December 4, 2024 • 32 mins
Elevate your understanding of modern agriculture with Travis Yeik, a trailblazer in precision agriculture technology. From his ranching roots in southeastern Wyoming to his role as software developer at GK Technology, Travis' journey is a testament to creativity and hard work. Learn how his agricultural background has shaped his success in programming, blending cutting-edge technology with real-world farming needs.
Explore the fascinating world of remote sensing and its innovative applications in farming and beyond. Travis shares insights on identifying and monitoring species like sagebrush and Phragmites, leveraging the unique wavelength signatures of different vegetation. Dive into the history of remote sensing, the importance of NDVI, and how these technologies unlock critical insights during the growing season to evaluate crop maturity and health.
Helpful formulas:
Normalized Difference Vegetation Index (NDVI)
NDVI = (NIR - RED) / (NIR + RED)
Green NDVI (GNDVI) = (NIR - GREEN) / (NIR + GREEN)
GK Technology, Inc. https://gktechinc.com/
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Sarah (00:00):
And now it's time for a Geek Speak with GK Technology's,
Sarah and Jodi friends and Ican't wait to get in the fields
(00:30):
again.
No, I can't wait to get in thefields again.
Welcome back to Ag Geek Speak.
We have another amazing guesthailing from the home office of
GK Technology, remotely locatedall over the United States, and
today the GK Technology manbehind the curtain, Travis Yeik,
is joining us.
He's a programmer with us hereat GK Technology.
Honestly, he does a lot ofthings and I don't think people
(00:53):
really know him well at all, sovery excited to have you on
today.
Thank you very much, Travis,for joining us.
Travis Yeik (00:59):
I am excited to be here and excited to share some
of the background thatabsolutely, as you say, no one
sees.
So maybe we can shed some lighton that and let people know
about the programs andtechnology behind everything
that they use for precision ag.
Jodi (01:16):
Absolutely.
Hopefully we can pull back thatcurtain a little bit today and
shed some light.
We mentioned that you'relocated remotely.
Where are you at right now andwhere is that in relation to
where you grew up?
Travis Yeik (01:27):
I am in Sheridan, Wyoming, and, let's see, we got
a little bit of snow on themountains here last weekend, so
everything's super green andgreat.
I am about four and a halfhours or so from where I grew up
, which is in southeasternWyoming, on a ranch outside of
Veteran, Wyoming with about apopulation of seven people.
So I'm remote, yep.
Jodi (01:48):
Fantastic.
Okay, and you mentioned youcame from a ranch background, so
you have an agriculturebackground.
How did you go from that agbackground to becoming what you
are now a programmerextraordinaire?
Travis Yeik (02:00):
Yeah, I think that's really important because
you don't see many people whogrew up on the ag and the
background there to go intobecoming the nerds that do all
the development behindeverything.
Sarah (02:15):
It's so true, right?
I mean, you don't think aboutbeing out there just doing like
the day-to-day monotonous thingsand then all of a sudden you
end up becoming this huge geek,this huge nerd, developing new
products all the time.
Travis Yeik (02:30):
Yeah.
So I want to step back.
I want to go into why I chosethe career, I guess, and the
path that I did, and how thatrelates to the bigger picture of
Precision Ag.
Just to put it into perspective.
You know, I think Precision Agare optimizing our returns on
some of the inputs Ag Just toput it into perspective.
I think Precision Ag areoptimizing our returns on some
of the inputs that we have to beable to preserve our resources.
And as a developer, that meansthat I have the responsibility
(02:54):
to have the means and themethods to make this practical
to everyday people such as ourfarmers and our consultants and
whatnot.
That's where it started and Ithink Precision Ag started.
Everybody thinks of PrecisionAg.
They think of auto steer.
You know that's the basic.
Where it kind of started evenwas trying to reduce overlap
when they were applyingchemicals and save money that
(03:17):
way.
You know that's what I hope.
I think more auto steer wasprobably developed from a hired
man who falls asleep at thewheel.
Sarah (03:25):
But you're right, Travis, everybody loves auto steer If
for no other reason than justnot having to actually
physically drive.
Travis Yeik (03:31):
It's very practical and everybody needs it now in
this day of age, but I don'tknow if we ever think about
where it comes from in thebackground.
There had to have been some guybehind it that developed it,
put it all together.
So we talked to Darin Johnson,the owner of GK Technology,
several weeks ago, and kind ofhave the same background as him,
where we had to developsomething that we needed to
(03:52):
utilize because it wasn't outthere.
That's how it all got started.
Sarah (03:56):
You're right, though, Travis.
It takes somebody to developthose technologies like auto
steer.
I think at this point in timein agriculture it's just really
easy to take that for granted.
You get in the tractor cab andit just it happens, the auto
steer is there, the autoguidance is there, the section
control is there, and we justtake it for granted.
But somebody had to be in thebackground with that First of
(04:20):
all, dreaming up crazy ideaslike hey, let's not actually
drive the tractor.
Can you imagine how nutso thatmust have sounded the first time
around?
It is important for developersto have that imagination so that
we can think outside the boxand dream bigger things.
From that standpoint, it'ssuper interesting the work that
(04:40):
you do at GK.
Jodi (04:41):
Technology.
The one thing I was thinking aswe discussed your background
too is like so you are adeveloper for GK Technology.
What exactly does that mean?
Like at the end of the day,what does that mean for you in
terms of what you do on a dailybasis?
And like where do you startwith what you're working on and
where do you end?
Travis Yeik (04:57):
It always starts with a cup of coffee, right, and
it always ends with a beer, andit always ends with a beer.
That's where it has to go.
Yeah, that's a good question.
It usually is just a crunchfrom one day to the next, just
like everybody else.
(05:17):
But it's great because it'sbroken down into projects where
it is something different.
Every day it's a differentproject, every day it's a
different problem.
The good thing about it is thatthere is a solution at the end.
Which makes it great is thatyou get to sit down and come and
look back at what you did forthe day and say, hey, I
accomplished this.
It'll help, whether it's theprogram or the people who use it
.
That's the great thing about it.
Sarah (05:37):
That's pretty fun and let's talk a little bit more
about some of the things andsome of the projects that you're
doing at GK Technology.
But I want to take a stepfurther back.
You know you mentioned that youwere raised on a ranch.
Talk to us about where did yougo to school, what did you study
and what got you into computerprogramming along the way.
Travis Yeik (05:56):
Yeah, so growing up on the ranch, you know, we
raised our own cattle there andwe did dairy and we had some
beef cows.
We didn't milk, we just bredthem and then sold them when
they're bred.
We didn't have precision agso-called on the farm where we
did corn and alfalfa.
I don't know about you guys.
Did you guys grow up inprecision ag?
I mean?
Jodi (06:16):
this is a funny story.
So, like I now farm with my dadand brother in Western North
Dakota.
We've used light bar maybesince like 2003, which has been
fantastic.
That is a godsend just having alight bar to tell you whether
or not you're on your AB line todrive and steer straight.
But the second year that my dad, brother and I were farming
together which is the year 2022,we had bought an auto steer
system an EZ Steer from Trimbleto put in our planting tractor.
(06:38):
No offense to my dad, not tothrow him under the bus, but he
didn't really think about itbeing economical for us to use
that.
But after the second pass hemade with it, he texted my
brother and I and said "man,this thing is awesome, why
didn't we do this years ago?
So we don't use much of it, butwe do use now and we're
continuing to grow that everysingle year with investments,
but from my experience growingup on the farm, precision
(07:01):
agriculture is pretty new.
