Editors' Notebook

Our 2018 Digi-Tour

Greg D Horstmeier
By  Greg D Horstmeier , DTN Editor-in-Chief
Connect with Greg:
The DTN Digital Yield Tour is based on state-of-the-art modeling. You can use links in state articles to visit the Gro Intelligence site and look at live yield estimates at the county level. (Map courtesy of Gro Intelligence)

We hope you've found valuable the articles around The DTN/Progressive Farmer 2018 Digital Yield Tour, powered by Gro Intelligence, which have been running on various DTN/PF platforms this week. If you've missed any articles, they are housed on our digital tour website at https://www.dtn.com/…

This has been an exciting project. It also is the realization of a dream DTN has had for many years: To share multi-state yield and crop condition information based on broad remote-sensing technology. DTN was the original digital information technology company in agriculture. It always felt a bit out of step for us to tromp around in random farm fields doing yield checks by hand. Not to slight the importance of getting into the field. It's important to touch the crop, ground-truth the computer models. We've attempted to do a bit of both in this year's digital tour.

All methods of estimating crop yields have their plusses and minuses.

Anyone who has been part of an on-the-ground tour knows how addictive they can be. Every field stop is like a treasure hunt. What will this field reveal? What bugs or disease can you find? Who in the car is the ace at pegging yield estimates?

At least that's the vibe at the beginning of such tours. By 4 p.m. on the third day, you've been coated with enough pollen, suffered enough leaf cuts, and had enough what-just-slithered-between-my-feet moments to last a while.

A common farmer complaint of traditional tours is the limited number of fields checked. Even with multiple teams of crop scouts, only a fraction of the fields in any state get visited. Since most scouting teams grab a quick yield check at one spot in the field and then motor on, yield estimates and pod counts come from a tiny portion of those few acres.

P[L1] D[0x0] M[300x250] OOP[F] ADUNIT[] T[]

"Must not have come to my fields," is often heard when farmers comment about traditional yield tour results. "You must have gone to the best (or worst) spots in the fields," is another common retort.

Those downsides point to the allure of yield estimates that include satellite imagery and other remotely-sensed data. The "birds" are looking at every square foot of every field of every county in every state. Unlike USDA survey data, there's no sandbagging on yield answers or holes in the results because too few farmers responded in an area.

While they were spared the physical hazards of walking fields, the greatest challenge to our team of digital "scouts" was the discipline to stop digging deeper and deeper into the Gro yield maps and finally turn their stories in. The comment of "Hey, look at that spot on the NDVI map, wonder what's going on there," was not infrequent.

Satellite-driven models do provide that broader, more encompassing look. But this method also has its detractors. Many readers feel the "black box" could be manipulated to push the market one way or the other. That has been part of the behind-the-scenes enjoyment of working with the folks at Gro. They welcomed questions and criticisms. They are confident their models are smart, but also know the key to machine-learning models is they continue to get smarter with time. There's been a great, open conversation each day around that as we discussed various state conditions and differences in the Gro numbers versus USDA and other predictions.

Is this digital tour the de facto final word on the 2018 crop? Alone, no more so than traditional crop tours or reports from other consultancies and prognosticators. There are still critical weeks yet to go. There are a lot of promising bean pods to fill and corn kernels to pack in, or not, that extra bit of test weight. Then we need to get the crop tucked away in the bins.

From our perspective, the non-farmer players in the grain market game have been looking at satellite data, sending their legions of interns out into fields, and making trading decision based on information that farmers never see. So, we think providing another view on the market, publically, and adding some insights around that view, is well worth the effort.

We want to really thank the folks at Gro, who came to us with the idea of a digital tour, not knowing all the pent-up want for such a thing on our end. They opened up their very sophisticated yield models and results. They answered a stream of pointed questions from DTN's Katie Dehlinger, Emily Unglesbee and Pam Smith, who -- as the good journalists they are -- wanted to be sure of the validity of the data they were sharing before they did so.

Most importantly, they opened up their very proprietary and confidential insight pages for any and all to poke around in. These days, that sharing and openness was pleasantly refreshing.

Let us know what you think. I'd also encourage some yield checks of your own. There's a short story with common methodology for taking corn yield estimates and soybean pod counts on that DTN tour website I mentioned above. We'd love to see what you find. Send any and all comments and yields to us at talk@dtn.com. Let us know what county and state, whether the field is dryland or irrigated, and any other weather or pest conditions to note.

Just watch out for things slithering underneath the bean canopy.

Greg D. Horstmeier can be reached at greg.horstmeier@dtn.com

Follow him on Twitter @greghorstmeier

(ES/AG)

P[] D[728x170] M[320x75] OOP[F] ADUNIT[] T[]
P[L2] D[728x90] M[320x50] OOP[F] ADUNIT[] T[]

Comments

To comment, please Log In or Join our Community .