Change in nitrogen (N) cycle is one of the causes of surface-earth eco-environmental and climatic changes in the Anthropocene, which is related to the quality of water-soil-air environments, food production, ecosystem structure and function, and climatic feedback. Synthetic ammonia (NH₃) is the main material for chemical N fertilizers and NH₃ fuel is also a key material to achieving "carbon neutrality" in the field of energy and power. However, population growth and economic development since the industrial revolution have led to continuous increases in the concentration, pool, and deposition flux of atmospheric NH₃ and ammonium (NH₄⁺). A better understanding of the main terrestrial sources of atmospheric NH₃ is critical for making emission mitigation strategies and assessing the effects of anthropogenic NH₃ pollution.

Due to its high solubility and chemical reactivity, NH₃ has a short atmospheric lifetime, but the atmospheric transport of NH₄⁺ is obvious and extensive, resulting in complex and highly mixed sources of atmospheric NH_x (the sum of NH₃ and NH₄⁺) at specific monitoring sites and underlying surfaces. The NH₃ volatilization from fertilizers and wastes (denoted as v-NH₃) has long been assumed to be the primary NH₃ source at regional and global scales. In recent years, evidence from laboratory simulations, in-situ observations, satellite observations, and more accurate emission inventories points to that fossil-fuel combustion sources dominated by coal combustion (industrial coal combustion in the urban and surrounding areas, and domestic coal combustion in non-urban areas) and oil combustion (urban traffic as a regional hotspot, and vehicles are also widely distributed in non-urban areas), as well as biomass burning dominated by crop straw and wildfires, emit a large amount of NH₃. However, it is difficult to accurately constrain the emission factors and intensity of the widely-existing combustion-related NH₃ (c-NH₃), leading to uncertainty in the fractional contribution and amount of c-NH₃ to regional NH₃ emissions.

Since the 1950s, stable N isotope (δ¹⁵N) has gradually been recognized as an effective tool for tracing or differentiating NH₃ emission sources. Until now, site-based δ¹⁵N observations of atmospheric NH_x have been conducted widely in high N-pollution regions of East Asia, North America, and Europe (Fig. 1). However, it remains a challenge to constrain the post-emission δ¹⁵N changes of NH₃ due to transformations in the atmosphere, which prevents a quantitative assessment of source contributions to NH₃ emissions.

Based on the above background, we constrained the N isotope effect between atmospheric NH₃, particulate NH₄⁺ (p-NH₄⁺), or precipitation NH₄⁺ (w-NH₄⁺) and the initial NH₃ mixture (i-NH₃). Then, we reconstructed δ¹⁵N of i-NH₃ (δ¹⁵N_i-NH3) for the isotopic observation sites in East Asia, North America, and Europe. We found that the δ¹⁵N_i-NH3 was higher in North America than in Europe and East Asia and generally decreased with time in the three study regions. These results indicate the relative proportion of c-NH₃ in North America is higher than that in the other two regions, but the relative proportions of v-NH₃ have been increasing during the past decades. Finally, in combination with the δ¹⁵N of major v-NH₃ and c-NH₃ emission sources and amounts of regional v-NH₃ emissions, we estimated the relative contributions and amounts of major c-NH₃ emissions in the three regions using δ¹⁵N mass-balance methods, showing 40±21% and 6.6±3.4 Tg N yr^-1 in East Asia, 49±16% and 2.8±0.9 Tg N yr^-1 in North America, and 44±19% and 2.8±1.3 Tg N yr^-1 in Europe (Fig. 1). All these results reveal that c-NH₃ is an important source of NH₃ emissions and provide new data for regional c-NH₃ and total NH₃ emissions (Fig. 1). This work provides new scientific evidence for making current and future mitigation strategies of NH₃ emissions, evaluating NH_x deposition fluxes and effects.