🔬 physics

Mapping ammonia emission plumes using shortwave infrared imaging spectroscopy

This paper demonstrates that shortwave infrared imaging spectroscopy, utilizing reflected sunlight rather than thermal emission, can effectively quantify atmospheric ammonia emissions from industrial sources, as validated by Tanager-1 satellite data from Pakistan and Uzbekistan.

Original authors: Nicholas Balasus, Daniel H. Cusworth, Jinsol Kim, Daniel J. Varon, Charles E. Miller, Riley M. Duren

Published 2026-07-15

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Nicholas Balasus, Daniel H. Cusworth, Jinsol Kim, Daniel J. Varon, Charles E. Miller, Riley M. Duren

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the atmosphere as a giant, invisible ocean of air that we all share. Sometimes, this ocean gets polluted with invisible gases that can hurt our health and damage nature. One such gas is ammonia. You might know it from the sharp smell of cleaning supplies, but in the sky, it mostly comes from farms and factories. When too much ammonia floats around, it can turn into tiny, harmful dust particles that we breathe in, or it can fall onto forests and lakes, acting like a toxic fertilizer that upsets the delicate balance of nature.

For a long time, scientists have tried to track these invisible ammonia clouds using special cameras that look at "heat." Think of it like trying to spot a warm cup of coffee in a cold room; the camera sees the heat difference. But this method has some big problems. It only works well if the ground is a different temperature than the air, it struggles to see small or hidden sources, and the cameras on satellites are often too blurry to see anything smaller than a whole city block. It's like trying to read a newspaper from a mile away using a telescope that only sees in the dark.

Now, a team of researchers has found a clever new way to spot these ammonia clouds. Instead of looking for heat, they decided to look at how the gas blocks sunlight. They used a satellite called Tanager-1, which acts like a super-powered camera that sees light in a part of the spectrum we can't see with our eyes. By analyzing how sunlight bounces off the ground and gets "stolen" by ammonia molecules, they were able to map the pollution plumes with incredible sharpness. They didn't just guess; they measured the gas coming out of specific factories in Pakistan and Uzbekistan, proving that this new "sunlight detective" method works and could help us see pollution that was previously hidden in the shadows.

The Sunlight Detective: Catching Invisible Clouds

For years, tracking the invisible clouds of ammonia pollution has been like trying to find a ghost in a foggy room using only a thermal camera. If the ghost (the ammonia) isn't a different temperature than the room (the air), the camera can't see it. Plus, the old cameras on satellites are so blurry they can only spot massive, isolated clouds, missing the smaller, sneaky leaks from factories and farms.

But in this new study, the researchers decided to change the game. Instead of hunting for heat, they decided to hunt for shadows. They realized that ammonia has a special "fingerprint" in the shortwave infrared part of sunlight. When sunlight hits the ground and bounces back up, ammonia molecules in the air act like tiny, invisible sunglasses, blocking specific colors of that light. By measuring exactly how much light is missing, the scientists can figure out exactly how much ammonia is there.

The New Tool: Tanager-1
To test this idea, the team used data from a satellite named Tanager-1. Imagine this satellite as a high-speed scanner that takes pictures of the Earth's surface in tiny 30-meter squares. That's about the size of a large house or a small park. This is a huge upgrade from the old satellite cameras, which might only see a whole city block at once.

The scientists used a mathematical trick called a "matched filter." Think of this like having a specific "wanted poster" for the ammonia fingerprint. The computer scans every single pixel of the satellite image, comparing the light it sees against the "wanted poster." If the light matches the pattern of ammonia blocking the sun, the computer flags it. They didn't just look at the whole picture; they zoomed in to find the exact shape of the pollution plumes drifting away from the sources.

The Big Discovery: Finding the Hidden Leaks
The team put their new method to the test in two places: Pakistan and Uzbekistan. They were looking at industrial fertilizer plants, which are known to release ammonia.

In Pakistan: They found a massive plume coming from a fertilizer factory. They calculated that this single plant was spewing out 837 kg h⁻¹ of ammonia.
In Uzbekistan: They spotted two nearby sources. One was releasing 781 kg h⁻¹, and the other 908 kg h⁻¹, for a total of 1689 kg h⁻¹.

Checking the Work
Now, you might wonder: "Could this be a mistake? Maybe the satellite just confused ammonia with water vapor or methane?" The researchers were very careful. They checked their work by looking at different windows of light (specifically 1425–1575, 1850–2100, and 2175–2300 nm). In every single case, the ammonia plumes still showed up clearly, proving they weren't just seeing other gases.

How Does This Compare to the Old Way?
The researchers compared their new numbers to what older, blurry thermal satellite cameras had reported in the past.

For the Pakistan site, older studies using thermal cameras estimated the leak was between 3212–8035 kg h⁻¹.
For the Uzbekistan site, an older study estimated 2714 kg h⁻¹.

The new numbers are lower. The authors suggest this might be because the old thermal cameras were sensitive to assumptions about how long the gas stays in the air, or they might have accidentally counted gas from nearby sources as one big blob. The new method, with its sharp 30-meter vision, allows for a much more precise look at exactly where the gas is coming from.

