Precision Agriculture Utilizing UAVs04/21/2017
With recent technology advancements in small UAS and with reasonable access to affordable sensors, the farmers and researchers of the agriculture industry are starting to adopt UAS technology. In theory, it allows the user to access a precise location, observe it with a plethora of sensors, and create maps of multivariate data. Growers are using them to e.g. detect pests, map weeds, assess growth, analyze moisture content to assist with irrigation or drainage, and optimize nutrient management. This helps not only to increase yield, but also to conserve valuable resources or limit the use of dangerous chemicals. The fact that the sensors are able to gather and display data that can impact decision making in real-time is a huge advantage.
Discussing what kind of impact UAVs can have on farming is about more than just talking about the data made available. For most agriculture professionals, it is about how the data gathered by UAVs can be properly utilized and what actions it will lead to.
If it was as easy as theory suggests we would likely have seen growers adopting this technology on a large scale. It is however much more complex. If we look at just a few of the parameters that affect the data,
- Soil type
- Water composition
- Type of crop
- Climate zone
we quickly see that not only is the problem multivariate, it will also require large number of statistical models to support the analysis in order to present a more generic solution, and thus enable larger scale adoption.
There are at least 570 million farms worldwide, of which more than 500 million can be considered family farms. Most of the world’s farms are very small, 84 percent of the farms are smaller than 2 hectares and they operate about 12 percent of the farmland. Conversely, 16 percent of the world’s farms are larger than 2 hectares and they represent 88 percent of the world’s farmland. In fact, about 2% of all the farms in the world are larger than 20Ha (or 50 acres). That is a lot of sizeable farms, 11.4 Million to be exact. These are farms that immediately could benefit from UAV operations, given that proper analysis tools and statistical models are available.
From an UAV technology perspective, there is a view of agriculture as a homogeneous marketplace. But treating it as such is futile, since there are no generic solutions to be found. By homogeneous we mean that the UAV industry tend to treat a variety of crops the same way, ignoring the vast differences between wide acre crops, vegetables, grapes or even fruits. The reality being that their practices differ on many levels. UAV technology providers would do well to consider the complexities and build statistical models for particular types of crops.
Crop specificity, and other physical dependencies are however only some of many aspects needed to consider, you also have business related parameters like e.g.
- Economic models
Being able to find ROI in the utilization of a UAV is dependent on many factors, and trying to appeal to a wide selection of different type of farmers’ crops and business setups is a matter of choice for manufacturers and service providers, but these distinctions are important to recognize. Talking about how UAVs can and will impact precision agriculture while also ensuring that the conversation doesn’t become too general or too specific is essential.
Some of the benefits of Agriculture UAVs are:
- Increased Yields
- Time and Cost Savings
- Integrated GIS Mapping
- Crop Health Imaging
- Environmental benefits
With use cases such as:
- Crop health monitoring: The ability to inspect in-progress crops from the air with Normalized Difference Vegetative Index (NDVI)
- Irrigation monitoring: Managing multiple irrigation pivots. Inspections of functionality of irrigation equipment delivering water
- Weed identification: Using NDVI sensor data and post-flight image processing to create a weed map
- Variable-rate fertility: NDVI maps to direct in-season fertilizer applications on crops. By using UAV-generated, Variable-Rate Application (VRA) maps to determine the strength of nutrient uptake within a single field
New technologies are revolutionizing the use of remote sensing in agriculture. The widespread availability of UAVs enables agricultural professionals to cost-effectively gather crop health information. UAVs have a unique advantage in being able to provide useful, high resolution, data from a range of sensors, such as optical, multispectral, NIR and thermal sensors to perform in-depth analysis of crop health and more.
Historically, researchers’ forays into employing UAVs for this end involved equipping them with thermal cameras to take measurements of canopy-air temperature above the crops. Canopy-air temperature being the basis of a widely-used index for water stress in plants.
Today we see established correlation between a number of sensor types and different agriculture parameters:
- Thermal, e.g.: Irrigation scheduling, plants disease detection, estimating fruit yield, and evaluating maturity of fruits
- Multispectral, e.g.: Identify pests, disease and weeds, provide data on soil fertility, count plants and determine population or spacing issues, estimate crop yield, and measure irrigation. NDRE (leaf chlorophyll content, plant vigor, stress detection, fertilizer demand, Nitrogen uptake)
- RGB + NIR, e.g.: NDVI, plant health, Nitrogen mineralization (measured as plant Nitrogen uptake)
Through the use of specialized software and statistical analysis it will be possible to draw conclusions from sensory data, but as agriculture is not a homogenous operation there will be need for significant statistical models covering combinations of different soil types, water compositions, crop types, climate zones and seasons.
Limitations of precision agriculture utilizing UAVs
Precision agriculture utilizing UAVs has its benefits but is today mostly used on a small scale. Under current regulations, which are more or less the same worldwide, all measurements need to take place within Visual Line Of Sight (VLOS) of the operator. The problem being that many fields and farms are bigger than VLOS. Operators could of course conduct several operations in a day by moving from section to section and stitching the results together into larger data maps for large farms, but this is not as effective as doing all at once.
Finding out if major technology providers have the expertise to offer products and services that are specifically related to a particular crop is a separate issue. It can be difficult, as these services and tools cannot become so specific that they’ll end up being completely unique for every single grower, and conversely, not so generic that the solution is no longer able to provide specific data.
From the ability to collect data from, and analyze, individual plants, to getting information on the moisture content of soil, the industry has seen how UAVs can deliver ROI for both growers and agronomists alike, and being able to see a financial return on an investment in UAS technology is a major consideration for many growers.
ROI really doesn’t become attractive until you start thinking about the unmanned systems as just a part of a solution. It takes an investment of time from the growers to see how the sensory data gathered from a UAV relate to what is happening on their fields. Growers need that data to provide context and create a correlation, and then identify cause and effect leading to informed decision making. This has led to growers just scratching the surface around how the technology can be utilized as they try and come up with value propositions that make financial sense.
ROI can be realized with drones as part of a solution, but anyone who plans to solely rely on drones to make decisions is likely to be disappointed. That said, as part of an ecosystem, UAVs can provide key pieces of information which will directly influence decision making.
ROI as a primary concern for growers is directly related to how UAVs are being integrated. As all of the variables are being considered, does it make sense for a grower to seek out a service provider or to try sort out these logistics for themselves? Initially, the easiest and lowest risk for the grower and provider themselves is probably to enter through a service.
The market potential for drones in precision agriculture needs more vetting. Despite the ROI studies by e.g. the American Farm Bureau Federation and Measure, it’s not yet clear how a UAS can deliver more usable data to a farmer or provide a cost benefit over the existing manned aircraft or the satellite image solutions available to them today.
As long as UAV technology remains a customized solution it will likely not get mass adoption. It has to become more generic. So there’s some work to be done. Once growers see the benefits, and the entire community is able to take advantage of them, that’s when things get really exciting.
Figuring out the ways UAVs can impact precision agriculture, and how to solve the multivariate nature of agriculture data analysis is essential, and asking questions around how growers can and should approach doing so is more important and relevant than ever.
That said, conditions are changing and continues to evolve at an ever faster pace. One may conclude that we are now probably right at the beginning of what in time might be called an Agro-drone revolution. All the early adopters are contributing to increased understanding of the multivariate complexity and a build-up of the needed statistical models. Sensors are evolving and regulatory conditions are being put in place enabling wider use, the latter which also continues to progress towards the full integration of UAVs into existing aviation systems.
Source: Roger Ohlund, SmartPlanes