Sarah (07:02):
I graduated from college in 2004 and I think GK
Technology was incorporated in2006.
I can remember being in atractor and it had the John
Deere Brown box in it and thatwas great.
But a lot of this stuff waspretty new and ironically I
would say that my number oneteacher in precision agriculture
(07:24):
was Kelly Sharpe.
So really learning how to mapand all of that started for me
probably around 2006.
And it was GK that reallystarted teaching me how to do
things.
I didn't know what a shapefilewas.
I didn't know why you needed aboundary, Just to let you know
where I started.
Call me the old fart in theconversation, but there it is.
Travis Yeik (07:42):
We are all in the same shoes.
Yeah, we use it for our beef.
You know, for example, there isa spread between your select
and your choice carcass weights,and usually it can be, you know
, 10 to $20 or whatnot, and sowe would ultrasound each and
every one of our steers beforewe sent them off to the market,
so that we knew the marblingthat was in there.
(08:05):
And if they weren't up tochoice yet, then we could leave
them in the feedlot for a littlebit longer.
And I think that's even rightthere.
You know, that's precision agin the beef industry.
I guess I started in FFA.
I was.
I had a project.
I decided that I wanted to mapout the fields of our farm, just
kind of digitize whereeverything is and how things and
(08:29):
keep records, I guess,throughout the years, of how we
were doing stuff.
And at that time there was nosoftware for doing that so, and
I was in high school, so I didthis all in Microsoft Paint, you
know, back in 2004.
That's amazing, that was myfirst actual precision ag
(08:49):
project that I probably had.
Sarah (08:51):
So you?
How did you?
You made a map out of yourfields in paint.
Travis Yeik (08:58):
Yes.
Sarah (08:59):
Were you able to get them geo-referenced eventually?
Travis Yeik (09:02):
At that time in high school I didn't have access
to, you know, to imagery.
I might've overlaid it onto aDRG map.
It was more for record keepingand so I had all the numbers and
stuff there, but it wasn'tprobably accurate with the geo
reference.
That's so cool.
Jodi (09:22):
You know that's really interesting, such a great idea,
because I think about how wekeep records on our farm, which,
again, I know it's 2024 and weprobably don't do the best job
of it.
But I know it's so much easierfor us to sit it down with a map
, like a printed out map of thearea, and just write on it what
we did, and it's so much moreimpactful having that map next
to the action we took versusjust having it in a list or even
(09:44):
like a spreadsheet.
I don't know why, but havingthat visual it helps us a lot.
That's so smart that beforehaving those digital maps you
would have it in paint, right?
You have that visual record ofthe map field that you could
then reference or keep actionsand keep records of things too.
Travis Yeik (09:58):
I'm sure my FFA teacher thought I was nuts
because it was something new atthat time and I was like, hey,
we need this.
Sarah (10:06):
Yeah, you crazy kid who's ever going to want to have a
map of their field.
Travis Yeik (10:11):
Yeah, so after that I went to the University of
Wyoming, which is Wyoming's onlyfour-year university, in
Laramie.
I went into GIS, which isGeographic Information Science.
That was really new at thistime.
People thought, maybe peoplethought that everything in the
world is already mapped out.
Why do we need more maps?
(10:32):
And GIS isn't just aboutcreating maps, it's about
modeling the world that we have,putting that in maps, so
whether it's your UPS driver andgoing around to find the best
routes or, as what we do,creating models of agriculture
in our fields.
That way In the universitythere at Wyoming I got
introduced to a guy who didremote sensing guy who did
(11:00):
remote sensing and that's, Iguess, when my career really
took off and I gained.
My interest was in the remotesensing aspect of it.
And if you don't know, remotesensing is looking at imagery in
different band waves of lightsand understanding how that
relates to our real world,whether it's vegetation or soils
or any geographic aspect ofwhat we look at.
And so with that project we ranaround and we mapped out
(11:23):
sagebrush in Wyoming.
And then after that I gothooked up with another guy there
at the university that wasdoing starting with remote
sensing and agriculture.
That was, I think my firstproject with Precision Ag was
looking at these pivots inWorland, wyoming and, without
ever going to the field, usingthe different wavelengths and
the vegetation indices to mapout what the production and dew
(11:46):
zones of those fields clear.
Back in I guess that would havebeen 2008 or so.
Sarah (11:51):
In that irrigation situation you said you were
mapping out differentproductivity areas of the
irrigated field.
Did it have anything to do withirrigation scheduling or how
did that work?
Travis Yeik (12:02):
It was strictly zone management and going back
through using Landsat, we lookedat the different dates and the
different years and to go backand say, hey, these are the
zones in that field.
Do micromanaging for thedifferent zones and parcels in
that field?
Jodi (12:17):
Was there an interest in doing that on irrigated land?
First because there could be apotential of higher ROI because
there was the rainfall there.
Why was the target under theirrigation fields in the first
place?
Travis Yeik (12:29):
So that's a good question and it probably plays
into the next part of my story,which is that, going from
Wyoming to Nebraska, everythingin Wyoming is irrigated is why
and that's the major reason Idon't know why we're farming in
Wyoming, but yes, that's asimple answer and I like that a
(12:51):
lot.
Jodi (12:52):
And can I ask you about that sagebrush remote sensing?
Does sagebrush like the speciesof plant itself?
Does it have a specificwavelength or like color
reflectance that makes it easyto identify what's considered
sagebrush or not?
Travis Yeik (13:05):
Yeah, and there's like four or five different
types of sagebrush here inWyoming.
We have Wyoming Ansis and BlackSage and Mountain Sage, and
then there's some otherdifferent types Silver Sage, and
so, yeah, each of thesedifferent types of sagebrush has
different wavelengths and weused a model to go through and
we would say this in this areawe did it by hectare.
(13:26):
This area was this density ofthis type of sagebrush or a mix
of these types of sagebrushes,and the goal was to look at the
sage-grouse which was becomingan endangered species at that
time and to go through.
But there's probably specificsignatures of like greenness for
a soybean field versus a cornplant that just looking at like
the difference in how those twocrops reflect light.
Jodi (14:05):
You could figure out what crop it is.
Travis Yeik (14:08):
The profile or whatnot of the phenology, even
of the plant and how it changes,but, yeah, the spectrum of
light visible in the infraredand the wavelengths.
After college I went to and gota graduate degree and I was
looking around for I knew Iwanted to go into agriculture,
remote sensing, and I waslooking around at the different
colleges to get a graduatedegree.
There was two colleges, Purdueor the University of Nebraska.
(14:31):
They were specialized inagriculture, remote sensing, and
so I went to Lincoln, Nebraska.
The advisor that I had there wasthe guy who pretty much started
, one of the first guys whostarted remote sensing in
agriculture.
I remember him telling us astory of NASA came to them with
a grant to say, hey, we havethis Landsat program and it
(14:53):
makes images and what can you dowith it?
And I remember they sent himthe data and he had just a big
sheet of I think it was at thattime probably eight bit data of
Lansat, so numbers one throughtwo, three, four, five, six,
seven, eight was each pixel andthey had to go through and kind
of just circle areas and say,hey, this is forest or this is
agriculture areas, and that washe made it so that NASA saw the
(15:17):
worth the use in satelliteimagery for agriculture.
Sarah (15:21):
And that was your major professor that did that.
Yeah, yeah, that's huge, that'samazing.
And so you worked with himdoing remote sensing and working
with, like, different bands oflight.
Travis Yeik (15:33):
Yeah, and so another professor that we had
there at Nebraska his name wasAnatoly Gittleson and he is the
guy who created the green NDVI.
