The best laid plans of mice and men often go awry; and, for anyone who has spent a season in agriculture, you know how true this may be.
For researchers and product developers, this can present unique challenges for plot trials. The only predictable aspect about the seasonality of our business is just how unpredictable it can be – and this can create new challenges for measuring the success of a trial.
These challenges can negatively impact results and skew the data being collected in research plots, making in-season research plot analysis vital in upholding data quality. By analyzing the quality of each research plots, deactivation plans can be put into place to ensure there aren’t significant areas affecting the overall quality of the data.
So, how and when can you gain this deeper insight into what is happening in your plots to ensure that the correct decisions are made?
Conducting a Research Plot Quality Analysis
Gathering data on individual plots throughout the season serves as imperative for most research trials, helping to understand and validate how traits or treatments respond to different variables. But, sometimes it helps to take a step back to analyze plots as a whole – and that’s when a full-scale plot quality analysis can be helpful.
A plot quality analysis looks at all plots and determines which ones may not be performing well, helping to remove them from analysis. What does “all plots” mean? That really varies on the research study at hand; it could be looking at all within a particular field, a region, or it could be any tied to a particular research trial.
With a plot quality analysis, it’s possible to measure different aspects to understand if there are any plots that have quality issues. Key points of validation may include:
- Yield and quality
- Track different trait performance
- Various growth stages
- Monitor crop stress
- Nutrient status
- Product efficacy
These help to understand what may be going well and not so well throughout your research plots. The measurements collected during an analysis can also help fuel decisions surrounding plot deactivation – if necessary – helping to preserve high-quality data and limiting potential skewed data.
Getting the Data You Need: Using Today’s Technology
With hundreds to thousands of plots, using manual methods – or having people walk the fields to get data on each and every one is extremely time-consuming – and real-time and efficient data is crucial. This is where remote sensing with aerial imaging can be extremely helpful and beneficial.
With ag drone technology, we can get a clear picture of the entire plot, which can then be translated into quantifiable data. This technology brings a high level of accuracy and precision, meaning that you can rely on it to make critical decisions this year – and use it to compare year to year data and performance or even with more accurate comparisons.
So, if there’s interest in conducting a few plot quality analysis checks throughout the year, getting a drone in the air can capture 100% coverage across plots, delivering imagery (RGB or multispectral) and detailed measurements to inform decisions. What’s more, if you have plots for the same trial in different regions, it eliminates the possibility of having differences in reporting key measurements and performance indicators. The drone will always capture the same data, even if it’s flown in different states, countries, or even at different points in the year (or over a course of a few years).
Once the data is captured, using an agronomic insights platform like FieldAgent can help detail what’s going on in the field, plot by plot. Take this example, which showcases Canopy Cover and uses two bin colors to indicate problematic plots:
As you can see, it displays a mostly green side to the right and gap ridden plots to the left in gray. Within the platform, researchers have controls to adjust the baseline measurement to determine which plots fall beneath that – in this case, understanding the ideal canopy measurement and finding areas where plants fell short (gray). This can help decide which variation performed best; as well as determine if deactivation needs to occur.
And, this data can be normalized to apply to your rating scale. For instance, if you have a rating scale like the below that focuses on 1-5, it may require creating bands to align based on the measurement scale in the platform:
This specifically highlights Sentera’s Canopy Cover analytic which is displayed by a linear scale to match up traditional research scoring systems like the visual above. This match up opens the door for quantifying the measurements captured, upping their value.
The Sample Rating Scale above best displays how our analytics are translated to match up with the scales being used in the ag research and development realm. This specifically highlights Sentera’s Canopy Cover analytic which is displayed by a linear scale to match up traditional research scoring systems like the visual above. This match up opens the door for quantifying the measurements captured, upping their value.
Plot Quality Analysis: Harnessing the Power of Your Data
In the image above, you can see an analyzed version of different research plots with the Canopy Cover highlighted. The middle section displays a mostly green side to the right and gap ridden plots to the left. This is done by adjusting the ideal canopy set by a researcher and the green depicts areas in the plots that met said requirements where the grey gaps fell short. This can help decide which variation performed best; as well as determine if deactivation needs to occur.
With the end goal of validating outcomes with objective data, truly understanding the data at hand is critical. After data capture, leveraging software tools for analysis helps paint a more tangible picture into plot performance. Key analytics are generated to understand the measurements being captured and to better apply them to your research.
This also makes matching and understanding which analytics to look at comes more naturally. With different analytics being designed for different times of year, matching the correct ones to growth stage is critical in understanding insight into growth, health, and performance – both to understand what’s happening within your plots, but to conduct wider analyses to optimize performance and truly understand what’s going on within your trials.
After all, even if the best laid plans go awry – showcasing resiliency can truly make a difference; and for researchers and product developers in agronomy, data serves as a critical piece to evaluate, respond, and measure response mechanisms to the challenges at hand.