Evaluation of potential topsoil productivity A.S. Rogowski USDA-ARS, Northeast Watershed Research Center, 110 Research Building A, University Park, Pennsylvania 16802, USA

ABSTRACT

Examples of soil profiles from mimng areas in several states will be used to illustrate potential problems and to suggest an approach to topsoil management. The approach outlined will involve the computation of a soil profile productivity index based on rather simple and readily available chemical and physical parameters. These parameters include soil properties such as pH, bulk density, texture, plant available water, aeration porosity, electrical conductivity, sodium absorption ratio, and an estimate of root distribution. Computations will show how the values of productivity index for a reclaimed spoil profile will be modified by a biomass productivity potential, which is a function of mean annual temperature and precipitation at a particular site. The productivity-controlling zone in any reclamation INTRODUCTION operation will be the topmost layer of a mined profile, Much as t h e primary goal of any mining operations which consists of a relatively thin upper crust of a must remain the profitable recovery of a deposit, the regolith that has formed by weathering of the underneed to restore land to previous or better use follov~ing lying rock mantle or sedimentary deposits. The commining cannot be overemphasized. Traditionally, position and behaviour over time of this topmost layer exploration and exploitation of mineral deposits has will be examined in the hope of reconstituting as closely been the province of mining engineers and geologists. as possible the premined properties and structure of Only recently have soil and plant scientists been called soils following reclamation or, should the opportunity upon to make a contribution in this field, The nation present itself, improving on natural conditions. The has realized that it can no longer afford to exploit its soils that existed before mining have formed during a resources without meaningful restoration. Such res- long period of time. The challenge of present-day rectoration constitutes a great challenge to ingenuity, lamation is the ability to reconstruct the premining because there may be an opportunity to create produc- conditions, or preferably improvement of the soil or the tive agricultural complexes where none existed before. landscape that existed prior to mining. In the humid Equally, if not more important is the preservation of eastern United States, concern with coal mine drainage existing agricultural land by restoring it to its original tends to dominate much of the reclamation effort with productivity. Novel approaches to reclamation have less emphasis placed on surface conditions; however, demonstrated that a mined area can be handsomely great areas of western United States could be made reclaimed, revegetated, and subsequently used for a more suitable for plant growth by modifying present variety of purposes, ranging from farming and ranching soils. to recreation and wildlife habitat. Reclaimed mined Soils have formed in response to varying climate, land can produce food and fibre for years to come different parent materials, topography, and vegetative despite a temporary mining disturbance. Although cover conditions. In some areas the nutrients have been initially the new "soil" will need to be supplemented leached out by abundant rainfall; in others, extensive with fertilizer and organic matter, eventually a full- decomposition of organic matter has created a nutrientfledged soil profile will form, perhaps not unlike the rich plant growth medium. Some soils are too dry to original profile it replaced. support much plant life, others have developed impedAlthough coal reserves constitute probably the ing layers that obstruct water movement and root largest areal deposits (Fig. 1) to be stripmined, many growth. Soil properties vary from field to field and from other mineral resources produce large volumes of waste point to point. Soi~ descriptions, available from the that are, or will be, in need of reclamation and U S D A Soil Conservation Service or similar organizarevegetation, Some of the more important deposits in tions elsewhere, contain much pertinent information, the United States are indicated in Fig, 1, Significant y e t say little in a quantitative way of how productive a mining for uranium in Wyoming; copper in Utah, Ari- given soil is prior to mining and how productive it may zona, and Montana; phosphate in Florida; and clay in be following mining. Consequently, operators have no Georgia result in areas that will need to be reclaimed in guidance on how a given site should be handled for a manner similar to that ~ow used by the coal industry. optimum results. A biomass productivity model based on soil properties and climate is developed from literature and used to evaluate and compare potential effects of mining and reclamation on several soils in the continental United States. Soil productivity is assumed to vary as a product of root distribution function modified by five soil properties: available water, aeration porosity, bulk density, electrical conductivity, and pH. Yield limiting property levels are derived from literature and input soil data are obtained from available USDA Soil Conservation Service information on typical profiles. Modelled values of potential productivity after mining, indicate problems and limitations to be expected. The proposed model can be used as a guide to reclamation strategy, to restore the land to premining conditions, or at times to enhance productivity of a reclaimed area.

Environmental Geochemistry and Healtb~ 1985, 7(3). 87-97.

Evaluation of potential topsoil productivity

88

It might therefore be desirable to show objectively how mining affects soil productivity, and propose two, or more ways of topsoil handling which will attempt to minimize the impact of mining operations. To do so we need to model anticipated soil productivity before and after the site is reclaimed. SOIL PRODUCTIVITY POTENTIAL Climatic index

Some currently available biomass productivity models estimate potential site productivity based on climate alone. In Table 1 biomass productivity (B) is calculated using mean annual values of temperature (equation 1) and precipitation (equation 2) in the Miami Model, BT = 3000/(1 + exp(1.315-0, l19T) Bp = 3000{1 - exp(-0.000664P)} B = BT B = Bp

(1) (2)

B T ~< Bp Bp < BT

where: B, B T and Bp = biomass productivity in (g/mZ/yr) T = mean annual temperature (~ P = mean annual precipitation (ram)

I~

~

.

.~

The site productivity is taken as the lesser of temperature (BT) or precipitation (Be) dependent values (these are the starred values in Table 1). The major drawbacks of the model are that it does not take soil into account and that it computes productivity on a rather gross scale. Other models approach the productivity potential through soil effects alone. Kiniry et al. (1983) assumed that the productivity was a function of certain physical and chemical soil properties which either are or may become limiting during the growing season. The primary soil properties of interest are bulk density, aeration porosity, available water, pH, and electrical conductivity. The model is weighted with depth by root distribution. When climatic variables of temperature and precipitation are included in the computations, Kiniry et al. (1983) findings showed a higher correlation between the calculated index and field measured yields than when they were not. The Kiniry et al. (1983) approach has been modified here, and combined with Lieth's (1975) Miami Model for assessment of site productivity potential before and after mining. This combined model, available on request,* is designed to guide the user in handling *A Biomass Productivity Approach to Topsoil Handling, A.S. Rogowski and B.E. Weinrich, 110 Research Building A, University Park, PA, USA 16802.

