There is no one silver bullet when it comes to image sources, advises Steven Ward, director, Geospatial Sciences, The Climate Corporation.
· Think about use and timing. How quickly do you need the image back?
· Consider if ground calibration is needed, as Ward advises with thermal imaging.
· Track developing imagery such as LIDAR, which has great potential for plant health. While currently cost prohibitive, it is getting cheaper and easier to work with.
If you aren't using imagery today, get started with a multitude of free or low cost systems currently available through retailers and entry-level subscriptions to field management services.
Satellite imagery is exploding with higher resolution, greater frequency and more availability than ever before, and all three factors are increasing at record rates. One image provider, Planet, whose images are used by FarmLogs, launched 88 Dove satellites in January and another 40 in May with the goal of capturing every acre on earth every day – except where clouds occur.
The optical resolution they offer has jumped from five square meters per pixel (smallest area in one color in an image) to 3.5 square meters, notes Tracy Blackmer, vice president, Science, FarmLogs. Even more dramatic is the possibility to get an image of your fields every single day of the year.
"Last year we were concerned about the frequency of image capture, and now the question is how to take advantage of the higher frequency without getting swamped by the number of images available," says Blackmer.
Image frequency, also referred to as temporal resolution or temporal cadence, is vital for comparing rapid changes in a field. Multiply daily or nearly daily imaging (depending on cloud cover) by 20 fields or more and the problem has flipped to being overwhelmed with images to review. That is where companies like FarmLogs come into the picture.
"We are close to having an image every day (depending on cloud cover) combined with a tremendous amount of data from a wide range of sources, including images going back to 2009," says Blackmer. "Using our model, our system notes how a field of corn looks and compares it to previous years when it was in corn. If it looks different, an email is sent to the grower alerting him where to look for the anomaly in the field."
Farmer compares imagery
Rob Sharkey is used to looking for anomalies. The Illinois farmer has tried imagery and field management software from multiple sources and has currently settled on Farmers Edge, in part for the quality of their images. "I like to look at different images like temperature and vegetation (normalized difference vegetative index -- NDVI) and put them together," he says. "You may see a bad spot, and when you go out to the field, you realize it is a small area of gravel you had never noticed before."
Sharkey compares today's satellite images to yield maps of even five years ago and notes the difference. "The spot may have shown up, but because yield maps were not gathered in real time, it may have shown up 100 yards from where it actually was," he says.
The satellite images even raise questions about the value of combine yield maps. "With today's technology, I can take an NDVI satellite image from late in the previous season, type in the total bushels harvested, and it will give me a yield map more accurate than the combine's yield monitor," adds Sharkey.
While Sharkey gives images a value just for scouting, it is when they are tied to other data that they really offer a return on investment." Combine an NDVI image with nitrogen rates, weather station collection of rainfall and growing degree days, and the Farmers Edges model can tell you how much more N to apply," says Sharkey. "This year my Farmers Edge recommendation was to not apply any more [nitrogen]. I tend to err on the side of putting more on, so I did some test strips and will find out at harvest if they were right."
This is the first year Farmers Edge has offered the N recommendation to growers through their comprehensive corn management system – Corn Manager. The model uses field-centric variables and NDVI image data to identify the growth patterns of the crop, its vigor and how it differs from other fields, taking moisture, fertility and drainage into account. A host of other factors, both current and historic, including tillage, inputs, soil properties (physical and chemical), historical weather data and more are also taken into account.
"Growers like Rob are using our model to decide when to apply, how much to apply and what to apply," says Solomon Folle, senior cropping system modeler, Farmers Edge. "This spring, we trialed it over thousands of acres, collecting over 36 million data points to provide our growers with accurate zone-based nitrogen recommendations. This fall, we will conduct further soil testing to collect information on residual nitrogen to continue the validation."
Imagery helps field zones
While imagery alone can be informative, it is the integration with all available factors that is invaluable, explains Steven Ward, director, Geospatial Sciences, The Climate Corporation. He adds that the more imagery, the better.
"Bringing in more spectral diversity lets us uncover more insights related to harvest timing, grain moisture, N stress and prescription recommendations in our Climate FieldView platform," says Ward. "Imagery (current and historical) allows us to create zones for farmers without previous year's data. Imagery gives us more data to work in context with other data layers, and that is what is invaluable."
Ward compares imagery alone to the cover of a book, whereas a model helps tell the whole story. Like Blackmer, he is very excited about the innovations in sensors and equipment and what the near-term future holds.
"In the next two to three years, we will see an absolute tidal wave of information coming in from space in increased resolution, cadence and spectral diversity," says Ward. "It will unlock much current research using sensors on airplanes that will soon be on satellites. It is an exciting time."