1. Introduction
The ammonia (NH3) cycle in the natural environment is a significant contributing factor to nitrification, eutrophication, air quality, the planetary radiation budget, and the health of humans, animals, and plants. These effects are engendered both by the agency of the free gas itself and as a result of its conversion to fine particulates through reaction paths with acidic vapors in the atmosphere [1,2]. Ammonia plays a crucial role in the formation of particulate matter (PM) through its reactions with atmospheric acids to form ammonium sulfate and ammonium nitrate aerosols. An accurate measurement of ammonia emissions is therefore essential for understanding and managing PM2.5 pollution, which has significant health implications [3]. Previous studies have indicated that Federal Reference Method (FRM) measurements may underestimate PM2.5 concentrations due to volatility-related losses of semi-volatile components like ammonium nitrate during sampling [4]. This underestimation poses challenges for air quality management, especially in regions where secondary nitrate constitutes a substantial fraction of ambient PM.
The salience of ammonia in regard to all of these micro- to macroscale phenomenologies is evidenced by the convening of dedicated discussion workshops, e.g., [5], and mounting recommendations for increased vigilance on ammonia emissions, especially with regard to PM pollution, e.g., [6]. Hence, the recent tightening of the National Ambient Air Quality Standards (NAAQS) for PM2.5 to 9 μg/m3 imposes even stricter requirements for attainment [7] and highlights the need for improved ammonia measurement techniques that can accurately capture spatial and temporal variations. The regional- and global-scale distribution of ammonia is now routinely tracked by a variety of space-based sensors [8,9,10,11]. Although such products go some way to addressing the ammonia measurement deficiencies identified previously [12,13], the coarse 10–50 km pixel size of these satellite measurements cannot resolve the local-scale variability in ammonia abundance and consequently cannot distinguish the often compact sources of emissions with any degree of fidelity. This hampers regulators and local authorities as they consider methods for controlling the most egregious sources of ammonia [14].
This work demonstrates the utility of airborne longwave-infrared (LWIR) hyperspectral imagery with the requisite spatiospectral resolution and radiometric sensitivity to characterize both diffuse and discrete sources of ammonia emission. The locations selected to carry out this demonstration were areas surrounding California’s Salton Sea, specifically in the Eastern Coachella Valley and Northern Imperial Valley, a region known for its intensive agriculture and animal husbandry. Largely as a result of the copious ammonia deposited into the atmosphere from these activities, the Imperial Valley has been one of the most consistent PM non-attainment regions in the United States [15]. High ambient ammonia concentrations are known to be detrimental to the health of plants [16,17] and animals [18,19] and thus adversely impact the productivity of both agriculture and animal husbandry operations. In this respect, the Imperial Valley shares many similarities with California’s San Joaquin Valley (SJV), which has been identified as one of the single largest sources of atmospheric free ammonia worldwide [20,21]. It has been variously estimated that between 70% [22] and 90% [14] of the atmospheric ammonia emitted by the SJV is sourced from animal waste and fertilizer application. Given the parallels in land use with the SJV, we may expect the Imperial Valley to be similarly afflicted, but it is also subject to significant ammonia contributions produced by the adjacent Salton Sea [23], hydrothermal vents [24], and as a waste product of geothermal power generation [25].
2. Materials and Methods
The data used for this study were acquired as part of a larger investigation into multi-scale ammonia transport that will be the topic of a separate paper. Airborne data acquisition operations were coordinated with ground measurements carried out with stationary and mobile sensor systems.
2.1. Airborne LWIR Hyperspectral Imager
The airborne segment of this study was carried out with The Aerospace Corporation’s Mako imager. Mako is a 3-axis-stabilized, whiskbroom hyperspectral LWIR imager [26,27] whose relevant performance specifications are given in Table 1. The flight altitude for the current study was ~3750 m above ground level (AGL), such that the along-track ground sample distance (GSD) was nominally 2.1 m. Standard operating practice is to carry out a radiometric calibration against hot and cold National Institute of Standards and Technology (NIST) traceable blackbodies immediately prior to and following each imaging line [27].
2.2. Mobile Sensor Suite
The mobile ground measurements for this work were obtained with an instrumented vehicle (Figure 1) built and operated by the University of California, Riverside (UCR), whose full instrument suite has been described elsewhere [28].
The mobile observatory was used to continuously measure geolocated atmospheric ammonia levels on-road. The platform is a Mercedes Sprinter van with dimensions approximately 7.3 m long, 2.1 m wide, and 3.0 m tall. The mobile lab is equipped with a cavity ring-down spectrometer (G2123 Analyzer for NH3/H2O; Picarro, Sunnyvale, CA, USA) to measure ammonia over the range 0–50 ppmv (
2.3. Ground-Based Static Air Quality Station
An additional series of airborne acquisitions was carried out centered on a static air quality monitoring station located at 33.57202°N, 116.06382°W in the Mecca township. Operated by the South Coast Air Quality Management District (AQMD) [29], this monitoring facility records ambient ammonia concentrations with a Picarro G2301 instrument (Picarro, Sunnyvale, CA, USA) at 5 m AGL, along with standard meteorological parameters at 9.7 m AGL, on a 1 min cadence. Quality assurance procedures for this instrument involve monthly zero and span checks.
3. Results
The principal region selected for this study is the area of the Imperial Valley southeast of the Salton Sea. The area is one of mixed arable cultivation, fallow plots, and cattle feedlots. Waste produced by the cattle is collected in holding ponds from whence it is applied to fertilize adjacent crop land, resulting in a diffuse ammonia background, within which is embedded a profusion of discrete sources that emit plumes above the background level.
