Nitrogen balance and groundwater nitrate contamination: comparison among three intensive cropping systems on the North China Plain

The annual nitrogen (N) budget and groundwater nitrate-N concentrations were studied in the field in three major intensive cropping systems in Shandong province, north China. In the greenhouse vegetable systems the annual N inputs from fertilizers, manures and irrigation water were 1358, 1881 and 402 kg N ha(-1) on average, representing 2.5, 37.5 and 83.8 times the corresponding values in wheat (Triticum aestivum L.)-maize (Zea mays L.) rotations and 2.1, 10.4 and 68.2 times the values in apple (Malus pumila Mill.) orchards. The N surplus values were 349, 3327 and 746 kg N ha(-1), with residual soil nitrate-N after harvest amounting to 221-275, 1173 and 613 kg N ha(-1) in the top 90 cm of the soil profile and 213-242, 1032 and 976 kg N ha(-1) at 90-180 cm depth in wheat-maize, greenhouse vegetable and orchard systems, respectively. Nitrate leaching was evident in all three cropping systems and the groundwater in shallow wells (<15 m depth) was heavily contaminated in the greenhouse vegetable production area, where total N inputs were much higher than crop requirements and the excessive fertilizer N inputs were only about 40% of total N inputs.     

Ammonia emission, deposition and impact assessment at the field scale: a case study of sub-grid spatial variability

A local ammonia (NH3) inventory for a 5x5 km area in central England was developed, to investigate the variability of emissions, deposition and impacts of NH3 at a field scale, as well as to assess the validity of the UK 5-km grid inventory. Input data were available for the study area for 1993 and 1996 on a field by field basis, allowing NH3 emissions to be calculated for each individual field, separately for livestock grazing, livestock housing and manure storage, landspreading of manures and fertiliser N application to crops and grassland. An existing atmospheric transport model was modified and applied to model air concentrations and deposition of NH3 at a fine spatial resolution (50 m grid). From the mapped deposition estimates and land cover information, critical loads and exceedances were derived. to study the implications of local variability for regional NH3 impacts assessments. The results show that the most extreme local variability in NH3 emissions, deposition and impacts is linked to housing and storage losses. However, landspreading of manures and intensive cattle grazing are other important area sources, which vary substantially in the landscape. Overall, the range of predicted emissions from agricultural land within the study area is 0-2000 kg N ha(-1) year(-1) in 1993 and 0-8000 kg N ha(-1) year(-1) in 1996, respectively, with the peak at a poultry farm located in the study area. On average, the estimated field level NH3 emissions over the study area closely match the emission for the equivalent 5-km grid square in the national inventory for 1996. Deposition and expected impacts are highly spatially variable, with the edges of woodland and small "islands" of semi-natural vegetation in intensive agricultural areas being most at risk from enhanced deposition. Conversely the centres of larger nature reserves receive less deposition than average. As a consequence of this local variability it is concluded that national assessments at the 5 km grid level underestimate the occurrence of critical loads exceedances due to NH3 in agricultural landscapes.     

Nutrient losses by surface run-off following the application of organic manures to arable land. 2. Phosphorus

Phosphorus (P) surface run-off losses were studied following organic manure applications to land, utilising a purpose-built facility on a sloping site in Herefordshire under arable tillage. Different rates and timing of cattle slurry, farm yard manure (FYM) and inorganic nitrogen (N) and P fertiliser were compared, over a 4-year period (1993-97). N losses from the same studies are reported in a separate paper. The application of cattle FYM and, especially slurry, to the silty clay loam soil increased both particulate and soluble P loss in surface water flow. Losses via subsurface flow (30 cm interflow) were consistently much lower than via surface water movement and were generally unaffected by treatment. Increased application of slurry solids increased all forms of P loss via surface run-off; the results suggested that a threshold for greatly increased risk of P losses via this route, as for N, occurred at ca. 2.5-3.0 t/ha solids loading. This approximates to the 50 m3/ha application rate limit suggested for slurry within UK 'good agricultural practice'. The studies also provided circumstantial evidence of the sealing of the soil surface by slurry solids as the major mechanism by which polluting surface run-off may occur following slurry application on susceptible soils. Losses of total and soluble P, recorded for each of the 4 years of experiments, reached a maximum of only up to 2 kg/ha total P (TP), even after slurry applications initiating run-off. Whilst these losses are insignificant in agronomic terms, peak concentrations of P (up to 30,000 micrograms/l TP) in surface water during a run-off event, could be of considerable concern in sensitive catchments. Losses of slurry P via surface run-off could make a significant contribution to accelerated eutrophication on entry to enclosed waters, particularly when combined with high concentrations of NO3(-)-N. Restricting slurry application rates to those consistent with good agronomic practice, and within the limits specified in existing guidelines on good agricultural practice, offers the simplest and most effective control measure against this potentially important source of diffuse pollution.     

