A Bird's Eye View of Farm Management: Drones in Agriculture

Drones have many applications, are considered by many fields and disciplines to be incredibly useful tools. In my first post, I provided an overview of many issues associated with drones. In closing my first post I hinted I would author future posts, offering some current or potential uses. Some of the applications I identify are currently uses, with potential for growth; other applications are more future-looking, not in place today, but perhaps in the not-so-distant future.

In my second post, I discussed another field where drone use is opening doors to new exploration techniques. Archaeology is finding many uses for a remotely-piloted aircraft, or unmanned aerial systems (UAS). Kites are being equipped with digital cameras. Balloon kits allow researches to hoist digital cameras aloft. More detailed archaeological surveys mandate stability unattainable from kites and balloons. The survey of archaeological sites must be systematic. Flight lines need to be planned in advance of a survey to ensure the capture of the area of interest (AOI). Scale of acquired imagery needs to be established. Scale and resolution of acquired imagery must be considered before flying a drone. When the details of the mission have been thoroughly accounted, then imagery is acquired. With regards to UAS, several missions might be flown. The first mission might result in imagery, visually inspected, to help plan the geographic scope of future flights. Sometimes, interesting landforms which imply a cultural imprint can only be seen from above, which might impact the scale and scope of later drone missions. In Europe, especially in the U.K., France, and Eastern Europe, drone use has improved the knowledge of culture imprints on the landscape, aided in the inventory of archaeological sites, aided in the detection, analysis, and interpretation of archaeological sites.

I ask students in my all of my courses what color they see the most when they are outside in the late spring or early summer. Sometimes, they need some prompting; "What is the color of the vegetation you see when you go outside?" Usually, the answer is "green," which is sort of wrong. I'll explain why.

From high school physics, we learned about the Electromagnetic spectrum (EM). The EM spectrum is the range of energies emanating from the Sun and other objects. Our eyes, great as they are, are sensitive to a mere fragment of the entire spectrum of possible energies. The rods and cones in our eyes are stimulated by energies associated with colors ranging from violet to red. A simply mnemonic to remember the colors is ROY G BIV (red, orange, yellow, green, blue, indigo, violet). The arrangement is from longest wavelength to shortest wavelength.

Our eyes are not sensitive to other energies, like radio waves, cosmic waves, x-rays, microwaves, or even energies just beyond the normal range, infra-red, or ultra-violet. Our natural environment, however, is very adept at reflecting energies beyond the range of our eyes to see.

At left, is a graph illustrating the reflectance of energy from the leaves of healthy vegetation, stressed vegetation, and severely stressed vegetation. "Stressed" means the vegetation has been affected by something, lack of water, lack of nutrients, presence of insects, or some form of blight.

Note the range of reflectance energies along the y-axis, and the range of reflectance wavelengths along the x-axis. Along the x-axis, we see the "visible" range, the range our eyes react to, the "near infra-red" range which we cannot see, and the short wave infra-red, which corresponds to thermal (heat) related energy.

Along the y-axis the reflectance values are indicated. The greater the reflected energy, the higher the reflectance value. Now, note the "visible" range (lower-left). What I mean by saying "green" is sort of a wrong answer is vegetation is obviously reflected far more energy in the near IR and the short wave IR spectrum. I didn't say "green" was "completely wrong," and clearly the reflectance of energy from vegetation is more complex than most people imagine.

To the left is a graphic illustrating what happens when incoming solar radiation strikes the surface of a leaf. The red, green, and blue wavelengths of energy pass through the leaf epidermis. Red and blue are absorbed, while the green wavelength is reflected back by the mesophyll and reaches our eye, and we see the leaf as being "green" as the cones in our eyes are stimulated by color.

When the structure of the leaf changes, the reflectance of the leaf changes. If the leaf lacks water, the epidermis will change color. When the leaf changes "color," we have to ask ourselves, "what is really happening?" What is really happening is the structural change in the leaf is modifying the absorption and reflectance of energy. These changes then manifest as color changes.

Often, when our human eyes detect stress, the plant has already experienced considerable stress. The stress has only just reached the threshold of our eyes to detect. The lack of water, or the presence of too much water, or the blight, acted on the plant well before the ability of our eyes to detect the problem.

One comment before I continue. The "dips" in reflectance values correspond to moisture absorption. Water does not reflect energy well. In fact, one way to find water on remotely-sensed imagery is to look for dark areas in imagery, places where infra-red energy is absorbed. These "dips" are called "water absorption bands."

Knowing something about the EM spectrum AND knowing how different vegetation types reflect energy can lead us toward discovering characteristics about vegetation. We can build sensors to measure reflectance energies in wavelengths we cannot see. These sensors capture and store data, data which can be input into software packages for analysis and interpretation.

