Robert Wellman Campbell
Biology Department, Augustana College
December 1997

South Dakota Vegetation Terms on Maps

This paper describes how South Dakota's natural vegetation has been organized, in words and maps, by some map authors.

Please note that map labels' colors match their respective regions, to accommodate small regions with big labels. Map colors are not coordinated from map to map. The maps are Albers projection, at 1 km/pixel. The vegetation zones were made by tracing vectors over registered scans of the source maps.

Vegetation classifications

All the maps presented here (except the EPA ecoregions map, included here as a reference) show something which does not exist-- South Dakota covered by grass. Even in 1884, the date of the earliest map, the grasses of eastern South Dakota were being replaced by cropland, towns, and exotic species not shown in the maps. So all the classifications are subjunctive.

But the similarity mostly ends there. These maps are difficult to compare because they all use different classification systems. Most are of "natural" vegetation, left undefined by the author. It could mean the plants existing before European settlement, or the plants surviving into the settlement period, among the crops, pastures and towns.

Kuchler mapped what he called potential natural vegetation, defined as the result of instantaneous disappearance of people and succession to climax. This is unusual mainly just for being clear and explicit, rather than vague "natural vegetation". But potential vegetation has its problems. Geographer Philip Gersmehl has criticized it as being abstract and therefore not useful for management. It is also a snapshot of a moving target, just like a map of actual vegetation, since every species is evolving and every population is migrating or dying out. It may be that potential vegetation is both hard to determine and hard to use.

Some of these maps are floristic, some structural, and some both. A floristic classification is of dominant species; a structural classification is of what Kuchler calls "life forms" such as shortgrass, tallgrass, and forest. The floristic system is normally more detailed. Shantz 1923, SCS 1942, and Kuchler 1964 aggregate the floristic classes into structural categories. Kuchler's, though, are so broad (e.g. "grasses") that effectively he has a purely floristic system of Latin-named genera (listed below) with English labels added on for the sake of easy reference.

Vegetation classes

In his study of Kansas vegetation maps, Kuchler compiled a master legend containing all the classes of all the maps. This is not possible with this wider variety of maps, but below is a fairly comprehensive list of classes.

Forests (all):
   Forest (Aikman 1935)
   Coniferous forest (Sargent 1884)
   Conifers (Westin 1967)
   (not Great Plains) (Shantz 1923) 
   yellow pine - douglas fir (Shantz and Zon 1923)
   Black Hills (Aikman 1935, SCS 1942)
   Black Hills pine forest (Kuchler 1964, SCS 1976)
   Northern floodplain forest (Kuchler 1964)
   oak-hickory (Shantz and Zon 1923)
   Eastern ponderosa forest (Kuchler 1964)

Grasses (all structural classes):
   plains - prairies (Sargent 1884)
   shortgrass - tallgrass (Shantz 1923)
   shortgrass (plains) - tallgrass (prairie) (Shantz & Zon 1923)
   shortgrass plains - mixed prairie - true prairie (Aikman 1935)
   shortgrass - midgrass - tallgrass (SCS 1942)
   short & mid - mid & short - mid & tall - tall (Westin 1967)

Grasses (shared floristic classes)
   western wheatgrass (Shantz 1923, "wheatgrass" in SCS 1976)
   wheatgrass-needlegrass (Kuchler 1964, SCS 1976)
   wheatgrass-grama-buffalograss (Kuchler 1964, SCS 1976)
   sandhills prairie (Kuchler 1964, SCS 1976)
   wheatgrass-bluestem-needlegrass (Kuchler 1964, SCS 1976)
   bluestem prairie (Kuchler 1964, SCS 1976)

The structural classes provide little framework for agreement or disagreement. Even with the Black Hills, which are relatively simple and discrete, some authors label landforms, some life forms, and some taxons. Shantz mapped only the Great Plains and the SCS mapped only grass, so on these maps forests got grass labels or none at all.

Sargent, a forester, differentiated "treeless areas" by how treeless they were, rather than by describing the grasses present; prairies were less than 20% woodland, and plains were treeless. This is an example of the terminology objected to by James Malin, the Kansan ecological historian. Malin challenged terms such as subhumid and marginal by asking, "Marginal for what? Subhumid for what?" He said these terms were relative to the wet east, which could equally be considered "marginal" for cactus, "superhumid" for wheat, and, especially, "grassless".

Almost the only common floristic classes are between Kuchler and the later SCS map which is obviously based on Kuchler (perhaps to its credit). At least this floristic disagreement is real, factual disagreement about grass and not logical meta-grass disagreement about definitions.

Vegetation zones

The few structural classes common to two or more maps can be compared geographically.

"Tallgrass" appears on four of these maps. Two maps, SCS 1942 and Westin 1967, show tallgrass in only the extreme southeast portion of the state. The two other maps, both by Shantz, extend tallgrass to the western edge of the James River valley. This western limit agrees closely with the limit of Kuchler's two eastern classes, and it agrees fairly with the western limit of northern glaciated plains in EPA 1997. This might approach some loose consensus for a tallgrass border, but the SCS survey conducted in 1940 shows no transition anywhere near that border, and this survey appears to have been more detailed, systematic, focused on South Dakota and supported by real ground data. Perhaps, though, the Shantz-SCS difference is a reflection of real change between 1923 and 1940, drought having pushed the tallgrass east and south.

