Asa-s . ~ aw 01‘ I_ our " * {truchigan Sew: University I This is to certify that the thesis entitled SUSCEPTIBILITY OF SOME MICHIGAN SOILS TO WIND EROSION presented by Janice Ruth Stone has been accepted towards fulfillment of the requirements for Master of Science degree in Crop and Soil Sciences Mai/”kg Major professor 0-7639 InlayllwillWMWWM If; ovenou: Ewes; . I.. .c"‘ ::\ ‘ k‘ ~:.Ié’h‘?§: 25¢ per day per item RETURNING LIBRARY MATERIALS: Place in book return to renow -‘ L - .331)!” . ' charge from circulation recon mac 0 SUSCEPTIBILITY OF SOME MICHIGAN SOILS TO WIND EROSION By Janice Ruth Stone A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Crop and Soil Sciences 1980 ABSTRACT SUSCEPTIBILITY OF SOME MICHIGAN SOILS TO NIND EROSION By Janice Ruth Stone The wind erodibility of some Michigan soils was studied. Susceptibility to wind erosion was related to Wind Erodibility Groups, particle size distribution, organic matter and calcium carbonate. Forty sites comprising seven surface soil textures were sampled. An alternate version of Chepil's rotary sieve was developed. For each texture, the measured percentage of dry fractions >0.84 mm was larger than the percentage assigned to the Wind Erodibility Group of that texture. Over the ranges studied, organic matter and calcium carbonate by themselves did not have a significant affect on wind erodibility. It was shown for the first time that the effects of clay, silt and sand on wind erodibility are the same in Western Canada, the Great Plains and in Michigan - three widely different geographical regions of North America. On the basis of polynomial regression, soil with the greatest resistance to wind erosion was found to be medium textured, with 24-30 percent clay, 30-40 percent silt and 3l-45 percent sand. These percentages agree well with the optimum percentages for Western Canada and Great Plains soils. This commonality of data indicates the alternate sieve is a valid method fer determining wind erodibility of soils. ACKNOWLEDGEMENTS I wish to express my appreciation to the following people who have made this study possible. Dr. D. L. Mokma, for his guidance, encouragement and helpful suggestions during all phases of this thesis. My parents, for their own special guidance, encouragement and helpful suggestions. Dr. L. S. Robertson and Dr. G. E. Merva for serving on the graduate committee and for sharing their knowledge and experience. Mr. Shawn McBurney, for constructing the alternate sieve, and Ms. Jeanne Goericke, for her invaluable assistance in data analysis. ii TABLE OF CONTENTS INTRODUCTION ........................ LITERATURE REVIEW ...................... Airflow Near the Ground ............... Process of Wind Erosion ............... Initiation of soil movement ............ Soil transport .................. Sorting and Deposition .............. Factors Affecting Wind Erosion ............ Soil cloddiness ................. Surface roughness ................ Wind and soil moisture .............. Field length ................... Wind Erosion Control ................. Soil cloddiness ................. Ridges ...................... Barriers ..................... Vegetation .................... Wind Erosion Equation ................ Description of the variables ........... Soil erodibility index, I! .......... Soil ridge roughness factor, K' ........ Climatic factor, C' .............. Field length, L ............... Equivalent quantity of vegetative cover, V . . Development of the equation ........... The Non-erodible Fractions .............. Nature of dry soil structure ........... Factors influencing I fraction .......... Particle size distribution .......... Vegetation and residues, organic amendments and microbial activity ........... Free calcium carbonate ............ Weathering ................ .. . Tillage .................... METHODS AND MATERIALS .................... Experimental Design ................. Textures Sampled ................... 50 Site Location and Soil Sampling ........... 53 Preliminary site location ............ 53 Site location in the field ............ 56 Soil sampling ................... 57 Development of Sieve ................. 58 The standard method ................ 58 The alternate sieve ................ 60 Description of alternate sieve ......... 60 Development of alternate sieve ......... ’ 6l Development of Sieving Procedure ........... 62 Calibration ................... 63 Sample placement ................. 63 Sieving time ................... 64 Final Sieving Procedure ............... 71 Associated Analyses ................. 72 RESULTS AND DISCUSSION ................... 73 Comparison of WEG Percent 0.84 mm and Actual Percent 0.84 mm ..................... 73 Effect of Soil Properties on Wind Erodibility . . . . 82 Organic matter and calcium carbonate ....... 82 Clay, sand and silt ............... 82 Clay ..................... 82 Sand ..................... 88 Silt ..................... 89 SUMMARY AND CONCLUSIONS ................... 92 APPENDIX I ......................... 95 APPENDIX II ..................... -. . . 96 BIBLIOGRAPHY ........................ ll6 iv LIST OF TABLES Table 1 Wind Erodibility Groups and Associated Percentages . . . 25 2 Soil Erodibility I for Soils with Different Percentages of Non-erodible Fractions as Determined by Standard Dry Sieving ..................... 26 3 Factors for Conversion of Wind Tunnel Erodibility to Natural Erodibility on a Field Scale Basis ...... 31 4 Estimation of Annual from Seasonal Soil Loss on the Basis of Number and Intensity of Dust Storms at Garden City, Kansas, During l954-56 ......... 33 5 Location and Classification of SOils Sampled ...... 54 6 Assumed Textures - Results of Tukey's Test ....... 74 7 Assumed and Actual Textures, by Site .......... 76 8 Sites Reorganized by Actual Texture ........... 78 9. Actual Textures - Results of Tukey's Test. . ...... 79 Appendices I Table l - Calibration Results .............. 95 Table 2 - Sieving Results - 21 Test Samples ....... 95 II Table l - Particle Size Distribution and Organic and Inorganic Contents of Soils Sampled . . . . 95 Fig. l LIST OF FIGURES Diagrammatic representation of the relative position of the ground and vegetative roughness elements above the ground ................... 8 The alternate sieve ................... 61 Change in 4 mm fraction ...... . ............. 70 Relationship between organic carbon and >O.84 mm fraction 83 Relationship between inorganic carbon and >O.84 mm fraction ....................... 84 Relationship between clay and >O.84 mm fraction ..... 85 Relationship between sand and >O.84 mm fraction - ..... . 86 Relationship between silt and >O.84 mm fraction ..... 87 vi INTRODUCTION Wind erosion is the disintegration and movement of soil material by wind (Chepil, T944). It occurs when forces holding soil particles in place are overcome by forces tending to move the soil particles (Chepil, 1959a). This movement takes place when soils susceptible to soil blowing are devegetated through construction, overgrazing or cultivation (Kimberlien et al., 1977). The "dust bowl" of the l930's is a frequently mentioned example of wind erosion on a large scale (Chepil and Woodruff, 1963; Kimberlien et al., T977; Noodruff, l975). Serious wind erosion did, however, occur before the l930's (Call, 1936) and has occurred since (Soil Conservation Service, l980). The United States Soil Conservation Service reports that wind erosion in the l0 state Plains area has damaged an estimated 3.l million acres from the period of November l979 to February l980. This is an increase from the 1.5 million acres reported damaged during the same period the previous year. This increase was attributed to low summer and fall precipi- tation and a lack of winter snow cover. Of the 322 million acres of soil susceptible to wind erosion nationwide (Kimberlien et al., l977), 1,500,000 are located in Southern Michigan, and encompass both mineral and organic soils (Drullinger and Schmidt, 1968). Crops grown on the mineral soils include beans, potatoes, corn, tomatoes and sugarbeets. High value truck crops such as onions, carrots, celery, radishes, and head lettuce are grown on the organics. Many of these crops are sensitive to the abrasive action of windblown soil particles. Wind erosion has social, environmental and economic affects. Dust storms contribute to environmental degredation for they probably add more particulate matter to the atmosphere than all other sources combined (Kimberlien et al., 1977). The visibility reduction resulting from suspended dust has been responsible for multiple car crashes, with accompanying death and injury (Hagen and Skidmore, 1977). Particu-i late matter in the air slows or halts air traffic, clogs machinery, is deposited in buildings and causes respiratory problems. Wind erosion also affects the agricultural community. Many crops are sensitive to the abrasive action of windblown soil particles (Skidmore, 1966; Armbrust, 1968; Fryrear and Downes, 1975). Partial or even total loss may result. Market value may be lowered as surface lesions facilitate insect damage (Skidmore and Siddoway, 1978). In rural areas where wind erosion is a problem, roads and fences may be buried, and irrigation ditches filled (Woodruff, 1975). Wind erosion also affects soil productivity and water holding capacity. Organic matter, clay and silt are the most valuable portions of the soil from a productivity standpoint. These are also the parts of the topsoil most readily removed by the wind. The water holding capacity of the soil decreases over time, for coarser soil fractions are untouched by the action of the wind (Daniel, 1936; Daniel and Langham, 1936; Lyles, 1975, 1977). Conditions in the Great Plains in the 1930's provided the impetus for wind erosion research. The causes and effects of wind erosion, the erosion process, and the control of soil blowing have been intensively studied (Woodruff, 1975). Investigations into the control of wind erosion have yielded the following principles of effective wind erosion control (Woodruff et al., 1977): 1. Promote an aggregated or cloddy condition of the soil surface. Clods must be large enough to resist the force of the wind. 2. Roughen the soil surface to reduce wind velocity and trap eroding soil particles. 3. Establish barriers or crop strips to reduce field length along the direction of the prevailing wind. 4. Keep the soil surface vegetated. These precepts of wind erosion control are reflected in a wind erosion equation, developed by W. S. Chepil and his associates. Users of the equation include researchers and soil conservationists. The equation serves two purposes (Woodruff and Siddoway, 1965). It is used to estimate the average potential soil loss in Tons/Acre/Year that may occur from a given area. The equation is also used to arrive at an approximate sequence of management practices necessary to reduce wind erosion losses to an acceptable level. The present form of the equation is (Woodruff and Siddoway, 1965): E=f(I,K,C,L,V) where: E = average potential soil loss in Tons/Acre/Year I. = soil erodibility index (knoll present) in Tons/Acre/Year K. = soil ridge roughness factor CI = climatic factor r ll field length along prevailing wind direction in feet < ll equivalent quantity of vegetative cover in equivalent lbs/acre The soil erodibility index I' is determined from the percentage of dry fractions >0.84 mm in diameter. Since aggregation >0.84 mm in diameter varies inversely with wind erosion, 1' decreases as the percentage of such dry aggregates increases (Chepil, 1958). The dry aggregate state of a soil, and hence its erodibility index, is deter— mined using a standard dry sieving procedure (Chepil, 1952). This sieving has been performed for many soils of the Great Plains, where the research has been conducted (Chepil, 1959b, 1960). Two basic tables were then generated from this data (Woodruff and Siddoway, 1965; Gillette, 1978a). One related percent dry aggregation >O.84mm in diameter to I (without knoll). The other related surface soil texture to average values of percent dry aggregation and I. The Agricultural Research Service of the USDA is the agency that assigned soil textural classes into Wind Erodibility Groups (WEG's) (Gillette, 1978a). The I values in Tons/Acre/Year assigned to WEG's are used by the Soil Conservation Service in planning wind erosion control programs (Hayes, 1972). Due to a lack of dry sieving data, WEG's are used outside the Plains area, where they were developed. Michigan is one of the states for which no sieving data is available (Quisenberry, personal communication). Michigan soil conservationists are using WEG's to assist farmers in planning wind erosion control programs. The objec- tives of this research are to: Inventory the 1' values of selected surface soil textures of Michigan soils. Relate the percent dry aggregates >O.84 mm in diameter to percentages of sand, silt, clay, organic carbon and CaCO3 . LITERATURE REVIEW Scientific interest in wind erosion dates to the late 19th and early 20th centuries. In 1894, the University of Wisconsin published a bulletin concerning the wind erosion problems on coarse- textured Wisconsin soil (King, 1894). The United States government officially recognized wind erosion as a problem in 1911, when the U. S. Department of Agriculture published The Movement pf_Soi1 py_ Wind by E. E. Free and its accompanying Bibliography pf Eolian Geology by S. C. Stuntz and E. E. Free. This detailed bulletin and its extensive bibliography still did not inspire research activity (Wood- ruff, 1975). That inspiration came from the diastrous effect of the dust bowl of the 1930's. Wind erosion research has dealt with the process of wind erosion, the factors influencing wind erosion, the control of wind erosion, and the development of the wind erosion equation. Airflow Near the Ground Knowledge of wind characteristics near the ground is important in the discussion of wind erosion. The air flow involved in wind erosion is always turbulent, characterized by multidirectional eddy flow (Chepil and Woodruff, 1963). The transporting power of the wind changes with eddy flow. For this reason eddies are more important than average wind velocity in the wind erosion process. The pattern of windspeed with height above the ground is the wind speed profile (Rosenberg, 1974). The change in velocity per unit of height is the velocity gradient (Chepil, 1961). No matter what the gradient is, wind speed increases with height in an exponential manner (Chepil and Woodruff, 1963). Over a surface roughened, for example, by a crop or soil clods, wind speed at any height, Z, above the roughness elements is described by the equation (Skidmore and Siddoway, 1978): u = 4%1nflggd > (1) where: u = wind speed at height Z u* = friction velocity k = Von Karman's constant (.4) Zd = displacement height 20 = roughness parameter The factor Zd, the displacement height, is introduced for wind flow over a rough surface (Rosenberg, 1974). It is also known as the effective roughness height (Lyles, 1977) and as the zero displacement height (Chepil and Woodruff, 1963). It is the average height of the roughness elements and varies directly with the height of the elements. It separates the fast moving "free air" above the roughness elements, from the slow moving "restricted flow" below the roughness elements (Chepil and Woodruff, 1963). Figure 1 illustrates the interrelationships between Zd, lo, the roughness elements and the ground surface. The hatched area of Figure 1 represents the roughness elements. level Where Wind Profile ”Aerodynamic Surface J Extrupolales to Zero -7 ///1. v //////,////I, //z. 2:. :4/24 ///,. A ~ A / Dis laced Reference Plain Zd ' l I 4 ‘ I”, r / ‘ '0’, I, \ w ‘\ \ ll,¢ us “1 ’,‘/"‘ ‘ \\\ ' Fig. 1 Diagrammatic representation of the relative position of the ground and vegetative roughness elements above the ground. (After Chepil and Woodruff, 1963; and Skidmore and Siddoway, 1978). As seen in Figure 1, the wind speed above a rough surface ideally extrapolates to zero at some point below the tops of the roughness elements (Chepil and Woodruff, 1963; Rosenberg, 1974; Skidmore and Siddoway, 1978). Wind speed is zero at height Zd + Z0 if the surface is impervious. Over a porous surface, such as that covered by vegetation, the velocity at Zd + 20 is somewhat greater than zero. The porous nature of the roughness elements permits some air movement (Chepil and Woodruff, 1963; Skidmore and Siddoway, 1978). The protective nature of vegetation or other roughness is clear, however. The distance Zd increases with height of roughness elements so if the erodible soil is located below Zd, no erosion should occur (Skidmore and Siddoway, 1978). 