Ag Geek Speak

11.5 Tiny Bytes: NDVI: Turning Plant Color into Actionable Farm Data

• A Podcast for Precision Agriculture Geeks • Season 2 • Episode 11

Share episode

NDVI (Normalized Difference Vegetative Index) helps farmers detect how green and/or productive different areas of their fields are by measuring the ratio of near-infrared to red light reflected from plant canopies. This technology creates reliable productivity zone maps when analyzed consistently over multiple growing seasons, enabling targeted input management across variable field conditions.

In this Tiny Byte Sarah and Jodi cover:
• NDVI measures plant canopy density by comparing near-infrared light reflection to red light absorption
• Chlorophyll absorbs visible light while plant cell structures reflect near-infrared light
• NDVI values range from -1 to 1 (or ~0-100 in ADMS) with higher values indicating denser vegetation
• Creating reliable management zones requires multiple years of imagery, not just single snapshots
• Best imagery comes from periods when crop canopy is developing or starting to senesce
• NDVI data often correlates with yield potential and can identify consistently productive field areas
• Zone management allows targeted input application based on productivity potential
• The most valuable NDVI images show variation in greenness rather than complete saturation

Check out our backlog episode "Choosing Quality Imagery" in Season 1 for more information on selecting the best imagery for creating management zones. https://www.buzzsprout.com/2297340/episodes/16135793

https://gktechinc.com/


Jodi:

And now it's time for a tiny bite of knowledge.

Sarah:

It ain't easy being green, being the colors of the leaves.

Jodi:

It might not be easy being green, but what is relatively easy is detecting how green something is, and that comes in the form of NDVI, or normalized difference vegetative index. So that's what NDVI stands for. But let's break this down. What is a vegetative index? Even what is?

Sarah:

it. A vegetative index is when we use light, specifically using different bands of light within the electromagnetic spectrum, to understand vegetation. So when we think about NDVI normalized difference vegetative index it's an index where we have a ratio and a specific formula that compares near infrared light to red light and that helps us understand how densely green the plant canopy is. Basically, in dumbed down terms, when our human eyes see green, it is because red light is getting absorbed, and so plants are absorbing red light. Green is being reflected, and so when we take into comparison near infrared light and red light, we are able to understand how green and how densely green the plant canopy is across the landscape.

Jodi:

So what does that mean for us? How do we use that number in a practical way?

Sarah:

So chlorophyll strongly absorbs visible light and plant cell structures on the leaves strongly reflect near-infrared light. So NDVI, specifically, is a way of calculating the amount of light that is being absorbed and used for photosynthesis and also how much light is reflected. This is important because it's going to help us understand how green and how densely green a plant canopy is across a landscape. For our purposes in agriculture, we're really thinking about where is a field of crop more or less green than other areas, because this is really an indication of photosynthesis. This can be a great way of thinking about the productivity potential across the landscape. And actually there's many places and many times where NDVI can actually correlate to yield data Not always, but many times it does correlate to yield data.

Jodi:

So does that mean a lot of what we use in terms of imagery? We'll extract the NDVI value from that imagery and then use that to build zones.

Sarah:

Absolutely Jodi. Many times here at GK Technology we actually use a lot of satellite imagery to help understand the productivity potential of a field across a landscape. It can help us understand where the crop has more potential to photosynthesize and have greater yield potential, productivity potential and areas of the field where there's less photosynthesis happening and therefore less productivity potential. So from that, when we can take a look at different satellite images over the years, we can understand across a given field where those areas are consistently happening, where we consistently seeing greater amounts of NDVI throughout the growing season and lesser NDVI and photosynthesizing photosynthesis potential across that field. Once we can figure out where that is happening consistently over a number of years, we can create zone maps that can help us understand where we should be investing in more inputs into areas where the greater production potential is happening and where maybe we should try to manage that input cost so that we have less inputs going into lower productivity areas.

Jodi:

So basically what I hear you saying is we want to find out like where the NDVI value is lowest, consistently the lowest. We want to those would be what we consider like a low productivity zone and the areas where the NDVI value is the highest. Those are hypothetically and probabilistically our areas of highest productivity and we want to manage those, like those are our highest producing zones or areas with highest potential.

