Thousands of years ago, agriculture began as a highly site-specific activity. The first farmers were gardeners who nurtured individual plants, and they sought out the microclimates and patches of soil that favored those plants. But as farmers acquired scientific knowledge and mechanical expertise, they enlarged their plots, using standardized approaches—plowing the soil, spreading animal manure as fertilizer, rotating the crops from year to year—to boost crop yields. Over the years, they developed better methods of preparing the soil and protecting plants from insects and, eventually, machines to reduce the labor required. Starting in the nineteenth century, scientists invented chemical pesticides and used newly discovered genetic principles to select for more productive plants. Even though these methods maximized overall productivity, they led some areas within fields to underperform. Nonetheless, yields rose to once-unimaginable levels: for some crops, they increased tenfold from the nineteenth century to the present.
Today, however, the trend toward ever more uniform practices is starting to reverse, thanks to what is known as “precision agriculture.” Taking advantage of information technology, farmers can now collect precise data about their fields and use that knowledge to customize how they cultivate each square foot.
One effect is on yields: precision agriculture allows farmers to extract as much value as possible from every seed. That should help feed a global population that the UN projects will reach 9.6 billion by 2050. Precision agriculture also holds the promise of minimizing the environmental impact of farming, since it reduces waste and uses less energy. And its effects extend well beyond the production of annual crops such as wheat and corn, with the potential to revolutionize the way humans monitor and manage vineyards, orchards, livestock, and forests. Someday, it could even allow farmers to depend on robots to evaluate, fertilize, and water each individual plant—thus eliminating the drudgery that has characterized agriculture since its invention.
ACRE BY ACRE
The U.S. government laid the original foundations for precision agriculture in 1983, when it announced the opening up of the Global Positioning System (GPS), a satellite-based navigation program developed by the U.S. military, for civilian use. Soon after, companies began developing what is known as “variable rate technology,” which allows farmers to apply fertilizers at different rates throughout a field. After measuring and mapping such characteristics as acidity level and phosphorous and potassium content, farmers match the quantity of fertilizer to the need. For the most part, even today, fields are tested manually, with individual farmers or employees collecting samples at predetermined points, packing the samples into bags, and sending them to a lab for analysis. Then, an agronomist creates a corresponding map of recommended fertilizers for each area designed to optimize production. After that, a GPS-linked fertilizer spreader applies the selected amount of nutrients in each location.
Over 60 percent of U.S. agricultural-input dealers offer some kind of variable-rate-technology services, but data from the U.S. Department of Agriculture indicate that in spite of years of subsidies and educational efforts, less than 20 percent of corn acreage is managed using the technology. At the moment, a key constraint is economic. Because manual soil testing is expensive, the farmers and agribusinesses that do use variable rate technology tend to employ sparse sampling strategies. Most farmers in the United States, for example, collect one sample for every two and a half acres; in Brazil, the figure is often just one sample for every 12 and a half acres. The problem, however, is that soil can often vary greatly within a single acre, and agricultural scientists agree that several tests per acre are often required to capture the differences. In other words, because of the high cost of gathering soil information, farmers are leaving productivity gains on the table in some areas of the field and overapplying fertilizer and other inputs in others.
Researchers are beginning to tackle the problem, developing cheap sensors that could allow farmers to increase their sampling density. For example, one new acidity sensor plunges an electrode into the soil every few feet to take a reading and records the GPS coordinates; manually sampling on that scale would be far too costly. Such sensors have not yet arrived at most farms, however. Some haven’t proved reliable enough, breaking after a few acres of use, whereas others aren’t accurate enough. But several research groups around the world are working on developing sturdier ones.
More practical are sensors that look at the color of plants to determine their nutritional needs. Plants with too little nitrogen, for example, tend to turn pale green or yellow, whereas those with enough appear dark green. Several U.S. and European companies have developed sensors that detect greenness, generating measurements that can be used to generate a map recommending various amounts of nitrogen to be applied later. Alternatively, the measurements can be linked directly to the nitrogen applicator to change the application rate on the go. A tractor may have a sensor mounted on the front and an applicator on the back; by the time the applicator reaches a point that the sensor has just passed, an algorithm has converted the readings into settings for how much fertilizer to apply. Because research in this area has focused mainly on small grains, such as wheat, barley, rye, and oats, the technology is mostly limited to the parts of the United States and Europe that grow those crops. According to a 2013 survey by Purdue University, only seven percent of agricultural-input dealers offer plant-color sensors. Given the number of start-ups in this area, however, it is clear that many investors see the technology as a potential gold mine.
FIELDS AND YIELDS
The government’s GPS decision also enabled another revolutionary technology to emerge: yield monitoring. Most harvesters in the United States and Europe are outfitted with special sensors that measure the flow rate of grain coming in. An algorithm specific to the crop then converts the resulting data into a commonly used volume or weight, such as bushels per acre or kilograms per hectare. That information is then turned into colorful maps that show the variation within fields.
(Source – https://www.foreignaffairs.com/articles/united-states/2015-04-20/precision-agriculture-revolution#cid=soc-twitter-at-essay-the_precision_agriculture_revo-000000)