20, which Chepil and Woodruff (1963) called k_is the height above the displaced reference plane where the wind velocity is zero. It may be thought of as an index of aerodynamic surface roughness for the value of lo increases as the roughness of the aerodynamic surface increases (Chepil and Woodruff, 1963). Z0 is not related to the height of the roughness elements but to variability in height, flexibility and density of the elements. The friction velocity u* of Equation 2 is defined as (Skid- more and Hagen, 1977): w = (mi: (2) where T surface drag p density of air Friction velocity is the same as Chepil's drag velocity V* (Chepil and Milne, 1941a). It is considered by many workers (Chepil and Milne, 1941a; Lyles, 1977; Skidmore and Hagen, 1977) to be an index of the capacity of the wind to erode. This is because part of the momentum of wind flowing over a surface is transferred to that surface (Skid- more and Hagen, 1977). This transfer of momentum causes the shearing stress, T, on the surface. A soil particle becomes more susceptible to movement by wind as the stress on it increases. The Process of Wind Erosion The process by which a particle susceptible to wind erosion is actually moved consists of three parts: 1. Initiation of soil movement 2. Transportation and 3. Sorting and Deposition (Chepil, 1945b). Initiation of Soil Movement Movement begins when the pressure of the wind on the soil particles overcomes the force of gravity holding them in place (Chepil, lO 1959a). The wind speed required to overcome gravity and initiate movement is the threshold velocity. It varies according to soil, crop and other environmental conditions (Chepil, 1945b; Gillette, 1978b). A fluid in motion, e.g. wind, exerts three types of pressures on a particle (Chepil, 1959a). The first type is velocity or jmpggt_ pressure. This is a positive pressure exerted on that part of the particle facing into the wind. It is due to the impact of the wind on the particle. The second type is called viscosity pressure. It is a negative pressure on the lee side of the grain. Magnitude of the viscosity pressure depends on the density, velocity and viscosity of the wind. The third type of pressure is called static, isotropic or internal pressure. This is a pressure on the top of the particle which is negative when compared to the pressure on the bottom of the particle. The pressure difference is caused by the Bernoulli effect. According to the Bernoulli law, pressure on a surface is reduced when a fluid flowing over that surface is increased in velocity. The wind speed at the top of a soil particle is generally higher than at the bottom of the particle. This pressure difference causes a ljjt_on the particle, increasing its tendency to rise (Chepil, 1945a). The sum of the impact and viscosity pressures exerted on a particle is called drag, Pressure differences between the top and bottom of the particle constitute a lift, while the force of gravity tends to counteract the lift (Chepil, 1961). The forces acting on a soil particle before movement is initiated include: drag, lift and gravity. The threshold drag and lift required to initiate soil movement 11 are influenced by particle diameter, shape, density, closeness of packing and by the angle of repose of the particle with respect to the average drag level of the wind (Chepil, 1959a; Chepil and Woodruff, 1963). Soil Transport After the forces of lift and drag initiate soil movement, the second phase of the wind erosion process, soil transport,begins. There are three types of soil transportation: saltation, suspension and surface creep (Chepil, 1945a; Lyles, 1977). Of the three, saltation is the most common, for between 50 and 80 percent of the soil moved is transported in this way (Chepil and Milne, 1939; Lyles, 1977). The average size of the soil particles moved by each of the three forms increases from suspension, through saltation, to surface creep (Lyles, 1977). The proportion of soil moved by the three forms varies with texture (Stallings, 1957). In general, coarse textured soils move by saltation and surface creep, while fine textured soils move mainly by saltation and suspension. Soil particles airborne due to the effects of lift and drag move in saltation (Chepil and Woodruff, 1963). Saltation is a series of short jumps by which a particle moves across the soil surface. When the effects of lift and drag initiate movement, soil particles leap into the air at an angle ranging from 75 to 90 degrees (Chepil, 1945a). The height of rise varies directly with the initial velocity of rise from the ground and with the velocity gradient (Chepil, 1961). Salta- ting particles do not, however, rise more than a few feet above the ground. More than 90 percent stay below one foot. Chepil also found 12 the almost vertical rise is followed by a straight line path of descent, striking the surface at an angle of 6-120. This straight- line descent path is due to the accelerating action of both wind and gravity (Chepil, 1945a; Bisal and Nielsen, 1962). Chepil (1945b) also investigated the fate of particles in saltation when they hit the ground. Upon hitting the surface, the saltating particle either rebounds in another jump, or loses its kinetic energy, thereby becoming part of the soil mass on the ground. A particle in saltation can lose its kinetic energy by striking another particle when it impacts, thus setting the second particle in motion. During the course of a wind erosion episode, a given particle can move via saltation, come to rest, and have its movement reinitiated many times (Chepil and Woodruff, 1963). If the particle set in motion by the striking action of a saltating grain is in the correct size range, it too will move by saltation (Chepil, 1945a). Soil particles can, then, begin movement by saltation in one of two ways. The first is by the direct force of the wind,and the second is due to the impact of another saltating particle. Chepil (1945 a, b) elucidated the basic relations between particle size and saltation. Soil grains moved by saltation range from 0.1 mm to 0.5 mm in diameter. Those from 0.1 mm to 0.15 mm in diameter are most susceptible to movement. Their threshold velocities are 8-9 mph at 6 inches (Chepil, 1945b). Similarly, particles smaller than 0.1 mm and larger than 0.5 mm were found to be unaffected by the direct force of winds ordinarily encountered. For the small particles, this increase in the threshold velocity is due to both their cohesive nature and to their small size. 13 They are too small to protrude above the laminar layer of slow moving air that exists at the surface over which a wind flows. Soil particles larger than 0.5 mm in diameter do protrude above the laminar layer, but their larger diameters increase the threshold velocity necessary for initiation of soil movement (Chepil, 1945 a, b, c). Particles O.84 mm in diameter (Wood- ruff and Siddoway, 1973). The usefulness of clods in controlling wind erosion is, however, dependent on soil texture. Coarse-textured aggregates are more susceptible to the disintegrating effects of saltation, weathering and field traffic (Woodruff et al., 1977). Secondary aggregates are formedduring tillage. Their size and strength depends on texture, moisture and soil density at time of tillage (Lyles and Woodruff, 1961). Ridges The usefulness of ridges also depends on soil texture. Many times, the ridges of weakly granulated soils abrade too quickly to be of any real use in controlling erosion (Woodruff et al., 1977). Roughen- ing the surface to control wind erosion can be useful if properly done on soils of suitable texture (Chepil and Woodruff, 1963). Tilling the soil surface to bring up moist, cloddy soil is an emergency control measure used when little or no vegetation protects the surface. This is called emergency tillage, and is done when erosion is either imminent or actively occurring (Woodruff et al., 1957). Barriers A more permanent set of control practices involves establishing barriers such as shelterbelts, snow fences and crop strips at right angles to the prevailing wind direction (Chepil and Woodruff, 1963). A given barrier has a drag, and by this drag it exerts a force upon the incident wind. In accordance with Newton's second law (Plate, 22 1971) the air loses momentum and its velocity is reduced. This decrease in velocity then decreases shear stress at the surface. The leeward extent of lowered velocity averages 20 to 30 times the height of the barrier (Woodruff, 1956), but in general, effective velocity reduction extends only 10 times the height of the obstruction (Woodruff, 1956). In addition to lowering velocity, barriers reduce avalanching by trapping saltating particles (Chepil and Milne, 1941b). The width of the "trap strip" effective in reducing erosion (Chepil, 1945a) varies with the density of the crop comprising the strip, and with the height of jump of the saltating particles. Vegetation The last of the four principles of wind erosion control is the establishment of vegetation or maintaining vegetative residues. They act as roughness elements by absorbing much of the total drag (Lyles et al., 1974). Crops and residues raise the zero velocity level of the wind (Chepil and Woodruff, 1963), decrease wind velocity at the surface, and trap eroding particles (Chepil, 1944). Like barriers and ridges, vegetation is more effective if it is oriented perpendicular to the prevailing wind direction (Siddoway, Chepil and Armbrust, 1965). The importance of a vegetative cover in wind erosion control is illustrated by what happened in the Great Plains when the cultural practice of summer fallow was introduced. In summer fallow, the soil is kept bare of vegetation. This conserves water in the rooting zone, but also increases wind erosion (Fenster, 1975). Minimum tillage practices that maintain crop residue while increasing water infiltration 23 and storage are now widely used (Fenster and Wicks, 1977). Other erosion control techniques using vegetation are: mulching, cover crops and establishment of permanent vegetation on marginal lands (Siddoway et al., 1965; Fenster and Wicks, 1977). Wind Erosion Equation A wind erosion equation integrating these principles of wind erosion control has been developed through work done in the Great Plains (Chepil and Woodruff, 1954; 1959; Chepil, 1960). It is used to estimate the potential for soil loss from a given agricultural field and to arrive at a sequence of management practices needed to reduce wind erosion to an acceptable level (Chepil and Woodruff, 1963). Five tons/acre/year is the maximum tolerable soil loss generally thought acceptable (Woodruff and Armbrust, 1968). The present form of the erosion equation is (Woodruff and Siddoway, 1965): I"! ll '93 A v—r . 7: . n ' r U < V (3) where M II potential average soil loss in T/Acre/Year H II soil erodibility index in Tons/acre/year soil ridge roughness factor climatic factor I" o 7< 11 field length along the prevailing wind direction in feet < ll equivalent of quantity of vegetative cover in equivalent lb/Acre The potential average soil loss, E, is expressed as a function of the 5 equivalent variables, for the complexity of the interrelationships 24 between the variables prohibits a simple mathematical solution (Chepil and Woodruff, 1963). Because of this, charts and tables were developed to permit a graphical solution. The charts and tables were cumbersome, so researchers in Kansas developed a computer program to solve the equation. In the field, Soil Conservation Service personnel use a wind erosion equation "slide rule" that permits easy solution of the equation (Chepil and Woodruff, 1954, 1959; Woodruff and Siddoway, 1965; Skidmore et al., 1970). Following is a brief description of each variable and its role in the equation. The infor- mation is adapted from Woodruff and Siddoway (1965). Description of the Variables Soil Erodibility Index I. 1', the soil erodibility index in Tons/Acre/Annum, is the potential soil loss from a "wide, unsheltered isolated field, with a bare, smooth, non-crusted surface." The value of I'is dependent on the cloddiness of the soil, and increases as the percentage of dry fractions >0.84 mm in diameter increases. The percentage dry fractions >0.84 mm in diameter can be determined two ways. The preferred method is the standard dry sieving procedure (Chepil, 1952; 1962). Where the sieve is not available, the percentage is determined from a table relating Wind Erodibility Groups to an average percentage of dry factions >0.84 mm in diameter (Table l) (Quisenberry, 1978). After the percentage of dry fractions >0.84 mm is determined for, or assigned to a soil, soil erodibility I in tons/acre is read from Table 2. In the solution of the wind erosion equation, soil erodibility, I, is multiplied by knoll erodibility, Is, to give erodibility yea. 4L 25 TABLE 1 WIND ERODIBILITY GROUPS AND ASSOCIATED PERCENTAGES (from Hayes, 1972) Soil Texture Class Very fine sand, fine sand, sand, coarse sand Loamy very fine sand, loamy fine sand, loamy sand, loamy coarse sand, sapric organic materials Very fine sandy loam, fine sandy loam, sandy loam, coarse sandy loam Clay, silty clay, noncalcareous clay loam, and silty clay loam with > 35% clay Calcareous loam and silt loam, calcareous clay loam and silty clay loam with <35% clay Noncalcareous loam and silt loam with <20% clay, sandy clay loam, sandy clay, hemic organic materials Noncalcareous loam and silt loam with >20% clay, noncalcareous clay loam with <35% clay Silt, noncalcareous silty clay loam with <35% clay, fibric organic material Soils not