determination of the properties of saline and alkali soils€¦ · chapter 2 determination of the...

Chapter 2

Determination of the Propertiesof Saline and Alkali Soils

This chapter discusses determinations that give in-formation on the chemical and physical properties ofsaline and alkali soils and thus serve as a basis for theirdiagnosis, treatment, and management. The status ofknowledge on this subject is such that it is not yet pas-sible to prepare a brief handbook containing a fewsimple measurements that will give all the necessary in-formation. A number of different types of measure-ments are presented. Some of these must be regardedas tentative and subject to change and improvement.In some cases alternate procedures are proposed, andthe individual worker will need to decide what kindand how many measurements will be required for theproblem at hand. The purpose, application, and inter-pretation of the various determinations are discussed inthis chapter. Detailed directions for making themeasurements are given in chapter 6.

Soil Sampling

There is no standard procedure for obtaining soilsamples for appraising salinity and alkali. Usually thedetails of procedure will depend upon the purpose forwhich the sample is taken. If the objective is to obtaina general evaluation of salinity in a given area, theaverage salt content of a number of samples providesan index for the over-all appraisal. The variationamong samples gives an index of the variation in saltcontent that may be encountered in the field. Thelarger the number of samples, the more accurate theappraisal will be. Too few samples may give a com-pletely erroneous index of the salinity status. Thedeviation between the actual conditions existing in anarea and the evaluation of the situation from thesampling procedure is designated as the “samplingerror.” It is evident that the larger the number ofsamples and the more carefully they are selected, thesmaller the sampling error will be.

Salt concentration in soils may vary greatly withhorizontal or vertical distance and with time. The na-ture of the soil, microrelief, and the cause and sourceof salinity should be considered. Factors that causemigration of salt, such as seasonal precipitation, irriga-tion, and phase in the crop cycle, should be taken intoaccount in relation to the time of sampling. In culti-vated areas, soil management history may be the most

important single factor in determining salinity status,and field boundaries may enter the problem of whereto sample and how to composite the samples.

The interpretation and use of salinity and alkalimeasurements necessarily depend on the completenessand accuracy of observational data recorded at the timeof sampling. A record of the species and condition ofthe plant cover is of particular importance. When at-tempting to correlate crop conditions in the field withsoil-salinity measurements, it is necessary to takesamples from the active root zone of the plants.

The following suggestions are offered on where andhow to sample:

(a)

(b)

(cl

(4

(e?

(f)

Visible or suspected salt crusts on the soilsurface should be sampled separately and theapproximate depth of sample recorded.If the soil shows evidence of profile develop-ment or distinct stratification, samples shouldbe taken by horizons or layers.In the absence of profile development or dis-tinct stratification, the surface samples (ex-cluding the surface crust) should be taken tothe plow depth, usually to a depth of 6 or 7inches.Succeeding samples may be taken at intervalsof 6 to 18, 18 to 36, and 36 to 7.2 inches, orother convenient depths, depending on thedepth of the root zone, the nature of the prob-lem, and the detail required.Sometimes soil samples taken for salinityand alkali determinations may be compositedto reduce analytical work.The size of samples will depend on the meas-urements that are to be made.

Detailed suggestions on taking and handling soilsamples along with a sample of the field data sheet usedat the Salinity Laboratory are given in Method 1.

Estimation of Soluble Salts From ElectricalConductivity

The choice of a method for measuring salinity de-pends on such things as the reason for making the meas-urements, the number of samples to be handled, and thetime and effort available for doing the work. Accurate

7

8 AGRICULTURE HANDBOOK 60, U. S. DEPT. OF AGRICULTURE

methods usually require more time and, therefore,limit the number of determinations.

Electrical-resistance measurements can be madequickly and accurately and have long been used for esti-mating soluble salts in soil (Whitney and Means, 1897) ;however, electrical conductance, which is the recipro-cal of resistance, is more suitable for salinity measure-ments, because it increases with salt content, thussimplifying the interpretation of readings. Moreover,expressing results in terms of specific conductance orconductivity makes the determination independent ofthe size and shape of the sample.

Electrical conductance is expressed in mhos, i. e.,reciprocal ohms, while electrical conductivity has thedimensions of mhos per centimeter. In this handbook,the symbol “EC” is used to represent electrical conduc-tivity.3

The salt content of the soil can be estimated roughlyfrom an electrical-conductivity measurement on a satu-rated soil paste or a more dilute suspension of soil inwater. A better estimate of soluble salts can be ob-tained from the conductivity of a water extract of thesoil. In general the higher the moisture content, theeasier it will be to obtain the extract, but the less repre-sentative the extracted solution will be of the solutionto which plant roots are exposed in the soil.

Soil so!utions in the field-moisture range can be ex-tracted for study and analysis by the displacementmethod (White and Ross, 1937) or with the pressure-membrane apparatus (Method 3d). These methods areused mainly for research and special chemical studies.

