Geoderma xxx (2014) xxx–xxx

GEODER-11660; No of Pages 11

Contents lists available at ScienceDirect

Geoderma

j ourna l homepage: www.e lsev ie r .com/ locate /geoderma

The historical role of base maps in soil geography

B.A. Miller a, R.J. Schaetzl b

a Leibniz-Centre for Agricultural Landscape Research (ZALF) e.V., Institute of Soil Landscape Research, Eberswalder Str. 84, 15374 Müncheberg, Germanyb Michigan State University, Dept. of Geography, 673 Auditorium Rd., East Lansing, MI 48912, USA

E-mail addresses:miller@zalf.de (B.A. Miller), soils@m

http://dx.doi.org/10.1016/j.geoderma.2014.04.0200016-7061/© 2014 Elsevier B.V. All rights reserved.

Please cite this article as: Miller, B.A., Schae10.1016/j.geoderma.2014.04.020

a b s t r a c t

a r t i c l e i n f o

Article history:Received 7 March 2014Received in revised form 8 April 2014Accepted 16 April 2014Available online xxxx

Keywords:Soil mappingGeographic technologyHistoryEnvironmental correlationModel parametersHistory of mapping

Soil mapping is a major goal of soil science. Soil maps rely upon accurate basemaps, both for positional referenceand to provide environmental data that can assist in the prediction of soil properties. This paper reviews thehistorical development of base maps used for soil mapping, and evaluates the dependence of soil mapping onbase maps. The availability of geographic technology for producing base maps has both constrained and directedthe geographic study of soil. The lack of accurate methods for determining location limited early geographicdescriptions of soils to narratives, or to listings of attributes for property-based map units. The first real basemaps available for soil mapping were outline maps produced in the late 18th century, fueled by governments'interests in documenting national boundaries and popular interest inworld atlases. These early soilmaps primar-ily used outline maps as a positional reference onto which soil-related thematic detail was added. Eventually,additional spatial information, in the form of topographic maps and later aerial photographs, increased thepredictive role of base maps in soil mapping. In the current digital, geospatial revolution, global positioningsystems and geographic information systems have nearly replaced the positional reference function of basemaps. Today, base maps are more likely to be used as parameters in soil-landscape models for predicting thespatial distribution of soil properties and classes. Formerly, as a reference for spatial position, paper base mapscontrolled the cartographic scale of soilmaps. However, this relationship is no longer true in geographic informationsystems. Today, as parameters for digital soil maps, base maps constitute the library of predictive variables andconstrain the supported resolution of the soil map.

© 2014 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 02. Early soil geography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

2.1. Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 02.2. Early maps with soil attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

3. The emergence of topographic maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 03.1. Surveying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 03.2. The rise of thematic maps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 03.3. Use of topographic maps for soil mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

4. Introduction of aerial photography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 05. Base maps in the age of geographic information systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

5.1. More spatial data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 05.2. GPS becomes primary source of spatial referencing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 05.3. Challenging definitions of scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

6. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

su.edu (R.J. Schaetzl).

tzl, R.J., The historical role of base maps in soil geography, Geoderma (2014), http://dx.doi.org/

2 B.A. Miller, R.J. Schaetzl / Geoderma xxx (2014) xxx–xxx

1. Introduction

Soil geography and mapping are indispensable components of soilscience, because soils are inherently spatial (Arnold, 1994; Campbelland Edmonds, 1984; Fridland, 1974; Goryachkin, 2005; Hole, 1953,1978; Hudson, 1992). The processes for creating soil maps, with usesranging from scientific study of soil pattern to applied use and manage-ment decisions, are firmly rooted in the concepts and methodologies ofgeography (Bushnell, 1929; Florinsky, 2012; Helms et al., 2002). There-fore, it can be beneficial to review the historical development of soilmapping, as it relates to evolving concepts in geography.

The on-line Dictionary of Cartography (2014) defines a base map as a“map onwhich informationmay be placed for purposes of comparison orgeographical correlation.” From the beginning of accurate surveying tech-niques (17th century), until the widespread completion of topographicmaps, base maps were often limited to outline maps, which provideonly positional reference. Outline maps tended to consist mainly of polit-ical borders, because these were of greatest interest to the governmentswho funded their compilation (Harley, 1988). Early soil maps usedthese base maps because the mapper needed a positional reference onwhich to plot observations, and these basemapswere the only ones avail-able. When other kinds of base maps - containing information useful be-yond positional reference - became available for soil mapping, theadditional information in them could be used to predict known soil pat-terns (Dokuchaev, 1883/1967; Florinsky, 2012; Hole and Campbell,1985; Milne, 1935). Thus, historically, the base map has served as botha positional reference and a source of soil-landscape model parametersfor the production of soil maps.

Although positional reference is a key function of all maps, recentdevelopments in global positioning systems (GPS) and geographicinformation systems (GIS) have nearly replaced the need for separatemaps to provide positional reference. Because of the heavy reliance onbase maps for positional reference in the past, these new geographictechnologies may lead some to consider the term ‘basemap’ as outdated.However, base maps still have use in “geographic correlation.”

In the modern definition of base maps, geography is recognizing thecontinuing role of basemaps for spatial association. Indeed, GPS data donot actually replace base maps, because GPS only provides informationon location. In contrast, base maps provide more information thanlocation alone; they include spatially associated attributes and context(Abler, 1993). Base maps’ significance to soil mapping has, in fact,grown over the past century, as new technologies have provided newand more data-rich base maps with critical information about the soilenvironment. In digital soil mapping applications, base maps with thesekinds of data have been commonly referred to by terms such as ‘parame-ters’, ‘covariates’, or ‘input variables’ (e.g. Behrens and Scholten, 2006;Grunwald, 2009; McBratney et al., 2003). These terms have been carriedover from non-spatial modelling applications. However, maps used asinput variables are a special kind of parameter used in soil-landscapemodelling (Dobos et al., 2000; Levi and Rasmussen, 2014), with impor-tant spatial characteristics. Therefore, the continuing influence of basemap properties on soil geography and soil mapping needs to be clarified,explained and hence, appreciated. Acknowledgment of basemaps’ role inGIS-based methods of soil mapping provides a key link betweentraditional soil mapping endeavors of the past, and the new wave ofdigital soil mapping (McKenzie and Ryan, 1999). Thus, in this paperwe examine the changing role of the base map in the creation of soilmaps throughout history, guiding this discussion within the context ofgeographic technology’s evolution over time.

2. Early soil geography

2.1. Background

Ever since early peoples began sowing crops, and perhaps before,humans have had a vested interest in the geography of soil. Dating

Please cite this article as: Miller, B.A., Schaetzl, R.J., The historical role o10.1016/j.geoderma.2014.04.020

back to 3,000–2,000 BCE in central India, archeologists have foundevidence of farming on the fertile, black soils formed in the Deccan basalt,but not in areas to the north where these soils are absent (Shchetenko,1968). Around the same time, the people ofMesopotamiawere adjustingcropping patterns in response to observed differences in soil fertility(Brevik and Hartemink, 2010; Krupenikov, 1993). In eastern Sweden,where farming systems date back to 500 BCE, patterns of settlements,fields, meadows, and pastures are correlated to soil type (Widgren,1979). Similarly, archeological sites in the United States (U.S.) havebeen shown to be preferentially located on well-drained soils (Almy,1978). In each of these cases, a mental map of where certain soil charac-teristics existedwas likely used indecidingwhere to live, settle, andplant.Although such information about soilswas probably gathered by trial anderror, eventually some rational patterning must have emerged so thatnumerous excavations were no longer required – if at all. Early peoplesmost likely used information observable from the surface, e.g., plants,slope, wetness, etc., as reference data with which to build their mentalmap of desired soil properties.

