To find more resources for your business, home, or family, visit the College of Agricultural, Consumer and Environmental Sciences on the World Wide Web at aces.nmsu.edu

The Desert Project—An Analysis of AridlandSoil-Geomorphic Processes Bulletin 798 H. Curtis Monger, Leland H. Gile, John W. Hawley, and Robert B. Grossman1

Agricultural Experiment Station • College of Agricultural, Consumer and Environmental Sciences

Department ofPlant and Environmental Sciences

INTRODUCTIONIn August 1957, R. V. Ruhe and L. H. Gile moved to Las Cruces, NM, to begin the Desert Soil-Geomor-phology Project (the Desert Project) of the U.S. Soil Conservation Service (SCS) (Appendix A). At that time, little was known about soils in arid and semiarid regions of the American Southwest away from alluvial valley floors. This study was undertaken to learn more about the morphology, classification, genesis, and occurrence of desert soils and their relation to late Cenozoic land-scape evolution, and to assist in understanding, classify-ing, and mapping soils in similar geomorphic settings elsewhere. The hallmark of this investigation was the integration of classic pedologic and geologic approaches in the study of a desert and semidesert soil-geomorphic system. Another distinctive feature of the project was the large number of formal and informal field seminars held in the study area. Hundreds of scientists and stu-dents from many parts of the world participated in these programs, and a series of field trip guidebooks evolved over a period of 46 years (1961–2007). Desert Project research between 1957 and 1972 is described in detail in Hawley (1975a, 1975b), Gile and Grossman (1979), and Gile et al. (1981).

Guy D. Smith, Director of the SCS’s Soil Survey Investigations Division, provided overall administrative and technical supervision of the Desert Project until his retirement in 1972. R. V. Ruhe conducted initial geologic and geomorphic studies as a full-time partici-pant from August 1957 to August 1960, and during short-term assignments several times a year through 1965 (Ruhe, 1967). F. F. Peterson was responsible for geomorphic investigations and some aspects of soils research from 1960–1962. J. W. Hawley conducted

Introduction ...................................................................... 1Setting .....................................................................................................3

Physiography .......................................................................................3

Climate ...............................................................................................4

Vegetation.. .........................................................................................6

Landforms and Stratigraphy .............................................. 6Parent materials, basins, and river valley...................................................6

Cenozoic stratigraphy and the ancestral Rio Grande ................................6

Hydrogeology ..........................................................................................10

Atmospheric Additions ...................................................... 10Geomorphic Surfaces and Morphostratigraphic Units ...... 10

Geomorphic surfaces as a mapping concept .............................................10

Morphostratigraphic units as a mapping concept .....................................11

The valley border stepped sequence .....................................................11

Effect of landscape dissection on geomorphic surfaces and soils ...............11

Constructional surfaces versus structural benches ....................................15

Paleoclimate ....................................................................... 17Late Quaternary climates .........................................................................17

Cyclic entrenchment of the Rio Grande ...................................................17

Cyclic sedimentation and buried soils of the fan piedmont ......................17

The Gardner Spring radiocarbon site ...................................................17

Polygenetic soils .......................................................................................23

Soil Chronology................................................................. 23Land survey notes ....................................................................................23

Buried charcoal ........................................................................................25

Radiocarbon ages of pedogenic carbonate ................................................25

Radiocarbon differences between carbonate and organic carbon ..............26

Volcanic ash .............................................................................................30

Paleontology ............................................................................................30

Relative ages based on pedogeomorphic tracing and soil development .....30

Soils and Soil-Geomorphic Relations ................................ 30Soil horizons and conventions .................................................................30

Soil taxonomy ..........................................................................................32

Horizons of carbonate accumulation........................................................32

Origin .................................................................................................32

The morphogenetic sequences and stages of carbonate accumulation ..32

Effect of increasing precipitation on development of the sequences .....35

The calcic and petrocalcic horizons .....................................................36

Micromorphology of carbonate accumulation .....................................37

The K horizon .....................................................................................37

1Respectively, Professor, Department of Plant and Environmental Sciences, New Mexico State University, Las Cruces; Soil Scientist (retired), Natural Resources Conservation Service; Senior Environmental Geologist (retired), New Mexico Bureau of Mines and Mineral Resources; and Soil Scientist (retired), Natural Resources Conservation Service.

Table of Contents

Bulletin 798 • Page 2

geologic, geomorphic, and hydrogeologic studies from 1962–1972, when SCS-sponsored project activities for-mally ended. L. H. Gile was in charge of soil research for the duration of the project. Field research was con-ducted in close collaboration with the SCS Soil Survey Laboratory staff, particularly R. B. Grossman, then head of the Lincoln (NE) Regional Laboratory. Other signifi-cant collaborators included W. C. Lynn, J. Cady, and R. C. Vanden Heuvel, who conducted studies in clay min-eralogy; F. E. Kottlowski, W. R. Seager, and W. E. King, who conducted studies in geology and hydrogeology; and C. H. Herbel, A. L. Metcalf, and C. E. Freeman, who conducted studies in ecology and paleoecology.

The 1,024 km2 (400 mi2) site for the Desert Project (Figure 1) was selected to be located in the Rio Grande Valley area of the southeastern Basin and Range Prov-ince because it fulfilled many of the requirements of the study. First, this part of southern New Mexico is typi-cal of large parts of the Southwest in terms of terrain, parent material, soil age, and general climatic history. Second, distinctly different soils formed on various landforms and parent materials. For example, the area contains a river valley as well as intermontane basins with internal surface drainage, which meant that soils and landscapes of each could be compared. In addi-tion, the area has both non-calcareous parent materials (e.g., rhyolite) and highly calcareous parent materials (e.g., limestone). Third, the area contains both semiarid mountains and arid basins, thus making it possible to study the effects of climatic and vegetation differences on soils and landscapes of the same age. Fourth, the area contains soils that range widely in age, from less than 100 years old to more than 2,000,000 years old. Pedo-logically, the Desert Project provided a place to study the effects of the five soil-forming factors and interac-tions between them. In addition, the region contained a land-grant university, New Mexico State University, where the project could be given library access, labora-tory space, and offices (Ruhe, 1967; Hawley, 1975b).

In this report, we use soil-geomorphic maps, cross sections, three-dimensional diagrams, tables of soil characterization data, and photographs to illustrate a variety of soil-geomorphic features in the Desert Project. Because of limited space, much information has been omitted, including additional soil-geomorphic features, such as those in the semiarid mountains as well as else-where in the arid zone, and laboratory data for some of the illustrative sites. Instead, reference is made to publi-cations where this information can be found. References for the Desert Project as a whole are The Desert Project Soil Monograph (Gile & Grossman, 1979); Soils and Geomorphology in the Basin and Range Area of Southern

Totals of pedogenic carbonate in soils ................................................37

Horizons of silicate clay accumulation .....................................................38

Evidence of clay illuviation and the argillic horizon .............................39

Micromorphology of the argillic horizon .............................................41

Obliteration of the argillic horizon ......................................................41

Horizons of organic carbon accumulation and soils of the semiarid mountains ...........................................................................41

Soils and soil-geomorphic boundaries ......................................................42

Boundaries caused by development of the argillic horizon ...................42

Boundaries caused by differences in soil age and carbonate content of parent materials (Figure 27) ..........................................................47

Difference in age (high-carbonate parent materials); boundary between Torrifluvents and Haplocalcids (Figure 27B) ....................47

Difference in age (low-carbonate parent materials); boundary between Haplocambids and Calciargids (Figure 27C)—initial development of the cambic and argillic horizons ............................47

Difference in carbonate content of Holocene parent materials; boundary between Haplocambids and Torrifluvents .....................49

Difference in carbonate content of late Pleistocene parent materials; boundary between Calciargids and Haplocalcids (Figures 27B, 27C)—initial development of the argillic horizon in high-carbonate parent materials ....................................49

Boundaries caused by differences in soil age and particle size (low-carbonate parent materials) (Figures 29A, 29B) .........................49

Difference in soil age; boundary between coarse-loamy Haplargids and fine-loamy Calciargids (Figure 29B) ......................49

Difference in particle size and age; boundary between coarse-loamy Haplargids and loamy-skeletal Calciargids (Figure 29B) ..................................................................................52

Boundaries caused by obliteration of the argillic horizon .....................52

Boundaries caused by landscape dissection ..........................................52

Difference in dissection; boundary between Argids and Calcids without argillic horizons ....................................................54

Difference in dissection; boundary between Calcids and Orthents (Figure 30A) ...................................................................54

Boundaries caused by differences in landscape dissection and soil moisture .....................................................................................54

Difference in dissection; boundary between Petroargids and Torriorthents ...........................................................................54

Difference in soil moisture; boundary between Haplargids or Calciargids and Argic Petrocalcids (Figure 15) ...........................55

Boundaries caused by differences in faunal activity (Figure 32) ............55

Related Research and Research Subsequent to Formal Closing of the Desert Project in 1972 ...........................55

Concluions and Dissemination of the Research ...................................59

References ...............................................................................................59

Acknowledgement ..................................................................................67

Appendix A–Map ...................................................................................69

Appendix B–Desert Project Tour, 2000 ................................................70

Appendix C–A Pilgrimage to the Desert Project ..................................71

Bulletin 798 • Page 3

physiographic subdivisions for southern New Mexico, northern Mexico, and West Texas (Figure 2). In this physiographic classification, the Desert Project lies in the Mexican Highlands Section of the Basin and Range Province. It is a region whose terrain is largely the prod-uct of late Cenozoic landscape evolution, consisting of broad desert basins with local volcanic fields, and nar-row mountain ranges created by the Rio Grande Rift tectonic system since the mid-Tertiary period (Hawley, 1978; Seager & Morgan, 1979). Many of the desert basins are sites of Quaternary pluvial lakes, dune fields, and basaltic volcanoes and lava flows (Hawley & Kot-tlowski, 1969; Seager et al., 1987). The only through-flowing drainage is the Rio Grande system that occupies valleys entrenched well below the basin floors. Eleva-tions in the Desert Project range from 9,012 ft (2,747

New Mexico—Guidebook to the Desert Project (Gile et al., 1981); Supplement to the Desert Project Soil Mono-graph, Volume I (Gile & Ahrens, 1994); Supplement to the Desert Project Guidebook, with Emphasis on Soil Micromorphology (Gile et al., 1995b); Supplement to the Desert Project Soil Monograph, Volume II (Gile & Ahrens, 1996); Supplement to the Desert Project Soil Monograph, Volume III (Gile et al., 2003); and A 50th Anniversary Guidebook for the Desert Project (Gile et al., 2007).

Setting

PhysiographyBuilding on the work of geologists and geographers investigating the U.S.-Mexico border region since the mid-19th century, Hawley (1969, 1975a) defined the

Figure 1. Block diagram of the 400 mi2 Desert Project area showing major landforms and locations of the Gardner Spring (1), Isaacks’ (2), and Shalam Colony (3) radiocarbon sites. Vicinities I through V on the diagram locate regions for the dis-cussion of soil boundaries (modified from Gile, 1975a, used with permission).

Bulletin 798 • Page 4

m) in the Organ Mountains to 3,870 ft (1,180 m) in the Rio Grande floodplain.

Landform names for the intermontane basin and river valley regions in the Desert Project are shown in Figure 3. The intermontane basin, or bolson, consists of mountain uplands, piedmont slopes, and basin floors. The entrenched river valley consists of valley rims, struc-tural benches, slopes of stepped sequences of geomor-phic surfaces, inner-valley scarps, and the valley floor (Figure 3).

ClimateBoth arid and semiarid climates occur in the Desert Project area. In the lower elevations of the Rio Grande valley and basin floors, the annual precipitation is about

20 cm (Table 1). Short-term records of precipitation at several places in the mountains suggest that above an elevation of about 5,000 ft (1,524 m) the climate is semiarid, with precipitation ranging from approximately 25 to 40 cm (Gile et al., 1981). Annual evaporation is approximately 10 times the annual precipitation.

Soils and vegetation also reflect a change from arid to semiarid conditions. Soils above approximately 5,000 ft have darker and thicker A horizons and have deeper zones of clay and carbonate accumulation (Gile, 1977). Vegetation above about 5,000 ft increases in density and contains vegetation characteristic of higher elevations, such as blue grama (Gile et al., 1981, p. 19).

Figure 2. Physiographic subdivisions of the Desert Project region of southern New Mexico, West Texas, and northern Chi-huahua, showing major stream systems, pluvial lake basins, dune fields, and basalt terranes (modified from Hawley, 1975a, used with permission; see also Hawley, 1986).

Bulletin 798 • Page 5

Figure 3. Profiles of landforms in the Desert Project. Top profile is of intermontane landforms; bottom profile is of river valley landforms (from Gile et al., 1981, used with permission).

Table 1. Average Precipitation in the Rio Grande Floodplain of the Valley Border Near Las Cruces (Elevation 3,881 ft) and Boyd’s Ranch in Organ Mountains (Elevation 6,200 ft) (from Gile et al., 1981) (values are in inches except for centimeters in parentheses, annual column.)Jan. Feb. March April May June July Aug. Sept. Oct. Nov. Dec. Annual

Valley border (University Park, 1892–1966)

0.36 0.41 0.36 0.21 0.30 0.59 1.47 1.70 1.22 0.71 0.44 0.49 8.26

(21)

Valley border (University Park, 1948–1957)

0.37 0.70 0.35 0.12 0.24 0.35 1.31 1.02 0.58 0.84 0.12 0.29 6.18

(16)

Mountain fronts (Boyd’s Ranch, 1948–1957)

0.82 0.82 0.37 0.34 0.49 0.57 3.29 1.71 0.83 1.68 0.25 0.34 11.49

(29)

Bulletin 798 • Page 6

Vegetation Grasses, trees, and forbs observed in the Desert Proj-ect are listed in Table 2. Shrubs are dominant in the arid part of the study area, and the most common are creosotebush, mesquite, tarbush, Mormon tea, ocotillo, snakeweed, whitethorn, ratany, mariola, four-wing salt-bush, prickly pear, Yucca elata, and Yucca baccata. Des-ert willow, brickellbush, Apache plume, burro brush, sumac, and mesquite are common in arroyos. Most of the shrubs listed above also occur at higher elevations (5,000–6,000 ft), along with catclaw, century plant, mountain lilac, indigo bush, sotol, turpentine bush, cholla, barrel cactus, juniper, squawbush, and a few piñon at the highest elevations.

Grasses in the arid zone occur only in scattered ar-eas where moisture conditions are favorable. They are mainly three-awn, tobosa, burro grass, fluffgrass, black grama, bush muhly, and dropseed. These grasses also occur at elevations above about 5,000 ft, where sideoats grama and blue grama are also found.

LANDFORMS AND STRATIGRAPHY

Parent materials, basins, and river valleyAs with many desert regions, dust is a constant source of new parent material for soil formation (Reheis et al., 1995). In the Desert Project, the average yearly ac-cumulation of dust was 23 g/m2 from 1962–1972 (Gile & Grossman, 1979). Although dust plays an important role in soil formation, it provides only minor volumes of parent material compared to alluvium derived from local mountain bedrock and deposits of the ancestral Rio Grande.

Local mountain bedrock is the source of a variety of residual, colluvial, and alluvial parent materials. The major bedrock types are rhyolite, andesite, monzonite, and marine carbonate rocks mainly composed of lime-stone and dolostone with interbedded cherts and clastic rocks comprising shale, siltstone, and sandstone (Seager et al., 1987).

The bedrock has reacted differently to weathering, transport, and deposition (Ruhe, 1964). Rhyolite resists comminution, and alluvium derived from it tends to be gravelly. In contrast, monzonite rounds in short dis-tances and displays a systematic decrease in size propor-tional to greater distances from its source (Ruhe, 1967). Consequently, soils developed in monzonite alluvium are low in gravel, except for steeper slopes. Limestone alluvium tends to be high in silt and clay (Gile et al., 1981, p. 63).

Deposits of the ancestral Rio Grande, beginning in the Rocky Mountains of southern Colorado and flowing

southward through New Mexico, contain a mixture of resistant, dominantly silicic, rounded, pebble-size rocks derived from distant and local sources. Clasts of marine limestones and other carbonate rocks are rare. Sands are quartz-rich, feldspathic, and usually free of a silt-clay matrix (Gile et al., 1981, p. 35). These non-calcareous deposits are the parent materials of soils formed on ba-sin floors and along the valley border (Figure 1).

In the intermontane basins, the basin floors are bounded by piedmont slopes that consist of alluvial fans and coalescent fan piedmonts descending from the mountains (Ruhe, 1964; Gile et al., 1981; Peterson, 1981). Distinct fans separated by interfan valleys exist in the upper piedmont. Downslope, alluvial fans spread out and coalesce laterally as fan piedmonts that descend to basin flats. Buried soils are extensive in this deposi-tional environment (Gile & Hawley, 1966). Toward the basin, sediments and their soils thin and in places merge as one thick, strongly developed soil (Ruhe, 1964).

Unlike the intermontane basins, which are aggrad-ing, the Rio Grande river valley is entrenching. This entrenchment has not been continuous, but has been punctuated by cyclic periods of downcutting and partial backfilling (Hawley, 1965; Metcalf, 1967, 1969; Hawley & Kottlowski, 1969). As a result, the river valley has a stepped sequence of surfaces that “ascends like a stair-case” (Ruhe, 1967, p. 34). Although the intermontane basin system has aggraded while the river system has cut a valley, both have experienced punctuated periods of sedimentation from upslope sources. As a result, both the river valley and intermontane basins have a similar series of geomorphic surfaces.

Cenozoic stratigraphy and the ancestral Rio GrandePapers by Ruhe (1962) and Hawley (1965, 1975a) dis-cussed the genesis of landforms in the context of late Cenozoic evolution of the Rio Grande basin, integrat-ing conceptual models of previous workers, notably Lee (1907), Dunham (1935), Bryan (1938), Kottlowski (1958), and Strain (1959). In particular, the formation of structural basins of the Rio Grande Rift is recognized to be the dominant feature in the Cenozoic record (Hawley, 1978; Seager & Morgan, 1979). The forma-tion of these basins began in the late Tertiary period, and they have continued to fill with detritus from sur-rounding slopes (Hawley et al., 1969). The detritus con-sist of piedmont slope alluvium, basin floor alluvium, stream channels, lacustrine deposits, and eolian deposits.

By Pliocene time, however, sediment derived from southern Colorado to central New Mexico was being delivered to the Desert Project region by the ancestral Rio Grande. During the past 4.5 Ma, this fluvial system

Bulletin 798 • Page 7

deposited the fluvial, deltaic, and lacustrine sedimentary sequence that forms the uppermost (Plio-Pleistocene) basin-fill of southern New Mexico, northern Mexico, and West Texas (the Camp Rice Formation of Strain [1966] and Hawley et al. [1969]). All late Cenozoic basin-fill deposits that predate early Pleistocene en-trenchment of the present river valley system comprise the Santa Fe Group (Hawley, 1978, charts 1 and 2). Ruhe (1962, 1964, 1967) did not recognize the sys-tem as conceived by subsequent workers, but he did describe a widespread basin-fill unit of fluvial and/or lacustrine derivation that spilled over from basin to ba-

sin (Ruhe, 1962). To Hawley et al. (1969), the ancestral Rio Grande system was a time-transgressive system of aggrading distributary channels with shifting loci of deposition that prograded southward into closed-basin (bolson-plain) areas of Chihuahua, Mexico, and West Texas.

The question of when the ancestral Rio Grande sys-tem stopped aggrading and started entrenching has been an intriguing and important question throughout the Desert Project. Maps of ancestral Rio Grande sediments suggest that the river system fed large lakes in southern New Mexico, northern Mexico, and West Texas until a

Perennial grassesAristida divaricata .....................................................Three-awnAristida pansa ...........................................................Three-awnBouteloua curtipendula ..............................................Sideoats gramaBouteloua eriopoda ....................................................Black gramaBouteloua gracilis .......................................................Blue gramaBouteloua hirsuta ......................................................Hairy gramaEnneapogon desvauxii ................................................Spike pappusgrassEragrostis spp. ...........................................................LovegrassHilaria mutica ..........................................................Tobosa grassLeptochloa dubia .......................................................SprangletopMuhlenbergia emersleyi ..............................................BullgrassMuhlenbergia porteri .................................................Bush muhlyPanicum obtusum ......................................................Vine mesquiteScleropogon brevifolius ...............................................Burro grassSetaria macrostachya ..................................................BristlegrassSporobolus airoides.....................................................Alkali sacatonSporobolus cryptandrus...............................................Sand dropseedSporobolus flexuosus ...................................................Mesa dropseedStipa eminens ............................................................Needle grassTrichachne californica ................................................CottontopTridens muticus .........................................................Slim tridensTridens pulchellus ......................................................Flufgrass

Annual grassesAristida adscensionis ..................................................Three-awnBouteloua barbata .....................................................Six weeks grama

ForbsAllionia incarnata .....................................................Trailing four o’clockAstragalus allochrous ..................................................MilkvetchAthysanus pusillus ......................................................MustardBahia spp. ................................................................Wild chrysanthemumBaileya pleniradiata ...................................................Desert marigoldDithyrea wislizeni......................................................Spectacle-podEriogonum abertianum ..............................................Desert buckwheatPectis angustifolia ......................................................Fetid marigoldPerezia nana .............................................................Desert hollyPhacelia spp. .............................................................Scorpion weedSalsola kali ................................................................Russian thistleVerbena spp. .............................................................VervainVerbesina encelioides ..................................................Golden crown-beard

Shrubs and treesAcacia constricta ........................................................WhitethornAcacia greggii ............................................................CatclawAgave palmeri............................................................Century plantArtemesia filifolia ......................................................Sand sageAtriplex canescens ......................................................Four-wing saltbushBacharis pteronoides ...................................................Yerba de pasmoBrickellia laciniata ....................................................BrickellbushCeanothus greggii .......................................................Mountain lilacCeltis reticulata .........................................................Desert hackberryCroton corymbulosus ..................................................CrotonChilopsis linearis ........................................................Desert willowColdenia canescens .....................................................Shrubby coldeniaCondalia lycioides ......................................................BuckthornCondalia spathulata ..................................................Mexican crucilloDalea formosa ...........................................................Indigo bushDalea scoparia ...........................................................BroomdaleaDasylirion wheelerii ...................................................SotolEchinocactus wislizenii ...............................................Barrel cactusEphedra torreyana ......................................................Mormon teaEphedra trifurca ........................................................Mexican teaEurotia lanata ...........................................................WinterfatFallugia paradoxa ......................................................Apache plumeFlourensia cernua ......................................................TarbushFouquieria splendens ..................................................OcotilloGutierrezia sarothrae .................................................SnakeweedHaplopappus laricifolius .............................................Turpentine bushHelianthus ciliaris .....................................................BlueweedHolacantha emoryi ....................................................Crucifixion thornHymenoclea monogyra ...............................................Burro brushJuniperus monosperma ...............................................JuniperKoeberlinia spinosa ....................................................Crucifixion thornKrameria parvifolia ...................................................RatanyLarrea tridentata .......................................................Creosote bushLippia wrightii ..........................................................OreganilloLycium berlandieri.....................................................Desert thornNolina microcarpa .....................................................BeargrassOpuntia spp. .............................................................Cholla, prickly pearParthenium incanum .................................................MariolaPinus edulis ...............................................................PiñonProsopis glandulosa ....................................................MesquiteQuercus spp. .............................................................OakRhus microphylla .......................................................SumacRhus trilobata ............................................................SquawbushSenecio filifolius .........................................................Threadleaf groundselYucca baccata ............................................................YuccaYucca elata ................................................................YuccaZinnia pumila ...........................................................Desert zinnia

Table 2. Scientific and Common Names of Perennial Grasses, Annual Grasses, Forbs, Shrubs, and Trees in the Desert Project Region (from Gile et al., 1981)

Bulletin 798 • Page 8

lower river system, which extended northward through the bolson region between the Big Bend and El Paso, became integrated with the ancestral upper Rio Grande in late early Pleistocene time. A complex process involv-ing upper basin geomorphic changes, climatic factors, and tectonic deformation resulted in partial drainage of preexisting lake basins, valley incision, and establish-ment of through-going Rio Grande drainage from Colo-rado to the Gulf of Mexico (Bryan, 1938; Kottlowski, 1953; Hawley & Kottlowski, 1969). Ruhe (1962) rec-ognized that entrenchment must have occurred during or after “Kansan time” based on the presence of Cuviero-nius (short-jawed mastodon) and Equus (horse) in sedi-ments of the youngest basin-fill sequence. These faunas are now recognized as dating back to early Pleistocene (i.e., pre-“Kansan time”) (Morgan et al., 1998).