Plants in saline soil are responsive to the concentra-tion of the soil solution, and the relation of concentra-tion to the normal field-moisture range is sometimesoverlooked. There is more than a tenfold range in thewilting percentage of various soils. Consequently, theGeld-moisture range may vary greatly from one soil toanother. For example, a sand and a clay could havethe same soluble-salt content expressed as percent, dryweight basis, but the soil-solution concentration whennear the wilting percentage could be 10 times as highfor the sand as for the clay.

’ The standard unit for conductivity (mho,‘cm.) is a large unit,so that most solutions have a conductivity that is much less thanone unit. For instance, a measurement on one sample of waterfrom the Rio Grande at the Elephant Butte Dam gave EC=0.000694 mho/cm. For such cases, with physical and chemicalmeasurements, it is customary to choose a small subunit thatgives a more convenient location of the decimal point whenrecording or expressing data. For example, the unit ECXIO”is called the millimho per centimeter. This is a convenient,nractical conductivitv unit for most soil salinitv work. TJntilrecently ECXIO' (or KXlO") has been in common use.ECX 10” designates conductivity expressed in micromhos percentimeter. This is the unit most generally used for expressingthe conductivity of waters. The conductivity of the Rio Grandesample mentioned above, when expressed in these various units,is:

EC=0.000694 mho/cm.ECX 10:‘=0.694 millimho/cm.ECX 10”==69.4 (=KX lo”)EC X lo’=694 micromhos/cm.

Conductivity of the Saturation Extract and theSaturation Percentage

The conductivity of the saturation extract is recom-mended as a general method for appraising soil salinityin relation to plant growth. The method is somewhatless rapid than a res’istance measurement of the soilpaste, but the result is easier to relate to plant response.The procedure involves preparing a saturated soil pasteby stirring, during the addition of distilled water, untila characteristic endpoint is reached. A suction filteris then used to obtain a sufficient amount of the extractfor making the conductivity measurement.

The special advantage of the saturation-extractmethod of measuring salinity lies in the fact that thesaturation percentage is directly related to the field-moisture range. In the field, the moisture content ofthe soil fluctuates between a lower limit representedby the permanent-wilting percentage and the upper,wet end of the available range, which is approxi-mately two times the wilting percentage. Measure-ments on soils indicate that over a considerabletextural range the saturation percentage (SP) is ap-proximately equal to four times the 15-atmospherepercentage (FM), which, in turn, closely approximatesthe wilting percentage. The soluble-salt concentrationin the saturation extract, therefore, tends to be aboutone-half of the concentration of the soil solution at theupper end of the field-moisture range and about one-fourth the concentration that the soil solution wouldhave at the lower, dry end of the field-moisture range.The salt-dilution effect that occurs in fine-textured soils,because of their higher moisture retention, is thus auto-matically taken into account. For this reason, theconductivity of the saturation extract (EC,) can beused directly for appraising the effect of soil salinity onplant growth.

Table 1 gives some of the experimental data sup-porting the foregoing statements. Since the 15atmos-phere percentage appears to be the most significantmoisture property that can be readily measured, thisretentivity value was used to separate soil samples intothree textural groups: Coarse, medium, and fine (tableI). The FAP ranges arbitrarily selected to designatethese textural groups were: Coarse, 2.0-6.5; medium,6.6-15.0; and fine, greater than 15.1. The numbers inthe FAP column of table 1 are the actual FAP valuesfor the available samples in the various textural groups.The SP,/FAP ratio of the medium-textured group,which is largest in number, is approximately 4 and thestandard deviation is small; whereas the ratios forthe fine-textured and high organic matter groups aresomewhat lower (Campbell and Richards, 1950).

The saturation percentage for sands, when determinedby the standard procedure, gives values that, relativeto the field-moisture range, are higher than for othersoils. This occurs because in sands the large pores’ thatare filled with water at the saturation-paste conditiondo not correspondingly retain water under field condi-tions. Consequently, EC,X lo3 for sands, when re-ferred to the regular saturation-extract scale, gives an

SALINE AND ALKALI SOILS 9

T A B L E l .-Relation of saturadon percentage (SP) to 15atmosphere percentage (I’AP) as inf luencedtexture

by soil

T FAP

Soil group

coarse. ............Mediom ...........Fine ...............0rgamc.............

Soilsamples

Number

8:1118

Mini- Maxi- Aver- Mini- Maxi- Aver- Mini- Maxi- Aver-Ill”lIl mum age mum mum age mllm mUm age

3.4 6. 5 5.06. 6 14.2 10. 8

16. 1 21.0 18. 527. 6 51.3 37. 9

-.

