GLO Soil Classification General Land Office Research Department of Landscape Architecture Iowa State University
	For more information, see Frequently Asked Questions or Visit Paul Anderson's summary page

 GIS Descriptive Modeling of Soil Suitability for Agriculture in Polk County, Iowa

Comparing Modern and Historic Soil Classifications

Paul F. Anderson

Iowa State University

 

Polk County

GLO surveys of townships in what is now Polk County were completed by five deputy surveyors.  Surveys were completed during the period June 25 to November 19, 1847 (Union Historical Company 1880, p. 398).

The portion of T77N R22W north of the Des Moines River is included in Polk County.  However, it was not included in this modeling project.

Polk County was part of the "New Purchase," the Sauk and Fox Cession of 1842 (Lokken 1942, p. 15), also called the Royce Cession 262 (Swierenga 1968, p. 18).  However, because it was west of the "Temporary Indian Boundary 1843-1845" (Red Rock Line), this part of the cession was not available for settlers until midnight of October 11, 1845 (Pratt 1972, p. 61).  Polk County was established and incorporated on April 22, 1846 (Pratt 1972, p. 62).  Swierenga (1968, p. 231) reported that land in Polk County __??__ Franklin Township was first offered for sale in 1848.  According to Pratt (1974, p. 66), Franklin Township was not organized as a civil (political) township until March 6, 1854.

Because the Fort Des Moines Land Office was not established until 1852 (Swierenga 1968, p. 8) or 1853 (Pratt 1972, p. 65), the early land sales in Polk County were entered at the Iowa City land office (Union Historical Company 1880, p. 390).

The map of Polk County in the Andreas Atlas of Iowa (1875) shows the location of 307 early farmsteads, each with a point symbol and the owner's name.  Thirty-six farmsteads shown without a point symbol on the county map were not included in the modeling.

Soil ratings.  Deputy Surveyors Durham, Evans, Jacobs, and Moorehead surveyed a total of approximately 968.4 miles of subdivision section lines in Polk County.  Deputy Surveyors Higbee and Marsh surveyed another 243.1 miles of township section lines in Polk County.  Surveyor Marsh rated two miles 1st and 2nd rate, without specifying which parts were 1st rate and which parts were 2nd rate.  Soils in the county were rated predominantly 1st rate (56.9 percent) and 2nd rate (29.2 percent).  The surveyors did not list the soil rating for 6 percent of the miles surveyed.  Surveyors split 39 section lines, rating the soils in one portion differently than another.  This was typically done when one portion was bottomland and another portion upland.   Of the sections lines that bounded the first parcels sold, most (75.2 percent of the total length) were described by GLO surveyors as 1st rate and most of the remainder (20.3 percent) were 2nd rate.  For this descriptive modeling, the "first parcels sold" include those described above as the early land sales in Franklin Township (Union Historical Company 1880, p. 390).  "Early farmsteads" include those described above from the Illustrated Historical Atlas of the State of Iowa (Andreas 1875).  The distribution of soil ratings for section lines near early farmsteads showed less 1st rate (55.4 percent) than for first parcels sold (75.2 percent), but more 3rd rate (16.0 percent) than for first parcels sold (0.7 percent).

GLO vegetation types.  Deputy Surveyors mapped 13 vegetation types in Polk County.  For comparison with Franklin Township, data for prairie, timber, grove, and field are shown in the table below.  (Not included in the table are the nine vegetation types that comprised 0.4 percent of the county:  city, lake, marsh, pond, scattering trees, slough, swamp, thicket, and wetland.)  The area mapped as timber covered 20.1 percent of Polk County.  However, a slightly higher proportion (20.9 percent) of the section lines surveyors rated 1st rate soils intersected areas that they mapped as timber.  Therefore, section lines rated as 1st rate soils intersect a disproportionately (slightly) larger percentage of timber.  Conversely, section lines rated as 2nd rate and 3rd rate soils intersect disproportionately smaller percentages of timber.  Section lines without a soil rating ("not rated") intersect a disproportionately (slightly) larger percentage of timber (20.3 percent), similar to 1st rate soils (20.9 percent).  The first parcels sold in 1848 included a relatively small amount of prairie (40.3 percent), but a relatively large amount of timber (55.8 percent) and grove (0.7 percent).  Early farmsteads (1875 Andreas Atlas) had a disproportionately large percentage of prairie (82.7 percent) and a disproportionately small percentage of timber (14.7 percent).