Fe

5o

col ~ _

K

Figure 1 Distribution of coal reserves (shaded areas); of metals: A G-silver, A U-gold, CU-copper, FE-iron~ MO-molybdendum, PB-lead, V-vanadium, U-uranium and ZN-zinc; of minerals: CL-clay, K-potash, Ml-mica, MR-marble and P-phosphate; and of Oilshale-O; and location of typical profiles: 1-Grays Harbor Co., WA; 2-Blanc Co., MT; 3-Bowman Co., ND; 4-Otter Trail Co., MN; 5-Elko Co., NV," 6-Salt Lake Co., UT; 7-Shelby Co., IA; 8-Logan Co., IL; 9-San Pete Co., UT; lO-Ottawa Co., OK," ll-Johnson Co., 1L," 12-Jefferson Co., KY; 13-Cochise Co., AR; 14-Collin Co., TX; 15-Williamson Co., TN and 16-Brantley Co., GA.

A.S. Rogowski

89

Table 1 Biomass productivity potential as a function of mean annual temperature (T), or precipitation (P) for selected areas in Fig. 1.

County

Grays Harbor Blaine Bowman Otter Trail Elko Salt Lake Shelby Logan Sanpete Ottawa Johnson Jefferson Cochise Collin WiUiamson Brantley

State

Washington Montana North Dakota Minnesota Nevada Utah lowa Illinois Utah Oklahoma Illinois Kentucky Arizona Texas Tennessee Georgia

Location

Temperature (T)

SW NC SW C NE N S C C NE S NW SE W C SE

Precipitation re)

~

mm

10 3 6 4 10 10 9 11 10 15 14 13 16 18 14 20

1524 381 381 603 178 254 787 965 254 1016 1219 1143 178 1016 1321 1270

Biomass (T)

Productivity (e)

g/mZ/yr 1406" 832 1062 905* 1406 1406 1318 1496 1406 1846 1961 1673 1929 2087 1761 2231

1909~ 670" 670* 990 334* 466* 1221" 1419" 466* 1472" 1665" 1596" 334" 1472" 1752" 1709"

~Starred values are taken as site biomass productivity.

topsoil. It wiIt allow the user to choose which soil horizons, he wishes to save and what soil properties he needs to maintain, improve, or amend in a reconstituted profile. The model m a k e s a realistic appraisal of the site's reclamation potential. H o w e v e r , it should be realized that values used in the computations of productivity factors, such as bulk density, p H , water availability, as well as m e a n annual values of t e m p e r a t u r e and precipitation needed for the Miami Model, are in fact distributions in space and time, spanning a range of values with different probabilities of occurrence. Since field personnel ( U S D A Soil Conservation Service or similar organization) generally know what range of values is reasonable for a given area, users should consult them on the selection of realistic p a r a m e t e r s for use in the model. Users should also bear in mind that for a given year or a given point within a site results may differ considerably from those predicted, although on the average the model will be correct. Consequently, the model p r o p o s e d here should be used as a relative rather than an absolute predictor of productivity. In this sense it is particularly suitable for use in mining operations where the user generally wants to know how pre- and postminiog conditions relate to one another.

Soil index Relative productivity of a site (P) in g/mZ/yr can be written, m n P,= B 5" W 17 xj i=1 j=l

(3)

where: B = y = W = 17 = xj .... = Xj=l ~ xj=2 = xj=3 ----Xj= 4 = xj=5 = xj_-6 =

biomass productivity, previously defined in equation 1 or 2 summation operalor over i= 1,2.:.m horizons, or layers relative root distribution weighting function to 1 metre depth, dimensionless product operator productivity factors, n=6 available water, by volume moist bulk density at 1/3 bar g/cm3 aeration porosity, by volume pH, in 1:1 H20 electrical conductivity (EC), ds/m other factors, such as sodium absorption ratio, topography, or nitrogen content

Undisturbed field soil profiles usually consist of several distinct horizons in a solum of different depths. Similarly, plant root distributions vary. Some plants are shallow rooted, others have roots going down to considerable depth. H o w e v e r in reconstituted topsoil root distributions should be quite similar and reasonably constant for most profiles. To simplify the procedure we will consider here a 1 metre deep soil profile, assuming that such a profile approximates an average depth of a reclaimed topsoiled profile available for ptant growth. We have f u r t h e r m o r e assumed that below this one metre of topsoil, there is an unconsolidated spoil material compacted a t the surface and consisting of rock fragments or unweathered and mixed alluvial, lacustrine or aeolian deposits with little or no potential to support plant life. For comparison purposes the 1 m e t r e depth will be used throughout, realizing that some topsoiled profiles m a y be shallower and others deeper than the ones discussed here. Taylor and Terrell (1982) compiled information on rooting depth

Evaluation of potential topsoil productivity

90

of over 200 plant species. Their data show that, on the average, 1.5 + 1 metre is the depth reached by many roots, or their branches, to which the absorption of water and nutrients takes place. The structure of the model given by equation 3 is fairly simple. Productivity factors based on readily available data are computed individually for each horizon by considering the minima, or at times the maxima required for plant growth. These values are then multiplied together to show where the potential problems are. The products are subsequently multiplied by an appropriate weighting factor for each horizon, approximating plant root distribution, summed, and multiplied by a site biomass productivity potential (Table 1) giving an overall site productivity index. Weighting factor (roots)

