Case study:

Drones — a Soybean Grower’s Best Friend

DroneDeploy's Blog
Published in
8 min readAug 11, 2015


Depending on which side of the omnivorous line you stand, soy evokes a specific type of feeling inside of you.

Soybeans, soja, tofu, soy milk, edamame, soy ice cream, soy sauce — cattle are even often fed soy. Love it or hate it, soy is here to stay.

And soybean growers, like all in the ag industry, face a unique set of challenges when trying to make their living.

NIR orthomosaic layer, Norm’s white truck in center-right quarter of image

The information below was given to us during an interview with DroneDeploy mapper Norm L. of Eagle Scout Imaging who've operated an AgEagle drone in Canada since April 2015. Norm has approximately 50 successful DroneDeploy flights under his belt so far and expects to reach 100 by the end of the current growing season for a total of 10,000+ acres of mapped fields.

Click and open the below maps in a new tab to explore and reference them for the purposes of this case study:

The story

How much soy would a soy grower grow if a soy grower could grow soy?

This is the question most often asked by the grower before the start of the next growing season. The question, however, affects every single type of grower out there.

When ramping up for the next season, various types of different tests are run that compare different improvements — all with the end goals being to increase yields and efficiency.

With respect to this case study, we'll compare three:

  1. How should I apply my fertilizer to achieve maximum yields?
  2. What should I do to manage weeds?
  3. When exactly should I plant — earlier in the season or later?

Referring to NDVI Soybean Map 1 above we can get the lay of the land:

4 key areas in this map

There are around 55 acres of soybeans in this field, which Norm has flown multiple times.

The lighter-colored, vertically-oriented area North of the barn shows an area of weeds with bare ground making up the rest of the red area it is situated in (marked as ‘Weeds’).

The area immediately south of the barn is soybeans. There are two 5 foot wide, lighter-colored stripes running east-to-west (marked as ‘Fertilizer lines’). This area was an area that received a fertilizer overlap (or twice the fertilizer amount).

The flight

Using an AgEagle, powered by DroneDeploy, Norm mapped this 210 acre field comprised of soybeans, corn, hay, and wooded areas, in one flight of 37 minutes.

Setup was accomplished in less than 10 minutes.

The ‘L’ is the landing zone, with the takeoff point being just adjacent to it

The AgEagle collected 347 images of the field which were uploaded in mid-flight, via cellular modem, to DroneDeploy’s map engine where they were stitched together in real-time.

Flight plan showing Air speed vs. Ground speed vs. Altitude and actual flight path (identifying features purposely blurred)

Assorted vanity metrics for data lovers:

Flight efficiency, Battery performance, GPS performance, Altitude, and Network performance graphs — auto-generated by DroneDeploy

Total flight time, 37 minutes

Total image processing time, 290 minutes

Analyzing the data

The below insights were not apparent at all until after flying the drone and analyzing the resulting map. They would not have been noticeable on foot or by land-based methods.

When we point our attention south of the barn, we notice the two 5 foot wide horizontal stripes running roughly east-to-west.

When zoomed in and viewed in the interactive NDVI Soybean Map 1, you can see that the color (or health) of these rows are not proportional to the other strips in this section of field.

Fertilizer was applied to these areas of the soybeans via foliar application (applying fertilizer directly to the leaves).

The sprayer typically sprays in 20 foot sections. However, a miscalculation ended with two five foot overlapping sections — the two light colored strips shown in the map.

The overlapping fertilization of this field on these two horizontal rows was an unintentional result which caused a double foliar application of fertilizer to the soybeans.

This unintentional double foliar application resulted in a much healthier series of soybean plants when compared to the adjacent rows.

Setting our sights north

When we look north of the barn we are greeted with a pesky nuisance: weeds.

Seen in the NDVI Soybean Map 1 (and in a bit of a different coloring perspective in the Orthomosaic Soybean Map 1 version), the weeds are rampant and thriving.

However, this is not as bad as it may seem.

Last season, on this farm, the grower ran an intentional test for weed mitigation. The results of his work are fruitful.