Agronomic measures of P, Q/I parameters and lysimeter-collectable P in subsurface soil horizons of a long-term slurry experiment

Soils from a long-term slurry experiment established in 1970 at Hillsborough, Northern Ireland, were used in the experiment. The site has a clay loam soil overlying Silurian shale. Seven treatments were used with three replicate plots per treatment under the following manurial regimes: (1) mineral fertiliser supplying 200 kg N, 32 kg P and 160 kg K ha(-1) yr(-1); (2)-(4) pig slurry applied at 50, 100 or 200 m3 ha(-1) yr(-1); (5)-(7) cow slurry applied at 50, 100 or 200 m3 ha(-1) yr(-1). Agronomic measures of P determined on subsurface layers down to 90 cm were compared with sorption isotherm data and rates of desorption. Adsorption isotherms were fitted using a standard Langmuir model. Data were compared with soluble (molybdate-reactive) P levels in soil water collected at 35 and 90 cm using PTFE suction cup lysimeters. Agronomically available P was concentrated in the top 30 cm of soil in all treatments. The accumulation of P in surface layers of the plots was significantly greater in the pig slurry treatments compared to the cow slurry, reflecting the history of P amendments. Nevertheless, over a period of a year, molybdate-reactive phosphorus (MRP) concentrations in lysimeter collections was consistently higher at 35 cm depth in the highest cow slurry treatment (7) compared to the equivalent pig slurry treatment (4). Either the movement of soluble P down the profile is facilitated by the higher organic content of cow slurry or P movement is not directly related to P accumulation in the soils. In addition, it is hypothesised that P movement down the soil profile depends upon two separate mechanisms. First, a 'break' point above which the accumulated P in the surface horizons is less strongly held and therefore amenable to dissolution and movement down the profile. Second, a mechanism by which some solute P from the surface horizons can travel rapidly through horizons of low P status to greater depth in the soil, i.e., by preferential flow.     

Phosphorus quantity-intensity relationships and agronomic measures of P in surface layers of soil from a long-term slurry experiment

Soils from a long-term slurry experiment established in 1970 at Hillsborough, Northern Ireland, were used in the experiment. The site has a clay loam soil overlying Silurian shale. Seven treatments were used with three replicate plots per treatment. Control plots were treated with mineral fertiliser supplying 200 kg N, 32 kg P and 160 kg K ha(-1) yr(-1). Slurry treatment plots were in two blocks and treated with either pig or cow slurry supplied at 50, 100 or 200 m3 ha(-1) yr(-1). Agronomic measures of P determined on 10-cm soil cores were compared with measured P quantity/intensity (Q/I) parameters from fitted sorption and desorption isotherms. Phosphorus affinity constant was found to be significantly and negatively correlated with P loading of soils. Desorption rate coefficient also increased significantly with increase in P loading from slurry, although there was no significant difference between slurry types (cow vs. pig). In contrast, while agronomic measures of P (water-soluble P, Olsen P, calcium chloride-extractable P, degree of P saturation (DPS)) also correlated significantly with P loading and total P (TP) in the soils, there was a separation and significant differences between the cow and pig slurry treatment blocks, with the former being much lower. Phosphorus inputs to pig slurry treated plots were much higher than to equivalent cow slurry plots over the first 15 years of the study but declined sharply over the most recent 10 years to more or less par. Conventional measures of agronomic P such as Olsen P and DPS, measure only P accumulation over the longer term and indicated only the higher content of P accumulating in soil of pig slurry treatments. Risk of P loss estimated by Q/I parameters appeared to show very similar behaviour between the two slurry types in line with more recent manurial additions but in contradiction of P accumulation statistics.     