To compute the roots distribution weighting factor a method proposed by Horn (1971) and adapted by Kiniry et al. (1983) was used. Horn (1971) computed root distribution from depletion measurements of water holding capacity in a 3.5 m soil profile under maple trees. Kiniry et al. (1983) assumed that the same approach can be used in soil profiles other than 3.5 m deep, provid~ed that the rooting depth of growing plants was known. Horn's key assumption was that the depletion of soil water holding capacity corresponded to density (actually length) of root distribution. The profile was assumed to be uniformly moist, lacked a water table, and had no physical or chemical barriers to root growth. The profile water holding ~apacity was taken as the difference between the soil moisture in the spring and at 1500 kPa (15 bar). In here a 1-m rooting depth profile is assumed which satisfies the other requirements, i.e., the reconstituted profile will have no physical or chemical barriers to root growth and no shallow water table. This appears to be a reasonable assumption for engineered profiles where an operator has a considerable latitude in selections of desirable horizons and in handling the soil in such a manner as to minimize adverse effects of compaction. The principal assumption of this model is that the biomass production is a function of root growth which in turn depends primarily on available water, bulk density, pH, and where applicable, on electrical conductivity, sodium adsorption ratio, and nutrient status of the soil. Equation 4 is an inverse hyperbolic sine curve, which gives a fractional depletion of profile moisture (W) at a depth r R W = 0.152

~

J

log R + ~ / R 2 + 6 . 4 5 dr

(4)

r + ~ r=o

where: R = rooting depth (cm) r = depth of layer in the profile When integrated with depth and divided by total profile depletion the equation gives the root distribution

weighting function (W). Others (Gardner, 1964) have described root distribution as decreasing logarithmically with depth. U s e of the inverse hyperbolic sine for reconstituted profiles appears reasonable, since the highest concentration of nutrients and water which materially affect productivity will probably occur near the surface but will decline less rapidly in a uniform reclaimed soil than a logarithmic distribution curve may indicate. Available water capacity

Available water capacity (AWC) in mm should realistically describe the amount of water that could be available for plant use during the growing season. This value will vary depending on soil texture, and water use efficiency of a particular plant species. Consequently, we must choose as a critical soil water content a value that will most likely limit plant growth under the particular site conditions. Working in Missouri, Kiniry et al. (1983) chose a value of 0.20 cm/cm as the limiting available water content and represented PAWC (available water capacity productivity factor) as, PAWC = AWC/0.20

(5)

For AWC >/0.20 cm/cm, PAWC = 1.0. For a 100-cm profile PAWC of 1.0 would amount to 200 mm of available water. The values of AWC, for a particular soil, can be estimated (Kiniry et al., 1983); Peterson et al., 1968) from texture (Table 2). AWC can also be expressed as the difference between water contents of 33 and 1500 kPa. Such values may be satisfactory for use in the model, provided they reflect actual field conditions. Table 2 Potential available water capacity estimated from soil texture.

Texture

Sand Coarse sand Medium sand Fine sand Loamy sand Loam Sandy loam Fine sandy loam Loam Silt loam Silt Very fine sandy loam Clay Clay Sandy clay Silty clay Sandy clay loam Clay loam Silty ~lay loam

Estimated water holding capacity cm/cm

mm I

0.016 0.030 0.066 0.070

16 30 66 70

0.115 0.130 0.180 0.190 0.200 0.200

115 130 180 190 200 200

0.100 0.110 0.115 0.125 0.145 0.145

100 110 115 125 145 145

~In 100 cm profile; for textures where 0.15 ~ clay ~< 0.40 estimated water holding capacity was computed as the average of textural limits for a given texture class from, AWC = 0.25-0.35 clay.

A.S. Rogowski Bulk density (BD) and aeration porosity (P) Since both bulk density and aeration porosity can be modified during topsoil handling they offer an opportunity to improve a reclaimed profile as compared to its natural condition. Assuming that reserves of moisture and nutrients are adequate, soil productivity depends largely on how well plant roots can access these reserves. It has long been established that dense soil horizons impede root growth by preventing root elongation, limiting respiration, and at times contributing to water logging. Soil bulk density (BD) is commonly used as an index of compaction. Its effects, however, should be evaluated relative to soil texture, moisture and moisture content at the time the soil is handled. Soil moisture content is particularly critical on reclaimed profiles where dense layers can often be produced if topsoil is over compacted during handling at or near the optimum moisture content (Terzaghi and Peck, 1948). If however, a seedbed is not firm enough to be in good contact with a planted seed, poor germination and excessive droughtiness of a soil may result. To account for these limitations, a bulk density productivity factor (D) suggested by Kiniry et al. (1983) was used, D = 1.00, < 1.30 D = 1.88- 0.68 BD, 1.30 ~< BD ~< 1.55 D = 5.96- 3.31 BD, 1.55 < BD ~< 1.80 D = 0 , B D > 1.80

(6)

where BD is the bulk density of moist soil at 1/3 bar. Users should remember that for particular conditions either lower or higher values of critical bulk density may be indicated, but in general BD suitable for plant growth will range from 1.30 to 1.80 g/cm3. A good discussion of the effects of bulk density is given by Pearson (1965) and again by Bowen (1981). In the model presented here, most physical and hydraulic effects of bulk density are assumed to be incorporated into the factor D. Other effects, such as adequate aeration of root growth medium necessary for proper respiration and functioning of the roots, are grouped under productivity factor aeration porosity. The two are related through the equafionl P = 1.0- (BD/2.65) - 0

(7)

where P is the aeration porosity of a fully recharged profile (at field capacity), 0 is the moisture content (or the volume of small pores), and 2.65 is the particle density of a mined soil, Values of P will vary with soil texture, structure and the amount of organic matter. In general P refers to macro pores (> 0.66 mm diameter) that will drain at low tensions (less than 1/3 bar). Actual values of the aeration porosity can be measured in the field, or can be approximated as a fraction of total porosity, P = (1-BD/2.65) c~

91

other soils most of which have some limitations it is taken as porosity at 0.8 of saturation (a = 0.2). The value of c~is related to pores that drain at low tensions, and equation 8 is an attempt to arrive at a potential value of aeration porosity that may at some time be limiting in a given soil. Literature suggests (Cannetl and Jackson, 1981; Pearson, 1965) that critical aeration porosity (Pcrit) when root growth may become restricted normally ranges from 0.05 to 0.15 pore space by volume, in the model it is therefore set at 0.10 pore space by volume. Realistically, aeration porosity effect should also include (but does not) a built-in dependence on time, a geometry factor to describe degree of continuity between air-filled pores and concentration level of CO2. In some of the mine spoils where root respiration may compete for pore oxygen with oxidation of pyrite or iron in the profile, or when heavy additions of organic material (such as sludge) place an additional demand on pore oxygen, it may well be advisable, particularly for deeper layers, to meet Pcrit higher than the recommended value. Productivity factor aeration porosity Pa was estimated in the model as an integral with depth of aeration porosity reciprocal divided into and integral with depth of Pcrit reciprocal, r=R