3.1. Ground-Based Mobile Measurements
The mobile ground-based data collection occurred from 13:30 to 00:40 UTC (local time +7 h) on 25–26 September 2023, covering an area from the southern boundary of the Salton Sea, south to Brawley township. Driving routes were planned in order to sample the environs of ammonia plumes that were identified during previous aircraft campaigns and included cattle feedlots and geothermal power plants. During the sampling period, the driving speed was controlled to 8–16 km/h.
3.2. Airborne Measurements
The Imperial Valley study area was imaged in three overlapping flightlines (Figure 2) on 25 September 2023 under clear skies in two stages, from 17:28 to 19:25 UTC (a.m. local time), with a repeat acquisition over the same area from 22:54 to 00:27 UTC (p.m. local time). The Mecca locale was imaged prior to each acquisition of the Imperial Valley area, also as depicted in Figure 2. These collection intervals were 16:35–17:11 and 22:06–22:39 UTC.
Included in Figure 2 are the a.m. route of the ground-based mobile observatory and the locations of the South Coast AQMD air quality monitoring station and Niland-English Rd. meteorological station (Section 3.3).
3.3. Airborne Retrievals
The ubiquity of ammonia signals throughout the scenes examined meant that the in-scene atmospheric compensation approach [30] would be subject to unacceptable levels of error in the retrieved column content. Ammonia column densities were therefore computed using the custom tool SAGE (Scene-based Algorithm for Gas Estimation) [31]. SAGE uses a nonlinear radiative transfer model to treat the interactions between the gas(es) of interest, the background scene underlying the air mass under analysis, and the atmospheric column between Mako and the target air volume. A scene substantially devoid of the target gases must be identified to compute background conditions for the SAGE retrieval algorithm. Such scenes were generally found at the extrema of the flightlines depicted in Figure 2. The modeled backgrounds are determined using principal components analysis, and MODTRAN®6 [32] is used to model the atmospheric column, following which matrix regression is applied to estimate the column density of the target gas in each image pixel. SAGE also reports the t-statistic, which describes the quality of the overall spectral fit to assess a confidence level for a given retrieval. Also output are the fit residuals and thermal contrast, ΔT, between the near-surface air parcel and underlying surface for each pixel. Column density values for each scene pixel are filtered with a t-statistic threshold ≥3.0 and |ΔT| ≥ 2.0 °C in order to reject retrievals with high uncertainty.
Ammonia column content over the areas surveyed was ascertained on a per-pixel basis. The surface brightness temperatures required for determining ΔT were computed from the atmospherically compensated radiance imagery using a standard inverse Planck treatment. The near-surface air temperature field corresponding to the imagery was derived from the NOAA High-Resolution Rapid Refresh (HRRR) model [33,34], which outputs values at various elevations that can be used to interpolate values at the desired 10 m elevation. Although not used in the analysis, the HRRR wind field outputs over the study areas are provided in Appendix A for reference purposes.
Errors in ΔT are one of the principal sources of retrieval inaccuracy, and since neither the spatial nor temporal resolutions of the HRRR data are commensurate with those of the field data, an intercomparison of the HRRR output was made with air temperature measurements from the UCR mobile laboratory and the nearest static weather stations to the two domains surveyed. These were the Niland-English Rd. meteorological station located at 33.21350°N, 115.54514°W (operated by the Imperial County Air Pollution Control District,
From Figure 3, it can be seen that the near-coincident HRRR air temperatures for the most part agree notably well with the corresponding UCR mobile measurements, providing confidence that HRRR is accurately modeling the atmospheric state within the study area. Figure 3 also indicates that during the morning airborne collection window, the static air temperature readings diverge significantly (2–5 °C) from the contemporaneous HRRR data and UCR mobile measurements, indicating that the ABL conditions had not yet stabilized, i.e., were not geographically uniform across the study area. This contrasts with the situation prevailing during the afternoon airborne collection window, where air temperatures from all sources agree to ~1 °C, indicative of a stable, uniform ABL.
Of the two Imperial Valley airborne surveys, the morning flight provided the greatest degree of overlap with the UCR mobile measurements within the most prolific ammonia-generating areas. It was therefore selected for the purpose of the current study and processed using the corresponding HRRR model output. Figure 4 shows the resulting ammonia column density map, in which the light gray areas signify that the ammonia signal strength (t-statistic) and/or the assessed ΔT did not meet the retrieval acceptance criteria. Figure 5 highlights the Mako/mobile laboratory overlap region and identifies the principal sources of ammonia within the scene.
The five compact sources (G) located in the western portion of Figure 5 originate from geothermal power plants, which expel unwanted ammonia that occurs as a byproduct of hypersaline brine extraction [35]. The strong composite plume to the east results from the coalescence of multiple distributed emissions from an extensive concentrated animal feeding operation (CAFO), denoted by ‘O’ in Figure 5, with the probable addition of other diffuse emissions from recently fertilized arable plots downwind of the CAFO.
Note that many of the data voids apparent in Figure 4 and Figure 5 correlate with cultural modification of the underlying terrain. This is due to the wide variability in land surface temperature between land use classes [36], which leads to a corresponding variance in the ΔT. Hence vegetated fields, which exhibit a considerably lower ΔT than fallow plots, more frequently fail to meet the retrieval acceptance criteria, resulting in a higher incidence of null records.
