Livestock Ammonia Emissions: Impacts on Air Quality, Animal and Human Health, and Mitigation Strategies

1. Introduction

Ammonia (NH₃) occupies a distinctive position in livestock production because it is simultaneously a nutrient loss, an indoor air contaminant and an atmospheric precursor. Nitrogen enters the production system mainly as dietary protein, is partitioned into animal product and excreta, and can be transferred rapidly to air when urinary urea or avian uric acid is converted to ammoniacal nitrogen. The emitted molecule may irritate animal and human airways close to the source, react with atmospheric acids to form secondary inorganic aerosol, deposit to sensitive ecosystems, or be captured and returned to productive use. This mobility makes NH₃ a defining component of the wider reactive-nitrogen cascade rather than an isolated odour problem [1,2].

Agriculture dominates anthropogenic NH₃ emissions in most intensively farmed regions, with livestock manure and fertilizer application providing the largest source categories. Historical inventories established the global importance of animal excreta, while satellite observations later revealed numerous agricultural hotspots that were absent from, or underestimated by, bottom-up inventories [3,4]. A global supply-chain assessment showed that livestock nitrogen losses are distributed across feed production, housing, manure management and field application; a measure applied at one stage can therefore conserve nitrogen only to permit its loss later [2]. Long-term modelling further indicates that agricultural NH₃ emissions and their contribution to nitrogen deposition have increased markedly since 1980 [5].

The air-quality consequence is not restricted to the farm boundary. NH₃ neutralizes sulfuric and nitric acids to form ammonium sulfate and ammonium nitrate, both important fractions of fine particulate matter (PM₂.₅). The amount formed depends on precursor availability, temperature, relative humidity and atmospheric transport; a kilogram of NH₃ emitted at one location cannot be converted to a fixed kilogram of PM₂.₅ without accounting for these conditions [6,7]. Global and regional modelling nonetheless shows that agricultural NH₃ abatement can materially reduce population exposure to PM₂.₅ and may be more cost-effective than additional nitrogen-oxide control in ammonia-rich regimes [8,9,10]. A laboratory reactor study published in 2026 found that approximately 76.6% of recovered particulate mass formed from controlled NH₃-SO₂ reactions fell within the PM₂.₅ size range; the authors appropriately described their derived factors as screening estimates rather than predictions of ambient concentrations [11].

Livestock buildings are also occupied environments. Animals can be exposed continuously at breathing height, whereas workers experience repeated, task-dependent peaks. NH₃ dissolves readily in ocular and airway lining fluid and occurs with organic dust, endotoxin, microorganisms, carbon dioxide, hydrogen sulfide and volatile organic compounds. Controlled studies isolate the toxicological contribution of NH₃ but simplify the commercial mixture; occupational studies capture the mixture but often cannot separate its correlated components [12,13]. These complementary designs show why a single concentration cannot serve simultaneously as an animal no-effect level, a worker exposure limit and an emission standard.

Recent work has increased both the relevance and the complexity of this field. A 2025 national spatial analysis in the United States identified 8763 cattle and 6963 hog operations and reported PM₂.₅ concentrations 28% higher in census tracts containing cattle operations and 11% higher in tracts containing hog farms than in tracts without them; the spatial design establishes association at the tract level, not the ammonia-specific contribution of an individual facility [14]. In 2026, global modelling projected that warming alone could increase agricultural NH₃ emissions by 5-22% late in the century and reduce the effectiveness of linked mitigation packages [15]. Climate adaptation, air-quality protection and nutrient management must consequently be evaluated together.

This review examines livestock NH₃ across a source-pathway-receptor-control continuum (Figure 1). Its objectives are to explain the transformation of feed nitrogen into airborne NH₃; distinguish the measurements needed for indoor exposure, facility emissions and regional air quality; synthesize consequences for animals, workers, communities and ecosystems; critically compare mitigation from nutrition to land application; and define research designs capable of detecting pollution transfer. The organizing principle is that successful mitigation must both reduce dose at the intended receptor and reduce whole-chain reactive-nitrogen loss.