16.0 43. 1 31. 8 4.6826.4 60.0 42. 5 3. 1541.8 78. 5 59. 5 2.0381.0 255 142 2.53-_ _.~

8. 45 6.375. 15 3.954.26 3.204. 97 3.66

1. 15

.z.75

optimistic index of salinity, i. e., underrates the salinity weight of a known volume of saturated paste has beencondition. Method 3b gives a tentative procedure for described by Wilcox (1951) and is included asestimating the upper limit of the field-moisture range. Method 27~.From this, a moisture content for extraction is deter- The endpoint for mixing a saturated soil paste ismined and a procedure for obtaining a conductivity reasonably definite; and, with a little training, goodvalue that can be used on the regular saturation-extract agreement can be obtained among various operators.scale is suggested. This new procedure is tentative Slight variations in technique, such as adding prac-because it has not been subjected to extensive testing,but it has given good results for soils with SP values

tically all the water to the soil sample before stirringor adding the air-dry soil to a known amount of water,

of approximately 25 or less. do not appreciably affect the saturation percentage ofIt would be more reliable to appraise salinity by using most soils. Special precautions, however, must be

measurements of extracts of the soil solution in the taken with very fine and very coarse textured soils.field-moisture range. However, difficulty of obtainingsuch extracts would make them prohibit&e for routine

For example, in some clay soils the amount of waterthat must be added to bring about saturation can be

use. The next higher feasible moisture content appears varied 10 percent or more, depending upon the rate ofto be the saturation percentage. The following scale adding water and the amount of stirring. The moreis recommended for general use in appraising the rapid the rate of water addition in relation to stirring,effect of soluble salts on crops. It shows the relation the lower the saturation percentage may be. The lowerof crop response to soil salinity expressed in terms value is desirable to reduce the time and effort duringof the conductivity of the saturation extract. mixing and also to minimize puddling of the soil.

Use of the conductivity of the saturation extract as Campbell and Richards (1950) found that the con-an index of soil salinity was introduced at the Rubidoux ductivity of the saturation-extract method is applicableLaboratory in 1939 for the Pecos River Joint Investi- also for the measurement of salinity in peat soils. Withgation. The salinity scale given in the earlier draft of air-dried peats, an overnight wetting period is necessarythis handbook was substantially the same as the scale to obtain a definite endpoint for the saturated paste.originally proposed by Scofield in his report on thePecos River Joint Investigation (United States National Relation of Conductivity to Salt Content andResources Planning Roard, 1942, pp. 263334). The Osmotic Pressurescale given here has been modified somewhat fromthose previously used.

The relation between the electrical conductivity and

It is often desirable, because of the extra informa-the salt content of various solutions is shown graphically

tion provided on soil texture and moisture retention,in several figures. The curves (fig. 2) for the chloride

to determine the soil-moisture content at saturation,salts and Na,SO, almost coincide, but MgSO,, CaSO,,and NaHCO, have lower conductivities than the other

i. e., the saturation percentage (SP) when saturatedsoil paste is prepared for salinity measurements. A

salts at equivalent concentrations. When the concen-tration is given in percent salt or parts per million,

rapid procedure for SP determination based on the the curves (fig. 3) are more widely separated.

___~ .~_____

Salinity effects mostlynegligible

Yields of very sensitive Yields of many cropscrops may be restricted restricted

Only tolerant cropsyield satisfactorily

Only a few very tolerantcrops yield satisfactorily

SP~.

SPfFAP

sStand-ard de-viation

0 2 4 8 16

Scale of conductivity (millimhos per centimeter at .25’ C.)259525 0 - 54 - 2


i

8

6

IO

8

6

0.2 .4 .6 .8 I 2 4 6 8 IO 2 4 0

CONDUCTIVITY - MILLIMHOS / C M . (ECxl03) A T 25O G.

FIGIJRE 2.-Concentration of single-salt solutions in milliequivalents per liter as related to electrical conductivity.

With soils from widely separated areas in westernUnited States, the concentration range was higher(fig. 4%) than that shown in figures 2 and 3 ; conse-quently, the electrical conductivity is expressed in mil-limbos per centimeter. This is a convenient unit to use .for extracts from saline soils. Soils represented bypoints that are considerably above the average lineusually contain a relatively high amount of calcium ormagnesium sulfate. Information on the salt contentof irrigation water in relation to electrical conductivityis given in chapter 5.

Experimental work conducted at the Salinity Labora-tory by Hayward and Spurr (1944), Wadleigh andAyers (1945)) and workers elsewhere indicates that theosmotic pressure of the soil solution is closely relatedto the rate of water uptake and growth of plants insaline soils. The osmotic pressure (OP) of solutionsexpressed in atmospheres is usually calculated fromthe freezing-point depression, in degrees C., AT, inaccordance with the relation, OPc12.06 A T -0.021 A T?, given in the International Critical Tables.

The relation between osmotic pressure and electrical

4

I

8

6

0

6


GONDUCTIVltY - YILLIMHOS /GM.(ECx103) AT es0 c.

FIGURE J.-Concentration of single-salt solutions in percent as related to electrical conductivity.


6 0 0 0

j 1 0 0 0

\ 0

6

1008

6

2 4 6 8 IO 2 4 6 0 100

C O N D U C T I V I T Y - MILLIMHOS/CM. (ECxiO3) A T 25O C .

FIGURF. 4.-Concentration of saturation extracts of soils in milliequivalents per liter as related to electrical conductivity.