3.4. Airborne Retrieval Validation
In order to compare the surface and airborne measurements, it is necessary to convert the latter from column density to absolute concentration. Because the airborne measurements lack vertical resolution, it is necessary to rely on auxiliary information, albeit this constitutes an unavoidable source of potential uncertainty. The capping inversion that frequently marks the top of the ABL inhibits gas and aerosol exchange between the ABL and the free troposphere, and this is especially true for flat terrain uninfluenced by topographical variability [37]. The regions under study here meet these criteria, so that to first order the ammonia measured by Mako was assumed to be confined within the ABL. For regions not immediately downwind from identified sources, the ABL depths from the HRRR model were used to convert the measured column densities to mean absolute concentrations. For locations directly downwind of identifiable plumes, a Gaussian plume model for continuous sources [38] was utilized to estimate the plume depth, which in turn yielded mean concentration estimates. The model attributed to Briggs [39] provides horizontal and vertical spread profiles, σy and σz, for Gaussian plumes between 100 m and 10 km in length as functions of both the distance from the plume source and the atmospheric stability type [40]. Pasquill [40] classifies plume turbulence into six categories from “moderately stable” to “extremely unstable,” parameterized by wind speed and solar insolation. Since the cross-sectional width of the plume was easily determined by visual inspection of the Mako ammonia detection imagery, the atmospheric stability type could be elucidated by inverting the Briggs horizontal spread equation. Plume distance and width at the in situ downwind location were measured by overlaying the Mako geolocated column density imagery onto a Google Earth map of the region. A detectable cloud width of 4σy for Gaussian plumes was used to infer the atmospheric stability class. This parameter, along with the distance from the source to the UCR mobile lab downwind location, yielded the vertical spread of the plume from the Briggs equations. A cloud height of 4σz was used for the plume depth.
To account for variances in geolocation accuracy between the participating sensor systems, a square array of pixels nominally centered on the target in situ data point was selected. In addition, the pixel acceptance criteria required that the correlated airborne and in situ measurements be contemporaneous within 15 min.
3.5. Imperial Valley Surveys
For the reasons discussed above, it transpired that much of the Imperial Valley data could not be used to validate the airborne ammonia retrievals. The differences between the airborne data and corresponding mobile lab measurements were found to range widely by up to an order of magnitude, as shown in Figure 6. The Mako plots in this figure were generated using the full ABL depth output by the HRRR model. The times given are for the in situ data. The corresponding Mako data points were acquired within 30 min of the in situ data point to which they are coregistered. The worst agreement with the in situ measurements occurs during the mobile lab transects of the CAFO plume around 17:18 and 17:27 UTC. In those instances the proximity of the mobile lab to the plume sources means that uniform vertical distribution of ammonia within the ABL cannot be assumed, as the actual plume depth could be much less at such close proximity to the source. This situation would be further aggravated by the complexity and dynamic nature of the sources in the study area.
Examining these cases further, Table 3 provides a set of data where the UCR mobile lab had transected the CAFO plumes highlighted in Figure 5, where it was possible to estimate plume depth using the procedure described above. It is apparent from this figure that there are very few mobile lab data points downwind of the CAFO. During this timeframe, the mobile monitor deviated from its nominal measurement cadence, resulting in significant data gaps. For locations downwind of a plume source, the randomly shifting winds during a 30 min timeframe would impact concentration measurements, hence a 25 × 25 pixel array (corresponding to a 50 m × 50 m box) was used to provide statistics on concentration values in the vicinity of the mobile lab location. Table 3 shows three measurements made by the in situ instrumentation approximately 250 m downwind of the CAFO (locations labeled A–C in Figure 5), along with corresponding Mako statistics for the overhead passes. The Mako retrieval results are reported as the median, standard deviation (σ), and peak ammonia concentration values within the 25 × 25 pixel array. It is clear that there are marked discrepancies between the in situ measurements and the nearest available airborne retrievals, with mobile ground observations exceeding the median and even maximum Mako inferred ammonia concentrations. The Gaussian spread equations estimated a depth of 120 m for these plumes at the location of the mobile lab, which was traveling along the road that marks the eastern boundary of the CAFO.
Three additional sites (designated D–F in Figure 5 and Table 3) were analyzed, which were sufficiently distant from ammonia sources that the modeled plume depth exceeded the HRRR ABL depth. Accordingly, in each of these cases the assumed plume depth was set equal to the HRRR ABL depth for the purpose of estimating the remotely inferred absolute concentrations. Sites D and F were located within the far downwind plumes from geothermal power plants, while site E was located in the downwind extension of the composite plume emitted by the CAFO. The analysis results are given in Table 3, from which it can be seen that none of these sites show agreement between the airborne and in situ measurements. These observations emphasize the difficulties entailed in correlating in situ with remotely sensed measurements and underscore the importance of stable, uniform atmospheric conditions when attempting such intercomparisons.
3.6. Mecca Area Surveys
The area around the South Coast AQMD air-monitoring station in Mecca is considerably more quiescent with respect to ammonia generation and therefore represents a more well-mixed airmass better suited to the validation exercise, though the atmospheric abundances of ammonia are much lower than in the Imperial Valley. However, neither of the Mecca airborne surveys overlapped with the mobile ground-based survey, so that the only opportunity to intercompare against in situ records is with the measurements acquired by the South Coast AQMD monitoring station.