Nutrient losses by surface run-off following the application of organic manures to arable land. 1. Nitrogen

Research was conducted on nitrogen (N) surface run-off losses following organic manure applications to land, utilising a purpose-built facility on a sloping site in Herefordshire under arable tillage. Different rates and timing of cattle slurry, farmyard manure and inorganic N and phosphorus (P) fertiliser were compared, over a 4-year period (1993-97). P losses from the same studies are reported in a separate paper. The application of cattle slurries to the silty clay loam soil increased the loss of solids and NH4(+)-N in surface water flow compared to control plots receiving inorganic fertiliser only, or no treatment, but had little effect on NO3(-)-N losses by this route. Results were consistent with other observations that rainfall events immediately after manure applications are particularly likely to be associated with nutrient run-off losses. Losses via subsurface flow (30 cm interflow) were consistently much lower than via surface water movement and were generally unaffected by treatment. Increasing slurry application rate and, in particular, slurry solids loading, increased solids and NH4(-)-N losses via surface run-off. The threshold, above which the risk of losses via surface run-off appeared to be greatly increased, was ca. 2.5-3.0 t/ha slurry solids, which approximates to the 50 m3/ha limit suggested for slurry within UK 'good agricultural practice'. Sealing of the soil surface by slurry solids appears to be a possible mechanism by which polluting surface run-off may occur following slurry application on susceptible soils. Total losses of NH4(+)-N and NO3(-)-N during the 4-year monitoring period were insignificant in agronomic terms, but average soluble N concentrations (NH4(+)-N + NO3(-)-N) in run-off, ranging from ca. 2.0 mg/l, up to 14.0 mg/l for the higher rate slurry treatments. Peak concentrations of NH4(+)-N > 30 mg/l, are such as to be of concern in sensitive catchments, in terms of the potential for contribution to accelerated eutrophication and adverse effects on freshwater biota.     

Nutrient run-off following application of livestock wastes to grassland

Three types of farm waste (cattle slurry, dirty water and farm yard manure (FYM)) were applied to hydrologically isolated grassland plots on a sloping poorly draining soil. Two applications were made, the first in October and the second in February. Application rates were 50 m(3) ha(-1) of slurry and dirty water and 50 t ha(-1) of FYM. Volumes of run-off following rainfall events and concentrations of N, P and K in run-off were measured. Losses of nutrients were higher following applications made with the soil at field capacity and rainfall soon after application. In terms of percentage loss of applied nutrients, losses were generally low. Concentration of N in run-off from the dirty water and FYM treated plots following the first application and the slurry treated plots following the second application exceeded 11.3 mg dm(-3) (a recommended limit for drinking water) although the maximum concentration recorded was 15 mg dm(-3) following FYM application. Concentration of P in run-off only exceeded 1 mg dm(-3) following the second application of cattle slurry. Concentration of K exceeded 10 mg dm(-3) following the first application of FYM and the second application of cattle slurry.     

Nitrate in waters from sewage-sludge amended lysimeters

Nitrate nitrogen was measured in runoff and tile-drainage during two years of operation of instrumented, large-scale lysimeters planted to corn (Zea mays L.) and amended with sewage sludge which was applied at rates supplying total N amounting to 2292 kg ha(-) in 1972 and 3286 kg ha(-1) in 1973. Other lysimeters were amended with inorganic fertiliser at the rate of 336 kg N ha(-1) year(-1). Annual losses in runoff and tile-drainage from sludge treatments were 0.9 and 5.1 and 371 and 663 kg NO(3)(-)-N ha(-1). Losses from lysimeters treated with inorganic fertiliser were 1.1 and 3.3 kg NO(3)(-)-N ha(-1) year(-1) in runoff and 31 and 79 kg NO(3)(-)-N ha(-1) year(-1) in tile-drainage. Given the nitrogen inputs accounted for in the study design, unaccounted for losses of 1800 to 2400 kg ha(-1) year(-1) were calculated for sludge and 277 kg ha(-1) year(-1) for inorganic fertiliser treatments. For one year there was a 300 kg ha(-1) increase in N in the lysimeters receiving inorganic fertiliser. Median NO(3)(-)-N concentrations ranged from 8.9 to 14.0 mg litre(-1) in runoff from sludge-treated lysimeters and 3.6 to 5.9 mg litre(-1) in runoff from lysimeters receiving inorganic fertiliser. In tile-drainage the median NO(3)(-)-N concentrations were 148 to 223 mg litre(-1) and 24 to 44 mg litre(-1) for sludge and inorganic fertiliser treatments, respectively. Highest runoff levels occurred in early summer storms, whereas highest tile-drainage concentrations occurred in late winter and early spring.     