Later, the Lava Creek B ash (0.66 Ma; Reynolds & Larson, 1972; Izett, 1981; Izett et al., 1992) was dis-covered in the Jornada Basin, just north of the Desert Project (Figure 2). This ash appeared to be stratigraphi-cally beneath the La Mesa surface, which is the surface that marks the end of Rio Grande aggradation (Hawley et al., 1969). However, additional mapping showed that the Lava Creek B ash was inset below the La Mesa sur-face in the Jornada Basin, and valley entrenchment there must have begun before 0.66 Ma. The Bishop ash (0.76 Ma; Sarna-Wojcicki & Pringle, 1992) occurs beneath sediments inset against La Mesa sediments at Rincon Arroyo, a tributary of the Rio Grande in the Jornada Basin. It is not certain whether the ash is in the inset alluvium or in the La Mesa sediment; in either case, downcutting of the Rio Grande began very near or at the Matuyama-Brunhes boundary at 0.78 Ma (Mack et al., 1998).

Three distinct ages of La Mesa surface occur in the Desert Project (Figure 1, Table 3), where magneto-stratigraphy at upper La Mesa (the oldest of the three) indicates that the soils there are about 2,000,000–2,500,000 yr old (Mack et al., 1993). Magnetostratig-raphy for an area of lower La Mesa (the youngest of the three) at the Desert Project has been identified as Matuyama (reversed polarity; Mack et al., 1998). South of the Desert Project, another area of lower La Mesa has been identified as Brunhes (normal polarity; Vanderhill, 1986). The true age of lower La Mesa could therefore be close to 780,000 yr, the boundary between Brunhes and Matuyama. This age also agrees with the approxi-mate time of valley entrenchment at Rincon Arroyo, in the Jornada Basin north of the Desert Project (Mack et al., 1998). This would initiate soil development in the abandoned lower La Mesa floodplain at about that time.

The third area of La Mesa in the Desert Project is JER La Mesa (Figure 1). As will be discussed, JER La Mesa is intermediate in age between upper and lower La Mesa. JER La Mesa is very extensive in the Jornada Basin, north of the Desert Project, and was so named because so much of the Jornada Experimental Range (JER) occurs on it.

Paleomagnetic studies by Mack et al. (1993) indicate that La Mesa in the Jornada Basin (JER La Mesa) ranges from 780,000 to 900,000 yr old. However, several fac-tors indicate that parts of JER La Mesa are much older than this. For reasons discussed below, pedogenic car-bonate in soils of lower La Mesa is good for comparison with carbonate in soils of JER La Mesa. Two soils at level sites on lower La Mesa average 1,080 kg/m2 of pe-dogenic carbonate. Using this value and the estimated age of 780,000 yr for lower La Mesa gives an average accumulation rate of 1.4 kg/m2/1,000 yr.

In contrast to an average of 1,080 kg/m2 of total carbonate for lower La Mesa, a soil (pedon 95-4) with dated pumice on JER La Mesa has 1,861 kg/m2 of total carbonate. Quoting from Gile (2002, p. 6):

Applying the same carbonate accumulation rate as for soils of lower La Mesa (1.4 kg/m2/1000 yr) gives an estimated age of about 1,300,000 yr. This is much older that the upper limit of 900,000 yr assigned for JER La Mesa by Mack et al. (1993). Pedon 95-4 also has the distinctive stage V morphology, common in JER La Mesa but seldom occurring in lower La Mesa, which is dominated by stages III and IV.

An estimated age of about 1,300,000 years for pe-don 95-4 would not be unreasonable because pumice dated at 1,600,000 yr occurs at a depth of only 3 m (Mack et al., 1996). However, if sedimentation above 3 m continued at about the same rate as for the thick layer with pumice, pedon 95-4 could be close to 1,600,000 yr old. The accumulation rate for this soil was calculated to check this. Using an age of 1,600,000 years and 1861 kg/m2 of pedogenic carbonate, the accumulation rate is 1.2 kg/m2/1000 years. This is very close to the accumulation rate of 1.4 kg/m2 for lower La Mesa, and is strong evidence that the soil at the pumice site is about 1,600,000 yr old.2

In addition, soils of several ages with stage V mor-

phology occur along and above fault scarps. Some of these soils may be older than 1,600,000 yr, and an over-all age range of 780,000–2,000,000 yr has tentatively

2Italicized text indicates direct quotation from cited work. Bracketed material refers to material in this report (e.g., figures and tables) or updates of terminology.

Bulletin 798 • Page 9

Table 3. Geomorphic Surfaces and Carbonate Accumulation in Soils of the Valley Border, Piedmont Slope, and Basin Floor North of Highway 70

Geomorphic surface† and carbonate accumulation Carbonate stage

(kg/m2)

Nongravelly Gravelly Estimated soil age Valley border Piedmont slope Basin floor materials materials (years B.P. or epoch)

Coppice dunes Coppice dunes Whitebottom Historical (since 1850 A.D.)

Lake Tank present to 150,000

Middle to late Holocene

Fillmore (5) Organ (8–20) 0, I I 100–7,000

III I I 100(?)–1,000

II I I 1,100–2,100

I I I 2,200–7,000

Leasburg (23–186) Isaacks’ Ranch II II, III Latest Pleistocene

(22–108) (10,000–15,000)

Butterfield (111) Baylor III III Late Pleistocene

(15,000–100,000)

Picacho (220) Jornada II Petts Tank III III, IV Late to middle Pleistocene

(213–300) (100,000–250,000)

Tortugas Modoc III IV Late middle Pleistocene

(250,000–500,000)

Jornada I Jornada I Jornada I III IV Middle Pleistocene

(751, 834) (795–1,080) (500,000–700,000)

Doña Ana IV > 700,000

Buried surfaces and soils 700,000–2,000,000

Lower La Mesa (992–1,168) III, IV Middle to early Pleistocene

(780,000)

JER La Mesa (1,861, 2,296) IV, V Early Pleistocene to Late Pliocene

(780,000–2,000,000)

Upper La Mesa V Late Pliocene

(2,000,000–2,500,000)

†Geomorphic surfaces after Ruhe (1967), Hawley and Kottlowski (1969), Gile et al. (1981), and Gile (2002). Materials genetically related to constructional phases of a geomorphic surface are designated by the geomorphic surface name (e.g., Fillmore alluvium; Hawley & Kottlowski, 1969). Lower and upper La Mesa and JER La Mesa are not formally considered a part of the valley border, but are included here because they form part of a stepped sequence with the valley border surfaces. Coppice dunes have not been formally designated a geomorphic surface, but are considered separately here because of their extent and significance to soils of the area. Buried surfaces and soils refer to surfaces and soils that are stratigraphically between the Jornada I soil and alluvium of the ancestral Rio Grande, north and south of Tortugas Mountain. Numbers after the surface names are single values or ranges of values of totals of pedogenic carbonate (kg/m2) in soils of the indicated geomorphic surfaces, from Gile et al. (1981), Monger et al. (1991), and Gile (1993, 1994, 1995). The true value for pedogenic carbonate in the late Picacho pedon would be greater than 111 kg/m2 because the pedon is on a ridge crest that has undergone some erosion (see Gile & Grossman, 1979, p. 331–338 for discussion). Values for JER La Mesa are from soils north of the Desert Project, in the Jornada Experimental Range. Carbonate stages after Gile et al. (1966) and Birkeland et al. (1991). Morphologies are best expressed where “nongravelly” soils contain less than about 20% by volume of gravel and “gravelly” soils contain more than about 60% by volume of gravel. Soils that have between 20% and 60% by volume of gravel have intermediate morphologies. Soils of the Picacho and lower La Mesa surfaces illustrate initial development of the stage IV plugged and laminar horizons in gravelly and nongravelly materials, respectively.

Bulletin 798 • Page 10

been assigned for these soils (Table 3). Thus, a combina-tion of paleomagnetism, dated pumice, carbonate mor-phology, and totals of pedogenic carbonate indicate that JER La Mesa is intermediate in age between upper and lower La Mesa at the Desert Project.

The ancient Rio Grande once flowed east of the Doña Ana Mountains (Figure 1). The river sediments of JER La Mesa, although buried in places, may be con-tinuously traced southward from the Jornada Basin to its occurrence on the east side of the valley in the Desert Project (Figure 1).

HydrogeologyThe structural basins and their sediments have recently taken on great societal importance because of the groundwater they contain, which is the main source of water for El Paso; Ciudad Juárez, Mexico; and Las Cru-ces (Hawley et al., 1969; Peterson et al., 1984). Ground-water yield, water table configuration, and groundwater quality are linked to the Cenozoic evolution of each particular basin (King et al., 1971; King & Hawley, 1975). For instance, clay strata, which may have formed as playa sediments, are now low-permeability zones for groundwater supply (Hawley et al., 1969). In contrast, well-sorted fluvial sands deposited by the ancestral Rio Grande have high transmissivities and high storage coef-ficients. In terms of water quality, gypsum deposits that formed in ancient playa lakes now degrade water quality. Fault zones that were once paths of high conductivity commonly become plugged with cement and become aquicludes.

Carbonate deposited from laterally moving ground-water is intimately mixed with pedogenic carbonate in some areas. However, combined studies of geomorphic history of the valley incision and hydrogeology of basin and valley fills show that high water table (phreatic and capillary-fringe) processes have not contributed to pe-dogenic carbonate in the Desert Project area except for very local sites of seeps and springs and beneath present and former floodplains of the Rio Grande.

ATMOSPHERIC ADDITIONSIn the pursuit to understand the origin of carbonate horizons, it was essential that the source of calcium in CaCO

3 be determined. Chemical weathering as a source

of calcium was considered unlikely. Ruhe (1967, p. 57) calculated that there was insufficient calcium in rhyo-lite alluvium parent material to account for the large amount of calcium in calcic or petrocalcic horizons formed in the rhyolitic parent material. Moreover, even if the rhyolitic parent material had contained enough calcium, it was only slightly weathered.

After the groundwater and chemical weathering hy-potheses were excluded as sources of calcium (see later section on carbonate horizons), atmospheric additions emerged as the most plausible hypothesis (Gile et al., 1966). To test this hypothesis, dust traps were set up across the Desert Project. Ten years of dust measure-ments revealed that calcareous dust fell ubiquitously on the landscape, ranging from 0.2 g/m2/yr in a grassy ba-sin floor area to 1.1 g/m2/yr in a sandy bare-ground area (Gile & Grossman, 1979, p. 82). The calcareous dust, combined with water-soluble calcium extracted from the dust, yielded a range from 0.35–1.3 g/m2/yr for all traps. If one trap with high carbonate content related to strongly calcareous surface horizons and another trap with a large dust catch because of local bare ground were omitted, the CaCO

3 equivalent ranged from 0.35–0.55

g/m2/yr that could potentially have been derived from airborne particles.

However, rain as a source of calcium became rec-ognized as an even more important source than dust. Chemical analysis of rain by Junge and Werby (1958) and Lodge et al. (1968) revealed that calcium from rain-water could produce an estimated CaCO

3 equivalent

of 1.5 g/m2/yr, assuming 200 mm of annual rainfall. Therefore, if ample bicarbonate is generated by roots and microbes, carbonate from rain is roughly two to three times more abundant than in calcareous dustfall (Gile et al., 1981, p. 63).

GEOMORPHIC SURFACES AND MORPHOSTRATIGRAPHIC UNITS

Geomorphic surfaces as a mapping conceptThe geomorphic surface as a mapping concept played a major role in the Desert Project by linking soil forma-tion to landscape evolution. To Hawley (1972, p. 25), the ideal geomorphic surface is a mappable two-dimen-sional feature that comprises the surface of genetically related landforms that developed when climatic and base-level controls were relatively constant. To Ruhe (1975, p. 136), a “geomorphic surface is a part of the land that is specifically defined in space and time.” A common example is a continuum of erosional and depo-sitional surfaces; however, a geomorphic surface could also be the surface of a lava flow, a glacial outwash plain, or a loess sheet.

Geomorphic surfaces and any genetically related deposits (following section, “Morphostratigraphic units as a mapping concept,” q.v.) were delineated according to whether they are inset below, inset against, overlap, or cut into adjacent surfaces (Ruhe, 1962, 1964; Hawley,

Bulletin 798 • Page 11

1965; Gile & Hawley, 1968; Gile et al., 1969; Gile 1975c). Quoting from Gile & Grossman (1979, p. 37):

The geomorphic surfaces were found to be useful in studies of soils because they serve as a chronologi-cal framework, and provide a common thread that makes the soil patterns easier to understand. Most geomorphic surfaces in the study area are extensive; and their surficial sediments [morphostratigraphic units], in which the soils have formed, can range widely in mineralogy, texture, and climatic occur-rence. Many kinds of soils can therefore occur on a single geomorphic surface.... However, a common feature to the various soils of a given surface is the degree of soil development, taking into account effects of changes in parent materials and climate.... Thus soil morphology can supply important evidence for the identification of geomorphic surfaces.

Morphostratigraphic units as a mapping conceptThe morphostratigraphic unit, now allostratigraphic unit (e.g., Frye & Willman, 1962; North American Com-mission on Stratigraphic Nomenclature, 1983), was the basic unit in geologic-geomorphic mapping of the Desert Project (Hawley & Kottlowski, 1969; Hawley, 1972, 1975a; Gile et al., 1981). It consists of earth ma-terial genetically related to the constructional phase of a geomorphic surface. The morphostratigraphic unit thus provided a way to combine a geomorphic surface with its underlying alluvium. This was useful for geo-logic mapping because individual alluvial fills, such as fan and terrace deposits, that could not be separated on a lithologic basis could be separated on the basis of their relative topographic position (Hawley & Kot-tlowski, 1969, Appendix A). Morphostratigraphic units were also very useful for paleoclimate and paleoecologic studies (Hawley, 1975a). This is because each unit, con-tained between unconformities, is (1) a sedimentary re-cord of landscape instability as indicated by erosion and concomitant sediment deposition and (2) a pedologic record of landscape stability as indicated by soil forma-tion.

Table 3 presents geomorphic surfaces, genetically-related morphostratigraphic units, and their estimated ages in the Desert Project. Figure 4 is a cross section of geomorphic surfaces along the eastern valley border in the vicinity of New Mexico State University. Based on geomorphic position, soil development (especially carbonate stages, see Table 3 and Figure 5), and radio-carbon dating, geomorphic surfaces of the intermontane basin correlate with surfaces of the valley border (Gile et al., 1981; Figure 1).

The valley border stepped sequenceGeomorphic surfaces along the valley border commonly occur in a terraced terrain in which age of the geomor-phic surfaces and their soils increases with increasing elevation of the terraces (Figure 4). This arrangement of geomorphic surfaces was termed a stepped sequence by Ruhe (1962, p. 165).

For example, with increasing elevation of the terraces, shown in Figure 6A as the Fillmore, Picacho, and Jor-nada I surfaces, soils increase markedly in development with age (see Gile et al., 1981, p. 109; Gile et al., 1995, p. 8). However, there are three situations where the age-elevation relationship does not hold. The first situation occurs in the valley border where colluvium from high ridges of ancestral Rio Grande alluvium partly buries the older Picacho surface and its soils (Figure 6B). The second situation occurs east of the valley border where an Organ fan of Holocene age partly buries a soil of Jornada II (late Pleistocene) age (Figure 6C). Sediments of the Organ morphostratigraphic unit occur as a ridge that is higher than the unburied part of Jornada II al-luvium. In the third situation, the deposits are at similar elevations, are isolated, and the stratigraphic relations cannot be determined. In this situation, soil morphol-ogy is a valuable (and virtually the only) tool for identi-fying and differentiating the deposits.

Effect of landscape dissection on geomorphic surfaces and soilsLandscape dissection and associated soil erosion have a ma-jor effect on soil properties and the identification of geo-morphic surfaces. Quoting from Gile et al. (1981, p. 99):

Dissection significantly affects geomorphic surfaces and soils, particularly adjacent to large river valleys such as the Rio Grande [Figure 7]. As a large valley evolves and drainage networks of the valley border expand, the main areas affected by tributary-stream dissection are the interfluve-summit, shoulder and backslope components of a typical hillslope sequence....

Boundaries between geomorphic surfaces may be readily identified in stable areas little affected by dis-section. In dissected terrains the situation is different because large volumes of sediment have been eroded from the soils and sediments associated with a given surface; and the amount that has been removed varies from one place to another. An example is the geomorphic change from Picacho to the Fillmore surface, identified by a combination of geomorphic and soil evidence [Figure 8]. Dissec-

Bulletin 798 • Page 12

Figure 4. Generalized diagram showing the stepped sequence of geomorphic surfaces (constructional surfaces) and a structural bench (an erosional surface) in the vicinity of Tortugas Mountain. Two older members of the stepped sequence (the relict basin floors, lower and upper La Mesa) on the west side of the valley are shown in Figure 1 (from Gile et al., 2007, used with permission).

Figure 5. Schematic diagram and descriptions of the diagnostic morphology for the stages of carbonate horizon formation in the two morphogenetic sequences. Carbonate accumulations are indicated in black (from Gile et al., 1966, fig. 5, used with permission). The stages will also be further discussed in the section “Horizons of carbonate accumulation.”

Bulletin 798 • Page 13

Figure 6. Relation of relative ages of geomorphic surfaces and elevation. A, terraced terrain in which age of the geomorphic surfaces and their soils increases with increasing elevation of the terraces. B & C, examples in which younger geomorphic surfaces are elevationally higher than older surfaces (modified from Gile et al., 1981, used with permission).

Bulletin 798 • Page 14

tion has resulted in a landscape dominated by ridges. The boundary between the Picacho and the Fillmore surfaces (also the boundary between Haplocalcids and Torriorthents) coincides with truncation of the calcic horizon [Figure 8].

The change from one surface to another and its relation to soils is quite distinct on the sides of ridges but is less obvious on ridge crests [Figure 8]. Rem-nants of the Picacho surface and its soils are preserved only on broadest ridge crests. The ridge crests are nearly level transversely, have a consistent longitudi-nal slope of approximately 2 percent, [and represent the relict constructional surface of an alluvial fan of mid-to late-Pleistocene age (Picacho alluvium)]. This slope accords with the general projection of the gradi-ent of stable arroyo-terrace remnants of the Picacho

surface to the east. The interfluves are crossed by small drainageways that truncate upper horizons of the soils and of the uppermost part of Picacho alluvium. However, saddles (which would indicate breaching of the Picacho surface and all horizons of its soils) are absent. The scattered small drainageways do not com-pletely truncate the lowermost diagnostic horizon (the calcic horizon).

Removal of all evidence of the Picacho surface and its soils is accomplished by continuation of the dissection process. Further dissection...is marked by rounding of ridge crests, narrowing of ridge remnants, development of saddles, and truncation of the calcic horizon. Then the Fillmore surface and its soils occur on ridge crests as well as on ridge sides.

Figure 7. Illustration of a dissected landscape east of the Rio Grande floodplain in the southern part of the Desert Project. Figure 8 shows some of the soils on the alluvial fan at left.

Bulletin 798 • Page 15

Constructional surfaces versus structural benchesStructural benches are common in the valley border and were once thought to be part of the stepped sequence of geomorphic surfaces (e.g., Ruhe, 1967, Plate 2). Quot-ing from Gile & Grossman (1979, p. 37):

A distinction is made between a constructional sur-face and a structural bench. The term, constructional, is ‘said of a landform that owes its general character to the processes of upbuilding, such as accumulation by deposition’ (Glossary of Geology). Constructional geomorphic surfaces form the major portion of the stable older land surface remnants along the Rio Grande Valley. Such surfaces were formed as a result of continued accumulation of alluvium, followed by a halt in deposition and the start of soil formation. Sta-bilization of the Picacho surface, for example, must have occurred, and soil formation must have started,

at about the same general period of time during late Pleistocene.

In contrast, structural benches (Figure 9) form by erosion instead of deposition (see also Gile et al., 1981, p. 29). The structural benches have formed in exhumed, erosion-resistant gravelly sediments deposited by the ancestral Rio Grande, and are marked by level to gently undulating ridge crests of gravelly sediments. Exhuma-tion of resistant layers of fluvial gravel from beneath overlying, less gravelly piedmont slope deposits and buried soils began in early mid-Pleistocene time and is continuing today. The gravelly sediments have also been variably dissected since their exhumation. For these rea-sons, age of soils on the benches ranges widely. In many areas, a low, sinuous scarp marks the boundary between the structural bench and the present limit of erosional stripping of the piedmont slope deposits (Figure 9).

Figure 8. Block diagram of soils and geomorphic surfaces. The Calciargids and Haplocalcids occur on the late Pleistocene Picacho surface and have pedogenic evidence (argillic horizons and/or calcic horizons in stage III of carbonate accumula-tion) of soils of that age. The Calciargids occur on small, remnant ridge crests, level transversely, that have been relatively little affected by the dissection. The Haplocalcids occur on rounded ridge crests in and near drainageways where the argil-lic horizon has been truncated and/or carbonate engulfed. The Torriorthents occur on the Holocene Fillmore surface of ridge sides, where both the calcic and argillic horizons have been truncated and the soils have stage I carbonate horizons typical of Fillmore age (modified from Gile et al., 1981, used with permission).

Bulletin 798 • Page 16

Figure 9. Illustration of formation of structural benches along the valley border. The rectangle locates a study area and cross section. Both land-surface and buried soils formed in the fan-piedmont sediments have loamy textures and little gravel. These soils do not slump readily, and a discontinuous scarp marks the area of their outcrop and contact on the un-derlying river gravel.

Bulletin 798 • Page 17

PALEOCLIMATE

Late Quaternary climatesSeveral lines of evidence indicate that a different late Quaternary climate existed in the now arid and semiarid regions of southern New Mexico. Fossil pollen suggests times of greater effective moisture during the last full-glacial interval from about 17,000–23,000 B.P. (Martin & Mehringer, 1965). Small glaciers were discovered to have formed on nearby mountain peaks (Richmond, 1963), and it was recognized that numerous perennial lakes had occupied many closed basins in the region (Herrick, 1904; Hawley et al., 1976). Packrat evidence indicates that the last full-glacial interval was followed by warmer and drier intervals interrupted by short peri-ods of greater effective moisture (Van Devender, 1977). More recent studies have supplemented these findings (e.g., Betancourt et al., 1990; Hawley, 1993).

Understanding the late Quaternary climates is im-portant for understanding soil and geomorphic features that, in part, were produced by those climates. Soil and geomorphic features themselves became sources of information on late Quaternary climates. Three soil-geomorphic categories were particularly useful for study-ing paleoclimate: (1) cyclic entrenchment of the Rio Grande, (2) cyclic sedimentation and buried soils of the fan piedmont, and (3) polygenetic soils.