Figure 7 shows the ammonia column density map for the Mecca region during the afternoon survey, where again the gray zones signify that the ammonia signal strength (t-statistic) and/or ΔT failed to meet the retrieval acceptance criteria. Note that the South Coast AQMD monitoring station was located at the periphery of a large ammonia cloud, in an area where the ammonia column density was relatively low.
Figure 8 shows the airborne ammonia column density retrievals in the vicinity of the South Coast AQMD monitoring station during both the morning and afternoon overpasses, the contrast between which is striking. The absence of retrievals in the morning passes is attributable to the significantly reduced ΔT and ABL depth in the morning relative to the afternoon. These factors combine to raise the minimum detectable concentration (MDC) of ammonia above the ambient levels recorded by the South Coast AQMD monitoring station during the morning overpasses.
The procedure for calculating the MDC derives from the formalism detailed in [27]. For the 2 m GSD and urban clutter statistics applicable to the Mecca locale, the Noise-Equivalent Concentration Length for unity ΔT (NECL) of ammonia at NTP is 16 ppm-m °C. Using the computed values of ΔT and ABL depths (H) from the HRRR model, it is straightforward to estimate the detection threshold values of ammonia concentration for all overpasses of the South Coast AQMD monitoring station site. The resulting MDCs are expressed in terms of absolute concentration (ppbv):
MDC = 3[NECL]/(H·ΔT),(1)
in which the factor of 3 reflects the adjustment to 3σ introduced in [27].The results are provided in Table 4, in which the Mako retrieval results are reported as the median, standard deviation, and peak ammonia concentration values within the 11 × 11 pixel array (22 m × 22 m box) centered on the South Coast AQMD station. It is evident that the ambient ammonia concentration was in all cases well below the Mako MDC during the morning overpasses; note, however, that σ values are biased high by the nulls representing sub-threshold pixels.
The afternoon ammonia concentrations recorded by the South Coast AQMD monitoring station are of a similar order to those in the morning. However, a doubling of the ΔT results in a concomitant doubling of Mako detection sensitivity that, combined with the near quadrupling of the ABL depth, results in a ready observation of the ambient ammonia burden (Table 4). Detection proved to be more difficult for the final afternoon overpass. Ammonia is suspected to have been prevalent in this flightline, and it was not possible to identify a background portion of the scene sufficiently free of ammonia, so that the SAGE retrievals were underestimated in this particular instance. It is apparent from Table 4 that the best overall agreement with the in situ measurements is obtained with the median ammonia concentration retrievals (for those cases not dominated by null pixels).
It should not be inferred from the foregoing discussion that there is a direct dependence of system sensitivity on ABL depth. Rather, the impact on MDC described is attributable to the assumption of a uniform ammonia concentration within the ABL, so that, even though the in situ ammonia concentration measurements remained fairly constant, the increase in ABL depth results in a larger putative ammonia column density. This argument is summarized in Figure 9, in which it can be seen that as ΔT → 0, the MDC → ∞.
4. Discussion
The interspersed CAFOs and intensive agriculture operations that dominate California’s Imperial Valley give rise to high levels of ambient atmospheric ammonia that make this region an ideal natural laboratory for evaluating ammonia remote sensing methodologies. A series of airborne LWIR spectral imagery acquisitions was conducted over the most active portions of the Imperial Valley and coordinated with mobile ground-based in situ measurements. LWIR spectral imaging is a proven high-sensitivity technique for ammonia measurement, but the few published attempts to validate retrieval accuracies serve to highlight the difficulties entailed in doing so, e.g., [41,42]. The Imperial Valley data set is a valuable resource for the investigation of ammonia transport and fate, and it is therefore important to validate its accuracy. However, the profuse quantity of discrete and diffuse ammonia sources, and the variability and inhomogeneity of their emissions, combined with the divergent spatial and temporal characteristics of the airborne and ground-based measurements to preclude use of the Imperial Valley data set for validation purposes.
A separate series of surveys conducted over a South Coast AQMD air-monitoring station located in Mecca, Calif., approximately 65 km to the northwest of the Imperial Valley study site, proved more amenable to the validation task. Although the ammonia abundances measured by this monitoring station were considerably lower than those observed in the Imperial Valley, they were much less influenced by a highly dynamic source environment and provided a more homogeneous, stable airmass with which to carry out the validation intercomparisons. Despite the profusion of factors contributing to the overall uncertainty that accrued to this process, in those cases where atmospheric stability and uniformity were optimal, agreement between the airborne ammonia quantitative retrievals and the corresponding in situ concentration measurements was in the 16–37% range.
Conceptualization, D.M.T., S.H. and O.K.; methodology, D.M.T., C.S.C., S.H. and O.K.; flight planning and airborne data acquisition, E.R.K.; airborne data processing, C.S.C. and K.N.B.; data visualization and plume dispersion analysis, C.S.C.; ground mobile data acquisition and processing, Y.M.; meteorological model data, M.A.; static monitoring station measurements, P.P. and M.S.; writing—original draft, D.M.T.; writing—review and editing, all authors. All authors have read and agreed to the published version of the manuscript.
The original data presented in this study are openly available in Mendeley Data at DOI:10.17632/bc2j7wbwsw.1. Data from the South Coast AQMD air quality monitoring station at Mecca used in the validation are publicly available through the South Coast AQMD Continuous Air-Monitoring Trends Analysis Dashboard:
This research was carried out with support from The Aerospace Corporation’s Independent Research and Development program and Research and Technology Collaborations Office, the University of California, Riverside, and the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration. All trademarks, service marks, and trade names are the property of their respective owners; their appearance does not constitute endorsement by the authors or their institutions.