The large red area with no weeds is actually barren earth — just soil. There are very little (if any) weeds to be found. In that large red area, the grower instituted a weed program (powered by a residual herbicide) last year that successfully prevented weeds coming up this year.

Grower: 1, Weeds: 0.

Part 2, Soybean Redux

Part 2 of our soybean case study references the exact same farm as part 1 above. However, the maps below are from 3 weeks later in the season. Do the same as you would in any of our interactive case studies — open up the maps in a new tab to reference them as you read along.

Explore these maps in new tabs:

A lot changed in just 3 weeks

Planting was well-underway and there were some key insights revealed to the grower after another drone flight.

Animated GIF, 3 week timeframe

More data analysis

After the 3 week wait, Norm flew his AgEagle once more for his client. He duplicated the exact same flight by using his saved flight plan in the DroneDeploy app.

The grower wanted to test staggered planting times so he planted one section of his field one week before the other.

Typical farm logic suggests that planting time doesn't make a big enough difference to pay too much attention to.

Norm and the grower discovered that was false.

The same soybean field from part 1 but flown 3 weeks after the 1st mission — left side is annotated for clarity, right side is the raw NDVI image

In the image above you can see that the grower initially planted soybeans in just the small southwestern plot, along with only the perimeter of the larger northeastern plot (1st).

One week later, the grower planted the interior of the larger northeastern plot (2nd). Again, this was done purposely for comparison.

While looking at the raw image (or interacting with NDVI Soybean Map 2) we can clearly see that the soybeans that were planted first are much more healthy (lighter color, closer to green) than the soybeans that were planted just one week later.

The grower could never have known this without the drone flight and the soybeans were a lot healthier than initially predicted.

Repeating the above, a common thought among soybean growers is that planting a few days — or even a week later — into the growing season doesn't make much of a difference. But this confirms that the planting time has indeed made a sizeable difference.

A young soybean sproutling

Earlier planted crops will have larger root masses and are better able to deal with moisture, weeds, and stress than their counterparts planted a week behind.

Norm and the grower never would’ve been able to prove this without seeing the NDVI map.


At this time , we won't know the exact figures saved until the grower puts his combine on the field and harvests.

However, we can certainly see that the soybean plants’ vitality is exceeding expectations — weed management was a success and a heavier, unintentional foliar application of fertilizer resulted in healthier plants in two of the crop rows south of the barn.

All of these incremental improvements lead to more flowers, which lead to more pods, which lead to more soybeans, which culminates into more profit for the grower and, of course, a job well-done by Norm.

These findings will be confirmed by yield maps via the yield monitor in the grower’s combine. Then the grower generates a shapefile based on that data. Norm will later layer the data from DroneDeploy with the generated shapefile to truly get a sense of the correlation of data.

For perspective’s sake, let’s get some numbers.

Let’s assume the grower increases yield by just 8% on this 55 acre section of soybeans and take today’s daily rate per bushel of soybeans at $10.69 for reference.

Assuming the grower produces an average of 58 bushels per acre, a fairly modest 8% yield increase equates to revenue increase of $2,728.09.


Extrapolating this 8% increase over to the average U.S. farm size of 441 acres (an 8x increase on our 55 acres of soybeans in this case study) results in a revenue increase of $21,874. Not bad at all.

The takeaway

Norm and the grower have successfully verified that the weed prevention program and staggered plant-time tests they had been running are effective. They were also quite surprised to learn that the accidental foliar fertilizer application improved overall plant health greatly in relation to the surrounding plants.

This insight would’ve never been realized without a drone.

The grower will be able to quantify and track the improvements he’s made to his field for this season and many seasons into the future.

DroneDeploy proved its worth by revealing insights into the soybean field in an extremely timely and efficient manner. The data is now stored in the cloud forever and will be used to correlate savings/profit growth with the results from the combine later in the season.

By Ian Smith, Sales & Marketing @DroneDeploy

Want to test this yourself?

Start your no risk, no credit card, 30-day DroneDeploy trial at

Not sure which drone is right for you?

Click here to see our latest drone and device compatibility



DroneDeploy's Blog

DroneDeploy is the leading cloud software platform for commercial drones.