He was from Israel and he wentinto the quantitative
understanding of how and whylight reflects and interacts
with vegetation, which was superinteresting, and learning how,
(15:54):
why at that red edge rightbetween the red reflectance and
the near-infrared reflectance,why it shifts to a different
wavelength as that vegetationgrows, and how that relates with
the chlorophyll within theplants.
Sarah (16:07):
So essentially, if that's the case at the red edge, can
we sense differences in like thecrop maturity If you could do
enough research with aparticular plant, can you see
how like that crop is growing?
And mat maturity If you coulddo enough research with a
particular plant, can you seehow like that crop is growing
and maturing?
Can you actually see thedifferences?
Travis Yeik (16:24):
Yeah, for sure.
And then the health of the crop, and I think that's why a lot
of times people use NDVI andthey use NDVI throughout the
entire life of the crop, which,from a remote spencing aspect,
is good for certain times of theyear but not for the entire
growing season.
Jodi (16:41):
Can you go into more depth about like what is NDVI, what
is green NDVI, what is red NDVI,and like where are some good
uses for each of those?
Travis Yeik (16:50):
I think that one of the very first indices that was
ever used for remote sensing inagriculture, or even just
remote sensing of vegetation ingeneral, is just the simple red
index, which is just red overnear infrared.
From there then it expanded alittle bit and a guy named
Tucker he developed NDVI.
It is a normalized differencevegetation index which means
(17:13):
that if, given that the red andthe near infrared, if it's a
really sunny day or it's acloudy day, if you take out some
of the differences, that it'snormalized in that way.
And so what happens?
If you look at vegetation andwhy is it green?
Well, it's because it reflectshigh in the green and if we were
(17:33):
to see in the near infraredvegetation would be near
infrared color if they had acolor, because it reflects a lot
in the near infrared, up to 50,60%, and so it absorbs all the
lights in the blue infraredwhich is being reflected and
(17:55):
said, as that plant ages and asit starts to senesce, that red
is what most changes in theplant because it goes from being
let's see it's hard to describeit without showing papers or
graphing, to show it over withwords here but as the plant
grows, the red reflectancechanges from being soil which is
(18:19):
kind of brown and high in red,and so as the plant starts to
grow, then that red slowlystarts to decrease and so the
amount of absorption of red isreally high.
In the amount of reflectance inthat red then is about 2% and so
it's really low.
As the plant starts to age andsness, or to stress, even, then
(18:40):
that red goes up, and so thenthat difference between that red
and that near-infrared, we cancorrelate that with how well the
plant is doing.
Jodi (18:48):
Wow, okay.
So if I'm thinking about this,right, if I'm somebody that say
is looking at NDVI?
When I say NDVI, what is thatassumed to be?
Is that usually red NDVI?
Travis Yeik (18:59):
That is.
Yeah, that's the red, otherwisethey'll specify that as green
NDVI.
Jodi (19:04):
Okay, that is really good to know.
And so like thinking about that, if a plant gets more stressed,
it reflects more red NDVI.
So, like, if I'm going weeklyand taking a look at red NDVI
values for a crop throughout thegrowing season, if I've got
some spot that's getting drownedout, like, say, in July,
because of a rainstorm, I cantell that where those wet spots
are, because they would have ahigher red NDVI value.
(19:27):
You know, after the rainstorm,where it's getting puddled,
where it's more stressed versus,like, the areas around it, am I
thinking about?
Travis Yeik (19:33):
that right, kind of just the opposite, but a lower
NDVI yeah.
Jodi (19:38):
This wasn't on the list of like areas to get into the
weeds on, but we all use thesenumbers but I don't think we
think enough about how theyactually work or like how they
actually mean right.
I always look at differentresources to like remind myself
how to use the number instead ofjust like knowing and
understanding, and I thinkpeople feel more comfortable
with these values when theyunderstand how they work.
Sarah (19:56):
So we talked about what NDVI is and you said that your
major professor invented greenNDVI.
So what is green NDVI and wheredo you think it has its
strengths?
Travis Yeik (20:06):
So when you look at vegetation as it increases in
density, that it is kind oflogarithmic in that as
vegetation increases about 60%of vegetation fraction or the
amount of coverage then youstart to get these leaves
underneath of the canopy thatstart stacking up against one
each another and as that happensthe red is already being
(20:28):
reflected only at like two or 3%anyhow.
So you're not going to see anychange in red reflectance hardly
.
But what you will see is thatthings become darker green, and
so at that point is when greenNDVI becomes important, because
you see more change in the greenthan you do the red.
Jodi (20:46):
Because you're like the thicker canopy would be
absorbing more of the greenlight than, say, an area that
has a less thick green canopy.
So even though, like a reallythick or like a soybean field
for instance, it might alreadyonly be reflecting like two
percent red, so you're not goingto see much changes of it.
But what really starts tochange is how much it picks up
(21:07):
and absorbs the greenwavelengths.
So we can use the green ndvithen to measure differences in
like canopy.
Travis Yeik (21:14):
Thickness.
Yeah, exactly you got it,nailed it, yep do we use this
for white mold mapping?
Sarah (21:18):
yeah, would we use this for, like really super thick
green dense canopies?
Is this the place, the bestplace, for us to be using?
Travis Yeik (21:26):
Yeah, so and that's kind of I think what a lot of
the research has shown is thatpast 60% of vegetation, when you
can't see any more soil, that'skind of where the greening and
DVI.
So it's especially important,like when you get to corn, like
right before it tassels, and itrelates to how much yield you're
going to get.
That the green NDVI is superimportant to show how the
vegetation, or the health ofthat vegetation is changing you
(21:53):
know, once it's fully developed.
Sarah (21:54):
I think when we take a look in the software we see
opportunities for downloadingNDVI and green NDVI and I think
sometimes our customers reallyhave some big questions about
where using each one of thoseproducts fits in better,
especially into making zones.
So those are some really goodideas for how we can think about
those things differently andquite frankly, from somebody
who's actually studied in theremote sensing area quite a bit,
(22:15):
who knew that our computerprogrammer actually is a remote
sensing geek.
Jodi (22:21):
And on that same subject like what did you focus on when
you were a remote sensinggraduate student at UNL?
Travis Yeik (22:27):
Yeah, my thesis was actually about a weed called
Phragmites australis.
If you know, it's one of theinvasive wetland weeds which was
super interesting learning itand I didn't add it to my thesis
.
But it actually does shift inthat red edge in that
chlorophyll and at some point itstarts producing chlorophyll B
(22:47):
instead of chlorophyll A and youcan see that in that
reflectance.
One of the really cool thingsin Nebraska is that they had
their own plane for doinghyperspectral imagery.
And what hyperspectral imageryis?
That it splits in thewavelengths into one or two
nanometer differences, and soyou might have 900 or 1,000
different wavelengths instead ofour typical red, green, blue.
(23:11):
It splits it up and so you cansee how each wavelength inside
of that changes, which is neatto be able to understand
quantitatively how the differentvegetation reflects in each of
those wavelengths.
Sarah (23:23):
So this might be a bit of a loaded question, but you were
mentioning that in the weedthat you studied, chlorophyll A
and chlorophyll B kind ofhandled the red edge differently
, or you could kind of tell thatyou're getting later in the
season because the plant woulduse chlorophyll A and
chlorophyll B at different times, right?
Or move to the production ofeach one.
(23:43):
We could see that in the light.
And then we're talking aboutmultispectral images and how
that might play into weed ID.
So I have long thought that atsome point in time and this is
coming from somebody who's cropscouted and the number one thing
we need to do is positivelyidentify those weeds, and we
need to do so when they're veryyoung.