Distance to nearest timber/grove, river, field.  To measure proximity of each soil rating class to timber, grove, river, and field, distances were measured from each section line midpoint to the nearest timber/grove polygon edge, nearest point along the river, and nearest field polygon edge.  The river location used in this analysis is not the current channel, but rather the one shown on the surveyors' township plat maps.  Because both timber and grove could provide shelter, building materials, and fuel, they were aggregated together for this analysis.  Midpoints of section lines rated 1st rate are closer to timber/grove polygons (mean distance of 949 meters) than those of 2nd rate soils or 3rd rate soils (1,347 and 2,547 meters, respectively).  The same pattern is evident in mean distances to the Des Moines, Raccoon, and Skunk Rivers.  Midpoints of section lines rated 1st rate are closer to rivers (mean distance of 3,856 meters) than those of 2nd rate soils or 3rd rate soils (4,768 and 5,955 meters, respectively).   The same pattern is also evident in mean distances to the 58 fields mapped the GLO surveyors.  These 58 fields are even closer to timber/grove (418 meters) than section line midpoints with 1st rate soils (949 meters) or soils not rated (1,126 meters).  Overall, the first parcels sold (1848) have the shortest mean distance to timber and grove (204 meters), to fields (2,318 meters), and to river (2,575 meters).  Early farmsteads (1875) have mean distances similar to 2nd rate soils, much farther from timber/grove, river, and field than first parcels soils (1848).

Soil description.  GLO Deputy Surveyors included description of the soil for 87 of the 1,352 section lines in Polk County.  Almost all (xxx percent) of the xxx section lines described as "rich" or "sandy" were "not rated."

Valley bottom delineation.  As instructed by the Surveyor General in the 1846 General Instructions to Deputy Surveyors, Evans included on his township map the edges of the river valley bottom:  "On this diagram you are to represent, as you progress with your survey, the crossing and courses of all streams of water and of the bottom land through which they meander."  Even though approximately half (51.5 percent) of the 1st rate soils are located in the bottomland area delineated by Evans, only 18.9 percent of the fields are located in the bottomland area.  An even smaller amount (7.8 percent) of the 2nd rate soils are located in the bottomland area.  None of the 3rd rate soils are in bottomland.  In contrast, approximately three-fourths (75.8 percent) of the soils not rated are located in the bottomland area delineated by Evans.  The first parcels sold are located predominantly on the upland (83.1 percent) rather than within the valley bottom delineated by Evans (16.9 percent).  An even higher proportion of early farmsteads (92.3 percent) were located on the upland.

Land surface.  Surveyor Evans included description of the land surface for all but five section lines.  Nearly half (44.9 percent) of the 1st rate soil lines were described as "bottom."  Only 20.6 percent of the 1st rate soil lines were described as rolling.  In contrast, approximately three-fourths of the 2nd rate soils and 3rd rate soils were described as "rolling."  None of the seven fields mapped by Evans were described as being in "bottoms."  Rather, fields were in areas described as "level," "rolling," and "not listed."  The first parcels sold are also in areas described by Evans as "level," "rolling," and "not listed."  Though most (53.8 percent) of the early farmsteads were in areas that Evans described as "rolling," a higher proportion than other features were in areas described as "broken" (23.1 percent).