suitable for cultivation due to coarse fragments or wetness, wind erosion not a problem % >0.84 mm l 10 25 25 25 4O 45 45 SOIL ERODIBILITY I FOR SOILS WITH DIFFERENT 26 Table 2 PERCENTAGES OF NONERODIBLE FRACTIONS AS DETERMINED BY STANDARD DRY SIEVING* (from Woodruff and Siddoway, 1965) Percentage . of dry soil Unlts fractions >0.84 mm 0 l 2 3 4 5 6 7 8 9 tens tons/acre 0 --- 310 250 220 195 180 170 160 150 140 10 134 131 128 125 121 117 113 109 106 102 20 98 95 92 90 88 86 83 81 79 75 3O 74 72 71 69 67 65 63 62 60 58 40 56 54 52 51 50 48 47 45 43 41 50 38 36 33 31 29 27 25 24 23 22 60 21 20 19 18 17 16 16 15 14 13 70 12 ll 10 8 7 6 4 3 3 2 80 2 -- -- -- -- -- -- -- -- -- *For a fully crusted soil surface, regardless of soil texture, the erodibility I is, on the average, about 1/6 of that shown. E1 = I x Is = I . This accounts for the presence of a significant knoll. The value of Is depends on the slope of the knoll,. For a flat field, the value of IS is set at 1.0. The soil erodibility values in Table 2 give the loss that would occur from a "wide, unsheltered, isolated field, with a bare, smooth, non-crusted surface," as if it were located at Garden City, Kansas during the severe wind erosion years of 1954, 1955 and 1956 (Woodruff and Siddoway, 1965). modify the I value to reflect local conditions of roughness, field length, vegetative cover and climate. The other factors in the equation 27 1 Soil Ridge Roughness Factor, K The soil ridge roughness factor is determined from Kr, a linear measurement of roughness elements of the soil, K. evaluates the effect of surface roughness other than that caused by vegetative residue or clods. A chart is used to determine K from Kr. The chart reflects the fact that the effectiveness of ridges in reducing wind erosion decreases as ridge height increases or decreases beyond certain limits (Woodruff et al., 1977). Erodibility E1 = I is then multiplied by K' to give Erodibility £2 = 1' x K'. Climatic Factor C. The climatic factor C. is related to two subvariables, wind velocity V, and the P-E index of Thornthwaite. C. is given a value of 100 percent at Garden City, Kansas. Climatic conditions at other locations either increase or decrease its value. To account for local climatic conditions, erodibility E2 = II X K. is multiplied by CI to give erodibility E3 = 11 X K'.X C'. Field Length Along Prevailing Wind Direction, L Avalanching makes L',the length of unsheltered field along the prevailing wind direction, an important consideration (Chepil, 1946b). L. is composed of two subfactors. They are Of, the total distance along the prevailing wind direction, and Db’ the distance along the prevailing wind direction protected by a barrier. To evaluate the effect of L', the angle of deviation of the prevailing wind direction from normality to the field must be determined. In 1965, when the article was written,data on prevailing wind direction was 28 available only for the Great Plains (Woodruff and Siddoway, 1965). This has now been expanded to include much of the United States (Skidmore and Woodruff, 1968). Of is determined using an alignment chart. The chart relates angle of deviation, and field width to Df. If a barrier is present, it is accounted for by multiplying its height by 10 to give Db. Db, subtracted from 05, gives L'. There is no simple relationship between E and L , so a graph is used to determine E4 = I X K X C X f(L ). Equivalent Quantity of Vegetative Cover V Equivalent quantity of vegetative cover, V, is related to three subfactors: quantity of cover, R'; kind of cover S, and orienta- tion of cover, Ko. RI is determined at the location in question using a standardized procedure (Cepil and Woodruff, 1954). Kind of cover, S, reflects the influence of cross sectional area of the vegeta- tion, while orientation of cover, Ko, takes into account the effective- ness of standing vs. flat cover in reducing wind erosion. As with field length, a graphical solution is necessary to evaluate E5 = E = I. X K. X C. X f(L.) X f(V). One of three charts is used to determine V from R'. The choice of charts depends on the kind of cover. The value for V read from the chart reflects orientation, Ko. A final chart is used to arrive at E5 = E = I. X K. X C. X f(L.) X f(V) in tons/acre/year. Development of the Equation Development of the wind erosion equation began when Chepil (l956b)determined there was a relationship between erosion and the 29 size and proportion of dry clods in the soil. This study was the beginning of investigations into the I factor (Chepil and Woodruff, 1963). The first wind erosion equation was developed to "estimate the relative susceptibility of field surfaces to erosion by wind, or conversely, to evaluate the effectiveness of crop residues and tillage practices in reducing erosion" (Chepil and Woodruff, 1954). The equation was developed from wind tunnel studies, and had the form: I X = 491.3. 4 FT—RK) .835 l ) where: X = wind tunnel erodibility in Tons/Acre I = soil erodibility index, based on percent dry fractions >0.84 mm in diameter R = crop residue in lbs/acre K = ridge roughness equivalent in inches (Chepil and Woodruff, 1954) The soil erodibility index I was a dimensionless expression of wind tunnel erodibility (Chepil and Woodruff, 1959). It was equal to X2/X], where X1 was the amount of erosion occurring under wind tunnel conditions from a soil with 60 percent of its clods >0.84 mm in diameter, and X2 was the amount eroded from the same soil, in the same wind tunnel, when the percent of clods >0.84 mm in diameter is not 60. The size of 0.84 mm in diameter is the approximate dividing line between erodible and non-erodible soil fractions (Chepil and Woodruff, 1963). 30 To solve the equation, an alignment chart was used. The data needed to determine wind tunnel erodibility were the percentage dry fractions >0.84 mm, the amount of crop residue in lbs/acre, and the ridge roughness equivalent in inches. Since a wind tunnel must be used to determine K directly, K was estimated for general use from photographs of fields with known K values. Through additional research, Chepil and Woodruff (1959) then revised this method to take the effect of the surface crust on erodibility into consideration. The constants of equation 4 were changed, from 491.3 and 0.835, to 400 and 1.26 respectively. They found the crust too fragile to be determined by dry sieving, so surface texture was used as an index of crusting. Chepil and Woodruff (1959) felt they were justified in including a crusting parameter, for they found a crusted surface to be common on cultivated soils. In this revision, wind tunnel erodibility, X, was determined as before (Chepil and Woodruff, 1954), but with the alignment chart modified for the new constants. The wind tunnel erodibility was then multiplied by a factor, F, to give natural erodibility: E = FX (5) where: E = natural erodibility, defined as the relative erosion occurring under field conditions from a comparable series of winds F = a factor, whose value depends on textural class of the surface soil X = wind tunnel erodibility (Chepil and Woodruff, 1959) 31 The value of the factor F increased as the soil crust became more fragile. This is shown in Table 3. TABLE 3 FACTORS FOR CONVERSION OF WIND TUNNEL ERODIBILITY TO NATURAL ERODIBILITY ON A FIELD-SCALE BASIS (from Chepil and Woodruff, 1959) Soil textural class Factor F Fine sand ................... 6 Loamy fine sand ................ 4 Fine sandy loam and clay ........... 2 Loam, silt loam, clay loam, silty clay loam. . 1 Measurements of the rate of soil movement downwind at varying distances across eroding fields was the source of the data used in the next modification of the equation (Chepil, 1959b). This revision provided the method to evaluate the influence of: 1) deviation of the prevailing wind direction from normality to the field or to a barrier, 2) field length along the prevailing wind direction and 3) barriers. The revised method also helped the soil conservationist determine the width of field necessary to control wind erosion. At this time, the form of the wind erosion equation was: E = IRKFBWD (6) where E = relative field erodibility I = soil cloddiness factor R = ridge roughness factor K = soil abradibility factor (formally factor F) B = wind barrier factor 32 W width of field factor 0 wind direction factor The solution of this equation involved determining natural erodibility as before (Chepil and Woodruff, 1959), and then using new graphs and alignment charts to arrive at relative field erodibility E. Equations 4, 5 and 6 gave only a relative indication of erodibility (Chepil and Woodruff, 1954, 1959; Chepil, 1959b),so the next step in the development of the equation was to convert relative erodibility, E, to annual soil loss in Tons/Acre/Year (Chepil, 1960). To do this, a field study was conducted during the severe wind erosion seasons of 1954, 1955 and 1956. Sixty nine sites in western Kansas and eastern Colorado, mostly fields sown to winter wheat, were evaluated for soil loss due to wind erosion. Soil loss was measured for the wind erosion season for each of the three years. The wind erosion season was defined as beginning January lst and extending through April. Chepil used two methods to estimate the average depth of soil removed. 1. Measuring the depth to which the wheat crowns were exposed, and 2. Measuring the difference in depth to the plow pan from the beginning of the wind erosion season to the end of the wind erosion season. Average depth of soil removed was converted to seasonal loss in Tons/Acre, assuming 2,000,000 lbs. for an acre furrow slice 6 inches deep (Chepil, 1960). Soil loss per season, which was measurable on only 24 of the 69 sites, was then converted to annual soil loss. This was done using an analysis of the intensity and 33 frequency of dust storms occurring at Garden City, Kansas during the years 1954, 1955 and 1956. Dust storm intensity was measured using the relationship of visibility to dust concentration determined by Chepil and Woodruff (1957). On the basis of the data in Table 4, a conversion factor relating seasonal to annual soil loss was calculated. Seasonal soil loss from each plot was multiplied by the conversion factor (1.293) to convert to annual soil loss (Chepil, 1960). TABLE 4 ESTIMATION OF ANNUAL FROM SEASONAL SOIL LOSS ON THE BASIS OF NUMBER AND INTENSITY 0F DUST STORMS AT GARDEN CITY, KANSAS DURING 1954-56 (from Chepil, 1960) January 1 to April 30 Quantity of Total storms Number Of dUSt storms dust at 6 feet times dust Visibility 1954 1955 1956 Total above ground concentration miles . mg./cu. ft.- 0-0.5 5 9 1 15 5.0 75.0 0.5-1 2 '1 2 5 1.2 6.0 1-3 13 3 2 18 0.5 9.0 Total 20 , 13 5 . 38 90.0 Calendargyear 0-0.5 7 9 2 18 5.0 90.0 0 5-1 2 2 3 7 1.2 8.4 1-3 23 4 9 36 0.5 18.0 To 1 32 15 14 61 116.4 Conversion factor from seasonal to annual soil loss therefore is 116.4/90.0 = 1.293. After the relative erodibility for each plot was calculated (Chepil and Woodruff, 1959; Chepil, 1959b) annual soil loss in Tons/Acre was plotted against relative erodibility. The resulting curve was the 34 tool wind erosion researchers needed to convert relative field erodibility to annual soil loss. However, Chepil (1960) wrote: In view of great inaccuracies in measuring rela- tively small annual soil losses from depth of soil removal, conversion of the relative field erodibility to annual soil loss based on the curve of . . . must be regarded only as highly approximate. The next step in the development of the wind erosion equation was development of the climatic factor (Chepil et al., 1962). The climatic factor is a wind velocity-surface soil moisture parameter. For any area other than Garden City, Kansas, it is equal to v3 c = 100 2 / 2.9 (7) (P-E) - where: C = climatic factor V = mean annual wind velocity at 30 feet P-E = potential evaporation index of Thornthwaite 2.4 = average value for C at Garden City, Kansas Since 2.9 is the average value for C at Garden City, C is expressed as a percent of the climatic factor C at Garden City (Wood- ruff and Siddoway, 1965). Climatic factor C is directly related to the cube of wind velocity, and inversely related to the square of P-E, because the rate of soil movement also varies directly as the cube of velocity (Chepil and Milne, 1941a) and inversely as the Square of effective moisture (Chepil, 1956). The P—E index was used instead of effective moisture, for data to determine effective moisture was not widely available (Chepil et al., 1962). A new form of the equation was subsequently published. It reflected both the inclusion of the climatic factor, and the 35 consolidation of barrier, wind direction and field lengths factors into the factor L - equivalent length of field (Chepil and Woodruff, 1963). E = f (I, C, K, L, V) (8) where: E = average annual soil loss in Tons/Acre/Year I = soil erodibility = local wind erosion climatic factor soil surface roughness = equivalent width of field < '- 7< ('5 11 = equivalent quantity of vegetative cover In addition, the soil cloddiness factor I of equation 6, a relative value, (Chepil, 1959b) was changed to the soil erodibility I in Tons/Acre/Year of Equation 8. The field studies conducted by Chepil in 1954-1956 made this conversion possible (Chepil and Woodruff, 1963). Another change from equation 6 to equation 8 was in the soil abradi- bility factor F. It was discarded from equation 8, for surface crusts were thought to be too transient when considering erosion on an annual basis (Chepil and Woodruff, 1963). Since the amount of wind erosion occurring from a knoll is higher than that for level terrain (Doughty and Staff, 1943 as cited by Chepil et al., 1964a) the I factor was modified accordingly in 1964.The isovelocity lines of wind flowing over knolls with slopes greater than 1.5 percent and lengths less than about 500 feet are compressed,with the amount of compression directly related to steep- ness of slope (Chepil et al., 1964; Woodruff and Siddoway. 1965). 36 Using an analysis based on this information, Chepil et al., computed the amount of erosion that would occur from the crest or from the slope of any significant knoll relative to the amount occurring from a level surface. With the relative soil loss from a level surface equal to 100 percent, the relative soil loss Of a knoll crest or its slope,Is, was shown to be greater than 100 percent (Chepil et al., 1964). When a significant knoll is present in a field, the I factor, or potential soil loss in Tons/Acre/Annum for a flat surface, must be multiplied by IS giving I , the soil erodibility index (Chepil et al., 1964; Woodruff and Siddoway, 1965). In their 1964 paper, Chepil et al., (1964) presented a chart whereby IS could be determined if the windward knoll slope is known. This evaluation of the effects of slope on wind erosion was incorporated into the wind erosion equa- tion, giving the form in present use: E = f(I', K', c',L', v) (3) where the variables are defined as before (Woodruff and Siddoway, 1965). The final modification came in 1968; Woodruff and Armbrust published a paper in which they demonstrated the importance of monthly variations in wind velocity and soil moisture on wind erosion. They derived a monthly climatic factor using average monthly wind velocity in place of average annual wind velocity. Woodruff and Armbrust (1968) recommended use of the monthly climatic factor for accurate results. Values fOr the monthly climatic factor for most areas of the United States have been published (Skidmore and Woodruff, 1968). 