2. Review Approach

This article was developed as a critical integrative review rather than a systematic review or meta-analysis. Peer-reviewed studies and authoritative technical guidance were selected to represent the complete livestock-ammonia pathway: nitrogen intake and excretion, manure chemistry, housing and ventilation, measurement, controlled animal exposure, occupational and community health, atmospheric transformation, and mitigation at housing, storage and field stages. Targeted searches were updated through 31 August 2026 using combinations of ammonia, livestock, poultry, swine, cattle, housing, emission, air quality, health, PM₂.₅, manure and mitigation. Titles, author lists, publication years and DOI links were checked against the publication record.

Interpretation followed four safeguards. First, air concentration was not treated as emission: concentration is affected by source strength, airflow, mixing and sampling position, whereas emission requires a mass-flow or dispersion calculation. Second, excreted nitrogen was not treated as emitted NH₃ because hydrolysis, acid-base partitioning and management intervene. Third, health associations near livestock operations were not assigned to NH₃ alone when dust, endotoxin, microorganisms and other gases co-varied. Fourth, a percentage reduction at one manure stage was not described as a whole-farm benefit unless downstream nitrogen fate and co-pollutants were considered. These safeguards provide the basis for the measurement framework in Table 1.

3. Formation and Emission Dynamics

The potential NH₃ source is established upstream by nitrogen intake, digestibility, amino-acid balance, animal requirement and productive retention. Nitrogen not retained in growth, milk or eggs enters faeces and urine. In cattle and pigs, urinary urea is rapidly hydrolysed after contact with faecal urease. In poultry, uric acid and other nitrogenous compounds undergo sequential microbial mineralization. Faecal organic nitrogen is generally less immediately volatile but can replenish total ammoniacal nitrogen (TAN) during storage. Species, production stage and manure system therefore determine the temporal pattern of substrate supply [16,17,18].

Urea hydrolysis can be summarized as CO(NH₂)₂ + H₂O → 2NH₃ + CO₂, although aqueous systems contain both NH₃ and ammonium (NH₄⁺). Their equilibrium is NH₄⁺ ⇌ NH₃ + H⁺. Only the un-ionized fraction readily transfers to the gas phase, so increasing pH sharply raises volatilization potential. Temperature both changes the dissociation equilibrium and accelerates biological conversion and mass transfer. Figure 2 calculates the equilibrium fraction of TAN present as un-ionized NH₃ across representative temperatures using the relationship reported by Emerson et al. [19]. At pH values typical of acidified manure, the fraction is small; as pH approaches and exceeds the dissociation constant, the available gaseous fraction rises nonlinearly. The curves describe thermodynamic partitioning, not an emission rate, because substrate supply, buffering, boundary-layer resistance and ventilation still control flux.

Moisture has similarly non-linear effects. Water enables enzyme activity, solute diffusion and contact between nitrogenous substrates and microbes; wet or caked poultry litter commonly sustains NH₃ generation. Very dry material can suppress microbial conversion, whereas excess liquid may dilute concentrations but enlarge emitting area and promote anaerobic zones. Drinker leakage, condensation and inadequate litter renewal therefore interact with stocking density, bedding type and ventilation. In slurry systems, urine-feces mixing, pit surface area, crust formation and the interval between removal events are decisive. Reviews of pig and cattle housing consistently identify diet, floor design, manure-removal method, temperature, ventilation and cleanliness as major determinants [20,21].

Mass transfer occurs when the partial pressure of NH₃ at the manure surface exceeds that in the overlying air. Turbulence and air speed reduce boundary-layer resistance and can increase release even while diluting the indoor concentration. This distinction creates the ventilation paradox: greater airflow generally lowers the concentration inhaled inside the building, but may increase or leave unchanged the mass exported outdoors. A simplified steady-state balance is Cᵢ ≈ Cₒ + S/Q, where Cᵢ and Cₒ are indoor and inlet concentrations, S is source rate and Q is ventilation rate. The relation is useful conceptually but assumes adequate mixing and a stable source, conditions rarely satisfied throughout a production cycle. Emission is instead estimated as E = Q(Cₒᵤₜ - Cᵢₙ), with synchronized concentration and airflow measurements and explicit treatment of uncertainty.

Historical multi-country monitoring illustrates the scale of between-system variation. Mean concentrations in Northern European buildings were generally below 8 ppm in cattle housing, approximately 5-18 ppm in pig housing and approximately 5-30 ppm in poultry housing, with strong effects of building and management [22]. Those values are descriptive of the monitored systems and period, not universal emission factors. Contemporary datasets reinforce the need to normalize emissions to animal place, live mass, product and time and to report housing, ventilation and manure-management metadata [23,24].