SALINE AND

conductivity (fig. 5) is useful for some agriculturalpurposes. This measurement is in general use and canbe more readily measured than freezing-point depres-sion. The relation between OP and EC for salt mixturesfound in saline soils is indicated in figure 6 from datareported by Campbell and coworkers (1949). The OPvalues were calculated from freezing-point measure-ments. In the range of 6C that will permit plantgrowth, the relation OPcO.36 X EC, X 103 can be usedfor estimating the osmotic pressure of soil solutionsfrom conductivity measurements.

Conductivity of 1: 1 and 1: 5 Extracts

For soil: water ratios of 1: 1 and 1: 5, the extractis obtained by filtering without the use of vacuum orpressure. The conductivity of these extracts is some-times used for estimating salinity from the line infigure 4 or, preferably, from special curves that applyfor the salts and soil in question.

Salinity estimates based on the conductivity of 1: 1and 1: 5 extracts are convenient for rapid determina-tions, particularly if the amount of soil sample is lim-ited, or when repeated samplings are to be made in thesame soil to determine the change in salinity with timeor treatment. The reliability -of such estimates dependsupon the kind of salts present. For chloride salts, theresults will be only slightly affected by moisture con-tent, but, if sulfate or carbonate salts, which haverelatively low solubility, are present in appreciablequantities, the apparent amount of soluble salt will de-pend on the soil : water ratio (table 2). In an experi-ment conducted by Wadleigh, Gauch, and Kolisch(1951) to &termine the salt tolerance of orchardgrass,the salts shown in the table were individually addedto a loam soil. During the course of the experiment,many samples were taken to check distribution of thesalt in the soil and conductivity measurements weremade of the saturated soil (EC,), the saturation extract(EC,), the 1: 1 extract (EC,), and the 1: 2 extract( EC3). The regression coefficients, which are the slopesof the best fit straight lines, were calculated for variouscomparisons among the data (table 2) .

The theoretical values given in the table are basedon the saturation percentage of 30 for the soil used.Except for small changes in the activity coefficientsof the ions with dilution, the conductivity ratios should

ALKALI SOILS 13

be inversely proportional to the moisture contents of thesoil at extraction if the total dissolved salt is inde-pendent of the moisture content at which the extractionis made. The average measured conductivity ratioswere always greater than the theoretical. The dif-ferences were not large for the chloride salts, but whenNaHCO,, Na2S04, o r MgSO, were added to this soil.in which the exchange complex was largely saturatedwith calcium, some CaSO, and CaCO, were precipitated.It is evident from the table that the regression coeffi-cients are quite different for extracts obtained at highmoisture contents if the less soluble salts are present inthe soil. This example illustrates why the estimationof salinity from the conductivity of the extract at 1: 1or at higher moisture contents is not recommended forgeneral use. These higher moisture contents may beused to advantage in certain cases, but the limitations ofthe method should be clearly understood.

Salinity Appraisal From the Electrical Resistanceof Soil Paste

Salinity determinations based on the electrical resist-ance of a standard sample of wet soil have been in usefor many years (Whitney and Means, 1897; Briggs,1899). The Bureau of Soils cup and the data pub-lished by Davis and Bryan (1910) have been widelyused by various agencies in this country for es’timatingthe percentage of soluble salts in soils. The apparatusis simple and rugged, the measurements can be quicklymade, and the results are reproducible.

To obtain the relation between wet-soil resistanceand percent salt, Davis and Bryan made measurementsusing 4 soil samples representing the textural groupsof sand, loam, clay loam, and clay. These samples ofsoil were composited from various types of nonsalinesoils. A mixture of chloride and sulfate salts was usedto obtain 5 levels of added salt ranging from 0.2 to3 percent, and resistance values were obtained on thesaturated pastes. Making use of these 20 readings onthe synthetic soil and salt mixtures, Davis and Bryanused graphical interpolation to obtain the relation ofsoil-cup resistance to percent salt for mixed sulfates andchlorides. The Davis and Bryan procedure for theBureau of Soils method of determining soluble salt insoil is given in Method 5. The method is also describedin the Soil Survey Manual (1951, p. 343).

T ABLE 2.-Regression coejicients (b) between various criteria for evaluating soil salinity by a conductanceprocedure

Soils containing- b_m.m. bEc~.Ece b3-EC,

NaCl ......................................... 0.359 * 0.0070 0.185 + 0.0037 0.514 + 0.0069CaCl2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .356f . 011 .191+ .0028 534+ .0046M&l* . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .376h .OlO 192zk .0042 :507dz .012NaHC03. ..................................... .379zt .027 :227+ .017 .589f .OllNazSO, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .590f .023 .355+ .OlO .600f .OllMgSOa ........................................ .600f .068 .471f .060 .780f .027

Theoretical .................................... -333 . 167 .5

_

_

bEC..ECe

0.235 + 0.0066.242 f .0078.237i .019.222* .013.217f .015.226 f .0054

..,.._..........


‘ti

87

6

5

4

2 3 4 5 6 7 8 9 1 0 2 3 4 5 6 7 0

CONDUGTIVITY - MlLLIMHOS/CM.(ECxl03) A T 25O C .

FNXRE 5.-Osmotic pressure of single-salt solutions as related to electrical conductivity. (Data from International Critical Tables.)