Cyclic entrenchment of the Rio GrandeThe modern arroyo system is graded to the current base level of the Rio Grande. Sediments of the arroyo chan-nels are generally freshly deposited and show essentially no evidence of pedogenesis. However, progressively higher geomorphic surfaces have progressively greater pedogenic modification, indicating (together with their geomorphic position) that they are progressively older. The fact that the geomorphic surfaces can be mapped upstream and downstream from their type locations in the Desert Project (Hawley et al., 1969), plus their simi-larity to surfaces in other parts of the Southwest (Ruhe, 1964), suggests that some regional factor controlled their development. Cyclic climate change is thought to be the regional factor responsible for the cyclic en-trenchment and formation of the observed stepped-sequences of valley-border erosional and constructional surfaces (Kottlowski, 1958; Ruhe, 1964; Hawley, 1965; Hawley & Kottlowski, 1969; Metcalf, 1965, 1969).

The hypothesis of cyclic climate-controlled entrench-ment is as follows: 1) The Rio Grande downcut during glacial periods when greater precipitation gave rise to more water in the river channel and caused denser veg-etative cover across the landscape. During this time, the river channel carried a lower sediment load as a result of

greater vegetative cover. 2) During waning glacial and early interglacial times, the river valley partially back-filled as a result of less water in the river channel and greater amounts of sediment supplied by erosion due to the decreased amount of vegetative cover on the land-scape. 3) Aggradation ceased and the base level stabilized during later interglacial times, until the cycle began again with downcutting with the renewed waxing phase of the next glacial cycle (Hawley, 1975a).

This hypothesis has yet to be disproven and is not in conflict with fossil molluscan faunas or chronologic, sedimentologic, and pedologic information. The hy-pothesis also fits well in Schumm’s (1965) scheme of river activity for the semiarid, continental-interior mid-section of river systems heading in glaciated mountains (Hawley & Kottlowski, 1969, p. 99). Cyclic sedimentation and buried soils on the fan piedmontRelative ages of soils in the valley-border stepped se-quence are determined by their position in the sequence (Figure 4), whereas the relative ages of soils on the fan piedmont are determined by their position in a section of buried soils (Figures 10 and 11). As paleoclimatic in-dicators, buried soils reveal that each period of sediment deposition was followed by a period of landscape stability and soil formation. The theoretical link of buried soils to climate change was explained by Ruhe (1962, p. 163):

By analogy with present conditions in the region and on a theoretical basis....Past pluvial environ-ments in present arid regions, correlative of glacial episodes elsewhere, should have resulted in greater vegetative cover on relatively stable landscapes....and soil formation. Interpluvial environments, as at pres-ent, should have resulted in increased aridity, lesser vegetative cover, unstable landscapes subject to severe erosion and to sediment transport. The best example of the interrelationships of buried soils and paleoclimate occurs in the northeast part of the Desert Project, where buried charcoal was found in Organ alluvium at the Gardner Spring radiocar-bon site (Figure 1).

The Gardner Spring radiocarbon siteThe Gardner Spring radiocarbon site (Ruhe, 1967; Gile & Hawley, 1968; Hawley & Kottlowski, 1969; Gile, 1975c; Gile et al., 1981) occurs in a valley fill between two large Pleistocene fans below the San Andres Moun-tains (Figure 12). The valley fill contains Organ allu-vium of mid and late Holocene age (Table 3).

Bulletin 798 • Page 18

The Organ deposits have been cut by Gardner Spring Arroyo, the banks of which (quoting from Gile et al. [1981], p. 158 et seq.):

...provide excellent exposures of silty to gravelly valley-fill alluvium associated with several ages of the Organ surface. Buried soils associated with the Jornada II surface can be observed at several places below Organ alluvium. This older surface forms the floor and side-slopes of the interfan valley [Figure 12]. A number of lenticular beds of charcoal, possibly associated with buried hearths, have been found in sections of Organ alluvium exposed along this arroyo and two tributaries. Radiocarbon dates obtained from charcoal collected from eight beds and physical evidence of several distinct breaks in sedimentation indicate that at least two major, and one minor, epi-sodes of alluviation occurred during development of the Organ surface in Holocene time [Figures 12–14].

The three Organ alluviums are thought to have regional paleoclimatic significance. Quoting from Gile (1975c):

Organ I may be of regional significance; Haynes (1968) indicated extensive channel filling between

6000 and 4000 B.P. in the western U.S. Haynes (1968, pp. 612, 613) states ‘...alluvial deposits of many tributary streams can be correlated on the basis of soils, fauna and archaeology over a wide area.... Some erosional contacts are correlative within 200 or 300 years at widely separated localities.’ Irwin-Wil-liams and Haynes (1970) relate patterns of human occupation in the Southwest to the Altithermal period and to the Fairbanks and Whitewater Droughts of Antevs (1955). The Holocene events at Gardner Spring may be related to these events [Figure 14].

The date of 6400 B.P. came from charcoal in the lower part of Organ alluvium, which here rests on Jornada II alluvium of late-Pleistocene age (Gile and Hawley, 1968). Hence the initiation of Organ allu-viation may approximately coincide with the start of the Altithermal, which began about 7500 B.P. (An-tevs, 1955). The Altithermal ended about 4000 B.P. according to Antevs (1955), but in some areas there is evidence that it ended about 5000 B.P. (Mehringer, 1967; Irwin-Williams and Haynes, 1970). The start of such a warm, dry period could have reduced the vegetative cover enough that erosion started in par-ticularly susceptible areas such as steep slopes adjacent to drainageways.

Figure 10. Block diagram of a fan-piedmont drainageway. The generalized diagram illustrates the relative age of soils of six geomorphic surfaces. The soils of the Jornada I surface are the oldest, while soils of Modoc, Jornada II, Baylor, Isaacks’ Ranch, and Organ are successively younger. The stipples designate Bt horizons. Horizons of carbonate accumulation occur beneath the Bt horizons. The soil morphology and occasional presence of C horizon material between the sets of genetic horizons demonstrate that the clay and carbonate horizons of Jornada I and II alluviums are those of buried soils (formed when they were at the land surface) and not the result of deep illuviation or groundwater phenomena (from Gile et al., 1981, used with permission).

Bulletin 798 • Page 19

Figu

re 1

1. B

urie

d so

ils (

and

mor

phos

trat

igra

phic

uni

ts)

in n

orth

wal

l of

Hig

hway

70

gully

, whe

re “

a” is

Jor

nada

I, “

b” is

Jor

nada

II,

and

“c”

is I

saac

ks’

Ran

ch. S

cale

in f

eet.

Pho

togr

aphe

d M

arch

196

5 (f

rom

Gile

et

al.,

1981

, use

d w

ith

perm

issi

on).

Bulletin 798 • Page 20

Figure 12. Block diagram of Gardner Spring radiocarbon site (top) and generalized diagram of the chronostratigraphic rela-tions (bottom). Holocene soils are in alluvium of an Organ valley fill between two large Pleistocene fans downslope from the San Andres Mountains, which consist mostly of Paleozoic carbonate rocks (from Gile et al., 1981, used with permission).

Bulletin 798 • Page 21

Figu

re 1

3. P

rofi

le o

f O

rgan

II

mor

phos

trat

igra

phic

uni

t (A

, B, C

ca, C

hor

izon

s) o

verl

ying

the

Org

an I

uni

t (A

b, B

b ho

rizo

ns)

just

bel

ow t

he 3

ft

mar

k. T

he g

rave

lly s

oil f

orm

ed in

Org

an I

I al

luvi

um is

at

leas

t 1,

100

yr o

ld b

ut n

ot o

lder

tha

n 2,

100

yr. T

he b

urie

d so

il is

less

tha

n 4,

600

yr o

ld a

nd

has

been

bur

ied

abou

t 2,

200

yr. V

eget

atio

n is

cre

osot

ebus

h (f

rom

Gile

, 197

5c, fi

g. 1

4, u

sed

wit

h pe

rmis

sion

). H

oriz

on n

omen

clat

ure

from

Soi

l Sur

vey

Staf

f (1

951)

.

Bulletin 798 • Page 22

Figure 14. Chronologic illustration of events at the Gardner Spring radiocarbon site in comparison to events described by Antevs (1955) and Haynes (1968). W.D. = Whitewater Drought. F.D. = Fairbanks Drought. Triangles indicate radiocarbon dates (from Gile, 1975c, fig. 15, used with permission).

Figure 15. Block diagram of a pipe extending into a petrocalcic horizon. Such pipes are thought to be the result of pluvial pe-riods in the Pleistocene, when greater amounts of water percolated deeper than current depths of wetting (from Gile, 1975b, used with permission).

Bulletin 798 • Page 23

Additional evidence for climatic change as a cause of Organ alluviation is the remarkable ubiquity of the deposits in areas accessible to Holocene sedimenta-tion (e.g., in valley fills between Pleistocene fans); and Organ alluvium occurs downslope from all mountain ranges in the area, even the small ones.

All dated charcoal horizons between 4000 and 5000 B.P. are well below the surface of Organ I. Hence, sedimentation of Organ I could have ended about 4000 B.P. The return of more moist conditions should have stabilized the land surface so that soils could form in Organ I alluvium. Pollen evidence sug-gests that this may be the case. Freeman (1972) found a decrease in cheno-ams and an increase in grass pol-len toward the upper part of Organ I alluvium. This would fit Antevs’ model of an end to the Altithermal about 4000 B.P.

Organ III alluvium, which filled a gully nearly 20 m wide and 2 m deep, has been observed only in the vicinity of the dated site. This gully cuts Organ II al-luvium, and must have developed between 1100 and 2100 B.P. Old filled arroyo channels record droughts; they were cut by severe floods during dry periods and filled during periods of transition to more moist cli-mates (Bryan, 1925; Antevs, 1955; Haynes, 1968). Irwin-Williams and Haynes note a brief period (about 1250 B.P.) of drastically decreased effective moisture and indicate that this period may corre-spond to the Whitewater Drought of Antevs (1600 B.P.). Such a drought may also have caused the Organ III gully at Gardner Spring.

The chronology at Gardner Spring is also in rea-sonable agreement with Haynes’ alluvial chronology for the southwestern United States [Figure 14].

No Bt horizon has formed in the Organ sediments because of the high-carbonate parent materials (the later section “Soil and soil-geomorphic boundaries,” q.v.). However, weak stage I carbonate, consisting of thin, discontinuous carbonate coatings on sand grains and pebbles, occurs in the gravelly Organ III sediments. More distinct stage I carbonate, occurring as continuous carbonate coatings, occurs in gravelly Organ II sedi-ments. In the finer-textured Organ I sediments, stage I carbonate filaments occur on surfaces of some peds.

Polygenetic soilsSoils with morphological changes thought to be due to major changes in climate, soil erosion, or sediment deposition are termed polygenetic (Gile et al., 1966).

This contrasts to soil evolution, in which morphological changes are caused by continuation of some genetic soil process, such as carbonate accumulation. Rates of soil evolution can be greatly changed by polygenetic factors.

Polygenetic soils are common in the Desert Project area, some having endured several arid interglacial-gla-ciopluvial cycles (Hawley, 1975a). As reviewed by Gile (1966b), the Desert Project region was cooler and wetter during the Pleistocene as based on the occurrence of gla-ciers in nearby mountains and the presence of lakes both north and south of the study area. In contrast to the in-terglacial Holocene soils, which must have formed in a climate similar to the present one, Pleistocene soils have thicker and stronger horizons, probably due to deeper leaching in pluvials (Gile, 1966b, 1970).

Vertical, funnel-shaped pipes (Figure 15) that extend downward through calcic and petrocalcic horizons may also be the result of greater and deeper leaching during Pleistocene pluvials. The presence of Bt material in the pipes and continuous laminar layers descending from atop petrocalcic horizons into the pipes indicate flushing with water.

The change from pluvial back to arid conditions may be recorded as shallow stage I carbonate filaments and coatings overlying laminar petrocalcic horizons. Depths of stage I carbonates are known from soils of Holocene age (Gile et al., 1966, 1981). Stage I carbonates at similar depths above calcic and petrocalcic horizons in Pleisto-cene soils are presumed to have accumulated during Ho-locene arid conditions (Gile et al., 1969). In soils with pipes, shallow stage I horizons of carbonate filaments extend above both calcic and petrocalcic horizons and as-sociated pipes. This suggests an upward shift in the depth of wetting in response to aridity (Gile et al., 1966).

SOIL CHRONOLOGYSoil age, as one of the five soil forming factors, has had a profound impact on the nature of soils in the Desert Project. Several techniques have been employed to measure soil age. For soils of Historical age, land sur-vey notes have been very useful. For soils of prehistoric age, radiocarbon, tephrochronology, paleomagnetism, and paleontology have been useful. The relative ages of soils have been determined by their superposition and cross-cutting relationships as well as the degree of soil profile development.

Land survey notesDrastic landscape changes resulting from desertification in the American Southwest have occurred during the past 150 years (Buffington & Herbel, 1965). Soils bur-ied by coppice dunes of Historical age reveal buried A

Bulletin 798 • Page 24

Figure 16. Position of Holocene soils in the vicinity of the Isaacks’ radiocarbon site (Figure 1). The soils are in ridges on the fan-piedmont downslope from the San Agustin Mountains, which are dominantly monzonite. The Isaacks’ radiocarbon site is in the slight ridge of Holocene alluvium (center left) inset against a slightly higher ridge of late Jornada II alluvium and soils (modified from Gile, 1975c, used with permission).

Bulletin 798 • Page 25

horizons beneath wind-blown C horizon material in the dunes (Gile, 1966a). The pre-dune A horizons buried and preserved by the coppice dunes are absent in the interdune areas. This stratigraphic evidence, combined with landscape notes recorded by surveyors in 1857 and 1885, reveals that extensive changes from grass to shrub vegetation were associated with wind erosion of sandy soils during the late 19th century. Other land survey notes suggest that some of the major arroyos in the Des-ert Project were not present in 1858, but were present in 1922 (Gile & Hawley, 1968).

Buried charcoal Charcoal radiocarbon dates have provided the best evi-dence for soil ages at the Desert Project. Charcoal ages range from 9,360 B.P. in Leasburg alluvium (Hawley & Kottlowski, 1969) to 1,130 B.P. in Organ III alluvium (Gile & Hawley, 1968). For pedology studies, radiocar-bon ages of buried charcoal establish a maximum age for genetic soil horizons above the charcoal. For example, a reddish-brown, non-calcareous Bt horizon and under-lying stage I carbonate at the Isaacks’ radiocarbon site (Gile, 1975c; Gile et al, 1981; Figure 16) overlie char-coal dated at 4,200 B.P. Therefore, the Bt horizon and stage I carbonate must be less than 4,200 yr old (Gile, 1975c).

Organic and inorganic carbon dates were also ob-tained from a nearby pedon (Gile et al., 1981). Charcoal from the Ck horizon was dated at about 4,000 B.P. Two dates were obtained from carbonate that, theoretically, must be younger than 4,000 yr because it was above the charcoal: (1) Carbonate adhering to pebbles dates at about 4,400 B.P. and (2) carbonate in the 0.2–0.002 mm fraction dated at about 1,600 B.P. The older date from the pebble coatings would be expected because such coatings represent the first carbonate to be precipi-tated. But the date is older than it could be according to the underlying dated charcoal. The general relation of older dates for inorganic rather than associated organic carbon dates also holds for much older samples (as dis-cussed later).

These C-14 charcoal dates and soil-geomorphic trac-ing from the dated sites demonstrate the extensive areas of soils of mid and late Holocene age. In particular, the Gardner Spring and Isaacks’ radiocarbon sites illustrate soils and sediments of this age along the mountain fronts and fan piedmont; the Shalam Colony radiocar-bon site illustrates Holocene soils along the valley border (Figures 1 and 17).

Figure 17 is a block diagram showing the soils and surfaces in the vicinity of the Shalam Colony radiocar-bon site (Ruhe, 1967; Hawley & Kottlowski, 1969; Gile, 1975c; Gile et al., 1981). Figure 18 shows a soil

map and a cross section of a Fillmore terrace and col-luvial wedge. Two charcoal lenses were dated at the Shalam Colony site (Ruhe, 1967). The uppermost lens, at about 1-m depth, was dated at 2,850 B.P. and was beneath the genetic horizons. The soils must therefore be less than 2,850 yr old and must have formed under a climate very similar to the present one. The most promi-nent evidence of pedogenesis is the stage I carbonate horizon (Figure 5). The lower charcoal lens, at 2.4-m depth, was dated at 4,900 B.P.

Fillmore terraces are common next to and near the Rio Grande floodplain, occurring between the much higher fan remnants of the Picacho surfaces of late Pleistocene age. Colluvial wedges occur on the sides of the Picacho remnants and grade to the Fillmore ter-races (Figure 18). Soils of the colluvial wedges occur in a nearly continuous band around margins of the Picacho remnants (Figure 18). The stage I carbonate horizon in the soil of the wedges is substantially thicker than in the soil on the Fillmore terraces. The colluvial wedges may therefore be older than the terraces, and may represent the oldest part of the Fillmore erosion and deposition, which presumably began at the start of the Altithermal about 7,500 years ago (Antevs, 1955).

Although 15 charcoal lenses have been found in soils of Holocene age, charcoal has not been found in soils and parent materials of late Pleistocene age and older (Gile & Grossman, 1979). This may be the result of the type of vegetation in the Pleistocene, probably dominat-ed by grasses, and the absence or sparseness of human populations in the region, since many of the charcoal lenses are hearths.

Radiocarbon ages of pedogenic carbonateCarbonate C-14 ages can vary considerably in the same soil (Ruhe, 1967; Gile et al., 1966; Table 4). The vari-ability is well illustrated by the soil shown in Figure 3 of Gile et al. (1966). This soil is considered to be highly significant because the carbonate morphology was stud-ied in detail and related to its radiocarbon chronology. This was the first soil so studied in the Desert Project and the first to be reported in the scientific literature on genesis of carbonate horizons. Table 4 gives the radio-carbon dates obtained from the various horizons (for an illustration of the plugged and laminar horizons of this soil, see the later section “Horizons of carbonate accumulation”). Reasons for the variety of radiocarbon dates obtained were discussed in Gile et al. (1966, pp. 356–357).

Older detrital carbonate, either from eroded soils up-slope, limestone, or calcareous dust, may change C-14 ages. Dissolution and re-precipitation is another compli-cation. Quoting from Gile et al. (1981, p. 76):

Bulletin 798 • Page 26

[In theory]....exchange with modern carbon [can make carbonate C-14 ages younger by] solution-precipitation of calcite. During each cycle of solution-precipitation, half of the carbon in the precipitated CaCO

3 originates from CO

2. If the CO

2 is modern,

during each cycle the radiocarbon activity increases by half the decrement from that for modern carbon. Isotopic exchange of carbon in solution with solid car-bonate is not considered an important factor; change in activity is mainly the result of solution and pre-cipitation. A small proportion of recently precipitated carbonate has a large influence on the C-14 age (Kim et al., 1969). The carbonate in most of the horizons exceeding 200 kyrs in age must have undergone ex-change with environmental C-14 after emplacement because all but one has an activity high enough to obtain a date. Frye and others (1974) present C-14 ages for carbonate from middle Pleistocene and older carbonate accumulations in east-central New Mexico that are within the datable range. Apparently most carbonate accumulations associated with the land surfaces in the region have an activity within the dat-able range.

In addition to the possibility of solution-precipitation of existing carbonate, biogenic carbonate precipitated by roots and microorganisms would also make carbonate C-14 ages younger (Monger et al., 1991b). Neverthe-less, certain aspects of soil chronology can be made clear from measuring carbonate C-14 ages. Quoting again from Gile et al. (1981, p. 76):

[Figure 19] compares age based on carbonate ra-diocarbon (C-14) to soil age based on independent estimates. Carbonate C-14 ages appear useful to cor-roborate the relative ages of soils from late Pleistocene through Holocene but are of little value in distin-guishing between late Pleistocene and older soil horizons....

Radiocarbon differences between carbonate and organic carbonAlthough it may seem logical on a theoretical basis that occluded organic matter would be older than the car-bonate that engulfs it (Ruhe, 1967, p. 60), radiocarbon dates consistently show occluded organic carbon to be younger than carbonate. Quoting from Gile et al. (1981, p. 77):

Figure 17. Block diagram of the geomorphic setting in the vicinity of the Shalam Colony radiocarbon site. A, soils of the Rio Grande floodplain; B, Fillmore and arroyo-channel surfaces; C, Fillmore ridge side, colluvium; D, Picacho surface; E, Torriorthents and sedimentary rock outcrop (mountain slopes and summits); F, location of charcoal horizons. Geologic units: 1, bedrock (sedimentary); 2, upper Camp Rice Formation, fluvial facies (ancient river alluvium); 3, Picacho river alluvium; 4, Picacho alluvium; 5, Fillmore colluvium; 6, Fillmore alluvium; 7, Rio Grande deposits (late Quaternary river alluvium) and soils (from Gile et al, 1981, used with permission).

Bulletin 798 • Page 27

Figu

re 1

8. S

oil-

geom

orph

ic m

ap a

nd c

ross

sec

tion

of

mor

phos

trat

igra

phic

uni

ts in

the

vic

init

y of

the

Sha

lam

Col

ony

radi

ocar

bon

site

. A, T

ypic

Tor

rior

then

ts

(Fill

mor

e an

d ar

royo

-cha

nnel

sur

face

s); B

, Typ

ic T

orri

orth

ents

(Fi

llmor

e ri

dge

side

s); C

, Typ

ic P

etro

calc

ids

(Pic

acho

sur

face

); D

, Tor

rior

then

t ro

ck o

utcr

op

(mou

ntai

n sl

opes

and

sum

mit

s). X

, loc

atio

n of

dat

ed c

harc

oal h

oriz

ons.

I-I

I lo

cate

s cr

oss

sect

ion.

Rig

ht v

iew

is a

dia

gram

mat

ic c

ross

sec

tion

of

soils

, sur

face

s,

and

sedi

men

ts (

I-II

on

soil-

geom

orph

ic m

ap)

(fro

m G

ile, 1

975c

, fig.

4, u

sed

wit

h pe

rmis

sion

).

Bulletin 798 • Page 28

Table 4. Radiocarbon Ages of Various Forms of Pedogenic Carbonate from a Petrocalcid (Pedon 59-16; Gile & Grossman, 1979)

Horizon Depth (cm) Sample Radiocarbon Age Years B.P. Isotope, Inc. Lab Number

B22tca 15–28 Whole soil 5,725 ± 200 I-374

Pebble coatings 10,226 ± 400 I-309

Fine earth (calculated) 1,300

K21m 28–30 Soft upper laminae 4,575 ± 170 I-375

Hard lower laminae† 13,850 ± 600 I-392

Adherent plugged horizon 18,300 ± 600 I-391

K22m 30–64 Plugged horizon, whole soil 15,300 ± 400 I-376

†Organic carbon in the hard lower laminae gives an age of 9,550 ± 300 years (I-616). The separated > 2-mm material was gently washed in water, dried, and passed through a 2-mm sieve. Carbonate adhering to the pebbles after washing is referred to as pebble-coating carbonate. The radiocarbon age of the fine-earth carbonate was calculated from the measured ages for the whole-soil and pebble-coating carbonate, plus the weight percentages of pebble-coating (63%) and fine-earth (37%) carbonate. The fine-earth carbonate is 14% sand size, 58% silt size, and 28% clay size.

Bulletin 798 • Page 29

The C-14 ages from organic and inorganic car-bon differ, and the difference becomes greater with increasing age [Table 5]. In the first two soils (both of late Pleistocene age) the C-14 ages from inorganic carbon are 3 and 4 kyrs, respectively, older than C-14 ages from organic carbon in the same horizon. One soil developed in non-calcareous rhyolitic alluvium and the other in alluvium from calcareous sedimen-tary rocks of Paleozoic age. Similarity in ages of the organic carbon suggests that the laminar horizons were formed at approximately the same time. The presence of limestone apparently did not affect the age of the authigenic laminar carbonate....