The authors declare no conflicts of interest.
Footnotes
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Enlarge this image.Figure 2. Geographical setting of the two field study domains considered in this work. The colored rectangles represent the ground coverage extent acquired during the Mako flightlines. The width of each swath is 7.25 km for the Imperial Valley and 7.50 km for Mecca. The Mecca flightlines are centered on the South Coast AQMD air-monitoring station, denoted by the red star. The yellow star marks the location of the Niland-English Rd. meteorological station, and the magenta track shows the route of the UCR mobile laboratory during the a.m. Mako Imperial Valley acquisition.
Enlarge this image.Figure 3. Intercomparison of HRRR air temperatures against UCR mobile laboratory and static meteorological station readings for the a.m. and p.m. survey periods.
Enlarge this image.Figure 4. Retrieved ammonia column densities across the study area for the 17:28–18:13 UTC Mako survey on 25 September 2023. Also shown is the track of the UCR mobile laboratory (concentration key at upper right) and the Niland-English Rd. meteorological station (yellow star). Red rectangles denote image whisks used to compute background conditions.
Enlarge this image.Figure 5. The Mako/mobile laboratory overlap region excerpted from Figure 4, indicating the principal sources of ammonia (G, O) and measurement intercomparison locations (A–F). The yellow star denotes the location of the Niland-English Rd. meteorological station.
Enlarge this image.Figure 6. Comparison between in situ measured ammonia concentrations and corresponding retrieved remote sensing values for the a.m. Imperial Valley acquisition period.
Enlarge this image.Figure 7. Retrieved ammonia column densities in the Mecca region for the 22:06–22:39 UTC Mako survey. The red star marks the location of the South Coast AQMD air-monitoring station. Red rectangles denote image whisks used to compute background conditions.
Enlarge this image.Figure 8. Mako ammonia column density retrievals around the South Coast AQMD air-monitoring station (red star): (a) morning pass, (b) afternoon pass, (c) visible scene image (Google Earth).
Enlarge this image.Figure 9. Dependence of Mako ammonia MDC on ΔT for selection of ABL depths. Uniform ammonia concentration within ABL is assumed.
Operating specifications of the Mako sensor.
| Parameter | Specification |
|---|---|
| Spectral coverage | 7.6–13.2 μm |
| Spectral resolution (128 channels) | 44 nm |
| Instantaneous field of view | 0.56 mrad |
| GSD (12 kft AGL) | 2 m |
| Base frame rate | 3255 Hz |
| Noise-equivalent spectral radiance (10 μm, 4 coadds) | <0.5 μW cm−2 sr−1 μm−1 |
| Noise-equivalent temperature difference (10 μm, 300 K) | 0.03 K |
Operating specifications of mobile lab instrumentation.
| Instrument | Observation | Frequency |
|---|---|---|
| GPS 16X, Garmin Ltd. global positioning unit | Geolocation (latitude, longitude) | 1 Hz |
| METSENS500, Campbell Scientific, Inc. 1 weather station with 2D sonic anemometer | Ambient temperature, relative humidity, pressure, wind speed, and wind direction | 1 Hz |
| Picarro G2123 extended-range NH3 concentration in air analyzer | NH3 mole fraction and % H2O in air | ~0.3 Hz |
1 Campbell Scientific, Inc., Logan, UT, USA.
Plume intercomparisons.
| Site | In Situ Time (UTC) | In situ NH3 (ppbv) | Overpass Time (UTC) | Mako Ammonia Retrieval (ppbv) | Plume Depth (m) | ||
|---|---|---|---|---|---|---|---|
| Median | Std Dev | Peak | |||||
| A | 17:18 | 418 | 17:31 | 62 | 44 | 259 | 120 |
| B | 17:27 | 1146 | 17:52 | 311 | 211 | 1041 | 120 |
| C | 17:28 | 575 | 17:52 | 108 | 65 | 302 | 120 |
| D | 17:45 | 80 | 17:52 | 17 | 6 | 26 | 637 |
| E | 17:55 | 71 | 17:31 | 13 | 5 | 41 | 371 |
| F | 18:11 | 29 | 17:30 | 31 | 11 | 65 | 445 |
South Coast AQMD air-monitoring station overpass specifics.
| Overpass Time (UTC) | In Situ NH3 (ppbv) | Mako Ammonia Retrieval (ppbv) | ΔT (°C) | H (m) | MDC (ppbv) | ||
|---|---|---|---|---|---|---|---|
| Median | Std Dev | Peak | |||||
| 16:40 | 4.98 | 0.00 | 0.00 | 0.00 | 3.64 | 400 | 33.0 |
| 16:53 | 5.29 | 0.00 | 0.00 | 0.00 | 4.05 | 430 | 27.6 |
| 17:09 | 4.87 | 0.00 | 0.00 | 0.00 | 4.46 | 482 | 22.4 |
| 22:11 | 3.81 | 3.21 | 2.17 | 6.47 | 9.35 | 1662 | 3.09 |
| 22:23 | 4.20 | 2.64 | 1.70 | 7.56 | 9.00 | 1690 | 3.15 |
| 22:38 | 4.40 | 0.00 | 0.50 | 2.82 | 7.30 | 1723 | 3.82 |
Appendix A
Enlarge this image.Figure A1. HRRR wind field over the study areas during the 25 September 2023 a.m. Mako acquisitions.