Seasonal concentration and emission can move in opposite directions. In a Korean laying-hen house, winter NH₃ concentration reached 9.33 ppm, whereas the NH₃ emission factor was greatest in summer and lowest in winter because ventilation volume changed substantially [25]. This finding is an especially clear demonstration that a high indoor reading does not necessarily identify the season of greatest exported mass. A 2026 analysis of the DATAMAN database likewise reported lower poultry emission factors in European than North American datasets and lower post-2010 than pre-2010 values; design, litter practice, climate, measurement method and technological change remain intertwined, so regional factors should not be transferred without qualification [26].

Within a building, spatial and temporal variability can be as important as the flock or herd mean. Freshly wetted zones, manure channels, pits and exhaust paths form hotspots. Animal growth, litter age, outside weather, controller setpoints and minimum-ventilation schedules produce diurnal and production-cycle patterns. Measurements should therefore capture breathing-zone means, peaks and cumulative duration as well as ventilation-weighted mass. Low-cost electrochemical or metal-oxide sensors can support continuous surveillance, but drift, humidity interference, cross-sensitivity, calibration range and placement must be reported. Reference-grade analysers remain necessary for calibration and periodic validation.

The mechanistic determinants, measurement consequences and corresponding control points are summarized in Table 2. The table emphasizes that no single variable is sufficient: TAN indicates substrate, pH and temperature indicate partitioning, surface and airflow indicate transfer, and downstream management determines whether conserved nitrogen remains conserved.

4. Air Quality and Health Consequences

4.1. Animal Health, Welfare and Productivity

NH₃ is highly soluble and reacts with moisture on ocular, nasal and respiratory surfaces. Local alkalinity can disrupt epithelial integrity, impair mucociliary clearance and increase susceptibility to secondary challenges. The dose received by an animal depends on concentration, duration, respiratory rate, life stage, microclimate and co-exposure. Controlled chambers are valuable for isolating concentration and duration, but commercial buildings add dust, pathogens, endotoxin, variable humidity and behavioural avoidance. Findings should therefore be interpreted as biological response ranges rather than a universal threshold.

Broilers provide the most developed controlled-exposure literature. Exposure to 0, 25, 50 or 75 ppm from days 1 to 28 produced concentration- and duration-dependent ocular lesions; abnormalities developed even at 25 ppm, more slowly and less severely than at 50 or 75 ppm, and recovery after exposure was incomplete at the higher concentrations [27]. Studies at 0, 30 and 60 ppm reported conjunctival injury and poorer feed-use efficiency at the highest concentration, while separate work linked sustained exposure to impaired modern-broiler performance [28,29]. These experiments do not imply that every commercial flock at a particular reading will experience the same loss, but they establish dose-responsive injury and show that ocular endpoints may change before mortality or gross clinical disease.

Subclinical responses extend beyond the eye. High atmospheric NH₃ altered hepatic protein expression and breast-muscle transcription in broilers, including pathways related to metabolism, oxidative stress and growth [30,31]. Such molecular findings strengthen mechanistic plausibility but should not be translated directly into farm-level production loss without replicated field validation. The critical management implication is that normal average daily gain cannot be treated as proof of biological neutrality.

Pigs show a comparable separation between overt production effects and earlier physiological change. Young pigs exposed to high concentrations in classic chamber work developed respiratory signs and growth impairment [32]. At 35 or 50 ppm, nursery-pig studies examined haematology, stress, behaviour and performance over acute and prolonged periods, demonstrating that endpoint and duration materially shape interpretation [33]. In a 15-week factorial experiment, pigs exposed to approximately 20 ppm showed lower salivary cortisol, larger adrenal cortices and less early play, while productivity was unchanged; the direction of cortisol should not be interpreted simplistically as reduced stress because chronic hypothalamic-pituitary-adrenal adaptation was plausible [34]. More recent work associates chronic NH₃ exposure with pulmonary oxidative stress, enhanced glycolysis, hindgut dysbiosis and altered microbial metabolites [35,36]. A 2024 long-duration study also reported reduced performance and multiple pathological lesions in exposed pigs [37].