Funny, as I write this post, Chris Anderson gave a talk at the San Francisco Bay area MakerFaire a short while ago. The title of his talk, "Farm Drones, Feed the World, Save the Environment, Try Hard To Not Become Self-Aware," gives away the  premise of my post.

The narrative I created earlier was to explain some of the basic theories and technologies. For decades, since the early 1970s, NASA and the USDA have spent hundreds of thousands of man-hours developing technologies for the study of agricultural areas. Satellites LandSAT 1-7 were placed in orbit specifically to capture large amounts of land cover / land cover imagery to support USDA efforts to measure amounts of certain crops. Alongside the efforts to study and measure crops were efforts to analyze and study forest cover. To many, the forest is a crop, silvaculture. In the southern portion of the United States, and the U.S. Northwest, timber is an important industry.

Not only has Landsat been used to help generate data on U.S agricultural products but on global agricultural production. Landsat data measures acreages of crops in other countries. The USDA can estimate crops in the field, acreages, and yields for every country on Earth. Knowing global crop data can help in planning global food production, and also assist in crop pricing.

Satellite imagery is a brilliant data source for large tracts of land. However, if the area of interest is small, such a single farm, the resolution of some satellite imagery is too crude, too course, for proper analysis. For medium- to large-scale farms, drones represent a new technology for farmers and ranchers for farm management.

Drones outfitted with appropriate technology could assist farm managers in the collection of a multitude of different data. For instance, a drone equipped with airborne sensors capable of sensing the near and short infra-red could detect vegetation in various states and types of distress. Further research could determine the type of stress, whether the stress is the result of water, insects, or some form of disease. Data collected by the drone could then be used to direct spraying, or watering.

Drones could assist in the determination of soil chemistry. Just as plants reflect energy which can be used to determine plant health, soils reflect solar radiation which can be used to determine the soil chemistry and moisture. Data could then be used for the direct application of fertilizer.

Drones could even be used in ranching. While I have not heard of such technology in use, yet, the technology exists today to make the following possible. Today, cattle have ear tags. Some ear tags are radio-frequency identification tags (RFID). I can envision a day not too far away where a farm drone is sent aloft to "count" cattle, or to locate lost cattle, or simply map the distribution of cattle on a ranch. If RFID tags are sophisticated enough, cattle from different herds could be distinguished on a map. Even more sophisticated, what if the RFID telemetry was connected to a database table. Perhaps a software app is developed where a farm manager could watch the flight of drone. The head of cattle in view would appear in the viewer with "tags" over their heads tied to the animal's record in the database table. A mouse-over of the tag would fire a database query resulting in a pop-up window illustrating the animal's data, owner, herd number, age, species, etc.

Flight software for mission planning already exists. ArduPilot (left) is perhaps the best known flight mission planning software for the do-it-yourself drone makers. The drone, equipped with WIFI and GPS, would be capable of being programmed to fly in a specific pattern using a set of waypoints. The drone, equipped with an altitude sensor would also be able to fly at specific altitudes. Therefore, a suitably proficient farm manager, or employee, could use a drone to monitor and manage farm assets.

Visit "DIY Drones" for enthusiastic information on drones.

Drone technology for commercial use in only in its nascent form. The technology has yet to be tested, explored, and analyzed. Drone companies in Europe are paving the way to dominate the global commercial drone market. Drone use in the United States is facing an uphill battle. However, there are many fields and disciplines where drone use is fair, appropriate, educational, informative, and perhaps even necessary. I will explore some of these other fields in future posts.

Shout/Out to readers "Jo,"  "WalkStx," and BKraxberger for their attention.

PAX

Finding Old Stuff with New Technology: Drones in Archaeology

Drones have many applications, are considered by many fields and disciplines to be incredibly useful tools. In my first drone post, I provided an overview of many issues associated with drones. In closing my first post I hinted I would author future posts, offering some current or potential uses. Some of the applications I identify are currently using, with potential for growth; other applications are more future-looking, not in place today, but perhaps in the not-so-distant future.

Drones have been widely used throughout Europe beginning in 2009, perhaps earlier. Prior to the use of drones, manned aircraft were used for evaluating and cataloging archaeological sites. Some of these early imagery collection efforts date back to the 1930s, with the use of fixed-wing aircraft and photography. Some early efforts also used manned balloons and photography.

Early efforts to gather imagery of archaeological sites were costly and time-consuming. A plane had to be hired, a pilot acquired perhaps paid, along with associated fuel costs. Photographs taken might not be adjusted for problems with scale arising from the camera's look-angle, or influences of terrain. Storage of film materials becomes problematic over time, as the film can degrade over time and rendered useless.