Measurement of ammonia volatilization from a field,in upland Japan,spread with cattle slurry

Ammonia volatilization from livestock manure is one of the most important pathways of nitrogen loss from agricultural cultivated fields. In this paper, we report the measurement of ammonia emission from cattle slurry manure applied to upland in Miyazaki, Japan. It has been determined that after the cattle slurry was sprayed on the upland surface, the emission flux of the first day was 110 microg N ha(-1) s(-1). The loss of NH4(+) -N in the applied slurry was 60% after 5 days following the spraying of cattle slurry.     

Management options to limit nitrate leaching from grassland

Nitrate leaching can be reduced by the adoption of less intensive grassland systems which, though requiring a greater land area to achieve the same agricultural output, result in less nitrate leaching per unit of production than do intensively managed grasslands. The economic penalties associated with reductions in output can be partly offset by greater reliance on symbiotic nitrogen fixation and the use of clover-based swards in place of synthetic N fertilisers. Alternatively, specific measures can be adopted to improve the efficiency of nitrogen use in intensively managed systems in order to maintain high outputs but with reduced losses. Controls should take account of other forms of loss and flows of nitrogen between grassland and other components of the whole-farm system and, in most instances, should result in an overall reduction in nitrogen inputs. Removing stock from the fields earlier in the grazing season will reduce the accumulation of high concentrations of potentially leachable nitrate in the soil of grazed pastures but will increase the quantity of manure produced by housed animals and the need to recycle this effectively. Supplementing grass diets with low-nitrogen forages such as maize silage will reduce the quantity of nitrogen excreted by livestock but may increase the potential for nitrate leaching elsewhere on the farm if changes to cropping patterns involve more frequent cultivation of grassland. Improved utilisation by the sward of nitrogen in animal excreta and manures and released by mineralisation of soil organic matter will permit equivalent reductions to be made in fertiliser inputs, provided that adequate information is available about the supply of nitrogen from these non-fertiliser sources.     

Factors affecting ammonia volatilisation from a rice-wheat rotation system

Some of the major factors influencing ammonia volatilisation in a rice wheat rotation system were studied. A continuous airflow enclosure method was used to measure NH3 volatilisation in a field experiment at an agricultural college in Jiangsu Province. The five treatments comprised application rates of 0, 100, 200 or 300 kg N ha(-1) as urea, per growing season with rice straw amendment when wheat was sown, and 200 kg N ha(-1) without rice straw amendment. There were three replicates in a randomised block design. Ammonia volatilisation was measured immediately after urea application in the three consecutive years 1995 to 1997. The results show that N losses through NH3 volatilisation accounted for 4-19% of N applied during the wheat growing season and for 5-11% during the rice growing season. Ammonia volatilisation was affected significantly by soil moisture and temperature before and after fertiliser application during the wheat growing season. The ratio of volatilised NH3-N to applied N after urea application during the rice growing season was as follows: top-dressing at the onset of tillering > top-dressing at the start of the booting stage > basal fertilization. The results also show that the amount of N lost through NH3 volatilisation increased with increasing N application rate, but the ratio to applied N was not affected significantly by N application rate. Amendment with rice straw had no significant effect on NH3 volatilisation.     

A UK inventory of nitrous oxide emissions from farmed livestock

A UK inventory of the nitrous oxide (N₂O) emissions from farmed livestock was compiled to identify areas where potential abatement practices may be effective. Where possible, emission factors based on direct experimental data gathered under UK conditions were used, but published data were used when this was not feasible, together with statistical information, which included details of numbers of animals within each category of a species, animal liveweights, number of days housed, excretal rates and volumes of manures in stores. Total N₂O emissions were calculated for each component of livestock production systems, i.e. animal houses, manure stores, following application of manures to land and during grazing. Emissions were also estimated from land used for forage conservation and tillage. Total annual N₂O emissions from UK farmed livestock, based mainly on 1996 animal census data, were estimated to be 38.27 kt. The two main terms were 22.66 kt N₂O from mineral fertilisers after application to soils and 5.61 kt N₂O from stored manures (mainly in the form of farmyard manure). Within buildings, poultry were the largest contributors of N₂O, 2.97 kt, followed by cattle, 1.62 kt. Within the total emissions from stored manures, cattle were the largest contributors of N₂O, 3.58 kt, followed by poultry, 1.86 kt. Dietary manipulation and a move from solid manure based systems to slurry based systems appear to be promising abatement practices.     