Evidence for calves is less complete. A systematic review found only a small number of studies that formally examined NH₃ in relation to bovine respiratory disease and concluded that heterogeneity and risk of bias prevented a defensible threshold [38]. One longitudinal field study associated cumulative barn exposure with ultrasonographic lung lesions, airway inflammation and infection, but the measured air-quality variables and group-housing conditions were correlated [39]. For calf housing, the strongest conclusion is methodological: continuous duration metrics, thoracic ultrasound, clinical scoring, ventilation, bedding, microbial load, dust and endotoxin should be evaluated together.

Table 3 compares representative animal and human studies while keeping direct endpoints separate from inference. Across species, four patterns recur: concentration and duration interact; ocular, behavioural, inflammatory or molecular responses may precede growth loss; commercial mixtures can modify the response; and experimental-unit limitations, particularly one chamber per treatment, can overstate precision.

4.2. Human Health, Community Air Quality and Environmental Fate

Livestock workers encounter complex bioaerosol mixtures. Swine-facility studies linked ammonia, endotoxin and dust metrics with chronic symptoms and pulmonary-function changes, while repeated poultry studies demonstrated cross-shift or prospective changes associated with the mixed exposure profile [40,41,42,44]. Experimental and occupational synthesis indicates that dust and NH₃ may interact, yet their correlation and the presence of endotoxin, microorganisms and disinfectants constrain single-pollutant attribution [13,45]. Worker protection should therefore control the mixture at source, monitor tasks and peaks, and use respiratory protection as a residual barrier rather than as the primary emission-control strategy.

Occupational limits also require careful interpretation. The NIOSH recommended exposure limit for NH₃ is 25 ppm as a 10-hour time-weighted average and 35 ppm as a 15-minute short-term limit [46]. These values were developed for adult workers and specific exposure periods; they are not animal no-effect concentrations, community ambient standards or substitutes for good housing management. Animals may be exposed continuously, including during vulnerable developmental stages, and welfare endpoints such as ocular injury or play behaviour differ from the endpoints used to establish occupational limits.

Community studies extend concern beyond the workplace but add greater source-attribution uncertainty. Repeated measurements in an agricultural community linked 24-hour ambient NH₃ variation with short-term reductions in lung function among children with asthma [43]. Studies near concentrated swine operations reported associations between livestock-related air pollution, symptoms and pulmonary function, and a systematic review found that proximity-based findings were suggestive but heterogeneous [47,48]. In the Netherlands, livestock-related air pollution was associated with airway obstruction in neighbouring residents and with acute respiratory responses in a panel of people with chronic obstructive pulmonary disease [49,50]. These studies support exposure reduction but do not justify assigning all observed risk to NH₃ because farm density, endotoxin, primary particles, microbes and secondary aerosol co-vary.

The 2025 United States spatial analysis adds national-scale resolution by locating cattle and hog operations and comparing census-tract PM₂.₅ burdens [14]. It’s finding that socially vulnerable populations were disproportionately represented in affected tracts makes distributional impact a legitimate mitigation criterion. However, tract-level association remains distinct from a facility-specific causal estimate. Fence-line monitoring, source tracers, dispersion modelling, particulate chemical speciation and longitudinal health measurements are needed to connect source, pathway and receptor.

Atmospheric processing determines how far the impact travels and in which chemical form. NH₃ has a comparatively short lifetime where dry deposition is efficient, but reaction with nitric or sulfuric acid can extend transport as particulate ammonium. The PM response to NH₃ control is larger in ammonia-rich, acid-limited regimes and smaller where other precursors or thermodynamics limit aerosol formation. Scenario studies estimate substantial health benefits from agricultural control, but their numerical results depend on emissions inventories, chemical transport, exposure surfaces, concentration-response functions and valuation [9,51,52]. The WHO PM₂.₅ guideline provides a receptor-based health benchmark for ambient fine particles, not a direct conversion factor for farm NH₃ [53].

Deposition transfers reactive nitrogen to terrestrial and aquatic systems. Chronic nitrogen enrichment can acidify soils, alter plant competition and reduce species richness, particularly in nutrient-sensitive grasslands and habitats [54,55]. Source apportionment using concentration gradients and nitrogen-isotope ratios has demonstrated livestock plume transport across landscapes, but fractionation, background mixing and deposition complicate quantitative allocation [56]. Integrated farm and regional policy must therefore address both the directly inhaled gas and the deposited or particle-bound nitrogen formed after emission.

