NH₃ release during the evaporation of different types of atmospheric precipitation: A case study in Changchun, China

Abstract

Ammonia (NH₃) is a key precursor of secondary inorganic aerosols. During precipitation, NH₃ in the atmosphere can be captured by rain and converted to NH₄⁺, whereas during evaporation, NH₄⁺ can become NH₃ and be released again. The northeastern region of China experiences diverse precipitation types, making the study of the NH₃ release flux and its influencing factors during evaporation highly significant. In this study, precipitation samples of haze (HZ), dust (DS), convective (CC), and monsoon (MN) events were collected three times in Changchun from March to September 2024 (a total of twelve rain events), and indoor simulation evaporation experiments were conducted. The results revealed significant differences in the NH₄⁺ conversion rate (R), NH₃ release flux (F) and release rate (V) across the precipitation types (P < 0.05). The NH₃ flux released from precipitation evaporation was 20.33 µg/m² in spring and 64.53 µg/m² in summer, accounting for approximately 4.14% and 7.70%, respectively, of the corresponding atmospheric NH₃ concentrations. Meteorological factors influenced NH₃ release similarly across precipitation types. R peaked and then decreased with increasing temperature and was significantly negatively correlated with wind speed and precipitation amount (P < 0.05). In addition, this study calculates the temperature coefficient (K₁), wind speed coefficient (K₂), and precipitation amount coefficient (K₃) by considering these factors. These findings provide valuable insights for estimating NH₃ release fluxes from precipitation evaporation in different regions.

Introduction

Ammonia (NH₃), the most abundant alkaline gas in the atmosphere¹ readily neutralizes acidic gases in the air, forming secondary inorganic aerosols such as ammonium sulfate, ammonium bisulfate, ammonium nitrate, and ammonium chloride^2,3. Ammonium (NH₄⁺) accounts for up to 57% of fine particulate matter (PM_2.5)^4,5,6 significantly affecting human health⁷ climate models⁸ nutrient cycling⁹ the atmospheric radiation balance and air quality¹⁰. Recent satellite monitoring has indicated a rise in global atmospheric NH₃ concentrations, driven by factors such as decreased emissions of acidic gases (e.g., SO₂ and NO₂), rising temperatures, and increased fertilizer use¹¹. This trend has transformed urban atmospheres into NH₃-rich environments, raising widespread concern regarding NH₃ sources and concentrations.

Previous studies have identified agricultural activities, such as fertilization and livestock farming, as the primary sources of atmospheric NH₃>^12,13. Other significant contributors include vehicle emissions¹⁴ coal and biomass combustion¹⁵ industrial and power plant NH₃ emissions¹⁶ and dew volatilization¹⁷. NH₃ released from these sources enters the atmosphere, and owing to its water-soluble properties, it can be absorbed during precipitation formation and settle to the surface with rainfall^18,19 leading to high concentrations of NH₄⁺ in atmospheric precipitation. For example, NH₄⁺ was the most abundant ion in precipitation from Ya’an¹⁵ and Dalian, China²⁰ with concentrations of 169.6 µEq/L and 107.76 µEq/L, respectively, accounting for 36% and 19.75% of the total ion concentration. High concentrations of NH₄⁺ are likely to be re-emitted as NH₃ during the evaporation of precipitation. Previous studies reported a rapid increase in near-surface NH₃ concentrations 1–2 h after specific rainfall events in northern Colorado, USA¹⁷ as well as in Beijing¹ and Nanchang, China²¹.

Precipitation serves as an indirect indicator of local air pollution levels, with its chemical composition and amount varying significantly on the basis of meteorological conditions, air mass trajectories, and pollution sources. For example, precipitation in Guiyang is influenced primarily by atmospheric circulation and terrain, with the majority occurring in summer as convective rain. In contrast, spring is dominated by persistent precipitation caused by stratiform clouds. The major ions in precipitation are SO₄²⁻ and Ca²⁺, with NH₄⁺ concentrations of approximately 50 µEq/L²². In comparison, Beijing experiences frequent dust storms in spring, and precipitation is mostly of the dust type²³. The main ions in precipitation are SO₄²⁻ and NH₄⁺, with the NH₄⁺ concentration reaching as high as 346 µEq/L²⁴ approximately seven times higher than those observed in Guiyang. Unlike Guiyang and Beijing, precipitation in Changchun is mainly haze-type and dust-type during spring, whereas convective and monsoonal rainfall prevails in summer. Precipitation in Changchun contains high concentrations of SO₄²⁻ and Mg²⁺, with an average NH₄⁺ concentration of 104.54 ± 35.97 µEq/L²⁵.

Although NH₄⁺ concentrations in rain are relatively high in many regions^15,20 systematic studies on the emission flux of NH₃ during the evaporation process remain limited. Most existing research has focused on specific geographic areas²² with insufficient consideration of the combined effects of precipitation chemistry, precipitation amount, and key meteorological factors such as temperature and wind speed. In addition, some methods used to estimate NH₃ fluxes lack validation under real environmental conditions, resulting in uncertainties. Therefore, a more comprehensive investigation is needed to characterize the emission patterns and driving mechanisms of NH₃ under diverse precipitation types and meteorological conditions. The objectives of this study are as follows:

• 1. To analyze the chemical composition of different types of precipitation in Changchun.

