How to Build VR Seed Maps

Have you ever wished you could build a variable rate seed map for your farm but not known what to base it on? Well, you've come to the right place! I'm going to walk you through a simple (well... kind of simple) How To and then look for feedback from you.

First, purchase and learn a GIS software program that can enable custom equations based on multiple layers of data.

Second, build management zones based on everything you can get your hands on. Soil types are only the beginning. Consider using EC or EM to refine soil texture and water holding capacity maps. Layer up RTK elevation data to get slope and surface water flow. Create a soil OM layer (or soil nitrate or similar) from GPS soil test data. It's easiest if you create all of these in a polygon format and do a polygon site aggregation so that the intersection of the layers create a file that looks nice and has every attribute compiled into a single layer. This will be useful to run your equations on in the next step. Aim to get representation of water holding, texture, OM, Soil type and Elevation data all in one layer. (or keep it at site level with a surface... I recommend no narrower than 90 ft squares as it can lock up your controller in the field if too small, or too many rates.)

Thirdly, create a normalized yield map for goals and economic purposes. Remember, we're trying to balance soil characteristics with probable paybacks here so past yields are as much about risk tolerance as they are future goals. If you don't have yield maps you may opt for aerial imagery or other remote sensing estimation of variability. Farmers Edge has a nice tool to do this, as does Winfield's R7. If working with either of those is out of the question you can ad-hoc your own "zones" by downloading free satellite imagery and outlining color changes manually. (WARNING: only attempt this if you have very highly variable soils and a crop season that shows it off quite well.)

Next, build an equation in Excel to test your hypothesis. Don't hesitate to use lots of buried IF/THEN statements and some assumptions about impact of various metrics on yield. I've seen and done numerous, each with little iterations unique for the client. For example, multiple the OM value by the average OM value of this soil type and then apply that percentage to the seed rate determined prior by the EC value. Adjust all of this by the normalized yield. Some of these can get pretty hairy, but ultimately it is doable with the right combination of people and assumptions. (SIDE NOTE: this is only a little tongue in cheek. I'm actually quite serious. If you really want a prescription that matches up exactly with your goals, you are going to have to invest in understanding this process and working through it with people that can help you.)

Or, you can simply click on a soil type and drop in a seed rate. Precision Planting was first to enable this. This is actually a pretty respectable way to guess and inject your knowledge at the last moment. The trouble here is that not many have rate trials on every soil and variety combinations and many farmers are aware that their soil type zones are not perfect. Maybe a custom zone that you tap would help, but then how do you remember every zone?

One of the biggest issues I hear from clients is that they want rates relative to elevation. When you're planting you simply cannot stand adjusting a rate any direction but up when going into the low ground valley. Keep this in mind when your prescriptions suggest that you should cut back on that wet hole that tends to flood. Your brain will play tricks on you in the field and you need to study these maps ahead and imagine yourself out in the field... and be okay with the logic. If you see areas that don't make sense, simply override them. Play your trump card.

So, I mentioned earlier the lack of seed variety influence. One solution I've used is building 2 or 3 or more optional rates for every field. This way the high-low range matches up with your flex or determinant ear type. Remember, you as the farmer can always bump up rates in the field at a percentage level, too (or at least you can with some effort.) Use the percentage up or down based on exactly how early you are planting and/or your expectation for rain and heat.

Finally, what about nitrogen and weather? These play a huge role in the probable success rate of your variable rate seeding plans and odds for success. You already have different rates of nitrogen if you have variable soils. If your high OM soils are releasing more N, they can feed more plants than your thinner, high ground. If you plan to vary Nitrogen however you need to take your seed rates into account, else you accidentally over compensate and assume flat rate seed inappropriately for your nitrogen and apply too little nitrogen for the increased seed rate. I mention this because it can have an impact on seed rates to a very large degree.

I've been doing this for years and testing rates and believe we have some great equations. I know our clients that get committed and work with us find a solution that meshes with their goals. However, I also realize this process is not for everyone (nay, majority.)

So, instead of asking everyone to step up their game we set about building software that would do it precisely how we wanted, automating most steps and giving the grower something really tight that they could trump. At the risk of looking salesy, here is a link to read more: www.Optmzr.co

In the end, export your rate files and test them in your monitor early. We've even built test files for the nearby hay field to make sure everything was working smoothly!

I really look forward to hearing how you do things and maybe answer a few questions. It's never too late to discuss!