The interactions between plant growth,vegetation structure and soil processes in semi-natural acidic and calcareous grasslands receiving long-term inputs of simulated pollutant nitrogen deposition

Regular applications of ammonium nitrate (35-140 kg N ha(-1) year(-1)) and ammonium sulphate (140 kg N ha(-1) year(-1)) to areas of acidic and calcareous grassland in the Derbyshire Peak District over a period of 6 years, have resulted in significant losses in both overall plant cover, and the abundance of individual species, associated with clear and dose-related increases in shoot nitrogen content. No overall growth response to nitrogen treatment was seen at any stage in the experiment. Phosphorus additions to the calcareous plots did however lead to significant increases in plant cover and total biomass, indicative of phosphorus limitation in this system. Clear and dose-related increases in soil nitrogen mineralization rates were also obtained, consistent with marked effects of the nitrogen additions on soil processes. High nitrification rates were seen on the calcareous plots, and this process was associated with significant acidification of the 140 kg N ha(-1) year(-1) treatments.     

Effects of increased deposition of atmospheric nitrogen on an upland moor: leaching of N species and soil solution chemistry

This study was designed to investigate the leaching response of an upland moorland to long-term (10 yr) ammonium nitrate additions of 40, 80 and 120 kg N ha(-1) yr(-1) and to relate this response to other indications of potential system damage, such as acidification and cation displacement. Results showed increases in nitrate leaching only in response to high rates of N input, in excess of 96 and 136 kg total N input ha(-1) yr(-1) for the organic Oh horizon and mineral Eag horizon, respectively. Individual N additions did not alter ammonium leaching from either horizon and ammonium was completely retained by the mineral horizon. Leaching of dissolved organic nitrogen (DON) from the Oh horizon was increased by the addition of 40 kg N ha(-1) yr(-1), but in spite of increases, retention of total dissolved nitrogen reached a maximum of 92% and 95% of 80 kg added N ha(-1) yr(-1) in the Oh and Eag horizons, respectively. Calcium concentrations and calcium/aluminium ratios were decreased in the Eag horizon solution with significant acidification mainly in the Oh horizon leachate. Nitrate leaching is currently regarded as an early indication of N saturation in forest systems. Litter C:N ratios were significantly lowered but values remained above a threshold predicted to increase leaching of N in forests.     

Effects of increased deposition of atmospheric nitrogen on an upland moor: nitrogen budgets and nutrient accumulation

This study was designed to investigate the effect of long-term (11 years) ammonium nitrate additions on standing mass, nutrient content (% and kg ha(-1)), and the proportion of the added N retained within the different compartments of the system. The results showed that more than 90% of all N in the system was found in the soil, particularly in the organic (Oh) horizon. Added N increased the standing mass of vegetation and litter and the N content (kg N ha(-1)) of almost all measured plant, litter and soil compartments. Green tissue P and K content (kg ha(-1)) were increased, and N:P ratios were increased to levels indicative of P limitation. At the lowest treatment, most of the additional N was found in plant/litter compartments, but at higher treatments, there were steep increases in the amount of additional N in the underlying organic and mineral (Eag) horizons. The budget revealed that the proportion of added N found in the system as a whole increased from 60%, 80% and up to 90% in response to the 40, 80 and 120 kg N ha(-1) year(-1) treatments, respectively.     

Anaerobic digestion of livestock and poultry manures spiked with tetracycline antibiotics

Abstract

We investigated the anaerobic degradation of tetracycline antibiotics (tetracycline [TC], oxytetracycline [OTC] and chlortetracycline [CTC]) in swine, cattle, and poultry manures. The manures were anaerobically digested inside polyvinyl chloride batch reactors for 64?days at room temperature. The degradation rate constants and half-lives of the parent tetracyclines were determined following first-order kinetics. For CTC the fastest degradation rate was observed in swine manure (k?=?0.016?±?0.001 d^?1; half-life = 42.8?days), while the slowest degradation rate was observed in poultry litter (k?=?0.0043?±?0.001 d^?1; half-life = 161?days). The half-lives of OTC ranged between 88.9 (cattle manure) and 99.0?days (poultry litter), while TC persisted the longest of the tetracycline antibiotics studied with half-lives ranging from 92.4?days (cattle manure) to 330?days (swine manure). In general, the tetracyclines were found to degrade faster in cattle manure, which had the lowest concentrations of organic matter and metals as compared to swine and poultry manures. Our results demonstrate that tetracycline antibiotics persist in the animal manure after anaerobic digestion, which can potentially lead to emergence and persistence of antibiotic resistant bacteria in the environment when anaerobic digestion byproducts are land applied for crop production.     

