Field Structure of Cultivated Soils with Special Reference to ...

Field Structure of Cultivated Soils with Special Referenceto Erodibilitv by Wind'ABSTRACTA studv of ~nrtc~re and erodibility nf wit hv wind was carriednut on croded and residual soil materials in a conditiont~sunllv existing in the field and alwr san~pling, qdtivating, anddry sieving.Fkld structure and erodibility of soils varied greatly with asIictlc ns I inch of simdated rainfall. In soils undisturbed byc illti\ a tion af ter rain, fowr distinct phases of structure wereIound, all of which possess different degrees of erodibility by'Conlrihution 30. 472, Department of .\gmiomy. liansa? Agric~alr~cralExperinlcnt Station. Manhattan. Kansas, mcl the SoilC:onrcrvation Service, I!. S. Departnwnt of Agricrrlturt.. Cooprraliveinvestigarions in the mechanics of wind erosion. PresentedIwlore Division I, Soil Scienre Society of America, Cincinnati,Ohio, November 20, 1952. Received for publication Decemher 8,1952.aProfessor of soils, Kansas Agricultnral Experiment Station,and Agent, Soil Conservation Service, U.S.D.A.wand. These phases are the primary aggregates (water-stable~!::;~egatn), the secondary aggregate? (granules and clods), thewrface crust, and the consolidated soil material between thesecondary aggregates.The abrasiw artkm of wind erosion was shown to be one ofthe moat serious aspects of erodibility by wind.The mechanical stability, that is, the resistance of a soil'Qwt~breakdown by mechanical forces such as cultivation or sieving,was found to vary directly with the resistance of the soil. toabrasion by wind-blown sand. Mechanical stability was greatestfor clrif t particle* (sand *grains and water-stable aggregates mostly),less fur the secondary aggregates, followed in order by thesurface crust, the consolidated materials between the secondaryaggregates, and lastly the consolidaterI materials, if any, whichheld drifted particles together after they were wetted and dried.The amount of dispersed fine silt was found to be a primaryfactor influencing the formation of the surface crust and theconsolidation of the soil body after it was wetted and dried. AEraction of clay size, being much less dispersible than silt, wasfound tq.? the primary factor of secondary aggregate forma-

186 SOIL SCIENCE SOCIETY PROCEEDINGS 1953tion. Waterdispersible silt and clay were responsible in largemeasure for the resistance of the soil to erosion by wind. Althoughthe amount of erosion was limited to some degree bythe presence ,of waterrtable aggregates too large to be movedby wind, it was found that cultivated dryland soils lack theseaggregates in sficient amounts. The resistance of these soils towind action was found to depend primarily on their ability toform secondary aggregates, or clods.The secondary aggregates prese~ed their identity below thesurface even after repeated wetting and drying in the field. Itwas concluded, therefore, that the physical condition of the soilis indicated as well or better by methods such as dry sieving,which primarily measure the state of the secondary aggregates,than by methods that measure the state of the primary aggregates.RESHLY cultivated soils are composed of a more orFless loose mixture of particles and aggregates ofwidely variable dimensions. These may range fromlarge clods several inches in diameter to particles ofdust. Erodibility in such a condition is dependent primarilyon the proportion and bulk density of erodiblefractions contained in the soil when it is dry (3). Onlydry soil materials are eroded by wind.An entirely different regimen of conditions exists ina cultivated soil after it has been wetted by rain anddried. A surface crust is almost invariably formed dueto impacts of raindrops on the ground. Often the crustis hardly discernible but has a profound influence onerodibility. Except at fhe immediate surface, the primary(water-stable) aggregates and the secondary aggregates,including clods, usually undergo little transformationby individual aet,ting from rain and drying. *A $-eater change occurs in' the degree of compactnessand cementation between the various Ytc~gnizable ag- ,gregates. This type of cementation has an importantinfluence on erodibility by wind, but the degree ofcementation is generalli tdo weak to be detecgble bywet or dry sieving. Thus, wet or dry aggregate analysesbased conventionally on sieving or elutriation in wateror air omit entirely some important aspects of soilstructure. A study was conducted to gain more