"Shortgrass" also appears on four maps. The two Shantz maps have a simple tallgrass/shortgrass border. SCS 1942 has pockets of tallgrass and shortgrass separated by a large expanse of "midgrass". (One of these pockets, labeled on SCS 1942 as "western wheatgrass - niggerwool - needle-and-thread", was reproduced in Kuchler 1964 and SCS 1976 as "wheatgrass - grama - buffalo grass".) The shortgrass zones are even more confused than the tallgrass, with only four maps to compare and no agreement between authors.

Non-grass areas of different types are also compared here. The authors vary in how many riparian forests they show, and how far up each river they map forest. Kuchler and especially Zon show the forests most fully. SCS 1976 has the most forest detail in the west, but shows no forest along rivers. The Black Hills appear on all maps, with modest variations.

Notes on the source maps

Sargent 1884

In Vegetation Mapping, Kuchler cited this map as the first of its kind, as well as the last for several decades. Nevertheless this map was very hard to obtain. It was missing from its portfolio in some libraries. My brother hunted it down for me and got a partial, color photocopy from the University of Minnesota Wilson Library, at the Government Documents reference desk. The library unfortunately ripped it during copying. There should be a reliable digital version of this map available somewhere.

This map has a simple legend with only four types; the only one not appearing in South Dakota is "Coniferous Forests". Prairies and the two forest types are distinguished only by shade of green, but the patch in northeastern South Dakota does indeed appear to be coniferous. In "The Vegetation of Kansas on Maps", Kuchler may have mislabeled a similar patch of coniferous as deciduous. This map also includes rivers and red "wooly worm" relief shading.

Shantz 1923

Shantz explicitly distinguished eight vegetation regions in South Dakota, which he categorized into tallgrass, shortgrass and sagebrush. No forest appears on this map. This map of the Great Plains also implicitly delineates two more regions of South Dakota as lying outside the Great Plains: the Black Hills, and the eastern margin of the state.

Shantz sought to map only climax natural vegetation. He disregarded grazed or cultivated vegetation, but also vegetation not on "well-developed loam soils". "Climax types of vegetation," he wrote, "occur only on older soils, those soils which come into equilibrium with the climatic conditions." (Shantz 1923, p. 87) But his map seems not to follow his article in this regard; he wrote that "the sand hills on the Great Plains are constantly tending toward the normal short-grass type of the region," but he mapped the Nebraska Sandhills as tallgrass.

Kuchler (1969), comparing vegetation maps of Kansas, praised Shantz' map as "an important refinement of our knowledge of the Kansas grasslands", showing detailed grass types which are still accepted. Shantz told Kuchler that the map manuscript was complete by 1912.

Shantz and Zon 1923

Shantz mapped the grass, Zon mapped the forests, and the U.S. Department of Agriculture published the map in its Atlas of American Agriculture. Both authors worked for the USDA, in the Bureau of Plant Industry and the Forest Service respectively.

This is a map of "natural vegetation". Text below the map briefly indicated the actual state of each vegetation region, by intensity of cultivation or clearing.

This map is surprisingly different from Shantz' earlier map. Zon's forests are added, in some detail: broad margins of oak-hickory along almost all rivers, and yellow pine - douglas fir on the Black Hills and scattered spots north. But Shantz' previous grass detail is gone; in contrast to Zon's floristic (i.e. species) and spatial detail, the grasses are simply divided into two broad structural regions, tallgrass and shortgrass, which do not match Shantz' earlier tallgrass and shortgrass regions. Kuchler (1988) criticized this map's inconsistent mix of floristic and physiognomic classes.

The loss of grass detail may or may not be surprising; this map has to cover 48 states, but it is at twice the scale of the earlier map.

Aikman 1935

This map was produced by the U.S. Forest Service, to show the zone where New Deal shelterbelts would be planted, and what vegetation existed in the six states the zone passed through. The shelterbelt zone was a 100-mile band running from central Texas to the Canadian border; over that extent it shifted slightly east or west of the "mixed prairie" as shown on this map. In the Dakotas it was shifted perhaps 50 miles east.

SCS 1942

This map, from the early days of the Soil Conservation Service, appears here as it was printed in a 1942 bulletin of the South Dakota Agricultural Experiment Station, Regrassing Areas in South Dakota. Other versions of this map may exist elsewhere. About the map the bulletin says:

"A carrying capacity survey was completed by the State Agricultural Experiment Station and the Soil Conservation Service in 1940, which formed the basis for the grassland type map (Fig. 1) and indicated the most important grasses and grasslike plants in each of 17 sub-areas."

Of all maps considered here, this map has by far the most floristic detail. The 1940 survey collected even more information, recording at least five dominant species for each region. The Black Hills are labeled only as "Black Hills Area", with no vegetation information. The 17 grass types are categorized physiognomically into tallgrass, midgrass and shortgrass.

