Ammonia Volatilisation from Urea in Paddy Soils — How Nitrogen Escapes Before Crops Can Use It

Urea is the world's most widely used nitrogen fertilizer. It is cheap, concentrated at 46% nitrogen by weight, and easy to handle and transport. It is also surprisingly easy to lose. Apply urea to a flooded paddy field on a warm morning, and a substantial fraction of the nitrogen you applied may have volatilised as ammonia gas before the rice roots ever encounter it.

The previous posts on this site have covered nitrogen cycling broadly and the case for slow-release alternatives. This post goes deeper on the specific mechanism that drives the largest single-pathway loss in tropical paddy agriculture: ammonia volatilisation. Understanding it at a chemical level is what makes the engineering of coated fertilizers and urease inhibitors more than guesswork.

The Chemistry: What Happens to Urea After It Enters the Floodwater

Urea dissolves readily in water. Once in solution, it is hydrolysed by the enzyme urease, produced by soil bacteria. The reaction proceeds in two stages. First, urea is converted to ammonium carbonate:

The intermediate ammonium carbonate is unstable and rapidly breaks down:

Urease drives this hydrolysis remarkably fast under warm conditions. In tropical soils at 30°C, hydrolysis can be 80 to 90% complete within 48 to 72 hours of application. The nitrogen does not sit in the urea molecule waiting to be taken up — it is transformed within days into a pool that is partly plant-available and partly vulnerable to loss.

The critical step for volatilisation is the equilibrium between ammonium and free ammonia in the floodwater:

At pH values below 7, almost all the ammonium-nitrogen exists as NH₄⁺, which is non-volatile and can be taken up directly by roots or held by the cation exchange complex of the soil. As pH rises above 8, the equilibrium shifts and a growing proportion exists as free NH₃, which can escape from the floodwater surface into the atmosphere. At pH 9, roughly half is in the NH₃ form. Above pH 9, the majority is free ammonia.

Why Flooding Makes Everything Worse

The flooding that defines paddy cultivation creates conditions that maximise volatilisation losses. Three mechanisms operate simultaneously.

The pH spike. When urea hydrolyses, the process consumes hydrogen ions and produces hydroxide equivalents, raising the pH of the floodwater around the urea granule. In a closed container, this spike can push pH from 7.0 to above 9.5 within the first 24 to 48 hours. In a real paddy field, algal photosynthesis during daylight hours further raises pH — sometimes to 9 or above by mid-afternoon — because photosynthesis removes dissolved CO₂ from the water, pushing the carbonate equilibrium toward higher pH. The combination of urea hydrolysis and afternoon photosynthesis creates a daily window during which volatilisation rates peak.

Warm floodwater. Temperature directly increases both the rate of urease activity and the volatility of NH₃. The Henry's law constant for ammonia rises steeply with temperature — the same ammonium concentration in 35°C floodwater produces roughly twice the atmospheric ammonia pressure as at 20°C. Bangladesh's paddy seasons bracket temperatures that drive high volatilisation: the boro (dry) season occupies cooler months but floodwater can still reach 25–30°C, while aman (monsoon) conditions push temperatures higher.

Shallow water with a large surface area. Volatilisation occurs at the floodwater-atmosphere interface. A shallow water layer presents a proportionally large surface relative to volume, and the NH₃ that escapes from this surface is replaced by upward diffusion from the bulk of the water. Deep water would dilute the ammonium concentration — but paddy fields are typically maintained at shallow depths of 5 to 10 cm, which maximises the problem.

What the Literature Says About Loss Rates

Estimating how much nitrogen actually leaves the field as ammonia is not straightforward, and published values vary considerably depending on measurement method, soil type, temperature, water management, and urea application rate. The range in tropical and subtropical paddy literature runs from below 10% to above 50% of applied nitrogen lost via volatilisation in a single season.

For South Asia and Bangladesh specifically, the picture is reasonably consistent. Freney and colleagues, working across Asian paddy systems in the early 1990s, found losses ranging from 10 to 40% of applied urea-N depending on conditions, with the higher end associated with alkaline soils, high temperatures, and direct surface application to floodwater. Choudhury and Kennedy (2005), reviewing fertilizer nitrogen losses in lowland rice, cited 20 to 30% as a typical volatilisation loss for surface-broadcast urea in tropical paddy under standard management.

Islam and colleagues, working in Bangladeshi conditions, have reported losses broadly consistent with this range. NUE values below 40% are commonly cited for urea in flooded boro rice — meaning that at minimum 60% of applied nitrogen is either lost (via volatilisation, leaching, and denitrification) or immobilised in organic matter. Volatilisation is generally the dominant loss pathway when urea is applied directly to floodwater surface, as is common practice.