The Greeks of the Hellenistic Period were particularly astute atobservation. Among these observations, theywere the first to recognizethe information that could be gained from examining a soil profile,which led to improved ideas related to pedogenesis (Krupenikov,1993). Beginning with Parmenides and Eudoxus (ca. 540–470 BCE),they also recognized the connections among similar soils, vegetation,and climates on large scales - in belts on the Earth (Isachenko, 1971;Krupenikov, 1993). Theophrastus (ca. 371–287 BCE) observed thatdifferent soils react differently to the vagaries of weather, and thatdifferent crops were better suited for different soils (Theophrastus,1916). Extending the soil geography of Theophrastus to predictiveindicators, Marcus Cato (ca. 234–149 BCE) recalled a point made byDiophanes of Bithynia that, “you can judge whether land is fit for cultiva-tion or not, either from the soil itself or from the vegetation growing on it”(p. 205, Cato, 1934). Even though the concepts of spatial and functionalassociationwere in place at this time, an actual soil mapwas still missing.

Many of the techniques for making the kind of paper map we knowtoday began to appear after 600BCE (Eratosthenes, 2010). However, thepreferred method for describing soil geography remained the writtennarrative. There are two primary reasons for the lack of interest inproducing a physical soil map at this point in history: (1) knowledgeof soil was minimal and generalized, and (2) base maps of any kindwere extremely crude. To be sure, themanual measurement of distanceacross land was relatively reliable. However, because a large amount ofthe geographic data was based on the narrative descriptions of explorers,assembling locations together on a two-dimensional format, i.e. the map,was not as reliable. For example, Herodotus (ca. 484–425 BCE)was highlycritical of maps, because of their potential to be misleading. He believedthat a map showing Persia far away from Lacedaemon (modern Greece)convinced the Spartans not to assist in the Ionian Revolt. In place ofmaps, he felt that his table of distances provided more accurate informa-tion (Eratosthenes, 2010). Of course, the potential formaps to bemislead-ing still exists today (Monmonier, 1996), but the primary problem withmaps of any kind during this early time was accurately representing theposition of all features consistently.

Due to the difficulty of accurately measuring position on the Earth,map scale was very inconsistent on early maps. Even though earlymaps essentially did not have a cartographic scale, their relationshipbetween extent and generalization remain consistent with the modernuse ofmap scale terms. Specifically, for a given size ofmedia (i.e. paper),themap could either includemore detail about a small extent or it couldbe more generalized for covering a larger extent.

Because of the limitations in geographic technology, maps before1500 CE were either theoretical maps about the Earth with a smallcartographic scale, or maps created for civil purposes with a largecartographic scale. Small cartographic scale maps were popular withnatural philosophers for the purpose of illustrating conceptualizationsabout the known world (Fig. 1). Large cartographic scale maps were

f base maps in soil geography, Geoderma (2014), http://dx.doi.org/

3B.A. Miller, R.J. Schaetzl / Geoderma xxx (2014) xxx–xxx

generally produced for individual cities and for more practical purposes.For example, the Romans surveyed grid systems of “centuries” as a basemap for city planning and levying of taxes (Dilke, 1987). Advances byPtolemy (ca. 100–178 CE) in map projections and coordinate systems(including the prototype for latitude and longitude), greatly improvedthe utility of maps. However, it would take technological developments,namely the printing press, in the Early Modern period (ca. 1500–1800CE) to popularize the use of paper maps (Brown, 1979).

2.2. Early maps with soil attributes

The earliest known spatial representations, i.e., maps, of soil proper-ties were tied to the assessment of land valuation at the analysis scale ofparcels or fields. In general, the valuation of land - fromwhich the levy-ing of taxes was based - was assessed by an index of soil productivity(Kain and Baigent, 1992). Therefore this soil attribute was associatedwith documents of land ownership. Cadastral recordings of soil qualitywere fairly common for levying taxes in the feudal system of westernEurope. From the 6th to the 15th century, cadastrals in the ArabianCaliphates used soil productivity as a factor (with crop type) to levytaxes (Krupenikov, 1993). From approximately 300 to 1951 CE, Chinacompiled data on land quality by crop fields, as well as with otherphysiographic properties, into special geographic descriptions calleddifanchzhi (Lee, 1921; Zaichikov, 1955). Although not a soil map perse, the linking of soil data with the spatial information of propertyboundaries was an advancement toward the creation of a real soil map.

Although cadastral maps advanced the geographic collection of soildata, they lacked a fundamental theory of prediction. Also, geographicinformation was limited, because soil boundaries often were simplyfield boundaries. It is likely that land assessors had some type of mentalmodel for interpreting the fields they viewed, but like the ancients, theylacked a systematic approach for testing those relationships. Moreimportantly, the cadastral maps summarized soil information with ananalysis scale based on field and property delineations, instead of as a

Fig. 1. A map that predates the use of cartographic scale – Eratosth

Please cite this article as: Miller, B.A., Schaetzl, R.J., The historical role o10.1016/j.geoderma.2014.04.020

natural body or continuous surface. As this example shows, the approachto soil science was, at least in part, directed by the available base maps.

3. The emergence of topographic maps

3.1. Surveying

Before thematic maps (i.e. geologic and soil maps) would be consid-ered viable endeavors, mappers needed more reliable base maps of thelandscape. Among the key ingredients for producing a reliable basemapwas the accurate determination of position. Early maps, with smallcartographic scales, were so speculative that a basic outline of landmasses was generally sufficient. Nonetheless, during ancient timeseven the configuration of the continents was known only generally.The scientific basis of continent outlines steadily improved during theEarly Modern period (ca. 16th–19th centuries). In 1554, GerardusMercator improved Ptolemy’s estimation of the Mediterranean Sea’slength from 62° to 53° longitude, which was corrected to 41° longitudebyGuillaumeDelisle in 1700 (Brown, 1979). The technology to geograph-ically plot observations and examine generalized spatial patterns becamemore available later in the 18th century. Although the same technologywas needed to produce accurate maps with large cartographic scales, itwould take much longer to map large areas of extent at the associatedlevel of detail. For this reason, thematic maps used for scientific studystill tended to use small cartographic scales, up to and through the 19thcentury.

Although the principles for determining position were mostlyworked out by the ancients, itwould take inventions of the 17th centuryto produce instruments accurate enough to determine positionwithin afew meters (Bennett, 1987). The development of this technology washastened by the Age of Exploration, which began in the 15th century(Martin andMartin, 2005). At that time, to calculate latitude oneneededonly the angularmeasurement of a celestial body at itsmaximumheightin the sky and the assistance of mathematical tables based on previous

enes’s map of the known world, ca. 194 BCE (Bunbury, 1883).

f base maps in soil geography, Geoderma (2014), http://dx.doi.org/

4 B.A. Miller, R.J. Schaetzl / Geoderma xxx (2014) xxx–xxx

observations. However, until devices measuring fractions of angularseconds were available, the measurement of latitude commonly hadan error range of a couple of degrees (equivalent to approximately100 km per degree of latitude, in the mid-latitudes). Important innova-tions for increasing both the accuracy and precision of measuring celes-tial angles were filar micrometers and the replacement of alidades withtelescopes fittedwith cross-hairs (Bennett, 1987). After the invention ofaccurate devices for measuring angles, the determination of longitudeon land only required the establishment of a celestial clock (Brown,1979). By 1676, sufficient data had been collected on the eclipses ofJupiter’s satellites to be used as a consistent timepiece anywhere inthe world.

