Thermal Data
Beyond NDVI

Trust Backed by Data. Confidence Backed by Results. Experience the Difference.

Thermal | VS | NDVI | VS | Thermal | VS | NDVI

NDVI is commonly used and Hydrosat applies it, but not on its own… 
NDVI = Normalised Difference Vegetation Index, a measure of vegetation cover, which is used as a proxy for general crop health. However, relying on this measurement alone can lead to many misjudgements:

NDVI does not account for the crop growth cycle. 

If a plant is just growing, NDVI will be at 10 – 20%, values which would normally be cause for concern, however on the ground it is understood that the plant is fine and just doesn’t cover a large area due to its infancy.


However, NDVI can still be used to detect problems, as if you have experience seeing this pattern of NDVI values growing over the crop cycle, any anomalous values, such as decreases or plateaus when you know it should be reaching it’s peak, would be cause for concern. However, the danger with this is that when NDVI drops, the damage has already been done. It cannot detect oncoming stress, it tells you that the plant is already stressed, not that it is going to become stressed; it’s a reactive measure not a preventative one.
  When water stress begins, the first thing the plant does is close the stomata to prevent water vapour from escaping and         conserve water. But the leaves are still green; NDVI is still high. Only when the stress lasts for several days, the leaves                 begin to change colour as chlorophyll drops, and then the NDVI goes down. But by that time, they have already lost                   production, there is already damage. That's why temperature is much more sensitive to detecting stress as the                           temperature of the leaf rises immediately when you close the stomata. So quick that it can be detected in the same day         as the temperature increase occurs within a matter of hours.

You could have two fields with the same NDVI – both very green – but very different temperatures. The one that is hotter is more stressed, even if they both look equally green. That's the advantage of working with thermal data – it anticipates the problem. It sees stress before there is visible damage, allowing us to act preventively.

NDVI does not account for different crop types.

Broadleaf crops, like sugar cane or corn, can easily reach over 90% but things like onions or garlic rarely reach 60%. This doesn’t mean the crop isn’t healthy it just is a result of the crop cover from its size. 

NDVI isn’t measured daily.

NDVI comes from visible images, mainly Sentinel satellites, which only update every 5 days, not daily like thermal imagery. Plus, if there are clouds, this 5-day cycle could be extended. And if there are clouds, it may take longer.

The classic solution for ET is based on data from weather stations and crop coefficients...

ETa = Real/ Actual Evapotranspiration

This is the one actually produced in the existing conditions of the crop.

Kc = Crop Coefficient 

FAO-56 Kc are a standardised, dimensionless factor which converts reference evapotranspiration (ET₀) into crop-specific water use (ETc) at each growth stage.


The FAO-56 tables provide three key values per crop:

  -  Kc initial: from planting to ~10% ground cover; low, dominated by soil evaporation

  - Kc mid: full canopy, peak water demand; typically ranges from 1.0 to 1.25

  - Kc end: late season through senescence; drops as leaves die off

These are general defaults for a sub-humid climate with moderate wind. 

ETp = Potential Evapotranspiration

The maximum ET a given surface could achieve considering the energy and atmospheric conditions, assuming the soil is always fully wet/ water was not a limiting factor.

ETo = Reference Evapotranspiration

(Penman- Monteith recommended by the FAO)
Standard evapotranspiration parameter based on an area that assumes grass as a reference crop, well irrigated, growing and continuously covering the soil surface, with uniform height of 0.12m and a surface strength of 70 sm-1 and albedo of 0.23.

ETc = Crop Evapotranspiration

This is a theoretical ET value of the crop under standard conditions, from experimentation (healthy, well-fertilized crop and in optimal soil moisture conditions).

These calculations aren’t very accurate as you generalize too much.

   Actual evapotranspiration (ETa) is dynamic in from one area to another and from one hour to the next (ET varies spatially and temporally)
         So if Kc is time and place specific, it is therefore not universally applicable
   NDVI measurements aren’t good enough as NDVI does not capture the actual dynamics of crops
         If you use NDVI to get a Kc value, it will be too vague as there is a linear relationship between NDVI and Kc, which is the basal Kc value
         If you use SEBAL, it uses the energy balance and brings in far more factors in order to calculate ET which accommodate for the vast number of dynamic factors influencing the crop
NDVI doesn't capture the actual dynamics of the crop, temperature does.

Remote sensing is the future as you’re not just talking about water requirement, it’s about the context of the crops surroundings in semi-real time (the microclimate around the crop, soil moisture, how much the crop is actually using) and therefore you can account for the dynamic conditions the crop is actually under and get the crop what it actually needs on that day and not what the crop generically needs, without accounting for the dynamics that the crop is under… what if there is a heatwave, or increased high windspeed – the crop will detect this and still try and protect itself/ react/ be more stressed even if there is enough moisture in the soil and this needs to be accounted for.

  • Irrigation cannot be treated just from a soil perspective when there are so many other factors influencing the crop
  • Some of these factors you cannot see just by looking at the crop or looking at NDVI, you need the thermal data

Want to learn more? Click below to read about another linked topic

Help us
Improve

Your feedback helps us make our Academy more useful. Tell us about your experience and what you'd like to see next.

Share With
Your Network

Know someone who could benefit?
Share the IrriWatch Academy with your colleagues and network.