But do you ever think thatthere'll be a way that we'll be
(24:04):
able to harness multispectralimaging and these different
concepts of, like differentamounts of chlorophyll A and B
and different nuances withinweeds and how weeds will reflect
light, so that we can actuallyjust run some light across there
and then be able to identifythe weeds?
Travis Yeik (24:20):
Boy, that is a loaded question.
Sarah (24:22):
Isn't it good?
Travis Yeik (24:24):
And I think there's a lot of research into that and
I don't know.
That is one of the things thatfrustrated me about the
university is that a lot ofresearch goes into understanding
how remote sensing works andwavelengths and doing research
projects, but it never actuallygets out to the real world, to
production, to how businessesactually use it.
It stays in the university Tosay that we're actually going to
(24:48):
get to the point where we cando that.
I don't know.
Sarah (24:51):
Is it a bridge between actually getting that academia
theoretical concepts actuallyapplied into the real life, then
, or is it just that it's thatintense of research that it's
complicated to get from point Ato point B?
Travis Yeik (25:04):
Yeah, I think it is just complicated in the way of
the nature things, or that youmight study what happens in
North Dakota for soybeans mightnot apply to everywhere else.
Yeah, just like anything elsein farming, there's so many
variables involved To make abroad spectrum analogy that will
happen for everything.
That's tough to do.
Sarah (25:26):
And it is expensive research to conduct if you can't
get it applied across a broadacreage.
So that's an interesting thingto think about.
I knew this was going to be afun and in-depth conversation
and so far this is notdisappointed.
Jodi (25:39):
I am so excited because I'm a weed scientist by training
, so I didn't know you worked ona weed for your master's right.
Travis Yeik (25:45):
Yes, or was?
Jodi (25:47):
it that is so freaking cool and like okay, the
chlorophyll A versus chlorophyllB doesn't one absorb like a
different.
It's about light absorption,right?
Or like what difference betweenthe two types of chlorophyll?
Is there any understanding ofwhy Phragmites does?
Travis Yeik (25:59):
that I couldn't tell you because I didn't add it
to my thesis and so okay, yeah,you mentioned that.
Yeah, really.
Maybe just you know hypothesisthat this is what's going on.
Sarah (26:10):
Wow, so that's pretty fun .
So we've gotten so far, thatwe've gotten to grad school, and
that's how long theconversation has gone.
So that's pretty fun.
Let's go from there.
So you get done with yourmaster's degree.
Travis Yeik (26:28):
Yeah, so out of graduate school I lived in Omaha
for a bit and my first job waswith Valley Irrigation in Valley
, Nebraska, and I was hired astheir variable rate irrigation
agronomist, which is goodbecause it worked in with my
background, you know, withlooking at so I have miners and
soils and they needed someone touse the remote sensing skills
that I had to determine how muchwater needed put into each
specific part of the field foruse with their variable rate
(26:49):
irrigation.
So it went back to kind of whatI did is getting out of field
without ever going out to it andsaying, hey, this is how much
water is needed here, which isextremely tough to do, and they
did not have a program to beable to do that.
So that's kind of where I pickedup my coding skills.
I needed to create some method,some program for me to be able
(27:11):
to determine how to apply thecorrect amount of irrigation
water to these fields.
Sarah (27:16):
So wait a second.
All of this time you're inschool, you're working on all of
this stuff and you hadn't takenhad you taken a lot of computer
programming courses at all.
Travis Yeik (27:26):
I had taken one yeah.
Sarah (27:29):
So you really are like Darin Johnson in the fact that
you're pretty much a self-taughtcomputer coder then as well.
Travis Yeik (27:35):
And that's what it is.
It's like you have a use or youhave a need for something and
there's no one out there that isgoing to create that program
for you to be able to do that,and so you yeah, you got to make
ends meet and do it.
Sarah (27:49):
That is just amazing.
I had no idea.
I honestly thought thatsomewhere along the line you
actually had formal training incomputer coding.
That is super interesting thatall the powerful tools that
we've got are basically createdout of a couple guys' brains
that are self-taught computercoders.
Jodi (28:05):
That is no dig, by the way , because what they build is
amazing.
And I think too the folks thatare coding and putting together
the pieces of the software thatyou use to make decisions on
your farms.
They're basically agronomistsby training or farmers that
understand the practical side ofit so that when you press the
buttons together in ADMS to makeyour maps they know that you
(28:28):
want to do something practical.
They understand my whole pointhere.
My long little side here isthat Darin and Travis are a very
special breed of computerprogrammer and I think that has
helped them so much to createamazing software for
agriculturalists across.
Sarah (28:47):
North America.
Quite frankly, from a humblestandpoint, it is the most
powerful software on the markettoday.
I mean, we really do allow oursoftware does allow people to
actually do stuff with data,agricultural data.
Sorry, that was my humbleinterpretation of having the
most powerful software on themarket today.
Travis Yeik (29:02):
Thank, you, Travis.
Sorry, that was my humbleinterpretation of having the
most powerful software on themarket today.
Thank you, Travis, Thank you.
Sarah (29:05):
Darin, so that's where you learned how to computer code
is when you're at ValleyIrrigation.
Travis Yeik (29:11):
That's right.
That's right, yep.
And so I created a GIS programfor them and I was trying to
actually create an easy buttonfor me that I could just input
satellite images, maybe somesoils data, and get water
holding capacities and put inthe different root depths of the
different crops and just make ascript that I could put in and
just push it and I could sitback at my desk and say, yep,
(29:35):
there we go, done.
And I didn't quite make it thatfar.
I was with Valley for a year,my wife and I.
We wanted to get back closer tofamily, and so we moved back to
Wyoming or back to I guess.
We moved to Montana, we movedto Bozeman.
I couldn't find a job there fora couple of weeks.
I really wanted something inremote sensing and research
because I liked that aspect ofit.
(29:55):
A guy that I was working withat Valley, Jeff Branch he's up
in Canada.
He told me about this companycalled GK Technology and that
they were looking for a softwaredeveloper, and so I got Darin's
number from him and I rememberI called up Darin and I told him
I have a background inagriculture and I'm remote
sensing and GIS and that I knowhow to code and it just so
(30:18):
happened it was the samelanguage as what coding language
is what Darin does, and Darinhired me on the spot without
even looking at a resume.
Sarah (30:26):
I didn't know that either .
When I see the products thatcome out of Darin and Travis's
brains and what we actually getto work with, it's pretty cool.
Travis Yeik (30:37):
I think Darin had been looking for a programmer
for quite a while, maybe forfive years or so.
There's not a lot of guys outthere that has the special niche
market where Darin and I are.
Sarah (30:48):
Well, and to Jodi's point , you do have to understand a
little bit of something that'sactually going on out in the
field.
I do think that there are a lotof companies out there they
forget about the field aspect ofit.
They forget about what it'slike to actually try to farm to
make a profit out there.
They forget what it's like tomake the agronomic
recommendation and that thesetools are supposed to enhance
(31:09):
all of that.
I feel like sometimes intoday's world, precision
agriculture companies get socaught up in what's going on in
the stock market and making surethat their investors are happy
that they forget about what'simportant at the ground level on
the farm, and I feel like we'revery much so in touch with that
.
Our primary programmers.
Again, Darin had managed tofertilize your plant, and
obviously Travis has big historyin agriculture as well.
(31:33):
It's pretty interesting to hearabout how you got to GK.
It's really fun to pull thecurtain back on this For this
episode.
This is a good place to wrap itup.