In the soil of middle Pleistocene age, the difference between organic and inorganic C-14 ages is even greater [Table 5]. The differences exceed any reported by Williams and Polach (1969, 1971). Initial age of authigenic carbonate may be partly responsible. The differences for the middle Pleistocene soil, however, ex-ceed 5.7 kyrs, which is the theoretical age of dead car-

bonate after going through one solution-precipitation cycle. The difference, moreover, is much greater than the initial age of 3 kyrs for carbonate.... Occurrence of preexisting carbonate in the laminar zone is not a ten-able explanation. A minimum of 30 percent allogenic dead carbonate would be needed to increase the ini-tial age from 5.7 kyrs (the maximum for authigenic carbonate) to 9 kyrs. Such a proportion of allogenic carbonate is inconsistent with the morphology of the laminar subhorizon (Gile and Grossman, 1979). An alternative explanation for the difference is that the organic carbon has had its C-14 age reduced more than the carbonate since emplacement. Polished sections of laminar horizons reveal occasional cracks that have been sealed by carbonate. Roots may have entered these cracks and contributed modern organic carbon. The explanation assumes that the carbonate, which later seals the cracks, has had less effect on the C-14 ages than the organic carbon added.

Figure 19. C-14 ages of carbonate versus soil age independently evaluated. Maximum carbonate ages were used; laminar horizons were excluded (from Gile et al., 1981, used with permission).

Bulletin 798 • Page 30

Recent studies indicate that dissolved colloidal organic matter may also youthen the C-14 age of organic carbon occluded in pedogenic carbonate. This occurs as a result of capillarity pulling the soil solution and organic col-loids into the carbonate-impregnated material (Wang et al., 1996).

Volcanic ashFor soils of middle Pleistocene age and older, volcanic ashes have provided the most conclusive evidence for bracketing soil ages. Two ashes occur in the region: the Lava Creek B (0.61–0.67 Ma) and Bishop ash (0.76 Ma) (Hawley et al., 1976; Izett et al., 1992; Sarna-Wojcicki & Pringle, 1992). Both are stratigraphically beneath the Jornada I surface, which was the last pied-mont surface to form before the incision of the ancestral Rio Grande.

PaleontologyFossils of Pleistocene megafauna have also been impor-tant for determining the age of soils older than middle Pleistocene. Paleontological remains provided some of the earliest clues to the ages of soils in the Desert Project region. Equus, Cuvieronius, Mammuthus (mammoth), and Stegomastodon (mastodon) were thought to be of mid-Pleistocene “Kansan age” in Camp Rice fluvial de-posits just beneath the La Mesa surface (Ruhe, 1962, p. 161; Hawley et al., 1969). Subsequent work shows this faunal assemblage to be of early Pleistocene age (Ted-ford, 1981; Morgan et al., 1998). Fossil mollusks in de-posits along the Rio Grande (Metcalf, 1967, 1969) and fossil pollen in alluvial fan deposits along the mountains (Freeman, 1972) helped identify periods when ecologi-cal zones were elevationally depressed (indicating pluvial conditions).

Relative ages based on pedogeomorphic tracing and soil development Tracing soil horizons and geomorphic surfaces laterally provides some of the most conclusive evidence about the relative ages of soils. A soil can be determined to be younger than a neighboring soil if its geomorphic surface and the associated sediments bury, cut, or are inset against the neighboring surface and its sediment (Figure 20).

If, however, a deposit and its soils are isolated and cannot be directly traced, and where the morphological range of soils on the various surfaces has been deter-mined, the geomorphic surface can be identified by soil properties alone (Gile, 1977).

Another indicator of relative soil age is the degree of profile development. Prominent profile indicators of relative soil age in the Desert Project are solum thickness and kind and expression of soil horizons (Gile, 1970). These include, on one extreme, highly developed pet-rocalcic horizons that can represent a couple of million years of pedogenesis. On the other extreme, and perhaps the most rapidly forming horizon, is the vesicular A ho-rizon, which was observed to have formed in a deposit partially filling a tire-track then known to be less than two years old (Gile & Hawley, 1968, p. 713).

SOILS AND SOIL-GEOMORPHIC RELATIONS

Soil horizons and conventions In general, the horizon designations in the Desert Proj-ect literature followed the Soil Survey Division Staff (1993), with the notable exception of the K horizon nomenclature (Gile et al., 1965) and the designations for buried soils, which are placed at the end of the des-ignation to handle more than one buried soil. The K horizon continues to be used because, as noted by Birke-land (1999), “Most pedologists and geologists working in arid lands find it a very useful term.” Although the K horizon is not formally recognized as a master horizon by the National Cooperative Soil Survey, the “kk” suffix, which is derived from the K horizon concept and cor-responds to the stage III plugged horizon or higher of the carbonate morphogenetic stages, has been formally adopted (Soil Survey Staff, 2006).

Horizons of carbonate accumulation have been des-ignated stages I–IV as described in Gile et al. (1966). Later, stages V and VI were proposed by Bachman and Machette (1977) and Machette (1985). These additions were found to be useful because they recognize advanced evolutionary changes beyond formation of the stage IV laminar horizon. Machette (1985) proposed that stage IV be limited to laminae or laminar layers that are less

Table 5. Comparison of C-14 Ages of Carbonate and Oc-cluded Organic Carbon in the Laminar Horizon of Three Soils with Petrocalcic Horizons Formed in Different Par-ent Materials (from Gile et al., 1981) C-14 ageSoil, age Carbonate Organic carbon Parent material (kyrs) (kyrs)

Hachita 59-16, late Pleistocene

Rhyolitic alluvium 14 10

Upton 66-5, late Pleistocene

Calcareous sedimentary rock 15 11

Cruces 61-7, middle Pleistocene

River alluvium

Upper half 29 21

Lower half 30 21

Bulletin 798 • Page 31

Figure 20. Illustrations of relative soil age based on pedogeomorphic tracing. Upper block diagram locates the middle and lower block diagrams. Middle diagram shows a Holocene valley fill burying soils formed on Pleistocene fans. Lower dia-gram shows Pleistocene soil buried by a ridge of Holocene sediments (from Gile, 1975a, used with permission).

Bulletin 798 • Page 32

than 1 cm thick, and that stage V be distinguished by laminae or laminar layers thicker than 1 cm, and in some cases also by pisoliths. Birkeland et al. (1991) and Birkeland (1999) do not limit stage IV to laminar layers that are less than 1 cm thick, and require that pisoliths, as well as laminae, be present for stage V. That usage has been followed in the Desert Project. In stage VI hori-zons, the pisolitic zone has multiple generations of brec-ciation and recementation. As pointed out by Machette (1985) stage VI horizons have higher bulk densities and CaCO

3 contents than do stages IV and V. Large areas of

stages I–V occur in the Desert Project; well-developed examples of Stage VI are on the Rincon surface, north of the Desert Project (Gile et al., 1996). The K horizon and the stages of carbonate accumulation are further discussed in the section on carbonate accumulation.

Soil taxonomyIn addition to understanding soil-geomorphic relations, soil classification was an important objective of the Des-ert Project. This included the refinement of definitions for diagnostic horizons and soil taxonomic names. Table 6 gives soils classified to the subgroup level in the Desert Project area. For a complete classification of Desert Proj-ect soils to the series and phase level see Gile et al. (2007).

Horizons of carbonate accumulation

OriginVoluminous literature on the origin of calcium car-bonate in soils, involving lacustrine, fluvial, ascending groundwater, and pedologic hypotheses, had been gen-erated by the late-1950s when the Desert Project began (e.g., Bretz & Horberg, 1949; Brown, 1956; see Gile & Grossman, 1979, pp. 139–142 for further discus-sion). Examples of lacustrine, fluvial, and groundwater carbonate were found in the Desert Project region, but these were limited to small areas. In contrast, the K horizon, which grades downward to Ck horizons, cov-ered vast areas of the landscape. Moreover, because K horizons formed in sloping soils of various textures, a geologic origin for the laterally extensive K horizon was problematic. In many areas, for example, the great depth of water tables precluded capillary rise as a source of soil carbonate. Quoting from Gile et al. (1966, p. 348):

The soils are well drained and do not at present have shallow water tables. The reasoning of Bretz and Horberg (1949, p. 509) can be followed to argue that these soils probably never were affected signifi-cantly by capillary rise from a shallow water table. The soils occur on dissected land surfaces. Each land surface, if it had had a shallow water table, would

have been drained during the succeeding erosion cycle. It would be highly coincidental if capillary rise were responsible for the pattern of increasing carbon-ate accumulation, and its nicely graded sequence, in the soils on progressively higher and older surfaces along the valley. The carbonate morphological sequences, to be presented, suggest strongly that the authigenic carbonate is illuvial.

Field evidence for a pedogenic-illuvial origin of carbonate accumulation was enumerated by Gile et al. (1965). The carbonate horizons (1) are parallel to the soil surface, (2) have upper boundaries within several inches to about 2 feet (0.61 m) of the soil surface, (3) have distinctive morphologies that show lateral continu-ity and that differ markedly from morphologies of over-lying and underlying horizons, (4) occur between hori-zons containing little or no carbonate, (5) occur across sediments of various compositions and textures, and (6) form in a developmental sequence related to time.

Later, two other lines of evidence were introduced. One was the depth of carbonate as a function of eleva-tion (Figure 21). In this case, the upper boundary of the carbonate horizon deepens with the gradual increase in elevation (i.e., toward mountains) and precipita-tion (Gile, 1977). A second line of evidence was found among soils of about the same elevation, where soils in a runoff position had shallower carbonate than soils of a more stable landscape position. For example, a soil on a runoff position on a narrow ridge had a carbon-ate horizon beginning at 20 cm. But tracing the soil and its geomorphic surface from the narrow ridge to a broad one showed that the top of the carbonate horizon deepened to 32 cm in the soil on the broad ridge, even though the soil texture remained about the same (Gile, 1977, p. 114).

The morphogenetic sequences and stages of carbonate accumulationHorizons of carbonate accumulation are very useful for assessing the relative ages of soils. For soils of progres-sively older surfaces, Gile et al. (1966) recognized a progressive increase in carbonate, which they expressed as stages of carbonate accumulation that formed in two morphogenetic sequences. Quoting from Gile et al. (1981, p. 66 et seq.):

The development of carbonate horizons of pedogen-ic origin is closely related to soil age. Two sequences of carbonate morphology associated with increasing soil age and amount of authigenic carbonate have been ordered in stages [Figure 5]. One of the sequences is in low-gravel (less than approximately 20 percent

Bulletin 798 • Page 33

Table 6. Classification (to the Subgroup Level) and Diagnostic Horizons and Features of Soils Discussed in Paper†Order Suborder Great Group Subgroup

Aridisols Argids Calciargids Typic Calciargids Have argillic but no petrocalcic horizon Have calcic horizon within 150 cm Have aridic moisture regime within 100 cm Haplargids Typic Haplargids No calcic horizon between 100 and 150 cm Have aridic moisture regime Petroargids Typic Petroargids Have petrocalcic horizon between 100 and 150 cm Have aridic moisture regime Calcids Petrocalcids Typic Petrocalcids Have calcic or petrocalcic horizon Have petrocalcic horizon between 100 and 150 cm Have aridic moisture regime within 100 cm; no argillic horizon within 100 cm unless petrocalcic horizon is within 100 cm Argic Petrocalcids Have argillic horizon within 100 cm

Haplocalcids Typic Haplocalcids Have calcic horizon within 100 cm Have aridic moisture regime

Cambids Haplocambids Typic Haplocambids Have cambic horizon within 100 cm Have aridic moisture regime

Entisols Psamments Torripsamments Typic Torripsamments Have < 35% by volume of rock fragments Have torric (aridic) moisture regime and texture finer than loamy fine sand

Orthents Torriorthents Typic Torripsamments Have ≥ 35% by volume of rock fragments Have torric (aridic) moisture regime and texture finer than loamy fine sand

Fluvents Torrifluvents Typic Torriorthents Organic carbon decreases irregularly with Have torric (aridic) moisture regime depth and/or is 0.2% or more at a depth of 125 cm, and slope is less than 25%

Mollisols Ustolls Argiustolls Aridic Argiustolls Have an ustic moisture regime or an aridic Have an argillic horizon moisture regime that borders on ustic Calciustolls Pachic Argiustolls Have calcic horizon within 100 cm or Have a mollic epipedon ≥ 50 cm petrocalcic horizon within 150 cm with no overlying argillic Aridic Calciustolls

Petrocalcic Calciustolls Have petrocalcic horizon within 100 cm

Haplustolls Cumulic Haplustolls Have mollic epipedon with irregular decrease in organic carbon

Pachic Haplustolls Mollic epipedon ≥ 50cm

Torriorthentic Haplustolls

Aridic Haplustolls

Paleustolls Petrocalcic Paleustolls Have argillic horizon with no clay decrease of Have petrocalcic horizon within 150 cm 20% and hue of 7.5 YR or redder, or 35% or more clay in its upper part

Vertisols Torrerts Haplotorrerts Chromic Haplotorrerts Have torric (aridic) moisture regime Have color value 4 (moist) or 6 (dry) or chroma ≥ 3

†Definitions are simplified. Refer to Soil Survey Staff (1999) for complete definitions.

Bulletin 798 • Page 34

gravel) materials and the other is in high-gravel (more than approximately 50 percent gravel) materials. Materials with intermediate contents of gravel have intermediate morphologies....

With continued carbonate accumulation, most or all pores and other openings in the soils become filled by carbonate; primary grains have been forced apart; bulk density has increased; and infiltration rate has markedly decreased. This process results in the plugged horizon, which develops in the last part of stage III [Figure 5].

After development of the plugged horizon, the lam-inar horizon forms on top of it. Differences between the laminar and plugged horizons are so great that the fabrics differ in kind. The laminar horizon has much more carbonate that the plugged horizon and essentially no allogenic skeletal grains. Rather than the carbonate being a filling between skeletal grains, it occupies almost the entire horizon and the skeletal grains are incidental. The laminar horizon is a new soil horizon in the sense that it consists almost entirely

of authigenic material and hence thickens the soil by its own thickness.... Infiltrating water concentrates at the top of the carbonate-plugged horizon to the extent that a thin zone of free water results. This water collects in hol-lows along the top of the plugged horizon and moves downward along faces of vertical prisms if they are present. Deposition of carbonate as these water films evaporate explains the thinness of the laminae and the filling of low spots in the upper surface of the plugged horizon [Figure 22A]. It also explains car-bonate coatings on prism faces and the occurrence of the laminar horizon only on a material of low perme-ability (other dense materials can substitute for the plugged horizon).

The numerous laminae suggest that accretion of carbonate in the laminar horizon is an episodic pro-cess that reflects many wettings and subsequent dry-ings.... Clay and organic matter suspended in down-ward-moving water are probably deposited on top of, and then incorporated in, the upward-developing laminar horizon.

Figure 21. Illustrative thicknesses of non-calcareous zone in Holocene soils as a function of increasing precipitation and elevation (from Gile, 1977, used with permission). The non-calcareous zone is not to scale.

Bulletin 798 • Page 35

The high genetic significance of the transition from the plugged horizon (which forms in the latter part of stage III) to the stage IV laminar horizon and the rela-tion to particle size are shown in dramatic fashion by abrupt facies changes from nongravelly to gravelly ma-terials (Figure 23). Such abrupt changes from stage III to stage IV horizons are caused by the abundant gravel, which greatly speeds the process of carbonate-plugging and formation of the stage IV laminar horizon. Associ-ated with the change from stage III (calcic) to stage IV (petrocalcic) horizons is a major change in soil classifi-cation, from Typic Haplocalcids to Typic Petrocalcids, respectively. Stage V petrocalcic horizons, which occur in soils of the JER and upper La Mesa surfaces, contain pisoliths (Figures 22A, 22B).

Effect of increasing precipitation on development of the sequencesA comparison of soils of the semiarid mountain canyons with soils of the arid valley border illustrates the effect of greater precipitation on development of the stages of carbonate accumulation. For example, soils of Jornada I—which have thick stage IV carbonate horizons in very gravelly materials of the arid valley border—have only stage I horizons in the semiarid mountain canyons. This stage I carbonate is thought to be of Holocene age because it is morphologically similar and occurs in similar textures and at similar depths as the carbonate of Holocene soils. It is thought that most or all carbon-ate leached from the Jornada I soils was removed in Pleistocene pluvials, since these soils lack the prominent carbonate accumulations in soils of the same age traced downslope.

Figure 22. A, landscape view of JER La Mesa just south of Goat Mountain (on the skyline). The top of the stage V carbonate horizon is exposed in the road. B, polished section of a fragment from the stage V horizon exposed in the road. Carbonate pisoliths are common, and some of them have coatings of laminar carbonate. The fragment is completely coated with laminar carbonate.

Bulletin 798 • Page 36

The character of the carbonate transition between the arid zone and the semiarid mountain canyons provides additional evidence of the effect of gradually increasing precipitation mountainward. The morphological change from stage IV in gravelly Jornada I sediments of the arid zone to stage I in the semiarid canyons does not take place all at once but instead occurs as a very gradual weakening and deepening of the carbonate accumula-tion towards the mountains.

In contrast, the ancient soils of Doña Ana surface, which stands above the adjacent Jornada I surface (Ruhe, 1967, p. 25) in the semiarid mountain canyon regions, do contain a stage IV carbonate horizon. This is attributed mostly to the ridge-crest occurrence of the Doña Ana surface and its effect on infiltration and penetration of the wetting fronts. In contrast to the Jor-nada I surface, which is relatively level transversely and little-eroded in many places, the ridge-crest position of the Doña Ana surface would tend to reduce both infil-tration and penetration of the wetting fronts over time,

thus lifting the zone of carbonate accumulation in soils of the Doña Ana surface.

The calcic and petrocalcic horizonsCarbonate morphology for initial development of the calcic and petrocalcic horizons differs according to car-bonate content and texture of the parent materials. In this context, high-carbonate parent materials are alluvial parent materials with more than about 15% CaCO

3

equivalent derived mainly from limestone bedrock (Gile, 1975a). In contrast, low-carbonate parent materi-als are alluvial parent materials with less than about 2% CaCO

3 equivalent.

In high-carbonate parent materials with finer texture and in non-gravelly materials, stage I horizons, with car-bonate occurring mostly as filaments, commonly qualify as calcic horizons. This is because the high-carbonate parent materials already have enough carbonate to meet the CaCO

3 requirements of a calcic horizon (Soil Survey

Staff, 1999).

Figure 23. Diagram showing the transition from nongravelly to gravelly sediments, and its relation to carbonate-plugging, formation of the stage III and IV horizons, and soil classification. The illustrative soils have formed in high-carbonate ma-terials, but the same morphogenetic relations are shown in soils formed in other materials. No boundary between the two soils is evident at the land surface, and the soils have formed in sediments of the same age (late Pleistocene) (modified from fig. 5 in Gile, 1975a, used with permission).

Bulletin 798 • Page 37

In low-carbonate parent materials, stage III horizons are commonly needed to meet the CaCO

3 requirements,

depending on texture. See definition of the calcic hori-zon (Soil Survey Staff, 1999) for the amount of carbon-ate required for different textures.

Stage III plugged horizons and stage IV, V, and VI horizons all qualify as petrocalcic horizons.

Micromorphology of carbonate accumulation Figure 24 illustrates the appearance of pedogenic car-bonates in thin section and with scanning electron microscopy. In crossed-polarized light, the carbonate is revealed to be subhedral and euhedral crystals generally ranging from 1–5 µm in diameter. In thin sections that are 30 µm thick, carbonate crystals display very high birefringence that produces a golden color. Carbonate is also identified by taking on a red color when stained with Alizarine Red S (Dickson, 1965) and by its remov-al with dilute HCl.

In gravelly soils, carbonate accumulation begins as stage I coatings on the surfaces of grains (Figure 24A). With increased accumulation, carbonate progressively fills the pores until a matrix is produced that is com-posed of an essentially continuous fabric of carbonate crystals (K-fabric). This occurs in the stage III morpho-genetic sequence (Figure 24B). At stage IV, laminae ap-pear in thin section as wavy layers composed of almost pure carbonate (Figure 24C).

Scanning electron microscopy observation of carbon-ate reveals that the crystals commonly occur as tubular filaments (Figure 24D) and sometimes as needle fiber calcite. These microbiological forms of carbonate, in addition to the common occurrence of carbonate on roots, suggest the important influence that microbes and plants have on the formation of pedogenic carbonate.

The K horizonThe first paper to be published from the Desert Project was on classification of ca horizons (Gile, 1961). At that time, horizons of carbonate accumulation were designated Cca or Ccam (Soil Survey Staff, 1951). The K horizon concept (Gile et al., 1965) was stimulated by the prominence of highly developed carbonate horizons, of which Ruhe (1967, p. 55) remarked, “A feature com-mon in the soils in the area, whether on alluvial fans, piedmont, aprons, or on basin or valley-border surfaces, is a subsoil horizon of calcium carbonate concentration. In most soils, this horizon is so prominent that it is the first feature to catch an observer’s eye.” The K horizon is based on the concept of K-fabric, in which “fine-grained authigenic carbonate coats or engulfs skeletal pebbles, sand, and silt grains as an essentially continuous me-

dium” (Gile et al., 1965, p. 74). Quoting from Gile et al. (1965, p. 75):

A minimum ranging from about 15 to 40 percent authigenic carbonate (CaCO

3 equivalent calculated

as percentage of fine earth material) is required for the development of K-fabric; least carbonate is re-quired in gravelly, coarse materials; most is required in fine-textured materials. Presence of limestone fragments or other allogenic carbonate confounds the relationship between chemically determined carbon-ate percentage and expression of K-fabric properties, because authigenic carbonate cannot be distinguished clearly from allogenic carbonate by chemical analysis.

High carbonate content is an important character-istic of material exhibiting K-fabric. Specific carbon-ate percentage is not a diagnostic property because it cannot be determined in the field, because authigenic and allogenic carbonate cannot be distinguished ana-lytically, and because wide variations in the necessary carbonate content occur. The properties which most simply define the K horizon are those of its soil fabric. These fabric properties are essentially the same for soil materials with or without allogenic carbonate grains, since such grains tend to act as skeletal material. Thus, the K horizon may be defined on the basis of approximate volumes of K-fabric, that is, volumes of material in which carbonate coats or engulfs skeletal grains and forms an essentially continuous medium.

The K2 horizon was defined as the most prominent subhorizon of the K horizon, with the K1 horizon (above the K2) and the K3 (below the K2) being transi-tional to the K2 in many soils. On the other hand, some users of the K horizon simply number the subhorizons consecutively without designation of transitional hori-zons per se (e.g., Ruhe, 1967, p. 58–59; Birkeland et al., 1991, p. 3; Machette, 1997). See Gile et al. (1965) for a more detailed discussion of the K horizon.

Totals of pedogenic carbonate in soilsMany soils in the Desert Project and in other arid re-gions contain very large amounts of carbonate despite having formed in parent materials with little calcium. For these soils, virtually all of the carbonate must have been derived from atmospheric additions. Because the amount of carbonate in the soils increases with increas-ing age (Figure 25), totals of pedogenic carbonate can be a very important indicator of soil age (Gile et al., 1971, 1981, 1998; Gardner, 1972; Bachman & Machette, 1977; Machette, 1978, 1985; Mayer et al., 1988; Mari-on, 1989; Gile, 1987b, 1990, 1993, 1994a, 1995, 1999).

Bulletin 798 • Page 38

The calculation for pedogenic carbonate is for a vol-ume element one square meter in horizontal cross sec-tion and of variable thickness according to the formula

CaCO3 (kg/m2) = {L × Db × [1 - (> 2mm vol %/100)] × CaCO

3}/10

where L is the thickness of the horizon in cm, Db is the bulk density of the fine-earth fabric, 1 - (> 2mm vol %/100) is a correction for the volume occupied by the > 2mm material, and CaCO

3 is carbonate content

of the horizon minus the carbonate content of the parent material.