Pa= f

r=R 1/Pcritdr /

f

r=o

1/Pdr

(9)

r~o

Using equation 7 it is possible to derive values of porosity at planting from field measured values of bulk density and moisture content. Workable estimates of P can be obtained by computing anticipated moisture storage in the spring for a given profile and by substituting its volumetric equivalent into equation 7. To be practical, however, an operator should ask himself if at any time during the growing season, particularly when soil is wet, dr following a heavy incorporation of organic matter, the soil oxygen concentration is likely to drop near critical level for even a short period of time (Cannell and Jackson, 1981). In the event of water logging, for instance, factor Pa for a particular profile may become quite critical. Consequently, in the model the use of equation 8 is recommended. At times, however, the operator may wish to minimize organic matter conversion or nutrient leaching in the topsoil that is stockpiled before being spread. Under these conditions he may want to manipulate bulk density or aeration porosity so as to make values of D in equation 6 and Pa in equation 9 as small as possible.

(8)

The approximations we suggest attempt to relate capability classes as used by USDA Soil Conservation Service to possible effects of field compaction. The model assumes that in free draining soils that are stony, channery, or shaley P is equal to about 1/2 of the total porosity (a = 0.5). For soils likely to be water logged at some time during the growing season P is set equal to porosity at 0.9 of saturation (a = 0.1), while for all

Soil reaction Soil reaction (pH) values appear particularly well suited to characterize productivity response of reconstituted minesoil profiles in the eastern United States, but the user should realize that the approach proposed here is rather simplistic. Critical pH (Spurway, 1941) varies among soils and plant species, and with time. The response to pH on acid soils may result from H toxicity,

Evaluation of potential topsoil productivity

92

A1 toxicity, Mn toxicity, Ca deficiency, or Mo deficiency. Thus, soils at the same value of pH could have limited yields for different reasons and the limiting factors would operate at different intensities in time and space (Pearson, 1965; Adams, 1981). Consequently, the proposed model, which follows Neill's (1979) original formulation, should be used with caution and can be adjusted if sufficient information about a particular site or plant species is readily available. In the meantime, the model will provide sufficient guidance for the potential user in differentiating between the layers that may cause him problems and those that will not. The pH productivity factor (pHv) (10) pHv = 0 , pH ~ 5.4 where pH denotes a measured value of pH in the 1:1 aqueous solution. These values are generally available from USDA Soil Conservation Service, or University personnel, or can readily be measured. Although Neill (1979) and Kiniry et al. (1983) have used in their model pH values in 0.01M CaC12 slurry, and although there are advantages to measuring pH in 1M KCL or in 0.01M CaCI2 solutions, these advantages are outweighed by convenience of measuring it in 1:1 soilwater slurry. Since the majority of published date in the USA report pH in water slurry and most soil testing laboratories recommend (Adams, 1981) lime application based on those values, the soil pH is used in the model proposed here.

Electrical conductivity ( EC) An argument similar to that offered above for pH applies as well to the effects of salinity on plant growth (Hoffman, 1981). Salinity effects may vary spatially with soil type, texture and moisture status while different plant species will exhibit different tolerances. Salinity associated problems are almost certain to be present if reclamation is carried out in arid regions when original soils contain sufficient soluble salts derived either from marine deposits or from soil weathering. The problems can also arise in semiarid regions whenever rainfall is approximately equal to evapotranspiration; as a result of upward artesian flow from aquifers; from overirrigation and from resulting saline seeps in adjacent areas; or if high water tables are present. Consequently, particular attention needs to be paid to the water regime and potential flow pathways in reconstituted profiles: Hoffman (1981) rates plants (his Table 9.3, p.315) according to their salt tolerance, and suggests an appropriate form of the productivity factor, ECF ECF = 1.0 - B ( E C - A )

(11)

where A is salinity threshold value, B is the yield reduction per unit fo salinity increase and EC is the electrical conductivity of soil saturation extract in ds/m. Selected values abstracted from Hoffman (1981) table are shown in Table 3. Equation 11, which separates moderately sensitive (MS) and moderately tolerant (MT) plant species with A = 3.0 and B = 0.0769, can be written,

ECv = 1.0 , EC ~ 16

(12)

Inspection of Table 3 will tell us that ECv can vary considerably depending on the crop used. Nevertheless, the model will alert the user that the salinity problem may exist in the reclaimed profile. Table 3 Salt tolerance of some agricultural crops (Hoffman,

1981). Crop

Threshold (A)

Yield reduction (B)

ds/m

per ds/m

1.7 1.7 3.9 6.9 7.7

0.16 0.12 0.53 0.64 0.52

Orange Corn Tall fescue Bermuda grass Cotton

Tolerance 1

S MS MT T T

~S=sensitive, MS=moderately sensitive, MT=moderately tolerant, T=tolerant.

Sodium saturation ratio (SAR) Soils or soil horizons that have excess sodium in their exchange complex are known as sodic soils or sodic horizons. Such soils when leached with low electrolyte content waters may show a marked decrease in permeability (Frenkel et al., 1978; Reeve and Bower, 1960). Many studies (Rhoades, 1982) have dealt with reclamation Of sodic soils and with evaluation of the irrigation water quality (Oster and Rhoades, 1976; Rhoades, 1972). Guidelines, regarding the suitability of irrigation waters for agriculture, based on the type of predominant clay mineral, have been proposed (Ayers and Westcot, 1976) and questioned by more recent findings (Frenkel et al., 1978; Shainberg et al., 1981; Suarez et al., 1984). Currently the consensus appears to be that the permeability of sodic soils may decrease with increasing pH and in general, depends on the electrolyte level a soil maintains; substantial decreases having been observed with low electrolyte contents (Rhoades, 1982; Shainberg et al., Suarez et al., 1984). Such variations in soil permeability would have a significant effect on infiltration and subsequently on the amount of water available to plants. Figure 2 from Rhoades (1982) summarizes the current situation for some of the more sensitive arid land soils and is used here to derive a productivity correction factor for high sodium soils. The SAR values in Fig. 2 are the adjusted SARa values (Bower et al., 1968; Oster and Rhoades, 1976) in the topsoil and the electrical conductivity is that of infiltrating water, here assumed to be in equilibrium with the soil solution (EC~). Accordingly, the curve in Fig. 2 is broken into two straight line segments at SARa = 10 and ECe = 1, ECc = 0.07 SARa + 0.30 SARa ~ 10

(14)

A.S. Rogowski

30

I

25

/

Area of likely permeability / hazard

/

O

Q.