• 2. To quantify the conversion rate (R), release flux (F), and release rate (V) of NH₄⁺ to NH₃ during evaporation.

• 3. To assess the influence of key meteorological (temperature, wind speed, and precipitation amount) and chemical (ionic composition and pH) factors and to propose corresponding correction factors.

Materials and methods

Experimental site and sample collection

The experimental study was conducted in Changchun City, Jilin Province (Fig. 1(a)). Changchun, located in northeastern China (43.88°N, 125.35°E), is a central city in the Northeast Asia economic zone and is a former industrial hub. The city covers a total area of 24,744.6 km² and has a population exceeding 9.1 million.

On the basis of air quality monitoring data collected during the 24 h preceding precipitation events (PM_2.5 and PM₁₀), backward air mass trajectories, and precipitation characteristics, the precipitation events in Changchun were classified into four types: haze type (HZ), dust type (DS), monsoon type (MN), and convective type (CC).

• HZ: PM_2.5 > 50 µg/m³ and PM_2.5/PM₁₀ > 0.7;

• DS: PM₁₀ > 100 µg/m³ and PM_2.5/PM₁₀ < 0.4;

• MN: precipitation duration > 5 h, total rainfall > 10 mm, with moisture trajectories originating from oceanic sources;

• CC: precipitation duration < 1 h, accompanied by strong convection or thunderstorm activity.

Changchun is located in the high-latitude northern temperate zone and experiences a typical continental monsoon climate. Both precipitation types and meteorological conditions exhibit pronounced seasonal variations. In spring, inland airflow from the Eurasian continent dominated, leading to a higher frequency of DS and HZ events. In contrast, summer precipitation is influenced primarily by Pacific-sourced moisture transport and thermally induced surface convection, resulting in more frequent MN and CC events.

Precipitation samples were collected from March 15, 2024, to September 15, 2024, with three sampling events for each precipitation type. Although the number of events was limited, the sampling strategy was designed to reflect the dominant precipitation types and seasonal variation in Changchun during this period. This approach enabled a controlled comparative analysis under a consistent classification scheme and standardized indoor simulation conditions. For each precipitation type, the average outdoor temperature and wind speed during the evaporation period, as well as the total precipitation amount were recorded (Table 1). Specifically, the post-rainfall evaporation period was set to 5 h for DS and HZ events and 3 h for CC and MN events.

Table 1 Outdoor meteorological conditions during the evaporation phases of precipitation.

Precipitation samples were collected from a lawn located approximately 50 m from the main road on the campus of Jilin Jianzhu University. No industrial pollution sources or buildings exceeding 30 m in height were located in the vicinity, and the surrounding area lacked dense vegetation canopies or open water bodies, making this location representative of typical urban precipitation conditions. Importantly, the samples were collected directly during rainfall events using pre-cleaned glass beakers, without any contact with terrestrial or aquatic surfaces (e.g., forest litter, rivers, soil). This approach ensured that the chemical composition of the samples solely reflected atmospheric wet deposition, without influence from post-depositional surface interactions. Prior to use, the beakers were thoroughly rinsed with deionized water (18.25 MΩ cm) and dried in an oven to minimize dilution effects from residual water. The beakers were placed within 30 min before the onset of each precipitation event, and samples were collected within 30 min after the event ended. Under these conditions, the influence of dry deposition on the samples can be considered negligible. In addition, to meet the analytical requirements, the collected volume of each sample exceeded 20 mL.

Sample analysis and laboratory simulations

The samples were filtered within 30 min of collection via a 0.45 μm microporous membrane filter. The pH and electrical conductivity (EC) were measured using a Leici DZS-706-A multiparameter analyzer immediately after filtration to prevent alterations caused by contamination or storage (Fig. 1(b)). The concentrations of four anions (Cl⁻, NO₂⁻, NO₃⁻, and SO₄²⁻) and five cations (NH₄⁺, Na⁺, K⁺, Mg²⁺, and Ca²⁺) in the precipitate were determined via ion chromatography (Thermo Fisher, USA). The anions were separated via an AS-19 column with 20 mM KOH as the eluent. Cations were separated via a CS-15 column with methanesulfonic acid as the isocratic eluent. All the samples were stored in the dark at ~ 4 °C until further analysis.

After analyzing the chemical compositions of the four types of precipitation, we quantified the NH₃ release flux and rate through complete evaporation experiments. A single-factor experimental design was employed to examine the factors influencing the NH₄⁺ conversion rate. On the basis of the seasonal characteristics of the precipitation types presented in Table 1, different evaporation conditions were set to compare the effects of temperature and wind speed on the NH₄⁺ conversion rates during precipitation evaporation. The effect of the rainfall amount on NH₃ release was assessed by adjusting the sample volume in the glass test tube to simulate different rainfall amounts.