3 0

6

SALINE AND ALKALI SOILS

0 . 6

I 3 6 IO 3 0 6 0

CONDUCTIVITY - MILLI MHOS/ CM.(ECx103) A T 25O C .

FICCRE 6.-Osmotic pressure of saturation extracts of soils as related to electrical conductivity.


TABLE 3.-Comparison of measured and culculated values of EC, after correctiqfor eflect of S P

Soils from-

L o w e r R i o Grande V a l l e y , T e x a s .G r a n d J u n c t i o n d i s t r i c t , C o l o r a d o .Tucumcari d i s t r i c t , N e w M e x i c o .G e m C o u n t y , I d a h o . .F o u r W e s t e r n S t a t e s .

A similar procedure was used by Davis and Bryanto obtain calibration data for “carbonate” s’alts, pre-sumably sodium carbonate. Tests at the Laboratory,however, indicate that table IV of Davis and Bryanfor carbonate salts is unreliable and should not be used.The unreliability of the calibration data for thes’e saltsis a result of cation-exchange reactions that were notgenerally understood at the time the original work wasdone.

The conductivity of the saturation extract (EC,) isrecommended in this handbook as a measurement forgeneral use for indicating soil salinity, but the methodbased on the soil-paste resistance (R,) is still com-monly used. The electrical conductivity of the soilpaste (EC,) is related to paste resistance by the relationEC,=O.25/R,, where 0.25 is the constant for theBureau of Soils electrode cup. In a study by Reitemeierand Wilcox (1946)) it was found that the relation be-tween EC, and EC, is markedly influenced by variationsin the saturation percentage, the salinity, and the con-ductivity of the soil minerals. From unpublished workat the Laboratory, Bower concluded that there is noeasy method for simplifying the relation of EC, (or R,)to EC,. He equilibrated a group of western soils withvarious concentrations of a 1: 1 mixture of sodium andcalcium chloride and found that on the averageEC,/EC,=5.4- 0.07 (SP) . Using this average rela-tionship and SP values calculated from the weight ofthe soil paste as described by Wilcox (1951)) he calcu-lated values for EC, based on R, measurements. Thedegree of correspondence between measured and cal-culated values is indicated by the data in table 3.

The calculated average values for EC, are somewhathigh but are acceptable except for the soils from GemCounty, Idaho. These soils had a low salinity level hutwere high in exchangeable sodium. The large dis-crepancy here and for some other locations apparentlyis owing to conduction by the clay minerals, when theycontain exchangeable sodium. Bower found, for ex-ample, that the electrical conductivity of a 5-percentsuspension of calcium-saturated montmorillonite was0.072 mmhos/cm., but when saturated with sodium,the conductivity was 0.446 mmhos/cm.

NO method has been found for improving the reli-ability of the paste-resistance method that does not de-stroy its simplicity. The method may be acceptable

6 9.9312 8.64II 9.63

7 5. 7312 10.25

Numberof

samples

Average EC, X lo3Standarddevia Lion

of

Measured Calculaleddiffe&ces

11.10 1. 17 1. 309.45 .81 .85

11.85 2.22 1. 1616.24 10.51 2 . 9 013.05 2.80 1. 76

for estimating salinity for purposes of soil classification,but for soils like those of Gem County, Idaho, it doesnot have acceptable reliability.

Conversion of Conductivity Data to a StandardReference Temperature

The electrical conductivity of solutions and of soilscontaining moisture increases approximately 2 percentper degree centigrade increase in temperature. Tosimplify the interpretation of salinity data, it is cus-tomary either to take the measurements at a standard-reference temperature or to determine the temperatureat which the measurement is made, and then, by meansof correction tables or a correction dial on the bridge, toconvert the measurement to a standard-reference tem-perature.

Whitney and Briggs (1897) measured the resist-ance of 9 soils at 13 temperatures and calculated theaverage relation of resistance to temperature. Whitneyand Means (1897) used these temperature data to con-struct a table used in converting resistance measure-ments of saturated soil to the standard temperature of60” F. Data from this table, which has been widelyused since its publication 50 years ago, are given intable 16 in chapter 6, along with instructions for its use.

More recently a study was made by Campbell, Bower,and Richards (1949) to determine the effect of tempera-ture on the electrical conductivity of soil extracts.Saturation extracts from 21 soils were measured at 5temperatures, ranging from 0” to 50” C. The tempera-ture coefficient of the electrical conductivity for theserepresentative soil extracts varied somewhat with tem-perature, but in the range from 15” to 35” it was veri-fied that for each degree centigrade increase in tem-perature the conductivity increased very nearly 2 per-cent of the value at 25”. The details of the proceduresfor measuring electrical conductivity and making tem-perature corrections are given in Method 4.

Comparison of Percent Salt in Soil and ExtractMeasurements

The diagram shown in figure 7 facilitates the inter-pretation of salinity in relation to crop response. It isbased on the following assumptions: P,,=p. p. m./10,000=0.064XECX103; P,,=(P,,XP,)/lOO; OP


OSMOTIC PRESSURE OF SATURATION EXTRACT - ATMOSPHERES5.76

CONDUCTIVITY OF SATURATION EXTRACT - MILLIMHOS/CM.