Figure 3 assembles quantitative anchors from four different intervention and exposure contexts. Panel A shows full-cycle pig-house emission reductions under amino-acid-balanced crude-protein reduction; Panel B shows high short-term removal by a full-scale indoor acid-scrubber prototype; Panel C shows the concentration-duration pattern of broiler ocular injury; and Panel D shows primary-PM reduction accompanied by increases in NH₃ and odour in manure-drying-tunnel operation. The panels are deliberately not pooled: their denominators, system boundaries and outcomes differ. Together they demonstrate why effect magnitude must always be reported with experimental context.

5. Mitigation Across the Nitrogen Cascade

5.1. Nutritional, Litter and Manure Controls

The first mitigation barrier is to avoid creating surplus excreted nitrogen. Precision and phase feeding, formulation on digestible amino acids, appropriate protein quality, reduced feed wastage and maintenance of animal health can lower the substrate available for NH₃ formation. In finishing pigs, lowering dietary crude protein changed manure composition and NH₃ emissions, and full-cycle room monitoring later showed that amino-acid-balanced crude-protein reduction decreased mean NH₃ emissions by 33.0% and 57.2% in moderate and deeper reduction treatments [57,60]. The response is not a licence for indiscriminate protein restriction: indispensable amino-acid adequacy, intake, growth, carcass quality, reproductive performance and diet cost must be retained.

Across cattle and pig manure studies, meta-analysis supports lower NH₃ potential when dietary crude protein is reduced, although the size of the response varies with species, baseline protein, nitrogen partitioning and manure management [61]. Poultry formulation can likewise reduce nitrogen excretion, but excretion should not be presented as measured atmospheric emission unless litter conversion and airflow were quantified [62]. Nutritional trials should report analysed diet, intake, animal product nitrogen, urinary and fecal nitrogen where feasible, litter or slurry TAN, and directly measured emission.

A 2026 commercial pig-house study illustrates both promise and interpretive caution. A novel feed-based intervention reduced high-frequency indoor NH₃ concentration by 81.7% in an emptied weaning unit but by 20.9% in a growing unit containing pre-existing slurry [63]. The contrast suggests that legacy manure can buffer or delay a fresh-excreta intervention. Because the primary endpoint was indoor concentration, ventilation-weighted mass emission and downstream nitrogen fate remain necessary before the result can be generalized as whole-system abatement.

After excretion, rapid separation or removal can interrupt the contact between urinary nitrogen and fecal enzymes. Partly slatted floors that remain clean, reduced pit area, frequent scraping or flushing, urine-feces separation and manure belts can reduce residence time and exposed area. Their efficacy depends on reliable operation: fouled solid floors, stagnant channels and infrequent removal can negate the design [20]. Mechanical removal also transfers nitrogen to another stage, so storage must be ready to receive the higher ammonium load.

In poultry, litter moisture, pH and caking are central. Drinker management, ventilation sufficient for moisture control, appropriate bedding depth and removal of wet zones reduce hydrolysis and volatilization. Acidifying amendments lower the un-ionized NH₃ fraction, whereas adsorbents and mineral amendments may bind ammonium or alter water activity. A 2026 experiment with 476 broilers found that a 40% zeolite-clay-rice-husk-ash supplement reduced end-of-cycle in-house NH₃ by about 50%, improved stress and footpad indicators, and did not impair growth [64]. The formulation is promising for humid systems, but commercial adoption requires assessment of amendment mass, dust, spent-litter nutrient value, soil effects and cost.

Windrowing, turning and drying require multi-pollutant attention. Manipulation can accelerate stabilization and pathogen reduction but may expose new surfaces and generate short emission peaks. Broiler-house downtime windrowing has been associated with NH₃ and nitrous oxide emissions that vary strongly during the management cycle [65]. Similarly, manure-drying tunnels evaluated as dust filters reduced primary PM₁₀ while increasing NH₃ and odour relative to a calculated no-tunnel comparator [59]. A control device should not be selected from one pollutant column alone.