We will have more, though, nextweek with the man behind the
curtain, Travis Yeik.
So thank you so much for Travisbeing here and we'll talk to
you next time.
And GK Technology we have a mapand an app for that, and I
(32:00):
can't wait to get in the fieldsagain.
Jodi (32:03):
No, I can't wait to get in the fields again.
And now it's time for a Geek Speak with GK Technology's,
Sarah and Jodi friends and Ican't wait to get in the fields
(00:30):
again.
No, I can't wait to get in thefields again.
Welcome back to Ag Geek Speak.
We have another amazing guesthailing from the home office of
GK Technology, remotely locatedall over the United States, and
today the GK Technology manbehind the curtain, Travis Yeik,
is joining us.
He's a programmer with us hereat GK Technology.
Honestly, he does a lot ofthings and I don't think people
(00:53):
really know him well at all, sovery excited to have you on
today.
Thank you very much, Travis,for joining us.
Travis Yeik (00:59):
I am excited to be here and excited to share some
of the background thatabsolutely, as you say, no one
sees.
So maybe we can shed some lighton that and let people know
about the programs andtechnology behind everything
that they use for precision ag.
Jodi (01:16):
Absolutely.
Hopefully we can pull back thatcurtain a little bit today and
shed some light.
We mentioned that you'relocated remotely.
Where are you at right now andwhere is that in relation to
where you grew up?
Travis Yeik (01:27):
I am in Sheridan, Wyoming, and, let's see, we got
a little bit of snow on themountains here last weekend, so
everything's super green andgreat.
I am about four and a halfhours or so from where I grew up
, which is in southeasternWyoming, on a ranch outside of
Veteran, Wyoming with about apopulation of seven people.
So I'm remote, yep.
Jodi (01:48):
Fantastic.
Okay, and you mentioned youcame from a ranch background, so
you have an agriculturebackground.
How did you go from that agbackground to becoming what you
are now a programmerextraordinaire?
Travis Yeik (02:00):
Yeah, I think that's really important because
you don't see many people whogrew up on the ag and the
background there to go intobecoming the nerds that do all
the development behindeverything.
Sarah (02:15):
It's so true, right?
I mean, you don't think aboutbeing out there just doing like
the day-to-day monotonous thingsand then all of a sudden you
end up becoming this huge geek,this huge nerd, developing new
products all the time.
Travis Yeik (02:30):
Yeah.
So I want to step back.
I want to go into why I chosethe career, I guess, and the
path that I did, and how thatrelates to the bigger picture of
Precision Ag.
Just to put it into perspective.
You know, I think Precision Agare optimizing our returns on
some of the inputs Ag Just toput it into perspective.
I think Precision Ag areoptimizing our returns on some
of the inputs that we have to beable to preserve our resources.
And as a developer, that meansthat I have the responsibility
(02:54):
to have the means and themethods to make this practical
to everyday people such as ourfarmers and our consultants and
whatnot.
That's where it started and Ithink Precision Ag started.
Everybody thinks of PrecisionAg.
They think of auto steer.
You know that's the basic.
Where it kind of started evenwas trying to reduce overlap
when they were applyingchemicals and save money that
(03:17):
way.
You know that's what I hope.
I think more auto steer wasprobably developed from a hired
man who falls asleep at thewheel.
Sarah (03:25):
But you're right, Travis, everybody loves auto steer If
for no other reason than justnot having to actually
physically drive.
Travis Yeik (03:31):
It's very practical and everybody needs it now in
this day of age, but I don'tknow if we ever think about
where it comes from in thebackground.
There had to have been some guybehind it that developed it,
put it all together.
So we talked to Darin Johnson,the owner of GK Technology,
several weeks ago, and kind ofhave the same background as him,
where we had to developsomething that we needed to
(03:52):
utilize because it wasn't outthere.
That's how it all got started.
Sarah (03:56):
You're right, though, Travis.
It takes somebody to developthose technologies like auto
steer.
I think at this point in timein agriculture it's just really
easy to take that for granted.
You get in the tractor cab andit just it happens, the auto
steer is there, the autoguidance is there, the section
control is there, and we justtake it for granted.
But somebody had to be in thebackground with that First of
(04:20):
all, dreaming up crazy ideaslike hey, let's not actually
drive the tractor.
Can you imagine how nutso thatmust have sounded the first time
around?
It is important for developersto have that imagination so that
we can think outside the boxand dream bigger things.
From that standpoint, it'ssuper interesting the work that
(04:40):
you do at GK.
Jodi (04:41):
Technology.
The one thing I was thinking aswe discussed your background
too is like so you are adeveloper for GK Technology.
What exactly does that mean?
Like at the end of the day,what does that mean for you in
terms of what you do on a dailybasis?
And like where do you startwith what you're working on and
where do you end?
Travis Yeik (04:57):
It always starts with a cup of coffee, right, and
it always ends with a beer, andit always ends with a beer.
That's where it has to go.
Yeah, that's a good question.
It usually is just a crunchfrom one day to the next, just
like everybody else.
(05:17):
But it's great because it'sbroken down into projects where
it is something different.
Every day it's a differentproject, every day it's a
different problem.
The good thing about it is thatthere is a solution at the end.
Which makes it great is thatyou get to sit down and come and
look back at what you did forthe day and say, hey, I
accomplished this.
It'll help, whether it's theprogram or the people who use it
.
That's the great thing about it.
Sarah (05:37):
That's pretty fun and let's talk a little bit more
about some of the things andsome of the projects that you're
doing at GK Technology.
But I want to take a stepfurther back.
You know you mentioned that youwere raised on a ranch.
Talk to us about where did yougo to school, what did you study
and what got you into computerprogramming along the way.
Travis Yeik (05:56):
Yeah, so growing up on the ranch, you know, we
raised our own cattle there andwe did dairy and we had some
beef cows.
We didn't milk, we just bredthem and then sold them when
they're bred.
We didn't have precision agso-called on the farm where we
did corn and alfalfa.
I don't know about you guys.
Did you guys grow up inprecision ag?
I mean?
Jodi (06:16):
this is a funny story.
So, like I now farm with my dadand brother in Western North
Dakota.
We've used light bar maybesince like 2003, which has been
fantastic.
That is a godsend just having alight bar to tell you whether
or not you're on your AB line todrive and steer straight.
But the second year that my dad, brother and I were farming
together which is the year 2022,we had bought an auto steer
system an EZ Steer from Trimbleto put in our planting tractor.
(06:38):
No offense to my dad, not tothrow him under the bus, but he
didn't really think about itbeing economical for us to use
that.
But after the second pass hemade with it, he texted my
brother and I and said "man,this thing is awesome, why
didn't we do this years ago?
So we don't use much of it, butwe do use now and we're
continuing to grow that everysingle year with investments,
but from my experience growingup on the farm, precision
(07:01):
agriculture is pretty new.
Sarah (07:02):
I graduated from college in 2004 and I think GK
Technology was incorporated in2006.
I can remember being in atractor and it had the John
Deere Brown box in it and thatwas great.
But a lot of this stuff waspretty new and ironically I
would say that my number oneteacher in precision agriculture
(07:24):
was Kelly Sharpe.
So really learning how to mapand all of that started for me
probably around 2006.
And it was GK that reallystarted teaching me how to do
things.
I didn't know what a shapefilewas.
I didn't know why you needed aboundary, Just to let you know
where I started.
Call me the old fart in theconversation, but there it is.
Travis Yeik (07:42):
We are all in the same shoes.
Yeah, we use it for our beef.
