The U.S. Department of Agriculture (USDA) measures two kinds of soil erosion: wind erosion, and what is called "sheet and rill" erosion. Wind erosion is self-explanatory; everyone recalls images of the "dust bowl" during the 1930s, when high winds blew away tons of drought-parched topsoil in the heartland of the nation. Wind erosion is prevalent in the arid western states that have dryer soil and less natural ground cover, while many eastern and southern states experience no measurable wind erosion at all. Sheet erosion is the removal of thin layers of soil over the whole surface chiefly through raindrop splash and surface water flow. Rills are channels small enough to be obliterated by normal tillage operations.
The USDA measures soil erosion every five years as a part of its National Resources Inventory (NRI). The USDA released the most recent data in November 1999. They show that wind erosion, which had been increasing during the late 1970s and 1980s, has declined in the 1990s (see Chart 21).
The NRI findings show that sheet and rill erosion has experienced a consistent decline at a rate of slightly more than two percent per year, or about 40 million tons per year (see Chart 22).
Soil erosion leads to a decline in agricultural productivity, and the sediment from erosion contributes to water degradation. The EPA regards soil erosion as one of the most important national environmental problems, though experts disagree about the severity of the issue. The rate and severity of erosion depends on local conditions and soil types. Hence it is difficult to make national generalizations. The title of a 1987 report from the U.S. Department of Agriculture is instructive: Soil Erosion: Dramatic in Places, But Not a Serious Threat to Productivity.48
Government policy aims to reduce soil erosion to "tolerable levels" (known as T-values) by 2010, and to zero by 2025. T-values from cropland are designed to be the maximum amount of erosion that will indefinitely support agricultural productivity. T-values range from one to five tons per acre per year, depending on soil type and location. The good news is that the NRI figures show soil erosion is occurring at the low end of the T-values on two thirds of the cropland acreage in the United States. Only about one percent of cropland is experiencing a high rate of soil erosion.
There is no uniform natural rate of soil erosion, and one leading textbook argues that "there is no solid basis for tolerance soil loss values."49 The normal rate of soil erosion under natural vegetation is thought to be in approximate equilibrium with the rate of soil formation. It is helpful to consider some numbers. One ton of soil per acre (the low end of the T-values) is equal to uniform depth of .007 inches (.18 mm). At one ton lost per acre per year, it would take 43 years to lose an inch of topsoil. This is approximately the rate of erosion on uncultivated cropland. Even this rate, however, may be faster than the rate at which new soil is formed. Readers should not conclude, however, that U.S. farmland is in any serious danger.
The USDA itself, in a 1994 report, said that "loss of farmland poses no threat to U.S. food and fiber production."50 The chief reason to be concerned about soil erosion is that it is a factor in "non-point" water pollution, i.e., soil erosion carries fertilizer and pesticides, as well as dirt, into streams and rivers.
Charts 23 and 24 show that the largest gains in reducing soil erosion have come on cultivated cropland, the result of educational efforts and improved technology in farming practices.
Charts 25 and 26 display average soil erosion findings for Michigan. Michigan has been below the national average for both kinds of erosion on cultivated cropland.