5. Descriptive modeling of modern soil characteristics

Each of the four classes of soil ratings (1st rate, 2nd rate, 3rd rate, and not rated) can be described by examining and measuring relationships with characteristics of soils mapped in the modern soil survey (Dideriksen and Radatz 2000).  These soil characteristics include the following:

• Landscape position
• Drainage class
• Hydric soils
• Corn Suitability Rating
• Estimated Corn Yield
• USDA Prime Farmlands classification

Landscape position.  Soils rated 1st rate by Evans are located predominantly on floodplain positions (51.6 percent) and upland positions (37.8 percent).  In contrast, soils rated 3rd rate are located predominantly on swale positions (55.2 percent) and upland positions (44.8 percent).  Of the soils not rated, most (70.5 percent) are located on floodplain positions.  The seven fields mapped by Evans are located predominantly on upland positions (50.3 percent) and floodplain positions (38.7 percent).  The first parcels sold are also located predominantly on upland positions (66.4 percent) and floodplain positions (21.2 percent).  Early farmsteads are located predominantly on upland ridge positions (73.1 percent) and upland swale positions (19.2 percent).

A data variable related to landscape position that also indicates topographic patterns is slope.  The 10-meter Digital Elevation Model (DEM) available from the US Geological Survey was used to derive a slope steepness grid theme.  The soil category with the lowest mean slope is 3rd rate soils (1.6 percent slope).  The soil category with the highest mean slope is 2nd rate soils (3.7 percent slope).  However, the first parcels sold have a higher mean slope than the 2nd rate soils (6.8 percent slope).

Drainage class.  Approximately half of the township (51.5 percent) has soils of the poorly drained class.  The percentage is similar for 1st rate soils, 2nd rate soils, soils not rated, and the seven fields mapped by Surveyor Evans.  Almost three-fourths (72.7 percent) of the 3rd rate soils area is poorly drained.  This area is on the upland west of the South Skunk River and north of Bondurant.  This area contains soils in depressions and swales, such as Okoboji, Harps, Canisteo, Webster, and Nicollet.  The lowest percentage of poorly drained soils is for the seven fields mapped by Surveyor Evans (45.9 percent) and the first parcels sold (27.3 percent).  The proportions are similar for early farmsteads.

Hydric soils.  The patterns in hydric soils percentages are similar to those for drainage class.  The highest percentage of hydric soils is for 3rd rate soils (55.2 percent).  A lower percentage of hydric soils is for 2nd rate soils (38.0 percent).  The lowest percentage of hydric soils is for the seven fields mapped by Surveyor Evans (33.7 percent) and the first parcels sold (24.6 percent).  This again suggests that the Euro-American settlers who preceded Evans' survey selected field locations that avoided soils with poor drainage.  This trend continued to 1875 with early farmsteads.

Corn Suitability Rating.  Corn Suitability Rating (CSR) is a measure of the soil's suitability for modern row-crop agriculture.  CSR is expressed as an index from 0 to 100.  It was developed at Iowa State University through the Iowa Agriculture and Home Economics Experiment Station and the Department of Agronomy (Fenton and Miller 1996, p. 9).

Corn suitability ratings provide a relative ranking of all soils mapped in the state of Iowa based on their potential to be utilized for intensive row crop production.  The CSR is an index that can be used to rate one soil's potential yield against another over a period of time.  The CSR considers average weather conditions as well as frequency of use of the soil for row crop production.  Ratings range from 100 for soils that have no physical limitations, occur on minimal slopes, and can be continuously row cropped to as low as 5 for soils with severe limitations for row crops.  The ratings listed in this table assume a) adequate management, b) natural weather conditions (no irrigation), c) artificial drainage where required, d) that soils lower on the landscape are not affected by frequent floods, and e) no land leveling or terracing.  The weighed CSR for a given field can be modified by the occurrence of sandy spots, local deposits, rock and gravel outcroppings, field boundaries, noncrossable drainageways, and so forth.  Even though predicted average yields will change with time, the CSRs are expected to remain relatively constant in relation to one another over time.

By the modern standards of CSR, 3rd rate soils have the highest suitability for row crop agriculture (mean CSR = 85.5).  This is the same area that contains a high percentage of poorly drained soils and hydric soils (see previous measures above).   Soils rated 1st rate and 2nd rate have a lower mean (area-weighted average) CSR (74.5 and 74.4 respectively).  The seven fields mapped by Evans have a mean CSR of 82.4, a value closer to 3rd rate soils than 1st rate soils.  The first parcels sold have the lowest area-weighted average CSR (65.2).  Early farmsteads are on soils with an average CSR of 74.0, similar to 1st rate and 2nd rate soils.