You know, for example, there isa spread between your select
and your choice carcass weights,and usually it can be, you know
, 10 to $20 or whatnot, and sowe would ultrasound each and
every one of our steers beforewe sent them off to the market,
so that we knew the marblingthat was in there.
(08:05):
And if they weren't up tochoice yet, then we could leave
them in the feedlot for a littlebit longer.
And I think that's even rightthere.
You know, that's precision agin the beef industry.
I guess I started in FFA.
I was.
I had a project.
I decided that I wanted to mapout the fields of our farm, just
kind of digitize whereeverything is and how things and
(08:29):
keep records, I guess,throughout the years, of how we
were doing stuff.
And at that time there was nosoftware for doing that so, and
I was in high school, so I didthis all in Microsoft Paint, you
know, back in 2004.
That's amazing, that was myfirst actual precision ag
(08:49):
project that I probably had.
Sarah (08:51):
So you?
How did you?
You made a map out of yourfields in paint.
Travis Yeik (08:58):
Yes.
Sarah (08:59):
Were you able to get them geo-referenced eventually?
Travis Yeik (09:02):
At that time in high school I didn't have access
to, you know, to imagery.
I might've overlaid it onto aDRG map.
It was more for record keepingand so I had all the numbers and
stuff there, but it wasn'tprobably accurate with the geo
reference.
That's so cool.
Jodi (09:22):
You know that's really interesting, such a great idea,
because I think about how wekeep records on our farm, which,
again, I know it's 2024 and weprobably don't do the best job
of it.
But I know it's so much easierfor us to sit it down with a map
, like a printed out map of thearea, and just write on it what
we did, and it's so much moreimpactful having that map next
to the action we took versusjust having it in a list or even
(09:44):
like a spreadsheet.
I don't know why, but havingthat visual it helps us a lot.
That's so smart that beforehaving those digital maps you
would have it in paint, right?
You have that visual record ofthe map field that you could
then reference or keep actionsand keep records of things too.
Travis Yeik (09:58):
I'm sure my FFA teacher thought I was nuts
because it was something new atthat time and I was like, hey,
we need this.
Sarah (10:06):
Yeah, you crazy kid who's ever going to want to have a
map of their field.
Travis Yeik (10:11):
Yeah, so after that I went to the University of
Wyoming, which is Wyoming's onlyfour-year university, in
Laramie.
I went into GIS, which isGeographic Information Science.
That was really new at thistime.
People thought, maybe peoplethought that everything in the
world is already mapped out.
Why do we need more maps?
(10:32):
And GIS isn't just aboutcreating maps, it's about
modeling the world that we have,putting that in maps, so
whether it's your UPS driver andgoing around to find the best
routes or, as what we do,creating models of agriculture
in our fields.
That way In the universitythere at Wyoming I got
introduced to a guy who didremote sensing guy who did
(11:00):
remote sensing and that's, Iguess, when my career really
took off and I gained.
My interest was in the remotesensing aspect of it.
And if you don't know, remotesensing is looking at imagery in
different band waves of lightsand understanding how that
relates to our real world,whether it's vegetation or soils
or any geographic aspect ofwhat we look at.
And so with that project we ranaround and we mapped out
(11:23):
sagebrush in Wyoming.
And then after that I gothooked up with another guy there
at the university that wasdoing starting with remote
sensing and agriculture.
That was, I think my firstproject with Precision Ag was
looking at these pivots inWorland, wyoming and, without
ever going to the field, usingthe different wavelengths and
the vegetation indices to mapout what the production and dew
(11:46):
zones of those fields clear.
Back in I guess that would havebeen 2008 or so.
Sarah (11:51):
In that irrigation situation you said you were
mapping out differentproductivity areas of the
irrigated field.
Did it have anything to do withirrigation scheduling or how
did that work?
Travis Yeik (12:02):
It was strictly zone management and going back
through using Landsat, we lookedat the different dates and the
different years and to go backand say, hey, these are the
zones in that field.
Do micromanaging for thedifferent zones and parcels in
that field?
Jodi (12:17):
Was there an interest in doing that on irrigated land?
First because there could be apotential of higher ROI because
there was the rainfall there.
Why was the target under theirrigation fields in the first
place?
Travis Yeik (12:29):
So that's a good question and it probably plays
into the next part of my story,which is that, going from
Wyoming to Nebraska, everythingin Wyoming is irrigated is why
and that's the major reason Idon't know why we're farming in
Wyoming, but yes, that's asimple answer and I like that a
(12:51):
lot.
Jodi (12:52):
And can I ask you about that sagebrush remote sensing?
Does sagebrush like the speciesof plant itself?
Does it have a specificwavelength or like color
reflectance that makes it easyto identify what's considered
sagebrush or not?
Travis Yeik (13:05):
Yeah, and there's like four or five different
types of sagebrush here inWyoming.
We have Wyoming Ansis and BlackSage and Mountain Sage, and
then there's some otherdifferent types Silver Sage, and
so, yeah, each of thesedifferent types of sagebrush has
different wavelengths and weused a model to go through and
we would say this in this areawe did it by hectare.
(13:26):
This area was this density ofthis type of sagebrush or a mix
of these types of sagebrushes,and the goal was to look at the
sage-grouse which was becomingan endangered species at that
time and to go through.
But there's probably specificsignatures of like greenness for
a soybean field versus a cornplant that just looking at like
the difference in how those twocrops reflect light.
Jodi (14:05):
You could figure out what crop it is.
Travis Yeik (14:08):
The profile or whatnot of the phenology, even
of the plant and how it changes,but, yeah, the spectrum of
light visible in the infraredand the wavelengths.
After college I went to and gota graduate degree and I was
looking around for I knew Iwanted to go into agriculture,
remote sensing, and I waslooking around at the different
colleges to get a graduatedegree.
There was two colleges, Purdueor the University of Nebraska.
(14:31):
They were specialized inagriculture, remote sensing, and
so I went to Lincoln, Nebraska.
The advisor that I had there wasthe guy who pretty much started
, one of the first guys whostarted remote sensing in
agriculture.
I remember him telling us astory of NASA came to them with
a grant to say, hey, we havethis Landsat program and it
(14:53):
makes images and what can you dowith it?
And I remember they sent himthe data and he had just a big
sheet of I think it was at thattime probably eight bit data of
Lansat, so numbers one throughtwo, three, four, five, six,
seven, eight was each pixel andthey had to go through and kind
of just circle areas and say,hey, this is forest or this is
agriculture areas, and that washe made it so that NASA saw the
(15:17):
worth the use in satelliteimagery for agriculture.
Sarah (15:21):
And that was your major professor that did that.
Yeah, yeah, that's huge, that'samazing.
And so you worked with himdoing remote sensing and working
with, like, different bands oflight.
Travis Yeik (15:33):
Yeah, and so another professor that we had
there at Nebraska his name wasAnatoly Gittleson and he is the
guy who created the green NDVI.
He was from Israel and he wentinto the quantitative
understanding of how and whylight reflects and interacts
with vegetation, which was superinteresting, and learning how,
(15:54):
why at that red edge rightbetween the red reflectance and
the near-infrared reflectance,why it shifts to a different
wavelength as that vegetationgrows, and how that relates with
the chlorophyll within theplants.
Sarah (16:07):
So essentially, if that's the case at the red edge, can
we sense differences in like thecrop maturity If you could do
enough research with aparticular plant, can you see
how like that crop is growing?
And mat maturity If you coulddo enough research with a
particular plant, can you seehow like that crop is growing
and maturing?
Can you actually see thedifferences?