37 The Non-erodible Fractions I is the most important of the five fractions comprising the wind erosion equation (Woodruff and Siddoway, 1965). For two soils under the same conditions of climate, roughness, vegetation and field length, the soil with the highest percentage of its dry structure and particles >0.84 mm in diameter is the least erodible. The para— meter of percent >0.84 mm in diameter is then, a simple index of erodibility (Lyles and Woodruff, 1962). The influence of soil structure on wind erosion arises from the sheltering effect of dry soil fractions large enough to resist the forces of the wind (Chepil, 1950b). The susceptibility of a soil to wind erosion therefore depends on the number and size distribution -of the non-erodible fraction. However, the size and numbers of these structural units are influenced by their resistance to the forces of breakdown (Chepil, 1951). These disintegrating forces include tillage, weathering, abrasion and raindrop impact. Resistance to break down, or mechanical stability, is in turn affected by soil factors such as particle size distribution, calcium carbonate content, soil moisture, and microbial activity (Chepil, 1953b). Conditions at the time Of aggregate and clod formation are also reflected in the characteristics of the non-erodible fraction (Lyles and Woodruff, 1962). Nature of Dry Soil Structure Chepil (1953a) described the nature of recently cultivated soil well when he wrote: Freshly cultivated soils are composed of a more or less loose mixture of particles and aggregates of widely varying dimensions. 38 These may range from large clods several inches in diameter to particles of dust. Since crusted soil is more common than the condition described above, Chepil (1953a) concentrated on characterizing the structure of a cultivated soil with a crust. ' He found four types of structure present in a crusted cultivated soil. The mechanical stability of the four types varied in a definite manner. The four phases Of dry structure arranged in order of decreasing structural stability are: (Chepil, 1953a) 1. primary aggregates (water-stable aggregates) secondary aggregates (granules or clods) surface crust boom consolidated soil material between secondary aggregates The primary aggregates consist of individual soil particles held together with water stable cements. They exhibit a high coherence and are very stable against weathering and abrasion (Chepil, 1953a, c). Primary aggregates are usually <1 mm in diameter, and most are of the Size range easily moved by wind (Chepil and Woodruff, 1963). This is shown by the composition of drifts resulting from the deposition of saltating particles against an obstruction. The drifts are composed primarily of water stable aggregates and individua1_sand grains (Chepil, 1953a). Any primary aggregates in the soil body that are >0.84 mm in diameter will reduce erosion just as secondary aggregates of that same size class (Chepil, 1953c). Primary aggregates less than 0.02 mm in diameter will also reduce erosion. They are too small to protrude above the laminar layer, and their cohesiveness inhibits movement of 39 larger particles. Secondary aggregates are the second phase Of dry soil structure. They are composed of both primary particles and primary aggregates held together with cements unstable in water (Chepil, 1953c). The cements consist of water dispersible particles <0.02 mm in diameter. Due to the nature of their cementation, secondary aggregates are unstable during wet sieving, and their quantity is determined using the dry sieving technique (Chepil, 1952; Chepil and Woodruff, 1963). Secondary aggregates >0.84 mm in diameter are an important aspect of the dry soil structure. They comprise the greater part of the soil resistant to wind erosion (Chepil, 1953a). Primary particles such as gravel and coarse sand, and water stable agregates >0.84 mm in diameter make up the rest of the non-erodible fraction. The third phase of dry soil structure is the surface crust. A crust is formed when the force of raindrops disintegrates surface aggregates. The soil disperses in the water, and forms the crust upon drying (Chepil and Woodruff, 1963). A major factor influencing crust formation is the amount of water dispersible fine silt in the soil (Chepil, 1953a). Silt disperses more readily in water then clay does, so medium textured soils with large amounts Of silt farm the thickest and most stable crusts (Chepil, 1953a; Chepil and Woodruff, 1963). A crusted soil is more resistant to wind erosion if no saltating particles abrade the crust. If a section of soil is not crusted, or if part of the crust is broken by mechanical means, salta- tion and subsequent abrasion can begin. Consolidated material between secondary aggregates is the 4O fourth type of dry structure. The erodibility of a soil decreases as the degree of consolidation increases (Chepil, 1953a). Consolida- tion occurs after wetting and drying, and this causes cementation between the secondary aggregates. The primary cause of consolidation was found to be the clay and silt dispersed during wetting. In general, silt is not considered an effective cement. It assumes more importance here though, because of its greater tendency to disperse in water (Chepil, 1953a; Chepil and Woodruff, 1963). Factors Influencing the I Fraction Although all phases of dry soil structure influence the erodibility of a soil, only that portion >0.84 mm in diameter, also known as the I fraction, is considered in assessing the susceptibility of a soil to wind erosion (Woodruff and Siddoway, 1965). This is because surface crusts and consolidated materials between secondary aggregates are too fragile to be sieved (Chepil, 1953a). The amount of water stable aggregates >0.84 mm in diameter is not-used as an index of erodibility because Chepil (1953a) found their numbers too limited to be of any importance in reducing erosion. This work was with dryland soils. Since the I value of a soil is based on its per- cent dry fraction >0.84 mm in diameter, factors influencing both the amount and mebhanical stability of that portion of the soil were considered., In a soil, the amount and stability of the dry aggregates >0.84 mm in diameter depends primarily upon (Chepil, 1953b): 1. Particle size distribution 2. Organic amendments 41 3. Vegetation and vegetative residue 4. Microorganisms and the products of microbial decomposition 5. Free calcium carbonate 6. Weathering 7. Tillage Particle Size Distribution The influence of particle size distribution on erodibility was investigated by Chepil (1955a). When soils of only one textural constituent were analyzed for erodibility, sand was the most erodible, followed by clay, and then by silt, 0.005-0.01 mm in diameter. Mix- tures of 95 percent sand and 5 percent of either silt or clay both had the same wind tunnel erodibility. For clay or silt contents above 5 percent, the silt produced more clods than clay, but they were of a softer nature. In general, erodibility decreased as the proportion of silt to sand increased. The effect of clay was more variable. Erodibility decreased as clay increased, but only if the clay content was between 20 and 30 percent (Chepil, 1955a). In an earlier study, soils became more erosive as their clay contents increased above 40 percent (Chepil, 1953b). How- ever, no mixture was more erosive than those containing >75 percent fine sand. It was concluded soil with 20-30 percent clay, 40-50 percent silt and 20-40 percent sand had the highest proportion of non-erodible clods. These findings have been confirmed by others (Schmidt and Triplett, 1967; Anderson and committee, 1966). Clay was shown to 42 increase aggregate size, particularly in soils low in organic matter (Hsieh and Wildung, 1969). Sands, fine sands, loamy fine sands, loamy sands and sandy loams are considered most erodible. They form unstable clods susceptible to abrasion and subsequent erosion (Anderson and committee, 1966). Loams, silt loams, clay loams and silty clay loams are considered most resistant to wind erosion. This is especially true if their silt ranges in size from 0.005 to 0.001 mm in diameter. Many of these effects have been attributed to the nature of the primary particles (Chepil, 1953b; Chepil and Woodruff, 1963; Hsieh and Wildung, 1969). Sand has little cohesiveness, so the clods that do form in sandy soils are more susceptible to disruption by mechanical forces. Clay and silt are viewed as aggregating agents for they exhibit the binding action important in aggregate formation. A model to explain the increase in mechanical stability seen with decreasing particle size was developed by Smalley (1970). He found mechanical stability directly related to tensile strength. Tensile strength in turn was inversely propOrtional tO the cube of the particle diameter, and directly proportional to packing density and interparticle bond strength. Vegetation and Residues, Organic Amendments and Microbial Activity Vegetation and vegetative residue, organic amendments, and microorganisms and their decomposition products are interrelated in their effects on the I fraction. For this reason, these factors shall be discussed together. Vegetation and vegetative residue affect the I fraction in two direct ways. In an erosion situation, a vegeta- tive cover reduces the number of particles in saltation (Chepil, 1957). 43 This decreases the amount of abrasion the I fraction is subjected to. A cover of vegetation or vegetative residue also decreases aggregate breakdown by absorbing the impact energy of raindrops (Stallings, 1957). The decomposition of vegetative matter and organic amendments affects the I fraction, but in more tenuous waySIChepil, 1955b). Qualitative Observations in Canada indicated soils high in organic matter, with good "tilth," and with a high nutrient level were very susceptible to wind erosion. Experiments were started in the Great Plains in 1955 to investigate the relationship between erodibility and organic matter. Varying amounts of straw were added to nine different Great Plains soils (Chepil, 1955b). Decomposition of the straw caused an initial increase in the percentages of both water stable and secondary aggregates >0.84 mm in diameter. This decreased wind tunnel erodibility. The effects were more pronounced as increasing amounts of organic matter were added. After the additions stopped, the beneficial effects disappeared in a year or two. Aggregates >0.84 mm in diameter then decreased in numbers, and the soils became more erodible. The soils stayed more erodible for 2 to 5 years, depending on the amounts of straw initially added. The effects lasted longer with greater quantities of straw. Based on this, Chepil recommended vegetative residue and organic amendments not be incorporated, but left on the surface. In this way, they would decompose more slowly, spreading out the initial benefits of decomposition over a longer period of time. The findings of Chepil agree with those of soil microbiologists. 44 AS microorganisms metabolize an energy source such as vegetative residue or organic amendments, they produce organic by-products. These organic materials act as soil cements and cause the initial increase in aggregation (Harris et al., 1966). When the residue or amendment is no longer available for metabolism, the organic binders causing the initial increase are used as energy sources and metabolized by soil microflora. After the initial binders are destroyed, they are replaced by secondary cements. The secondary cementing agents are more brittle, and are not as effective in main- taining large aggregates (Chepil and Woodruff, 1963). Free Calcium Carbonate The presence of free calcium carbonate is another soil property affecting the I fraction. The influence of differing amounts of CaC03 on some soils of the Central United States was studied by Chepil (1954a). Precipitated CaCO3 was shaken in water with each of three textures: silt loam, fine sandy loam, and loamy-fine sand. The soil-CaCO3 mixture was then dried. Amounts of calcium carbonate added were 0, l, 3 and 10 percent. The effect of CaCO3 depended on soil texture. For silt loam and fine sandy loam, CaCO3 increased soil erodibility by reducing both the proportion of the I fraction and the mechanical Stability of the I fraction. Maximum increase in erodibility resulted from the addition of 3 percent CaC03. Loamy fine sand responded differently to the addition of CaCO3. Calcium carbonate increased the proportion of its I fraction and increased the mechanical stability. The effects on all textures remained as long as there was lime in the soil. 45 Chepil and Woodruff (1963) explained the aggregating effect of CaCO3 on loamy fine sand by equating its effect to silt. Silt increased the aggregation of sandy soils in a manner very similar to that of CaC03. In addition, the crystals of precipitated CaCO3 are the size of silt when they are shaken in water. The decreases in non-erodible aggregates seen with silt loam and fine sandy loam was attributed to the flocculation phenomena. Weathering Another factor influencing the I fraction is not a soil property, but a process: weathering. Soil structural conditions are influenced by wind, by freezing and thawing, and by wetting and drying (Chepil and Woodruff, 1963; Gillette, 1977). The effect of wind blown abrasive particles has been discussed, but winds can also act alone, or in conjunction with rain to disintegrate aggregates. In Texas, the action Of wind alone was shown to decrease clod size and increase erodibility (Gillette, 1977). The effects of wind velocity and the intensity and duration of rainfall were investigated by Lyles et a1. 0969). For a given clod size and wind velocity, a 10 minute rain falling at the rate of 5.6 cm/hr was as disruptive as a 90 minute rain falling at a rate 1.6 cm/hr. When a rain was windblown, 66 percent more soil was lost from the clods. This was due to increasing drop size with increasing wind velocity. The second type of weathering is wetting and drying. It can be either a disruptive or a consolidating process (Chepil, 1953a, b). Wetting and drying can cause cementation of soil between the secondary aggregates. This is from the shrinkage of water films associated with 46 fine particles. The cementation increases the resistance of the soil to abrasion (Chepil, 1953a). The disruptive action of wetting is noticeable in fine textured soils, especially those with >40 percent clay (Chepil, 1953b; Chepil, 1954b). Air trapped in "dead end pores" during wetting can also break aggregates apart (Taylor and Ashcroft, 1972). The third type of weathering is freezing and thawing. The expansion Of soil water during the freezing process can cause aggregate breakdown (Chepil, 1954b). The effect of freezing water on soil structure was demonstrated over the course of two Kansan winters (Chepil, 1954b). One winter was moist, and the other was dry. Compared to the dry winter, frost action during the moist winter broke down more aggregates to a size 0.84 mm in diameter than other a one way disk or surface sweep. The clods produced by the moldboard plow also had a higher mechanical stability. Differences in the I fraction due to kind of implement were not lost by rain, but were wiped out by weathering and the effects of subsequent tillage (Lyles and Woodruff, 1962). Tillage Operations of the stubble mulch system of farming common in the Great Plains have also been studied (Woodruff et al., 1965). 