The way nutrients are applied to cropland has changed significantly over time, with adoption of various precision agricultural practices such as nutrient management planning, grid soil sampling and variable rate application technology. However, the way we handle manure nutrients has undergone little change.
Manure is a valuable asset in agriculture, as it is rich in the essential nutrients needed for plant growth and can return organic matter to the soil. However, as a fertilizer source, manure can be highly variable. Between farms, manure can have drastically different nutrient composition due to changes in animal type, feed materials, and manure collection and handling system, just to name a few. However, even within a manure lagoon, nutrients can vary. Settling of solids, also known as stratification, during storage periods can result in nutrient variations at different depths. This variability can be reduced by using proper agitation practices and taking numerous manure subsamples, but this is time consuming, expensive and variability will remain. To further improve nutrient management when applying manure, it would be highly beneficial to determine manure nutrient values on-site during application, so manure rates can be adjusted to meet nutrient needs.
Being able to determine manure nutrients quickly on-site without the need for complex wet chemical analysis is not a farfetched idea, and in fact academic and industry researchers have been evaluating near-infrared spectroscopy (NIRS) sensors as a mechanism for such analysis for years. The NIRS method uses the near-infrared portion of the electromagnetic spectrum to identify chemical bonds of a substance, including manure, which can then be used to predict nutrient composition. These sensors take thousands of readings per second and provide results instantaneously. Studies have found that NIRS sensors can be used to predict manure dry matter, organic matter, total nitrogen, ammonium, phosphorus and potassium of livestock manure. However, NIRS manure sensing has only been present in a laboratory setting, until now.
The John Deere HarvestLab 3000 is currently the emerging manure sensing technology on the market. The system can be incorporated onto manure tankers (Photo 1) or draglines to provide real-time estimates of manure nutrients.
Combining the sensing technology with tractor automation software, users can set nutrient prescription rates or limits, and the system will adjust the manure application rate on the go using speed or flow rate automation to match the desired nutrient rates.
Comparing real-time sensing with traditional methods
While the potential of the technology is appealing, does the technology improve nutrient application compared to traditional methods? There are several traditional methods used to determine manure nutrient values to make application recommendations. One of the most common methods used at livestock facilities is using manure samples collected during the previous application.
A recent trial at a Wisconsin dairy operation compared using the previous year’s data to the HarvestLab 3000 manure nutrient constituent sensing system for applying manure to meet nitrogen application rates. The traditional method used the previous fall’s manure sample data (consisting of 17 samples pulled throughout the hauling period) to calculate manure application rates to meet nitrogen application rates of 75 and 150 pounds of nitrogen per acre. For the NIRS application, the tractor automation software was set to apply manure to meet the 75 and 150 pounds of nitrogen per acre based on real-time NIRS total nitrogen data. Plots were then laid out, manure was applied, and pans were used to collect manure and evaluate if NIRS improved nitrogen application rates compared to the traditional sampling method (Photo 2).
Pan trial comparing prescribed versus actual nitrogen application rates
using manure sensing technology and traditional application methods. Photo provided by Joseph Sanford
Results from the small trial indicated that NIRS improved nitrogen application rates by reducing the percent difference from the intended application rate compared to using the previous year’s data (Table 1).
While the results from the study are promising, a lot more information is needed to further validate the system, including using other manure sources, comparing to other manure sampling and analysis techniques, and improving the sensing accuracy.
Potential uses for the sensing technology
One of the potential benefits to using manure sensing rather than traditional sampling is the possibility of reducing the variability caused by manure stratification in manure storages. Research has shown that despite proper agitation and taking numerous subsamples for manure analysis, variability can remain during field application. Using real-time measurements to apply manure rather than manure sampling for laboratory analysis could take this variability out of the equation.
Additionally, it could alter the way manure storages are agitated. By receiving real-time measurements, the need for completely mixed homogenous manure would no longer be necessarily needed, as the system could adjust application rates based on real-time data, potentially saving time and cost.
Finally, the system not only reads and adjusts manure application, but collects data on nutrient compositions for nitrogen, phosphorus and potassium. This data produced could have huge implications on nutrient management, allowing for better record keeping for regulatory purposes and helping farmers make more informed decisions on the need for additional fertilizer.
Considerations before using manure sensing technologies
While manure nutrient sensing has shown promise for improving the way manure nutrients are applied, a common question to ask is: Will it work for my manure? The answer is that it will depend. Systems using NIRS technology require significant calibration, using hundreds or thousands of test samples and complex analysis to fit statistical models to the data that can then be used to predict manure nutrients. However, these calibrations are only as good as the data used to establish them, and the predicted range of nutrient information can be highly limited.
The current calibrations have been developed and only valid for specific ranges of dry matter, total nitrogen, ammonium, phosphorus and potassium. If your manure nutrient composition is outside of these ranges, the data may be unreliable. Before using the manure sensing systems, one should ask what the current calibrations range is for the specific sensor and compare those ranges to the farm’s historical manure nutrient data to see if the sensor would work for their manure.
Another common question is: If using the system, will you no longer have to collect and send in manure samples to laboratories? While the systems on the market have shown promise to provide accurate data, they are not yet approved methods for manure nutrient analysis. Depending on state regulations or farm permitting, traditional sampling likely will still be required until the sensors become an acceptable method.
Joseph Sanford is also a Wisconsin Dairy Innovation Hub affiliate, department of biological systems engineering at the University of Wisconsin – Madison.
Iris (Xiaoyu) Feng is an assistant professor of agricultural and biosystems engineering at North Dakota State University and former Wisconsin Dairy Innovation Hub research associate, department of biological systems engineering at the University of Wisconsin – Madison.