A 6 c D E

PLANT RESPONSEFIGURE 7.-Relation of the percent salt in the soil to the osmotic pressure and electrical conductivity of the saturation extract and to

crop response in the conductivity ranges designated by letters. These ranges are related to crop response by the salinity scaleon page 9.

= 0.36 X EC X 10”. P,,=percent salt in water; P,,=percent salt in soil; I’,= percent water in soil ; andOP= osmotic pressure in atmospheres. The lower scalegives values for the conductivity of the saturation ex-tract. The top scale shows the osmotic pressure of thesaturation extract. The osmotic pressure of the soilsolution at the upper limit of the field-moisture rangewill be approximately double these values.

The diagonal lines help correlate the conductivityof the saturation extract with the percent salt contentfor various soil textures. For example, at EC, X lo3 = 4,nearly all crops make good growth and for a soil witha saturation percentage of 75, as seen in the diagram,this corresponds to a salt content of about 0.2 percent.On the other hand, 0.2 percent salt in a sandy soil forwhich the saturation percentage is 25 would correspondto EC, X lo”= 12, which is too saline for good growthof most crop plants. Partial lists of crop plants intheir order of tolerance to soil salinity are given inchapter 4.

The diagram indicates the growth conditions of crops

to be expected for various degrees of salinity in theactive root zone of the soil, i. e., the soil volume that ispermeated by roots and in which moisture absorptionis appreciable. Obviously, the diagram does not applyfor soil in which salt has been deposited after the rootshave been established and have become nonabsorbing,or to soil adjacent to the plant, either high or low insalt, that has not been permeated by roots. With ma-ture row crops, for example, salt may have accumulatedin the ridge to such an extent that the roots no longerfunction as moisture absorbers and, therefore, the ridgecannot be considered as characteristic of the activeplant-root environment.

Chemical Determinations

Soil Reaction-pH

The pH value of an aqueous solution is the negativelogarithm of the hydrogen-ion activity. The value maybe determined potentiometrically, using various elec-trodes (Method 21), or calorimetrically, by indicators

18 AGRICULTURE HANDBOOK 6 0,

whose colors vary with the hydrogen-ion activity.There is some question as to the exact property beingmeasured when methods for determining the pH valuesof solutions are applied to soil-water systems. Appar-ent pH values are obtained, however, that depend onthe characteristics of the soil, the concentration of dis-solved carbon dioxide, and the moisture content atwhich the reading is made. Soil characteristics thatare known to influence pH readings include: the com-position of the exchangeable cations, the nature of thecation-exchange materials, the composition and con-centration of soluble salts, and the presence or absenceof gypsum and alkaline-earth carbonates.

A statistical study of the relation of pH readings tothe exchangeable-sodium-percentages of soils of aridregions has been made by Fireman and Wadleigh(1951). The effect of various factors such as moisturecontent, salinity level, and presence or absence ofalkaline-earth carbonates and gypsum upon this rela-tionship was also studied. Some of the more pertinentstatistical data obtained are presented in table 4. Whileall the coefficients of correlation given in the table arehighly significant, the coefficients of determinationshow that at best no more than 54 percent of the vari-ance in exchangeable-sodium-percentage is associatedwith the variance in pH reading. The data on the effectof moisture content indicate that the reliability of pre-diction of the exchangeable-sodium-percentage from pHreadings decreases as the moisture content is increased.Similarly, the data on the effect of salinity indicate thatthe reliability of prediction is lowest when the salt levelis either low or very high. An increase in pH readingof 1.0 or more, as the moisture content is changed froma low to a high value, has been found useful in someareas for detecting saline-alkali soils. However, thereliability of this procedure should be tested before useon any given group of soil samples.

CT. S. DEPT. OF AGRICULTURE

Experience and the statistical study of Fireman andWadleigh permit the following general .statements re-garding the interpretation of pH readings of saturatedsoil paste: (1) pH values of 8.5 or greater almost in-variably indicate an exchangeable-sodium-percentageof 15 or more and the presence of alkaline-earth carbo-nates; (2) the exchangeable-sodium-percentage of soilshaving pH values of less than 8.5 may or may not exceed15; (3) soils having pH values of less than 7.5 almostalways contain no alkaline-earth carbonates and thosehaving values of less than 7.0 contain significantamounts of exchangeable hydrogen.

Soluble Cations and Anions

Analyses of saline and alkali soils for soluble cationsand anions are usually made to determine the compo-sition of the salts present. Complete analyses for sol-uble ions provide an accurate determination of total saltcontent. Determinations of soluble cations are used toobtain the relations between total cation concentrationand other properties of saline solutions, such as electri-cal conductivity and osmotic pressure. The relativeconcentrations of the various cations in soil-water ex-tracts also give information on the composition of theexchangeable cations in the soil.