49 Differences in the I fraction after initial and subsequent tillage were assessed. After initial tillage, the size and stability of secondary aggregates did vary with type of implement. The effects of subsequent tillage operations were quite variable, however. The investigators concluded they could make no statement concerning the effect of subsequent tillage on soil structure. Tillage is the last of the primary factors influencing the dry fraction >0.84 mm in diameter. The processes and soil properties discussed have both disrupting and consolidating tendencies. Dry clod structure is a constantly changing facet of the soil (Bisal and Furguson, 1968). It can be influenced by particle size distribution, vegetation and vegetative residue, organic amend- ments, microorganisms and the products of decomposition, calcium carbonate, weathering and tillage. The dry fraction is in a continual state of flux, depending on the net effect of the aggregating and dis- integrating forces acting upon it. invo >0.3 must 111V 81‘! C0 METHODS AND MATERIALS Analysis of a soil sample for susceptibility to wind erosion involves sieving the soil to determine the percentage of dry fraction >0.84 mm in diameter (Woodruff and Siddoway, 1965). This percentage must be known if the I factor for a soil can be determined with accuracy. The Agricultural Research Service of the U.S.D.A. has assigned I values to surface soil textures for use in areas where no sieving data is available (Gillette, 1978a). In Michigan, no soils had been sieved to determine the proportion of non-erodible dry fractions present. An inventory was then conducted to Obtain a clearer picture of the wind erodibility of some Michigan soils. Experimental Design The erodibility of five surface textures was studied. They covered Wind Erodibility Groups 1, 2, 3, 4L.5 and 6. Loam was sub- divided according tO specific soil properties. Textures were: 1. fine sand 2. loamy fine sand 3. loamy sand 4. sandy loam 5. loam a. with free CaCO3 b. without free CaCO3 . >20% clay O Q . <20% clay 50 51 All sampling sites were located in Michigan, within an area south Of a line extending from Bay to Oceana counties. Most of Michigan's agricultural activity is concentrated in the southern half of the lower peninsula (Michigan Department of Agriculture, 1980). For each texture and subdivision of loam, five sampling sites, each 25 m2, were located within this area. Ten subsamples were collected at each site. Percent dry fraction >0.84 mm in diameter was determined for each of the ten subsamples. Five of the 10 subsamples were analyzed for particle size distribution and organic carbon. Calcareous loams were also analyzed for inorganic carbon. The five sites of each texture were located as widely apart as possible across the sampling area, giving a good indication of the general erodibility of Michigan's soils. The relatively large number of subsamples were collected at each site to allow a better evaluation of the performance of the sieving method developed for use in the study. Where possible, the soils at the five sites for each texture were of the same soil type. This decreased variability from Site to site. When the extent of a soil type was limited in the sampling area, several soil types of similar properties were Combined to get five sites. The five sampling sites for a given soil type were ideally in five different counties, to achieve wide distribution. For some soil types, this was not possible, for they were concentrated in one part of the study area. Textures Sampled Four of the five textures sampled are Sandy. Sandy soils are important because of their susceptibility to wind erosion (Woodruff et al., 52 1977). In Michigan, most of the wind erosion problems occur on sandy soils (Drullenger and Schmidt, 1968). Fine sands were sampled instead of medium coarse sands. The water holding capacity of soil increases as texture becomes finer, so fine sands were more likely to be cultivated (Taylor and Ashcroft, 1972). This was especially true in the sampling area for fine sands - Macomb, St. Clair and Wayne counties. Higher value crops in those urbanizing counties make culti- vation of marginal soils profitable (Mokma, personal communication, 1979). Loamy fine sand, loamy sand and sandy loam were sampled for several reasons. The increasing contents of silt and clay from coarser to finer provided a range of values, so the effect of particle size distribution could be studied. These textures also comprise a large portion Of Michigan's erodible soils (Drullinger and Schmidt, 1968; Kimberlien et al., 1977). Their inclusion was necessary for a thorough overview of the erodibility of Michigan soil textures. Loams were sampled because they are widespread throughout lower Michigan where many areas of economically important crops are grown (Drullinger and Schmidt, 1968). The economics are especially important in the "Thumb" area, an important sugar beet region Of mainly loam soils. The abrasive action Of wind blown soil particles has damaged valuable crops in the "Thumb" area (Dush, 1966; Drullinger, 1968). The Agricultural Research Service assigned loams to WEG's based on differences in clay content and on the presence or absence of CaCO3 (Table 1). For this reason loams were subdivided to cover more WEG‘s. 53 Site Location and Soil Sampling Preliminary Site Location Sampling sites were located before field work could begin. A Soil Conservation Service computer printout was used to choose suitable soil types of wide distribution and large acreages. The printout listed Michigan soil types, gave the counties they were found in and the acreages in each county. Only those counties with a modern published soil survey were included in the printout. After the decision was made to sample a particular soil type in a county, the soil maps of the county soil survey report were scanned for possible sampling sites. A potential sampling site had to be large enough to be located at sampling time, in cultivation when the aerial photograph for the base map was taken, and near a road for easy access. The sites were chosen on the basis of aerial photographs taken some time before the study was initiated. Changes in agriculture and land use could make a potential site unsuitable for sampling. Therefore, more than one site for a particular soil type was located in a county. The soil types sampled, the classifica- tion of each soil series, and the location of each sampling site are given in Table 5. Morley loam and Miami loam were the two soil types sampled to assess the effects of clay contents >20 percent and <20 percent on wind erodibility. Both Miami and Morley have wide distributions in the sampling area, are cultivated, well drained and have a loam surface over clay loam subsoil (Soil Conservation Service, 1974, 1976). Morley parent material is finer than Miami parent material, presumably 4 5 ON.ou zoommcw>wu NNNZNN NN .umm xmzxzzxuzxmz mp .00 azmppo 3¢_szh .N .uam xzzxzmxzzx32 NF .ou ammacao NNNZNN N_ .umm xzmxmzxmzxmz . NF .00 mean: Nmmmmh N .uam xzzxzmxzmxzm cpazcacguo uP_Fou= op .00 awcoH 3Nmzoh mm .umm xzmxzmxmmem uwmae .uast .Nsaop-m=Nu macEapaz sacs Nucam nogpco_amz urc_< utmwac ; m_ .00 azmppo zepazoh Nm .amm xzmxzzxzzxwz .uaxwe .Nsmo_ cm>o Necmm mm=P2o=mz ¢_.ou capmmcp>ws mmmzap NN .umm xzmxzszmNZN cpmuapamz uwcmc< mp .00 mmmacmu NNNZNP MN .umm mexNzNNzx3m uvmms .umst Namo_-mmcmou ”mom: N_ .00 Nowuacu zNNzc_H eN .uam xzzxmzxmzxuz paacgooeozu P_ .00 wean: mammap ON .umm xwmszxzzxzm uw=c< ummms .umst .Nsmoo mouwcc_mm ucmm Name; pamgguogusm uw3c< o, .ou mamacmw Nomzmh a .Umm xzmxzzxzmxmz upmme waxwe .Nan_-mmcaou mocwz m .00 zacapgmmz NNNmNF eN .uam xNzxzzszxzz N .00 zmcmpgmaz NNNma» mm .umm xzzxzzxZWNNm N .8 2,33 “5%: : .uam £35333 22:3 2% o .00 mesa: ummmeh NN .umm meNZNNNm -awua uw=a< uwmme .umxwz zoccmp New; Name; eccpcopaa: m .ou tempo .pm Nmpmzmh eN .uam x3mxzzx3mx3m umocm acute» waxes .aucam sawmmsou a .00 neoumz NN_NZNN P .uam xzzxmmxzmxmm N .09 neouaz NN_szh F .uam xzmxzmxzmxmm N .00 mesa: Nwmmeh NF .umm xmmxmzxmmxmz p=a22mm p .00 mesa: mmmme» NN .umm mexNzxzszz -awua UPQNH uwmms .umxwz aFFP>¥mo team wave .02 ll :owumuog :owumqummmmpu mwNme mgzuxm» apwm .Pom auacczm amgazam mmmzma N .aam mmzmmzmmzmmz mamas .aaxas .NsaON-a=am NON .Eaam mN .OO azaaaO NNNmsz NP .aam mzzmzmmzmxzz aN .OO aaaaam mOmZNa Nm .aam xmmmmmmmmmzm NN .OO aazaaam mmmmma mN .aam mzmxmzmmmmmm NN .OO zaaaaaaaz mmmmmp NN .aam xzmmmmmmzmmz NNaOONOaI NaNa NN .OO aaaON NmNZNN NP .aam xmmxmmmzmmzm ONONN aaaas .ONONNNN .aaam NON .anm .02 cowpauom :oquUmemmapu wgsuxm» muwm muamcam vwzcwucou a m m4m0.84 mm in diameter was done with a variation of the rotary sieve first developed by Chepil and Bisal (1943). The original rotary sieve has been modified several times by Chepil (l952; l962), and once by Lyles et al., (l970). The rotary sieve was developed to replace the less accurate method of hand sieving (Chepil and Bisal, l943). The sieve had six concentric metal cylinders turned by an electric motor. Part of each cylinder had openings all of one size extending around the circum- ference of the cylinder. The Openings ranged from 38 mm to <0.42 mm resulting in separation of the dry fraction into seven size classes: >38mm; 38-12.7 mm; l2.7-6.4 mm; 6.4-2.0 mm; 2.0-0.83 mm; 0.83-0.42 mm and <.42 mm. When a soil sample was sieved, the first cylinder the soil contacted was the one with 38 mm holes (Chepil and Bisal, l943). Clods <38 mm fell through the openings to the next cylinder, while clods >38 mm slid down the cylinder to be collected in a pan. This continued until the sample was divided into the seven size classes. The cylinders had a 4 percent slope to facilitate the sliding action. Chepil and Bisal hoped to minimize structural breakdown by the gentle 4 percent slope and a slow rotation of 14 rpm. Provisions made to attach brushes to the finest sieves to prevent clogging proved unnecessary. Clogging of fine sieves was a disadvantage of hand sieving eliminated by the 59 rotary sieve. Elimination of the variable human factor inherent in hand sieving was another advantage of the rotary sieve. A modification of the rotary sieve increased the number of cylinders to l3 and reduced speed of rotation to seven rpm (Chepil, l952). A tapping device was added to dislodge dust caught on cracks and sieve Openings. The original rotary sieve required laborious hand feeding of the sample, so Chepil attached an automatic feeding device. It was a conveyer belt that fed soil into the sieve at a constant 60 in3/minute. Since the volume of soil flowing into and out of the sieve was constant, time of sieving and size of sample made no difference. In a later paper, Chepil (1962) reported on a modified rotary sieve, presented a detailed field sampling method and listed the advan- tages and disadvantages of the rotary sieve. The number of cylinders was reduced to five. It was recommended "reasonably dry" soil be sampled with a flat, square cornered spade. According to Chepil, sampling dry soil reduces aggregate breakdown. Method of sample drying had little effect on results. Oven drying at 70°C and air drying at room temperature produced no significant differences in sieving results. Advantages and disadvantages of the rotary sieve are (Chepil, 1962): . Advantages: l. Consistency Impartiality No sample size variability Less structural breakdown 01-wa Clogging of fine sieves eliminated 60 Disadvantages: l. Complexity of construction 2. Sieves not interchangeable. The rotary sieve was modified a third time when sieving errors were discovered (Lyles et al., 1970). A source of error in Chepil's rotary sieve was mesh length. This is the length of perforated area in each cylinder through which soil can fall. If the mesh length is too short, the soil particles do not have sufficient time on the mesh, and incomplete separation results. Lyles and his associates increased the mesh length of each cylinder of the rotary sieve. Changes in the feeding device and power transmission eliminated more sources of error. Although the rotary sieve represents the standard method of determining percent dry fractions >0.84 mm in diameter, it does have some disadvantages (Chepil, 1952; l962). It is a large, complex, heavy machine, so using it in field demonstrations of wind erodibility is not possible. The rotary sieve is not available commercially; each one must be individually constructed. A simpler, less expensive version was constructed for this study. The Alternate Sieve Description of Alternate Sieve The alternate sieve is a device for separating an air dry soil sample into two size classes. One class consists of dry soil fractions 30.84 mm in diameter. The other size class contains dry soil fractions m_w wh0.84 mm in diameter that may be encountered at a site. Sieving time was determined by investigating changes in the amounts of soil falling through the mesh vs. time. Data was graphed as cumulative percent 0.84 mm in diameter was also apparent. The aggregates were irregularly shaped at the start of sieving. As sieving progressed, the aggregates began to round and decrease in size. Only a very few rounded aggregates were retained on the sieve at the end of 10 minutes sieving. The l0 minute sieving time and 60 second interval were accordingly shortened to 240 seconds and 30 seconds respectively. Five 67 .Ampa>cmu:w muzcwe mcov FNOm Ocam NEOOF a mew>mwm mcmgzu cowpuagm as em.ov ca mmcago .m .mwm 552.3: 02.5.» No m2: (maaaaa ammnwno) nouavaa mwwo > 68 samples were sieved in this manner. The data was averaged and graphed (Fig. 4). The amount of soil falling through the sieve did not change appreciably after 90 seconds of sieving. Sieving time was then shortened to 90 seconds. The 30 second interval was still too long, so the time interval was set at 10 seconds. Five replications were sieved, using the 90 second sieving time and l0 second intervals. To eliminate the contribution of aggregate breakdown, a sample of the loamy sand bulk supply was passed through a flat 4 mm sieve. This removed many aggregates contributing to breakdown, for the stability of secondary aggregates is inversely related to their size (Chepil, 1952). The flat sieve was shaken as little as possible to lessen aggregate breakdown. Soil fractions <4.0 mm in diameter were then sieved for 90 seconds, using l0 second intervals. Three replications were done using this variation of the sieving technique. Subsamples of the loam and sandy loam bulk soils were then sieved. Three replications of each were sieved for 90 seconds, stopping the sieve every l0 seconds. Three more replications of each were passed through the flat 4 mm sieve, and the soil sieved for 90 seconds with l0 second intervals. The results for all three textures were averaged and graphed as cumulative percent 0.84 increased and aggregate breakdown decreased. On the basis of this graph, all soils were sieved for 40 seconds. The reasons were: 69 . .Ampa>gmucw Ocoumm omv Fwom Ocam NEOOF O mcw>mmm mcwgaw compost; as Ow.ov cw macacu .e .mwm A8233 025m.» “.0 52: SN 69. Our co co 1 q d a a u d (maaaaa amrnnwno) wouzwaa wwwo > L 02 70 .Oowauacw as Ox mcw>osmg paoguwz new saw: mFNOm anp uca anp Nucam .ucam Nsaop mcw>mam mcwgsu cowpoagw es em.o cw mmcagu .m .mwm 2.23va 02.33. mo NE: a . a, . a. a... a > .0 30:99.