Sarah:

Absolutely, and so we are obviously transitioning this conversation from NDVI to how we're using these in a practical sense for making zones so, jodi, would you ever use just one NDVI? To how we're using these in a practical sense for making zones. So, jodi, would you ever use just one NDVI image to create zones from one year?

Jodi:

Absolutely not. It is a snapshot in time and that doesn't necessarily reflect any sort of variance in weather patterns. Maybe, like a single snapshot is a culmination of conditions that occurred that year, but that doesn't necessarily mean that that's consistent over time. And so, no, I do not use a single image to create zones, and I do not promote that either, just because what we're trying to do when we're building zones is we're trying to predict variability and predict consistent variability, and so with just one snapshot, you aren't able to really create something that holds up over time.

Sarah:

I couldn't agree with you more, jodi. You just totally hit the nail on the head with that, and I do the same thing when I'm looking at imagery over time. I want to find those areas where we are having consistently lower productivity potential and areas of the field where we're having consistently higher productivity potential, and that just is not possible with only one image in one year.

Jodi:

Yeah, absolutely.

Jodi:

And there's a couple things to just mention about NDVI before we wrap this up.

Jodi:

The calculation for NDVI just the bare bones or textbook version of this it's going to produce a value that's either from negative one to plus one, whereas if you get closer to one that means that is the highest densely green you could have, whereas the whole negative one would be the most opposite of green you can have. But what I want to make this point of is that in ADMS, if you're an ADMS user and you're extracting NDVI images, what we do in the software is we do multiply that NDVI value by 100, meaning that if you're used to seeing and looking at NDVI values, this negative one to one concept is going to sound kind of weird, but that's because we multiply it by 100 to make it easier to remember. In our grand scheme of things, you probably are familiar with looking at values from like negative point something or like very lowly negative to up to like 83 to 90. And so those are kind of like the normal, what we would say range of NDVI values you might see in ADMS.

Sarah:

And I think in ADMS, when we're looking at our histogram, once you reach right around like 35 or 40 in there, we're starting to think about having, you know, some greenness on the landscape. You're kind of out of the area where green plant matter is starting to dominate that land versus soil or something that's non-vegetative, and it doesn't take very long to get up to that complete saturation. You know those numbers that are 80, 80, that's. It's really hard to find vegetative indices out there that are different from each other. You know, when I'm looking at imagery and I'm trying to make zones out of that imagery, something that has no vegetative indice to it at all, it's like less than that 30 value and you can't tell differences on the landscape from a vegetative perspective.

Sarah:

That's not very helpful to me when I'm considering vegetative uh indicee data only. But also once I get up to that place where it's completely saturated and I can't see any any differences in the in the green canopy, that's also not the most helpful image to me because I can't tell where those differences are and a lot of times we'll see the differences in the plant canopy earlier in the growing season and then there'll be that time when the plant canopy is completely saturated and you just won't be able to tell a whole lot of difference. And then, as you start to move more towards senescence, you'll start to see some variability again and perhaps how that plant canopy dries down again. So the images that I find the most difficult to use when it comes to NDVI are where there's absolutely no plant material out there at all, or the images that are absolutely 100% saturated with greenness and I can't tell any difference.

Jodi:

Right, we're trying to use these images to figure out where those consistent differences are, and Sarah and I could talk about this subject for a long time.

Sarah:

And in fact we have.

Jodi:

And if you want to hear more, too, about you know how we can choose imagery that helps us get better NDVI data to build those zones, do check our backlog. We do have an episode called Choosing Quality Imagery in Season 1, so please check that out. But what I think I've heard today, sarah, is that maybe it isn't always easy being green, but hopefully with zone management it's a lot easier to manage those areas that are green and hopefully improve the way that we're managing those less green areas to get them closer to that darker green, the higher NDVI value.

Sarah:

It ain't easy being green, being the colors of the leaves. Tune in next time for a tiny bite of knowledge from GK Technology, where we have a map and an app for that. I love Kermit the Frog. He makes me so happy.