Estimated Corn Yield.  Estimated Corn Yield (ECY) is expressed in bushels per acre (Fenton and Miller 1996, p. 9).

The yield estimate for each SMU is based on kind of parent material, slope class, erosion class, natural drainage class, and nature of the subsoil in terms of rooting environment to include limiting layers, soil depth, and plant available water capacity.  In addition, potential for periodic flooding and weather conditions are included.  Corn yields are estimated for high-level management and are normalized for a 5-year average.  High-level management includes the adoption of best available technology for crop production to include agronomic, engineering, and economic practices.

The patterns in measures of Estimated Corn Yield (areas weighted average) are similar to those described above for CSR.  The highest yield is for 3rd rate soils (147.6 bu/ac).  The lowest yield is for 1st rate soils (130.5 bu/ac).  The yield for the seven fields mapped by Evans (146.7 bu/ac) is almost as high as that for 3rd rate soils (147.6 bu/ac).  Of all the categories, the first parcels sold have the lowest area-weighted average yield (110.0 bu/ac).  Early farmsteads are on soils that have an average yield of 147.0 bu/ac, similar to GLO fields and 3rd rate soils.

USDA Prime Farmland.  The USDA Prime Farmland classification includes six classes (Fenton and Miller 1996, p. 8).

Prime farmland, as defined by the USDA, is the land that is best suited to food, feed, forage, fiber, and oilseed crops.  It may be cropland, pasture, woodland, or other land, but is not urban and built-up land or water areas.  It either is used for food or fiber or is available for these uses.  The soil qualities, growing season, and moisture supply are those needed for a well-managed soil to produce economically a sustained high yield of crops.  Prime farmland produces the highest yields with minimal inputs of energy and economic resources, and farming it results in the least damage to the environment.  Prime farmland usually has an adequate and dependable supply of moisture from precipitation or irrigation.  The temperature and growing season are favorable.  The level of acidity or alkalinity is acceptable.  Prime farmland has few or no rocks and is permeable to water and air.  It is not excessively erodible or saturated with water for long periods and is not frequently flooded during the growing season.  The slopes range mainly from 0 to 6 percent.  Some soils have a seasonal high water table and soils that are frequently flooded qualify for prime farmland only in areas where these limitations have been overcome by a drainage system or flood control.  The need for these measures is indicated by a number following the letter designation for prime farmland.  On-site evaluation is needed to determine whether or not these limitations have been overcome by corrective measures.

• P = Prime
• P2 = Prime, where drained
• P3 = Prime, if protected from flooding or does not flood more than once in 2 years during a growing season
• P5 = Prime, where drained and protected from flooding
• S = Statewide Importance.  These are soils that generally also can be highly productive for cropland, but occur on slopes greater than 6% or have limitations in drainage or flood control that are more difficult to overcome.  These soils are in capability class 3 or 4.  At this time, the soils identified as statewide importance are a potential listing as it has not yet been approved by the State of Iowa.
• L = Local Importance.  These are soils that generally are poorly suited or unsuited to cropland because of the steepness of slope or flooding and wetness limitations.  They may be important in the county, however, for other uses such as pasture, wildlife, or recreation.  The soils identified as local importance are a potential listing of soils that may be considered by county officials for this designation.

For this research, the four Prime classes were aggregated.  Again, the patterns in percentages of USDA Prime Farmland classes are similar to those described above for CSR and ECY.  Third rate soils are 97.2 percent Prime.  Only 81.3 percent of 2nd rate soils are Prime.  The seven fields mapped by Evans are 92.5 percent Prime (the second highest percentage).  The first parcels sold have the lowest proportion of Prime land (65.3 percent) and the highest proportion of land of Local importance (22.8 percent).  Early farmsteads have a higher proportion of Prime land (76.9 percent).