“. EEVA 30:33.. W 23.22“. as: .33-.. . ON m M .coqdorlilluWTIlIJOIli llmll zilli‘ . w” 911, i 1 I; . ov .0 .:09d nu >3 3 . m . 00 n 1: m An a. 301) )0) IIOII||||O| O . . . on m >509. a M g 71 l. The 90 second cumulative percentage for soil with aggregates >4.0 mm in diameter removed was assumed the correct value for percent dry fractions 4.0 mm removed, there was very little change in cumulative percent 4.0 mm in diameter present was only slightly greater than the 90 second percentage for soil without aggregates >4.0 mm in diameter. Final Sieving Procedure The final sieving procedure for erodibility determinations involved collecting the subsamples and transporting them to the lab as described previously. After the subsamples were air dried, residue was removed from the soil as suggested by Chepil (l962). Stones >3/4 inch in diameter were also removed. Stones of this size were taken out of the samples because they would have an abnormal crushing effect on the aggregates during sieving. A l000 9 sample was placed into the scoop, deposited into the cylinder, and the end opening closed. A tared piece of waxed paper placed under the cylinder caught the dry fractions 0.84 mm was: P = ——S—§—X—- (100) where: P = percent dry soil fractions >0.84 mm in diameter S = sample weight (g) X = weight dry fractions 0.84 mm and Actual Percent >0.84 mm The initial analysis of the sieving data was based on the assumption that the texture of the sample was the same as the texture indicated by the soil mapping unit. The >0.84 mm percentage was determined for the ten subsamples from each site (Table l, Appendix 2). A one way analysis of variance was used to determine if the sites for an assumed texture or subdivision of loam were significantly different in their percentage of dry fractions >0.84 mm in diameter. For an assumed texture and subdivision of loam, each of the five sites was considered a treatment. Every analysis of variance performed, one for each assumed texture and subdivision of loam, showed a significant difference between the treatment means. The results of Tukey's Multiple Comparison test are summarized in Table 6 (Steel and Torrie, 1960). The average >0.84 mm percentage for each assumed texture and subdivision of loam was then considered a treatment. An analysis of variance revealed significant differences in wind erodibility existed between textures and subdivisions of loam. The results of Tukey's 73 74 TABLE 6 ASSUMED TEXTURES-RESULTS OF TUKEY'S TEST Assumed >0.84 m (g) 1 Texture WEG Sieving, WEG Difference2 Site Mean(>0.84 mm)1 Fine Sand l ll l 10 l 8 AB* 2 4 A 3 ll BC 4 20 D 5 l4 C Loamy Sand 2 37 l0 27 ll 38 BC 12 28 AB l3 2l A 14 46 CD l5 53 D Loamy Fine Sand 2 30 lO 20 6 5 A 7 l2 A 8 4l B 9 49 C 10 44 BC Sandy Loam 3 77 25 52 l6 50 A l7 83 BC l8 89 C l9 87 C 20 76 D Loam, <20% Clay 5 78 40 38 26 89 . C 27 69 A 28 80 BC 29 79 BC 30 75 AB Loam, >20% Clay 6 82 45 37 2l 78 AB 22 83 BC 23 86 CD 24 9l 0 25 74 A Loam, non- 5 75 40 35 36 88 D calcareous 37 76 BC 38 52 A 39 73 B 40 84 CD Loam, calcareous 4L Bl 25 56 3l 78 B 32 95 D 33 67 A 34 87 C 35 77 B 9 Any two means of the same texture followed by the same letter are not significantly different from each other at P=.Ol by Tukey's test. ‘Mean of lo subsamples from the five sites. 2Difference between mean sieving percent >0.84 mm and WEG %. 75 test were: Assumed Texture or Subdivision Mean >0.84 mm (%) Fine Sand ll A* Loamy Fine Sand 30 B Loamy Sand 37 B Sandy Loam 77 C Loam, non-calcareous 75 C Loam, calcareous 8l C Loam, <20% clay 77 C Loam, >20% clay 82 C *any means followed by the same letter are not significantly different at P = .Ol using Tukey's test. Based on this analysis, the soil textural classes and sub- divisions of loam fell into three groups. Fine sand appeared the most erodible, with an average of ll percent of its dry fractions >0.84 mm in diameter. Loamy fine sand and loamy sand make up the next group. All the loams, including sandy loam, comprise the thirdgleast erodible group. However, particle size analysis of the soils revealed incon- sistencies between the mapping unit texture and the actual texture of the soil. Particle size data for each site is given in Table l of Appendix II. The five sampling sites of the soil assumed to be cal- careous loam were extremely variable in texture. Two sites were sandy clay loam, and there was one site each of silty clay, silty clay loam and clay loam. Textural analysis of the other soils revealed three sand sites and a total of three clay loam sites. It was felt a minimum of three sites were needed to provide a valid estimate of wind erodibility at the sampling time. Analysis of the erodibility of two additional textures, clay loam and sand, was then possible. A compari— son of assumed and actual textures, by site, are given in Table 7. The TABLE 7 ASSUMED AND ACTUAL TEXTURES BY SITE Site Assumed Actual l Fine Sand Sand 2 Fine Sand Fine Sand 3 Fine Sand Sand 4 Fine Sand Loamy Fine Sand 5 Fine Sand Sand 6 Loamy Fine Sand Fine Sand 7 Loamy Fine Sand Fine Sand 8 Loamy Fine Sand Loamy Fine Sand 9 Loamy Fine Sand Loamy Sand 10 Loamy Fine Sand Loamy Fine Sand 11 Loamy Sand Loamy Sand 12 Loamy Sand Loamy Sand 13 Loamy Sand Loamy Sand 14 Loamy Sand Sandy Loam 15 Loamy Sand Loamy Sand 16 Sandy Loam Sandy Loam 17 Sandy Loam Clay Loam 18 Sandy Loam Sandy Loam 19 Sandy Loam Loam 20 Sandy Loam . Sandy Loam 21 Loam (>20% clay) Loam (13% clay) 22 Loam (>20% clay) Loam (15% clay) 23 Loam (>20% clay) Loam (22% clay) 24 Loam (>20% clay) Loam (20% clay) 25 Loam (>20% clay) Loam (18% clay) 26 Loam (<20% clay) Loam (17% clay) 27 Loam (<20% clay) Loam (12% clay) 28 Loam (<20% clay) Loam (19% clay) 29 Loam (<20% clay) Loam (19% clay) 30 Loam (<20% clay) Loam (20% clay) 31 Loam, free CaCO3 Silty Clay Loam, free CaCO3 32 Loam, free CaC03 Clay Loam, free CaCO 33 Loam, free CaCO3 Silty Clay, free CaC 3 34 Loam, free CaCO3 Sandy Clay Loam, free CaCO 35 Loam, free CaCO3 Sandy Clay Loam, free CaCO3 36 Loam, no free CaCO3 Clay Loam, no free CaCO 37 Loam, no free CaCO3 Sandy Loam, no free CaC 38 Loam, no free CaC03 Sandy Loam, no free CaCO3 39 Loam, no free CaCO3 Sandy Loam, no free CaCO 40 Loam, no free CaCO3 Sandy Loam, no free CaC03 77 reorganization of sites by actual texture is given in Table 8. The textures included in Table 8 are those actually analyzed statistically for differences in erodibility. The two loam subdivisions of<:and >20 percent clay were analyzed together. Differences in clay between the two subdivisions were within error. For some textures, the sampling sites now encompassed more soil types 'than originally planned. It was imperative though to have the sampling sites of a given textural class actually of that texture. Texture is the basis for placing soils into WEG's. The sampling sites reorganized by actual texture were analyzed using the statistical procedure described previously. Table 9 summarizes the results of Tukey's test. For all textures, there were significant differences between two or more of the sites. This shows the effects differences in management and environment can have on erodibility. No attempt was made to control these factors in this study. In Kansas, the erodibility of soil sites also varied because of management and environmental differ- ences (Chepil, l953b). An analysis of variance performed on the textural Ineans of Table 9 revealed significant differences in wind erodibility toetween some textures. The results of Tukey's test are: Actual Texture Mean >0.84 mm (%) Sand 10 A* Fine Sand 7 A Loamy Sand 38 B Loamy Find Sand 35 B Sandy Loam 68 C Loam 81 0 Clay Loam 89 0 *Two means followed by the same letter are not significantly different from each other at P = .01 using Tukey's test. 78 TABLE 8 SITES REORGANIZED BY ACTUAL TEXTURE Texture Series Site No. Sand Oakville l Oakville 3 Oakville 5 Fine Sand Oakville 2 Tedrow 6 Tedrow 7 Loamy Fine Sand Oakville 4 Tedrow 8 Minoa 10 Loamy Sand Tedrow 9 Selfridge ll Selfridge 12 Metea 13 Menominee 15 Sandy Loam Metea l4 Metamora l6 Metamora 18 Parkhill 20 Parkhill 37 Parkhill 38 Parkhill 39 Parkhill 4O Loam Metamora l9 Morley 21 Morley 22 Morley 23 Morley 24 Morley 25 Miami 26 Miami 27 Miami 28 Miami 29 Guelph 30 Clay Loam Metamora l7 Tappan 32 Parkhill 36 79 TABLE 9 ACTUAL TEXTURES-RESULTS 0F TUKEY'S TEST >0.84 mm‘(%) % Texture WEG Sieving WEG Difference? Site Means (>0.84 mm)1 Sand 1 10 l 9 l 8 A* 3 11 AB 5 12 B Fine Sand 1 7 1 6 2 4 A 6 6 A 7 12 B Loamy Sand 2 38 10 28 9 50 CD 11 38 BC 12 28 AB 13 21 A 15 53 D Loamy Fine Sand 2 35 10 25 4 20 A 8 41 B 10 44 8 Sandy Loam 3 68 25 43 14 46 A 16 50 A 18 89 C 20 76 B 37 76 B 38 52 A 39 73 B 40 84 BC Loam, <20% 5 81 40 41 19 87 EF clay 21 78 BC 22 83 CDE 23 86 DEF 24 91 F 25 74 AB 26 86 DEF 27 69 A 28 80 BCD 29 79 BCD 30 75 AB Clay Loam, 6 89 45 44 32 95 C <35% clay 36 88 B 17 83 A *Any two means of the same texture followed by the same letter are not significantly different at P=.Ol using Tukey's test. 1Mean of 10 subsamples from the five sites 2 Difference between mean sieving percent >0.84 mm and WEG %. 80 The textures analyzed fall into four groups. Fine sand and sand make up the most erodible group, followed by loamy sand and loamy fine sand in the second group. The erodibility of sandy loam is significantly different from all the other textures. It is the sole member of the third group. Loam and clay loam make up the fourth, least erodible group. The textures fell into only three groups when erodibility was analyzed on the basis of the mapping unit. The removal of the clay loam and loam sites from the group of sandy loam sites caused the separation of sandy loam from loam and clay loam. Farm Advisors must remember this variability in texture when assisting cooperators with wind erosion problems. If the texture of a field is inconsis- tant with the mapping unit texture, the erodibility of the soil may be misjudged. I For the textures studied, the average >0.84 mm percentage was higher than the percentage of the WEG into which the texture was assigned (Table 9). The percentages were also higher than those found by Chepil (l953b; 1955a). Soils of irrigated Colorado sugar beet field and soils of the Southern Coastal Plains also had erodibilities lower than those reported by Chepil and lower than those of the WEG's (Carreker, 1966; Simmons and Dotzenke, 1974). The higher percentage of non-erodible clods in the Michigan soils studied is probably due to environmental and cultural differences between the Great Plains and Michigan. The higher average moisture contents of Michigan soils likely contribute to the decreased erodibility (Lyles and Woodruff, 1961; Soil Survey Staff, 1965). Cultivation forms a greater number of large clods as the moisture content of the soil at cultivation increases. 81 At time of sampling, sandy loam, loam and clay loam had on the average more than 66 percent of their dry fractions >0.84 mm in diameter. If 66 percent or more of the surface soil is composed of non-erodible fractions, it is considered resistant to wind erosion (Woodruff and Siddoway, 1973). The high percentages for sandy loam, loam and clay loam does not mean they will always be resistant to wind erosion. The wind erodibility of a soil can change from season to season or from day to day (Chepil, l953a; Bisal and Furguson, 1968). The soils in this study were sampled in a non-crusted, recently cultivated conditions. Soils were collected from May 8, 1979 to June l, 1979, so recent cultivation at Michigan's higher spring- time soil moisture contents may explain the extremely cloddy condition of the finer soils. The sandy soils sampled were also less erodible than the WEG's would indicate. The magnitude of the difference between the actual percent >0.84 mm and the WEG percent >0.84 mm decreased as soil texture became coarser than sandy loam. Sandy soils are not as susceptible to the aggregating effects of tillage as are finer textured soils (Woodruff et al., 1977). The lower the clay and silt content of a soil, the less its dry clod structure will be influenced by tillage. Sandy soils also have lower water holding capacities, so they are less likely to form large durable clods at tillage. These data indicate the present WEG classifications nay not adequately reflect the springtime erodibility of Michigan soils. 82 Effect of Soil Properties on Wind Erodibility Organic Matter and Calcium Carbonate The organic matter and calcium carbonate contents of the soils studied are given in Table 1, Appendix II. The affects of organic matter decomposition on water stable aggregates are well documented (Harris et al., 1966). Calcium carbonate and the decomposition of organic matter have also been shown to affect the dry aggregate structure of Great Plains soils (Chepil and Woodruff, 1963). Organic matter and calcium carbonate, by themselves, had no significant influence on wind erodibility of these Michigan soils (Figures 6 and 7). Over the range of organic matter and calcium carbonate contents studied, the effects of organic matter and CaCO3 may have been too small to detect by sieving. The influence of clay, silt and sand may have masked any effects organic matter and calcium carbonate had. Clay, Silt and Sand The percentages of clay, sand and silt were each plotted against percent dry fractions >0.84 mm in diamater (Figures 8, 9 and 10). Each point is the mean of a site. A second degree polynomial fit the data best. Clay Figure 8 illustrates the relationship between clay and the erodibility of a soil. As percent >0.84 mm increases, the erodibility of soil decreases. As clay increases from 3 percent (the smallest clay percentage measured) the erodibility of soil decreases. The 83 § I O T: O O Q E O O 9 so - O 0 :: (3(5) 3 ‘5 2 g so » < O as oo0 " O a ' Q 1- Linear. R2= .16 o 2-Quadratic. R2= .30 g 20 - 0 3— Cubic. R2= .35 2 <9 (23 g) 0 O 00 i z :1; 4 5 ORGANIC CARBON comem (PERCENT) Fig. 6. Relationship between organic carbon and >0.84 mm fraction. 84 10° - 80 OO 2 1: i! W ‘5’. .9: VI 2! Q 5 < E .. >- 8 40 - § . I- Linear. R2: .20 5 2' Quadratizc. R2=‘.20 g 20 _ 3- Cable. R = .22 § 0 1 1 O 1 2 INORGANIC CARBON CONTENT (PERCENT) Fig. 7. Relationship between inorganic carbon and >0.84 mm fraction. 100 ' 85 o A o g (fl 0 O Q U 80 a.“ % § 0 T: o o 2: g 60 - 0 E g 0 g 40 ' B Q g o g 20’ O Y2=-8.4 -I-7.4x-.14x2 Z ' c§> R = .86 00 IO 20 30 4o CLAY CONTENT (PERCENT) Fig. 8. Relationship between clay and >0.84 mm fraction. 