The experimental setup is shown in Fig. 1(c). The samples were placed in a glass test tube, with one end connected to an air compressor pump supplying clean zero air and the other end linked to a URG-2000-30*150-3CSS annular denuder to collect the NH₃ released during precipitation evaporation. The denuder was cleaned and coated before use, following a procedure detailed in previous studies²⁵. During evaporation, some NH₄⁺ can volatilize as gaseous NH₃ and N₂. The released NH₃ is captured by the annular denuder, while N₂ is released directly into the atmosphere. Moreover, some NH₄⁺ remains as ammonium salts in the glass test tube. Once the solution fully evaporated, 20 mL of deionized water was added to the test tube, and the tube was thoroughly shaken to extract residual components remaining after complete evaporation. Subsequently, 10 mL of deionized water was added to the denuder, which was slowly rotated for ~ 30 s to extract trapped substances. (Fig. 1(d)). Both extracts were analyzed for NH₄⁺ concentration via ion chromatography, and the data were used to calculate the NH₄⁺ release ratio and residual rate during the precipitation evaporation phase.

Fig. 1

Experimental process flow diagram.

(a) Sample collection location and method; (b) pretreatment prior to the experiment; (c) laboratory-simulated precipitation evaporation and experimental variable gradients (red bold text indicates the baseline experimental conditions used to investigate the effects of other variables on the NH₄⁺ conversion rate); (1) glass test tube; (2) annular denuder; (d) sample testing.

Data analysis

The ion balance (ΔC) of the samples was determined via the following formula on the basis of the ion concentration (µEq/L) determined via ion chromatography²⁶:

$$\:\begin{array}{c}\varDelta\:C=\frac{\left[{NO}_{3}^{-}\right]+\left[{Cl}^{-}\right]+\left[{SO}_{4}^{2-}\right]+{[NO}_{2}^{-}]}{\left[{NH}_{4}^{+}\right]+\left[{Ca}^{2+}\right]+\left[{Na}^{+}\right]+\left[{K}^{+}\right]+\left[{Mg}^{2+}\right]}\end{array}$$

(1)

The NH₄⁺ conversion rate (R) during evaporation was determined from the ratio of NH₄⁺absorption by the denuder to the corresponding ion concentration in the precipitation. This conversion rate can be calculated via the following Eq²⁷:

$$\:\begin{array}{c}R=\frac{{C}_{g}}{{C}_{i}}\times\:100\%\end{array}$$

(2)

$$\:\begin{array}{c}F=R\times\:\frac{{C}_{i}}{18.031}\times\:17.031\times\:I\times\:1000\end{array}$$

(3)

$$\:\begin{array}{c}V=\frac{F}{T}\times\:1000\end{array}$$

(4)

where C_g is the concentration of NH₄⁺ in the denuder (mg/L), C_i is the concentration of NH₄⁺ in the precipitation (mg/L), F is the daily flux of NH₃ released during precipitation evaporation (µg/m²), I is the rainfall depth (mm), 1000 is a conversion factor, V is the rate of NH₃ release during precipitation evaporation (ng/m²·s), and T is the evaporation duration (s).

The volume-weighted mean (VWM) of the chemical components of precipitation was calculated via the following formula²⁸:

$$\:\begin{array}{c}VWM=\frac{{\sum\:}_{i=1}^{n}\left({P}_{i}{M}_{i}\right)}{{\sum\:}_{i=1}^{n}{P}_{i}}\end{array}$$

(5)

where P_i represents the depth of each rainfall event (mm), M_i represents the concentration of a given chemical element (µEq/L), the pH value or electrical conductivity (EC; µS/cm), and n represents the total number of rainfall events.

The total dissolved solids (TDS) in precipitation can be estimated from EC via the following Eq²⁹:

$$\:\begin{array}{c}TDS\approx\:{k}_{e}EC\end{array}$$

(6)

where TDS is the total dissolved solids (mg/L), EC is the electrical conductivity (µS/cm), and K_e is an empirical correlation factor, which typically ranges from 0.55 to 0.8, with an average value of K_e ≈ 0.7.

A previous study showed that the proportion of NH₄⁺ released as a gas during evaporation depends on the difference in concentration between nonvolatile ions and the concentration of NH₄⁺itself. This value can be calculated via the following Eq¹⁷:

$$\:\begin{array}{c}Frac\left({NH}_{3}\right)=\frac{{\left[{NH}_{4}^{+}\right]}_{i}-\left(\sum\:\left(nonvolatile\:anions\right)-\sum\:\left(\text{nonvolatile\:}cations\right)\right)}{{\left[{NH}_{4}^{+}\right]}_{i}}\end{array}$$

(7)

where Frac (NH₃) is the theoretical fraction of NH₄⁺ that is converted to NH₃ during evaporation, [NH₄⁺]_i is the initial NH₄⁺ concentration in solution (µEq/L), ∑nonvolatile anions is the sum of the concentrations of nonvolatile anions (e.g., SO₄²⁻, Cl⁻, and NO₃⁻), and ∑nonvolatile cations is the sum of the concentrations of nonvolatile cations (e.g., Ca²⁺, K⁺, Mg²⁺, and Na⁺).

