Multispectral satellite imagery is becoming more and more ubiquitous throughout agriculture. And for good reason: it can provide data that helps farmers adjust nutrition plans, monitor crop health, improve yields, and maintain productivity on the farm.
But satellite imagery has a lot of moving pieces — and they all need to be optimized to actually provide farmers those benefits.
Unless your imagery program is:
Using the correct vegetation indices
At the correct time in the season
In a way that aligns with your specific crops, tillage practices, and yield goals
… then you’re not really maximizing the potential of satellite imagery. In fact, when these pieces are misaligned, it might not be very helpful at all.
The basics: Understanding multispectral satellite imagery
Multispectral satellite imagery isn’t just a mouthful to say: it can be a complex topic to wrap your head around. Before diving into choosing the right imagery program for your farm, it’s helpful to understand more about how the whole system works.
Whether you’re brand new to integrating imagery into your farming operation or you’re just looking to optimize your existing program, here’s a refresher on some of the most important pieces:
What is multispectral satellite imagery?
There’s a certain range of colors that are visible to the human eye. This range (RGB) is how your eyes process images in everyday life.
But there’s actually more light and energy around us than what we can see with the naked eye. That’s where multispectral imagery comes in — this technology can make the full electromagnetic spectrum visible by capturing images within specific wavelength ranges.
Those wavelengths of light can be combined into vegetation indices to measure different aspects of a crop field.
What is a vegetation index?
Depending on various stress factors (drought, heat, nitrogen deficiency, etc), crops can express differences in photosynthesis, leaf structure, and other health attributes throughout their life cycles. These differences can be measured using satellite imagery and vegetation indices.
Different vegetation indices can measure things like:
Crop biomass
Nitrogen content
Chlorophyll content
Drought stress
Yield potential
Other nutrient content (for nutrients like sulfur)
There are hundreds of vegetation indices out there. However, only a handful of them have been proven through scientific research to correlate with factors that are important to farmers. Some of the most popular include:
NDRE (Normalized Difference Red-Edge Index: highly-related to plant photosynthesis
NDVI (Normalized Difference Vegetation Index): measures vegetation density
SAVI (Soil Adjusted Vegetation Index): Separates soil reflectance from crop reflectance
OSAVI (Optimized Soil Adjusted Vegetation Index): Separates soil reflectance from crop reflectance
CCCI (Canopy Chlorophyll Content Index): measures a plant’s nitrogen
The usefulness of a particular vegetation index will also depend on what phase of a crop’s development that that index is being used. For example, NDVI is more suitable for corn early in the growing season. Later in the season, this index starts to saturate and doesn’t provide clear information.
What about image resolution?
There are a few types of resolution that can impact the quality and useability of satellite imagery.
Spatial resolution: Pixel size. Just like when you take a photo with a camera, the clarity/granularity (and therefore, precision) of satellite images is impacted by how pixelated they are.
Temporal resolution: How often images are captured. This is important for determining the pace at which crop conditions (photosynthesis, drought stress, etc) can change.
Spectral resolution: Range and precision of radiation captured. This helps determine the type of vegetation indices that can be computed from an image.
Why is using the right vegetation indices important?
Using the wrong vegetation indices for your specific crop or situation might not be harmful — but it also won’t be useful for you.
To improve your yields and maintain profitability, you need to monitor and measure crop health. And the appropriate vegetation indices will help you do exactly that. With accurate, relevant crop health information at your fingertips, you can make better, data-backed decisions to:
Secure better yields
Make in-season adjustments to nutrition programs
Improve efficiency and profitability
Using the wrong vegetation indices within your satellite imagery program can hinder all of those things.
So what exactly impacts the usefulness of a specific vegetation index?
Soil tillage type: No-tillage fields will lead to much different background reflectance in imagery than fields with regular tillage.
Crop type: Different crops will take different amounts of time to reach a uniform canopy, which impacts how helpful certain indices are.
Row spacing: Canopy is also impacted by how far apart your crop rows are. For example, corn usually reaches canopy at the V7 stage when it’s planted with 30-inch row spacing. But if you plant rows with 20-inch spacing, that canopy might happen closer to V6. With 36-inch rows, it might creep closer to V8.
All of these factors are important to consider when choosing which vegetation indices to prioritize in a satellite imagery program.
How do you choose the right vegetation indices?
The good news is that as a farmer, you don’t have to become an overnight expert on multispectral imagery and vegetation indices. (You have enough on your plate as it is!)
There are already experts in this field. You just have to ask the right questions to ensure the imagery platform you choose aligns with your operation, crop, and goals.
Here are a few starter questions to ask potential satellite imagery vendors before signing on to work with them:
Which vegetation index are you using to make recommendations?
Your vendor should be well informed on the pros and cons of many different indices. They should also be able to articulate why they’ve chosen to use the ones that they use.
What’s the range of vegetation indices that are available to be viewed in your platform?
If the imagery vendor only uses one index, for example, then it might not be as robust of a platform as you need. Make sure they pull from multiple sources so you can access as much information as possible to make better operational decisions.
What proof / scientific research do you have that the vegetation indices you’re using are appropriate for the recommendations you’re making?
The vendor should be able to provide science-backed reasoning for why they utilize the vegetation indices they’ve chosen — as well as how those indices connect back to the recommendations they provide you. If their recommendations feel like guesswork, that’s going to negatively impact your on-farm decisions down the road.
How do you account for the differences in crop / soil practices / phase of development, etc?
Satellite imagery is (quite literally) not black-and-white. Recommendations will — and should — change based on differences in tillage, crop phase, weather, etc. Any platform you use should make recommendations ONLY when they’re completely confident in their data. If they can’t account for the variety that’s inherent to the process, then it will be hard to trust those recommendations.
The Sentinel Fertigation N-Time platform uses satellite imagery (updated almost daily) that’s particularly sensitive to crop nitrogen status, to provide timely, relevant fertigation scheduling recommendations. Read more about how farmers have used this type of imagery to improve their yields — and profitability — by checking out our case studies.
If you have questions about vegetation indices or how to integrate the right satellite imagery program into your operation, reach out to the Sentinel Fertigation team.
Comments