Totals of pedogenic carbonate for soils of various ages are presented in Table 3. When using pedogenic carbonate for chronological purposes, it is essential that the parent material contain very little or no carbonate,

and that landscape position and parent material texture be similar; otherwise, substantial variations in total car-bonate can occur in soils of the same age (Gile, 1993, 1994a, 1995; Table 3).

Horizons of silicate clay accumulationIn addition to the petrocalcic and calcic horizons, the argillic horizon is an important diagnostic horizon for soil classification in arid regions. By definition, some of the clay in the argillic horizon must be illuvial and it must have more clay than the overlying eluvial horizon, depending upon clay content of the latter (Soil Survey Staff, 1998). The major criterion for recognition of the argillic horizon in the Desert Project region is the pres-ence of oriented clay on sand grains (and on pebbles if present) and maximum expression of the oriented clay

Figure 24. Thin section showing micromorphology of carbonate. A, carbonate coatings on grains in a Typic Petroargid on the lower La Mesa surface; B, carbonate filling interstitial spaces between grains; C, laminar layers composed mostly of car-bonate; D, scanning electron micrograph of petrocalcic horizons showing calcified filaments.

Bulletin 798 • Page 39

in the horizon of maximum clay content (Gile & Gross-man, 1968). Thin-section studies have shown that most argillic horizons in this area easily meet the requirement of at least 1% of oriented clay (Soil Survey Staff, 1998). Evidence of clay illuviation and the argillic horizonHorizons of silicate clay accumulation (which are often, but not always, synonymous with the argillic hori-zon) typically are just above or extend into horizons of carbonate accumulation. The clay accumulation was formerly thought to be due to in situ weathering rather than illuviation (Nikiforoff, 1937; Brown & Drosdoff, 1940; Agriculture Experiment Station-Soil Conservation Service, 1964). However, later work indicates an illuvial origin for much of the clay despite the absence of clay

skins (argillans) on peds and in pores (Gile & Gross-man, 1968; Smith & Buol, 1968; Nettleton et al., 1969; Nettleton et al., 1975). Several lines of evidence indicate that horizons of clay accumulation in the Desert Project contain illuvial clay and that the absence of clay skins is caused by unfavorable conditions for their formation or preservation rather than by lack of illuvial clay. Quoting from Gile et al. (1981, p. 72):

1) Soils of stable sites have a thin, grayish [E]-like horizon with less clay than the underlying red-dish-brown or red B horizon. The grayish color is most apparent when dry.

Figure 25. Soil carbonate as a function of the morphogenetic stages illustrating the effect that soil age has on the amount and thickness of carbonate accumulation in nongravelly materials. I, stage I of carbonate accumulation of the Fillmore surface; II, stage II of the Leasburg surface; IIIA, stage III (not plugged) of the Picacho surface; IIIB, stage III (plugged) of the lower La Mesa surface; V, stage V of the upper La Mesa surface (from Gile, 1970, used with permission).

Bulletin 798 • Page 40

2) Prominent coatings of oriented clay on sand grains and pebbles are a distinctive micromorpho-logical feature of all Bt horizons in the study area. These coatings do not of themselves demonstrate clay movement subsequent to initiation of soil de-velopment because some of the clay on the grains could be inherited from parent materials. In this respect the oriented coatings differ from clay skins on ped surfaces, which post-date the ped surface and hence must have formed after development started. The oriented coatings do provide evidence for illuviation where their maximum expression coincides with the clay maximum....

3) Prominent clay skins do occur in many pipes of Bt material that penetrate horizons of carbon-ate accumulation. Their presence deep in the soil where water content is nearly constant and roots and fauna seldom penetrate suggests that the absence of clay skins in Bt horizons at shallower depths may be due to physical disruption associ-ated with biotic activity and shrinking and swell-ing as the soil dries and then is wetted.

4) Clay skins also occur in certain soils of the semi-arid portion of the study area. This suggests that clay skins would form in soils of the arid part of the study area if soils were wetter.

5) Reddish coatings of silicate clay have been ob-served on and in cracks in the tops of petrocalcic horizons that underlie argillic horizons at shallow depths....This clay must have illuviated from the overlying B horizon. Similar but less prominent coatings occur in upper parts of many calcic hori-zons that underlie argillic horizons.

6) Distinct linear bodies of oriented clay occur within many peds. Some of these linear bodies are interpreted as former clay skins, now inside the peds because of the development of new faces as the soil wetted and dried. Buol and Yesilsoy (1964) and Nettleton et al. (1969) have made the same interpretation.

7) Wetting and drying of a mixture of sand grains and silt-size aggregates of clay did not produce coatings of oriented clay on the sand grains (Gile and Grossman, 1979). Formation of such coatings evidently required prior disaggregation of the clay, which would have been its state

during illuviation. This finding agrees with work of Thorp and others (1957, 1959). In leaching

experiments dealing with aspects of clay movement they state: “...clay is brought into suspension and moves through the soil as individual clay particles.”

8) Only slight weathering of primary minerals has been found in the oldest argillic horizons; little clay apparently has been produced by weathering in place.

9) Some clay was apparently derived from atmo-spheric additions, particularly in pervious sedi-ments....Such clay would be illuvial because it must have moved from the surface downward.

10) The clay increase from A to B consists largely of fine clay. Changes in the fine to total clay meet the requirements of the argillic horizon (Soil Survey Staff, 1975).

11) The positional relation of the horizon of silicate clay accumulation to the horizon of carbonate accumulation suggests illuviation. In Holocene soils (which are less complicated than older polygenetic soils), the horizon of silicate clay ac-cumulation is just above or extends slightly into the carbonate horizon. This arrangement would be expected on a theoretical basis if the accumu-lations were illuvial. Clay moves downward in suspension (Thorp and others, 1957, 1959) and would be deposited at the base of the zone that is wetted by water movement rapid enough to maintain the clay in suspension. Calcium bicar-bonate, being in solution rather than suspension, would be expected to move deeper than the clay and then to precipitate below it as the soil solu-tion dries.

12) In sandy and sandy-loam parent materials, clay increases with increasing soil age back to the late Pleistocene. Because little evidence of weathering has been observed, an illuvial origin for some of the clay is strongly implied. However, the increase in clay with age does not accord well in soils older than the late Pleistocene (Gile and Grossman, 1979).

Although clay illuviation occurs in the Desert Project, many soils do not have argillic horizons. Parent materi-als have an important control on clay illuviation because argillic horizons have not been found in materials that contain abundant fragments of calcareous rocks, such as limestone. For example, soils with limestone fragments

Bulletin 798 • Page 41

on the late-Pleistocene Picacho surface have a petrocal-cic horizon, but no argillic horizon, even in stable areas with very little evidence of erosion. In contrast, soils of the Picacho surface formed in rhyolite alluvium have both a petrocalcic horizon and a distinct argillic horizon (Gile & Hawley, 1972, p. 122). The explanation appar-ently lies in the flocculating effect carbonate has on clay movement, although clay illuviation can occur in calcar-eous soils with large pores (Goss et al., 1973).

Micromorphology of the argillic horizonFigure 26A illustrates coatings of illuvial silicate clay (grain argillans) on sand grains and pebbles. Like car-bonate crystals, grain argillans have a gold color in thin section, but generally are redder. Unlike carbonate crystals, however, they lack discrete crystal boundaries, occurring instead as continuous layers on grain surfaces showing optical extinction that runs along the coating when the stage of the petrographic microscope is rotat-ed. Also, unlike carbonate coatings, they are unaffected by staining with Alizarine Red S stain or HCl.

Grain argillans first occur as thin coatings in cambic horizons of Holocene age (Gile et al., 1995a). With suf-ficient clay accumulation to form a prominent argillic horizon, as in many Pleistocene soils, grain argillans are prominent (Figure 26A). Obliteration of the argillic horizonLandscape dissection is one of the major causes of de-struction of the argillic horizon (Gile, 1975a). This oc-curs where surfaces that were once stable enough for the formation of argillic horizons were later eroded. Climate change, loss of vegetative cover, and land use changes increase slope erosion and can cause truncation of argillic horizons.

Engulfment of argillic horizons by carbonate is an-other process that destroys argillic horizons. This engulf-ment is commonly associated with erosion and greater aridity. Another mechanism is an upward shift in the depth of wetting due to drier climate with resultant car-bonate precipitation in the Bt fabric of Pleistocene soils (Gile et al, 1969). Field evidence and thin sections indi-cate that the formation of carbonate crystals pushes clay particles and disrupts orientation of silicate clay films (Gile et al., 1981, p. 74). With continued carbonate ac-cumulation, grain argillans tend to become obliterated (Figure 26B).

Faunal mixing is another process that destroys argil-lic horizons. Tunnels and mounds made by kangaroo rats, badgers, and termites destroy Bt fabric (Gile et al., 1981, p. 74). This process occurs most extensively in soils that are sandy and have low amounts of rock frag-ments, and thus would presumably be the easiest soils to

dig. Faunal mixing will be discussed further in a later section.

Horizons of organic carbon accumulation and soils of the semiarid mountainsSemiarid mountains occur upslope of the arid basins. The mollic epipedon (Soil Survey Staff, 1998) and the Mollisols occur only in the semiarid mountains. How-ever, certain soils of the arid basin floors and lower pied-mont slopes have as much or more organic carbon than the Mollisols (Table 7). The high amount of organic carbon in these soils is the result of finer textures and/or run-in from upslope, but horizons of these soils gener-ally are not dark enough for a mollic epipedon.

The following is a summary discussion of some of the semiarid soils and soils of the arid-semiarid transition (from Gile, 1977; illustrative data, additional informa-tion, and diagrams may be found in this reference and in Gile et al., 1981).

The arid-semiarid transition is marked by increases in elevation and precipitation, thickening and dark-ening of A horizons, increase in depth of wetting, and a change in vegetation.

Nearly all Holocene soils in the arid basin and val-ley are Aridisols or Entisols; Mollisols do not occur in soils of any age. In the semiarid mountains, however, most Holocene soils are Mollisols with thick, dark A horizons.... The Aridisol-Mollisol transition occurs in several ways. One involves Holocene soils of stable sites in which the landscape is level or nearly level trans-versely. Organic carbon gradually increases and color darkens with increasing elevation; finally the Holo-cene Aridisols change to Holocene Mollisols [Table 8]. In another kind of Aridisol-Mollisol transition, Ho-locene Aridisols occur on narrow ridges well within the semiarid zone. On these narrow ridges the mollic epipedon has been truncated or did not form [Table 9]. Laterally the Holocene Aridisols change abruptly to Holocene Mollisols on adjacent terraces that are level transversely.

Another Aridisol-Mollisol transition involves Holocene soils and soils of Pleistocene age. In these cases, soils of Pleistocene fans have been truncated and Bt horizons with chromas too high for a mollic epipedon are at shallow depth. These soils are [Ustic Haplargids or Calciargids], which are intergrades to the Mollisols. Low, stable terraces with Holocene Mollisols are inset against the higher Pleistocene fans.

Bulletin 798 • Page 42

This results in an abrupt boundary between Holocene Mollisols and Aridisols of Pleistocene age.

Soil and soil-geomorphic boundariesBoundaries between soils are basic features that are closely related to the soil-forming factors of topography, time, parent materials, climate, and organisms. Thus, the occurrence of soil boundaries is of major significance in soil-geomorphic studies. The wide expression of the soil-forming factors in the Desert Project permitted holding certain factors constant while varying one or more of the other factors to determine their effect on soil properties and on soil boundaries. Two papers dis-cuss some of the main causes of soil boundaries (Gile, 1975a, 1975b). This section summarizes the main points of those papers.

Formation of geomorphic surfaces is directly related to the soil-forming factor of time (Table 3), and thus has a major effect on soil morphology, genesis, and clas-sification. Because soils of different geomorphic surfaces differ, often profoundly, some soil boundaries coincide with geomorphic surface boundaries. These coincident soil-geomorphic boundaries range from prominent and abrupt (e.g., see Figure 17) to indistinct and gradual (see Figure 10).

Boundaries caused by development of the argillic horizonPresence or absence of the argillic horizon distinguishes soils taxonomically at categorical levels ranging from order to subgroup (Table 6). Thus, identification of the

Figure 26. Thin section showing grain argillans. A, argillans in a non-calcareous Bt horizon formed in rhyolite alluvium in a Typic Calciargid on the Jornada I surface. B, argillan on quartz sand grain that has been partially obliterated by carbon-ate accumulation.

Bulletin 798 • Page 43

Tabl

e 7.

Lab

orat

ory

Dat

a Il

lust

rati

ng t

he R

elat

ivel

y H

igh

Am

ount

s of

Org

anic

Car

bon

and

Cla

y in

Soi

ls o

f th

e Jo

rnad

a B

asin

Flo

or N

orth

of

Hig

hway

70

(fro

m

Gile

et

al.,

1981

, p. 1

84)

C

arbo

nate

< 0.

002

Ext

ract

able

E

xtra

ctab

le†

Bul

k W

ater

O

rgan

ic

D

epth

Sa

nd

Cla

y

Fine

/tot

al

< 2m

m

mm

* so

dium

po

tass

ium

de

nsit

y re

tent

ion

carb

onA

lluvi

um

Hor

izon

(c

m)

(%)

(%)

clay

(%

) (%

) (%

) (m

e/10

0 g)

(%

) (g

/cc)

di

ffer

ence

(%

)

Ust

ollic

Cal

cior

thid

(R

eaga

n va

rian

t 66

-7)

Org

an

A

0–8

43

26

0.24

12

2

1.37

0.

19

1.17

A

8–

18

41

27

0.22

13

5

1.37

0.

19

0.66

B

21ca

18

–30

21

41

0.21

19

8

1.25

0.

21

0.66

B

22ca

30

–41

41

32

0.29

15

7

0.52

2C

41

–51

75

17

0.35

6

3

0.

26

Jorn

ada

II

3B2c

ab

51–6

1 45

38

0.

40

25

14

1.41

0.

11

0.44

3K

2b

61–8

9 62

24

0.

45

35

16

1.49

0.

15

0.30

3C

1cab

89

–112

25

1.

80

0.15

0.

12

3C

2cab

11

2–14

0

16

1.

75

0.12

4C

3cab

14

0–17

5

15

O

rgan

ic c

arbo

n, 6

.2 k

g/m

2 to

89 c

m

Ust

ollic

Cal

cior

thid

(R

eaga

n 66

-6)

Org

an

A

0–8

44

19

0.16

9

2

1.

43

0.16

0.

90

B

1 8–

15

22

30

0.11

16

5

1.32

0.

19

0.74

B

21

15–3

3 15

40

0.

11

18

7

1.

27

0.17

0.

70

B

22ca

33

–48

14

45

0.14

16

8

1.35

0.

15

0.66

B

23ca

48

–64

19

46

0.19

15

8

1.41

0.

12

0.69

Jorn

ada

II

B21

cab

64–8

1 48

32

0.

26

13

6

1.

59

0.10

0.

37

B

22ca

b 81

–114

43

36

0.

43

21

11

1.40

0.

14

0.34

K

2b

114–

135

45

28

0.50

35

18

1.

52

0.16

0.

23

C

1cab

13

5–16

8 45

25

0.

40

31

11

1.72

0.

13

0.10

C

2cab

16

8–18

5 58

23

0.

32

19

6

C

3b

185–

208

60

19

0.41

20

5

2C

4b

208–

259

27

30

0.33

28

8

1.39

0.

22

2C

4b

259–

300

25

34

0.24

28

8

1.29

0.

21

2C

5b

300–

325

21

38

0.23

28

8

3C

6b

325–

348

49

28

0.28

17

6

1.52

0.

17

Jorn

ada

I 3B

2cab

2 34

8–36

3 62

25

0.

32

13

6

3K

21b2

36

3–37

3 54

30

0.

61

44

23

3K

22b2

37

3–39

6 55

30

0.

66

50

28

1.45

0.

22

3K

23b2

**

396–

437

28

3C

1cac

sb2*

* 43

7–45

7

6

3C

2csb

2**

457–

518

3

O

rgan

ic c

arbo

n, 8

.8 k

g/m

2 to

114

cm

Bulletin 798 • Page 44

Tabl

e 7.

Lab

orat

ory

Dat

a Il

lust

rati

ng t

he R

elat

ivel

y H

igh

Am

ount

s of

Org

anic

Car

bon

and

Cla

y in

Soi

ls o

f th

e Jo

rnad

a B

asin

Flo

or N

orth

of

Hig

hway

70

(fro

m

Gile

et

al.,

1981

, p. 1

84)

C

arbo

nate

< 0.

002

Ext

ract

able

E

xtra

ctab

le†

Bul

k W

ater

O

rgan

ic

D

epth

Sa

nd

Cla

y

Fine

/tot

al

< 2m

m

mm

* so

dium

po

tass

ium

de

nsit

y re

tent

ion

carb

onA

lluvi

um

Hor

izon

(c

m)

(%)

(%)

clay

(%

) (%

) (%

) (m

e/10

0 g)

(%

) (g

/cc)

di

ffer

ence

(%

)

Ust

ollic

Cal

cior

thid

(R

eaga

n 60

-17)

Pett

s’ Ta

nk

A

0–8

26

34

0.13

10

5

tr

11

1.01

B

21

8–20

27

42

0.

18

12

6 0.

1 14

0.

84

B

22

20–4

3 35

34

0.

30

12

7 0.

1 13

0.

65

B

23ca

43

–76

5 55

0.

24

12

6 0.

2 12

0.

62

K

2 76

–112

19

52

0.

33

29

19

0.3

9

0.

24

C

1ca

112–

142

12

51

0.22

32

16

0.

5 8

0.12

C

2ca

142–

190

19

48

0.23

29

11

Jorn

ada

I B

tbca

19

0–20

6 50

36

0.

41

7 1

O

rgan

ic c

arbo

n, 8

.8 k

g/m

2 to

112

cm

Ust

ollic

Hap

larg

id (

Stel

lar

60-2

1)

Jorn

ada

I A

2 0–

8 32

37

0.

07

2

tr

12

1.22

A

3 8–

13

32

45

0.23

tr

tr

11

0.65

B

1t

13–2

5 32

47

0.

30

tr

tr

11

0.

57

B

21t

25–5

1 31

44

0.

29

3

tr

12

0.40

B

22t

51–7

9 31

49

0.

47

6 2

0.1

11

0.27

K

1 79

–99

36

48

0.51

24

14

0.

1 10

0.

17

K

21

99–1

30

59

29

51

tr

7

K

22

130–

155

64

25

46

0.1

7

K

23

155–

178

66

23

49

0.1

7

K

24

178–

216

68

21

36

0.1

8

K

3 21

6–25

4 70

20

12

0.

1 6

C

1ca

254–

284

78

15

9

0.

1 6

2C

2 28

4–30

5 87

9

1

0.

2 6

O

rgan

ic c

arbo

n, 6

.0 k

g/m

2 to

99 c

m

* P

erce

ntag

e in

< 2

mm

.

† P

erce

ntag

e of

cat

ion

exch

ange

cap

acit

y.

**G

ypsu

m, r

espe

ctiv

ely,

zer

o, 8

0, a

nd 7

0 pe

rcen

t, in

low

er th

ree

hori

zons

.

Bulletin 798 • Page 45

Tabl

e 8.

Lab

orat

ory

Dat

a of

Soi

ls F

orm

ed in

Rhy

olit

ic A

lluvi

um I

llust

rati

ng t

he I

ncre

ase

of O

rgan

ic C

arbo

n w

ith

Incr

ease

of

Ele

vati

on (

from

Gile

, 197

7)

Part

icle

siz

e di

stri

buti

on (

mm

)a

Sand

Si

lt

Cla

y >

2 vo

l (%

) C

aCO

3 equ

ivb

Org

anic

car

bonc

Hor

izon

D

epth

(cm

) 2–

0.05

(%

) 0.

05–0

.002

(%

) <

0.00

2

(%

)

Tran

siti

on: T

ypic

Hap

larg

id (

Pin

alen

o, N

o. 1

); e

lev.

4,7

30 f

t (1

,442

m)

A2

0–5

68

23

8 50

tr

(s)d

0.18

B1t

5–

18

67

22

11

50

tr(s

) 0.

28

B2t

18

–30

65

20

15

65

tr(s

) 0.

30

B3t

30

–51

68

19

13

65

tr(s

) 0.

23

C1c

a 51

–71

72

19

9 65

2

0.11

C2c

a 71

–94

75

17

8 65

1

0.07

C3

94–1

47

86

11

4

1 0.

03

Btb

14

7–17

8 45

21

34

1 0.

11

Sem

iari

d zo

ne: P

achi

c H

aplu

stol

l (Sa

nto

Tom

as, N

o. 2

); e

lev.

5,7

00 f

t (1

,737

m)

C

0–3

66

26

8 40

tr

(s)

1.00

A11

3–

8 59

30

12

50

tr

(s)

1.32

A12

8–

28

59

29

12

50

tr(s

) 1.

18

A13

28

–53

61

27

13

45

tr(s

) 0.

81

A3

53–7

9 60

28

12

45

tr

(s)

0.47

C

79–1

04

64

25

11

50

tr

0.30

a Met

hod

3A (

Soil

Con

serv

atio

n Se

rvic

e, 1

972)

. Car

bona

te n

ot r

emov

ed.

b Met

hods

6E

1b, 6

E2A

.c M

etho

d 6A

1a.

d Tra

ce, d

etec

ted

only

by

qual

itat

ive

proc

edur

e (e

volu

tion

of g

as b

ubbl

es o

bser

ved

unde

r bi

nocu

lar

mic

rosc

ope

on a

ddit

ion

of 6

N H

2SO

4) m

ore

sens

itiv

e th

an q

uata

tive

pro

cedu

re u

sed.

Bulletin 798 • Page 46

Tabl

e 9.

Lab

orat

ory

Dat

a Il

lust

rati

ng O

rgan

ic C

arbo

n D

iffe

renc

es B

etw

een

Nei

ghbo

ring

Sem

iari

d A

ridi

sols

and

Mol

lisol

s Fo

rmed

in M

onzo

niti

c Pa

rent

Mat

eria

l (f

rom

Gile

, 197

7)

Pa

rtic

le s

ize

dist

ribu

tion

(m

m)a

Sand

a Si

lta

Cla

ya

O

rgan

ic c

arbo

nc

Hor

izon

D

epth

(cm

) 2–

0.05

(%

) 0.

05–0

.002

(%

) <

0.00

2 >

2 vo

l (%

) C

aCO

3 equ

ivb

(%)

Mol

lisol

: Tor

rior

then

tic

Hap

lust

oll (

Ala

ddin

, No.

3);

ele

v. 5

,500

ft

(1,6

76 m

)

A11

0–

5 71

22

7

15

—(s

)d 0.

8

A12

5–

36

67

23

11

15

—(s

) 0.

6

A13

36

–53

70

19

11

15

—(s

) 0.

5

A14

53

–89

69

20

11

20

0.

4

A15

89

–117

73

17

9

20

0.

3

AC

11

7–14

7 68

20

11

15

tr

(s)

0.2

C1

147–

173

76

15

9 20

—

(s)

0.1

C2

173–

218

78

14

8 15

0.1

Mol

lisol

: Tor

rior

then

tic

Hap

lust

oll (

Haw

keye

, No.

4);

ele

v. 5

,125

ft

(1,5

62 m

)

A11

0–

3 72

20

8

5 tr

(s)

1.0

A12

3–

20

81

12

6 15

tr

(s)

0.7

A13

20

–41

81

13

6 15

tr

(s)

0.5

A14

41

–71

81

13

7 15

tr

(s)

0.3

C1

71–9

9 74

18

8

15

tr(s

) 0.

3

C2

99–1

32

73

19

9 15

0.

1 0.