Why This Matters
This study is a "proof of concept," meaning it's the first time anyone has successfully shown that you can map ammonia using reflected sunlight instead of heat. It's not just a theory; they actually measured real emissions from real factories.

The best part? This isn't just a one-trick pony. There are already other satellites and airplanes flying around with cameras that can see these same light wavelengths (like AVIRIS, EMIT, PRISMA, and EnMAP). This discovery means we don't need to wait for brand-new, expensive satellites to start seeing ammonia pollution clearly. We can use the tools we already have to build a much sharper, more frequent map of the world's ammonia emissions, helping us protect our ecosystems and our health.

Technical Summary: Mapping Ammonia Emission Plumes Using Shortwave Infrared Imaging Spectroscopy

Problem Statement
Atmospheric ammonia emissions pose significant risks to ecosystems and human health by introducing reactive nitrogen and acting as a precursor to hazardous particulate matter. While agricultural sources currently dominate, industrial emissions are often underestimated in global inventories. Traditional monitoring relies on thermal infrared (TIR) spectrometers, which exploit ammonia absorption features between 10–11 µm. However, this approach faces three primary limitations:

Thermal Contrast Requirements: TIR techniques require a temperature difference between the surface and the atmosphere, which limits sensitivity to surface-level emissions and necessitates knowledge of plume altitude, complicating estimation.
Spatial Resolution: Existing satellite sensors in the TIR range offer coarse spatial resolution (kilometer-scale), restricting their utility to large, isolated sources.
Measurement Frequency: High-resolution airborne TIR surveys exist but are infrequent.

Methodology
The authors propose a novel approach using shortwave infrared (SWIR) imaging spectroscopy to detect ammonia absorption features, thereby circumventing the need for thermal contrast. The study utilizes calibrated radiance data from the Tanager-1 spaceborne pushbroom imaging spectrometer, which covers a spectral range of 400–2500 nm with 5 nm sampling.

The retrieval process employs a matched filter algorithm to estimate path length ammonia concentration enhancements ( $\hat{\alpha}$ ) for each ground pixel. Key methodological steps include:

Spectral Range: Analysis is limited to the 1400–2500 nm bands.
Target Signature ( $s$ ): The algorithm models the at-sensor radiance spectrum using a non-scattering radiative transfer model with U.S. standard atmosphere profiles. Absorption coefficients are derived from HITRAN2024. The target signature is calculated as the logarithmic difference between modeled radiance with an ammonia enhancement ( $\alpha = 2000$ ppm·m) and the background atmosphere.
Plume Delineation and Quantification: Detected path length enhancements are converted to vertical column mass enhancements ( $\Delta\Omega$ ). Plumes are isolated using thresholding, median filtering, and dilation. Emission rates ( $Q$ ) are calculated using the integrated mass enhancement method, incorporating ECMWF IFS 10 m wind data and the geometry of the plume's convex hull.
Validation: To ensure identified plumes are not artifacts of other gases (e.g., methane, CO₂, water) with overlapping absorption features, retrievals were conducted over specific sub-windows (1425–1575, 1850–2100, and 2175–2300 nm).

Key Results
As a proof of concept, the study successfully mapped and quantified ammonia emissions from industrial fertilizer production plants in Pakistan and Uzbekistan at a 30 m spatial resolution:

Pakistan: A single industrial source was quantified with an emission rate of 837 kg h⁻¹. This is notably lower than previous TIR-based estimates (3212–8035 kg h⁻¹) from 2013–2017, though the authors note that TIR estimates are sensitive to assumed ammonia lifetimes and nearby sources.
Uzbekistan: Two nearby sources were identified with rates of 781 kg h⁻¹ and 908 kg h⁻¹, totaling 1689 kg h⁻¹. This is also lower than a previous TIR estimate of 2714 kg h⁻¹ (2008–2016).
Robustness: The ammonia plumes remained distinct across different spectral sub-windows, confirming that the detections were not artifacts of interference from other gases.

Significance and Claims
The paper presents this work as a proof of concept demonstrating that SWIR imaging spectroscopy can effectively map ammonia emission plumes. The primary significance of this capability lies in:

Overcoming Technical Barriers: It eliminates the requirement for thermal contrast, simplifying the retrieval process and removing the need for precise plume altitude knowledge.
Expanding the Observing System: It enables a broad class of existing and future imaging spectrometers (e.g., AVIRIS, EMIT, PRISMA, EnMAP) to contribute to ammonia monitoring.
High-Resolution Monitoring: It facilitates high-resolution (30 m) observations from both airborne and spaceborne platforms, which are more common than high-resolution TIR surveys.

The authors modestly conclude that while this method shows promise for quantifying industrial point sources, continued sampling is required for rigorous quantification assessment. They also note that next-generation instruments targeting methane absorption features in the 2200–2400 nm range could offer heightened sensitivity to ammonia.

The Sunlight Detective: Catching Invisible Clouds

Technical Summary: Mapping Ammonia Emission Plumes Using Shortwave Infrared Imaging Spectroscopy

More like this