When there\u2019s too much rain, or not enough, or another calamity strikes, US farmers and ranchers rely on insurance from the Federal Crop Insurance Corporation to provide a safety net. In 2023, crop insurance covered more than $207\u00a0billion in liability.<\/p>"},{"acf_fc_layout":"image","image":714292,"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"
Although <\/span>most farmers file claims for justified losses, occasionally fraudulent claims are filed. Investigators with the US Department of Agriculture\u2019s (USDA) Risk Management Agency (RMA) root out these schemes using modern crop monitoring. This integration of advanced technologies and data science techniques supports the investigative process.<\/p>\r\n The application of high-resolution aerial imagery to capture field conditions, machine learning algorithms to automate pattern detection, and GIS generates positive results. For example, RMA caught Colorado ranchers who tampered with rain gauges; Kentucky tobacco farmers who falsely reported hail damage; and a North Carolina farmer who produced crops, sold them under the names of other farmers, and claimed those crops were lost to natural disasters. Although it\u2019s only a small number of farmers and ranchers who look for ways to game the system, the financial savings of rejected claims adds up.<\/p>\r\n \u201cWe\u2019ve been able to document a cost avoidance through our Spot Check List program that amounts to more than $1.75 billion over the past 20-plus years,\u201d said Jim Hipple, a physical scientist in the Business Analytics Division of USDA RMA. \u201cCost avoidance is even better than cost recovery because we haven\u2019t paid anything out, so we don\u2019t have that added burden of trying to pull money back.\u201d<\/p>\r\n The work by Hipple and others at RMA also helps instill trust in the crop insurance system, which has an important role buffering farmers from major losses due to drought, excessive rain, hail, wind, frost, insects, and disease.<\/p>\r\n\r\n \u201cCrop insurance policies have gotten more specific about the field location,\u201d Hipple said. \u201cThat helped us better understand conditions on each farm.\u201d<\/span><\/p>\r\n The USDA Farm Service Agency, a sister agency to RMA, mapped the location of every field to the common land unit (CLU) level. [A CLU is an individual contiguous farming parcel.]<\/i> To accomplish this, more than 2,500 field service centers across the country were equipped with GIS.<\/p>\r\n At the start of a growing season, farmers report their planting intentions through acreage reports. Field boundaries from these reports are compiled into a database. Over nearly a decade, more than 36 million CLU boundaries have been recorded along with the associated land ownership, soil, and crop type.<\/p>\r\n These digital records, which replace paper maps, can be easily updated and analyzed to visualize agricultural trends. They let investigators ask location questions related to claims and speed processing of insurance payments after disaster strikes. Having digital records at the CLU level was a big improvement in geospatial awareness, but it required more computational power.<\/p>\r\n\r\n To handle big data processing at scale, the USDA works with the Center for Agribusiness Excellence at Tarleton State University in Texas. \u201cWe leverage the advanced analytics from the university effort to better understand the integrity of a policy, and to seek out waste, fraud, and abuse,\u201d Hipple said.<\/p>\r\n By adding tabular data to the map, crop insurance compliance investigators can spot patterns and irregularities that indicate potential insurance problems. The key, according to Troy Thorne, director of the Center for Agribusiness Excellence at Tarleton State University, is in identifying inefficiencies\u2014places where the connection between the land and what it produces seems odd or unusual.<\/p>\r\n Thorne cited the practice of yield switching as an example. Crop insurance is based on a field's yield history. If a field has produced the same crop with the same farming practices for three years, insurers average the output to determine an approved yield history. That figure becomes the baseline for insurance claims.<\/p>\r\n To raise the baseline, a farmer might record the accurate overall yield total for all fields but move the numbers around to inflate one field\u2019s total, thus raising that field\u2019s yield history. When that field produces a normal yield the next year, it will appear to have underperformed, providing the basis for a potential insurance claim.<\/p>\r\n \u201cYield switching is a big deal,\u201d Thorne said. \u201cYou improve the outcome of your insurance claim without actually suffering the losses.\u201d<\/p>\r\n The ability to see all related data on a map rather than in a tabular format has helped analysts and investigators find incidents of yield switching and other anomalies.<\/p>\r\n \u201cAs a tabular perspective, it kind of gets lost in the detail,\u201d Thorne said. \u201cBut when you add the geospatial layer and drop these things on a map, you can look at historical yields, and see that the farmer\u2019s yields are constantly fluctuating.\u201d<\/p>\r\n <\/p>"},{"acf_fc_layout":"image","image":714282,"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"sidebar","layout":"standard","image_reference":null,"image_reference_figure":"","spotlight_image":null,"section_title":"","spotlight_name":"","position":"Center","content":" The Tarleton Analytics Institute, part of the Center for Agribusiness Excellence at Tarleton State University in Texas, is one of two contractors RMA employs to examine crop insurance claims. Tarleton\u2019s expertise in data science has helped RMA save millions of dollars each year.<\/span><\/p>\r\n The center\u2019s director, Troy Thorne, has embraced GIS for its ability to \u201cpinpoint areas on a map and examine different layers to not only understand our own results but to identify what\u2019s being impacted, why it\u2019s important, and to get the message across.\u201d<\/span><\/p>\r\n Tarleton has been helping migrate RMA\u2019s systems to the cloud to meet the federal government\u2019s cloud-first pledge. At RMA, cloud-based processing is listed as part of the agency\u2019s Climate Adaptation Plan to reduce the risk of disaster events damaging its systems. Distributed computing is just one adaptation strategy the agency employs to harden its infrastructure.<\/p>\r\n For farmers, Thorne sees a need for more data-driven decision-making. \u201cHaving geospatial and weather data at your fingertips will help farmers make better decisions,\u201d he said. \u201cAs we continue to see our farmland diminish, it\u2019s imperative that we help farmers address these challenges.\u201d USDA recently updated the federal crop insurance program to include conservation and climate-smart activities such as good farming practices. These practices promote the conservation of soil, water, air, animals, and energy resources.<\/p>","snippet":""}],"references":null},"yoast_head":"\nThe Rise of Field-Level Awareness<\/h3>\r\n
A Data Science Partner<\/h3>\r\n
Climate Adaptation Strategy<\/span><\/h2>\r\n