2

t-

!

I

93

On the average, Frenkel et al. (1978) have observed an 83 percent reduction in permeability of montmorillonitic, kaolinitic and vermiculitic soils when leached with distilled water as compared with IN NaCI-CaCI solution. The model given here assumes a productivity reduction factor (SARF) proportional to the ECe/EC ratio, where E C is the electrical conductivity of the soil,

I

/

20

O

SARF = (EC~-0.3)/EC, 0 10 SARF = 1.0 , SARF > 1 SARF = 0 , SARF < 0

E "O O

o~

0

I i i m 0 1 2 3 4 5 6 Electrical conductivity in infiltrating water, dS/rn

Figure

2 Threshold values of adjusted sodium adsorption ratio of topsoil and electrical conductivity of infiltrating water (assumed ta be in equilibrium with soil solution) for maintenance of soil permeability (from Rhoades, 1982).

I

I

I

I

(16)

At this stage no p H dependence was incorporated, and the program/subroutine is activated only when a problem is thought to exist. The adjusted SARa values are calculated following the procedure outlined in Ayers and Westcot (1976), SARa = SAR(1 + (8.4-pHc))

(17)

where SAR = Na/X/(CA+Mg)/2

(18)

and pHc = (pK2-pK'c) + p(CA+Mg) + p(Alk)

100

(I5)

(19)

where (pK2-pK'c) = 1.86 + 0_23(Ca+Mg+Na) u4

(20)

p(Ca+Mg) = 3.29- 0.43 in(0.001 +Ca+Mg)

(21)

and p(Alk) = 2.99 - 0.43 ln(0.001+HCO3)

10

p H c - a theoretical p H of water in contact with lime and in equilibrium with soil CO2-is computed from equations 19, 20 and 21 fitted to curves in Fig. 3. The input required is the concentration in meq/1 of Ca, Mg, Na, CO3 and HCO3 in the soil solution extract. The computations outlined here may help to identify the problem if one exists. In the event of attempted irrigation, procedures given by Rhoades (1982) should be followed.

B

m

~r

E C

o

(22)

1.0

C C

o L)

Topography

A/

0.1

.01 0

Figure

C

t

1

t

1

1

2

3

4

5

Value

3 Plots of algorithms used for calculating the components (equation 19) of adjusted sodium adsorption ratio, p K2pg'c (A), p(Ca+Mg) (B), and p(Alk) (C) (from Ayers and Westcot, 1976).

There exists a great complexity in the soil system. Attempt has been made here to deal with this complexity in two dimensions, by considering the distributions of the productivity index identified with a particular soil series. However, it is a well known fact in pedology (Jenny, 1980) that local relief, aspect, and drainage, are among the most significant modifiers of the soil profile, particularly on a small scale. Thus, south facing slopes may have a different vegetation, moisture, or temperature regime from the north facing ones. Soils developed on the hilltops are likely to be shallow compared with those developed near the base. Properties such as cation exchange capacity or clay content may vary significantly with elevation and drainage and available moisture may range from excessive to impaired depending on position relative to the slope and degree of profile anisotropy (Zaslavsky and Rogowski, 1969).

94

Evaluation of potential topsoil productivity

Considerations such as the ones above should be incorporated into the productivity assessment primarily through a representative sampling of the area to be reclaimed and through use of p a r a m e t e r values that adequately reflect the area heterogeneity. This is of p a r a m o u n t importance when attempting to reclaim land in Appalachia and elsewhere where mountainous, rapidly changing conditions prevail.

Table 4

RESULTS Figure 1 shows the approximate location of the selected typical profiles; Table 1 gives the values of biomass productivity index for these profiles based on climate alone, Table 4 shows the computation of available water capacity, and Table 5 lists the Biomass Productivity Index values for the same profiles following the execution of the C o m p a r a t i v e Biomass Productivity Model. Figure 4 illustrates a typical computer program output sheet for a soil, here Profile # 3 , a Mollisol, from B o w m a n County, North Dakota, The lower table in Fig. 4 contains input soil p a r a m e t e r s as well as the mean annual t e m p e r a t u r e , precipitation, and clay content for each horizon, A place for user's soil number, and a space for a three-digit soil mapping unit name, although not utilized here, are also available. In computing available soil water holding capacity, values higher than 0.20 cm/cm were set equal to 0.20, The actual water holding capacity for this and other profiles studied was taken as the customary difference between water content at 33 and 1500 kPa. W h e n a profile is, or can be fully recharged in Spring, the above procedure may be correct, however soils can retain water at

1 2 SOIL ; NAME

3 3 3 3 3 3 3 3 3

SOIL

P87 P87 P87

P87 P87 P87 P87 P87 P87

NAME

3 DEPTH ( c m )

0 8 18 25 30 38 53 71 97

8 18 25 30 38 53 71 97 100

DEPTH ( r

4 HORIZON

5

PAWC

A A A A B B B B I

1.0000 1.0000 1.0000 0.9500 0.8000 1.0000 1.0000 0.9500 1,0000

HORIZON

CLAY

6 D

7 Pa

1.0000 0.9280 0.9348 0.9348 0.6680 0.8736 0.8464 0.6348 0.8532

1.0000 0.9935 0.9812 0.9760 0.9306 0,9167 0.8996 0.8657 0.8655

1.0000 1.0000 1,0000 1.0000 1.0000 1.0000 1,0000 1.0000 1.0000

DENSITY

WATER

pH

P87 P87 F87 FB7 PB7 F87 P87 P87 P87

0 8 IB Z5 30 38 53 71 97

8 18 25 30 38 53 71 97 100

A A A A B B B B I

O.09s 0.1050 0.0950 O.O610 O.l&O0 0.1950 0.0930 0.0890 0.1890

1.24 1.40 1.39 1.39 1.60 1.48 1,52 1.61 1.51

Depth (cm)