30 NONERODIBLE DRY FRACTIONS(PERCENT) '3‘ 80‘ 60- 40' 20- Fig. 86 Y=46.0 + 2.1x - .0st R2: .91 ' 20 ' 4O 1 m so 100 SAND CONTENT (PERCENT) 9. Relationship between sand and >0.84 mm fraction 100 - ch 0 on O O O a: CD NONERODIBLE DRY FRACTIONS (PERCENT) 87 Y: -219 + 6.3x - .IOx2 R2 = .39 #4 0 Fig. 10. IO 20 so do so SILT CONTENT (PERCENT) Relationship between silt and >0.84 mm fraction. 88 erodibility of soil decreases up to a clay content Of about 27 per- cent. Above 27 percent clay, the susceptibility of soil to wind erosion begins to increase. In soils of Western Canada, Kansas and Nebraska this same relationship was seen (Chepil, l953a; 1955a: Chepil and Woodruff, 1963). The initial decrease in erodibility with increasing clay was found to be due to the cementing effects of Clay in clod formation. That conclusion is supported by the results of this study. The increase in erodibility above a certain clay content was attributed to the more pronounced effects of freezing and thawing and wetting and drying (Chepil, 1954b). Soils high in clay have greater water holding capacities and generally, a higher shrink-swell poten- tial (Taylor and Ashcroft, 1972). Upon wetting, swelling causes stresses leading to clod breakdown. When a soil freezes, the expansion of freezing water fractures Clods. Freezing effects are probably very important in Michigan. When freezing occurs, Michigan soils are usually moist. Sand Figure 9 illustrates the relationship of sand and wind erodibility. At very high sand contents, eg. 90 percent, the soil is extremely erodible. As sand content decreases, so does the susceptibility of soil to wind erosion. Soil appears least erodible at about 38 percent sand. As the percentage of sand decreases below 38 percent, the soil erodibility again increases. The increase in erodibility seen with sand contents above 38 percent is due to the non-cohesive nature of sand grains (Chepil and Woodruff, 1963). Sand consists of non-collodial grains of quartz 89 and other inert minerals. Grains of sand do not have the cementing effects important in secondary aggregate formation (Chepil, l953a). The decrease in erodibility as sand contents fall to about 38 percent is probably due to the cementing effects of the greater percentages of silt and clay in the soil. These same relationships seen in Great Plains soils were also attributed to the cementing effects of clay and silt (Chepil, 1955a, b). The importance of silt and clay is well illustrated in the steep slope of the curve at very high sand contents. A small increase in silt or clay apparently causes a large increase in dry fractions >0.84 mm in diameter. This was demonstrated in Kansas by Chepil (1955a). The first increments of clay or silt added to sand were very effective in reducing the erodibility of the sands. As sand continues to decrease below 38 percent, the erodibility of soil again rises. This is probably due to the finer texture (higher clay and silt contents) of the soils with these low sand contents. As shown by Figure 8 the erodibility of soil increases as the percentage of clay rises above about 27 percent. Silt The relationship of silt and dry clod structure is shown in Figure 10. Wind erodibility decreases as silt increases up to about 35 percent. Erodibility then increases as silt increases above 35 percent. The initial decrease in erodibility associated with increasing silt contents may be caused by the cementing effects of silt. The importance of silt as a cementing agent in dry clod formation has been shown for soils of the Great Plains (Chepil, l953a). 90 Two factors may be contributing to the increase in erodibility above 35 percent silt. Of the soils sampled, most of those with the high silt contents also had clay percentages above 30. The decrease in dry aggregation above 35 percent silt may also be caused by the nature of aggregates formed when silt is the primary cementing agent. Silt forms softer aggregates when it is the primary cementing agent (Chepil, 1955a). Mechanical breakdown of the softer secondary aggregates during sieving may have contributed to the lower percentages. From these relationships it can be concluded sand has an overall negative effect on dry aggregation, while silt has an overall positive effect. Clay appears to influence wind erodibility in either a positive or negative manner, depending on the amount of clay present. The derivative of each equation was taken to arrive at the value of X (percentcflay, sand or silt) resulting in maximum Y (percent >0.84 mm). Maximum resistance to wind erosion occurred at 27 percent Clay, 35 percent silt and 38 percent sand. On the basis of Figures 8, 9 and 10 soil with maximum resistance to wind erosion has a clay content ranging from 24-30 percent, a silt content from 30-40 percent and a sand content from 31-45 percent. The mean percentages of each size fraction are: 27, 35 and 38 percent of clay, silt and sand respectively. The summation percentage equals 100. These optimum clay, silt and sand percentages are very Close to the two different sets of percentages estimated by Chepil (1940, as cited by Chepil, 1955a; l953b; 1955a). In 1940, Chepil studied the wind erodibility of some soils of Western Canada. The greatest degree of cloddiness occurred in soils‘ having about 20, 38 and 42 percent clay, silt and sand, respectively. These percentages are extremely 91 close to the percentages of 27, 35 and 38 estimated from Figures 8, 9 and 10. In 1955, Chepil found Nebraskan and Kansan soils gave Optimum resistance to wind erosion when they contained 20-30 percent 'clay, 40-50 percent silt and 20-50 percent sand. This agrees well with the clay, silt and sand ranges of 24-30 percent, 30-40 percent and 31-45 percent estimated from Figures 8, 9 and 10. This similarity demonstrates the validityof'the alternate sieve as a method of determining percent dry fractions >0.84 mm in diameter. Chepil (l955a)felt differences in parent material may be a factor in the discrepancy between the optimum silt contents for the Canadian and Nebraskan/Kansan soils. The Nebraskan and Kansan soils were loessial in origin, while most of the Canadian soils had glacial till parent material. The very Close Optimum percentages of the Canadian study and of this study support Chepil's contention. The majority of the soils in this Michigan study had glacial till parent material. SUMMARY AND CONCLUSIONS The wind erodibility of some Michigan soils was studied. Forty sites representing 13 soil series and seven surface textures were sampled. The soils were analyzed for their organic matter and calcium carbonate contents, and for their percentages of sand, silt and clay. An alternate version of the standard rotary sieve was developed. Percent non-erodible dry fractions >0.84 mm for each soil was determined using the alternate sieve. The variability in erodibility from site to site within each texture was analyzed. The average >0.84 mm percent for each texture was then compared to the >0.84 mm percent of the WEG to which each texture was assigned. Analysis of the effects of clay, silt and sand on wind erodibility revealed the particle size distribution resulting in maximum resistance to wind erosion. The following conclusions were made: 1. There were significant differences in the wind erodibility between sites of the same texture. Variations in environment and management from site to site were most likely responsible for the observed differences. 2. For every texture studied, the average >0.84 mm percentage, as determined by alternate sieving, was higher than the percentage for the WEG to which the texture was assigned. Differences in 92 93 soil moisture at time of tillage may have contributed to a large portion of the discrepancy between actUal and assigned textures. The magnitude of the difference between the actual and assigned textures decreased as soil texture became coarser. Coarser textured soils have lower water holding capacities and are generally drier when tilled. Based on these data, the present WEG assignments may not adequately reflect the springtime erodibility of Michigan's soils. Organic matter and calcium carbonate had no significant effect on wind erodibility over the ranges studied. For Michigan conditions it is possible the effects of organic matter and calcium carbonate may have been overshadowed by the influence of clay, silt and sand. There were significant curvilinear relationships between clay and wind erodibility, between silt and wind erodibility and between sand and wind erodibility. Second degree polynomial equations fit the data best. Sand had an overall negative influence on the non-erodible dry fractions, while silt had an overall positive influence. The effect of I clay on the non-erodible dry fractions was variable, depending on the clay content of the soil. An increase in clay up to about 27 percent decreased 94 erodibility of soil. Above 27 percent clay, wind erodibility increased as Clay content increased. These same clay, silt and sand relationships were described previously for soils of Western Canada and the Great Plains. A medium textured soil containing 24-30 percent clay, 30-40 percent silt and 31-45 percent sand was found most resistant to wind erosion. The percentages agree very well with the optimum percentages for soils of Western Canada and the Great Plains. These findings show for the first time that clay, silt and sand have the same effect on the wind erodibility of Michigan soils as they do on Western Canadian and Great Plains soils. This indicates the basic effects of the primary particles on dry clod structure are the same between areas with widely different environments and cultural practices. There is a commonality of conclusions between the Great Plains studies, where standard rotary sieving was used, and this Michigan study, where alternate sieving was used. This demonstrates the validity of the alternate sieve as a method of determining the wind erodibility of soils. APPENDIX I 95 TABLE 1 CALIBRATION RESULTS Empty 1000 9 Glass Beads Setting 45pm Setting rpm 0 O 0 O 1 O 1 O 2 O 2 O 3 l 3 l 4 2 4 1-2 5 4 5 4 6 5 6 5 7 6 7 6 8 6 8 6 9 6-7 9 6-7 10 7 10 7 TABLE 2 SIEVING RESULTS-21 TEST SAMPLES >0.84 mm Sample No. >0.84 mm Sample No. 1%) (cont) 1%) (cont) 1 16 12 18 2 13 13 9 3 20 14 10 4 16 15 20 5 20 16 19 6 18 17 17 7 18 18 14 8 16 19 18 9 17 20 19 10 14 21 17 11 13 x=16.25 s=3.06 APPENDIX II TABLE 1 PARTICLE SIZE DISTRIBUTION AND ORGANIC AND INORGANIC CARBON CONTENTS OF SOILS VF Texture* 0C M Sand Fraction(%)* Silt Sand Clay W (95) (74) (74) [\meNO 000000 96 LnLnLoanoo 000000 (74) >0.84 mm Site a Q) . C -I- q. >5 mmmmmm WWWUIU'IUI $- ‘I—‘I—‘I—‘I—‘I—‘I— g II A LL NCDNO‘I—Q‘ Nair-NV” > o o o o o o o o o o I o V Q'Q'Q'mmd’ wmmmmv “- > IDPOQ‘GDN O¢NNNO Q; QNLOMQ'V ONOOF-N C mmmmmm \DLOKD‘ONDLD a: mmmmmoo QNQ’O‘Q’N II mLOLDLDNr- mmoommo A NNNNF-N r—t—l—r—F-N Ll— v I... LOCOQ'NDO Ml—t—mNO‘ COMLDON owl-COLD“) +4 NNNNMN o- 0!- U" O‘COMOQ’ wound-mm ll OF'l—I—l—l— OOOOOO :53 'l- W“!- 'U “(D >55 oowwNo NOO‘OMN CU mmoooomm mmoommoo T)" 2 II Q 0:: IUD CU F u i’ mmmtxmm mmmmv—Ix Q’ I—> 0%" 'U :0 as: com VII- II " moowmoocnmmmvx '1 1| mmmmmmmwm II II .0 a '— '— lxw Ixm .3 3,. Cd 0 U C II r—NMQ‘WSDNwO‘I— FNMQ‘LO‘DNQO‘ II I I I I I I I I I I I I I I I I I O r- N U - Continued TABLE 1 Sand Fraction(%) Co Texture 0C VF >0.84 mm Clay Silt Sand VCO Site W W) L741 (76) (74) ONO‘NPNFE— OOl—F-l—l— mmmmmm wSOl-Owhw mmmmmm Nt—Nr—O‘bomr—QO O r—NMQ‘LDLDNQOF ('3' I I I I I I I I 97 moor—moo P LONNO‘ONNMMQ r—t—NF-NF-NNNE— O ,- I 4-1 -2 -3 -4 -s -6 -7 -8 -9 TABLE 1 - Continued Sand Fraction(%) 98 0;; MQVQ'KD camoosooun 0v r—r-r-I—r— 000000 0) L m mmmmmm 3 mmr- mm I+-"II-"II-"I-‘In'4- X G) .— LL QQ’SDMO‘ Nf—Q'O‘O‘Lo > NNNMN VVQ’Q‘Q‘Q’ NVwom GDP-Nmmw LL emmwm 0000000001 mmmmm Q'Q'Q'LDLOQ' t—OOG’Q’ O‘CONwN Z Or—INO'I q- Pm,— "MNNN NNNNNN Nmfi'mfi‘ SOQ'NQ'OM 8 00‘1me (BUILDOWON VCo .0 7 6 6 8 UA CB‘Q ONNQO‘ NCDOWQ'O‘LD 8" 0503000000 moooomoom “A r—BQ w-‘J [\lnNNLO mummom m l— >”'\ (BBQ ”MIDI-OLD mmmmmo PM U E Om '- ON V’N ”3'2 NMQ’O‘NNr—OQ'N II II LOKDLOQ'LDQ'LDCDQ‘O‘ II II 0v r-l—r— F-F—I—F-I—m o Ixm Ixm A C) O O +3 r—NMQ'LOSONQGF- r—NMQ'I-DSDNwOWP 'F I I I I I I I I I I I I I I I I I I I I U) LO LO - Continued TABLE 1 Sand Fraction (%) CO Texture OC VF Silt Sand VCO Clay >0.84 mm Site (5’6) W (74) (76) Q'F-MNLOND l-‘F'F'F—F'f— fs fs fs fs fs fs \DNNMO‘O @Q’Q’VQ’LO PQOMNW C‘r—omOII-I- mmommxo SOONQ'OO CNNmLDKD C‘QQMF—N Q'Q’d'éom Q'GDLDKDO‘N mmmmmcr mmNNNCMNLDw '— F—F'Nl- 0"!— O FNMQ’LDAONQOF 'LI I I I I I I I I 99 ”054006? NNNNNN 0.84 mm Clay Silt Sand VCO Site (76) (5’6) (74) (76) cap-h~a><- uausuaua0.84 mm Clay Silt Sand Site (74) (76) (76) (74) (76) Par—£00103 mwooo-zro MMMQ’VQ mmmsou—O ‘OLOQ’F-an NNNNNN Nmmwxooo QFONO‘F r-r—t— I‘— [\ww‘OI‘m 000000 75 76 82 82 80 85 13 13 11 8 17 8 13 11 8 10 3 8 63 51 43 29 16 36 -2 -3 -4 -5 -6 11-1 51 38 29 21 -7 -8 -9 -10 101 QVNNO‘ r—NNNN r—NNQ'I— NMNNN co'osme-o Qr-O‘md' VLDQ'Q'Q' \OO‘MO‘Q’ VO‘Nr—w Nf—NNF- O‘Nr—QN corxtxoom mommm OOOOO 82 83 82 82 84 10 10 13 10 8 26 29 34 28 32 21 -2 -3 -4 -5 -5 12-1 25 27 -7 -8 -9 102 OF M m 0e u.m mm oPI N0 mI mm wI P0 NI mm 0I Fm w.PF 0.0m m.w 0.0 o.~ on ON 0— vv mI m— o.pp N.mm 0.0F 0.0 0.0 0N mm P um «I Fm o.pp P.0m F.NP ¢.0 N.o 00 mm 5 mm mu Pm N.NF m.mm 0.m 0.¢ 0.0 mm Fm N oe NI Fm m.m~ m.mm w.m 0.0 m.o Om MN 0 me FIQP m H m —N u.m mm CPI KN mu mp wI pm NI pm 0: mp N.¢ m.~¢ N.¢N N.FF N.F 00 OP 0 mm mI m ¢.¢ m.N¢ m.m~ ¢.NF N.p um OF M «N eI mp N.0 F.N¢ n.0N m.FP u.o N0 mp 0 mp mu m ~.¢ w.~¢ N.NN w.pp ~.p mm 0 0 VP NI m— ¢.¢ F.Nv m.v~ 0.—F N.P mm or 0 MN PIMP :3 $3 :3 g weapxw» m> m z oo 00> ucmm upmm Nope 55 em.ox muwm mxvcowuuaem scam umzcwucou I F u4m0.84 mm Clay Silt Sand Site (%1 L%) W») W W») OOIDNO NNr—NN ls NNLDLOQ’ Qd'd'd’d' oooooooxe Ixmm-am m 2 OO OO> Oeam NNNm NaNO es Om.Ox aNNm vacowuuagu Ocam Omacwucou I N mqm N 2 cu 00> namm pNNm NmNu as em.oA mpN vacoNpumgu team umzcwwcou n F u4m0.84 mm Clay Silt Site (74) (7%) (93) W W MMNON F-P—F-r-F- mmoooo MMQMV KNOCK NMLOQ’O f—l—l-‘l-‘l— O‘C‘OMN NNszO momma) cmquSm LDKDOLDLD OOOOO [\momm MMVQ'V NVLDOO QQ’QQ’Q‘ WWI-OLD” NNv—r—l— movammomco— \DKDNNNNNNNQ O FNMQ'mDNme I I I I I I I I I I LO N 108 monomxo F-l-‘F'NF-l— OMOOOQ: \DLfi‘DOQ‘D C>c>c300c30.84 mm Site (74) W (74) 11 (M wNwNNO F-c—r-r—F-N ONTMPQ mmmdmm OOONNO oiomoioo F-Nf—F-NN NNQ’ONM dooocioio F-‘ooomo I\.I\.I\.I\.I\.I\ ONr-‘fl'd'm NI—I—l—f—l— 48 50 40 48 50 47 41 41 43 41 40 40 9 17 12 11 13 72 74 64 72 73 62 27-1 -2 -3 -4 -5 -6 109 cow-emc- OI—tomr- kOLDLDGSD «ammo.— onxcoNo LOLOQ’IOIT WDNON OOOLDF- ”N600 34 34 39 37 36 17 22 17 25 16 80 -10 - Continued TABLE 1 Sand Fraction(%) Texture 0C VF Silt Sand VCo 4n 4%) CTay >0.84 mm Site (74) (24) W 00050203 PI—‘OOO col—comm Q'LOQ'OLO Dowel-0&0 ”NMNN Nocomm NNr—r-N roommmd-Lnoco mmeI‘NNNNO O FNMQ’LO‘ONCDO‘I— I I I I I I I I I I 03 N 110 LOP-Q'Q'N NNNNN OP F'f—F' WI— moommm "IMO“;NQ" flomeV MQMI—N l—l-‘l—‘l—l— OLDMLDN [\oomd'oo @Nwlfim AQMMQ' LOLOLDLDO 000°C 29 33 30 28 39 50 48 49 so 44 19 21 22 17 21 84 77 84 80 67 30-1 -2 -3 -4 -5 80 86 84 78 78 -6 -7 -8 -9 ~10 111 mm o_- no a. em m- mm B. mm o- _u m.m o.e_ P.o_ m.op P._ _4 om mm “a m- _u o.m m.m_ m.o. e.m o._ mm Pm _m cm e- _u ~.m m.m. m.op m.m m.o on _m cm mm m- _u o.m m.mp ~.