During much of the history of modern cartography, the productionof base maps using the improved methods of location determinationrequired the patronage of national governments. In 1668, Francebecame the first country in Europe to sponsor a systematic survey of itslands using accurate methods. Its motivation was to improve infrastruc-ture for the purpose of increasing trade and tax revenue (Brown, 1979).After establishing a series of points in a line using astronomic observa-tions, detailed topographic surveying could be done, extending out fromthe established (meridian) line. Because of the time it took to producesuch an accurate survey, the project focused on surveying France’s borderfirst. In 1739, 70 years after the meridian line was established, the firstsurvey of France’s border was completed. The topographic survey ofFrance then ensued, requiring monies from private investors and localgovernments through time of war. After four generations of work bythe Cassini family on this project, the first general topographic map ofFrance was completed in 1815 (Konvitz, 1987) (Fig. 2).

Soon after the successful beginnings of France’s mapping endeavor,other countries began initiating their own topographic surveys. By1730, topographic surveys were underway in Bohemia, Moravia, Silesia,Lombardy, Transylvania, Austria, Hungary, and Germany (Brown, 1979).Astronomic and triangulation survey in the U.S. began with Meriwether

Fig. 2. A portion of thefirst systematically surveyed, topographicmap of France published betwenote the lack of elevation information. The cartographic scale was 1:86,400. (EHESS, 2014).

Please cite this article as: Miller, B.A., Schaetzl, R.J., The historical role o10.1016/j.geoderma.2014.04.020

Lewis andWilliam Clark’s expedition in 1803, and continued sporadicallywith subsequent military reconnaissance parties. However, a systematictopographic survey of the U.S. did not start until after 1885 (Wheeler,1885).

The transition between simple outline maps and topographic mapshas considerable relevance to the field of soil geography. Topographicmaps quickly evolved from being outline maps -usually the survey ofa political boundary- as a starting point, to containing additional infor-mation about the landscape (Konvitz, 1987). Although topographicmaps usually contain information about topography, they also regularlyinclude information on cultural features, water bodies, and vegetation.As more information beyond the outline of political boundaries cameto be included in these types of base maps, they provided the soilmapper with data that were useful beyond locational reference alone.This added information provided parameters useful to the soil mappers’mental model of spatial association. And yet, the thematic maps thatwere to come would prove to be a key ingredient to better soil maps.

3.2. The rise of thematic maps

From their inception, thematicmaps have relied on basemaps for po-sitional reference, because of the special equipment and time-consumingwork of accurately determining location on the Earth’s surface. In otherwords, a thematic map could have been created from astronomicallyreferenced positions, but it was farmore efficient to utilize the previouslymade positional references in an existing base map. Base maps werepurchased by publishing companies, who printed compilations of themas atlases. These atlases were widely popular, because they presentednew views of the world and reported the discoveries of Europeanexplorers. By the 1600s, hand-coloring of these maps was a fashionableactivity for the recreational study of geography (Brown, 1979). Basemaps were used like popular coloring books, which could be eithereducational or used to add whatever theme one wished.

en 1756 and 1815. Although this mapwas amilestone for accurate positioning of features,

f base maps in soil geography, Geoderma (2014), http://dx.doi.org/

5B.A. Miller, R.J. Schaetzl / Geoderma xxx (2014) xxx–xxx

During the Age of Exploration, atlases grew from being collections ofbase maps from around the world to also including thematic maps ofthose places. In 1890, Titov summarized the work of several mid tolate 17th century Russian geographers, who described the distributionsof climate, vegetation, and soils (Krupenikov, 1993). Also in typical atlasstyle of the time, the French Constantin-François Volney published Viewof the Soils and Climatic Pictures of the United States of America in 1804.This early physiographic description and map of North America com-pared Canada with Siberia and the prairies with the Russian steppes(Volney, 1804).

The reliance of thematicmaps onbasemaps focused the production ofthematic maps at the cartographic scale of the available base maps. Theprimary sponsors of base map production were national governmentsinterested in outlining the borders of their country (Harley, 1988). There-fore, the first available, accurate, basemapswere at the cartographic scalebest suited for fitting the area of a country on the largest piece of paperavailable. However, in order to add detail to the maps, their cartographicscales needed to be increased and their extents decreased. To expand themap extent again required drawing the maps on adjoining sections ofpaper (e.g. Fig. 3). In the 18th through 20th centuries, these physicallimitations set the cartographic scale ofmaps based on the desired extent.

3.3. Use of topographic maps for soil mapping

When agrogeologists met the newly available, accurate, topographicmaps, a new synergy was found. For the first time, a resource existedthat allowed them to record and visualize spatial information with areasonable amount of efficiency and accuracy. Scientists could simplyplot their observations on the base map. Further, they could connecttheir observations on the map based on experience alone or guided byadditional information in the base map (e.g. elevation). For example,William Smith’s 1815 geology and soil map of England was built onprintings of John Cary’s 1812 topographic map (Fig. 3). The geologicalattributes were added by hand, using water colors (University of NewHampshire, 2013). The linking of soil and geology attributes wasbased on the mapper’s experience of soil characteristics correspondingto the spatial distribution of geology as mapped at that cartographicscale (Boud, 1975; Brevik and Hartemink, 2010). Stanisław Staszic’s1815 geologymap of Eastern Europe used the same approach - coloringtopographic maps that were reproduced from an earlier cartographer’sefforts (Grigelis et al., 2011).

The famous map produced by V.V. Dokuchaev in 1883, showing thedistribution of Chernozems in European Russia, is another example ofthe reliance of early soils maps on basemaps (Fig. 4). It can be assumedthat Dokuchaev did not determine the location of the cities, rivers, andland boundaries included on his map. Instead, he plotted observationsof humus content on a pre-existing base map. This followed the styleof geographic study of the natural environment popularized byAlexandervon Humboldt (Hartshorne, 1958; Robinson and Wallis, 1967). Hence,the cartographic scale in Dokuchaev’s and other thematic maps of thenatural world was controlled by the available base map. Consequently,their analysis scale was also dictated by the level of detail that could beadded to those maps.

When the U.S. Soil Survey began in 1899, its primary goal was tocreate maps specific enough to provide guidance for crop selection(Whitney, 1900, 1909). However, the base maps available preventedthe use of larger cartographic scales and limited the amount of detailthat could be shown (Simonson, 1952). Furthermore, by this time, theU.S. Geological Survey had produced topographic maps for only asmall portion of the U.S., generally focusing on areas of interest formineral mining, irrigation, and land reclamation (Brown, 1979). As aresult, the typical kit for the first U.S. soil surveyors included a six footsoil auger with extensions, shovel, compass, protractor and scale,accompanied by “a copy of the usually inadequate county or other avail-able basemap” (Lapham, 1949, p. 12). Soil boundaries were determinedby frequent borings. In situations where a topographic map wasn’t

Please cite this article as: Miller, B.A., Schaetzl, R.J., The historical role o10.1016/j.geoderma.2014.04.020

available, soil boundaries would be sketched on a blank plat book. Theplat books were usually based on the General Land Office surveys,which commonly had many errors (Lapham, 1949). Thus, the endeavorof detailed soilmapping in the early 1900swas limited by the inadequacyof base maps.