6.40 6+90 7.70 ~.40 9.00 8.60 8.50 9.30 8.90

1

0 0 0.32 8 18 0.29 18 25 0.21 25 30 0.19 30 38 0.16 38 53 0.20 53 71 0.21 71 97 0.19 97 100 0.24 Potential Profile Capacity 216 mm Maximum Spring Profile Capacity 85

2

3

4

020 0.20 0.20 0.19 0.16 0.20 0.20 0.19 0.20

0.12 0.11 0.08 0.07 0.06 0.08 0.08 0.07 0.09

0.60 0.55 0.40 0.36 0.38 0.40 0.40 0.37 0.45

'l=Computed as water content at 0.33 kPa (l/3 bar) less water content at 1500 kPa (15 bar). 2=Values listed in Fig. 4. 3=Computed as vaIues under (1) reduced m the ratio of actual to maximum potential (Thorthwaite, 1948} water holding capacity 85/216 = 0.39. 4=Column 3/column 2.

1.0000 1.0000 1.0000 1.0000 0.9685 0.3303 0.8378 0.8378 0.2303

EC

(as/m) 0.2000 0.2000 0.2000 0,1900 0.1600 0,2000 0,2000 0,1900 0.2000

Computations of available water capacity Available water capacity (cm/cm/

PRODUCTIVITY FACTOR 8 9 pH F EC F

(oleo) 3 3 3 3 3 3 3 3 3

tensions lower than 33 kPa (1/3 bar), and some plants can utilize water at tensions higher than 1500 kPa (15 bar). Thus, a good practical estimate of soil water holding capacity (except for sandy soils) is the difference between water held at 0.9 saturation and i500 kPa (15 bar). In areas with insufficient rainfall, such as B o w m a n County, North D a k o t a , example in Fig, 4, a m o r e nearly correct estimate of A W C is to subtract

0.0 0.0 0,0 0.0 3.40 11.70 5.10 5,10 13.00

INDEX i0 SARF

1.0000 1.0000 1.0000 1.0000 1.0000 0.7458 1.0000 1.0000 0.6254

SAR

a 0.0 0.0 O.O 0.0 68.74 87.28 81.31 81.31 81,31

ii PRODUCT

12 W

1.0000 0.2653 0,9271 0.2108 0.9215 0.1104 0.8685 0,0662 0.4909 0.0886 0.1985 0.1218 0.6430 0.0889 0.4446 0.0476 0.1064 0.0005 CUMULATIVE

RAIN

TEMP

(H)

(~)

381.0 381.0 381.0 381,0 381.0 381.0 381.0 381,0 381.0

6,0 6.0 6.0 6.0 6.0 6,0 6.0 6,0 6,0

13 PRODUCTIVITY

14 EIOMASS (g/sq m/yr)

0.2653 0,1954 0.1018

177.9 131,0 68.2 38.5 29.2 16.2 38.3 14.2 0.0 513.6

0.0575 0.0435 0,0242 0.0571 0.O211 0,0000 0.7660

CA

MG

NA

NCO~

(~eqT1) 0.0 0.O 0,0 0.0 1.6 18.9 1,0 1,0 1.0

0.0 0.O 0.0 0.0 1.2 21.5 1,5 1.5 1.5

0.0 0.0 O.O 0.0 38.6 332.0 51,0 51,0 51.0

674.0 675.0 676A0 677.0 IO.0 9.0 5.8 5.8 5.8

Figure 4 Computer printout containing input and output of the comparative biomass productivity" model for a profile in Bowman Co., ND.

A.S. Rogowski

water content at 1500 kPa from an estimated profile water content in the Spring, to correct for maximum Spring profile capacity. As a last resort, when desorption data are lacking, but values for clay content are available, estimated profile water content can be computed from clay content and climatic data (Table 2). Productivity factor values given in the upper table of Fig. 4 for Profile #3 constitute the output which is pertinent to choosing an appropriate procedure for topsoil handling. The magnitude of the five productivity factors is listed by depth and horizon for this soil in column 6 through 10. The factors range from 0 (critical value exceeded) to 1 (no soil limitations to root growth). Their product (column 13) integrated in 1 cm increments, and multiplied by the root distribution function in column 12 shows the potential quality of each layer. Thus some layers (i.e., 38 to 53 cm and the bottom one) show low product values. Scanning individual products it can be seen that both layers have high EC values and SARa limitations. The high electrical conductivity (EC) values reduce the productivity factor in all layers below 30 cm, and are particularly serious in the above two layers in their potential effect on SAR. The analysis suggests that at this site the 0 to 30 cm layer is best for plant growth and therefore it should be segregated and used in reclamation. Column 12 in Fig. 4 gives the estimated profile root distribution which when multiplied by the values in column 11 (Product) gives the Productivity Index of each layer in column 13. The Productivity Index when multiplied by the Biomass Productivity (Table 1) gives the Biomass Productivity Index (column 14) of each horizon which when summed gives the Profile Productivity Index of 514 g/mZ/yr. Because of limited recharge at this site only part (39%) (Soil Survey Staff, 1975) of the available water capacity will be filled in Spring. Consequently, available water should generally be reduced by that amount Table 5 No.