Enlarge this image.Figure A2. HRRR wind field over the study areas during the 25 September 2023 p.m. Mako acquisitions.
References
1. Nowak, J.B.; Neuman, J.A.; Bahreini, R.; Middlebrook, A.M.; Holloway, J.S.; McKeen, S.A.; Parrish, D.D.; Ryerson, T.B.; Trainer, M. Ammonia sources in the California South Coast Air Basin and their impact on ammonium nitrate formation. Geophys. Res. Lett.; 2012; 39, L07804. [DOI: https://dx.doi.org/10.1029/2012GL051197]
2. Paulot, F.; Jacob, D.J. Hidden cost of U.S. agricultural exports: Particulate matter from ammonia emissions. Environ. Sci. Technol.; 2014; 48, pp. 903-908. [DOI: https://dx.doi.org/10.1021/es4034793] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24370064]
3. Xing, Y.F.; Xu, Y.H.; Shi, M.H.; Lian, Y.X. The impact of PM2.5 on the human respiratory system. J. Thorac. Dis.; 2016; 8, pp. E69-E74. [DOI: https://dx.doi.org/10.3978/j.issn.2072-1439.2016.01.19] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26904255]
4. Chiu, Y.T.T.; Carlton, A.G. Aerosol thermodynamics: Nitrate loss from regulatory PM2.5 filters in California. ACS EST Air; 2024; 1, pp. 25-32. [DOI: https://dx.doi.org/10.1021/acsestair.3c00013] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39166529]
5. Sutton, M.A.; Reis, S.; Baker, S.M.H. Atmospheric Ammonia: Detecting Emission Changes and Environmental Impacts; Springer: Dordrecht, The Netherlands, 2009; [DOI: https://dx.doi.org/10.1007/978-1-4020-9121-6]
6. Gu, B.; Zhang, L.; Van Dingenen, R.; Vieno, M.; Van Grinsven, H.J.; Zhang, X.; Zhang, S.; Chen, Y.; Wang, S.; Ren, C. et al. Abating ammonia is more cost-effective than nitrogen oxides for mitigating PM2.5 air pollution. Science; 2021; 374, pp. 758-762. [DOI: https://dx.doi.org/10.1126/science.abf8623]
7. Final Reconsideration of the National Ambient Air Quality Standards for Particulate Matter (PM); U.S. Environmental Protection Agency: Washington, DC, USA, 2024; Available online: https://www.epa.gov/pm-pollution/final-reconsideration-national-ambient-air-quality-standards-particulate-matter-pm (accessed on 18 December 2024).
8. Shephard, M.W.; Cady-Pereira, K.E.; Luo, M.; Henze, D.K.; Pinder, R.W.; Walker, J.T.; Rinsland, C.P.; Bash, J.O.; Zhu, L.; Payne, V.H. et al. TES ammonia retrieval strategy and global observations of the spatial and seasonal variability of ammonia. Atmos. Chem. Phys.; 2011; 11, pp. 10743-10763. [DOI: https://dx.doi.org/10.5194/acp-11-10743-2011]
9. Shephard, M.W.; Dammers, E.; Cady-Pereira, K.E.; Kharol, S.K.; Thompson, J.; Gainariu-Matz, Y.; Zhang, J.; McLinden, C.A.; Kovachik, A.; Moran, M. et al. Ammonia measurements from space with the Cross-track Infrared Sounder: Characteristics and applications. Atmos. Chem. Phys.; 2020; 20, pp. 2277-2302. [DOI: https://dx.doi.org/10.5194/acp-20-2277-2020]
10. Van Damme, M.; Clarisse, L.; Franco, B.; Sutton, M.A.; Erisman, J.W.; Kruit, R.W.; Van Zanten, M.; Whitburn, S.; Hadji-Lazaro, J.; Hurtmans, D. et al. Global, regional and national trends of atmospheric ammonia derived from a decadal (2008–2018) satellite record. Environ. Res. Lett.; 2021; 16, 055017. [DOI: https://dx.doi.org/10.1088/1748-9326/abd5e0]
11. Warner, J.X.; Wei, Z.; Strow, L.L.; Dickerson, R.R.; Nowak, J.B. The global tropospheric ammonia distribution as seen in the 13-year AIRS measurement record. Atmos. Chem. Phys.; 2016; 16, pp. 5467-5479. [DOI: https://dx.doi.org/10.5194/acp-16-5467-2016]
12. Pinder, R.W.; Adams, P.J.; Pandis, S.N.; Gilliland, A.B. Temporally resolved ammonia emission inventories: Current estimates, evaluation tools, and measurement needs. J. Geophys. Res. Atmos.; 2006; 111, D16310. [DOI: https://dx.doi.org/10.1029/2005JD006603]
13. Pinder, R.W.; Walker, J.T.; Bash, J.O.; Cady-Pereira, K.E.; Henze, D.K.; Luo, M.; Osterman, G.B.; Shephard, M.W. Quantifying spatial and seasonal variability in atmospheric ammonia with in situ and space-based observations. Geophys. Res. Lett.; 2011; 38, L04802. [DOI: https://dx.doi.org/10.1029/2010GL046146]
14. Ammonia: Supplemental Information for EPA in Support of 15 µg/m3 Annual PM2.5 Standard; California Air Resources Board: Sacramento, CA, USA, 2023; Available online: https://ww2.arb.ca.gov/sites/default/files/2023-04/AmmoniaSupplementalInformation.pdf (accessed on 18 December 2024).
