Prevailing Myths About Agricultural Abandonment and Forest Regrowth in the United States

Abstract

The classic story of historical land-cover change in the United States suggests that agricultural clearing in the 1800s was followed by agricultural abandonment at the turn of the twentieth century and subsequent forest regrowth—often referred to as a forest transition. Most descriptions present statistical data from historical censuses and surveys to make this case. Here we show that the historical data on cropland and forest area change for the United States need to be interpreted with care. Some earlier studies have exaggerated the extent of cropland abandonment and forest regrowth by failing to account for changes in definitions of croplands over time and changes in political boundaries in the case of forests. We reexamined the historical data to find that cropland and forest area for the United States as a whole have not undergone large-scale abandonment and regrowth but rather stabilized around the mid-twentieth century. Moreover, we find that, consistent with local and regional case studies, croplands were indeed abandoned in the eastern portions of the continent accompanied by forest regrowth, but there was compensating cropland expansion and forest clearing in the west. Our study suggests the need to exercise caution when using historical data to understand land-cover change and for developing theories such as forest transition. [Supplemental material is available for this article. Go to the publisher's online edition of Annals of the Association of American Geographers for the following free supplemental resource: (1) a table of cropland harvested area for the states of the United States from 1879 to 2002.]

El clásico recuento de los cambios históricos en la cobertura del suelo en los Estados Unidos sugiere que al desbrozo agrÍcola ocurrido durante el siglo XIX le siguió a la vuelta del XX el abandono de las tierras de cultivo y el subsiguiente recrecimiento del bosque—proceso al que a menudo se le conoce como transición forestal. La mayoría de las descripciones presentan datos estadísticos de censos y levantamientos de campo históricos en soporte de sus puntos de vista. En nuestro estudio mostramos que los datos históricos sobre el cambio de área de tierras cultivadas y bosques para los Estados Unidos deben interpretarse con cuidado. Algunos de los estudios anteriores han exagerado la extensión de campos de cultivo abandonados y de los bosques que las reemplazaron, al no tener en cuenta cambios en la propia definición de tierra de cultivo y las modificaciones que ocurrieron en los límites políticos en lo que se refiere a bosques a través del tiempo. Examinamos los datos históricos para así establecer que el área de tierras cultivadas y bosques de los Estados Unidos, en conjunto, no ha experimentado abandono en gran escala y recrecimiento forestal, sino que por el contrario se estabilizó a mediados del siglo XX. Aun más, encontramos que, consistente con estudios de casos locales y regionales, tierras cultivadas de las regiones orientales del continente en verdad fueron abandonadas y acompañadas de recrecimiento del bosque, pero que, en compensación, ocurrió una expansión del área cultivada y reducción de los bosques en el oeste. Nuestro estudio sugiere obrar con cautela cuando se utilizan datos históricos para entender los cambios de la cubierta del suelo y para desarrollar teorías como la de la transición forestal. [Hay disponible material suplementario para este artículo. Acceder a la edición online del publicista de Annals of the Association of American Geographers para el siguiente recurso suplementario gratuito: (1) una tabla del área de las tierras de cultivos cosechados en los estados de EE.UU. de 1879 a 2002.].

Key Words:

• farmland abandonment
• forest transition
• historical data
• land-cover change
• United States

关键词:

• 农田废弃
• 森林过渡
• 历史数据
• 土地覆盖变化
• 美国

Palabras clave:

• abandono de tierras de cultivo
• transición forestal
• datos históricos
• cambio de la cobertura de la tierra
• Estados Unidos

Acknowledgments

We would like to thank Lisa M. Curran, Jonathan A. Foley, and Billie L. Turner II for useful feedback on this article.

Notes

1. Historical data on population were available consistently back to 1800. Agricultural data were not available until 1880, so Clawson used the relationship between population and agricultural data from 1880 to 1975 to estimate cropland area between 1800 and 1880.

2. This volume presented data going back to 1880; it is not clear why Clawson did not use the 1880 figure but rather estimated the value.

3. Note that often the census report was issued one year after the data were collected. For example, the 1880 census reported data on agriculture in 1879. HisStatUS (and Clawson) assigned the values to the year of the report, whereas we assigned the values to the year of data collection in this article.

4. Note that cropland harvested denotes the area of land devoted to harvested crops; if a piece of land is multiple cropped, it is not counted more than once. Therefore, our data until 1920 are not entirely consistent because we simply added the harvested area of individual crops to get cropland harvested. However, multiple cropping was likely nonexistent in the early 1900s; even today, only about 4 percent of cropland harvested is multiple cropped (U.S. Department of Agriculture 2008).

5. Cultivated summer fallow.

6. Williams cited his data sources as Clawson (1979) and U.S. Forest Service (1982). It is clear that the legacy of Clawson's first estimates have been reproduced since in other studies.

7. Footnotes to the cropland data in HisStatUS are as follows. Footnote 5: “Cropland harvested only” for the data in years 1880 to 1920; Footnote 4: “Includes Alaska and Hawaii” for data in years 1930, 1940, and 1950; and Footnote*: “Except as indicated by footnote 4, denotes first year for which figures include Alaska and Hawaii.” The exclusion of Alaska and Hawaii in some earlier reports, although noteworthy, is insignificant because they contain only 7 percent of the total cropland in the United States currently, and contained less than that in the past.

8. According to a table footnote from the 1945 census, “The 1940 figures are not strictly comparable with those for 1945. The 1945 figures include land used only for pasture, which has been plowed within 7 years. The 1940 figures include land pastured, which could have been plowed and used for crops without additional clearing, drainage, or irrigation. This land may not have been plowed within 7 years prior to 1940” (U.S. Census 1947, as cited in Waisanen and Bliss 2002, 84–4).

9. Indeed, a USDA report suggests that there is an inverse relationship between cropland used for crops and cropland idled because of federal programs (USDA 2008).