The coefficients K_n (n = 1,2,3), which represent the influence of various meteorological conditions on NH₄⁺ conversion in precipitation, were calculated on the basis of the R value measured from the indoor simulated evaporation experiment and the theoretical conversion rate (Frac (NH₃)) calculated via Eq. (7). The general formula for calculating the meteorological influence coefficient is as follows:

$$\:\begin{array}{c}{K}_{n}=\frac{R}{Frac\left({NH}_{3}\right)}\end{array}$$

(8)

where K₁, K₂, and K₃ denote the temperature coefficient, wind speed coefficient, and precipitation amount coefficient, respectively.

Statistical analysis was conducted via SPSS 25.0. Q‒Q probability plots were used to determine the distributions of cations (Ca²⁺, Mg²⁺, Na⁺, K⁺, and NH₄⁺) and anions (NO₂⁻, Cl⁻, NO₃⁻, and SO₄²⁻). Given the limited sample size for each event type (n = 3), nonparametric statistical methods were applied to assess differences between groups. A significance level of P < 0.05 was adopted for reference.

The 24-hour concentration data of air pollutants, including PM_2.5 and PM₁₀, in Changchun are freely available from the China National Environmental Monitoring Center (https://quotsoft.net/air/). Meteorological parameters for the study area, such as the wind speed, temperature, and precipitation, were obtained from the automatic weather station (MK-111-LR) at Jilin Jianzhu University. Airflow characteristics near the study area are analyzed via the hybrid single-particle Lagrangian integrated trajectory (HYSPLIT) model provided by the National Oceanic and Atmospheric Administration (NOAA) (https://www.ready.noaa.gov/HYSPLIT_traj.php)^30,31. This model uses an altitude of 500 m as the boundary layer height to generate 24-hour backward trajectories of air masses. Data on the atmospheric quality and fire hotspot distribution along air mass pathways can be obtained from the Fire Information for Resource Management System (FIRMS) (https://firms.modaps.eosdis.nasa.gov/).

Results and discussion

Chemical composition of different precipitation types

According to statistics, there were 20 precipitation days in spring in Changchun, with a total rainfall of 15.2 mm. Among them, HZ and DS occurred three times each, generally accompanied by low rainfall amounts. In summer, precipitation was predominantly classified as MN, which was influenced mainly by East Pacific circulation and typhoons. In addition, the concentrations of PM_2.5 and PM₁₀ in spring were significantly higher than those in summer (P < 0.05). Different meteorological conditions and air quality levels can result in variations in the chemical compositions of different types of precipitation.

pH is a fundamental and widely recognized indicator for assessing precipitation acidity¹⁵. Fig 2 demonstrated that the pH values of the four types of rainfall ranged from 5.75 to 7.41, with significant differences (P < 0.05). The VWM pH, determined from rainfall data via Eq. (5), followed the sequence HZ < MN < CC < DS (P < 0.05). All the measured values exceeded the pH of natural precipitation in equilibrium with atmospheric CO₂ (pH = 5.6)²⁶. Consequently, the four types of precipitation were considered nonacidic and unlikely to cause acid damage to the environment, which is consistent with the findings of Wu et al.³².

EC is a crucial indicator of atmospheric pollution, with elevated values typically signifying increased pollution levels³³. Figure 2 revealed that the EC of various types of precipitation followed the order DS > HZ > MN > CC (P < 0.05), with all values surpassing the average EC of 14.8 µS/cm at Waliguan Mountain, a background site in China³⁴. Comparison analysis revealed notable seasonal differences, with the EC of spring precipitation being significantly greater than that of summer precipitation (P < 0.05). This trend reflects the improved air quality in summer compared with that in spring. On the one hand, the greater amount and frequency of rainfall in summer reduce the amount of suspended particulate pollutants in the atmosphere, resulting in lower EC values. On the other hand, the low temperatures and high atmospheric pressure characteristic of spring frequently create stagnant wind conditions, inhibiting the dispersion of atmospheric particulates and leaving more pollutants suspended in the air³⁵.

According to Fig. 2, the predominant water-soluble ions in HZ, DS, CC, and MN were SO₄²⁻, NO₃⁻, Ca²⁺, and NH₄⁺, accounting for 74%, 75%, 74%, and 67% of the total ion concentrations in these types of precipitation, respectively. Although the major cation and anion components were the same, the ion concentrations in HZ and DS were significantly greater than those in CC and MN (P < 0.05), reflecting marked seasonal variation. This disparity can be attributed to the difference in rainfall amounts between spring and summer, which resulted in varying levels of ion dilution. Additionally, the intensification of anthropogenic activities in spring, such as agricultural practices, urban construction, transportation, and biomass burning (e.g., straw or waste combustion), significantly enhances the emission of sulfur dioxide, nitrogen oxides, and particulate matter, thereby shaping the unique seasonal characteristics of the different precipitation types³⁶. A comparative analysis of precipitation types within the same season indicated that HZ and DS presented similar ion compositions, as did CC and MN. This similarity can be explained by the shared sources of pollutants. Spring haze formation is predominantly affected by meteorological drivers, pollutant sources, and atmospheric circulation patterns rather than by coal burning or heating emissions, distinguishing it from the haze formation mechanisms observed in autumn and winter^37,38. As a result, the ionic characteristics of HZ and DS were consistent. The ionic composition of precipitation driven by intense atmospheric convection under the strong influence of summer ocean monsoons is likely governed by marine air masses. Consequently, CC and MN exhibited analogous structural compositions.