1 2 3 5 6 7 8 9 10 11 12 13 14 15 16

95

(Table 4 column-marked 3). The profile productivity will then be reduced in the ratio of column marked (3) to column marked (2) as given by column marked (4) in Table 4. When this correction is applied the profile Biomass Productivity I n d e x drops to 221 g/m2/yr. At this site excess sodium in the profile below the 0.3 m depth was expected to affect permeability and reduce available water content. As a consequence results were calculated using the approach involving a computation of SAR. Thus, Fig. 4 also lists the input and output for this option including values of adjusted sodium adsorption ratio and concentrations of Ca, Mg, Na and HCO3 in the soil solution extract. The Biomass Productivity Index of 514 g/mZ/yr given here was only slightly less than if it were computed without correction for SAR. The reasons for the small effect of SAR on the overall profile productivity is the relatively deep location of the SAR affected layers. Had these layers been located at, or near the surface, a substantial reduction in predicted productivity could have occurred. The discussion of Fig. 4 illustrates the type of data the model will provide the user. This kind of information may guide the user in handling the topsoil so as to achieve optimum results, and tell the operator how his results will compare with results from other areas in the nation. Table 5 summarizes the output of the Comparative Biomass Productivity Model for some typical profiles. The input information for these profiles, chosen for soils from or near major mining areas, was extracted from a Soil Survey Staff (1975) Publication-Soil Taxonomy. The biomass productivity values in Table 5 range from a low of 110 g/mZ/yr for a dry Aridisol in Arizona (#13) to highs of over 1000 g/m2/yr for an Inceptisol in Washington (#1) and a Mollisol in Illinois (#3), (these numbers translate to potential biomass production ranging from 0.5 to 6 T/A/yr). Table 5 ranks these soils according to their productiv-

Biomass productivity and limitations for selected locations near or in mining areas. Location

Biomass 3 g/m2/yr

LimitationJ

Ordere

Grays Harbor Co., Washington Blaine Co., Montana Bowman Co., North Dakota Elko Co., Nevada Salt Lake Co., Utah Shelby Co., Iowa Logan Co., Illinois Sanpete Co., Utah Ottawa Co., Oklahoma Johnson Co., Illinois Jefferson Co., Kentucky Cochise Co., Arizona Collin Co., Texas Williamson Co., Tennessee Greenbrier Co., West Virginia

1169 211 514 277 465 938 1143 447 951 1332 976 110 495 1038 589

pH AW, BD, A, EC, SAR, C AW, BD, A, EC, SAR, C AW, BD, C C AW AW, BD AW, C AW, BD, A BD, A, pH AW, BD, A, pH AW, BD, A, C AW, BD, A, SAR AW, BD, A, pH BD, A, pH

I M M M M M M M A A A AR V A I

ILimitations: Limiting factors such as, AW=available water, BD=bulk density, A=aeration, pH=pH, EC=electrical conductivity, C=climate, SAR=sodium adsorption ratio. 2Soil order: I=Inceptisol, M=Mollisol, A=Alfisol, AR=Aridisol, V=Vertisol. 31 g/m2/yr *0.004467 = i T/A/yr.

96

Evaluation of potential topsoil productivity

ity. Thus, most care should be taken in handling the productive soil, like the profiles #1, #8, #11 or #15 to restore them to nearly the same productivity values. Less productive soils, on the other hand, afford an opportunity for modifying the profile so as to increase the productivity. Profile #14, a Vertisol from Texas, has high sodium content in the profile, high bulk density when dry, limited available water, and low aeration values in most horizons. The area needs to be managed properly so as to maintain favourable bulk density and increase aeration in the upper 56 cm. Profile #13 (Arizona) lacks water, has high density and low aeration values. Supplemental irrigation and proper management to decrease density and increase aeration could substantially increase the productivity, Even some of the better profiles can be improved. For inst~ ance, Profile #1 from Washington which is limited by pH and would benefit from liming. Profile #8 from Illinois which is somewhat dry and in part limited by bulk density and adequate aeration, could benefit from both an increase in water holding capacity, perhaps by adding organic matter, and from an overall improvement in management, by decreasing density and increasing aeration. Here the user will have to make a choice. If the site is reclaimed as is, the projected profile productivity value would be 1143 g/m2/yr. If the bulk density (BD) and aeration in the reconstituted profile were engineered so as to make their productivity factors equal to 1 (i.e., BD ~< 1.30) the Biomass Productivity Value would increase to 1160 g/m2/yr for a productivity gain of about 1 percent. If, on the other hand, water holding capacity (AWC) was increased and other factors were kept constant, a much larger increase to 1400 g/m2/yr (a 22 percent gain) would result. If both AWC and BD were improved, a Biomass Productivity increase to 1419 g/m2/yr could take place (a 24 percent gain). Under these circumstances the profile would have no apparent limitations to plant growth and would be controlled by climate alone. A similar situation appears to exist in the two Utah profiles (profile #6 and #9) where climate plays the dominant role in determining the productivity potential. Under such circumstances reclamation to status quo and supplemental irrigation may be advised. In much the same way one can analyze the other profiles listed. Their limitations to plant growth are expressed primarily in terms of available water (#2, #3, #5, #7, #8, #9, #10, #12, #13, #19 and #15), bulk density and aeration (#2, #3, #5, #8, #10, #11, #12, #13, #14, #15 and #16), pH (#1, #11, #12, #15), EC (#3), SAR (#2, #3, #14) and climate #2, #3, #5, #6, #9, #13). A different selection of soil profiles may perhaps show different trends: SUMMARY AND CONCLUSIONS How best to handle topsoil on a mined area? The user should answer this question by first placing it in the context of where the area is located. He will note that while abundant rain, erosion potential, and low pH values will create problems in one area, lack of water, salinity, excess sodium, and high bulk density will govern the use of reclamation techniques and topsoil handling in the other. The user should next examine the climate to see what kind of moisture budget he may

