What is NDVI and how to use crop imaging in remote sensing?

The most popular remote sensing measure used in agriculture is the NDVI index. NDVI stands for Normalized Difference Vegetation Index, which simply describes the plant’s development based on how a plant reflects different light waves. A healthy well-developed plant absorbs red light and reflects near-infrared light and the opposite happens to a diseased or poorly developed plant. NDVI measures the difference between near-infrared and red light. NDVI values range from 0 to +1. The value 0 indicates no vegetation, so most likely the surface is bare soil, snow or water etc. Values near +1 show healthy vegetation and dense coverage with green leaves.

NDVI palettes

Different palettes are used for remote sensing with NDVI. The Green palette is the more common NDVI palette, where light green shows areas with low NDVI (poor vegetation) and dark green with high NDVI (healthy vegetation).

Colour scale of green palette
NDVI: Colour scale of the green palette. (https://agromonitoring.com/api/images)

In order to clearly bring out problematic areas, there is a contrast palette available. The contrast palette shows NDVI values from 0 to 0,07 as grey (very low index values). Low vegetation areas are coloured as red (0,07 to 0,3) on this palette.

NDVI Colour scale of contrast palette
Colour scale of the contrast palette (https://agromonitoring.com/api/images).

NDVI images help a farmer to get a good overview of the development of the crop without actually having to visit the paddock. It is a tool that can help to identify problem areas within the field. For example, when lower NDVI areas are at the same place every year there might be a problem with drainage, pH or even compaction. Low vegetation areas can also be caused by diseases, insect damage, lack of nitrogen or drought-prone soil types. In addition they can simply turn out to be obstacles on paddocks, such as big rocks or piles of stones. NDVI is helpful for taking care of large areas when on-site controlling is time-consuming. Changes in NDVI might be due to mismanaging application operations- e.g not applying fertilisers to all the field or making mistakes while drilling.

NDVI Compacted area on the field (where machinery usually enters)
Compacted area on the field (where machinery usually enters).

Other crop imaging indexes

NRI, Nitrogen Reflectance Index is used to indicate the nitrogen level in plants. Similarly to NDVI, the green colour shows high level of nitrogen and red means low level of N. Seeing red areas on the map might indicate nitrogen deficiency.

NDVI: Classical headland effect: due to fertilisation laws or wrong settings of fertiliser spreader, edge of the field received less nitrogen than other parts of the field.
Classical headland effect: due to fertilisation laws or wrong settings of fertiliser spreader, edge of the field received less nitrogen than other parts of the field.

NDWI, Normalized Difference Water Index, shows the water level in the area. Green means high water level and red means low water level which might indicate drought. If there is a dark green area on the field, it might indicate poor drainage. For farmers who use irrigation systems, early recognition of plant water stress can be critical to prevent lower crop production or crop failure.

NDVI Water stress during June 2020 started to be seen on the NDWI map from the headlands
Water stress during June 2020 started to be seen on the NDWI map from the headlands.

Analysing weather data

Remote sensing includes analysing weather parameter data. Air temperature has a great influence on both physical (growth) and chemical processes of plants (photosynthesis). In the eAgronom NDVI module, the weather data is seen in the maps section of NDVI page. On every weather parameter tab, there is also yield history. This enables farmers to compare their yield data with weather parameters, which is necessary for deeper analysis of yield forming and to form opinions about why the yield number was as high, average or low as it was.

The first weather parameter to look at is the weather forecast, which is specific to the location of the paddock. It helps farmers plan their tasks- e.g when spraying wind speed, air temperature and predicted rainfall must be taken into account.

Soil temperature is measured at the surface and to a depth of 10 cm. Comparing temperatures with past year is possible from the graph’s legend. Soil temperature helps a farmer to decide the drilling time. For example, it is very important for an organic farmer to put the seed into warm soil, so that the crop can germinate immediately and get a good start to compete with weeds.

Soil temperature graph on spring barley field
Soil temperature graph on spring barley field

Soil moisture graphs also provide valuable data for drilling. When soil moisture is low and the weather forecast doesn’t show rainfall in the near future, then it might be reasonable to adjust drilling depth a bit deeper so that the seed will not be left in dry conditions where it will not germinate. Soil moisture also helps to plan irrigation scheduling.

Soil moisture graph
Soil moisture graph.

Historical weather charts provide farmers with data about temperatures and rainfall over the past 3 months (the period can be adjusted). When was the last time it rained? What were the maximum, minimum and average temperatures on a specific date? When was the average temperature above 5 degrees, so the crops actively grew? Answers can be found by analysing the weather history charts.

Historical weather data: temperature graph from last 3 months
Historical weather data: temperature graph from last 3 months

 

Historical weather data: daily rainfall (mm) in last 3 months- it has been a dry autumn
Historical weather data: daily rainfall (mm) for the last 3 months, nearly one month without precipitation

 

Accumulated graphs

Each plant species needs a specific sum of day degrees to reach a certain stage of development. The accumulated temperature chart below shows how much heat plants have accumulated by a specific date. This chart helps farmers to predict when crops will reach certain growth stages and can also help to predict harvesting time.

Accumulated temperature charts provide farmers with comparative data about previous years, so they can see where they stand in the current season. If there have been lower day degrees then the crops might mature later and harvesting will be delayed. The example below shows that autumn 2020 has been warmer than normal and this is reflected in the field where development growth of winter crops and NDVI levels are higher than normal. This is a positive indication that the 2021 winter crop harvest as of today has a higher potential than the last two years.

Accumulated temperature graph: at the moment temperatures have been higher than previous years
Accumulated temperature graph: at the moment temperatures have been higher than in previous years.

Accumulated precipitation graph also helps farmers to compare the current season with previous years. It helps farmers to see how much moisture has been available for crops and plan tasks according to that. It is again useful data for irrigation planning.

Accumulated precipitation graph: at the moment there has been less rainfall than previous years.
Accumulated precipitation graph: at the moment there has been less rainfall than in previous years.

 

Crop based NDVI reports

The crop based NDVI report provides indications of potential yield when comparing NDVI with historical data about a similar crop in a similar year. If the NDVI figures are higher than on the comparative year, this might indicate that there will be higher yield. The end result will always be affected by weather conditions.

Crop based NDVI report.
Crop based NDVI report.

In conclusion, remote sensing tools like NDVI and weather reports are useful tools for helping farmers to locate problems on paddocks, look for reasons behind the problems, plan tasks and predict their yield potential.

 

eAgronom has recently added NDVI to the platform. When used together with yield maps, soil type data, terrain maps, and weather data, the combination creates a powerful tool for assisting to pinpoint the reasons behind yield differences within fields and years.

Sign up and try it yourself!