The total length of section lines with 1st rate, 2nd rate, and 3rd rate soils (51.4 miles) were analyzed in terms of their USDA Prime Farmlands class.  (This analysis excludes the section lines for which the soils were not rated by Deputy Surveyor Evans.)

The GLO soil classification has almost equal amounts of 1st rate and 2nd rate soils (44.2 percent and 40.3 percent, respectively).  The USDA Prime Farmlands classification is quite different with a predominance of Prime farmland (86.1 percent). 

10. Literature cited

Anderson, Paul F.  1996.  GIS research to digitize maps of Iowa 1832-1859 vegetation from General Land Office township plat maps.  Des Moines: Iowa Department of Natural Resources, State Preserves Advisory Board; and Ames: Department of Landscape Architecture, Iowa State University.

Andreas, A.T. 1875. Illustrated historical atlas of the State of Iowa.  Andreas Atlas Company, Chicago. 590 p.

Bogue, Allan C.  1963.  From prairie to corn belt:  farming on the Illinois and Iowa prairies in the nineteenth century.  Ames: Iowa State University Press.  309 p.

Chalmers, Thomas.  1832.  Political economy in connection with the moral state and moral prospects of society.  Edinburg: W. Collins.

DeMers, Michael N.  2002. GIS modeling in raster.  New York: Wiley.  203 p.

Dideriksen,  Robert O., and Caryl A. Radatz.  2000.  Soil survey of Polk County, Iowa.  Washington, DC: USDA, Natural Resources Conservation Service.  290 p.

Dodds,  J.S.  1943. Original instructions governing public land surveys of Iowa.  Ames: Iowa Engineering Society.  565 p.

Duden, Gottfried. 1829.  Report on a journey to the western states of North America and a stay of several years along the Missouri (during the years 1824, '25, '26, and 1827).  Reprinted English translation by the State Historical Society of Missouri and University of Missouri Press (1980).  372 p.

Fenton, Thomas E., and Gerald A. Miller.  1996.  Iowa Soil Properties and Interpretations Database (ISPAID 6.0).  Ames: Iowa State University, Iowa Agriculture and Home Economics Experiment Station.  28 p.

Flint, Timothy.  1832.  The history and geography of the Mississippi valley:  to which is appended a condensed physical geography of the Atlantic United States, and the whole American continent.  Cincinnati : E.H. Flint and L.R. Lincoln.  464 p.

Lokken, Roscoe L.  1942.  Iowa public land disposal.  Iowa City: State Historical Society of Iowa.  318 p.

Minnick, Robert F.  1964.  A method for the measurement of Areal Correspondence.  Papers of the Michigan Academy of Science, Arts, and Letters 49: 333-343. 

Peterson, J.B. and A.J. Engelhorn.  1946.  The soil that grows the crops.  p. 18-31 in A century of farming in Iowa, 1846-1946.  Ames: Iowa State College Press.  357 p.

Pratt, LeRoy G. 1974. From cabin to capital:  a brief history of Des Moines and Polk County, Iowa. Iowa Department of Public Instruction, Des Moines.  87 p.

Prior, Jean C. 1991. Landforms of Iowa. University of Iowa Press, Iowa City.  153 p.

Ross, Earle.  1951.  Iowa agriculture:  an historical survey.  Iowa City: State Historical Society of Iowa.  226 p.

Sage, Leland L.  1974.  A history of Iowa.  Ames: Iowa State University Press.  376 p.

Stewart, Lowell O.  1935.  Public land surveys: history, instructions, methods.  Ames: Collegiate Press. 202 p.

Swierenga, Robert P.  1968.  Pioneers and profits: land speculation on the Iowa frontier.  Ames:  Iowa State University Press.  260 p.

Union Historical Company.  1880.  The History of Polk County, Iowa.  Des Moines: Birdsall, Williams, and Company.  1037 p.

Unwin, David.  1981.  Introductory spatial analysis.  Methuen, London.  212 p.

Last update:  19 November 2004