4. Introduction of aerial photography

The ability to locate features in soil surveys accurately and withgreater detail was markedly enhanced by the use of aerial photographyin the 1930s (Soil Survey Staff, 1951). Many advances during WorldWar I had improved the utility of aerial photographs for mapmakingin general (Smith, 1985). Soon thereafter, the U.S. Geological Surveybegan using aerial photography for making topographic maps (Davey,1935). Aerial photography for soil survey work in the U.S. had beenproposed as early as 1923 (Cobb, 1923), but the U.S. Bureau of Chemistryand Soils showed little interest in taking on the expense (Simonson,1989). Bushnell (1929) compared soil mapping with and without aerialphotography, and made four main points on the advantages of usingaerial photography. Aerial photographs made soil mapping (1) moreefficient, (2) of higher quality, (3) with improved consistency, and (4)with reduced cartographic demands for the determination of location.

The use of aerial photography in soil survey operations in the U.S.was gradually adopted as commercial aerial photography (Bushnell,1932), or state level aerial photographs, becamemore widely available.Michigan was the first state to report use of a mirror stereoscope in soilmapping, allowing for soil boundary determinations over areas thatwere thick in vegetation or otherwise difficult to traverse (Millar,1932). In 1951, standards for interpreting aerial photographs wereadded to the U.S. Soil Survey Manual (Soil Survey Staff, 1951). By theend of the 20th century it would be common practice for the U.S. SoilSurvey to use aerial photographs taken with a 153 mm lens, resultingin a 23 cm square image with a scale between 1:20,000 and 1:80,000(Simonson, 1989; Soil Survey Staff, 1993). An example of a soil mapthat used an aerial photograph as both a source of positional referenceand spatial association is provided in Fig. 5.

The paradigm shift caused by the implementation of aerial photogra-phy has yielded lessons about the use of base maps. For example, Brown(2006) criticized the U.S. Soil Survey for limiting the efficiency of soilmapping because it continued to use extensive ground observationsand field sampling of every delineation. However, when Pomereningand Cline (1953) compared results from soil mapping done using aerialphoto interpretation alone versus that done with field mapping, theyfound that field reconnaissance was still important for maintaining mapaccuracy. In their study of soil mapping procedures, the mean for soilseries mapped correctly was only 48% using aerial photographinterpretation alone, but increased to 84%when photo interpretationwas combined with field reconnaissance. A similar increase inmapping accuracy was obtained for other soil attributes such as drainageclass, parent material, and land use capability. In summary, althoughhigh-quality aerial photography improved the U.S. Soil Survey mapproducts by facilitating greater detail and precision, it did not replacethe need to understand the contextual information in aerial photographsthrough field observation.

By the late 1900s, even though aerial photography had greatlyincreased the level of detail that could be included in a soil map, limitsstill existed. The minimum delineation size on paper maps is relativeto the legibility of identifying symbols and colors, as well as levels ofspatial accuracy (Arnold, 2006; Hupy et al., 2004). For example, errorsof location and internal composition increase as delineation sizedecreases. Therefore, the decision of cartographic scale and associatedminimum delineation size for a map depended on the purpose, degreeof boundary precision possible, and themaximumamount of acceptableerror. In the U.S. Soil Survey program, those limits are expressed bydefinitions of map orders (Schoeneberger et al., 2012).

f base maps in soil geography, Geoderma (2014), http://dx.doi.org/

Fig. 3. The first geological/soil map of Britain, produced bywater coloring a pre-existing topographic map. Published byWilliam Smith in 1815. In order to increase the cartographic scale,the basemap needed to be printed onmultiple pieces of adjoining paper. The cartographic scalewas approximately 1:316,000. (Library Foundation, Buffalo and Erie County Public Library,2013).

6 B.A. Miller, R.J. Schaetzl / Geoderma xxx (2014) xxx–xxx

5. Base maps in the age of geographic information systems

The recent introduction of geographic information technologies,such asGPS andGIS, has instigated another revolution in soil geography.Similar to aerial photographs, and topographic maps before them, these

Please cite this article as: Miller, B.A., Schaetzl, R.J., The historical role o10.1016/j.geoderma.2014.04.020

new technologies have created opportunities in soilmapping by provid-ing critical spatial information and newmethods to analyze that data. Inshort, modern geospatial technologies are resetting the soil mappingparadigm yet again. New technologies beckon the reevaluation ofcurrent procedures and the development of new techniques for the

f base maps in soil geography, Geoderma (2014), http://dx.doi.org/

Fig. 4. Isohumus belts identified by Dokuchaev based on quantitative point observations, plotted on a topographic base map. The cartographic scale was 1:4.2 million.(Dokuchaev, 1883/1967).

Fig. 5. A portion of the Lincoln County, Wisconsin soil survey, conducted by the U.S. Natural Resources Conservation Service, which typically uses aerial photograph base maps (Mitchell,1996). Even though this is a forested, and not a bare-soil, image, the tone, texture, and value of the photograph provide clear clues about the location of soil map unit boundaries. Lineplacement on the photograph shows how the aerial photograph provides positional reference. The assignment of the same map unit to areas of similar tone is an example of the use ofthe aerial photograph for spatial association. The original cartographic scale was 1:20,000.

7B.A. Miller, R.J. Schaetzl / Geoderma xxx (2014) xxx–xxx

Please cite this article as: Miller, B.A., Schaetzl, R.J., The historical role of base maps in soil geography, Geoderma (2014), http://dx.doi.org/10.1016/j.geoderma.2014.04.020

8 B.A. Miller, R.J. Schaetzl / Geoderma xxx (2014) xxx–xxx

advancement science. For this reason, the current geospatial revolutionhas caused soil science to spawn a new sub-discipline, digital soilmapping (Scull et al., 2003). Many aspects of reality have remainedthe same, however, justifying the continued use of proven conceptsfrom the previous paradigm. New geospatial technologies are shiftingthe soil mapping paradigm in three primary ways: (1) by increasing thetypes of spatial information available as base maps, (2) by replacingbase maps with GPS as the primary source of positional referencing,and (3) by decoupling the relationships between different aspects ofmap scale.

5.1. More spatial data

Remote sensing technologies, beginning in the 20th century andcontinuing today, have greatly improved the quality and variety ofinformation that can be used as base maps (Lillesand et al., 2008). Theprocess of remote sensing is generally based on recording the soil’s orother related environmental properties’ interaction with the electro-magnetic spectrum. This interaction includes the reflectance/absorptionof radiation from the sun or man-made emitters. That information isthen analyzed for single time frames and bands, or processed together,to produce unique perspectives of the environment (Pohl and vanGenderen, 1998). New remote sensing technologies range from proxi-mal to aerial to satellite based sensors, each with unique capabilitiesfor measuring soil properties directly, or for providing indirect evidenceof soil properties by spatial association (Doolittle, 1987; Doolittle andBrevik, 2014; Lobell et al., 2010;Mulder et al., 2011). In addition, remotesensing can interpret the time required for a signal to travel betweenthe emitter, a landscape feature, and back to a sensor, and then usethis information to provide data about landscape morphology (e.g.LiDAR (Wehr and Lohr, 1999)). The spectral and spatial resolution,plus accuracy, of these data continue to improve, which regularly addsto the library of base maps available to soil-landscape modelers.

Basemaps, such as aerial photographs, have also provided importantcontextual information for the soil mapper. From an aerial photograph,interpretations can be made about geomorphology, landscape position,vegetation, soil wetness, erosion status, land use, among other parame-ters useful for predicting soil properties (Goosen, 1968; Soil SurveyStaff, 1993). Spatial analysis methods in GIS can quantify contextmuch more efficiently than traditional methods, but also offer newmethods of analysis. For example, multiple remote sensing bands canbe analyzed together to produce indicators such as the normalizeddifference vegetation index (Kriegler et al., 1969; Rouse et al., 1973).

