Balaji Sethuramasamyraja 2017-06-28 00:33:46
Precision agriculture is all about managing agricultural inputs to optimize yield while protecting the environment through site-specific crop management— in other words, profitability through sustainability. Precision agriculture involves significant applications of technology in day-to-day agricultural operations, including autosteer and autonomous machinery, on-the-go soil and yield mapping, differential harvesting, variable-rate fertilizer and pesticide treatments, precision irrigation, and unmanned aerial systems. The adoption of precision agriculture has varied across the agricultural regions of the country due to the different cropping patterns and cultural practices of each region. For example, adoption of precision agriculture in California’s Central Valley has been slow, but steady. Some of the reasons for the slow adoption are the diversity of crops, limited farm acreage, and specialty farm equipment. My professional journey has taken me from large-scale soybean and corn production in the Midwest to a faculty position at California State University’s Fresno campus, which serves one of the world’s most diverse agricultural regions. The CSU Fresno campus has a 405 ha (1,000 acre) farm, called the University Agricultural Laboratory (UAL), that’s modeled on an enterprise management system. Students in CSU Fresno’s Jordan College of Agricultural Sciences and Technology gain valuable experience through research and operations in the various UAL enterprises. The products of these enterprises, including fresh produce, meat, milk, wine, and others, are sold in the on-campus farm market and at retail locations around the city of Fresno. Differential harvesting of wine grapes One of the highlights of precision agriculture research at UAL has been the development of automated differential harvesting for wine grapes. This process uses near-infrared (NIR) spectrometers as sensors to determine the fruit quality in the field based on the fruit’s anthocyanin content. This research was widely reported in the global ag media because it resulted in the first-ever differential harvest of wine grapes based on fruit quality. The work was performed in collaboration with Constellation Wines and Oxbo Korvan International. Differential harvesting of wine grapes involves three steps: sensing the anthocyanin content in the fruit, using the data to produce a quality map based on a threshold anthocyanin level, and then using the quality map to direct the harvester. Using the NIR sensor data, a quality map of the vineyard was generated showing two zones (called quality A and quality B) based on an anthocyanin threshold value of 0.87 mg g-1 of fruit, as specified by wine makers. The quality map was then uploaded into the harvester’s InSight display (a product of Ag Leader Technology, Ames, Iowa) as a shape file to control the differential harvesting attachment, which was installed on a standard commercial wine grape harvester. Using the two-zone quality map of the vineyard, wine grapes that were identified as quality A were loaded into one of the two gondolas, and grapes that were identified as quality B were loaded into the other gondola. Sugar content (measured in °Brix) was used to determine the timing of the harvest. To verify the difference in fruit quality produced by differential harvesting, three 40-ton lots of harvested fruit— 40 tons each of quality A and quality B (which were differentially harvested) and 40 tons of control grapes (which were harvested as a normal load of wine grapes)—were sampled for analysis. The three lots were then fermented separately and subjected to further analysis. A wine tasting was conducted to determine the differences in aroma, flavor, and mouth feel between wines made from quality A and quality B (low versus high anthocyanin content) and control samples of Cabernet Sauvignon grapes harvested at the nearby Twin Creeks and Merjan vineyards. The tasting panel, which included 20 panelists and duo-trio tests (i.e., replicated five times) of two different test groups, found significant differences (99.4% confidence) in the taste and mouth feel between quality A and quality B. Initially, we needed to sample the vineyards by hand to delineate them into two zones (quality A and quality B). In our earlier study, we used an experimental plot and sampled all over the field. However, in real-world situations, manual sampling is expensive, and high-density data collection is not always feasible as in experimental studies. To find a solution to this challenge, we have developed geo-statistical tools with Geographic Information Systems (GIS) modeling to determine an optimum sampling protocol that can reveal the spatial variability of wine grapes and yet requires a minimum number of sampling points. Yield monitoring of almond harvest During the almond harvest season, the local farmers get quite excited as they ready their machinery for harvest operations. However, as soon as the harvest is over, they move on to the next task without taking time to see what actually happened during the harvest. To monitor the almond harvest more closely, we worked with industry to develop a sensor-based system for measuring the harvest yield across an almond orchard. Almonds are typically harvested using a three-step process: shaking, sweeping, and picking. Prior to harvest, the orchard is cleared to remove debris, such as twigs and fallen leaves. A mechanical fork shaking system shakes the almond trees at the trunks with a displacement of 10 mm at 20 Hz. Mature almonds fall to the ground during the shaking process and are raked into windrows using an almond sweeper. The almonds are then sun-dried on the ground to remove excess moisture. After two weeks of drying, an almond yield picker is used to gather the dried almonds for further processing. Yield monitoring systems have been widely adopted for other crops, including grains, fruits, vegetables, and other varieties of nuts, but none have been commercially available for almonds. Our almond yield monitoring system uses a mechanical load cell mounted on the axle of a yield implement, a GPS receiver for geo-referencing each yield data point, a controller, and a user interface with data storage. The developed system successfully collected yield data during almond harvests at the UAL and at other experimental farms in Fresno County. ASABE member and California/Nevada Section District 4 Chair Balaji Sethuramasamyraja, Associate Professor and Program Director of Agricultural Systems Management and Precision Agriculture Technology, Department of Industrial Technology, Jordan College of Agricultural Sciences and Technology, California State University, Fresno, USA, email@example.com.
Published by ASABE. View All Articles.