15. Freedman, F.R.; English, P.; Wagner, J.; Liu, Y.; Venkatram, A.; Tong, D.Q.; Al-Hamdan, M.Z.; Sorek-Hamer, M.; Chatfield, R.; Rivera, A. et al. Spatial particulate fields during high winds in the Imperial Valley, California. Atmosphere; 2020; 11, 88. [DOI: https://dx.doi.org/10.3390/atmos11010088]
16. Van der Eerden, L.J.M.; de Visser, P.H.B.; Van Dijk, C.J. Risk of damage to crops in the direct neighbourhood of ammonia sources. Environ. Pollut.; 1998; 102, pp. 49-53. [DOI: https://dx.doi.org/10.1016/S0269-7491(98)80014-6]
17. Krupa, S.V. Effects of atmospheric ammonia (NH3) on terrestrial vegetation: A review. Environ. Pollut.; 2003; 124, pp. 179-221. [DOI: https://dx.doi.org/10.1016/S0269-7491(02)00434-7]
18. Reece, F.N.; Lott, B.D.; Deaton, J.W. Ammonia in the atmosphere during brooding affects performance of broiler chickens. Poultry Sci.; 1980; 59, pp. 486-488. [DOI: https://dx.doi.org/10.3382/ps.0590486]
19. Chen, D.; Miao, Z.; Peng, M.; Xing, H.; Zhang, H.; Teng, X. The co-expression of circRNA and mRNA in the thymuses of chickens exposed to ammonia. Ecotoxicol. Environ. Safety; 2019; 176, pp. 146-152. [DOI: https://dx.doi.org/10.1016/j.ecoenv.2019.03.076]
20. Clarisse, L.; Shephard, M.W.; Dentener, F.; Hurtmans, D.; Cady-Pereira, K.; Karagulian, F.; Van Damme, M.; Clerbaux, C.; Coheur, P.-F. Satellite monitoring of ammonia: A case study of the San Joaquin Valley. J. Geophys. Res. Atmos.; 2010; 115, D13302. [DOI: https://dx.doi.org/10.1029/2009JD013291]
21. Burns, A.M.; Chandler, G.; Dunham, K.J.; Carlton, A.G. Data gap: Air quality networks miss air pollution from concentrated animal feeding operations. Environ. Sci. Technol.; 2023; 57, pp. 20718-20725. [DOI: https://dx.doi.org/10.1021/acs.est.3c06947]
22. Battye, W.; Aneja, V.P.; Roelle, P.A. Evaluation and improvement of ammonia emissions inventories. Atmos. Environ.; 2003; 37, pp. 3873-3883. [DOI: https://dx.doi.org/10.1016/S1352-2310(03)00343-1]
23. Amrhein, C.; Anderson, M.A.; Reese, B.K. Inorganic Carbon Precipitation and Anaerobic Reactions in Agricultural Wastewater; Soil Carbon and California’s Terrestrial Ecosystems Final Report: 2004218; Kearney Foundation of Soil Science University of California: Davis, CA, USA, 2007; Available online: https://kearney.ucdavis.edu/OLD%20MISSION/2004_Final_Reports/2004218Amrhein%20_FINALkms.pdf (accessed on 18 December 2024).
24. Tratt, D.M.; Young, S.J.; Lynch, D.K.; Buckland, K.N.; Johnson, P.D.; Hall, J.L.; Westberg, K.R.; Polak, M.L.; Kasper, B.P.; Qian, J. Remotely-sensed ammonia emission from fumarolic vents associated with a hydrothermally active fault in the Salton Sea geothermal field, California. J. Geophys. Res. Atmos.; 2011; 116, D21308. [DOI: https://dx.doi.org/10.1029/2011JD016282]
25. Tratt, D.M.; Young, S.J.; Johnson, P.D.; Buckland, K.N.; Lynch, D.K. Multi-year study of remotely-sensed ammonia emission from fumaroles in the Salton Sea Geothermal Field. Proceedings of the 8th IEEE Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing—WHISPERS; Los Angeles, CA, USA, 21–24 August 2016; [DOI: https://dx.doi.org/10.1109/WHISPERS.2016.8071692]
26. Hall, J.L.; Boucher, R.H.; Buckland, K.N.; Gutierrez, D.J.; Keim, E.R.; Tratt, D.M.; Warren, D.W. Mako airborne thermal infrared imaging spectrometer–performance update. Proc. SPIE; 2016; 9976, 997604. [DOI: https://dx.doi.org/10.1117/12.2239245]
27. Buckland, K.N.; Young, S.J.; Keim, E.R.; Johnson, B.R.; Johnson, P.D.; Tratt, D.M. Tracking and quantification of gaseous chemical plumes from anthropogenic emission sources within the Los Angeles Basin. Remote Sens. Environ.; 2017; 201, pp. 275-296. [DOI: https://dx.doi.org/10.1016/j.rse.2017.09.012]
28. Thiruvenkatachari, R.R.; Carranza, V.; Ahangar, F.; Marklein, A.; Hopkins, F.; Venkatram, A. Uncertainty in using dispersion models to estimate methane emissions from manure lagoons in dairies. Agric. For. Meteorol.; 2020; 290, 108011. [DOI: https://dx.doi.org/10.1016/j.agrformet.2020.108011]
29. Site Survey Report for Mecca; South Coast Air Quality Management District: Diamond Bar, CA, USA, 2024; Available online: https://www.aqmd.gov/docs/default-source/clean-air-plans/air-quality-monitoring-network-plan/aaqmnp-saul-martinez.pdf (accessed on 18 December 2024).