Although the list of spatial analysis methods available today is large,one of the most commonly used categories of analysis to produce basemaps for soil mapping is land-surface derivatives (Florinsky et al.,2002; Moore et al., 1993). Each of these scale-dependent products ofdigital terrain analysis (Albani et al., 2004; Hupy et al., 2004; Roeckerand Thompson, 2010;Wood, 1996) provides different types of informa-tion about the soil environment (Wilson and Gallant, 2000). The factthat land-surface derivatives can be calculated at many analysis scales(Fig. 6), and new methods of analysis are being regularly proposed,serves only to multiply the number of base maps of this type that areavailable to soil-landscape modelers.

5.2. GPS becomes primary source of spatial referencing

GPS provides an efficient means for accurately determining locationby triangulating for the distance between a GPS receiver and the knownlocations of satellites in orbit around the Earth. The distance betweenthe receiver and the satellites is calculated using the time it takes aradio signal to travel from the respective satellites to the receiver(Hofmann-Wellenhof et al., 2001). By this method, the location of thesurface-based receiver can be determined rapidly, accurately, and atrelatively low cost. Thus, the preferred method for spatial referencingfield observations is now with a GPS receiver (Schaetzl et al., 2002).

Please cite this article as: Miller, B.A., Schaetzl, R.J., The historical role o10.1016/j.geoderma.2014.04.020

Similarly, GPS enables the spatial referencing of remote sensing dataas they are collected, with only minor post-processing adjustments toimprove accuracy (Lillesand et al., 2008; Mostafa and Schwarz, 2000).

Despite the efficiency of GPS, the need for base maps for positionalreference has not been completely replaced. Data collected withoutGPS available need to be georeferenced before they can be used in aGIS. The majority of data requiring georeferencing are legacy data,created before the availability of GPS and GIS technologies (e.g. Pásztoret al., 2010). Such legacy maps can only be incorporated into a GIS bygeoreferencing points viewable in both the legacy map and a base map.Another example is the importing of points referenced by distance fromlandmarks, such as political boundaries (Veenstra and Burras, 2012).Although the role of base maps has been clearly altered by GPS, basemaps are still useful resources for positional reference when a GPS isunavailable or fails.

5.3. Challenging definitions of scale

A common problem for disciplines studying spatial phenomena isthe wide range of interpretations and definitions of scale (Dunganet al., 2002; Withers and Meentemeyer, 1999). Now that geographictechnologies have evolved to the point where scale can be examinedin a variety of ways, it is important to distinguish and standardizedefinitions for the different types of scale. This notion is especiallyrelevant to the consideration of how basemaps are used in soil mapping.The decoupling of the relationship between cartographic scale, analysisscale, extent, and resolution has been a subtle change that has challengedour definitions and understanding of scale (Goodchild and Proctor, 1997).

One of the most fundamental differences between traditional soilmapping and digital soil mapping is that the cartographic scale -oftenreferred to as map scale- of base maps no longer constrains thecartographic scale of the resulting soil map. For practical reasons, suchas legibility, the amount of information that can be placed on a papermap is limited by a finite space, i.e. the piece of paper. Put another way,the level of detail on a papermap is limited by cartographic scale. Becausebase maps provide both positional reference and parameters for spatialassociation, a paper base map controls the cartographic scale and theresolution simultaneously. Thus the resolution, or the spatial density ofinformation, has a direct relationship with the cartographic scale. Resolu-tion can also be equated to the minimum delineation size described inprotocols for traditional soil mapping (e.g., Schoeneberger et al., 2012),although other factors are also involved in setting the lower limit ofminimum delineation size. For example, in Fig. 5, the aerial photographclearly shows darker (muck) areas that the mapper knew existed, butcould not delineate them due to time and legibility issues. In contrast,the user-interface of a GIS allows the display size of a map (i.e. carto-graphic scale) to be freely changed by the viewer, without affecting thedensity of information (i.e. resolution) stored in the spatial database.Because of this characteristic, GIS has eliminated the issue of legibility asa constraining or limiting factor in the determination of the minimumdelineation size. However, it has not eliminated issues of spatial accuracyand resolution (Frank, 2009). Therefore, base maps no longer controlcartographic scale. However, the resolution of the base map is still thelower limit of resolution possible for soil maps, without interpolation ordownsampling. In order to properly convert lessons from the past andutilize opportunities of the newdigitalmedium, it is essential to recognizethe changing relationships between these classic geographic properties.

Further adding to confusion in the new world of GIS, there aremultiple meanings for the term ‘scale’ used in popular and scientificliterature. Geography distinguishes between cartographic scale, andthe concept of analysis scale as a tool for identifying phenomenonscale (Montello, 2001). Analysis scale differs from resolution by beingable to summarize an area containing multiple data points (Miller,2014). A common example of resolution is the size of raster cells,which provides a single value for what has been observed or estimatedabout the space represented by each cell. In this way, the resolution

f base maps in soil geography, Geoderma (2014), http://dx.doi.org/

Fig. 6. The same portion of Lincoln County, Wisconsin as shown in Fig. 5, analyzed for slope gradient. The analysis was conducted on a 3 m resolution DEM (USDA-NRCS, 2014). Thecalculation of slope gradient was conducted at twodifferent analysis scales: (a) 9m and (b) 90m. Although there is a gradient between analysis scales, the different analysis scales providedifferent base maps that could be used for soil mapping.

9B.A. Miller, R.J. Schaetzl / Geoderma xxx (2014) xxx–xxx

describes the level of spatial detail available. The analysis scale can beleft as equivalent to the resolution, or it can consider a greater neighbor-hood of data. Traditionally, changing the analysis scale used differentzones, such as soil delineations or political delineations, to summarizeneighborhoods of data (Gehlke and Biehl, 1934). However, withmodern raster analysis, the context of a cell can be quantified by summa-rizing or otherwise analyzing an area larger than the cell without alteringthe spatial resolution. The change in results due to the change in analysisscale has been termed the modifiable areal unit problem, where thiseffect is divided into the zoning problem and the scale problem (JelinskiandWu, 1996;Openshaw, 1983).When the spatial patterns of an analysisscale coincide with the spatial patterns of a target variable, the analysisscale is described as detecting the scale of the phenomenon.

To illustrate this difference between resolution and analysis scale,consider a cell in a digital elevation model (DEM). The DEM resolutionis the cell size. For the original grid of elevation values, the analysis scaleis equivalent to resolution. If the mean elevation within soil delineationsis calculated from that elevation grid, the analysis scale is changed to alevel equivalent to the soil delineations, even though the information isbased on data collected at the original DEM’s resolution. The DEM alsocan be analyzed for slope, but must use a neighborhood of cells in orderto determine a change in elevation over a distance. The size of thatneighborhood is the analysis scale. The cell is the center point and theneighborhood is shifted respectively for each of the other cells. Thus,the resolution is unchanged, despite a larger analysis scale being usedto calculate new values for each cell location.

Although the definition of scale has long been a point of confusion,using the definitions of the three types of scale set by geography offera basis for clarifying ideas, definitions, and meanings, as we transitionto a new soil mapping paradigm. Cartographic scale, which was once

Please cite this article as: Miller, B.A., Schaetzl, R.J., The historical role o10.1016/j.geoderma.2014.04.020

paramount for describing the characteristics of a map, is now only adescription of how the map is being displayed. Analysis scale is notnecessarily viewable in the map, but is a key part in the process ofcreating the map. And finally, phenomenon scale is the scale at whichspatial patterns occur in reality, which can be discovered or best repre-sented by appropriately matching the scale at which the data areanalyzed.