The ratio of total anions to total cations (ΔC) is commonly used as an indicator of the relative amounts of major ionic components³⁹. The calculations indicated that the total ion concentration ratio of anions to cations in all the precipitation samples was 0.93 ± 0.07, suggesting a deficiency of anions. This anion deficiency could be attributed to unmonitored inorganic anions (e.g., F⁻, PO₄³⁻) and organic anions (e.g., HCOO⁻, CH₃COO⁻) present in the precipitation⁴⁰. Additionally, construction dust from nearby areas and the naturally alkaline properties of northern China’s soil likely contributed to the enrichment of cations such as Ca²⁺ and Mg^2+41,42. Fig 2 demonstrated that the VWM of the total ion concentration in spring precipitation was approximately double that in summer precipitation. This observation was consistent with Rao et al. (2016), who reported that analyte concentrations were higher during dry periods than during wet periods, emphasizing the critical impact of climatic conditions on the composition of precipitation.

Fig. 2

VWMs of ion concentrations (µEq/L), pH, EC (µS/cm), and total ion concentrations for various types of precipitation (numbers in circles).

To examine the characteristics of airflow movement near the study area and identify the primary sources of ions in different precipitation types, specific dates were chosen to represent HZ, DS, CC, and MN: April 12, March 27, August 20, and July 22, respectively. Starting at an altitude of 500 m—selected to represent regional air mass transport while minimizing the effects of surface-layer turbulence—24-hour backward trajectories were calculated, as illustrated in Fig. 3. The trajectories for different precipitation types showed significant differences and distinct seasonal patterns: in spring, terrestrial input predominated with poorer air quality, whereas summer air masses mainly originated from marine sources. Air masses generally acquire pollutants as they pass through regions with emission sources⁴³. The air masses associated with the HZ (Fig. 3(a)) followed long trajectories from the northeast, passing through areas with poor air quality, resulting in high concentrations of SO₄²⁻ and NO₃⁻. The air masses for DS (Fig. 3(b)) predominantly originated from eastern Russia, with long-range transport of soil dust resulting in the enrichment of terrestrial components such as Ca²⁺ and Mg²⁺⁴⁴. CC air masses (Fig. 3(c))involve primarily localized transport between northeastern cities, with significant contributions from anthropogenic pollution⁴⁵. MN air masses (Fig. 3(d)) are influenced by typhoon weather, characterized by long trajectories and fast-moving, clean marine airflow. After passing through regions of poor air quality in the Sea of Japan, these air masses carry some anthropogenic pollutants⁴⁶.

Fig. 3

Backward trajectories of air masses and air quality along their pathways for various types of rainfall events: (a) HZ (4.12); (b) HD (3.27); (c) CC (8.20); (d) MN (7.22).Map data source: Tianditu Map Import service via the original plugin (version 2021b (9.85)) (https://www.originlab.com).

NH₄ ⁺ conversion rate, release flux and rate of NH₃ in different precipitation types

The transformation of NH₄⁺ to NH₃ during evaporation is governed primarily by the acid–base equilibrium between NH₄⁺ and NH₃ (NH₄⁺ ⇌ NH₃ + H⁺, pKa ≈ 9.25), which is highly sensitive to changes in pH and temperature^47,48. Under near-neutral or slightly acidic conditions (pH < 7), inorganic nitrogen predominantly exists as NH₄⁺, making NH₃ volatilization thermodynamically unfavorable. However, as evaporation progresses, localized increases in pH may occur due to preferential water loss and ionic concentration effects, slightly shifting the equilibrium toward NH₃. This shift can increase the volatilization of NH₃. Additionally, the thermal instability and semi volatility of NH₄NO₃ at temperatures above 25 °C may further contribute to NH₃ emissions⁴⁹. Therefore, the observed unrecovered NH₄⁺ (26.90 ± 13.21%) may have resulted from the combined effects of the NH₄⁺/NH₃ acid–base equilibrium and NH₄NO₃ volatilization under the experimental conditions.

The R values for the different precipitation types, ranked from high to low, were DS > HZ > CC > MN (P < 0.05), with average values of 61.67 ± 9.71%, 61.00 ± 1.73%, 45.67 ± 9.02%, and 38.67 ± 6.35%, respectively (Fig. 4). This indicates that approximately half of the NH₄⁺ in the precipitation in the study area was released back into the boundary layer, facilitating nitrogen deposition. Assuming complete evaporation of all atmospheric precipitation, the NH₃ release flux (F) and release rate (V) were calculated via Eqs. (3) and (4), considering the different precipitation amounts and evaporation durations. The results revealed significant differences in both the release flux (F) and release rate (V) among the different precipitation types. In spring, the F values for HZ and DS were 1714.85 ± 1662.88 µg/m² and 4010.58 ± 2184.05 µg/m², respectively, with corresponding V values of 93.34 ± 112.32 ng/m²·s and 216.43 ± 132.37 ng/m²·s, respectively. In summer, the F values for CC and MN were 905.39 ± 662.89 µg/m² and 11192.50 ± 6580.20 µg/m², respectively, whereas the V values were 52.45 ± 39.03 ng/m²·s and 714.44 ± 408.37 ng/m²·s, respectively. These differences may result from variations in precipitation characteristics and evaporation durations among the different rain types. Notably, within the same precipitation type, large variations in the amount of precipitation among individual events can lead to substantial fluctuations in F values, thereby contributing to the large standard deviations observed.