This map deals only with native grasses, though the bulletin notes that Canada and Kentucky bluegrasses would be considered native, and sedges would be "discussed as grasses". The two regions of Missouri/tributary bottomlands are labeled only by their dominant natural grasses, though colored on the map as "River Breaks and Bottomlands".

Kuchler 1964

This popular map has been published several times since 1964. On the 1967 USGS version, Kuchler defined potential natural vegetation as the result of human removal and instantaneous succession. South Dakota is divided into eight classes, each labeled by 1-3 dominant genera. Each class is also given a descriptive English name.

Eastern ponderosa forest         Pinus
Black Hills pine forest          Pinus
Northern floodplain forest       Populus-Salix-Ulmus    
Wheatgrass-needlegrass           Agropyron-Stipa
Wheatgrass-bluestem-needlegrass  Agropyron-Andropogon-Stipa
Wheatgrass-grama-buffalo grass   Agropyron-Bouteloua-Buchloe
Bluestem prairie                 Andropogon-Panicum-Sorghastrum 
Sandhills prairie                Andropogon-Calamovilfa
Kuchler grouped these types into broad physiognomic units; in South Dakota these include grassland, western forests (Black Hills and the northwestern bit), and eastern broadleaf forests (river bottoms). Kuchler said these were not used as hierarchical classes, but only as a device to locate types quickly.

SCS 1976

This map is apparently from the Soil Conservation Service in Lincoln, Nebraska, made with and for the Old West Regional Commission. It is labeled 5,P-35,738, and a note reads, "Source: family of maps SCS DRWG no. 5,S-32,929 (4/74) and information from SRM/OWRC Advisory Board."

This map in fact appears to be Kuchler's map with the following changes:

EPA 1997

This map appears here as a reference, since its biotic/abiotic ecoregions are very different from the vegetation regions of all the other maps. Ecosystems are defined by the authors as "areas of general similarity in ecosystems and in the type, quality and quantity of environmental resources".

These are the Environmental Protection Agency's Level III ecoregions. Each of these regions is further divided into Level IV regions.


Maps included in this study

Aikman, J. M., 1935, Principal vegetative zones of the prairie-plains region: in Possibilities of shelterbelt planting in the Plains Region: Washington, U.S. Forest Service, p. 157, scale 1:9,000,000.

Environmental Protection Agency, 1997, Level III Ecoregions of the Conterminous United States. The linework shown here came from digital files on the EPA's FTP server.

Kuchler, August W., 1964, Potential natural vegetation of the conterminous United States: in U.S. Geological Survey, 1970, The national atlas of the United States of America: Washington, USGS, scale 1:7,500,000.

Sargent, C.S., 1884, Forests, prairie, and treeless regions of North America exclusive of Mexico: Washington, D.C., 10th Census of the United States, vol. 9, Atlas, map no. 1.

Shantz, Homer Leroy, 1923, The natural vegetation of the Great Plains region: Annals of the Association of American Geographers, v. 13, p. 83, scale 1:16,000,000.

Shantz, Homer Leroy and Raphael Zon, 1923, Natural vegetation of the United States: in U.S. Department of Agriculture, 1923, Atlas of American Agriculture: Washington, USDA, scale 1:8,000,000.

United States Soil Conservation Service, [1940], South Dakota Native Grass Map: in Franzke, C.J. and A.N. Hume, Regrassing areas in South Dakota: Brookings, S.D., South Dakota State College, Agr. Exp. Sta., Bull. no. 361, p. 5, scale 1:3,400,000.

United States Soil Conservation Service, 1976, South Dakota natural vegetation: in Baumberger, Rodney, 1977, South Dakota rangeland resources: Rapid City, S.D., Society for Range Management, Old West Regional Range Program, scale 1:2,150,000.

Westin, Frederick C., [1967], [Natural vegetation of South Dakota]: in Westin, Fred C., 1967, Soils of South Dakota: Brookings, S.D., Agronomy Dept., Agr. Exp. Sta., SDSU, scale [?].

Other references

Gersmehl, Philip J., 1980?, Maps in landscape interpretation: chapter in unknown volume, p. 79-115.

Kuchler, August W., ed., 1965-1970, International bibliography of vegetation maps: Lawrence, University of Kansas Libraries. Volume 1 North America.

Kuchler, August W., 1969, The vegetation of Kansas on maps: Transactions, Kansas Academy of Science, v. 72, no. 2, summer 1969, p. 141-166.

Kuchler, August W. and I. S. Zonneveld, eds., Vegetation mapping: Dordrecht, Netherlands, Kluwer, 1988. Historical sketch of U.S. vegetation maps, p. 4.

Malin, James C., Robert P. Swierenga ed., 1984, History and ecology; studies of the grassland: Lincoln, U. of Nebraska Press. "Grassland, 'Treeless' & 'Subhumid': A Discussion of Some Problems of the Terminology of Geography" p. 21.