2

C3

132–

173

79

15

7 20

0.

6 0.

2

Ari

diso

l: Ty

pic

Hap

larg

id (

Sono

ita,

No.

5);

ele

v. 5

,000

ft

(1,5

24 m

)

A

0–8

82

12

6 10

tr

0.

2

B21

t 8–

23

75

13

12

15

—

0.2

B22

t 23

–48

75

13

12

30

tr

0.2

B3

48–8

1 83

11

7

20

tr(s

) 0.

1

2Cca

81

–102

81

13

6

25

1

a Met

hod

3A1

(Soi

l Con

serv

atio

n Se

rvic

e, 1

972)

. Car

bona

te n

ot r

emov

ed.

b Met

hods

6E

1b, 6

E2A

.c M

etho

d 6A

1a.

d —

(s)

= N

one

dete

cted

by

sens

itiv

e qu

alit

ativ

e te

st.

Bulletin 798 • Page 47

argillic horizon is critically important to soil classifica-tion and boundaries. It should be stressed that many argillic horizons have prominent morphologies and are not difficult to identify. In the Desert Project and similar areas, the reddish-brown or red, relatively clayey horizons of silicate clay accumulation easily qualify as argillic horizons. It is in the transitional stages between argillic horizons and other horizons in the B position that identification is less certain. The occurrence of the argillic horizon in some areas is complex because of major environmental changes since an argillic horizon formed.

Boundaries caused by development of the argillic horizon are highly dependent on soil age and carbonate content of the parent materials. In the developmental scheme for soils of the Desert Project, the argillic hori-zon first appears in Holocene soils that have formed in low-carbonate parent materials. In these materials, slow illuviation of clay appears to start soon after deposition of parent materials has ceased. Progressive clay accu-mulation is accompanied by gradual reddening of the B horizon. By the time the clay increase from A to B is enough for an argillic horizon, it is commonly non-calcareous and has 5YR hue in sandy loams. In finer textured horizons (such as loam or sandy clay loam), colors tend to be somewhat less red in Holocene soils, and the argillic horizon may have hue of 7.5YR. Not all B horizons with these hues have enough clay increase for an argillic horizon, though most do.

Carbonate content of the parent materials, as dis-cussed above, is also important. The argillic horizon has not developed at all in sediments with abundant frag-ments of high-carbonate rocks, such as limestone. This is true even in soils of Pleistocene age that must have developed partly in times of greater effective moisture. Although clay may increase with depth in some of these soils, their high carbonate content precludes recognition of the required amount of oriented clay in thin section.An argillic horizon can develop in some soils of Pleis-tocene age (though not of Holocene age) if the high-carbonate parent materials contain few or no fragments of calcareous rocks and are on the low end of the high-carbonate range. In these materials, enough oriented clay for an argillic horizon is indicated if the parent ma-terials (of 10YR hue) have reddened to 5YR hue. If the reddening has proceeded only to 7.5YR, there is usually so much carbonate that the required amount of oriented clay cannot be identified in thin section.

Boundaries caused by differences in soil age and car-bonate content of parent materials (Figure 27)As illustrated in Figure 27A (vicinity I in Figure 1), there is a change from high-carbonate materials derived

from the San Andres Mountains to low-carbonate mate-rials (monzonite alluvium) derived from the San Agustin Mountains. On lower slopes of the fan-piedmont, the area of monzonite materials has little transverse relief ex-cept for occasional gullies and slight ridges. In contrast, the landscape in the high-carbonate materials common-ly has a distinctive terrain with vertical or near-vertical scarps up to 1 m in height (Figures 27B, 27C).

Difference in age (high-carbonate parent materials); boundary between Torrifluvents and Haplocalcids (Figure 27B)Torrifluvents and Torriorthents are dominant in high-carbonate alluvium of Holocene age. These soils have 10YR hue and are strongly calcareous throughout; no reddish-brown B horizons have formed because of the high carbonate content of the parent materials. Textures are silt loam, clay loam, or silty clay loam. These soils have structural B horizons but not cambic horizons since there is very little or no visible pedogenic carbon-ate (Soil Survey Staff, 1998). A Bk horizon (with car-bonate filaments on ped faces) is present in places and illustrates incipient development of the calcic horizon in high-carbonate parent materials.

The boundary between the Holocene soils and soils of late Pleistocene age is marked in many places by scarps (Figure 27B) that are cut mainly in the Holocene soils. Perpetuation of the vertical scarps is promoted by the silty textures, the vertical faces of prisms in the soil, and the thick grass cover that is usually present along the top of the scarp.

The soils of late Pleistocene age, buried in many places by the Holocene deposits, commonly emerge at the surface below the scarps. These soils have thick cal-cic horizons and are Haplocalcids (Figure 27B). They usually have brown cambic horizons with 7.5YR hue (where not truncated), and thus illustrate a reddening in the B horizon not found in Holocene soils formed in similar materials. This reddening is thought to have been caused by the increased effective moisture of the late Pleistocene pluvial. However, there is still so much carbonate in the illustrative soil that insufficient orient-ed clay for the argillic horizon (Soil Survey Staff, 1999) can be seen in thin section. The reddening reaches 5YR in places, and there is enough oriented clay for an argillic horizon.

Difference in age (low-carbonate parent material); boundary between Haplocambids and Calciargids (Figure 27C)—initial development of the cambic and argillic horizons The Holocene soils have non-calcareous, reddish-brown B horizons not found in high-carbonate parent materi-

Bulletin 798 • Page 48

Figure 27. Illustration of the effect of high-carbonate parent material on soil formation in Organ alluvium (late and middle Holocene) and Jornada II alluvium (late Pleistocene) (vicinity I in Figure 1). A, diagram of geomorphic setting illustrating the low-carbonate parent materials from the San Agustin Mountains on the right and high-carbonate parent materials from the San Andres Mountains on the left. Rectangle locates lower block diagram. B & C, comparison of land-surface and buried soils, soil horizons, and topography of high-carbonate parent material on left with low-carbonate par-ent material on right (from Gile, 1975a, used with permission).

Bulletin 798 • Page 49

als of the same age. The B horizon is a sandy loam and is underlain by a stage I carbonate horizon. Commonly, the B horizon has a clay increase and is a Bt horizon. The clay increase in the illustrative soil is too slight for an argillic horizon; the Bt is a cambic horizon and the soil is a Typic Haplocambid (Figure 27C). In other places, the increase in clay from A to B is enough for an argillic horizon; these soils are Typic Haplargids and illustrate initial development of the argillic horizon in these parent materials. The Haplargids occur where there is more clay in the parent materials and where the soils date from earlier in the Holocene.

Boundaries between the Holocene soils and the soils of late Pleistocene age occur at the margin of slight Holocene ridges, where the latter soils emerge at the surface (Figure 27). All the soils of late Pleistocene age have reddish-brown and red argillic horizons, prominent calcic horizons, and are Typic Calciargids. The distinct differences in morphology, compared to the Haplo-cambids, are attributed to greater age and development partly during a Pleistocene pluvial.

Difference in carbonate content of Holocene parent materials; boundary between Haplocambids and TorrifluventsIn places along the piedmont slope, Holocene soils in transitional areas between low- and high-carbonate par-ent materials contain a mixture of both. With increas-ing carbonate content of the parent materials, the zone of transition between these soils is characterized by a change from Haplocambids or Haplargids with reddish-brown B horizons (low-carbonate parent materials) to Torrifluvents lacking these horizons. Parent materials of the Torrifluvents are still dominated by monzonite, but enough carbonate is present to prevent development of the reddish-brown B horizon. The exact amount of carbonate required to prevent this development is not known, but is probably between 2% and 15% for Holocene soils.

Difference in carbonate content of late Pleistocene parent materials; boundary between Calciargids and Haplocalcids (Figures 27B, 27C)—initial development of the argillic horizon in high-carbonate parent materialsIn late Pleistocene alluvium, the transition from the prominent Calciargids in low-carbonate (monzonite) al-luvium to the Haplocalcids in high-carbonate alluvium (Figures 27B, 27C) corresponds to bedrock differences in the mountains upslope (Gile et al., 1981). With increasing carbonate in the parent materials, the bound-ary zone from Calciargids to Haplocalcids is marked by these changes, commonly in the following order: the

argillic horizon becomes calcareous throughout, macro-scopic carbonate appears in all subhorizons of the argil-lic horizon, and colors become less red, finally reaching the 7.5YR or 10YR hues characteristic of the cambic horizon in Haplocalcids of the area. Reddening of the B horizon reaches 5YR in places, illustrating initial and sporadic development of the argillic horizon and the Calciargids in these high-carbonate parent materials.

The precise location of the boundary between the Calciargids and the Haplocalcids is not well marked on the landscape, except that it usually occurs near the bor-der of the scarped terrain that characterizes much of the area of high-carbonate materials. The general location of the boundary between high- and low-carbonate parent materials is apparent in aerial photographs (e.g., Fig-ure 28). North and northeast of the playa is a band of mostly light-colored, high-carbonate sediments derived from the San Andres Mountains, which contain abun-dant carbonate rocks such as limestone. High-carbonate sediments tend to be finer-textured and to have lower infiltration rates than do the low-carbonate sediments to the south, and probably are more susceptible to erosion. The light colors in the band are due to sparsity of veg-etation and light color of the sediments. The scarcity of vegetation is apparently associated with recent erosion.

Boundaries caused by differences in soil age and particle size (low-carbonate parent materials) (Figures 29A, 29B)Figure 29 (vicinity II in Figure 1) illustrates soil bound-aries caused by differences in soil age and particle size in low-carbonate parent materials (monzonite alluvium). Large Pleistocene fans are major topographic features (Figure 29A). The fans represent several stages of fan de-velopment during late to middle Pleistocene, with only minor areas of Holocene age. A Holocene fan-piedmont is downslope from the Pleistocene fans. In this area, the Holocene alluvium overlies fan-piedmont deposits of late Pleistocene age that emerge farther downslope (in places, the Holocene alluvium rests on an alluvium intermediate in age between the units shown in Figure 29B). While this area illustrates boundaries due to soil age and particle size, slight increases in moisture have also affected soils nearest the mountains. This increase is indicated by a thickening of B horizons and an increas-ing depth to carbonate toward the mountains in Holocene soils. Difference in soil age; boundary between coarse-loamy Haplargids and fine-loamy Calciargids (Figure 29B)Coarse-loamy Haplargids have formed in the Holocene alluvium that occurs as slight ridges in the coalescent fan-piedmont (Figure 29A). These soils have sandy loam

Bulletin 798 • Page 50

Figure 28. Aerial photograph contrasting the appearance of soils formed in low-carbonate parent material (south and east of playa) with the lighter-colored high-carbonate parent material (north and northeast of playa).

Bulletin 798 • Page 51

Figure 29. Diagrams illustrating the effect of soil age and particle size on soil boundaries (vicinity II in Figure 1). A, illus-tration of the alluvial fans debouching from the Organ Mountains and the coalescent fan-piedmont that occurs downslope. Rectangle locates lower block diagram. B, illustration of buried soil and the progressive increase in coarse fragments up-slope (from Gile, 1975a, used with permission). The block at left illustrates soils and sediments of Holocene age that bury soils and sediments of Pleistocene age, used with permission.

Bulletin 798 • Page 52

argillic horizons and weak stage I carbonate horizons. The argillic horizons tend to be slightly more promi-nent here than at lower elevations. This prominence is thought to be largely due to slightly greater precipitation in this area. Also, some of these soils may have begun developing somewhat earlier in the Holocene, and thus had a longer time for clay illuviation.

The fine-loamy Calciargids of late Pleistocene age emerge along the downslope margins of the Holocene ridges. These Calciargids have stage III calcic horizons and reddish-brown and red argillic horizons with an average texture of sandy clay loam. The differences in morphology from the Holocene soils are attributed to substantially greater age development partly during a Pleistocene pluvial.

Difference in particle size and age; boundary between coarse-loamy Haplargids and loamy-skeletal Calciargids (Figure 29B) Soil patterns on the Pleistocene fans (Figure 29A) are complicated by wide variations in soil age and degree of landscape dissection. Major soils are loamy-skeletal, Ustic Calciargids with thick argillic and calcic horizons. Depending on soil age, the Argids on these fans have calcic or petrocalcic horizons, or only stage I or II car-bonate horizons.

The boundary between the coarse-loamy and loamy-skeletal Argids is apparent because of the distinct fan form and differences in slope. Location of the bound-ary between soils that are skeletal and soils that are not depends largely upon the character of the rock in the alluvium. Monzonite has the property of rapid commi-nution in transport and weathering. Hence, the skeletal soils are confined largely to the steeper slopes along the mountain fronts.

Boundaries caused by obliteration of the argillic horizonObliteration of the argillic horizon is usually the result of (1) landscape dissection, with its associated soil truncation and carbonate engulfment, and (2) faunal activity.

The Argids and the adjacent soils (in which the ar-gillic horizon has been obliterated) are mostly of Pleis-tocene age. The morphological change is usually from an argillic horizon that is readily identified (because of its distinct silicate clay maximum and reddish-brown and red colors) to a B horizon lacking these features. Carbonate accumulation is involved in the obliteration process since argillic horizons tend to become less red and lighter colored with increasing carbonate accumula-tion. Carbonate also affects the estimation of silicate

clay, which is needed to assess the requirement for clay increase. Some horizons feel fine-textured enough, but analyses show that the increase in fineness is due to fine-grained carbonate instead of silicate clay.

Acid, color, and macroscopic carbonate are useful for field distinction between Argids and their associ-ates with obliterated argillic horizons. If part of the horizon of clay accumulation is non-calcareous, then enough oriented clay generally remains for the horizon to qualify as an argillic horizon. If the argillic horizon is calcareous, it can still have enough oriented clay. The point at which the horizon in B position contains too little oriented clay is commonly marked by a shift in color. If part of the horizon of silicate clay accumulation is reddish-brown (approximately 5YR 5/4, dry) or red-der, then enough oriented clay usually remains for the horizon to qualify as an argillic horizon. As carbonate continues to accumulate and hues become yellower than about 6YR, in most soils there is so much carbonate that essentially all of the oriented clay has been obliterated, and this marks the shift to the Typic Calcids. At this point, macroscopic carbonate is visible as grain coatings. Also at this point, the carbonate content in these hori-zons is high enough that estimates of silicate clay are less reliable.

There are numerous degrees of obliteration of the reddish-brown and red argillic horizon material. If the horizon in B position has at least 10% by volume of argillic horizon material, the soil has been classified as an Argid. If the horizon has less than 10%, then the soil is classified as a Haplocalcid or a Typic Petrocalcid. The 10% figure has been used because, as seen in long profile exposures illustrating the boundary between soils with and without argillic horizons, much smaller per-centages of argillic material may occur at that transition zone. Such minor amounts might easily be missed with an auger or small pit.

Observations elsewhere in the Southwest and in des-ert regions of northern Mexico indicate that similar rela-tions for obliteration of the argillic horizon are extensive in other arid regions. The argillic horizon is generally less red in colder deserts, and the color changes caused by landscape dissection, soil truncation, and carbonate accumulation would probably differ accordingly.

Boundaries caused by landscape dissectionSoil change caused by landscape dissection (Figure 30) ranges from minor truncation with no change in classifi-cation to the truncation of all diagnostic horizons and a resultant change from Aridisols to Entisols.

Bulletin 798 • Page 53

Figure 30. Diagrams illustrating the effect of dissection on soil classification (vicinity III in Figure 1). A, with increased dissection soils change from Argids to Calcids to Orthents as the argillic and subsequently calcic horizons are truncated by erosion. Rectangle locates lower block diagram. B, in the early part of this dissectional sequence, erosion, carbonate engulfment, and/or biotic mixing obliterates the argillic horizon, which is thus transformed into a cambic horizon. As a result, soils change from Argic Petrocalcids to Typic Petrocalcids (modified from Gile, 1975b, used with permission)

Bulletin 798 • Page 54

Difference in dissection; boundary between Argids and Calcids without argillic horizonsIn Figure 30 (vicinity III in Figure 1), the transition between soils with and without argillic horizons occurs where the soil parent materials were derived largely or wholly from non-calcareous rocks, so that an argillic horizon had formed and is still preserved on the most stable sites. In these soils, which range in age from middle to late Pleistocene, the argillic horizon is usually underlain by a calcic or petrocalcic horizon. If a calcic horizon is present, loss of the Bt horizon causes a change from a Calciargid to a Haplocalcid. If a petrocalcic ho-rizon is present, the change is from an Argic Petrocalcid to a Typic Petrocalcid (Figure 30B).

The morphological change from Argids or Argic Petrocalcids to Typic Haplocalcids or Petrocalcids on a given ridge usually occurs over a distance of only a few meters. Downslope from the Argids of ridge crests that are level transversely, the argillic horizon first becomes calcareous, with little or no visible carbonate. With in-creasing distance downslope, macroscopic carbonate ap-pears and gradually rises in the soil as truncation of thin upper horizons brings partially carbonate-impregnated horizons closer to the surface. Truncation, by increasing slopes and runoff, also causes carbonate to accumulate at depths shallower than in adjacent areas that are level transversely. Finally, less than 10% of the reddish-brown, argillic horizon material remains, and the soils are classified as Haplocalcids or Typic Petrocalcids.

The transition between soils with an intact argillic horizon in terraced terrain differs in different parts of the terrain. Where the argillic horizon has been obliter-ated on the highest ridges, it may be well preserved on younger, less dissected terraces below. With increasing dissection, the argillic horizon is finally obliterated on these younger surfaces also.

The boundaries between Argids, Argic Petrocalcids, and Calcids without argillic horizons occur extensively in dissected terrains in many arid regions. Clearly, large areas of Calcids without argillic horizons owe their gen-esis to landscape dissection.

Difference in dissection; boundary between Calcids and Orthents (Figure 30A)The change from Calcids without argillic horizons to Orthents occurs in two general landscape positions, ridge sides and ridge crests, depending upon the charac-ter of the dissection. With increasing dissection, Orth-ents are first encountered on ridge sides, just below the point at which the calcic or petrocalcic horizons have been truncated, exposing the underlying parent mate-rials. The soils downslope from this point are usually

calcareous throughout and have stage I carbonate char-acteristic of Holocene soils. In more strongly dissected areas, there is gradation from Calcids to Orthents on ridge crests, with the Orthents predominant in drain-ageways that form saddles in the ridge crests, truncating the calcic or petrocalcic horizon. Psamments may be en-countered instead of Orthents if the materials are sandy and have little gravel (Soil Survey Staff, 1999). If buried soils have been exhumed by dissection, the classification is determined by the morphology of the exhumed soil.

Boundaries caused by differences in landscape dissection and soil moistureFigure 31 (vicinity IV in Figure 1) illustrates soil bound-aries caused by differences in dissection and past soil moisture. Large remnants of the relict basin floors occur along the margins of the valley. Soil parent materials are sandy, low-carbonate sediments deposited by the ances-tral Rio Grande. Coppice dunes occur in places (Figure 31B), illustrating boundaries caused by differences in age (Gile, 1966a).

Landscape dissection has caused the development of structural benches (Thornbury, 1969) bordering the scarp cut in the relict basin floor (Figure 31). Structural benches occur extensively along the river valley where surficial low-gravel materials have been eroded and un-derlying gravelly beds have been exposed. The gravelly beds are resistant to erosion and tend to form gravelly ridges. Slopes are much steeper on the west side of the valley than on the east side because the deeply en-trenched floodplain is close (Figure 1). For this reason, the soil boundary caused by dissection is from Argids to Entisols. The Calcids without argillic horizons, so com-mon on the gentler slopes on the east side of the valley, seldom occur on the west side.

Difference in dissection; boundary between Petroargids and Torriorthents Typic Petroargids, with thick argillic horizons and pet-rocalcic horizons, are dominant on the relict basin floor along the scarp shown in Figure 31B. Downslope from the scarp, Typic Torriorthents, sandy-skeletal, are domi-nant on the ridge crests (the structural bench). This illustrates the extreme effect of dissection. The strongly expressed diagnostic horizons have been truncated, and young soils are now forming in the underlying materi-als. This involves the soil-forming factor of age since the materials exposed by dissection have weak horizons typi-cal of Holocene soils in these materials. The young soils are Torriorthents with thin, brown B horizons and stage I carbonate horizons. The soil boundary is well marked by the scarp.

Bulletin 798 • Page 55

Difference in soil moisture; boundary between Haplar-gids or Calciargids and Argic Petrocalcids (Figure 15)In many soils, roughly funnel-shaped zones of reddish-brown, argillic horizon material (pipes) descend into carbonate horizons (Figure 15). These zones have been observed only in soils of Pleistocene age and must there-fore have formed primarily in the Pleistocene. Many pipes of mid-Pleistocene basin floors are particularly large. Pipes must have been deeply flushed with mois-ture in pluvials, moving carbonate to substantial depths. Adjacent petrocalcic horizons are plugged with carbon-ate and should funnel water into the pipes. Morphology indicates that the lower parts of the pipes are seldom if ever wetted at the present time (Gile et al., 1981). Typic Haplargids and/or Calciargids occur in the pipes, where the petrocalcic horizon is below 1.5-m depth or is absent. Argic Petrocalcids have a petrocalcic horizon within 1-m depth and are adjacent to the pipes. The boundary between the Haplargids or Calciargids and the Argic Petrocalcids is prominent in exposures, but cannot be seen at the land surface because the slope and landform are the same across both.

Pipes of JER and upper La Mesa tend to be complex, with thick laminar linings; calcic horizons have formed in the lower parts of some pipes. In contrast, pipes of the much younger soils of lower La Mesa have thin laminar linings, discontinuous laminar linings, or none at all.

Boundaries caused by differences in faunal activity (Figure 32)Figure 32 (vicinity V in Figure 1) illustrates soil bound-aries caused by faunal activity. The basin floor is level or nearly level, with occasional slight ridges. The soil parent materials are sandy, low-carbonate sediments deposited by the ancestral Rio Grande. The soils are of mid-Pleistocene age, and in these parent materials distinct argillic horizons would normally be expected.

Figure 32B shows part of a slight ridge in the basin floor. Typic Haplocalcids are dominant on the ridge; they have cambic horizons and thick calcic horizons. Textures of the cambic horizons are sandy loam and light sandy clay loam. Soils are calcareous throughout; sand grains and the few pebbles are thinly coated with carbonate. Very little oriented clay is visible because of the carbonate. The horizons are more yellow than 5YR (commonly they are 7.5YR), and color is quite uniform throughout. In places, however, remnants of the argillic horizon and the Calciargids are still preserved (Figure 32B).

Several factors indicate that argillic horizons were once present over the whole ridge, but now have been largely obliterated by soil fauna. (i) Argillic horizons are

still present in a few areas. (ii) In places, pipes of red-dish-brown Bt material are preserved in the thick calcic horizons, and descend below B horizons with numerous termite burrows. Soils with such pipes in other parts of the study area are commonly connected to argillic hori-zons. (iii) Analyses of particle size (carbonate-free basis) of A and B horizons show a silicate clay distribution unlike that of typical Calciargids in the area. In some instances there is more clay at the surface than in the B horizon (Gile et al., 1981). (iv) Termite burrows and mounds and tunnels constructed by rodents are present.

Although termite tunnels are less obvious than those constructed by rodents, they may have obliterated larger areas of argillic horizons. Lee and Wood (1971) state that “termites have developed the capacity to burrow and mould structures from soil and organic matter to a level unknown in any other group of soil animals.” They also note their importance in the physical disturbance and overturning of profiles, especially in movement of fine materials from deep horizons to the surface. Ter-mites in the study area are subterranean. Their workings include numerous tunnels in the cambic horizon, many of which are partly or completely filled with fine earth, and sheaths of fine earth on shrubs and twigs at the soil surface. Such movement of fine earth could account for the erratic distribution of silicate clay and would contribute carbonate to upper horizons. The boundary is not apparent at the land surface since the slope and landform continue smoothly across the change from Haplocalcids to Calciargids.