Ammonia emissions from cattle, pig and poultry wastes applied to pasture

In four field experiments, carried out in The Netherlands, small wind-tunnels were used to make direct measurements of ammonia (NH(3)) volatilization from different types of slurry and manure applied to the surface of grassland. During periods of up to six days following application, losses of NH(3)-N often amounted to more than 40% of the NH(4)-N applied. Percentage loss was highest (83%) from a poultry slurry and least (21%) from an air-dried poultry manure. Losses of NH(3)-N were generally greater from pig slurry (36-78%) than from cattle slurry (41%). In most cases 80% or more of the total NH(3)-N loss occurred within 48 h of application. Estimates were made of total annual NH(3) emissions from four systems of poultry housing. The highest total loss (50% of the N voided in droppings) occurred with a battery house producing a slurry with a low content of dry-matter; most of the loss took place after spreading. With a second battery house, in which the droppings were air-dried, the total loss was only 12%, with much lower emissions from the housing and during spreading.     

Effect of liquid manure on the mole fraction of nitrous oxide evolved from soil containing nitrate

The same emission factor is applied to fertiliser N and manure N when calculating national N2O inventories. Manures and fertilisers are often applied together to meet the N needs of the crop, but little is known about potential interactions leading to an increase in denitrification rate or a change in the composition of the end-products of denitrification. We used the 15N gas-flux method in a laboratory experiment to quantify the effect of liquid manure (LM) application on the fluxes of N2 and N2O when the soil contained fertiliser 15NO3-. LM increased the mole fraction of N2O from 0.5 to 0.85 in the first 12 h after application. More than 94% of the N2O was from the reduction of NO3-, probably due to aerobic nitrate respiration as well as respiratory denitrification.     

Nitrogen deposition and the biodiversity of boreal forests: implications for the nitrogen critical load

The critical load concept is used to establish the deposition levels which ecosystems can tolerate without significant harmful effects. Here we summarize work within the Swedish research program Abatement Strategies for Transboundary Air Pollution (ASTA) assessing the critical load of N for boreal forests. Results from both field experiments in an area with low background N deposition in northern Sweden, and from a large-scale monitoring study, show that important vegetational changes start to take place when adding low N doses and that recovery of the vegetation after ceasing N input is a very slow process. The data presented indicate that changes in key ecosystem components occur even at a lower rate of N input than the present recommended empirical critical load for boreal forest understorey vegetation of 10-15 kg N ha(-1) yr(-1). Based on the data presented, we suggest that the critical load should be lowered to 6 kg N ha(-1) yr(-1).     

The dynamics of dissolved nitrogen in a blanket peat dominated catchment

Precipitation, soil solution and drainage water were collected from a blanket peat catchment at Moor House National Nature Reserve in the Northern Pennine Uplands, UK, an area of moderately high N deposition. Two tributaries of the main stream were also sampled. Between 1993 and 1995 samples were analysed for NH4+ and NO3- and for part of the period for organic N. Inputs of N in precipitation exceeded outputs in stream water. Organic N represented a small proportion of N inputs while inputs of inorganic N averaged 10.2 kg ha(-1) a(-1). Soil solution from 10 cm depth in the peat was dominated by organic N whereas at 50 cm NH4+ slightly exceeded organic N. NO3- was rarely detected at either depth except during a period of exceptionally warm and dry weather in 1995. Output fluxes in stream water of organic N (5.7 to 6.5 kg ha(-1) a(-1)) were much greater than those of inorganic N (0.6 to 2.2 kg ha(-1) a(-1)). Inorganic N in streams was predominantly NO3- except in the smallest stream which had the largest concentrations of NH4+. This suggests that N transformations, particularly nitrification, may be taking place in the mineral soils adjacent to the streams or within the stream channel of the larger catchment.