On the basis of the total precipitation during the study period and an actual evaporation ratio of approximately 6%⁵⁰, the average NH₃ release flux from rainfall evaporation was estimated to be 20.33 µg/m² in spring and 64.53 µg/m² in summer. To further assess the potential contribution of this flux to atmospheric NH₃ levels, the atmospheric boundary layer was defined as 100 m, which more accurately represents the near-surface environment where NH₃ volatilization and deposition predominantly occur. By comparing the estimated release with the average atmospheric NH₃ concentrations in northeastern China⁵¹—4.83 µg/m³ in spring and 8.44 µg/m³ in summer—NH₃ released through rainfall evaporation accounted for approximately 4.14% of the total NH₃ concentration in spring and 7.70% of the total NH₃ concentration in summer. This further corroborated the phenomenon whereby NH₃ surges near the ground surface within 1–2 hours after precipitation.

Fig. 4

NH₄⁺ conversion rate (R), NH₃ release flux (F), and release rate (V) during the evaporation of various types of precipitation.Experimental conditions: 15 °C, 1 m/s, 1 mL.

The Frac (NH₃) for each precipitation event was calculated via Eq. (7). If Frac (NH₃) ≥ 1, all NH₄⁺ is released as NH₃ into the atmosphere. If 0 < Frac (NH₃) < 1, part of the NH₄⁺ is released as NH₃, while the remaining NH₄⁺ forms ammonium salts with SO₄²⁻, Cl⁻, and NO₃⁻, which are left as residues. If Frac (NH₃) < 0, all NH₄⁺ remains in the form of ammonium salts, with no NH₃ released. The experimentally obtained relationship between R and Frac (NH₃) is shown in Fig. 5. The fitting coefficient (R² was 0.92, indicating a strong correlation. The observed discrepancy may be due to the failure to account for small amounts of formate, acetate, and other low-molecular-weight organic acids in the samples, which could slightly overestimate the theoretical values compared with the actual values¹⁷.

Fig. 5

Actual and theoretical NH₄⁺ conversion rates during the evaporation of precipitation.

Factors influencing the NH₄ ⁺ conversion rate during the evaporation of different types of precipitation

Solution evaporation is a complex process involving both chemical and physical mechanisms, with outcomes varying significantly depending on environmental conditions. Takenaka et al. (2009) reported that when an ammonium acetate solution rapidly evaporates at high temperatures (100 °C), the solute tends to deposit as a solid on the contact surface. In contrast, when the solution evaporates slowly at room temperature, the solute is released primarily in the gas phase, with minimal solid residue remaining on the surface. These findings indicate that the drying rate of the solution significantly influences the final fate of the solute; evaporation rates are affected by various meteorological factors, including temperature, wind speed, and droplet size¹⁹. To further investigate the effects of temperature, wind speed, and droplet size on the NH₄⁺ conversion rate during the evaporation of different types of precipitation, this study utilized natural precipitation samples collected on April 15, April 24, August 22, and July 22, representing HZ, DS, CC, and MN, respectively. Indoor simulation experiments were conducted to measure the R value under various experimental conditions. The results provide new insights into the dynamic characteristics of nitrogen cycling during precipitation evaporation and its broader environmental implications.

Effects of temperature on the NH₄ ⁺ conversion rate during the evaporation of different precipitation types

As shown in Fig. 6, the temperature significantly influenced R during the evaporation of different types of precipitation, with a consistent pattern across all types. Overall, R increased to a peak and then decreased with increasing temperature, suggesting that temperatures within a specific range promoted the conversion of NH₄⁺ to NH₃. However, excessively high temperatures may trigger adverse reactions, further increasing the conversion rate. At low temperatures (T ≤ 10 °C), R was generally low. As the temperature increased from 10 °C to 25 °C, R increased significantly, peaking at 25 °C. This increase was likely due to the high concentrations of NO₃⁻ and NH₄⁺ in the precipitation, which, under optimal conditions, combined to form secondary ammonium salts (NH₄NO₃). These ammonium salts are thermally unstable and particularly sensitive to temperature changes. Studies have shown that when the temperature is between 5 °C and 25 °C and the relative humidity exceeds 40%, increasing the temperature accelerates the decomposition of ammonium salts, shifting the reaction equilibrium toward NH₃ release, which significantly enhances the ammonia flux and increases R⁵². From a molecular dynamics perspective, higher temperatures increase molecular motion in solution, increasing the distance between molecules and weakening intermolecular interactions. This reduction in NH₃ solubility in solution facilitated the conversion of NH₄⁺ into NH₃, allowing it to escape more readily from the solution⁵³. However, as the temperature continued to rise, R decreased, reaching a minimum at 35 °C. This was likely due to the increased evaporation rate at higher temperatures, which caused water molecules to evaporate more rapidly⁴⁸. Rapid evaporation increased the concentration of solutes in solution, leading to the precipitation of poorly soluble salts once their solubility limit was reached, removing them from the evaporation process. Concurrently, ions with higher solubilities, such as NH₄⁺ and NO₂⁻, remained dissolved and continued to react to form NH₄NO₂. NH₄NO₂ tends to decompose into N₂ and H₂O at relatively high temperatures¹⁹ leading to the consumption of NH₄⁺ and a further decrease in R.

