Gro Offers New Satellite Data on Crop Health

Talk to our our team about Gro's offering
Talk to our team

A new daily measure of the NDVI satellite-data series on the Gro Intelligence platform provides users with near-real-time insights of plant health and crop condition globally.

The new daily NDVI is a cousin of the 8-day and 16-day NDVI data series previously available in Gro. NDVI, short for normalized difference vegetation index, can be an early indicator of plant stress such as impending drought or can foreshadow a bumper crop. As a result, it's a major driver of all of Gro’s machine-learning-based crop yield models.

NDVI evaluates plant greenness, chlorophyll content, and photosynthetic activity. Satellite sensors measure the reflection of near infrared radiation (NIR) and visible light. In healthy plants, most visible light is absorbed for photosynthesis and NIR is reflected. But when a plant is stressed, inefficient photosynthesis means less NIR is reflected. Satellite sensors detect the intensity of NIR and visible red light reflected and these values are then used to produce NDVI.

Gro aggregates NDVI, an index of vegetative health, from the pixel to the district level to allow better comparisons with other data series. These charts show current growing conditions in Brazil’s largest soybean producing state of Mato Grosso, where soybeans are maturing and near harvest. The left chart shows the daily reading of NDVI for Jan. 17 in Mato Grosso. The chart on the right shows the same districts, with a focus on the quantity of soybeans produced in each district, based on data in Gro provided by government source IBGE. 

A limitation to NDVI: Heavy cloud cover inhibits a satellite’s ability to take measurements, which can result in an erratic data series. The 8- and 16-day NDVI series are designed to smooth the data. The highest NDVI reading from each period is selected, filtering out the volatility created by clouds and presenting a reliably consistent data series. The process does result in lower data frequency. However, the approach normally works well since vegetative conditions don’t often change significantly within those time periods. Still, since 8-day NDVI readings are updated every eight days, the resulting lag means the gap between NDVI readings can be as large as 15 days. The methodology also means that only the highest NDVI readings per period are available.

To overcome these reporting limitations, Gro has added daily NDVI to our platform. Daily NDVI allows more frequent updates and a shorter lag time between readings and reporting. It’s particularly helpful in assessing the impact of natural disasters, like floods and hail storms, in real time. And in parts of the world where data for crop planting and harvesting isn’t readily available, NDVI can signal crop growth and harvest times, generally within a day. One drawback to daily NDVI is that, to use it in modeling, it requires pre-processing to remove inconsistencies caused, for example, by cloud cover.

The daily and 8-day series of NDVI, an important indicator of crop condition, offer different advantages. As seen in this chart of NDVI in Brazil’s Mato Grosso state, 8-day NDVI data (green line) generates a more consistent trend, while the daily series (blue line) is more variable. But the daily series can signal changes in crop condition sooner. In late November, for example, the daily series showed an uptick in NDVI six days before it appeared in the 8-day series. 

Daily NDVI on Gro’s platform comes from the European Space Agency’s Sentinel-3 satellites, while the 8- and 16-day NDVI series are from NASA’s Terra and Aqua satellites. The additional European source provides source diversity to the Gro platform, and also means the satellite collection times vary for the different NDVI data series. Daily NDVI data is available in Gro back to April 2018. By contrast, Gro has historical readings for 8-day and 16-day NDVI from 2000. That long track record also means there’s a great deal of peer-reviewed research (including this and this) to provide guidance on the series' effectiveness in modeling vegetation across the world.

Related Insights from Gro Intelligence

This insight was powered by the Gro platform, which enables better and faster decisions about factors affecting the entire global agricultural ecosystem. Gro organizes over 40,000 datasets from sources around the world into a unified ontology, which allows users to derive valuable insights such as this one. You can explore the data available on Gro with a free account, or please get in touch if you would like to learn more about a specific crop, region, or business issue.

Get a demo of Gro
Talk to our enterprise sales team or walk through our platform