Travis Yeik (16:24):
Yeah, for sure.
And then the health of the crop, and I think that's why a lot
of times people use NDVI andthey use NDVI throughout the
entire life of the crop, which,from a remote spencing aspect,
is good for certain times of theyear but not for the entire
growing season.
Jodi (16:41):
Can you go into more depth about like what is NDVI, what
is green NDVI, what is red NDVI,and like where are some good
uses for each of those?
Travis Yeik (16:50):
I think that one of the very first indices that was
ever used for remote sensing inagriculture, or even just
remote sensing of vegetation ingeneral, is just the simple red
index, which is just red overnear infrared.
From there then it expanded alittle bit and a guy named
Tucker he developed NDVI.
It is a normalized differencevegetation index which means
(17:13):
that if, given that the red andthe near infrared, if it's a
really sunny day or it's acloudy day, if you take out some
of the differences, that it'snormalized in that way.
And so what happens?
If you look at vegetation andwhy is it green?
Well, it's because it reflectshigh in the green and if we were
(17:33):
to see in the near infraredvegetation would be near
infrared color if they had acolor, because it reflects a lot
in the near infrared, up to 50,60%, and so it absorbs all the
lights in the blue infraredwhich is being reflected and
(17:55):
said, as that plant ages and asit starts to senesce, that red
is what most changes in theplant because it goes from being
let's see it's hard to describeit without showing papers or
graphing, to show it over withwords here but as the plant
grows, the red reflectancechanges from being soil which is
(18:19):
kind of brown and high in red,and so as the plant starts to
grow, then that red slowlystarts to decrease and so the
amount of absorption of red isreally high.
In the amount of reflectance inthat red then is about 2% and so
it's really low.
As the plant starts to age andsness, or to stress, even, then
(18:40):
that red goes up, and so thenthat difference between that red
and that near-infrared, we cancorrelate that with how well the
plant is doing.
Jodi (18:48):
Wow, okay.
So if I'm thinking about this,right, if I'm somebody that say
is looking at NDVI?
When I say NDVI, what is thatassumed to be?
Is that usually red NDVI?
Travis Yeik (18:59):
That is.
Yeah, that's the red, otherwisethey'll specify that as green
NDVI.
Jodi (19:04):
Okay, that is really good to know.
And so like thinking about that, if a plant gets more stressed,
it reflects more red NDVI.
So, like, if I'm going weeklyand taking a look at red NDVI
values for a crop throughout thegrowing season, if I've got
some spot that's getting drownedout, like, say, in July,
because of a rainstorm, I cantell that where those wet spots
are, because they would have ahigher red NDVI value.
(19:27):
You know, after the rainstorm,where it's getting puddled,
where it's more stressed versus,like, the areas around it, am I
thinking about?
Travis Yeik (19:33):
that right, kind of just the opposite, but a lower
NDVI yeah.
Jodi (19:38):
This wasn't on the list of like areas to get into the
weeds on, but we all use thesenumbers but I don't think we
think enough about how theyactually work or like how they
actually mean right.
I always look at differentresources to like remind myself
how to use the number instead ofjust like knowing and
understanding, and I thinkpeople feel more comfortable
with these values when theyunderstand how they work.
Sarah (19:56):
So we talked about what NDVI is and you said that your
major professor invented greenNDVI.
So what is green NDVI and wheredo you think it has its
strengths?
Travis Yeik (20:06):
So when you look at vegetation as it increases in
density, that it is kind oflogarithmic in that as
vegetation increases about 60%of vegetation fraction or the
amount of coverage then youstart to get these leaves
underneath of the canopy thatstart stacking up against one
each another and as that happensthe red is already being
(20:28):
reflected only at like two or 3%anyhow.
So you're not going to see anychange in red reflectance hardly
.
But what you will see is thatthings become darker green, and
so at that point is when greenNDVI becomes important, because
you see more change in the greenthan you do the red.
Jodi (20:46):
Because you're like the thicker canopy would be
absorbing more of the greenlight than, say, an area that
has a less thick green canopy.
So even though, like a reallythick or like a soybean field
for instance, it might alreadyonly be reflecting like two
percent red, so you're not goingto see much changes of it.
But what really starts tochange is how much it picks up
(21:07):
and absorbs the greenwavelengths.
So we can use the green ndvithen to measure differences in
like canopy.
Travis Yeik (21:14):
Thickness.
Yeah, exactly you got it,nailed it, yep do we use this
for white mold mapping?
Sarah (21:18):
yeah, would we use this for, like really super thick
green dense canopies?
Is this the place, the bestplace, for us to be using?
Travis Yeik (21:26):
Yeah, so and that's kind of I think what a lot of
the research has shown is thatpast 60% of vegetation, when you
can't see any more soil, that'skind of where the greening and
DVI.
So it's especially important,like when you get to corn, like
right before it tassels, and itrelates to how much yield you're
going to get.
That the green NDVI is superimportant to show how the
vegetation, or the health ofthat vegetation is changing you
(21:53):
know, once it's fully developed.
Sarah (21:54):
I think when we take a look in the software we see
opportunities for downloadingNDVI and green NDVI and I think
sometimes our customers reallyhave some big questions about
where using each one of thoseproducts fits in better,
especially into making zones.
So those are some really goodideas for how we can think about
those things differently andquite frankly, from somebody
who's actually studied in theremote sensing area quite a bit,
(22:15):
who knew that our computerprogrammer actually is a remote
sensing geek.
Jodi (22:21):
And on that same subject like what did you focus on when
you were a remote sensinggraduate student at UNL?
Travis Yeik (22:27):
Yeah, my thesis was actually about a weed called
Phragmites australis.
If you know, it's one of theinvasive wetland weeds which was
super interesting learning itand I didn't add it to my thesis
.
But it actually does shift inthat red edge in that
chlorophyll and at some point itstarts producing chlorophyll B
(22:47):
instead of chlorophyll A and youcan see that in that
reflectance.
One of the really cool thingsin Nebraska is that they had
their own plane for doinghyperspectral imagery.
And what hyperspectral imageryis?
That it splits in thewavelengths into one or two
nanometer differences, and soyou might have 900 or 1,000
different wavelengths instead ofour typical red, green, blue.
(23:11):
It splits it up and so you cansee how each wavelength inside
of that changes, which is neatto be able to understand
quantitatively how the differentvegetation reflects in each of
those wavelengths.
Sarah (23:23):
So this might be a bit of a loaded question, but you were
mentioning that in the weedthat you studied, chlorophyll A
and chlorophyll B kind ofhandled the red edge differently
, or you could kind of tell thatyou're getting later in the
season because the plant woulduse chlorophyll A and
chlorophyll B at different times, right?
Or move to the production ofeach one.
(23:43):
We could see that in the light.
And then we're talking aboutmultispectral images and how
that might play into weed ID.
So I have long thought that atsome point in time and this is
coming from somebody who's cropscouted and the number one thing
we need to do is positivelyidentify those weeds, and we
need to do so when they're veryyoung.
But do you ever think thatthere'll be a way that we'll be
(24:04):
able to harness multispectralimaging and these different
concepts of, like differentamounts of chlorophyll A and B
and different nuances withinweeds and how weeds will reflect
light, so that we can actuallyjust run some light across there
and then be able to identifythe weeds?
Travis Yeik (24:20):
Boy, that is a loaded question.
Sarah (24:22):
Isn't it good?
Travis Yeik (24:24):
And I think there's a lot of research into that and
I don't know.