Unmanned aerial systems (UASs) provide several well-documented enhancements over conventional airborne collection, efforts, e.g. piloted planes. UASs can be carried within a suitcase, for example. Models have HD front- and bottom-facing cameras. Some models fly for perhaps 12-15 minutes. Many of the smaller, light-weight models are based on hobby airframes, made of carbon composite materials. The power cells are rechargeable NiCd batteries, powering tiny electric motors. Some models are capable of flying for long durations, 30-60 minutes. These cost more than small universities and college can afford, though.

The Parrot Drone 2.0

Smaller UASs are able to carry user-installed USB drives. The USB drive stores HD video aboard the UAS, extracted after the drone returns to ground. Many of the UASs available today can be flown from a tablet or smartphone. 

One such UAS, the Parrot Drone (Parrot, Inc.) weighs about 4lbs, carries two HD 720p cameras, and can be flown from an Android tablet (Kindle Fire), an Apple iPad, or an Android/iOS-powered smartphone. Live streaming video is sent to the remote control via a secure WIFI connection. The Parrot captures HD video on a user installed USB drive, while pics can be captured during flight straight to the remote control device. The Parrot documentation suggests maximum flight distance from the pilot is about 50m (150ft). Flight time is about 12 minutes with a full charge.

Aurora Flight Systems SKATE UAS

Some U.S. schools have engaged in inter-disciplinary cooperation. Vanderbilt University, in Nashville, TN., has developed a UAS project based upon the Aurora Flight Systems SKATE platform. In 2012, Vanderbilt garnered some publicity for the use of the SKATE over ruins in Peru.

Archaeologists in the United States are well-behind their European counterparts in the use of aerial systems for mapping, analysis, cataloging, and overall research of North American archaeological sites. Since at least 2009, Italy, France, and the UK have all initiated both academic and national archaeological research projects specifically with UASs in mind.

The Mirkovo Basin research utilized the Quest Systems 200 UAV

In 2012, research sponsored by the Field Museum, Canada, and conducted by Durham University (UK) and the New Bulgarian University (Bulgaria), gathered aerial surveys of important archaeological sites in the Mirkovo Basin (Bulgaria). The unique landscape of the Mirkovo Basin provides over 8,000 years of human settlement history, covering the Neolithic to the modern age of human habitation. Aerial surveys of archaeological sites promises to provide improvements to landscape analysis, cataloging of current sites, with potential use of finding additional unknown sites.

UAS offers fantastic potential for archaeology. Currently, most of the less-expensive UASs are equipped with only HD cameras. More expensive UASs offer thermal imaging capabilities. Thermal imaging technology has the potential to reveal archaeological sites not visible to the human eye. Soil conditions can betray hidden paths, buried foundations, or hint subsurface structures. UASs can be equipped with infrared or multispectral cameras. The human eye, while a wonderful apparatus, is limited in the ability to see wavelengths only between 0.4µm (blue) to 0.7µm (red). Humans cannot see the ultraviolet or the infrared.

When we go outside, we see green trees, grasses, etc., due to the absorption of green electromagnetic (EM) radiation by the 'cone" structures in our eyes. However, vegetation reflects more energy in the infrared portion of the EM. When I tell people vegetation is a poor reflector of green, they look at me like I'm crazy. Chlorophyll response is significantly greater in the infrared than in green, but the human eye cannot see into the infrared. We engineer sensors to reveal the infrared to us, and, in doing so, staying with the domain of archaeology, the sensors can indicate traces of human disturbance. Plants, because they reflect infrared so well, betray changes in soil moisture and chemistry our human eyes are unable to detect. Landscape changes can indicate heretofore undiscovered sites, forts, homes, villages, or cemeteries.

Very expensive drones can be equipped with LiDAR. Light Detection and Ranging (LiDAR) uses lasers to map terrain very accurately. Subtle changes in terrain, when the elevation data are processed and mapped, can be used to analyze historical human landscapes. In some cases, both in Central America and in Southeast Asia, archaeologists have literally walked over archaeological sites without realizing. Only after mapping and surveying, either using ground-based or aerial techniques, did the researchers realize how they had underestimated the size of their study areas.

UASs have the potential to revolutionize archaeology. When coupled with ground survey information, using levels and transits - traditional survey tools, UAVs can provide unparalleled detail for archaeological studies. Geographic information systems, as well as other remote sensing methods, promise to transform the field of archaeology from the stereotypical Indiana Jones stomping around in the jungle, dodging poisonous snakes and toxic darts, to a technology-driven, cutting-edge 21st century digital discipline.

[Archeometry, Department of Geography; Ghent University. The link implies no endorsement. I simply provide as a sample of the peculiar field of archeometry and to illustrate by example the nature of technology use in archaeology.]