specificinformation of the various phases of soil structure commonto cultivated soils in the field and how thesephases relate to erodibility by wind.,ProcedureComposite samples of soil when in a dry condition were taker.from 1 to 2 inches of depth from wind-eroded fields and fromfreshly accumulated drifts within or near the eroded fields. Fourfields were chosen, all from the black soil zone, representingLancaster sandy loam, Crete silt loam, Rokeby silty clay loam,and Sutphen clay. Henceforth, the soil will be referred to bytextural class only.Samples from the freshly accumulated drifts, still unaffectedby rain, represented that portion of the soil which had beenmoved about by the wind and deposited against obstructions ortraps, such as weeds, fence rows, or road ditches. Henceforth forconvenience, soil material obtained from there drifts will be refmdto as drifted soil material and soil material obtained fromihe wind-eroded fields as residual soil material.Determinations were made of the physical soil characteristicswhkh might have had some bearing on erodibility by wind.These determinations included mechanical composition by themethod of Bouyoucos (I), the size-frequency distribution of waterstableparticles and aggregates by the method of Yoder (9) modifiedin accordance with the latest Soil Conservation Servicerecommendations, the size-frequency distribution of dry aggregatesand their mechanical stability by the rotary sieve method(4), and the amounts of erosion determined in the laboratorywind tunnel used regularly in this work.Td measure the amount of erosion the soil was placed in aporous-bottom tray 5 feet long, 8 inches wide, 2 inches high,and e?posed to wind until soil movement cease'd. The, wind hada drag velocity of about 1.4 miles peE hour and a velocity of about25 miles per hour at a 6-inch height. The amounts of erosionfrom trays placed parallel with the direction of the wind weredetermined under three sets of conditions: (a) exposing wellmixeddry soils to wind; (b) exposing to wind after consolidatingthe soil by spraying with 1 inch of water and drying, and (c)exposing to wind and to sand abrasion after consolidating thesoil. Fine dune sand was used as abrader and was placed in atray immediately on the windward of the tray that containedthe soil.The surface crusts, caused by spraying, were tested for abradabilityby dune sand and also for resistance against breakdownby sieving. The crusts were scraped off gently with a spatulawhen the soil was dry and sieved on the rotary sieve.Another series of tests was conducted to determine more convenientlythe abradability of the different structural units ofthe soil. In these tests small volumes of clods 1/, to 1/, inch indiameter and cylindrical blocks of soil were exposed to abrasionby dune sand. The clods or blocks of soil were placed on theleeward of the tray containing the abrader. The clods wereplaced to cover the bottom of a trough sloping into the directionof the wind. The vertical moss-sectional area of that surface ofthe clods exposed to abrasion by dune sand was the same as thatof the cylindrical block of soil. The amount of abrasion wasdetermined by weighing the clods or block of soil before andafter exposure and adding to this difference the weight of thesoil fragments greater than 0.42 mm that were broken off theclod or block and lodged at the foot of it. The amount of thesoil fragments greater than 0.42 mm was determined by sieving.Consolidation of soil blocks, which were 2 inches in diameterand 2 inches high, was accomplished by placing a soil in a waxedpaper cylinder with open top and bottom and spraying the topof the soil with about 1 inch of water, followed by drying.Wetting and drying was repeated several times in some cases.In every, case the4consoJidated soil, even when composed of almostpure sand, held togqtber without crumbling.Finally, the degree pf,disporsion of the various fractions of',the soil in water was determined. Air dry soil was imdrsed ins'water in .a vacuum and the bbttIesL containing the soil. and thewater were turned end over end at the rate of 30 turns perminute for 2 minutes. The percentage ofTpar\icles smaller than0.02 mm contained in suspension was hetermined by the hydrometermethod.ResultsSIZE AND NATURE OF SOIL FRACTIONSTRANSPORTED BY WINDDry soil fractions transported by wind and depositedin drifts in the vicinity of the eroded fields were generallysmaller than 0.84 mm in diameter (table 1). Afew fractions exceeding this diameter were mainly