Fig. 6

Effects of temperature on the NH₄⁺ conversion rate during the evaporation of various rainfall types.

A comparison of different samples revealed significant systematic differences in R during evaporation across the entire temperature range. The conversion rates followed the order HZ > DS > CC > MN (P < 0.05), reflecting the varying effects of the chemical composition of different precipitation types on R. This difference can be attributed to the combined effects of pH, excess nonvolatile anions, and the initial concentration of NH₄⁺ in the precipitation⁵⁴. A relatively high pH typically favors the conversion of NH₄⁺ to NH₃. However, the presence of excess nonvolatile anions in precipitation shifts the chemical equilibrium, reducing R¹⁹. Additionally, the initial concentration of NH₄⁺ has a complex effect on the conversion rate²⁵. A higher initial concentration provides more NH₄⁺ but also amplifies the inhibitory effect of nonvolatile anions, further decreasing R. Although the pH of HZ was lower than those of the other precipitation types, the presence of excess nonvolatile anions and a moderate NH₄⁺ concentration contributed to its higher R value. In contrast, DS and CC exhibited a stronger inhibitory effect from excess nonvolatile anions, which prevented these samples from achieving high R values, despite having higher pH values. Finally, MN, with the lowest initial NH₄⁺ concentration and a low pH, presented the lowest R value during evaporation.

Effect of wind speed on the NH₄ ⁺ conversion rate during the evaporation of different precipitation types

As shown in Fig. 7, the R values during the evaporation of the four different types of precipitation ranged between 20% and 80%, exhibiting a significant decreasing trend with increasing WS. Specifically, R was negatively correlated with WS (P < 0.05). This trend indicated that wind speed had a pronounced inhibitory effect on the conversion of NH₄⁺ to NH₃ during precipitation evaporation. The main reason for this may be that an increase in the wind speed altered the microenvironment of the solution. On the one hand, higher wind speeds accelerated evaporation at the solution surface, and the cooling effect during evaporation lowered the surface temperature. Under these lower temperature conditions, NH₄⁺ tends to combine with acidic substances in solution to form ammonium salts, which remain in the liquid phase rather than being released as a gas. This resulted in a significant decrease in R with increasing wind speed⁵⁵. On the other hand, increased wind speed reduced the thickness of the stagnant boundary layer at the gas‒liquid interface, making it easier for water molecules to diffuse into the gas phase. This phenomenon accelerated evaporation, causing the concentration of solutes to gradually increase. As a result, some poorly soluble metal salts may have crystallized out due to supersaturation, limiting the escape of NH₄⁺ to the gas phase and causing R to decrease progressively.

Fig. 7

Effects of the wind speed on the NH₄⁺ conversion rate during the evaporation of various rainfall types.

In addition, all types of precipitation presented different rates of decrease in R across the range of wind speeds, which were influenced mainly by the initial NH₄⁺ concentration and solution saturation. Under calm wind or near-still conditions (wind speed = 0 m/s), the R value of CC was the highest and decreased most rapidly. When the wind speed increased to 3 m/s, R decreased to 19.12%, which was only one quarter of the rate under calm conditions. This suggested that CC was the most sensitive precipitation type to wind speed changes, with its evaporation highly dependent on disturbances at the gas‒liquid interface. Upon closer analysis, although CC had a low total ion concentration, resulting in low solution saturation, its significantly lower NH₄⁺ concentration reduced the inhibitory effect of excess NH₄⁺ on R²⁵ allowing R to vary significantly with changes in wind speed. In contrast, MN exhibited a more stable decreasing trend under different wind speeds, with a difference of only 29.01% between the maximum and minimum conversion rates, indicating that MN was the precipitation type least affected by wind speed changes. This was likely due to the low NH₄⁺ concentration and low solution saturation of MN, which weakened the impact of wind speed on its conversion rate. DS, with the highest solution saturation, showed a greater degree of wind speed-induced variation, although it was still lower than that of CC. Its decline pattern was characterized by an initial rapid decrease followed by a more gradual flattening. HZ, with NH₄⁺ and total ion concentrations intermediate between those of DS and MN, displayed a relatively uniform decreasing pattern, and its response to wind speed changes was also intermediate. Overall, owing to differences in initial NH₄⁺ concentration and solution saturation, the response of R to variations in wind speed differed among the precipitation types, with the order from high to low being CC > DS > HZ > MN (P < 0.05). These differences highlight the driving effect of wind speed on the conversion of NH₄⁺ to NH₃ during precipitation evaporation and its underlying mechanisms.