30. Young, S.J.; Johnson, B.R.; Hackwell, J.A. An in-scene method for atmospheric compensation of thermal hyperspectral data. J. Geophys. Res. Atmos.; 2002; 107, 4774. [DOI: https://dx.doi.org/10.1029/2001JD001266]
31. Hall, J.L.; Chang, C.S.; Westberg, K.; Tratt, D.M. Scene Analysis for Gas Estimation (SAGE): A tool for quantification of diffuse and point-source atmospheric trace gas abundance. Remote Sens. Environ.; 2025; under review
32. Berk, A.; Conforti, P.; Kennett, R.; Perkins, T.; Hawes, F.; van den Bosch, J. MODTRAN®6: A major upgrade of the MODTRAN® radiative transfer code. Proc. SPIE; 2014; 9088, 90880H. [DOI: https://dx.doi.org/10.1117/12.2050433]
33. Dowell, D.C.; Alexander, C.R.; James, E.P.; Weygandt, S.S.; Benjamin, S.G.; Manikin, G.S.; Blake, B.T.; Brown, J.M.; Olson, J.B.; Hu, M. et al. The High-Resolution Rapid Refresh (HRRR): An hourly updating convection-allowing forecast model. Part I: Motivation and system description. Wea. Forecast.; 2022; 37, pp. 1371-1395. [DOI: https://dx.doi.org/10.1175/WAF-D-21-0151.1]
34. James, E.P.; Alexander, C.R.; Dowell, D.C.; Weygandt, S.S.; Benjamin, S.G.; Manikin, G.S.; Brown, J.M.; Olson, J.B.; Hu, M.; Smirnova, T.G. et al. The High-Resolution Rapid Refresh (HRRR): An hourly updating convection-allowing forecast model. Part II: Forecast performance. Wea. Forecast.; 2022; 37, pp. 1397-1417. [DOI: https://dx.doi.org/10.1175/WAF-D-21-0130.1]
35. Bishop, H.K.; Bricarello, J.R. Scaling and corrosion in an experimental geothermal power plant. J. Petroleum Technol.; 1978; 30, pp. 1240-1242. [DOI: https://dx.doi.org/10.2118/6612-PA]
36. Nega, W.; Balew, A. The relationship between land use land cover and land surface temperature using remote sensing: Systematic reviews of studies globally over the past 5 years. Environ. Sci. Pollut. Res.; 2022; 29, pp. 42493-42508. [DOI: https://dx.doi.org/10.1007/s11356-022-19997-z]
37. Lehning, M.; Richner, H.; Kok, G.L. Transport of air pollutants from the boundary layer to the free troposphere over complex terrain. Phys. Chem. Earth; 1998; 23, pp. 667-672. [DOI: https://dx.doi.org/10.1016/S0079-1946(98)00108-6]
38. Hanna, S.R.; Briggs, G.A.; Hosker, R.P. Handbook on Atmospheric Diffusion; Rep. DOE/TIC-11223 Office of Health and Environmental Research, Dept. of Energy: Oak Ridge, TN, USA, 1982; [DOI: https://dx.doi.org/10.2172/5591108]
39. Briggs, G.A. Diffusion Estimation for Small Emissions; Rep. TID-28289 Atmospheric Turbulence and Diffusion Lab., National Oceanic and Atmospheric Administration: Oak Ridge, TN, USA, 1973; [DOI: https://dx.doi.org/10.2172/5118833]
40. Pasquill, F. The Estimation of the Dispersion of Windborne Material. Meteorol. Mag.; 1961; 90, pp. 33-49. Available online: https://digital.nmla.metoffice.gov.uk/IO_58fec883-2be8-42e2-a3b1-e29cbe13f19e (accessed on 18 December 2024).
41. Van Damme, M.; Clarisse, L.; Dammers, E.; Liu, X.; Nowak, J.B.; Clerbaux, C.; Flechard, C.R.; Galy-Lacaux, C.; Xu, W.; Neuman, J.A. et al. Towards validation of ammonia (NH3) measurements from the IASI satellite. Atmos. Meas. Tech.; 2015; 8, pp. 1575-1591. [DOI: https://dx.doi.org/10.5194/amt-8-1575-2015]
42. Guo, X.; Wang, R.; Pan, D.; Zondlo, M.A.; Clarisse, L.; Van Damme, M.; Whitburn, S.; Coheur, P.F.; Clerbaux, C.; Franco, B. et al. Validation of IASI satellite ammonia observations at the pixel scale using in situ vertical profiles. J. Geophys. Res. Atmos.; 2021; 126, e2020JD033475. [DOI: https://dx.doi.org/10.1029/2020JD033475]
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; Chang, Clement S 1 ; Keim, Eric R 1 ; Buckland, Kerry N 1 ; Alvarez, Morad 1 ; Kalashnikova, Olga 2
; Hasheminassab, Sina 2 ; Garay, Michael J 2 ; Miao, Yaning 3 ; Porter, William C 3 ; Hopkins, Francesca M 3 ; Pakbin, Payam 4 ; Sowlat, Mohammad 4 1 The Aerospace Corporation, Los Angeles, CA 90009, USA;
2 Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91011, USA;
3 Department of Environmental Sciences, University of California, Riverside, CA 92521, USA;
4 South Coast Air Quality Management District, Diamond Bar, CA 91765, USA;