organicfragments, considerably lighter than the mineralparticles or aggregates.It will also be seen from table 1 that soil marerialsdeposited in the drifts contained much less dust smallerthan 0.05 mm in diameter than corresponding soilmaterials from the eroded fields. Evidently much ofthis fine fraction was carried away from the vicinityof the eroded fields and only a portion of it was depositedin the drifts.Although the residual soil materials coritained somewater-stable aggregates too coarse to be moved bywind, the amount of these in the soils studied wasrelatively small. A great percentage of the water-stableaggregates was of a size readily erodible by wind(table 1).The erodibility of well-mixed, freshly cultivatedsoils (condition (a), table 1) varied with the amount oferodible fraction contained in the soil. In sandy loamand silt loam the erodible fraction was predominantlysmaller than 0.42 mm in diameter, but in the finertextured soils a considerable proportion of particlesfrom 0.42 to 0.84 mm in diameter also constituted anerodible fraction (table 1). Clay had the highest pro-

CHEPIL: FIELD STRUCTURE OF SOILS AND WIND EROSION 187r_SoilclassSoilmaterialTable 1.-Physical properties of wind-eroded (drifted) and residual soil materials.Dry fractionsWater-stable fractionsI I Amounts eroded*ISandyloamSiltloamSilty clayloamClayDriftedResidualDriftedResidualDriftedResidualDriftedResidual*Conditions described in "Procedure."portion of erodible fractions ranging from 0.42 to 0.84mm in diameter, and these fractions apparently contributedgreatly to the high erodibility (9.5 tons peracre).IINFLUENCE OF SOIL CONSOLIDATION AND OFABRASION ON ERODIBILITY BY WINDThe application of 1 inch of water sprayed on the soil followedby drying (condition b, table 1) produced a surface crdstand a cementation of the soil body which greatly reduced the :erodibility by wind. The crust was easily recognizable by its ,dense,"platy structure. This type of structure became less distinctwith depth till it merged with the mass of soil below at a depth .varying to % in&, de$ending on the soil. The most tenaciouscrust and the greatest degree of cementation below the crust wason silt loam and silty cl&y loam. A thin, fragile crust was formedon sandy loam and a thicker but even more fragile one on clay.Comequently, the greatest propprtional reduction in erodibilitydue to soil consolidation was qn silt loam and silty clay loam,less on sqpdy loam, and least on clay.The amount of erosion increased greatly when a stream ofdune sand was blown over the consol~dated soils in a dry state(condition c, table 1). The cutting, grinding, and impact actionof sand disintegrated the surface crust and some of the consolidatedmaterial below the crust into particles small enoughto be moved by wind. This type of structural disintegration isknown as abrasion and is a very important phase of the winderosion process on all soils (3). There are two main aspects ofabrasion: One is the disintegration of nonerodible or consolidatedsoil units into particles small enough to be moved bywind, and the other is the wearing-away of erodible soil unitsinto dust capable of being carried away from the vicinity of theeroded region, Methods of measurement described herein aredesigned primarily to measure the former aspect which, fromthe standpoint of wind erosion control, is perhaps a more importantaspect of the abrasion process.On the whole, the consolidated sandy loam was most subjectto abrasion. Clay had a very fragile crust which wore off readilyunder abrasion, but dry clay aggrega'tes were extremely resistantto abrasion. Silt ham and silty clay loam had a surface crust mostresistant to abrasion, yet their aggregates showed a moderatedegree of resistance to abrasion.It was evident from these preliminary tests that the differentphases of sail structure in a dry state possess different degreesof abradability. A study was continued to find possible relationshipsbetween the physical properties of the different structuralunits of the soil and their abradability.MECHANICAL STABILITY AND ABRADABILITY OFDIFFERENT STRUCTURAL UNITS OF WIND-ERODED AND RESIDUAL SOILSA separation d some structural units of the soil bysieving in a dry state was necessary to study more convenientlythe abradability of these different units. Allof the structural units broke down somewhat undersieving, but in various degrees. The rotary sieve wasused for this purpose.The breakdown of clods and other structural unitsof the soil bv sieving