How Researchers Measure It

Measuring ammonia volatilisation from paddy fields is methodologically tricky. The gas escapes at the water surface in a spatially variable way, rates change across the day (highest midday, lower at night), and any measurement device placed on the field surface inevitably alters the microenvironment it is trying to observe. Three approaches dominate the literature:

Closed-Chamber / Alkali Trap Method

A floating chamber placed on the floodwater surface captures the gas escaping from a defined area. Inside the chamber, a vessel containing dilute acid (typically sulphuric acid, H₂SO₄) absorbs the NH₃ as it accumulates. At timed intervals, the acid solution is collected and analysed for ammonium by Kjeldahl distillation or colorimetry. The loss rate is calculated from the mass of NH₄-N trapped per unit area per unit time.

This is the most common method in Bangladesh and South Asia research because the equipment is low-cost and field-portable. Its limitation is that the chamber disrupts the wind environment at the surface — real field conditions involve turbulent air movement that removes NH₃ faster than a static chamber captures it. As a result, closed-chamber methods may underestimate actual losses, though careful design minimises this bias.

Vented (Semi-Open) Chamber Method

A modification that draws air through the chamber at a controlled flow rate, passing it through an acid trap. This better replicates natural ventilation. The IPNI (International Plant Nutrition Institute) method using a semi-open system has been widely adopted in Asian paddy research because it gives more reproducible results than purely closed chambers while remaining practical at field scale.

Micrometeorological Methods

At larger spatial scales, micrometeorological approaches measure ammonia concentration at multiple heights above the canopy and use boundary layer theory to back-calculate flux. Techniques include the integrated horizontal flux method and eddy covariance systems. These are non-intrusive and capture field-scale behaviour, but require expensive instrumentation and technical expertise that limits their use in developing-country research contexts. They are most commonly seen in European and North American studies but have been applied in Asian rice systems in collaboration with international institutions.

The SPADE Connection: Monitoring What You Cannot See

One of the fundamental problems with nitrogen management is that loss is invisible. A field that loses 30% of applied nitrogen to volatilisation looks identical to one that loses 5%. There is no colour change, no physical evidence, no alarm. The crop may show early greenness from the initial dose, then gradually decline — but by the time yield loss is obvious, the loss has already happened.

This is partly what drives interest in tools like SPAD meters and spectral analysis platforms for crop nitrogen status. The SPADE open-source tool I have been involved with works by measuring leaf chlorophyll content as a proxy for nitrogen status, allowing researchers and farmers to detect nitrogen deficiency before it translates to yield loss. If floodwater volatilisation has already stripped a portion of the applied dose, SPADE-based monitoring can catch the deficit early enough for a corrective split-dose application.

The monitoring approach does not address the underlying loss mechanism — but it changes the decision point from "how much did I lose?" to "what does the crop need now?" In practical terms that shift can be meaningful for yield and for total nitrogen input across a season.

What Can Be Done

Three main strategies reduce ammonia volatilisation losses from paddy urea, each attacking the problem at a different point in the loss chain.

Deep placement. Moving the urea from the floodwater surface to the reduced zone below the soil surface (7 to 10 cm depth) is the single most effective intervention available under current technology. Nitrogen placed below the soil-water interface is protected from the high-pH floodwater environment. Deep placement can reduce volatilisation losses by 50 to 80% compared to surface broadcast. The barrier is labour: placement requires either manual insertion of urea supergranules or mechanical applicators, both of which add cost in systems where surface broadcasting is the norm.

Urease inhibitors. Compounds like N-(n-butyl) thiophosphoric triamide (NBPT) can be applied as coatings or surface treatments on urea granules. They temporarily block urease activity, delaying hydrolysis by 7 to 14 days. This delays the pH spike and spreads the ammonium release over a longer period when the crop is better positioned to capture it. NBPT is approved in several markets and has shown 20 to 50% reductions in volatilisation in controlled studies. Cost and availability remain constraints in South Asian markets.

Coated slow-release fertilizers. The approach at the centre of my BBCU research is physically encapsulating the urea in a matrix material that controls the rate of water ingress and urea dissolution, thereby slowing hydrolysis. A bentonite and biochar coating around each granule delays the release of urea into the floodwater, reducing the initial-burst loss that accounts for most of the seasonal volatilisation. The goal is not to eliminate hydrolysis but to spread it over a period — 10 to 14 days rather than 2 to 4 — that better matches the crop's nitrogen uptake window. A detailed post on how this works is available on the slow-release fertilizer page.

Why This Matters Beyond Bangladesh

Ammonia volatilisation from paddy systems is not only a productivity problem. It is an atmospheric chemistry problem. Ammonia deposited downwind of agricultural areas contributes to secondary particulate matter formation, eutrophication of sensitive ecosystems, and acidification of soils distant from the source. South and Southeast Asia are among the largest regional sources of agricultural ammonia globally, and paddy rice is a major contributor within that region.

Improving NUE in paddy systems therefore has a double return: more nitrogen stays in the crop, and less nitrogen enters the atmosphere. Researchers working on coatings, inhibitors, and management practices are working on both simultaneously, which is part of what makes this a productive area of investigation.