expect in his chosen area and simultaneously check the soils at the mine site for any possible problems that may arise because of their taxonomic classification. Having satisfied himself about the broad implications of location, climate, and soils, the user will need to get more specific. Execution of the Comparative Biomass Model (Adams, 1981), will point out where the problem areas are. Having identified his problem soils, the user may then want to select different options available, and to simulate possible outcomes. Having decided on the desired course of action, the user would want to consult with the Soil Conservation Service or University Extension personnel in his area, to select specific techniques recommended to overcome particular problems on the area of interest. It is hoped that the procedure outlined above may streamline the solving of problems associated with topsoil handling, may identify areas that need closer attention, and may suggest techniques to improve productivity of a mined site compared with original soil. The model should not be regarded as final. Much room for improvement and for incorporation of additional components exists. It however, can form a workable framework for assessing efficient comparative reclamation procedures. ACKNOWLEDGMENT This paper is a contribution from the U.S. Department of Agriculture, Agricultural Research Service, in cooperation with the Pennsylvania Agricultural Experiment Station, The Pennsylvania State University, University Park, Pennsylvania, supported by EPAARS Interagency Agreement Funds: EPA-IAG-D5E763. REFERENCES Adams, F. 1981. "Alleviating chemical toxicities," p.269 in G.F. Arkin and H.M. Taylor (ed.) Modifying the Root Environment to Reducing Stress, ASAE Monograph No. 4, American Society of Agricultural Engineers, 2950 Niles Road, St. Joseph, MO. Ayers, R.S. and Westcot, D.W. 1976. Water quality for agriculture, Irrigation and Drainage Paper No. 29, Food and Agriculture Organization of the United Nations, Rome, Italy. Bowen, H.D. 1981. "Alleviating mechanical impedance," p.2111 in G.F. Arkin and H.M. Taylor (ed.) Modifying the Root Environment to Reducing Stress, ASAE Monograph No. 4, American Society of Agricultural Engineers~ 2950 Niles Road, St. Joseph, MO. Bower, C.A., Ogata, G., and Tucker, J.M. 1968. Sodium hazard of irrigation waters as influenced by leaching fraction and by precipitation or solution of calcium carbonate, Soil Sci. 106, 29-34. Cannell, R.Q and Jackson, M,B, 1981. "Alleviating aeration stress," p.191 in G.F. Arkin and H_M. Taylor (ed.) Modifying the Root Environment to Reducing Stress, ASAE Monograph No. 4, American Society of Agricultural Engineers, 2950 Niles Road, St. Joseph, MO. Frenkel, H., Goertzen, J.O., and Rhoades, J .D. I978. Effects of clay type and content, exchangeable sodium percentage, and electrolyte concentration on clay disperson and soil hydraulic conductivity, Soil Sci. Soc, Am. J. 42, 32-39. Gardner, W.R. 1964. Relation of root distribution to water uptake and availability, Agronomy J. 56, 41-45. Hoffman, G.J. 1981. "Alleviating salinity stress," pages 305-341 in G.F. Arkin and H,M. Taylor (ed.) Modifying the Root Environment to Reducing Stress, ASAE Monograph No. 4, American Society of Agricultural Engineers, 2950 Niles Road, St. Joseph, MO.

A.S. Rogowski Horn, F.W. 1971. "The prediction o1 amounts and depth distribution of water in a well-drained soil," M.S. Thesis, Univ. of Missouri, Columbia, MO, Jenny, H. 1980, The Soil Resource: Ecological Studies 37, SpringerVerlag, New York. Kiniry, L.N., Serivner, C.L., and Keener, M.E. A soil productivity index based upon predicted water depletion and root growth, Research Bulletin 1051, Univ. of Missouri, Columbia, MO. Lieth, H. 1975, "Modeling the primary productivity of the world," in H. Lieth and R.H. Whittaker (ed.) Primary Productivity of the Biosphere, Springer-Verlag, New York. Neill, L.L. (L.N. Kiniry), 1979. An evaluation of soil productivity based on root growth and water depletion, Unpublished M.S. Thesis, Univ. of Missouri, Columbia, MO. Oster, J.D. and Rhoades, J.D. 1976. Various indices for evaluating the effective salinity and sodicity of irrigation waters, Proceedings of the International Salinity Conference, Texas Tech. Univ., Lubbock, TX, August 1976, 1-14. Pearson, R.W., 1965. "Soil environment and root development," p.9511, in N.H. Pierre (ed.) Plant Environment and Efficient Water Use, American Society of Agronomy, 677 South Segoe Road, Madison, WI. Peterson, G.W., Cunningham, R.L,, and Matelski, R.P. 1968. Moisture characteristics of Pennsylvania soils: moisture retention as related to texture, Soil Sci. Soc, Am. Proc. 32, 271-275. Reeve, R.C. and Bower, C.A, 1960. Use of high salt waters as a flocculent and source of divalent cations for reclaiming sodic soils, Soil Sci. 90, 139-144,

97

Rhoades, J.D. 1972. Quality of water for irrigation, Soil Sci. 113, 277-284. 9Rhoades, J.D. 1982. Reclamation and management of salt-deflected soils after drainage, Soil and Water Management Seminar, Lethbridge, Alberta, Canada, Nov. 29-Dec. 2. Shainberg, I., Rhoades, J.D., Suarez, D.L.~ and Prather, RA. 1981. E~fect of mineral weathering on clay dispersion and hydraulic conductivity of sodic soils, Soil Sci. Soc. Am. J. 45,287-291. Soil Survey Staff. 1975. Soil taxonomy, Agriculture Handbook No. 436, U.S. Government Printing Office, Washington, DC 20402. Spurway, C.H. 1941. Soil reaction (pH) preferences of plants, Michigan Agric. Exp. Sta. Spec. Bull, 306. Suarez, D.L., Rhoades, J.D., Lavado, R., and Grieva, C.M. 1984. Effect of pH on saturated hydraulic conductivity and soil disperson Soil Sci. Soc. Am. J. (in press), Taylor, H.M. and Terrell, E.E. 1982. "Rooting pattern and plant productivity," in Miloslav Rechligl, Jr. (ed.) Handbook of Agricultural Productivity, Vol. 1, CRC Press, Inc., Boca Raton, FL. Terzaghi, V. and Peck, R.B. 1948. Soil Mechanics in Engineering Practice, p.5611, John Wiley and Sons, New York. Thornthwaite, C.W. 1984. An approach towards a national classification of climate, Geographical Review 38, 55-94. Zaslavsky, D. and Rogowski, A.S. 1969, Hydrologic and morphologic implications of anisotropy and infiltration in soil profile development, Soil Sci. Soc. Am. Proc. 33, 594--599.