Other aspects of the digital map are best described by terms otherthan scale. Resolution is the spatial detail of the data, without dependen-cies on any type of scale. Further, extent describes the real-world areaencompassed by the map, which in a GIS is also independent of anytype of scale or the spatial data’s resolution.

We recommend the use of these definitions to disambiguate thevarious uses of the term ‘scale’ in the context of describing base mapsand the process of using them to create soil maps.

6. Conclusions

Soil maps have always relied on base maps for information neededto predict the spatial distribution of soil properties. Over time, thespatial information garnered from base maps has shifted in emphasisbetween positional reference in early maps, to attributes that are morereliably associated with soil properties. The only possible exceptions tothe reliance on base maps are spatial interpolation techniques that solelyrely on spatial autocorrelation. However, recent spatial interpolationmethods, such as co-kriging, reiterate the utility of base maps for makingaccurate soil maps. Even before paper maps, it seems likely that a mentalmap of above ground observations informed the creation of a mentalsoil map.

f base maps in soil geography, Geoderma (2014), http://dx.doi.org/

10 B.A. Miller, R.J. Schaetzl / Geoderma xxx (2014) xxx–xxx

Base maps determine many of the characteristics of the soil map,because base maps are the primary ingredients in the production ofthe soilmap.When soil mapswere produced on paper, the cartographicscale of the soilmap had tomatch the cartographic scale of the basemap,because the base maps were the most practical source for positionalreferencing. When base maps became available with information aboutspatially associated variables, such as geology, vegetation, or elevation,soil mappers were able to use that information to improve the qualityof the soil map. However, unless the soil mappers made additionalobservations, the resolution of the soil map could not reliably exceedthe resolution of information available in the base maps. Also becauseof the practical limits in detail that can be included on a paper map, asdetermined by cartographic scale, base maps also strongly influencedthe analysis scale of soil maps. In contrast, digital maps are free frommany of the constraints of paper maps. However, as the source of spatialdata, base maps are still the limiting factor in the density of spatialinformation available.

As the sophistication and format of base maps have evolved overtime, so have soil maps and the geographical concepts used to under-stand them. Modern definitions of scale in geography help clarifyconcepts that were once inextricably linked. Multiple map characteristicsthat at one time could be summarily described with the scale of the mapnowneed separate definitions to accommodate the new format and toolsof GIS. Cartographic scale, analysis scale, map extent, andmap resolution,must now be treated as independent concepts. In contrast to papermaps,the only link between these concepts in a GIS is the limit of computingpower for handling large file sizes. In this case, map resolution and mapextent could be related, but the relationship is weak and somewhatarbitrary. Therefore, the description of base maps used in the scientificstudy of soil mapping should always includemap extent, map resolution,and analysis scale when it differs from the resolution.

Acknowledgements

We appreciate the helpful comments on this manuscript from theanonymous reviewers.

References

Abler, R.F., 1993. Everything in its place: GPS, GIS, and Geography in the 1990s. Prof.Geogr. 45 (2), 131–139. http://dx.doi.org/10.1111/j.0033-0124.1993.00131.x.

Albani, M., Klinkenberg, B., Andison, D.W., Kimmins, J.P., 2004. The choice of window sizein approximating topographic surfaces from digital elevationmodels. Int. J. Geogr. Inf.Sci. 18 (6), 577–593.

Almy, M.M., 1978. The archeological potential of soil survey reports. Fla. Anthropol. 31 (3),75–91.

Arnold, R.W., 1994. Soil geography and factor functionality: Interacting concepts. Factorsof Soil Formation: A Fiftieth Anniversary Retrospective. Soil Sci. Soc. Am. Spec. Publ.,33, pp. 99–109.

Arnold, R.W., 2006. Soil survey and soil classification. In: Grunwald, S. (Ed.), EnvironmentalSoil-Landscape Modeling: Geographic Information Technologies and Pedometrics.Taylor & Francis Group, Boca Raton.

Behrens, T., Scholten, T., 2006. Digital soil mapping in Germany - a review. J. Plant Nutr.Soil Sci. 169, 434–443.

Bennett, J.A., 1987. The Divided Circle: A History of Instruments for Astronomy, Navigationand Surveying. Phaidon-Christie's, Oxford.

Boud, R.C., 1975. The early development of British geological maps. ImagoMundi 27, 73–96.Brevik, E.C., Hartemink, A.E., 2010. Early soil knowledge and the birth and development of

soil science. Catena 83, 23–33.Brown, L.A., 1979. The Story of Maps. Dover Publications, Inc., New York.Brown, D.J., 2006. A historical perspective on soil-landscape modeling. In: Grunwald, S.

(Ed.), Environmental Soil-Landscape Modeling: Geographic Information Technolo-gies and Pedometrics. Taylor & Francis Group, Boca Raton, pp. 61–98.

Bunbury, E.H., 1883. A History of Ancient Geography among the Greeks and Romans fromthe Earliest Ages till the fall of the Roman Empire. John Murray, London.

Bushnell, T.M., 1929. Aerial photography and soil survey. Am.Assoc. Soil Surv. Bull. 10, 23–28.Bushnell, T.M., 1932. A new technique in soil mapping. Am. Soil Surv. Assoc. Bull. 13, 74–81.Campbell, J.B., Edmonds, W.J., 1984. The missing geographic dimension to Soil Taxonomy.

Ann. Assoc. Am. Geogr. 74, 83–97.Cato, M.P., 1934. On Agriculture (Hooper, W.D., Trans.), Harvard University Press,

Cambridge, Massachusetts.Cobb, W.B., 1923. Possibilities of the airplane in soil survey work. Am. Assoc. Soil Surv.

Work. Bull. 4, 77–80.

Please cite this article as: Miller, B.A., Schaetzl, R.J., The historical role o10.1016/j.geoderma.2014.04.020

Davey, H.C., 1935. Brief outline of aerial photographic work in the U.S. Geological Survey.Photogramm. Eng. 1, 3.

Dictionary of Cartography, 2014. Base map. Geography-Dictionary.org. Available onlinehttp://www.geography-dictionary.org/base_map (Accessed [3/6/2014]).

Dilke, O.A.W., 1987. Roman large-scale mapping in the early empire. In: Harley, J.B.,Woodward, D. (Eds.), The History of Cartography. University of Chicago Press, Chicago.

Dobos, E., Micheli, E., Baumgardner, M.F., Biehl, L., Helt, T., 2000. Use of combined digitalelevation model and satellite radiometric data for regional soil mapping. Geoderma97, 367–391. http://dx.doi.org/10.1016/S0016-7061(00)00046-X.

Dokuchaev, V.V., 1967. The Russian Chernozem (Kaner, N., Trans.). Israel Program forScientific Translations, Jerusalem. (Original work published in 1883).

Doolittle, J.A., 1987. Using ground-penetrating radar to increase the quality and efficiencyof soil surveys. Soil Survey Techniques. Soil Science Society of America, Madison,Wisconsin, pp. 11–32.

Doolittle, J.A., Brevik, E.C., 2014. The use of electromagnetic induction techniques in soilsstudies. Geoderma 223–225, 33–45. http://dx.doi.org/10.1016/j.geoderma.2014.01.027.

Dungan, J.L., Perry, J.N., Dale, M.R.T., Legendre, P., Citron-Pousty, S., Fortin, M.J.,Jakomulska, A., Miriti, M., Rosenberg, M.S., 2002. A balanced view of scale in spatialstatistical analysis. Ecography 25, 626–640.