varied directlv with the abradabilityof th&e fraccons (figure 1). '~hus, the looselyconsolidated blocks of wind-eroded particles obtainedfrom fresh drifts were most subiect to breakdown bvsieving and also to breakdown dy abrasion with dunksand. Blocks of the same sieve-fraction from residual.., soils were less,subject to breakdown by sieving and'alsoby abrasion with dune sand. Pry clods were much lesssubject toL breakdokn by. sipving and also much less~ubject 'to disintegration Ijy abrasion into particlessmall enough to be moved by wind.The resistance oft the soil to breakdown by mechanicalforces, such as sieving or abrasioil by drifting sand,is known as mechafiical stabilitv. The mechanical stabilityis a fklative measure of cdherence or strength ofcementation ,between or within the particles or soilaggregates in a dry state and, as used in this study, isw1equal to 100 -, where W is the weight of individuaIWparticles or aggregates, or the weight of a consolidatedbody of these units before sieving and W1 is the weightafter the sieving. The speed of disintegration of theindividual particles or aggregates measures the stren~th --of cementation within these units; the speed w~thwhich consolidated bodies disintegrate to individualstructural units is a measure of cementation betweenthese units.A study of the abradability of bodies or blocks ofsoil by wind-blown soil grains showed that the soilbody is composed of several different types of structuralunits'havi%g widely different degrees of mechanicalstability or coherence and, consequently, of Bisistanceto abrasion by wind-blown soil grains. Soilgrains blown into drifts by wind had the greatestmechanical stability and, therefore, were least susceptibleto abrasion (table 2). As shown in table 1 thesegrains were, in the main, water-stable aggregates andindividual sand grains. Virtually complete resistanceto breakdown by sieving was exhibited by wind-erodedsand grains common in sandy loam soil and by aggregatesfrom clay. A somewhat lower mechanical stabilitywas exhibited by wind-eroded units (mostly waterstableaggreggtes) from silt loam and silty clay loam.

SOIL SCIENCE SOCIETY PROCEEDINGS 1953a-Consolidoted frodlon t0.42f r residual ~ soils-8-Consolidated residual soilsGrams abraded ta 0.42 mm. from fresh*t s ....................Dry aggregates and clods >0.42mm .....................Surface crust M toinchthick; ...................Fraction

the size of silt and clay'were washed down, leavingsome coarser grains loose on the surface crust. Theseloose grains moved readif in saltation and causedIsome abrasion of the sur ace crust, the amount ofabrasion depending on. the amount of the grqins, themechanical stability'of the surface crust, and the distancethat the grains moved along the ground. Evenwith a distance as short as 5 feet, it was evident fromcasual observation that the amount of abrasion of thesurface crust was much greater on the leeward thanthe windward end.Next in order of mechanical stability and resistanceto abrasion was the consolidated material initially composedof fractions smaller than 0.42 mm in diameter(table 2). These fractions are highly erodible if unconsolidatedand are found in various amounts betweenthe coarser ranules and clods. The cementing forceor cohesion % etween these particles in a consolidatedstate, as for the surface crust, was least in clay andsandy loam and greatest in silt loam soil. Consolidatedbodies composed of these fractions disintegrated at arelatively high rate under impacts of abrasion by dunesand, 1 gram of the abrader wearing away from 0.6 to1.5 grams of the consolidated material, depending onsoil texture (figure 1). Thus, the rate of abrasion ofthese consolidated bodies was 150 to 300 times as greatas that of granules and .clods.Blocks of consolidated soil composed of all structuralunits common in cuIt\vated soils abraded veryunevenly (figure 3). After ri ihort"duration of exposure .the granules and clods began? to protrude into the airstream indicating that after wetting they were merelyembedded in the fine, loosely consolidated portion ofthe soil. The strength of cementation between the aggregateswas much lower than that within these aggregates,hence the reason why consolidated blocks of soilabraded so unevenly.Last in order of mechanical stability and thereforeof resistance to abrasion were the consolidated bodiescomposed of particles less than 0.42 mm in diameterdrifted and deposited into dunes by wind (figure 1and table 2). The rate of abrasion