> ## Documentation Index
> Fetch the complete documentation index at: https://badixth-dc85e378.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Diagnose and Respond to Nitrogen Deficiency in Rice

> Identify nitrogen deficiency in paddy from uniform pale-green canopy and low NDRE, and correct it with split N applications guided by leaf color chart or SPAD readings.

Nitrogen is the nutrient rice needs in the largest amount and the one most commonly limiting yield. N deficiency reduces tillering, leaf area, and grain filling. It is almost universal on unfertilized rice and is where satellite indices add the most unique value.

## When rice is most vulnerable

From active tillering through panicle initiation. N demand peaks between mid-tillering and booting; a shortage in this window caps the maximum yield the crop can reach.

## How nitrogen deficiency shows in satellite data

* **NDRE is the leading indicator.** NDRE is sensitive to chlorophyll and leaf N content and responds before visible symptoms are obvious.
* **NDVI is also low**, but NDVI saturates in dense canopies while NDRE keeps discriminating - use NDRE from tillering onward.
* Pattern is **uniform across the field or across whole management zones**, not patchy - if it is patchy, suspect salinity, poor soil, or biotic stress instead.
* The whole crop looks pale green from the air; the darker green reference strip (if you leave one) stands out clearly.

## Field symptoms

* Uniform pale-green to yellow canopy, oldest leaves first
* Reduced tillering and narrow leaves
* Stunted growth and delayed heading
* Older leaves die back from the tip while younger leaves stay greener (N is mobile in the plant)
* Small panicles and reduced grain filling

## Reduce the damage by

* **Split N application** - avoid applying all N at basal; split into basal, active tillering, and panicle initiation.
* **Use a leaf color chart (LCC) or SPAD meter** to time top-dressing to actual crop need. **This documentation publishes no threshold to read either against** — see below.

## Leaf colour readings are recorded, not interpreted

The [field walk protocol](/guides/Crop/Rice/field-walk-protocol) asks a scout to take an LCC or SPAD reading at every stop and record the number. **The platform stores that reading and does not turn it into a verdict**, because the threshold it would be read against is not settled.

### Why there is no threshold here yet

Two Malaysian sources give different trigger panels, and neither states the one thing that would make a panel number portable.

| Source                                                | Trigger                                               | Urea rate                                         |
| ----------------------------------------------------- | ----------------------------------------------------- | ------------------------------------------------- |
| DOA, *Pakej Teknologi Padi* (2008), Jadual 11/12      | Not stated as a panel number                          | **50 kg/ha** off season, **75 kg/ha** main season |
| IADA Pulau Pinang, *Manual Tanaman Padi* (KPKM, 2019) | Panel **No. 3** direct-seeded, **No. 4** transplanted | **76 kg/ha**                                      |

<Warning>
  **Neither source states the chart's panel count.** LCC charts are published in 4-panel and 6-panel versions, and *"No. 3"* is a different green on each. A trigger panel without its chart is not a threshold, and applying one chart's number to the other chart's card would be a fabricated reading.

  Note also that both rates are kilograms of **urea product**, not kilograms of N.
</Warning>

Until the panel count and the source conflict are resolved, a leaf colour reading is **evidence for the agronomist reading the record**, not an input the platform acts on. Recording it is still worth doing: it builds the block history that a later calibration would need.

### An LCC reading may not be averaged

An LCC panel number is **ordinal**. Panel 4 is not twice panel 2, and the steps between panels are not equal, so a mean of panel numbers is arithmetic on a rank and means nothing.

Where several stops are summarised, the block carries a **distribution** — how many stops read at each panel — in the same way a scope spanning several bands carries a distribution rather than a band. See [How a band resolves](/snippets/band-resolution).

**SPAD is different and may be averaged.** The meter returns a continuous chlorophyll index, not a rank. The two are named together throughout this documentation because they answer the same question, but they are not the same kind of number and do not aggregate the same way.

* **Correct water management** - N loss through volatilization and denitrification is very high in flooded rice; incorporate urea into soil and maintain shallow standing water.
* **Use deep placement of urea briquettes** where available - major reduction in N loss.
* **Grow a legume in rotation or as green manure** to build background soil N.

## Set an alert for early warning

<Tip>
  Turn the diagnosis into a trigger. Go to the [Alerts](/guides/alerts-notifications) page and configure a rule so N-limited zones surface before panicle initiation, when top-dressing still recovers yield.
</Tip>

Recommended settings for nitrogen on rice:

* **NDRE threshold alert** - trigger when field-average NDRE falls below your fertile-strip baseline during tillering.
* **NDRE zone alert** - flag zones that stay one standard deviation below the field mean for two consecutive passes; these are candidates for variable-rate top-dressing.

## Related

* [Rice Diagnosis](/guides/Crop/Rice/rice-abiotic-stress)
* [Create variable-rate prescription maps](/guides/prescription-maps) - turn NDRE zones into an N prescription.
* [Set up automated crop alerts](/guides/alerts-notifications)