EHESS, 2014. Des villages de Cassini aux communes d’aujourd’hui. Available online http://cassini.ehess.fr (Accessed [3/7/2014]).

Eratosthenes, 2010. Geographika (Roller, D.W., Trans.), Princeton University Press,Princeton, New Jersey.

Florinsky, I.V., 2012. The Dokuchaev hypothesis as a basis for predictive digital soilmapping (on the 125th anniversary of its publication). Eurasian Soil Sci. 45, 445–451.

Florinsky, I.V., Eilers, R.G., Manning, G.R., Fuller, L.G., 2002. Prediction of soil properties bydigital terrain modelling. Environ. Model Softw. 17, 295–311.

Frank, A.U., 2009. Scale is introduced in spatial datasets by observation processes. In:Goodchild, H. (Ed.), Spatial Data Quality from Process to Decision. CRC Press, pp.17–29. http://dx.doi.org/10.1201/b10305-4.

Fridland, V.M., 1974. Structure of the soil mantle. Geoderma 12, 35–41.Gehlke, C.E., Biehl, K., 1934. Certain effects of grouping upon the size of the correlation

coefficient n census tract material. J. Am. Stat. Assoc. 29 (185A), 169–170. http://dx.doi.org/10.1080/01621459.1934.10506247.

Goodchild, M.F., Proctor, J., 1997. Scale in a digital geographic world. Geogr. Environ.Model. 1, 5–23.

Goosen, D., 1968. Aerial photo interpretation and its significance in soil survey[Interpretación de fotos aéreas y su importancia en levantamiento de suelos]. Boletínde Suelos, 6. United Nations Food and Agriculture Organization, Rome.

Goryachkin, S.V., 2005. Studies of the soil cover patterns in modern soil science:approaches and tendencies. Eurasian Soil Sci. 38 (12), 1301–1308.

Grigelis, A., Wójcik, Z., Narębski, W., Gelumbauskaitė, L.Z., Kozák, J., 2011. StanisławStaszic: an early surveyor of the geology of Central and Eastern Europe. Ann. Sci. 68(2), 199–228. http://dx.doi.org/10.1080/00033790.2010.511263.

Grunwald, S., 2009. Multi-criteria characterization of recent digital soil mapping andmodeling approaches. Geoderma 152, 195–207.

Harley, J.B., 1988. Maps, knowledge, and power. In: Cosgrove, D., Daniels, S. (Eds.), TheIconography of Landscape. Cambridge University Press, Cambridge, pp. 277–312.

Hartshorne, R., 1958. The concept of geography as a science of space, from Kant toHumboldt to Hettner. Ann. Assoc. Am. Geogr. 48 (2), 97–108.

Profiles in the History of the U.S. Soil Survey. In: Helms, D., Effland, A.B..W., Durana, P.J.(Eds.), Iowa State Press, Ames, Iowa.

Hofmann-Wellenhof, B., Lichtenegger, H., Collins, J., 2001. Global Positioning System:Theory and Practice, fifth ed. Springer. http://dx.doi.org/10.1007/978-3-7091-3293-7.

Hole, F.D., 1953. Suggested terminology for describing soils as three-dimensional bodies.Soil Sci. Soc. Am. Proc. 17, 131–135.

Hole, F.D., 1978. An approach to landscape analysis with emphasis on soils. Geoderma 21,1–13.

Hole, F.D., Campbell, J.B., 1985. Soil Landscape Analysis. Rowman & Allanheld, Totowa,New Jersey.

Hudson, B.D., 1992. The soil survey as a paradigm-based science. Soil Sci. Soc. Am. J. 56,836–841.

Hupy, C.M., Schaetzl, R.J., Messina, J.P., Hupy, J.P., Delamater, P., Enander, H., Hughey, B.D.,Boehm, R., Mitrok, M.J., Fashoway, M.T., 2004. Modeling the complexity of different,recently glaciated soil landscapes as a function of map scale. Geoderma 123, 115–130.

Isachenko, A.G., 1971. Development of Geographical Ideas [Razvitie geograficheskikhidei]. Mysl', Moscow.

Jelinski, D.E., Wu, J., 1996. The modifiable areal unit problem and implications forlandscape ecology. Lands Ecol. 11 (3), 129–140.

Kain, R.J.P., Baigent, E., 1992. The Cadastral Map in the Service of the State: A History ofProperty Mapping. The University of Chicago Press, London.

Konvitz, J., 1987. Cartography in France, 1660–1848: Science, Engineering, and Statecraft.The University of Chicago Press, London.

Kriegler, F.J., Malila, W.A., Nalepka, R.F., Richardson,W., 1969. Preprocessing transformationsand their effects on multispectral recognition. Proceedings of the Sixth InternationalSymposium on Remote Sensing of Environment. University of Michigan, Ann Arbor,pp. 97–131.

Krupenikov, I.A., 1993. History of Soil Science (Dhote, A.K., Trans.), A.A. BalkemaPublishers, Brookfield, Vermont.

Lapham, M.H., 1949. Crisscross Trails: Narrative of a Soil Surveyor. W.E. Berg, Berkeley,California.

Lee, M.P., 1921. The economic history of china: with special reference to agriculture. In:The Faculty of Political Science of Columbia University (Ed.), Studies in HistoryEconomics and Public Law, vol. 39. Columbia University, New York.

Levi, M.R., Rasmussen, C., 2014. Covariate selection with iterative principle componentanalysis for predicting physical soil properties. Geoderma 219–220, 46–57.

f base maps in soil geography, Geoderma (2014), http://dx.doi.org/

11B.A. Miller, R.J. Schaetzl / Geoderma xxx (2014) xxx–xxx

Library Foundation, Buffalo and Erie County Public Library, 2013. A Delineation of theStrata of England and Wales with part of Scotland: Exhibiting the Collieries andMines, theMarshes and Fen Lands Originally Overflowed by the Sea, and the Varietiesof Soil According to Variations in the Substrata, Illustrated by the Most DescriptiveNames. Available online at: http://en.wikipedia.org/wiki/File:Geological_map_Brit-ain_William_Smith_1815.jpg (Accessed [3/6/2014]).

Lillesand, T.M., Kiefer, R.W., Chipman, J.W., 2008. Remote Sensing and Image Interpreta-tion, sixth ed. Wiley.

Lobell, D.B., Lesch, S.M., Corwin, D.L., Ulmer, M.G., Anderson, K.A., Potts, D.J., Doolittle, J.A.,Matos, M.R., Baltes, M.J., 2010. Regional-scale assessment of soil salinity in the RedRiver Valley using multi-year MODIS EVI and NDVI. J. Environ. Qual. 39 (1), 35–41.

Martin, G.J., Martin, T.S., 2005. All Possible Worlds: A History of Geographical Ideas.Oxford University Press.

McBratney, A.B.., Mendonça Santos, M.L., Minasny, B., 2003. On digital soil mapping.Geoderma 117, 3–52. http://dx.doi.org/10.1016/S0016-7061(03)00223-4.

McKenzie, N.J., Ryan, P.J., 1999. Spatial prediction of soil properties using environmentalcorrelation. Geoderma 89, 67–94.

Millar, C.E., 1932. The use of aerial photographs in the Michigan land economic survey.Soil Sci. Soc. Am. J. 13, 82–85.

Miller, B.A., 2014. Semantic calibration of digital terrain analysis scale. Cartogr. Geogr. Inf.Sci. 41, 166–176.

Milne, G., 1935. Some suggested units of classification and mapping particularly for EastAfrican soils. Soil Res. 4 (3).

