I was recently talking to one of my local farmers who uses precision agriculture to manage his corn and soybean crops. His combine, sprayer and planter are all connected to the cloud, and he uses the data from the last 10 years to manage next year’s crop.
We were discussing ways to use the data he collects to better manage the farm. As we looked at his data, some fields had four years of yield history missing due to his hay business. He commented that it is strange how we are still managing hay fields on a field scale, but the rest of his crops are zone managed; yet alfalfa is often his most profitable crop. How do we improve hay production zones without yield and quality data? There are tools available to better manage forage production.
There has been a surprising amount of work done recently in this area, including remote sensing for yield, quality and stand evaluation; developing yield monitors and soil mapping. Last year started out very disappointing for alfalfa growers in my area, with large areas winter-killed. As we started to assess stand damage, walking every field and accurately documenting damage was not practical. Instead, we started using drone imagery to map damage while making a plan for 2019 forage needs.
One of our strategies was to plant annual grasses in the field to fill in the dead areas. This created even more variability in the field from forage quality to nutrient removal. To maximize grass yields, we needed additional nitrogen, except in the areas where the alfalfa survived. The other challenge we faced was the variation in forage nutritional value; the alfalfa sections needed sorted from the grass. If only we had a new baler that could have labeled the bales!
Satellite and drone imagery are being studied for use in determining forage yield and quality on both production fields and pastures. Many studies have found ways for this to be successful in pure stands. We used aerial maps last year to make variable-rate nitrogen applications only to the grass areas. Before remote sensory, the predictive equation for alfalfa quality (PEAQ) was developed. It uses height of alfalfa plants and maturity to predict neutral detergent fiber (NDF) and relative feed value (RFV). Using remote sensors to measure plant height, a quality map can be developed to help make management decisions of when to harvest each field.
The challenge with only sensing plant height is the effect of lodging on height. One method of avoiding this is to take readings between 14 and 23 days and predict future growth. Adding additional sensors, such as green normalized difference vegetation index (GNDVI), helped improve prediction of yield and quality but also increased data processing time, making it more expensive and time-consuming.
Baler yield monitors are the most reliable systems for yield mapping, when well-calibrated. For reliable yield data, the baler needs to at least be capable to measure mass flow and moisture content and then be calibrated at various pressure settings for bale density. The ideal solution would be a dynamic weight system built into the baler. This is an option on newer balers or as a retrofit system for some older balers. For calibration, individual bale weights can be obtained using an on-loader transducer system, which measures pressure in the hydraulic system to weigh bales. Uniformity is a challenge in the humid East with most fields being a mix of alfalfa and grass. This leads to different bale densities at the same bale pressure.
There are multiple yield monitoring systems available on the market for new and used balers. One of these systems, being developed by Clemson University, measures windrow size. It is designed to be installed economically on any size, shape or age of baler. It works well to define yield variability within fields. Windrow variability, such as patches being fluffier, can create errors that need cleaned from the data or marked in the field.
Another system available on the market for large square balers tracks yield and moisture with the ability to add preservative at variable rates as windrow moisture changes. This system will track RFV by averaging the weight of three bales on the accumulator and calculating bale density based on weight. It also helps keep the mass flow-calibrated. For this to work, when the field is mowed a sample is sent to a lab to determine RFV. Variation is calculated from this base value. Bales can be marked with a spray system or RFID tag to be sorted later.
As we get closer to accurate yield and quality maps, how are we going to use this information to improve our operations? The age-old use of yield maps to create fertility zones to replace nutrients removed by the crop is the first step. It is not uncommon to see yield variations of over a ton per acre, which is 13 pounds of P2O5 and 50 pounds of K2O. These zones will also be a great way to improve productivity in the field, by dialing in on what is causing the yield drag.
While it will vary greatly across the country, in my area the biggest yield drags have been due to soil variability, including pH, sulfur, soil diseases and drainage. In other areas, it may be salinity or water-holding capacity and irrigation needs. If water-holding capacity and irrigation are your limiting factors, installing soil moisture sensors in different yield zones may allow for variable rate irrigation, improving water-use efficiency and yield.
By finding the low-yielding zones, we can sample to help determine what is causing the problem. This could lead to choosing different varieties that will survive better in certain areas of the field. These varieties may not be the highest yielding but will have longer stand life in those soils because they are better adapted to them. Locally, some producers mix two varieties of alfalfa together because of field variability. Imagine if we knew the best variety for each zone of our fields and could vary varieties across the field, improving stand life and yield. Yield and quality monitoring in forage production will allow us to improve forage quality, environmental sustainability and profitability. How can you define and better manage variability in 2020?