At Its Core, Soil Sampling Can Be Improved

The soil testing process is ripe for improvements. ( Margy Eckelkamp )

The basic process of soil analysis as a fertility management tool has not changed much since the 1950s, but how that soil is farmed and what is expected from it have changed. At one time, pulling four or five soil cores from a 40-acre field, mixing them in a bucket and sending a single sample bag off to the local Extension office for testing was “good enough.”  

But just as fields and farms got larger, soil variability was recognized, and technology advanced. GPS introduced precision agriculture’s version of the classic game Battleship—it divided fields by 2.5-acre grids. It totally transformed how we approached soil sampling and field fertility. With the ability to precisely apply varying amounts of fertilizer in each grid within a field using that same GPS technology, the business of soil sampling ultimately shifted into overdrive.

Such a soil sampling renaissance only hyper-exposed the primary issue that has plagued the process from its very be-ginning. That issue is that it is still an intensively manual endeavor. Starting and stopping in a field every 330 feet, pulling six to eight cores and bagging and tagging each grid sample is mind-numbing and physically unsustainable on a large scale. So given that testing soil properties is now entrenched as the bedrock of most precision agriculture agronomy programs, it only makes sense that technologies are finally rising up to redefine what soil “sampling” and “analysis” look like in the 21st century.

A true robotic soil “sampler” is just one of those technologies. It goes directly after the “labor” issue at the heart of today’s soil sampling regime. Rogo Ag LLC based out of West Lafayette, Indiana, has developed its SmartCore autonomous soil sampling system. Using a tracked Bobcat skid-steer chassis, it navigates fields using digital field boundary algorithms and Lidar sensors, which transmit beams of light to detect objects in the surrounding environment. The robotic system collects samples through a hydraulic auger bit, which is configured so that no soil escapes the core. SmartCore is able to return to each location at any point in the future using RTK GPS and built-in navigational programming. This ensures accuracy, purity and depth control. The creators of SmartCore say such preciseness reduces sampling errors by as much as 20%, which can have significant implications on profit per acre.

But what if you didn’t need an actual soil lab? A small but significant step toward a virtual soil lab is a hybrid soil sampling model currently being deployed by SoilOptix, a company in Ontario, Canada. The system uses radiation sensors in conjunction with traditional soil sampling. It works by driving over a field with a small vehicle fitted with the scanner that measures the natural gamma radiation emitted by the soil. 

After the data are collected and “ground-truthed” with old-fashioned soil samples, the result is a resolution of more than 335 points per acre and up to 25 different layers of data. SoilOptix can map micro- and macronutrients, plant-available water, soil texture, pH and salinity at square-meter levels. 

But what if you could watch what’s going on with your soil 24/7 and not wait a year or even four years to get an update on its health? Teralytic is a New York-based soil analytics company that has developed what could be best described as a Fitbit for your fields. Each one of the company’s soil probes contains two sensors, including what they claim is the world’s first NPK sensor.

Through wireless technology, the probes can transmit weather, soil moisture, pH, salinity, light and even aeration and respiration taking place within the growing crop. It updates all that information every 15 minutes. 

Real-time data are also provided by Precision Planting’s SmartFirmer, which was introduced in 2017. The SmartFirmer is a sensor-ladened seed firmer that attaches to your planter’s row units. The firmer’s sensor measures temperature, moisture, furrow uniformity, residue and organic matter.  

Having such “real-time” soil information allows for “on-the-fly” variable-rate seeding and fertility prescriptions. The number of data points created by such sensors because of their continuous sets of streaming data is nothing short of mind-boggling. I’m sure there is much more planter or implement-based monitoring to come.

The soil testing process is ripe for improvements. Soil testing is the most basic of basics when it comes to producing a crop. And like one of my good precision ag colleagues likes to say, “If you don’t test, it’s just a guess.”