Faunal activity has obliterated the argillic horizon in some areas but not in others. The precise reasons are unknown, but there appear to be several contributing factors. (1) Textures of the B horizon are sandy loam and light sandy clay loam. Such textures are not as hard and are easier to burrow through than heavier textures. (2) A nonindurated, high-carbonate horizon is near the surface. This should favor faunal transport of carbon-ate from the top of this horizon to upper horizons. (3) The high-carbonate horizon may restrict the general zone of termite activity to shallower depths, resulting in maximum churning within the A and B horizons. The argillic horizon must have formed largely in Pleistocene pluvials, and the present Holocene conditions appar-ently do not favor its preservation.

RELATED RESEARCH AND RESEARCH SUBSEQUENT TO FORMAL CLOSING OF THE DESERT PROJECT IN 1972The Desert Project formally ended in 1972, and much of the subsequent research has been concerned with

Bulletin 798 • Page 56

Figure 31. Illustration of dissection along a prominent erosional scarp and recent deposition of eolian material as coppice dunes (Torripsamments) (vicinity IV in Figure 1). A, diagram of the geomorphic setting showing the modern Rio Grande floodplain and the relict basin floor. The rectangle to the right locates the block diagram below; the rectangle to the left locates Figure 15. B, truncation of the argillic and petrocalcic horizons results in the gradation from Typic Petroargids to Typic Torriorthents (modified from Gile, 1975b, used with permission).

Bulletin 798 • Page 57

Figure 32. Illustrations of the effect faunal activity has on soil morphology, classification, and boundaries (vicinity V in Figure 1). A, diagram of a slight ridge in an otherwise level or nearly level basin floor. Square locates block diagram below. B, illustration of the laterally discontinuous Bt horizon and its effect on classification. Destruction of the Bt is thought to be the result of faunal activity (modified from Gile, 1975b, used with permission).

Bulletin 798 • Page 58

(1) Plio-Pleistocene history of the ancestral Rio Grande, (2) hydrogeologic framework of basin and valley fills, (3) soil mineralogy, (4) carbon storage in arid soils, (5) aridland ecology related to water and root morphol-ogy, and (6) relating the Desert Project findings to nearby areas.

(1) Understanding of the arrival and entrenchment of the ancestral Rio Grande was advanced by dating Camp Rice sediments with magnetostratigraphy and Ar-Ar dating of interbedded pumice (Mack et al., 1993, 1996). These techniques, added to the K-Ar dates of basalt flows within and overlying the Camp Rice Formation (Seager et al., 1984), indicate that the ancestral Rio Grande arrived in the Desert Project region about 5 Ma and that river-valley incision was initiated very near or at the Matuyama-Brunhes boundary (0.78 Ma; Mack et al., 1998). In addition, the magneto-stratigraphy and dated pumice, together with more data on pedogenic carbonates, have greatly increased the precision of dating the ancient soils of lower, upper, and JER La Mesa.

(2) Hydrogeologic studies during the past two de-

cades (Peterson et al., 1984; Hawley & Lozinsky, 1992), as well as ongoing studies, done in cooper-ation with the New Mexico Water Resources Re-search Institute, the U.S. Geological Survey, and New Mexico Institute of Mining and Technology (New Mexico Tech), have emphasized develop-ment of detailed conceptual models that integrate available subsurface geological, geophysical, and geochemical data with surface information. These models, in turn, provide a much better “ground-truth” basis for numerical models of groundwater-flow systems. The architecture of mappable subdi-visions of basin and valley fills that can be defined in terms of aquifer and vadose-zone behavior is emphasized in the conceptual models. Fundamen-tal hydrogeologic units include basin-bounding and intrabasin structures and lithofacies assem-blages that are combined into hydrostratigraphic map units (Hawley & Haase, 1992; Hawley et al., 1995).

(3) Soil mineralogy research has focused on the mech-

anisms of carbonate and silicate clay formation. Pedogenic carbonate in the Desert Project, regard-less of the parent material in which it forms, is calcite (Kraimer et al., 2005) and, as revealed by laboratory experiments and electron microscopy,

can be precipitated by soil microorganisms in addition to abiotic precipitation (Monger et al., 1991a, 1991b). Silicate clay also has more than one origin. Much is derived from dust, which is dominated by kaolinite, illite, and smectite (Gile & Grossman, 1979). Still, some interstratified micas are weathered from sedimentary rocks, and some kaolinites and smectites are formed by in situ weathering, especially along the mountain fronts (Monger & Lynn, 1996). Palygorskite and sepiolite, common clay minerals associated with older petrocalcic horizons (Vanden Heu-vel, 1966), have also formed in situ (Monger & Daugherty, 1991a, 1991b).

(4) Large quantities of carbon are stored in arid soils as inorganic carbon (Schlesinger, 1982, 1985; Monger & Martinez-Rios, 2001; Serna-Perez et al., 2006; Liu et al., 2007). Analysis of Desert Project data, other than for soils developed in calcareous parent materials, indicates that soils contain eight times more inorganic carbon than organic carbon, with a range of 25 kg m-2 for in-organic carbon and 3.1 kg m-2 for organic carbon (Grossman et al., 1995).

(5) Aridland ecological patterns are very much related to soil-geomorphic patterns (Schlesinger et al., 1990; McAuliffe, 1994; Monger & Bestelmeyer, 2006; Duniway et al., 2007). Herbel et al. (1994) measured soil water suctions in the Desert Project region for two decades. Although the arid part of the Desert Project receives about 200 mm of annual precipitation, soils in topographically low areas that accumulate runoff receive more that 200 mm and support more and different vegetation. Moreover, root studies reveal that soil horizons affect water availability by (1) control-ling root morphology (Gile et al., 1995a) and (2) storing water in the microporosity of petrocalcic horizons (Hennessy et al., 1983). Additional stud-ies relating plant roots in arid lands to soil mor-phology, genesis, and classification were presented by Gile et al. (1997, 1998). Isotope studies reveal that shifts from C4 grasses to C3 shrubs occurred concomitantly with middle Holocene erosion (Monger et al., 1998; Buck & Monger, 1999) and late-19th century erosion (Connin et al., 1997a, 1997b).

(6) Research related to the Desert Project findings in nearby areas includes (i) analysis of stage V and

Bulletin 798 • Page 59

VI calcrete soils in the Rincon area of southern New Mexico (Gile et al., 1996), (ii) neotectonic analysis of Quaternary faults (Gile, 1986; Ma-chette, 1987; Gile, 1994b; Keaton et al., 1995), (iii) timing of volcanic activity (Gile, 1987a, 1987b, 1990), and (iv) geoarchaeological studies (Monger, 1995; Buck, 1996). The broad-scale context in which the Desert Project resides, both geomorphic and temporal, has been described by Hawley (2005).

CONCLUSIONS AND DISSEMINATION OF THE RESEARCH The Desert Project investigated pedologic, geomorphic, and geologic topics, varying from the regional scale con-cerned with the genesis of basins and Cenozoic stratigra-phy, to the microscopic scale concerned with the genesis of clay illuviation and carbonate formation. In terms of its original goal of providing a basic understanding of arid soil-geomorphic relationships, the Desert Project generated much knowledge about the nature of Basin and Range landforms, the relationship between soils and geomorphic surfaces, the genesis of horizons of car-bonate and clay accumulation, and the reasons for soil boundaries. It also provided a soil-geomorphic and geo-logic context for understanding rangeland ecology and vegetation change (Gibbens & Lenz, 2001; Bestelmeyer et al., 2006; Monger et al., 2009).

In addition to this basic research, the Desert Project generated voluminous soil characterization data, tax-onomy concepts, and some practical applications for soil-geomorphic studies. A central practical application is the prediction of landscape stability using geomorphic maps. A geomorphic map delineates the active areas of erosion and sedimentation, such as arroyo systems, and the areas that have been stable for many thousands of years, such as uneroded fan-terrace surfaces of Pleisto-cene age. Therefore, rather than indiscriminately laying out housing subdivisions across unstable areas in typical rectangular patterns, vast areas of the American South-west would benefit from planning in accordance with geomorphic surfaces (Hawley, 1972).

Results of the research have been presented in jour-nal articles, reports, meetings, slide sets, videos, and in numerous tours of the study area. The most recent tours were held in May 2000 (Appendix B) and May 2007, the 50th Anniversary Desert Project Tour (Gile et al., 2007). Guidebook to the Desert Project (Gile et al., 1981), which in 1983 received the Kirk Bryan Award from the Geological Society of America for a distinguished con-tribution to geomorphology, has been found to be very

useful in both study tours of the project area and in the classroom. Quoting from Gile et al., 1981, p. vii:

The Desert Project has been a good study and train-ing ground for a wide variety of workers. The project area is similar to many arid and semiarid regions in terms of terrain, parent materials for soils, range in age of soils, and general climatic history. Thus princi-ples of soil and landscape evolution worked out in the Desert Project also apply to many areas other than the southwest United States. A number of formal field-study tours were held during progress of the research. Participants included agronomists, anthropologists, archaeologists, biologists, foresters, geomorphologists, geologists, range scientists, and soil scientists. We have received many requests for copies of previous field guides, which were printed only in limited num-bers. Supplies of these earlier guides have long been exhausted, and we hope that this volume will be a permanent guide to many of the detailed study sites of the Desert Project.

REFERENCES Agricultural Experiment Stations-Soil Conservation

Service. (1964). Soils of the western United States (exclusive of Hawaii and Alaska). Pullman, WA: Washington State University.

Antevs, E. (1955). Geologic-climatic dating in the West. American Antiquity, 20, 317–335.

Bachman, G. O., & Machette, M. N. (1977). Calcic soils and calcretes in the southwestern United States [Open-File Report 77-794]. Reston, VA: U.S. Geological Survey.

Bestelmeyer, B. T., Ward, J. P., & Havstad, K. M. (2006). Soil-geomorphic heterogeneity governs patchy vegetation dynamics at an arid ecotone. Ecology, 87, 963–973.

Betancourt, J. L., Van Devender, T. R., & Martin, P. S. (Eds.). (1990). Packrat middens: The last 40,000 years of biotic change. Tucson: The University of Arizona Press.

Birkeland, P. W. (1999). Soils and geomorphology. New York: Oxford University Press, Inc.

Birkeland, P. W., Machette, M. N., & Haller, K. M. (1991). Soils as a tool for applied Quaternary geology [Utah Geological and Mineral Survey Misc. Pub. 91-3]. Salt Lake City: Utah Department of Natural Resources.

Bulletin 798 • Page 60

Bretz, J. H., & Horberg, L. H. (1949). Caliche in southeastern New Mexico. Journal of Geology, 57, 491–511.

Brown, C. N. (1956). The origin of caliche on the northeastern Llano Estacado, Texas: Journal of Geology, 64, 1–15.

Brown, I. C., & Drosdoff, M. (1940). Chemical and physical properties of soils and their colloids devel-oped from granitic materials in the Mojave Desert. Journal of Agricultural Research, 61, 338–344.

Bryan, K. (1925). Date of channel trenching (arroyo cutting) in the arid Southwest. Science, 62, 338–344.

Bryan, K. (1938). Geology and ground-water condi-tions of the Rio Grande depression in Colorado and New Mexico. In Regional planning: Part VI—The Rio Grande joint investigations in the Upper Rio Grande Basin in Colorado, New Mexico, and Texas, 1936-1937, vol. I (pp.196–225). Washington, D.C.: Gov-ernment Printing Office.

Buck, B. J. (1996). Late Quaternary landscape evolu-tion, paleoclimate, and geoarchaeology, southern New Mexico and west Texas. Unpublished Ph.D. the-sis, New Mexico State University, Las Cruces.

Buck, B. J., & Monger, H. C. (1999). Stable isotopes and soil-geomorphology as indicators of Holocene climate change, northern Chihuahuan Desert. Journal of Arid Environments, 43, 357–373.

Buffington, L. C., & Herbel, C. H. (1965). Vegetation changes on a semidesert grassland range from 1858 to 1963. Ecological Monographs, 35, 139–164.

Bullock, P., Federoff, N., Jongerius, A., Stoops, G., & Tursina, T. (1985). Handbook for soil thin section description. Wolverhampton, UK: Waine Research Publications.

Buol, S. W., & Yesilsoy, M. S. (1964). A genesis study of a Mojave sandy-loam profile. Soil Science Society of America Proceedings, 28, 254–256.

Cole, D. R., & Monger, H. C. (1994). Influence of atmospheric CO

2 on the decline of C4 plants during

the last deglaciation. Nature, 368, 533–536.Connin, S. L., Virginia, R. A., & Chamberlain, C. P.

(1997a). Carbon isotopes reveal soil organic matter dynamics following arid land shrub expansion. Oeco-logia, 110, 374–386.

Connin, S. L., Virginia, R. A., & Chamberlain, C. P. (1997b). Isotopic study of environmental change from disseminated carbonate in polygenetic soils. Soil Science Society of America Journal, 61, 1710–1722.

Dickson, J. A. D. (1965). A modified staining technique for carbonates in thin section. Nature, 205, 587.

Dunham, K. C. (1935). The geology of the Organ Mountains [Bulletin 11]. Socorro, NM: New Mexico Bureau of Mines and Mineral Resources.

Duniway, M. C., Herrick, J. E., & Monger, H. C. (2007). The high water-holding capacity of petrocal-cic horizons. Soil Science Society of America Journal, 71, 812–819.

Freeman, C. E. (1972). Pollen study of some Holocene alluvial deposits in Dona Ana County, southern New Mexico. Texas Journal of Science, 24(2), 203–220.

Frye, J. C., Glass, H. D., Leonard, A. B., & Coleman, D. D. (1974). Caliche and clay mineral zonation of Ogallala Formation, central-eastern New Mexico [Circular 144]. Socorro, NM: New Mexico Bureau of Mines and Mineral Resources.

Frye, J. C., & Willman, H. B. (1962). Morphostrati-graphic units in Pleistocene stratigraphy. American Association of Petroleum Geologists Bulletin, 46, 112–113.

Gardner, L. R. (1972). Origin of the Mormon Mesa caliche, Clark County, Nevada. Geological Society of America Bulletin, 83, 143–156.

Gibbens, R. P., & Lenz, J. M. (2001). Root systems of Chihuahuan Desert plants. Journal of Arid Environ-ments, 49, 221–263.

Gile, L. H. (1961). A classification of ca horizons in the soils of a desert region, Dona Ana County, New Mexico. Soil Science Society of America Proceedings, 25, 52–61.

Gile, L. H. (1966a). Coppice dunes and the Rotura soil. Soil Science Society of America Proceedings, 30, 657–660.

Gile, L. H. (1966b). Cambic and certain noncambic horizons in desert soils of southern New Mexico. Soil Science Society of America Proceedings, 30, 773–781.

Gile, L. H. (1970). Soils of the Rio Grande valley bor-der in southern New Mexico. Soil Science Society of America Proceedings, 34, 465–472.

Gile, L. H. (1975a). Causes of soil boundaries in an arid region: I. Age and parent materials. Soil Science Soci-ety of America Proceedings, 39, 316–323.

Gile, L. H. (1975b). Causes of soil boundaries in an arid region: II. Dissection, moisture, and faunal ac-tivity. Soil Science Society of America Proceedings, 39, 324–330.

Gile, L. H. (1975c). Holocene soils and soil-geomorphic relations in an arid region of southern New Mexico. Quaternary Research, 5, 321–360.

Gile, L. H. (1977). Holocene soils and soil-geomorphic relations in a semi-arid region of southern New Mex-ico. Quaternary Research, 7, 112–132.

Gile, L. H. (1986). Late Holocene displacement along the Organ Mountains fault in southern New Mexi-co—A summary. New Mexico Geology, 8, 1–4.

Bulletin 798 • Page 61

Gile, L. H. (1987a). A pedogenic chronology for Kil-bourne Hole, southern New Mexico: I. Soils in the tuff. Soil Science Society of America Journal, 51, 746–751.

Gile, L. H. (1987b). A pedogenic chronology for Kil-bourne Hole, southern New Mexico: II. Time of the explosions and soil events before the explosions. Soil Science Society of America Journal, 51, 752–760.

Gile, L. H. (1990). Chronology of lava and associated soils near San Miguel, New Mexico. Quaternary Re-search, 33, 37–50.

Gile, L. H. (1993). Carbonate stages in sandy soils of the Leasburg surface, southern New Mexico. Soil Sci-ence, 156, 101–110.

Gile, L. H. (1994a). Soils of an eolian analog of the Leasburg surface, southern New Mexico. Quaternary Research, 41, 191–199.

Gile, L. H. (1994b). Soils, geomorphology, and multiple displacements along the Organ Mountains fault in southern New Mexico [Bulletin 133]. Socorro, NM: New Mexico Bureau of Mines and Mineral Resources.

Gile, L. H. (1995). Pedogenic carbonate in soils of the Isaacks’ Ranch surface, southern New Mexico. Soil Science Society of America Journal, 59, 501–508.

Gile, L. H. (1999). Eolian and associated pedogenic features of the Jornada Basin floor, southern New Mexico. Soil Science Society of America Journal, 63, 151–163.

Gile, L. H. (2002). Lake Jornada, an early-middle Pleis-tocene lake in the Jornada del Muerto Basin, south-ern New Mexico. New Mexico Geology, 24, 3–14.

Gile, L. H., & Ahrens, R. J. (Eds.). (1994). Supplement to the Desert Project soil monograph, volume I: Soils and landscapes of a desert region astride the Rio Grande Valley near Las Cruces, New Mexico [Soil Survey Investigations Report No. 44]. Lincoln, NE: U.S. Department of Agriculture, Soil Conservation Service.

Gile, L. H., & Ahrens, R. J. (Eds.). (1996). Supplement to the Desert Project soil monograph, volume II: Soils and landscapes of a desert region astride the Rio Grande Valley near Las Cruces, New Mexico [Soil Survey In-vestigations Report No. 44]. Lincoln, NE: U.S. De-partment of Agriculture, Soil Conservation Service.

Gile, L. H., Ahrens, R. J., & Anderson, S. P. (Eds.). (2003). Supplement to the Desert Project soil mono-graph, volume III: Soils and landscapes of a desert region astride the Rio Grande Valley near Las Cruces, New Mexico [Soil Survey Investigations Report No. 44]. Lincoln, NE: U.S. Department of Agriculture, Soil Conservation Service.

Gile, L. H., Gibbens, R. P., & Lenz, J. M. (1995a). Soils and sediments associated with remarkable, deeply-penetrating roots of crucifixion thorn (Koeberlinia spinosa Zucc.). Journal of Arid Environments, 31, 137–151.

Gile, L. H., Gibbens, R. P., & Lenz, J. M. (1997). The near-ubiquitous pedogenic world of mesquite roots in an arid basin floor. Journal of Arid Environments, 35, 39–58.

Gile, L. H., Gibbens, R. P., & Lenz, J. M. (1998). Soil-induced variability in root systems of creosotebush (Larrea tridentata) and tarbush (Flourensia cernua). Journal of Arid Environments, 39, 57–78.

Gile, L. H., & Grossman, R. B. (1968). Morphology of the argillic horizon in desert soils of southern New Mexico. Soil Science, 106, 6–15.

Gile, L. H., & Grossman, R. B. (1979). The Desert Project soil monograph: Soils and landscapes of a desert region astride the Rio Grande Valley near Las Cruces, New Mexico. Lincoln, NE: U.S. Department of Agri-culture, Soil Conservation Service.

Gile, L. H., Grossman, R. B., & Hawley, J. W. (1969). Effects of landscape dissection on soils near Univer-sity Park, New Mexico. Soil Science, 108, 273–282.

Gile, L. H., Grossman, R. B., Hawley, J. W., & Monger, H. C. (1996). Ancient soils of the Rincon Surface, Northern Dona Ana County. In L. H. Gile & R. J. Ahrens (Eds.), Supplement to the Desert Project soil monograph, volume II [Soil Survey Investigations Re-port No. 44] (pp. 1–110). Lincoln, NE: U.S. Depart-ment of Agriculture, Soil Conservation Service.

Gile, L. H., & Hawley, J. W. (1966). Periodic sedimen-tation and soil formation on an alluvial-piedmont in southern New Mexico. Soil Science Society of America Proceedings, 30, 261–268.

Gile, L. H., & Hawley, J. W. (1968). Age and compara-tive development of desert soils at the Gardner Spring radiocarbon site, New Mexico. Soil Science Society of America Proceedings, 32, 709–719.

Gile, L. H., & Hawley, J. W. (1972). The prediction of soil occurrence in certain desert regions of the south-western United States. Soil Science Society of America Proceedings, 36, 119–124.

Gile, L. H., Hawley, J. W., & Grossman, R. B. (1971). The identification, occurrence and genesis of soils in an arid region of southern New Mexico [Training bulletin]. Washington, D.C.: U.S. Department of Agriculture, Soil Conservation Service.

Gile, L. H., Hawley, J. W., & Grossman, R. B. (1981). Soils and geomorphology in the Basin and Range area of Southern New Mexico: Guidebook to the Desert Project [Memoir 39]. Socorro, NM: New Mexico Bureau of Mines and Mineral Resources.

Bulletin 798 • Page 62

Gile, L. H., Hawley, J. W., Grossman, R. B., Monger, H. C., Montoya, C. E., & Mack, G. H. (1995b). Supplement to the Desert Project Guidebook, with emphasis on soil micromorphology [Bulletin 142]. Socorro, NM: New Mexico Bureau of Mines and Mineral Resources.

Gile, L. H., Monger, H. C., Grossman, R. B., Ahrens, R. J., Hawley, J. W., Peterson, F. F., Gibbens, R. P., Lenz, J. M., Bestelmeyer, B. T., & Nolen, B. A. (2007). A 50th anniversary guidebook for the Desert Project. Lincoln, NE: U.S. Department of Agricul-ture, Natural Resources Conservation Service, Na-tional Soil Survey Center.

Gile, L. H., Peterson, F. F., & Grossman, R. B. (1965). The K horizon: A master soil horizon of carbonate accumulation. Soil Science, 99, 74–82.

Gile, L. H., Peterson, F. F., & Grossman, R. B. (1966). Morphological and genetic sequences of carbon-ate accumulation in desert soils. Soil Science, 101, 347–360.

Goss, D. W., Smith, S. J., & Stewart, B. A. (1973). Movement of added clay through calcareous materi-als. Geoderma, 9, 97–103.

Grossman, R. B., Ahrens, R. J., Gile, L. H., Montoya, C. E., & Chadwick, O. A. (1995). Areal evalua-tion of organic and carbonate carbon in a desert area of southern New Mexico. In R. Lal, J. Kimble, E. Levine, & B. A. Steward (Eds.), Soils and global change (pp. 81–92). Boca Raton, FL: CRC Press, Inc.

Hawley, J. W. (1965). Geomorphic surfaces along the Rio Grande valley from El Paso, Texas, to Caballo Reservoir, New Mexico. In J. P. Fitzsimmons & C. L. Balk (Eds.), Guidebook: Sixteenth field conference, southwestern New Mexico II (pp. 188–198). Socorro, NM: New Mexico Geological Society.

Hawley, J. W. (1969). Notes on the geomorphology and late Cenozoic geology of northwestern Chihuahua. In D. A. Cordoba, S. A. Wengerd, & J. Shomaker (Eds.), Guidebook: Twentieth field conference, the bor-der region (pp. 131–142). Socorro, NM: New Mexico Geological Society.

Hawley, J. W. (1972). Geologic-geomorphic mapping to serve soil resource development. In Proceedings of the 27th Annual Meeting, Soil Conservation Society of America, 24–30.