Effect of the precipitation amount on the NH₄ ⁺ conversion rate during the evaporation of different types of precipitation

The factors influencing solution evaporation include not only meteorological conditions such as temperature, wind speed, relative humidity, and solar radiation but also solution factors such as droplet size and solute concentration⁵⁶. As shown in Fig. 8, the NH₃ release and R results for the different types of precipitation at different sample volumes revealed similar patterns across all precipitation types. Specifically, for 0.5 mL of precipitation, the amount of NH₃ released after evaporation ranged from 0.4 to 1.4 mg/L, while the R values remained between 60% and 75%. However, as the sample volume increased to 5 mL, the NH₃ release increased to 1.5–5.5 mg/L, whereas R significantly decreased to less than 30%. This suggested that increasing the sample volume significantly inhibited the conversion of NH₄⁺ to NH₃ during evaporation, with a negative correlation observed between the two factors (P < 0.05). By calculating the ratio of nonvolatile anions to nonvolatile cations in the samples, it was found that excess nonvolatile anions were present in all types of precipitation. As the sample volume increased, excess nonvolatile anions gradually accumulated in the solution, inhibiting the conversion of NH₄⁺ and resulting in a decrease in R. Moreover, as evaporation progressed, the reduction in water molecules caused metal cations in the precipitate to combine with nonvolatile anions such as SO₄²⁻, NO₃⁻, and Cl⁻, forming salts with low solubility that exited the evaporation system as crystals. This weakened the promoting effect of metal cations on NH₄⁺ conversion, leading to the observed decline in R with increasing sample volume. On the other hand, NH₃ release clearly increased with increasing sample volume, indicating a positive correlation between sample volume and NH₃ release (P < 0.05). This resulted from the increased NH₄⁺ concentration in the precipitate as the sample volume increased, leading to more NH₃ being released after evaporation. Furthermore, the NH₃ release levels for the different types of precipitation followed the order DS > HZ > CC > MN (P < 0.05), which was closely related to the initial NH₄⁺ content in the precipitation.

Fig. 8

Effects of the amount of precipitation on the NH₄⁺ conversion rate and NH₃ release during the evaporation of various rainfall types.

Temperature coefficient, wind speed coefficient, and precipitation amount coefficient affect NH₃ release

On the basis of the results of the indoor simulation experiments conducted under different meteorological conditions to determine R, as well as the Frac (NH₃) values calculated via Eq. (7), the temperature coefficient (K₁), wind speed coefficient (K₂), and precipitation amount coefficient (K₃) affecting NH₄⁺ conversion during precipitation evaporation were further calculated via Eq. (8) (Table 2). The results showed that R was controlled not only by the chemical parameters of precipitation, such as ion concentration and pH but also by meteorological conditions. Specifically, with the same chemical parameter values, the total NH₃ release during precipitation evaporation in moderate-temperature regions may be greater than that in colder or hotter regions, whereas the F value of precipitation in areas with mild winds and an arid climate may be lower than that in regions with strong winds and heavy rainfall. Therefore, the R values in different regions are constrained by both the chemical composition of precipitation and the local meteorological conditions, resulting in significant differences in the total NH₃ flux released after the evaporation of precipitation across various regions.

Table 2 Temperature coefficient, wind speed coefficient, and precipitation amount coefficient.

By comprehensively considering the impacts of temperature, wind speed, and precipitation amount and quantitatively analyzing these factors, the changes in NH₃ flux under different spatial and temporal conditions can be predicted more accurately. This study provides valuable data for subsequent research and serves as a scientific basis for optimizing environmental management measures.

Conclusions

This study revealed that all precipitation types exhibited nonacidic characteristics and were dominated by water-soluble ions such as SO₄²⁻, NO₃⁻, Ca²⁺, and NH₄⁺. Significant differences were observed in the NH₄⁺ conversion rates among the precipitation types, and the release of NH₃ exhibited clear seasonal pattern. Meteorological factors, including temperature, wind speed, and precipitation, play critical roles in influencing NH₃ volatilization. In addition, the NH₃ fluxes released from precipitation evaporation were 20.33 µg/m² in spring and 64.53 µg/m² in summer, accounting for approximately 4.14% and 7.70%, respectively, of the corresponding atmospheric NH₃ concentrations.

By integrating chemical composition analysis with meteorological analysis, this study offers a novel approach to quantify NH₃ release from precipitation evaporation at local scales. These findings underscore the importance of precipitation evaporation as a nonnegligible source of atmospheric NH₃ and provide a scientific basis for improving ammonia emission inventories and atmospheric nitrogen management strategies.

This study is limited by a relatively small sample size and the absence of surface-level interactions under field conditions. Key environments factors such as relative humidity and photochemical processes were not considered, which may constrain the spatial applicability of the results. Moreover, in natural settings, post-depositional interactions with land, vegetation, and water bodies can alter NH₄⁺ concentrations and pH, thereby influencing NH₃ volatilization. Future research should incorporate field observations, large datasets, and coupled modeling approaches to capture a broader range of environmental variables and improve the mechanistic understanding of NH₄⁺ to NH₃ conversion under complex atmospheric and surface-mediated conditions.