That is one of the things thatfrustrated me about the
university is that a lot ofresearch goes into understanding
how remote sensing works andwavelengths and doing research
projects, but it never actuallygets out to the real world, to
production, to how businessesactually use it.
It stays in the university Tosay that we're actually going to
(24:48):
get to the point where we cando that.
I don't know.
Sarah (24:51):
Is it a bridge between actually getting that academia
theoretical concepts actuallyapplied into the real life, then
, or is it just that it's thatintense of research that it's
complicated to get from point Ato point B?
Travis Yeik (25:04):
Yeah, I think it is just complicated in the way of
the nature things, or that youmight study what happens in
North Dakota for soybeans mightnot apply to everywhere else.
Yeah, just like anything elsein farming, there's so many
variables involved To make abroad spectrum analogy that will
happen for everything.
That's tough to do.
Sarah (25:26):
And it is expensive research to conduct if you can't
get it applied across a broadacreage.
So that's an interesting thingto think about.
I knew this was going to be afun and in-depth conversation
and so far this is notdisappointed.
Jodi (25:39):
I am so excited because I'm a weed scientist by training
, so I didn't know you worked ona weed for your master's right.
Travis Yeik (25:45):
Yes, or was?
Jodi (25:47):
it that is so freaking cool and like okay, the
chlorophyll A versus chlorophyllB doesn't one absorb like a
different.
It's about light absorption,right?
Or like what difference betweenthe two types of chlorophyll?
Is there any understanding ofwhy Phragmites does?
Travis Yeik (25:59):
that I couldn't tell you because I didn't add it
to my thesis and so okay, yeah,you mentioned that.
Yeah, really.
Maybe just you know hypothesisthat this is what's going on.
Sarah (26:10):
Wow, so that's pretty fun .
So we've gotten so far, thatwe've gotten to grad school, and
that's how long theconversation has gone.
So that's pretty fun.
Let's go from there.
So you get done with yourmaster's degree.
Travis Yeik (26:28):
Yeah, so out of graduate school I lived in Omaha
for a bit and my first job waswith Valley Irrigation in Valley
, Nebraska, and I was hired astheir variable rate irrigation
agronomist, which is goodbecause it worked in with my
background, you know, withlooking at so I have miners and
soils and they needed someone touse the remote sensing skills
that I had to determine how muchwater needed put into each
specific part of the field foruse with their variable rate
(26:49):
irrigation.
So it went back to kind of whatI did is getting out of field
without ever going out to it andsaying, hey, this is how much
water is needed here, which isextremely tough to do, and they
did not have a program to beable to do that.
So that's kind of where I pickedup my coding skills.
I needed to create some method,some program for me to be able
(27:11):
to determine how to apply thecorrect amount of irrigation
water to these fields.
Sarah (27:16):
So wait a second.
All of this time you're inschool, you're working on all of
this stuff and you hadn't takenhad you taken a lot of computer
programming courses at all.
Travis Yeik (27:26):
I had taken one yeah.
Sarah (27:29):
So you really are like Darin Johnson in the fact that
you're pretty much a self-taughtcomputer coder then as well.
Travis Yeik (27:35):
And that's what it is.
It's like you have a use or youhave a need for something and
there's no one out there that isgoing to create that program
for you to be able to do that,and so you yeah, you got to make
ends meet and do it.
Sarah (27:49):
That is just amazing.
I had no idea.
I honestly thought thatsomewhere along the line you
actually had formal training incomputer coding.
That is super interesting thatall the powerful tools that
we've got are basically createdout of a couple guys' brains
that are self-taught computercoders.
Jodi (28:05):
That is no dig, by the way , because what they build is
amazing.
And I think too the folks thatare coding and putting together
the pieces of the software thatyou use to make decisions on
your farms.
They're basically agronomistsby training or farmers that
understand the practical side ofit so that when you press the
buttons together in ADMS to makeyour maps they know that you
(28:28):
want to do something practical.
They understand my whole pointhere.
My long little side here isthat Darin and Travis are a very
special breed of computerprogrammer and I think that has
helped them so much to createamazing software for
agriculturalists across.
Sarah (28:47):
North America.
Quite frankly, from a humblestandpoint, it is the most
powerful software on the markettoday.
I mean, we really do allow oursoftware does allow people to
actually do stuff with data,agricultural data.
Sorry, that was my humbleinterpretation of having the
most powerful software on themarket today.
Travis Yeik (29:02):
Thank, you, Travis.
Sorry, that was my humbleinterpretation of having the
most powerful software on themarket today.
Thank you, Travis, Thank you.
Sarah (29:05):
Darin, so that's where you learned how to computer code
is when you're at ValleyIrrigation.
Travis Yeik (29:11):
That's right.
That's right, yep.
And so I created a GIS programfor them and I was trying to
actually create an easy buttonfor me that I could just input
satellite images, maybe somesoils data, and get water
holding capacities and put inthe different root depths of the
different crops and just make ascript that I could put in and
just push it and I could sitback at my desk and say, yep,
(29:35):
there we go, done.
And I didn't quite make it thatfar.
I was with Valley for a year,my wife and I.
We wanted to get back closer tofamily, and so we moved back to
Wyoming or back to I guess.
We moved to Montana, we movedto Bozeman.
I couldn't find a job there fora couple of weeks.
I really wanted something inremote sensing and research
because I liked that aspect ofit.
(29:55):
A guy that I was working withat Valley, Jeff Branch he's up
in Canada.
He told me about this companycalled GK Technology and that
they were looking for a softwaredeveloper, and so I got Darin's
number from him and I rememberI called up Darin and I told him
I have a background inagriculture and I'm remote
sensing and GIS and that I knowhow to code and it just so
(30:18):
happened it was the samelanguage as what coding language
is what Darin does, and Darinhired me on the spot without
even looking at a resume.
Sarah (30:26):
I didn't know that either .
When I see the products thatcome out of Darin and Travis's
brains and what we actually getto work with, it's pretty cool.
Travis Yeik (30:37):
I think Darin had been looking for a programmer
for quite a while, maybe forfive years or so.
There's not a lot of guys outthere that has the special niche
market where Darin and I are.
Sarah (30:48):
Well, and to Jodi's point , you do have to understand a
little bit of something that'sactually going on out in the
field.
I do think that there are a lotof companies out there they
forget about the field aspect ofit.
They forget about what it'slike to actually try to farm to
make a profit out there.
They forget what it's like tomake the agronomic
recommendation and that thesetools are supposed to enhance
(31:09):
all of that.
I feel like sometimes intoday's world, precision
agriculture companies get socaught up in what's going on in
the stock market and making surethat their investors are happy
that they forget about what'simportant at the ground level on
the farm, and I feel like we'revery much so in touch with that
.
Our primary programmers.
Again, Darin had managed tofertilize your plant, and
obviously Travis has big historyin agriculture as well.
(31:33):
It's pretty interesting to hearabout how you got to GK.
It's really fun to pull thecurtain back on this For this
episode.
This is a good place to wrap itup.
We will have more, though, nextweek with the man behind the
curtain, Travis Yeik.
So thank you so much for Travisbeing here and we'll talk to
you next time.
And GK Technology we have a mapand an app for that, and I
(32:00):
can't wait to get in the fieldsagain.
Jodi (32:03):
No, I can't wait to get in the fields again.
Popular Podcasts
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
Therapy Gecko
An unlicensed lizard psychologist travels the universe talking to strangers about absolutely nothing. TO CALL THE GECKO: follow me on https://www.twitch.tv/lyleforever to get a notification for when I am taking calls. I am usually live Mondays, Wednesdays, and Fridays but lately a lot of other times too. I am a gecko.