Hawley, J. W. (1975a). Quaternary history of Dona Ana County region, south-central New Mexico. In W. R. Seager, R. E. Clemons, & J. F. Callender (Eds.), Guidebook: Twenty-sixth field conference, Las Cruces country (pp. 139–150). Socorro, NM: New Mexico Geological Society.

Hawley, J. W. (1975b). The Desert Soil-Geomorphol-ogy Project. In W. R. Seager, R. E. Clemons, & J. F. Callender (Eds.), Guidebook: Twenty-sixth field con-ference, Las Cruces country (pp. 183–185). Socorro, NM: New Mexico Geological Society.

Hawley, J. W. (Compiler). (1978). Guidebook to the Rio Grande rift in New Mexico and Colorado [Cir-cular 163]. Socorro, NM: New Mexico Bureau of Mines and Mineral Resources.

Hawley, J. W. (1986). Physiographic provinces and land forms of New Mexico. In J. Williams (Ed.), New Mexico in maps (pp. 28–31). Albuquerque: The Uni-versity of New Mexico Press.

Hawley, J. W. (1993). Geomorphic setting and late Quaternary history of pluvial-lake basins in the southern New Mexico region [Open-File Report 391]. Socorro, NM: New Mexico Bureau of Mines and Mineral Resources.

Hawley, J. W. (2005). Five million years of landscape evolution in New Mexico: An overview based on two centuries of geomorphic conceptual-model develop-ment. In S. G. Lucas, G. S. Morgan, & K. E. Zeigler (Eds.), New Mexico’s ice ages [Bulletin No. 28] (pp. 9–94). Albuquerque: New Mexico Museum of Natu-ral History and Science.

Hawley, J. W., Bachman, G. O., & Manley, K. (1976). Quaternary stratigraphy in the Basin and Range and Great Plains provinces, New Mexico and western Texas. In W. C. Mahaney (Ed.), Quaternary stratig-raphy of North America (pp. 235–274). Stroudsburg, PA: Dowden, Hutchinson & Ross, Inc.

Hawley, J. W., & Haase, C. S. (Compilers). (1992). Hy-drogeologic framework of the northern Albuquerque Basin [Open-File Report 387]. Socorro, NM: New Mexico Bureau of Mines and Mineral Resources.

Hawley, J. W., Haase, C. S., & Lozinsky, R. P. (1995). An underground view of the Albuquerque Basin. In C. T. Ortega-Klett (Ed.), The water future of Albu-querque and middle Rio Grande Basin [WRRI Report No. 290] (pp. 37–55). Las Cruces, NM: New Mexico Water Resources Research Institute.

Hawley, J. W., & Kottlowski, F. E. (1969). Quaternary geology of the south-central New Mexico border region. In F. E. Kottlowski & D. V. LeMone (Eds.), Border stratigraphy symposium [Circular 104] (pp. 89–115). Socorro, NM: New Mexico Bureau of Mines and Mineral Resources.

Bulletin 798 • Page 63

Hawley, J. W., Kottlowski, F. E., Strain, W. S., Seager, W. R., King, W. E., & LeMone, D. V. (1969). The Santa Fe Group in the south-central New Mexico border region. In F. E. Kottlowski & D. V. LeMone (Eds.), Border stratigraphy symposium [Circular 104] (pp. 52–76). Socorro, NM: New Mexico Bureau of Mines and Mineral Resources.

Hawley, J. W., & Lozinsky, R. P. (1992). Hydrogeologic framework of the Mesilla Basin in New Mexico and western Texas [Open-File Report 323]. Socorro, NM: New Mexico Bureau of Mines and Mineral Resources.

Haynes, Jr., C. V. (1968). Geochronology of late Qua-ternary alluvium. In R. B. Morrison & H. E. Wright, Jr. (Eds.), Means of correlation of Quaternary succes-sions (pp. 591–631). Salt Lake City: University of Utah Press.

Hennessy J. T., Gibbens, R. P., Tromble, J. M., & Cardenas, M. (1983). Water properties of caliche. Journal of Range Management, 36, 723–726.

Herbel, C. H., & Gile, L. H. (1973). Field moisture regimes and morphology of some arid-land soils in New Mexico. In R. R. Bruce & M. Stelly (Eds.), Field-soil water regime [Special Publication No. 5] (pp. 119–152). Madison, WI: Soil Science Society of America.

Herbel, C. H., Gile, L. H., Fredrickson, E. L., & Gib-bens, R. P. (1994). Soil water and soils at soil water sites, Jornada Experimental Range. In L. H. Gile & R. J. Ahrens (Eds.), Supplement to the Desert Project soil monograph, volume I [Soil Survey Investigations Report No. 44] (pp. iii–592). Lincoln, NE: U.S. De-partment of Agriculture, Soil Conservation Service.

Herrick, C. L. (1904). Lake Otero, an ancient salt lake in southeastern New Mexico. The American Geologist, 34, 174–189.

Irwin-Williams, C., & Haynes, C. V. (1970). Climatic change and early population dynamics in the south-western United States. Quaternary Research, 1, 59–71.

Izett, G. A., & Wilcox, R. E. (1982). Map showing lo-calities and inferred distributions of the Huckleberry Ridge, Meas Falls, and Lava Creek Ash Beds (Pearl-ette Family Ash Beds) of Pliocene and Pleistocene age in the western United States and southern Canada [U.S. Geological Survey Miscellaneous Investigations Series Map I-1325]. Reston, VA: U.S. Geological Survey.

Izett, G. A., Pierce, K. L., Naeser, N. D., & Jaworowski, C. (1992). Isotopic dating of Lava Creek B Tephra in terrace deposits along the Wind River, Wyoming: Im-plications for post 0.6 MA uplift of the Yellowstone Hotspot. Geological Society of America Abstracts with Programs, 24(7), A102.

Junge, C. E., & Werby, R. T. (1958). The concentration of chloride, sodium, potassium, calcium, and sulfate in rain water over the United States. Journal of Meteo-rology, 15, 417–425.

Keaton, J. R., Barnes, J. R., Scherschel, C. A., & Mon-ger, H. C. (1995). Evidence for episodic earthquake activity along the East Franklin Mountains fault, El Paso, Texas. Geological Society of America Abstracts with Programs, 27(4), 17.

Kim, S. M., Ruch, R. R., & Kempton, J. P. (1969). Radiocarbon dating at the Illinois State Geologi-cal Survey [Environmental Geology Notes No. 28]. Champaign: Illinois State Geological Survey.

King, W. E., & Hawley, J. W. (1975). Geology and ground-water resources of the Las Cruces area. In W. R. Seager, R. E. Clemons, & J. F. Callender (Eds.), Guidebook: Twenty-sixth field conference, Las Cruces country (pp. 195–204). Socorro, NM: New Mexico Geological Society.

Kottlowski, F. E. (1953). Tertiary-Quaternary sediments of the Rio Grande valley in southern New Mexico. In F. E. Kottlowski (Ed.), Guidebook: Twenty-fourth field conference, southwestern New Mexico (pp. 144–148). Socorro, NM: New Mexico Geological Society.

Kottlowski, F. E. (1958). Geologic history of the Rio Grande near El Paso. In Guidebook, 1958 field trip, Franklin and Hueco Mountains (pp. 46-54). Midland: West Texas Geological Society.

Kraimer, R. A., Monger, H. C., & Steiner, R. (2005). Mineralogical distinctions of carbonates in desert soils. Soil Science Society of America Journal, 69, 1773–1781.

Lee, K. E., & Wood, T. G. (1971). Termites and soils. New York: Academic Press.

Lee, W. T. (1907). Water resources of the Rio Grande valley in New Mexico [Water-Supply Paper 188]. Reston, VA: U.S. Geological Survey.

Liu, X., Monger, H. C., & Whitford, W. G. (2007). Calcium carbonate in termite galleries—Biominer-alization or upward transport? Biogeochemistry, 82, 241–250.

Lodge, Jr., J. P., Pate, J. B., Basbergill, W., Swanson, G. S., Hill, K. C., Lorange, E., & Lazrus, A. L. (1968). Chemistry of United States precipitation—Final report on the National Precipitation Sampling Net-work. Boulder, CO: National Center for Atmospher-ic Research.

Machette, M. N. (1978). Dating Quaternary faults in the southwestern United States by using buried calcic paleosols. Journal of Research of the U.S. Geological Survey, 6, 396–381.

Bulletin 798 • Page 64

Machette, M. N. (1985). Calcic soils of the southwest-ern United States. In D. L. Weide (Ed.), Soils and Quaternary geology of the southwestern United States [Special Paper 203] (pp. 1–21). Boulder, CO: Geological Society of America.

Machette, M. N. (1987). Preliminary assessment of pa-leoseismicity at White Sands Missile Range, southern New Mexico: Evidence for recency of faulting, fault segmentation, and repeat intervals for major earth-quakes in the region [Open-File Report 87-444]. Reston, VA: U.S. Geological Survey.

Machette, M. N., Long, T., Bachman, G. O., & Timbel, N. R. (1997). Laboratory data for calcic soils in cen-tral New Mexico—Background information for map-ping Quaternary deposits in the Albuquerque Basin [Circular 205]. Socorro, NM: New Mexico Bureau of Mines and Mineral Resources.

Mack, G. H., McIntosh, W. C., Leeder, M. R., & Mon-ger, H. C. (1996). Plio-Pleistocene pumice floods in the ancestral Rio Grande, southern Rio Grande rift, USA. Sedimentary Geology, 103, 1–8.

Mack, G. H., Salyards, S. L., & James, W. C. (1993). Magnetostratigraphy of the Plio-Pleistocene Camp Rice and Palomas Formations in the Rio Grande rift of southern New Mexico. American Journal of Science, 293, 49–77.

Mack, G. H., Salyards, S. L., McIntosh, W. C., & Leed-er, M. R. (1998). Reversal magnetostratigraphy and radioisotopic geochronology of the Plio-Pleistocene Camp Rice and Palomas Formation, southern Rio Grande Rift. In G. H. Mack, G. S. Austin, & J. M. Barker (Eds.), Guidebook: Forty-ninth field conference, Las Cruces country II (pp. 229–236). Socorro, NM: New Mexico Geological Society.

Marion, G. M. (1989). Correlation between long-term pedogenic CaCO

3 formation rate and modern

precipitation in deserts of the American Southwest. Quaternary Research, 32, 291–295.

Martin, P. S., & Mehringer, Jr., P. J. (1965). Pleistocene pollen analysis and biogeography of the Southwest. In H. E. Wright & D. G. Frey (Eds.), The Quater-nary of the United States (pp. 433–451). Princeton: Princeton University Press.

Mayer, L., McFadden, L. D., & Harden, J. W. (1988). Distribution of calcium carbonate in desert soils: A model. Geology, 16, 303–306.

McAuliffe, J. R. (1994). Landscape evolution, soil formation, and ecological patterns and processes in Sonoran Desert Bajadas. Ecological Monographs, 64, 111–148.

Mehringer, Jr., P. J. (1967). Pollen analysis and the allu-vial chronology. Kiva, 32, 96–101.

Metcalf, A. L. (1967). Late Quaternary mollusks of the Rio Grande Valley, Caballo Dam, New Mexico, to El Paso, Texas [Science Series No. 1]. El Paso: University of Texas at El Paso.

Metcalf, A. L. (1969). Quaternary surfaces, sediments, and mollusks: Southern Mesilla Valley, New Mexico and Texas. In D. A. Cordoba, S. A. Wengerd, & J. Shomaker (Eds.), Guidebook: Twentieth field confer-ence, the border region (pp. 158–164). Socorro, NM: New Mexico Geological Society.

Monger, H. C. (1993). Soil-geomorphic and paleocli-matic characteristics of the Fort Bliss maneuver areas, southern New Mexico and western Texas [Historic and Natural Resources Report No. 10]. El Paso, TX: Cultural Resources Management Program, Director-ate of the Environment, U.S. Army, Fort Bliss.

Monger, H. C. (1995). Pedology in arid lands archaeo-logical research: An example from southern New Mexico-western Texas. In M. E. Collins, B. J. Carter, B. G. Gladfelter, & R. J. Southard (Eds.), Pedological perspectives in archaeological research [Special Pub-lication 44] (pp. 35–50). Madison, WI: Soil Science Society of America.

Monger, H. C., & Bestelmeyer, B. T. (2006). The soil-geomorphic template and biotic change in arid and semiarid ecosystems. Journal of Arid Environments, 65, 207–218.

Monger, H. C., Cole, D. R., Buck, B. J., & Gallegos, R. A. (2009). Scale and the isotopic record of C4 plants in pedogenic carbonate—From the biome to the rhizosphere. Ecology, 90, 1498–1511.

Monger, H. C., Cole, D. R., Gish, J. W., & Giordano, T. H. (1998). Stable carbon and oxygen isotopes in Quaternary soil carbonates as indicators of ecogeo-morphic changes in the northern Chihuahuan Des-ert, USA. Geoderma, 82, 137–172.

Monger, H. C., & Daugherty, L. A. (1991a). Neofor-mation of palygorskite in a southern New Mexico Aridisol. Soil Society of America Journal, 55, 1646–1650.

Monger, H. C., & Daugherty, L. A. (1991b). Pressure solution: Possible mechanism for silicate grain disso-lution in a petrocalcic horizon. Soil Science Society of America Journal, 55, 1625–1629.

Monger, H. C., Daugherty, L. A., Lindemann, W. C., & Liddell, C. M. (1991b). Microbial precipitation of pe-dogenic calcite. Geology, 19, 997–1000.

Monger, H. C., Daugherty, L. A., & Gile, L. H. (1991a). A microscopic examination of pedogenic calcite in an Aridisol of southern New Mexico. In W. D. Nettleton (Ed.), Occurrence, characteristics, and genesis of carbonate, gypsum, and silica accumula-tions in soils [Special Publication 26] (pp. 37–60). Madison, WI: Soil Science Society of America.

Bulletin 798 • Page 65

Monger, H. C., & Lynn, W. C. (1996). Clay mineralo-gy at the Desert Project and the Rincon Surface study area. In L. H. Gile & R. J. Ahrens (Eds.), Supplement to the Desert Project soil monograph, volume II [Soil Survey Investigations Report No. 44] (pp. 111–155). Lincoln, NE: U.S. Department of Agriculture, Soil Conservation Service.

Monger, H. C., & Martinez-Rios, J. J. (2001). Inor-ganic carbon sequestration in grazing lands. In R. F. Follett, J. M. Kimble, & R. Lal (Eds.), The potential of U.S. grazing lands to sequester carbon and mitigate the greenhouse effect (pp. 87–118). Boca Raton, FL: Lewis Publishers.

Nettleton, W. D., Flach, K. W., & Brasher, B. R. (1969). Argillic horizons without clay skins. Soil Sci-ence Society of America Proceedings, 33, 121–125.

Nettleton, W. D., Witty, J. E., Nelson, R. E., & Hawley, J. W. (1975). Genesis of argillic horizons in soils of desert areas of the southwestern United States. Soil Science Society of America Proceedings, 39, 919–926.

Nikiforoff, C. C. (1937). General trends of the desert type of soil formation. Soil Science, 43, 105–131.

North American Commission on Stratigraphic Nomen-clature. (1983). North American stratigraphic code. American Association of Petroleum Geologists Bulletin, 67, 841–875.

Peterson, F. F. (1981). Landforms of the Basin & Range Province defined for soil survey [Technical Bulletin 28]. Reno: Nevada Agricultural Experiment Station, University of Nevada, Reno.

Peterson, D. M., Khaleel, R., & Hawley, J. W. (1984). Quasi three-dimensional modeling of groundwater flow in the Mesilla Bolson, New Mexico [WRRI Re-port No. 178]. Las Cruces, NM: New Mexico Water Resources Research Institute.

Reheis, M. C., Goodmacher, J. C., Harden, J. W., McFadden, L. D., Rockwell, T. K., Shroba, R. R., Sowers, J. M., & Taylor, E. M. (1995). Quaternary soils and dust deposition in southern Nevada and California. Geological Society of America Bulletin, 107, 1003–1022.

Richmond, G. M. (1963). Correlation of some glacial deposits in New Mexico. In Geological Survey research 1962; Short papers in geology, hydrology, and topogra-phy; Articles 180-239 [Professional Paper 450-E] (pp. E121–E125). Reston, VA: U.S. Geological Survey.

Ruhe, R. V. (1962). Age of the Rio Grande valley in southern New Mexico. Journal of Geology, 70, 151–167.

Ruhe, R. V. (1964). Landscape morphology and alluvial deposits in southern New Mexico. Annals of the Asso-ciation of American Geographers, 54, 147–159.

Ruhe, R. V. (1967). Geomorphic surfaces and surficial deposits in southern New Mexico [Memoir 18]. So-corro, NM: New Mexico Bureau of Mines and Min-eral Resources.

Ruhe, R. V. (1975). Geomorphology: Geomorphic processes and surficial geology. Boston: Houghton Mifflin Com-pany.

Sarna-Wojcicki, A. M., & Pringle, Jr., M. S. (1992). Laser-fusion 40Ar/39Ar ages of the tuff of Taylor Canyon and Bishops Tuff, E. California-W. Nevada. Eos, 73, 633.

Schlesinger, W. H. (1982). Carbon storage in the caliche of arid soils: A case study from Arizona. Soil Science, 133, 247–255.

Schlesinger, W. H. (1985). The formation of caliche in soils of the Mojave Desert, California. Geochimica et Cosmochimica Acta, 49, 57–66.

Schlesinger, W. H., Reynolds, J. F., Cunningham, G. L., Huenneke, L. F., Jarrell, W. M., Virginia, R. A., & Whitford, W. G. (1990). Biological feedbacks in global desertification. Science, 247, 1043–1048.

Schumm, S. A. (1965). Quaternary paleohydrology. In H. E. Wright & D. G. Frey (Eds.), The Quaternary of the United States (pp. 783–794). Princeton: Princeton University Press

Seager, W. R., Hawley, J. W., Kottlowski, F. E., & Kel-ley, S. A. (1987). Geology of east half of Las Cruces and northeast El Paso 1° x 2° sheets [Geologic Map 57]. Socorro, NM: New Mexico Bureau of Mines and Mineral Resources.

Seager, W. R., & Morgan, P. (1979). Rio Grande rift in southern New Mexico, West Texas, and northern Chihuahua. In R. E. Riecker (Ed.), Rio Grande Rift: Tectonics and magmatism (pp. 87–106). Washington, D.C.: American Geophysical Union.

Seager, W. R., Shafiqullah, M., Hawley, J. W., & Mar-vin, R. (1984). New K-Ar dates from basalts and the evolution of the southern Rio Grande rift. Geological Society of America Bulletin, 95, 87–99.

Serna-Perez, A., Monger, H. C., Herrick, J. E., & Murray, L. (2006). Carbon dioxide emissions from exhumed petrocalcic horizons. Soil Science Society of America Journal, 70, 795–805.

Smith, B. R., & Buol, S. W. (1968). Genesis and rela-tive weathering studies in three semi-arid soils. Soil Science Society of America Proceedings, 32, 261–265.

Soil Survey Division Staff. (1993). Soil survey manual revised [U.S. Department of Agriculture Handbook 18]. Washington, D.C.: U.S. Department of Agricul-ture, Soil Conservation Service.

Soil Survey Investigations Staff. (1996). Soil survey lab-oratory methods manual [Soil Survey Investigations Report No. 42]. Lincoln, NE: U.S. Department of Agriculture, Natural Resources Conservation Service .

Bulletin 798 • Page 66

Soil Survey Staff. (1951). Soil survey manual [U.S. De-partment of Agriculture Handbook 18]. Washington, D.C.: U.S. Government Printing Office.

Soil Survey Staff. (1999). Soil taxonomy—A basic sys-tem of soil classification for making and interpreting soil surveys, 2nd edition [Handbook 436]. Wash-ington, D.C.: U.S. Department of Agriculture, Soil Conservation Service.

Soil Survey Staff. (2006). Keys to soil taxonomy, 10th edition. Washington, D.C.: U.S. Department of Ag-riculture, Natural Resources Conservation Service.

Strain, W. S. (1959). Blancan mammalian fauna from Rio Grande Valley, Hudspeth County, Texas. Geologi-cal Society of America Bulletin, 70, 375–378.

Strain, W. S. (1966). Blancan mammalian fauna and Pleistocene formations, Hudspeth County, Texas [Bulletin 10]. Austin: Texas Memorial Museum Bul-letin.

Thornbury, W. D. (1969). Principles of geomorphology. New York: John Wiley & Sons, Inc.

Thorp, J., Cady, J. G., & Gamble, E. E. (1959). Genesis of Miami silt loam. Soil Science Society of America Proceedings, 23, 156–161.

Thorp, J., Strong, L. E., & Gamble, E. E. (1957). Ex-periments in soil genesis—The role of leaching. Soil Science Society of America Proceedings, 21, 99–102.

Van Devender, T. R. (1977). Holocene woodlands in the southwestern deserts. Science, 198, 189–192.

Vanden Heuvel, R. C. (1966). The occurrence of sepi-olite and attapulgite in the calcareous zone of a soil near Las Cruces, New Mexico, in clays and clay min-erals. In W. F. Bradley & S. W. Bailey (Eds.), Proceed-ings of the thirteenth national conference on clays and clay minerals (pp. 193–207). New York: Pergamon Press.

Vanderhill, J. B. (1986). Lithostratigraphy, vertebrate paleontology, and magnetostratigraphy of Plio-Pleis-tocene sediments in the Mesilla Basin, New Mexico. Unpublished Ph.D. thesis, University of Texas, Aus-tin.

Wang, Y., McDonald, E., Amundson, R., McFadden, L., & Chadwick, O. (1996). An isotopic study of soils in chronological sequences of alluvial deposits, Providence Mountains, California. Geological Society of America Bulletin, 108, 379–391.

Williams, G. E., & Polach, H. A. (1969). The evalu-ation of C-14 ages for soil carbonate from the arid zone. Earth and Planetary Science Letters, 7, 240–242.

Williams, G. E., & Polach, H. A. (1971). Radiocarbon dating of arid zone calcareous paleosols. Geological Society of America Bulletin, 82, 3069–3086.

Bulletin 798 • Page 67

This paper was initially prepared for a symposium honoring Dr. R. V. Ruhe. Here we also honor Dr. Guy D. Smith (above), formerly the Director of Soil Survey Investigations, whose 100th birthday was June 20th, 2007. These two men were responsible for initiating the Desert Soil-Geomorphology Project.

ACKNOWLEDGEMENT

Bulletin 798 • Page 68

Bob Ruhe stands at the junction of the “airport trenches” in upper La Mesa. Runways of Las Cruces International Airport are beyond the trenches. The Sleeping Lady Hills are on the skyline at right. Photographed August, 1959.

Bulletin 798 • Page 69

APPENDIX A—MAP

Bulletin 798 • Page 70

Desert Project Tour group picture, May 2000 (top) and tour leaders (bottom, from left) John W. Hawley, Leland H. Gile, H. Curtis Monger, and Robert B. Grossman.

APPENDIX B—DESERT PROJECT TOuR, 2000

Bulletin 798 • Page 71

APPENDIX C—A PILGRIMAGE TO THE DESERT PROJECT

The following is a reproduction of an article by Terry Canup about one of the stops on the 2000 study tour in the Fall 2000 issue of New Mexico Resources, published by the NMSU College of Agricultural, Consumer and Environmental Sciences (reprinted with permission from the author).

Bulletin 798 • Page 72

Bulletin 798 • Page 73

Bulletin 798 • Page 74

Bulletin 798 • Page 75

Notes

Bulletin 798 • Page 76

New Mexico State University is an equal opportunity/affirmative action employer and educator. NMSU and the U.S. Department of Agriculture cooperating.

May 2009 Las Cruces, NM

Contents of publications may be freely reproduced for educational purposes. All other rights reserved. For permission to use publications for other purposes, contact pubs@nmsu.edu or the authors listed on the publication.