of these consolidatedbodies was from about 300 to 1,000 times as great asof granules and clods. It is evident that the cementingmaterial capable of holding these particles togetherafter consohdatjon, thereby resisting erosion by wind,was partly removed in the form of dust < 0.05 mm indiameter (table 1). The data in table 1 show also thatthere was a considerable removal of water-stable fractions< 0.02 mm from the soil material drifted anddeposited into dunes by wind. It is apparent that theseparticles were the cause of strong cementation foundbetween the secondary aggregates or clods.To determine the influence of these fine waterstablefractions on soil structure, soils were immersedand shaken in water for 2 minutes and the particlessmaller than 0.05 mm in diameter were removed byrepeated decantation after appropriate periods of settling.The water-stable grains or aggregates greaterthan 0.05 mm were then dried. There was virtually nocementation between the water-stable aggregates afterblocks of these were wetted and dried. The water-stableaggregates from all soils acted much like sand grainsin that they failed to cohere to each other. It is apparentthat the finely dispersed fraction of the soil isto some degree responsible for holding these water-CHEPIL: FIELD STRUCTURE OF SOILS AND WIND EROSIONstable fractions together to form secondary aggregatesor clods.Discussion and ConclusionsThe study indicates that the great majority of thewater-stable aggregates contained in Chernozem soilsare of the size readily eroded by wind. Since the waterstableparticles or aggregates represent the more stablestructural units to which the soil may be disintegratedin the field, it may be expected that the increase intheir size will increase the resistance of the soil to winderosion. The problem is how to create enough of thewater-stable aggregates sufficiently large to resist thewind.The resistance of these soils to wind erosion is dependentconsequently on their ability to form whatmay be considered as secondary aggregates, that is,coarse granules and clods. In fact, emergency methodsof wind erosion control are based principally on methodsof tillage that leave the soil surface rough andcloddv. ,The formation of secondary aggregates is shownfrom this study to be due in large measure to theamount of water-dispersible cements. Tillage, pressureat depth, repeated wetting and drying, and time prob- :ably all influence the dispersion of soil cements andcontribute to'the formation of secondary aggregates orclods. Since many soils undergo frequent wettings inthe field it has beead assumed that a useful measure ofsoil aggregation would be found by sieving the soil in:- water rather than in the air (a).+ The secondary aggre- s' - gates or clods were thus considered as a transient massthat slakes with repeated wetting and drying. Thisstudy shows, however, that a clod, the same as anyother structural aggregate, is "transient," but only indegree. Individual rains are shown to have little influenceon the form or compactness of clods below thesurface. The identity of these secondary aggregates ismore or less preserved even after repeated wetting anddrying in the field. Therefore, the field structure ofsoil is indicated as well or better by methods, such asdry sieving, which primarily measure the state of thesecondary aggregates, than by methods that measurethe state of the primary aggregates. Only within a narrowzone of the immediate surface do the secondaryaggregates become appreciably disintegrated by impactsof raindrops where the soil mass assumes a structuredifferent from that below.This study shows that the formation of secondaryaggregates is due primarily to the proportion of waterdispersiblesoil particles of the size of clay, perhaps ofboth mineral and organic origih, obtained when theaggregates ,are skaken in water. When this fraction togetherwith dispersed silt is removed from the soil, thewater-stable aggregates of which the secondary ag&gates are composed remain loose much like individualsand grains. Russell (6) affirms that micro-aggregates(water-stable aggregates) " . .may be considered as uhitsfrom which clods in arable soils are built-up." Thisstudy shows that the micro-aggregates are virtuallyincapable of forming the secondary aggregates or clodsunless a certain proportion of cementing fraction ofthe size of clay is dispersed among these aggregates.The importance of secondary aggregates in, soilstructure forqation is not fully recognized in agriculturalresearch. For example, a recent review OK soil