¢ m.~ m.o mm Fm mm em N- F0 _.m _e.m_ o.PF _.N m.o mm em em om F-~m m u m we 41: me op- om m- mm m. mm N- e“ m- we m- Poem ¢.e e.“ o._ ~.P N.o m_ he mm w“ e- _uwm m.¢ m.“ O.N ¢.P m.o op ee mm mu m- Fuwm N.¢ o.m m.o m.o F.o ¢_ Ne mm mm N- Poem ~.m F.“ m._ w._ ~.o N_ we Km om F-Fm idxv lldwv ARV ANV meaaxmh d> d 2 cu 00> ucmm “Fem »m_u as em.oA oppm vacowuumgm ucmm amazeucoo - P m4m<fi mwcoooo f—F-l—f—N 112 m "m “mum mm op: ow an em mu mm N: am on —um m.m “.mp m.m~ m.n 0.0 Fm em mm mm m: pumnp m.m ¢.mp O.NP m.m «.0 we Nm ON pm cu Pom o.¢ ¢.wp N.FP o.w N.o we am RN 5” mu pom o.¢ N.mp m.pp m.m “.0 we mm mm mm N: P0 m.m .e.m~ m.FF m.“ m.o m¢ RN mm mm anm m "m mo u.m we op: so m: mm w: No N: mm m: on mu omm m.m ¢.¢ 5.0 0.9 F.o m me m¢ en #1 0mm m.m m.¢ m.o m.o ~.o NP we ow mm M: uwm o.m ¢.m m.o 0.0 N.o NF me me #0 Nu uwm m.m F.¢ 0.0 n.o N.o op me me Nu Flmm Aav New; ARV ARV wezaxmp u> m 2 cu 00> ucmm “Fem xopu as em.OA muwm Aflvcowpuwed teem uwzcwucou n p mgm ucmm “Fem xmpo as em.OA mumm Navcowuumgu ucmm umacwpcou - F m4m d 2 cu 08> team “Fem ampu as ew.OA muwm Nxvcowpomee scam umscwucou n F u4m OONO ONNO NONO as ONOA ONNO ANOOONNOOLN OOOO OOOONOOOO - N NOOON BIBLIOGRAPHY Allison, L. E. 1965. Organic carbon. ln_C. A. Black (ed.) Methods of soil analysis. Part II. Agronomy: 9:1367-1378. Anderson, D. T., and Committee. 1966. Soil erosion by wind, cause, damage control. Pub. 1266. Canada Dept. Agr., Ottawa, Ont. 22 pp. Anderson, C. H., and A. Wenhardt. 1966. Soil erodibility, fall and spring. Can. J. Soil Sci. 46:255-260. Armbrust, D. V. 1968. Nindblown abrasive injury to cotton plants. Agron. J. 60:622-625. Bisal, F., and N. S. Furguson. 1968. Monthly and yearly changes in aggregate size of surface clods. Can. J. Soil Sci. 48:159-164. Bisal, F., and K. F. Nielsen. 1962. Movement of soil particles in saltation. Can. J. Soil Sci. 42:81-86. Bisal, F., and K. F. Nielsen. 1964. Soil aggregates do not necessarily break down over winter. Soil Sci. 98:345. Bisal, F., and K. F. Nielsen. 1967. Effect of frost action on the size of soil aggregates. Soil Sci. 104:268-272. Bundy, L. G. and J. M. Bremner. l972. A simple titrimetric method for determination of inorganic carbon in soils. Soil Sci. Soc. Am. Proc. 36:273-275. Call, L. E. 1936. Cultural methods of controlling wind erosion. Amer. Soc. Agron. Jour. 28:193-201. Carreker, J. R. 1966. Wind erosion in the Southeast. J. Soil Water Conserv. 21:86-88. Chepil, u. S. 1940. The relation of wind erosion to some physical soil characteristics. Unpublished Ph.D. thesis. University of Minneapolis, Minnesota. Chepil, N. S. 1941. Relation of wind erosion to the dry aggregate structure of a soil. Sci. Agric. 21:488-507. Chepil, N. S. 1944. Utilization of crOp residues for wind erosion control. Sci. Agric. 24:307-319. 116 Chepil, Chep11, Chepil, Chepil, Chepil, Chepil, Chepil, Chepil, Chepil, Chepil, Chepil, Chepi1, Chepil, Chepil, Chepil, 117 N. S. 1945a. Dynamics of wind erosion: I. The nature of movement of soil by wind. Soil Sci. 60:305-320. w. S. 1945b. Dynamics of wind erosion: II. Initiation of soil movement. Soil Sci. 60:397-411. u. S. 1945c. Dynamics of wind erosion: III. Transport capacity of the wind. Soil Sci. 60:475-480. N. S. 1946a. Dynamics of wind erosion: IV. The translocating and abrasive action of the wind. Soil Sci. 61:167-177. N. S. 1946b. Dynamics of wind erosion: V. Cumulative inten- sity of soil drifting across eroding fields. Soil Sci. 61: 257-263. N. S. 1946c. Dynamics of wind erosion: VI. Sorting of: soil material by the wind. Soil Sci. 61:331-340. w. 5. l950a. Properties of soil which influence wind erosion: I. The governing principle of surface roughness. Soil Sci. 69:149-162. u. S. 1950b. Properties of soil which influence wind erosion: II. Dry aggregate structure as an index of erodibility: Soil Sci. 59: 403- 414. u. S. 1951. Properties of soil which influence wind erosion: V. Mechanical stability of structure. Soil Sci. 72:465-478. N. S. 1952. Improved rotary sieve for measuring state and stability of dry soil structure. Soil Sci. Soc. Am. Proc. 16: 113-117. u. S. 1953a. Field structure of cultivated soils with special reference to erodibility by wind. Soil Sci. Soc. Am. Proc. 17:185-191. N. S. 1953b. Factors that influence clod structure and erodibility of soil by wind: 1. Soil texture. Soil Sci. 75:473-483. u. S. 1953c. Factors that influence clod structure and erodibility of soil by wind: 11. Water stable structure. Soil Sci. 76:389-399. w. S. 1954a. Factors that influence clod structure and erodibility of soil by wind: III. Calcium carbonate and decomposed organic matter. Soil Sci. Soc. Am. Proc. 77:473-480. u. S. 1954b. Seasonal fluctuations in soil structure and erodibility of soil by wind: IV. Soil Sci. Amer. Proc. 18: 13-16. 118 Chepil, H. S. 1955a. Factors that influence clod structure and erodibility by wind: IV. Sand, silt, and clay. Soil Sci. 80:155-162. Chepil, N. S. 1955b. Factors that influence clod structure and erodibility of soil by wind: V. Organic matter at various stages of decomposition. Soil Sci. 80:413-420. Chepil, M. S. 1956. Influence of moisture on erodibility of soil by wind. Soil Sci. Soc. Am. Proc. 20:288-292. Chepil, w. S. 1957. Width of trap strips to control wind erosion. Tech. Bull. 92, Kansas State College of Agric. and Appl. Sci. Chepil, u. S. 1958. Estimations of wind erodibility of farm fields. USDA, ARS Prod. Res. Rpt. No. 25. Chepil, N. S. 1959a. Equilibrium of soil grains at the threshold of movement by wind. Soil Sci. Soc. Am. Proc. 23:422-428. Chepil, u. S. 1959b. Wind erodibility of farm fields. J. Soil Water Conserv. 14:214-219. ,-// Chepil, N. S. 1960. Conversion of relative field erodibility to - annual soil loss by wind. Soil Sci. Soc. Am. Proc. 24:143-145. Chepil, N. S. 1961. The use of spheres to measure lift and drag on wind eroded soil grains. Soil Sci. Soc. Am. Proc. 25:343-345. Chepil, u. S. 1962. A compact rotary sieve and the importance of dry sieving in physical soil analysis. Soil Sci. Soc. Am. Proc. 26:4-6. Chepil, N. S., and F. Bisal. 1943. A rotary sieve method for determining the size distribution of soil clods. Soil Sci. 56:95-100. Chepil, N. S., and R. A. Milne. 1939. Comparative study of soil drifting in the field and in a wind tunnel. Sci. Agric. 19: 249-257. Chepil, N. S., and R. A. Milne. 1941a. Wind erosion of soil in relation to roughness of surface. Soil Sci. 52:417-431. Chepil, N. S., and R. A. Milne. 1941b. Wind erosion of soils in relation to size and nature of the exposed area. Sci. Agric. 21:479-487. Chepil, N. S., F. H. Siddoway, and D. V. Armbrust. 1962. Climatic factor for estimating wind erodibility of farm fields. J. Soil Water Conserv. 17:162-165. 119 Chepil, u. S., F. H. Siddoway, and D. V. Armbrust. 1963. Climatic index of wind erosion conditions in the Great Plains. Soil Sci. Soc. Am. Proc. 27:449-452. Chepil, w. 5., F. H. Siddoway, and D. V. Armbrust. 1964. Wind erodibility of knolly terrain. J. Soil Water Conserv. 19: 179-181. Chepil, N. S., and N. P. Woodruff. 1954. Estimations of wind erodibility of farm fields. J. Soil Water Conserv. 9:257-266. Chepil, N. S., and N. P. Woodruff. 1957. Sedimentary characteristics of dust storms: Visibility and dust concentration. Am. J. Sci. 255:104-114. Chepil, u. S., and N. P. Woodruff. 1959. Estimations of wind erodibility of farm fields. Prod. Res. Rept. No. 25, USDA. Chepil, u. S., and N. P. Woodruff. 1963. The physics of wind erosion and its control. Adv. Agron. 15:211-302. Daniel, H. A. 1936. The physical changes in soils of the Southern High Plains due to cropping and wind erosion and the relation between the sang];y51lt ratios in these soils. J. Am. Soc. Agron. 28:570-580. Daniel, H. A., and N. H. Langham.' 1936. The effect of wind erosion and cultivation on the total nitrogen and organic matter con- tents of soils in the Southern High Plains. Jour. Am. Soc. Agron. 28:587-596. Day, P. R. 1965. Particle fractionation and particle-size analysis. In_C. A. Black (ed.) Methods of soil analysis, Part I, Agronomy: 9:545-567. Doughty, J. L., and Staff. 1943. Report of investigations. Soil Research Laboratory, Canada Department of Agriculture, Swift Current, Sask. Drullinger, R. H., and B. L. Schmidt. 1968. Wind erosion problems and controls in the Great Lakes Region. J. Soil Water Conserv. 23:58-59. Dush, R. 1966. The wind that blew no good. Sugar Beet J. 29:4-5. Fenster, C. R. 1975. Erosion control tillage: method and application. Proc. Soil Cons. Soc. Am. 30th:l35-l39. Fenster, C. R., and G. A. Hicks. 1977. Minimum tillage fallow systems for reducing wind erosion. Trans. ASAE 20:906-910. 120 Free, E. E. 1911. The movement of soil material by wind, with a Bibliography of Eolian Geology by S. C. Stuntz and E. E. Free. USDA Bur. Soils Bull. 68. 272 pp. Fryrear, D. H., and J. D. Downes. 1975. Consider the plant in planning wind erosion control systems. Trans. ASAE 18: 1070-1072. Gillette, D. A. 1977. Fine particulate emissions due to wind erosion. Trans. ASAE 20:890-897. Gillette, D. A. 1978a. A wind tunnel simulation of the erosion of soil; effect of soil texture, sandblasting, wind speed and soil consolidation on dust production. Atm. Env. 12:1735-1743. Gillette, D. A. 1978b. Tests with a portable wind tunnel for deter- mining wind erosion threshold velocities. Atm. Env. 12:2309- 2313. Gillette, D. w., and T. R. Walker. 1977. Characteristics of airborne particles produced by wind erosion of sandy soil, high plains of West Texas, Soil Sci. 123:97-110. Hagen, L. J., and E. L. Skidmore. 1977. Wind erosion and visibility problems. Trans. ASAE 20:898-903. Harris, R. F., G. Chesters, and O. N. Allen. 1966. Dynamics of soil aggregation. Adv. Agron. 18:107-169. Hayes, N. A. 1972. Designing wind erosion control systems in the midwest region. RTSC Tech. Note LI-9, 1972. Hsieh, J. J. C., and R. E. Wildung. 1969. Bentonite stabilization of soil to resist wind erosion. Soil Sci. Soc. Am. Proc. 33: 637-638. Kimberlin, L. H., A. L. Hidelbaugh, and A. R. Grunewald. 1977. The potential wind erosion problem in the United States. Trans. ASAE. 20:873-879. King, F. H. 1894. Destructive effects of winds on sandy soils. Univ. Wisc. Bull. 42. Lyles, L. 1975. Possible effects of wind erosion on soil productivity. J. Soil Water Conserv. 30:279-283. Lyles, L. 1977. Wind erosion: processes and effect on soil productivity. Trans. ASAE 20:880-884. Lyles, L., L. A. Disrud, and N. P. Woodruff. 1969. Effects of soil physical properties, rainfall characteristics and wind velocity on clod disintegration by simulated rainfall. Soil Sci. Soc. Am. Proc. 33:302-306. 121 Lyles, L., J. D. Dickerson, and L. A. Disrud. 1970. Modified rotary sieve for improved accuracy. Soil Sci. 109:207-210. Lyles, L., and R. K. Krauss. 1971. Threshold velocities and initial particle motion as influenced by air turbulence. Trans. ASAE 14:563-566. Lyles, L., R. L. Schrandt, and N. F. Schmeidler. 1974. How aero- dynamic roughness elements control sand movement. Trans. ASAE 17:134-139. Lyles, L., and N. P. Woodruff. 1961. Surface soil cloddiness in relation to soil moisture at time of tillage. Soil Sci. 91: 178-182. Lyles, L., and N. P. Woodruff. 1962. How moisture and tillage affect soil cloddiness for wind erosion control. Agric. Eng. 43: 150-153, 159. Malina, F. J. 1941. Recent developments in the dynamics of wind erosion. Trans. Am. Geophys. Union. 1941:262-284. Michigan Department of Agriculture, 1980. Michigan Agricultural Statistics, Mi Agr. Rept. Serv., Lansing, MI. Plate, E. J. 1971. The aerodynamics of shelter belts.. Agric. Meterol. 8:203-222. Quisenberry, D. 1978. Michigan Technical Guide, Section III, Wind-1. USDA-SCS. Rosenberg, N. J. 1974. Microclimate: The biologiCal environment. John Wiley and Sons, N.Y. Schmidt, B. L., and G. B. Triplett, Jr. 1967. Controlling wind erosion. Ohio Rept. on Res. and Dev. 52:35-37. Siddoway, F. H., W. S. Chepil, and D. V. Armbrust. 1965. Effect of kind, amount and placement of residues on wind erosion con- trol. Trans. ASAE 8:327-331. Simmons, S. R. and A. D. Dotzenko. 1974. Proposed indices for estima- ting the inherent“wind erodibility of soils. J. Soil Water Conserv. 29:275-276. Skidmore, E. L. 1966. Wind and sandblast injury to seedling green beans. Agron. J. 58:311-315. Skidmore, E. L., P. S. Fisher, and N. P. Woodruff. 1970. Wind erosion equation: computer solution and application. Soil Sci. Soc. Am. Proc. 34:931-935. 122 Skidmore, E. L., and L. J. Hagen. 1977. Reducing wind erosion with barriers. Trans. ASAE 20:911-915. Skidmore, E. L., and F. H. Siddoway. 1978. Crop residue requirements to control wind erosion. Ig_Crop Residue Management Systems, Am. Soc. Agron., Madison, Wisc. 17-33. Skidmore, E. L., and N. P. Woodruff. 1968. Wind erosion forces in the United States and their use in predicting soil loss. Agr. Handbook 346. ARS, USDA. Smalley, I. J. 1970. Cohesion of soil particles and the intrinsic resistance of simple soil systems to wind erosion. J. Soil Sci. 21:154-161. Soil Conservation Service. National Cooperative Soil Survey. 1974. USDA-SCS - Lincoln, Nebr. Soil Conservation Service. National Cooperative Soil Survey. 1976. USDA-SCS - Lincoln Nebr. Soil Conservation Service. 1980. Cited in J. Soil Water Conserv. 35:102. Soil Survey Staff., 1965. Soil taxanomy. Agriculture Handbook No. 436. ‘SCS,USDA. . Stallings, J. H. 1957. Soil conservation. Prentice-Hall, Inc. Englewood Cliffs, N.J. Steel, R. G. D. and J. H. Torrie. 1960. Principles and procedures of statistics. McGraw-Hill, New York. Taylor, S. A. and G. A. Ashcroft. 1972. Physical edaphology. W. H. Freeman and Company. Woodruff, N. P. 1956. The spacing interval for supplemental shelter- belts. J. Forestry 54:115-122. Woodruff, N.P. 1975. Wind erosion research - past, present and future control programs. Proc. Soil Cons. Soc. Am. 30th: 147-152. Woodruff, N. P., and D. V. Armbrust. 1968. A monthly climatic factor for the wind erosion equation. J. Soil Water Conserv. 23:103-104. Woodruff, N. P., and W. S. Chepil. 1956. Implements for wind erosion control. Agric. Eng. 37:751-754, 758. Woodruff, N. P., W. S. Chepil, and R. D. Lynch. 1957. Emergency chiseling to control wind erosion. Kan. Agr. Exp. Sta. Tech. Bull. 90. 123 Woodruff, N. P., C. R. Fenster, W. S. Chepil, and F. H. Siddoway. 1965. Performance of tillage implements in a stubble mulch system: II. Effects on soil cloddiness. Agron. J. 57: 49-51. Woodruff, N. P., L. Lyles,F3 H. Siddoway, and D. W. Fryrear. 1977. How to control wind erosion. Agric. Inf. Bull. No. 354. ARS, USDA. Woodruff, N. P., and F. H. Siddoway. 1965. A wind erosion equation. Soil Sci. Soc. Am. Proc. 29:602-608. Woodruff, N. P., and F. H. Siddoway. 1973. Wind erosion control. Proc. National Tillage Conf. Des Moines, Iowa. Soil Cons. Soc. Am., Ankeny, Iowa, pp. 156-162. Zingg, A. W. 1951. Evaluation of the erodibility of field surfaces with a portable wind tunnel. Soil Sci. Soc. Am. Proc. 15: 11-17. Zingg, A. W., and W. S. Chepil. 1950. Aerodynamics of wind erosion. Agr. Engin. 31:279-282. MICHIGAN STATE UNIV. LIBRARIES 111111111111 111111111111” 111INIHWWIWHI 31293103784959