5.2. Housing, Exhaust Treatment, Storage and Land Application

Ventilation remains essential for heat, humidity, carbon dioxide and contaminant control. The goal is not simply maximum airflow, but demand-responsive ventilation that maintains thermal and respiratory conditions while limiting unnecessary turbulence over emitting surfaces. Source-oriented extraction, localized manure-channel capture and balanced inlet design can separate clean occupied zones from source zones. Continuous sensors can trigger corrective actions, but concentration control should be paired with outlet mass measurement to determine whether the intervention reduced or merely exported NH₃.

Residual NH₃ in mechanically exhausted air can be captured by acid scrubbers, biofilters or biotrickling filters. Acid scrubbing converts NH₃ to an ammonium salt that may have fertilizer value. An early ARS scrubber demonstrated the engineering concept, and a 2025 full-scale indoor prototype achieved approximately 87-99% removal during controlled shop tests across 10-100 ppm and two airflow rates [58,66]. These short-duration tests establish technical efficacy but not annual performance in an occupied, dusty building. Field evaluation must measure pressure drop, energy, reagent use, corrosion, fouling, water consumption, maintenance, captured-nitrogen quality, worker safety and uptime.

Biological treatment can reduce NH₃ and odour but may transform nitrogen into nitrous oxide. Literature-based assessment has shown that N₂O from biological treatment can materially erode greenhouse-gas benefits [67]. Acid systems shift the burden toward chemical supply and an ammonium-rich residual, whereas biological systems require control of oxygen, alkalinity, temperature and microbial pathways. The appropriate comparison is avoided receptor exposure plus recovered nitrogen minus energy, reagent, greenhouse-gas and residual-management burdens.

Storage and field application complete the intervention. Covers reduce wind exchange; cooling and acidification lower volatilization; and shorter storage may reduce cumulative loss where logistics permit. At land application, injection, trailing hose or shoe systems and prompt incorporation can reduce surface exposure, although soil type, slope, crop stage, compaction, fuel use and worker safety affect suitability. Reviews of manure-management chains show that technologies interact and that control of NH₃ can change methane or nitrous oxide [68,69]. Guidance therefore places feeding, housing, storage and field measures within a single sequence [70].

The conservation paradox is fundamental: successful barn control retains more ammonium in manure. If the material is stored uncovered or surface-applied outside crop demand, part of the conserved nitrogen can be re-emitted or lost as nitrate and N₂O. A defensible assessment follows nitrogen from feed to animal product, excreta, housing, storage, treatment, field, crop and residual loss. The numerator should be useful nitrogen retained in animal or crop product; the accompanying ledger should include NH₃, N₂O, nitrate, energy and cost.

Table 4 organizes interventions as a hierarchy rather than a menu of independent percentages. Prevention precedes suppression, exposure control, capture and downstream conservation. Each barrier is paired with a failure mode and a minimum verification metric, because local removal efficiency alone is insufficient.

6. Integration, Trade-Offs and Research Priorities

Four paradoxes define the next stage of livestock-ammonia science. The first is dilution: lower indoor concentration can coexist with unchanged or greater exported mass. The second is conservation: nitrogen retained at one stage remains available for loss at the next. The third is single-pollutant optimization: control of NH₃ may increase N₂O, odour, particulate matter, energy use or aqueous nitrogen, and control of primary dust may increase NH₃. The fourth is productivity neutrality: animals can maintain average gain while experiencing ocular, behavioural, inflammatory or molecular injury. Each paradox is resolved by widening the system boundary and measuring the intended receptor together with nitrogen mass and co-pollutants.

Climate change adds a fifth challenge. Temperature increases the fraction of TAN present as NH₃ and accelerates biological and physical transfer. Jiang et al. [15] estimated 5-22% greater global agricultural NH₃ emissions under plausible late-century warming and found that six linked mitigation measures would deliver smaller percentage reductions under warmer conditions. These are global model results rather than farm trials, but they identify a practical need: emission factors, housing controls, manure storage and national commitments should be stress-tested under future temperature and ventilation regimes.

Standardized reporting is essential. Housing studies should provide animal category, live mass, productivity, diet, manure system, floor and pit geometry, litter or slurry properties, weather, ventilation method, analyser calibration, averaging interval, missing-data treatment and uncertainty. Concentration should be reported separately from emission rate; emission factors should include their denominator and system boundary. Comparative databases can reveal large regional and temporal differences, but apparent improvement may partly reflect study design and should be confirmed with harmonized measurements [23,26].