The soluble cations and anions commonly deter-mined in saline and alkali soils are calcium, magnesium,sodium, potassium, carbonate, bicarbonate, sulfate, andchloride. Occasionally nitrate and soluble silicate alsoare determined. In making complete analyses, a de-termination of nitrate is indicated if the sum of cationsexpressed on an equivalent basis significantly exceedsthat of the commonly determined anions. Appreciableamounts of soluble silicate occur only in alkali soilshaving high pH values. In analys’es made by the usualmethods, including those recommended in this hand-

TARLE 4.-Coeficient of correlation (r)l and coeficient of determination (9) for the relation of pH reading toexchangeable-sodium.-percentage as influenced by moisture content, salinity level, and presence or absence ofalkaline-earth carbonates and gypsum

Moisture Salinity a8 Blkaline-content EC, x 103 earth

(percent) at 25’ C. carbonates

Saturation. .........500 . . . . . . . . . . . . . . . .1 , 0 0 0 . . . . . . . . . . . . . .6 , 0 0 0 . . . . . . . . . . . . . .

S a t u r a t i o n .D o .D o .D o .D o .Do.Do.Do. .

Variable. ............ do . . . . . . . . . ..... do . . . . . . . . . ..... do . . . . . . . . . .

o-4 . . . . . . . . . . . . .4 - 8 . . . . . . . . . . . . .8-15 . . . . . . . . . . . .

15-30 . . . . . . . . . . . .> 3 0 . ...........Variable. ............ do . . . . . . . . . ..... do . . . . . . . . . .

V a r i a b l e .d o . . . . . . . . . . . .d o . . . . . . . . . . . .d o .

d o . . . . . . . . . . . .d o . . . . . . . . . . . .d o . . . . . . . . . . . .do.d o . . . . . . . . . . . .

P r e s e n t ..do. _.

Absent.

Gypsum Samples r

V a r i a b l e .,. .do.

d od o . . . . . . . . . . . .

.2:::::::Y:::d o . . . . . . . : . . . .d o . . . . . . . . . . . .do,

P r e s e n tA b s e n t .

do ,

Number868271289346

349

1::

ti

237452154

0. 66.65

:G

.56

.z74

: 49

:E.41

z1495424

:i17

1 All values are significant at the l-percent level.


book, any soluble silicate present is determined ascarbonate.

As shown by Reitemeier (1946) and others, valuesobtained from determinations of the soluble-cation andsoluble-anion contents of saline and alkali soils aremarkedly influenced by the moisture content at whichthe extraction is made. The total dissolved quantitiesof some ions increase with increasing moisture content,while concurrently those of others may decrease; al-most invariably values obtained for total salt contentincrease with increasing moisture content at extraction.Processes that are responsible for the changes in therelative and total amounts of soluble ions which occurwith increasing moisture content include cation-ex-change reactions, negative adsorption of ions, hydrol-ysis, and the increased solution of silicate minerals,alkaline-earth carbonates, and gypsum. Ideally, thedetermination of soluble ions should be made on ex-tracts obtained at a moisture content in the field-moisture range. However, the preparation of suchextracts is time-consuming and requires the use ofspecial extraction equipment (Method 3d). Saturationpercentage is the lowest practical moisture content forobtaining extracts on a routine basis. Use of thesaturation extract is, therefore, recommended for thedetermination of soluble ions. Methods are availablethat permit determination of the electrical conductivityand the common soluble constituents on 10 to 50 ml.of saturation extract. As a rule, about one-fourth ofthe moisture in a saturated soil paste can be removedby ordinary pressure or vacuum filtration.

The choice of methods for the determination of thevarious cations and anions depends upon the equip-ment available and the personal preference of theanalyst. No attempt is made here to present all of themethods that are suitable. The methods given werechosen on the basis of their convenience and reliability.Owing to the fact that the amount of extract availablefor analysis is usually limited, most of the methodsselected are of the semimicro type. They generally in-volve the use of a centrifuge, a flame photometer, and aphotoelectric calorimeter. Where the amount of ex-tract is not limited, the macromethods employed forwater analysis given in chapter 8 may be used. Most ofthese methods do not require the use of a centrifuge orphotoelectric calorimeter.

Soluble Boron

The importance of soluble boron from the standpointof soil salinity lies in its marked toxicity to plants whenpresent in relatively small amounts. Toxic concentra-tions of boron have been found in the saturation ex-tracts of a number of saline soils. It is necessary,therefore, to consider this constituent as a factor in thediagnosis and reclamation of saline and alkali soils.High levels of boron in soils can usually be reduced byleaching. During the leaching process, boron may notbe removed in the same proportion as other salts. If theconcentration of boron is high at the outset, a consider-able depth of leaching water may be necessary to reduce

the boron content to a safe value for good plant growth.This is illustrated by a recent leaching test. At thebeginning of the test, the conductivity of the saturationextract of the top 12 inches of soil was 64.0 mmhos/cm.After 4 feet of irrigation water had passed through thesoil, the conductivity was reduced to 4.2 mmhos/cm.;after 8 feet, the conductivity was 3.4 mmhos/cm.; andafter I2 feet, it was 3.3 mmhos/cm. The concentrationof boron in the saturation extract at the start of the testwas 54 p. p. m. After the passage of 4 feet of water,the concentration was 6.9 p. p. m.; after 8 feet, it was2.4 p. p. m.; and after 12 feet, it was 1.8 p. p. m. Thus,leaching with 4 feet of water reduced the salinity to asafe level, but the boron content was still too high forgood growth of plants sensitive to boron.

