999). Ammonia volatilisation from nitrogen fertilizers with and without gypsum. Soil Use Manage. 15, 133–135. 622 S. G. SOMMER, J. K. SCHJOERRING AND O. T. DENMEAD ARTICLE IN PRESS 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838