aggregate formation (7) fails even to mention the secondaryaggregates, and the whole thesis of soil structureformation is based on the unfortunate assumptionthat the status of the water-stable aggregates as conventionallydetermined by sieving, elutriation, or sedimentationin water is a proper index of soil structure.However, soil structure as it exists in the field is acomplex condition. Consequently, no single methodcan be used to evaluate it completely.Primary and secondary aggregates are phases of soilstructure common under field conditions. There is yeta third phase that assumes major importance with respectto erodibility by wind and, no doubt, also withrespect to soil tilth and probably to erodibility bywater. This third phase is concerned with cohesiveforces that are exerted between the secondary aggregatesafter the soil has been wetted and then dried.The degree of cementation between the secondary aggregatesvaries greatly, as within the aggregates, dependingon the number and the nature of wettingand on the physical-chemical nature of the soil.The degree of cementation between the secondaryaggregates after the soil has been wetted and dried isdue in large measure to the amount of particles of thesize of silt and clay dispersed by the wetting. Dispersedsilt, although usually not considered as a soil cement,acts as a weak cement after the soil is wetted and dried.Silt disperses in water much more readily than particlesof the size of clay. The presence of large amountsof dispersed silt particles in a soil appears to cause theformation of a compact, massive structure which, whilequite resistant to wind erosion, may present a seriousstructure problem otherwise. Bradfield and Jamison(2) conclude that hard and intractible soils are usuallythose largely composed of fine silt ,having a singlegrain structure when dispersed in water. Kaspirov (5)further concludes that the main factor in crust formationis the presence in soils of water-dispersible particlesless than 0.01 mm in diameter, regardless of theamount and quality of humus or the amount of saltspresent in soil solution. The present study confirmsthe above conclusions.It is generally recognized that soil possessing a welldevelopedwater-stable structure is much more resistantto water erosion than one with a poorly developedwater-stable structure (6). Resistance to watererosion is probably primarily determined by soil permeability.The determining factor of resistance of soilto wind erosion, on the other hand, is the size of dryaggregates large enough not to be moved by the wind.Many soils that have a well developed water-stablestructure are extremely erodible by wind because the-%t~ Surface crust 0 ~econdory aggregatePtimary aggregates:& :,., Consoll&red materidbrtwren secondaryogpegatesFIG. 4. -Diagrammatic representation of field structure of soil.size of the aggregates is, on the whole, too small toresist the wind. Mineral soils resistant to wind erosiongenerally contain more than 60 per cent of dry aggregatesgreater than 0.84 mm in diameter as determinedby dry sieving, regardless of whether these aggregatesare stable in water or not.The fourth phase of field structure in cultivatedsoils is the surface crust, which results principally fromthe beating action of raindrops on the ground. Itcreates conditions of the soil surface, some of whichtend to reduce and some to increase: erodibility by I >, wind. The, beating action of raindrops consolidates~. some of the erodible soil fractions and tends to reduceerodibility. At the same,,timet it tends to smooth thesoil surface and this tends to increase the erodibility.The raindrop action also tends to leave some soilgrains or aggregates loose on the surface. These grainsare readily, poved in saltation and under certain conditionscause' considerable abrasion of the surfacecrust. The various phases of field structure in cultivatedsoils are shown diagrammatically in figure 4.Literature Cited1. Bouyoucos, G. J. Agron. Jour. 43:434 (1951).2. Bradfield, R., and Jamison, V. C. Soil Sci. Soc. Amer. Proc.(1937\ 3:70 (1938).3. ~he~il:~. S.'Soil ~ ci. 72:387 (1951).4. -. Soil Sci. Soc. Amer. Proc. (1951) 16:113 (1952).5. Kaspirov, A. I. 81. 1940.6. ~u&ell, E. W. Imp. Bur. Soil Sci., Tech. Comm. No. 37 (1938).7. Stallings, J. H. U.S.D.A.. Soil Cons. Service., SCS-TP-110,(1952). -8. Tiulin, A. F. Perm. Agric. Exp. Sta., Div. Agric. Chem., 2:77(1928).9. Yoder, R. E. Jour. Amer. Soc. Agron. 28:337 (1956).