Animal studies require genuine replication at the chamber or room level. Assigning many animals to one room per treatment creates pseudoreplication because the room, not the animal, receives the airborne treatment. Future experiments should emphasize lower, fluctuating concentrations representative of commercial conditions, include recovery phases, and measure ocular, respiratory, immune, behavioural and performance outcomes. Field studies should use longitudinal designs and combine sensor time series with pathology or validated clinical endpoints. For calves and pigs, hierarchical models should account for pen, room and farm clustering.

Human studies should move from proximity alone to source-pathway-receptor designs. Personal and fence-line NH₃, size-fractionated particles, ammonium, nitrate, sulfate, endotoxin, microbial markers, meteorology and work task should be measured concurrently. Repeated lung function, symptom, medication and vulnerability data can then be linked to time-resolved exposures. Community analyses should assess distributional impacts explicitly, following the spatial inequities identified in recent national work [14]. Causal attribution will remain limited without chemical speciation, dispersion analysis and suitable comparison populations.

Monitoring innovation should focus on traceability rather than novelty alone. Sensor networks need reference co-location, humidity and temperature challenge tests, documented drift correction, data-completeness thresholds and uncertainty propagation into emissions. Facility-scale inverse dispersion, mobile monitoring and satellite constraints can provide independent checks on bottom-up factors. Nitrogen-isotope ratios offer an additional means of tracing agricultural plumes across landscapes [56], and controlled sheep-manure incubation suggests that isotope discrimination can support estimation of cumulative NH₃ loss under defined conditions [71]. Such methods are promising verification tools but require source-specific calibration and cannot replace direct flux measurement without validation.

Intervention trials should report sequential remaining mass rather than add percentage reductions. If source prevention reduces NH₃ formation by 50% and a downstream device removes 80% of the remaining mass, the combined reduction is 90%, not 130%, provided that the second device treats the full residual flow under comparable conditions. Commercial studies should extend for at least one annual cycle, report uptime and maintenance, and follow retained nitrogen through storage and crop uptake. Techno-economic evaluation should include capital, energy, reagent, labour, corrosion, safety, fertilizer credit and farm-scale distribution.

Policy instruments act at different points and should not be conflated. Occupational limits protect workers; animal-welfare rules address housing conditions; permits and national commitments regulate emission mass; ambient standards address receptor concentration; and nutrient policies influence storage and land application. The European National Emission Reduction Commitments Directive establishes national NH₃ commitments, while the recast ambient-air directive strengthens PM₂.₅ protection and monitoring of rural-background ammonia and particulate ions [72,73]. Policy portfolios should reward verified whole-chain reduction and avoid incentives that shift nitrogen to another pollutant, location or farm stage.

Table 5 translates these needs into answerable research questions and minimum study designs. The priorities are deliberately cross-disciplinary because the most consequential knowledge gaps sit between nutrition, building engineering, toxicology, atmospheric science and agronomy.

7. Conclusions

Livestock NH₃ is best understood as a mobile form of reactive nitrogen whose impacts change across space, time and chemical form. Dietary surplus and incomplete retention create the source pool; species-specific excreta chemistry and microbial conversion generate TAN; pH, temperature, moisture, residence time, exposed area and airflow determine release. Indoors, the gas contributes to ocular and respiratory injury and occurs within a mixed exposure environment. Outdoors, it deposits to ecosystems or forms secondary ammonium-containing PM₂.₅, extending the health and environmental footprint beyond the farm.

No single technology can resolve this continuum. The most robust strategy prevents surplus nitrogen, separates or removes excreta rapidly, controls moisture and pH, uses ventilation to protect occupied zones while measuring mass export, captures residual NH₃ where feasible, and conserves the retained ammonium through storage and crop-timed application. Recent 2025-2026 findings reinforce the need for seasonal emission measurements, climate-sensitive policy, updated housing factors and commercially practical feed, bedding and scrubbing interventions. They also show why concentration reduction, device efficiency and regional association must each retain their correct interpretation.

Future progress depends on paired outcomes: nitrogen mass and receptor dose; NH₃ and co-pollutants; animal productivity and welfare; device efficacy and annual durability; barn reduction and downstream fate. When these endpoints are measured together, ammonia abatement becomes more than pollution control. It becomes a framework for improving nitrogen-use efficiency, air quality, animal and human health, ecosystem protection and fertilizer circularity.