Mitchell, M.J., 1996. Soil Survey of Lincoln County. Wisconsin. U.S. Department of Agricul-ture, Washington, D.C.

Monmonier, M.S., 1996. How to Lie with Maps. University of Chicago Press, Chicago.Montello, D.R., 2001. Scale in geography. In: Smelser, N.J., Baltes, P.B. (Eds.), International

Encyclopedia of the Social & Behavioral Sciences. Pergamon Press, Oxford, pp.13501–13504.

Moore, I.D., Gessler, P.E., Nielson, G.A., Peterson, G.A., 1993. Soil attribute prediction usingterrain analysis. Soil Sci. Soc. Am. J. 57, 443–452.

Mostafa, M.M.R., Schwarz, K., 2000. Amulti-sensor system for airborne image capture andgeoreferencing. Photogramm. Eng. Remote Sens. 66 (12), 1417–1423.

Mulder, V.L., de Bruin, S., Schaepman, M.E., Mayr, T.R., 2011. The use of remote sensing insoil and terrain mapping - a review. Geoderma 162, 1–19.

Openshaw, S., 1983. The Modifiable Areal Unit Problem. Geo Books, Norwich.Pásztor, L., Szabó, J., Bakacsi, Z., 2010. Digital processing and upgrading of legacy data

collected during the 1:25000 scale Kreybig soil survey. Acta Geodaet. Geophys.Hung. 45(1), 127–136.

Pohl, C., van Genderen, J.L., 1998. Multisensor image fusion in remote sensing: concepts,methods, and applications. Int. J. Remote Sens. 19 (5), 823–854.

Pomerening, J.A., Cline, M.G., 1953. The accuracy of soil maps prepared by various methodsthat use aerial photograph interpretation. Photogramm. Eng. 19 (5), 809–817.

Robinson, A.H., Wallis, H.M., 1967. Humboldt’s map of isothermal lines: A milestone forthematic cartography. Cartogr. J. 4 (2), 119–123.

Roecker, S.M., Thompson, J.A., 2010. Scale effects on terrain attribute calculation and theiruse as environmental covariates for digital soil mapping. In: Boettinger, J.L., Howell,D.W., Moore, A.C., Hartemink, A.E., Kienast-Brown, S. (Eds.), Digital Soil Mapping:Bridging Research, Environmental Application, and Operation. Springer, pp. 55–66.

Rouse, J.W., Haas, R.H., Schell, J.A., Deering, D.W., 1973. Monitoring vegetation systems inthe great plains with ERTS. Third ERTS Symposium, NASA SP-351. I, pp. 309–317.

Schaetzl, R.J., Drzyzga, S.A., Weisenborn, B.N., Kincare, K.A., Lepczyk, X.C., Shein, K.A.,Dowd, C.M., Linker, J., 2002. Measurement, Correlation, and Mapping of Glacial LakeAlgonquin Shorelines in Northern Michigan. Ann. Assoc. Am. Geogr. 92, 399–415.

Schoeneberger, P.J., Wysocki, D.A., Benham, E.C., Soil Survey Staff, 2012. Field Book forDescribing and Sampling Soils, Version 3.0. U.S. Department of Agriculture, NaturalResource Conservation Service, Lincoln, Nebraska.

Please cite this article as: Miller, B.A., Schaetzl, R.J., The historical role o10.1016/j.geoderma.2014.04.020

Scull, P., Franklin, J., Chadwick, O.A., McArthur, D., 2003. Predictive soil mapping: a review.Prog. Phys. Geogr. 27 (2), 171–197.

Shchetenko, A.Y., 1968. The Most Ancient Agrarian Cultures of the Deccan [Drevneishiezemledel’cheskie kul’tury Dekana]. Nauka, Leningrad.

Simonson, R.W., 1952. Lessons from the first half century of soil survey: II. mapping ofsoils. Soil Sci. 74, 323–330.

Simonson, R.W., 1989. Historical highlights of soil survey and soil classification withemphasis on theUnited States, 1899–1970. International Soil Reference and InformationCentre, Wageningen, The Netherlands.

Smith, C.B., 1985. Air Spy. Am. Society for Photogrammetry Foundation, Falls Church,Virginia.

Soil Survey Staff, 1951. Soil Survey Manual. U.S. Department of Agriculture Handbook, No.18 (Washington, D.C.).

Soil Survey Staff, 1993. Soil Survey Manual. U.S. Department of Agriculture Handbook, No.18 (Washington, D.C.).

Theophrastus, 1916. Enquiry into Plants and Minor Odours and Weather Signs (Hort, A.,Trans.), W. Heinemann, London.

University of New Hampshire, 2013. William Smith’s Geological Map of England andWales and Part of Scotland, 1815–1817. Available online http://www.unh.edu/esci/WilliamSmiths-StrataIdentified/i/explanatory.html (Accessed [4/15/2013]).

USDA-NRCS, 2014. Geospatial Data Gateway. United States Department of Agriculture,Natural Resources Conservation Service (Available online: http://datagateway.nrcs.usda.gov. Accessed [3/6/2014]).

Veenstra, J.J., Burras, C.L., 2012. Effects of agriculture on the classification of Black soils intheMidwestern United States. Can. J. Soil Sci. 92, 403–411. http://dx.doi.org/10.4141/CJSS2010-018.

Volney, C.F., 1804. View of the Climate and Soil of the United States of America (Brown, C.B., Trans.), J. Conrad & Co. Philadelphia, Pennsylvania.

Wehr, A., Lohr, U., 1999. Airborne laser scanning - an introduction and overview. ISPRS J.Photogramm. Remote Sens. 54 (2–3), 68–82. http://dx.doi.org/10.1016/S0924-2716(99)00011-8.

Wheeler, G.M., 1885. Facts Concerning the Origin, Organization, Administration,Functions, History, and Progress, of the Principal Government Land and MarineSurveys of the World, Being Extracts from the Report on the Third InternationalGeographical Congress and Exhibition. Government Printing Office, Washington,D.C.

Whitney, M., 1900. Field Operations of the Division of Soils. U.S. Department of Agricul-ture, Bureau of Soils, Washington, D.C.

Whitney, M., 1909. Field Operations of the Bureau of Soils. U.S. Department of Agriculture,Bureau of Soils, Washington, D.C.

Widgren, M., 1979. A simulation model of farming systems and land use in Swedenduring the early Iron Age (c. 500 B.C.-550 A.D.). J. Hist. Geogr. 5 (1), 21–32.

Wilson, J.P., Gallant, J.C. (Eds.), 2000. Terrain Analysis: Principles and Applications. JohnWiley & Sons, New York.

Withers,M.A.,Meentemeyer, V., 1999. Concepts of scale in Landscape Ecology. In: Klopatek, J.M., Gardner, R.H. (Eds.), Landsc. Ecological Analysis: Issues and Applications. Springer,New York, pp. 205–252. http://dx.doi.org/10.1007/978-1-4612-0529-6_11.

Wood, J.D., 1996. Scale-based characterisation of digital elevation models. In: Parker, D.(Ed.), Innovations in GIS 3: Selected Papers from the Third National Conference onGIS Research. Taylor & Francis, London, UK, pp. 163–176.

Zaichikov, V.T., 1955. Voyagers of Ancient China and Geographic Studies in the People’sRepublic of China [Puteshestvenniki drevnego Kitaya I geograpficheskie issledovaniyav Kitaiskoi Narodnoi Respublike]xd. Geografgiz, Moscow.

f base maps in soil geography, Geoderma (2014), http://dx.doi.org/