Permissible limits for boron in the saturation extractof soils can at present be given only on a tentative basis.Concentrations below 0.7 p. p. m. boron probably aresafe for sensitive plants (ch. 4) ; from 0.7 to 1.5 p. p. m.boron is marginal; and more than 1.5 p. p. m. boronappears to be unsafe. The more tolerant plants canwithstand higher concentrations, but limits cannot beset on the basis of present information. For land onwhich crops are being grown, a better appraisal ofboron conditions often can be made by an analysis ofplant samples (ch. 4) than can be obtained from ananalysis of soil samples.

Exchangeable Cations

When a sample of soil is placed in a solution of asalt, such as ammonium acetate, ammonium ions areadsorbed by the soil and an equivalent amount ofcations is displaced from the soil into the solution.This reaction is termed “cation exchange,” and thecations displaced from the soil are referred to as “ex-changeable.” The surface-active constituents of soilsthat have cation-exchange properties are collectivelytermed the “exchange complex” and consist for the mostpart of various clay minerals and organic matter. Thetotal amount of exchangeable cations that a soil canretain is designated the “cation-exchange-capacity,” andis usually expressed in milliequivalents per 100 gm. ofsoil. It is often convenient to express the relativeamounts of various exchangeable cations present in asoil as a percentage of the cation-exchange-capacity.For example, the exchangeable-sodium-percentage(ESP) is equal to 100 times the exchangeable-sodiumcontent divided by the cation-exchange-capacity, bothexpressed in the same units.

Determinations of the amounts and proportions ofthe various exchangeable cations present in soils areuseful, because exchangeable cations markedly in&r-ence the physical and chemical properties of soils. Theexchangeable-cation analysis of saline and alkali soilsis subject to difficulties not ordinarily encountered withother soils, such as those from humid regions. Salineand alkali soils commonly contain alkaline-earth carbo-nates and a relatively high concentration of solublesalts. They may have low permeability to aqueoussolutions and to alcohol. Solutions capable of displac-


ing exchangeable cations from soils dissolye most or allof the soluble salts and significant amounts of the carbo-nates of calcium and magnesium if they are present.The soluble salts should not be washed out of the soilprior to extracting the exchangeable cations, because ofsignificant changes that take place as a result of dilutionand hydrolysis. The dissolving of salts, therefore,necessitates independent determinations of soluble-cation contents and correction of the exchangeable-cation analysis for their presence, while the occurrenceof calcium and magnesium carbonates prevents accuratedetermination of exchangeable calcium and magnesium.Furthermore, the low permeability of many alkali soilsrenders the conventional leaching techniques for dis-placement of cations time-consuming and inconvenient.

Neutral normal ammonium acetate is the salt solutionmost commonly used for the extraction of exchangeablecations and for the saturation of the exchange complexin the determination of cation-exchange-capacity. Al-though this solution has many advantages for exchange-able-cation analysis, some saline and alkali soils fix ap-preciable amounts of ammonium as well as potassiumions under moist conditions. The fixation of ammo-nium does not interfere with the extraction of exchange-able cations, but values obtained for cation-exchange-capacity by ammonium saturation are low by amountsequal to the quantity of ammonium fixed. The desir-ability of using a cation not subject to fixation for thedetermination of cation-exchange-capacity is, therefore,evident.

As discussed in a previous section, the determinedvalues for the soluble-ion contents of soils vary withthe moisture content at which the extraction is made.liecause equilibria exist between the soluble and ex-changeable cations in soils, the changes in relative andlotal concentrations of soluble cations with variationsin moisture content are accompanied by changes in therelative composition of the exchangeable cations. Ina strict sense, therefore, values for exchangeable-cationcontents apply only at the moisture content used for theextraction of soluble cations. Owing to difficulties in-volved in the determination of soluble cations at mois-ture contents in the field range, it is convenient to deter-mine exchangeable-cation contents at the saturationpercentage.

Consideration of the various factors involved in thedetermination of the exchangeable cations and thecation-exchange-capacity of saline and alkali soils hasled to the adoption of the following scheme ofanalysis :

(ai Extract a sample of the soil with an excess ofneutral normal ammonium acetate solution and deter-mine the milliequivalents of the various cations removedper 100 gm. of soil.4

(b) Prepare a saturation extract of the soil and de-termine the milliequivalents of the various solublecations per 100 gm. of soil.”

(c) Calculate the exchangeable-cation contents ofthe soil by subtracting the amounts of the various cat-ions dissolved in the saturation extract from the

amounts extracted by the ammonium acetate solution.(d 1 Determine the cation-exchange-capacity by

measuring the milliequivalents of sodium adsorbed per100 gm. of soil upon treating a sample with an excessof normal sodium acetate solution of pH 8.2.

The difficulties encountered in leaching soil samplesof low permeability are overcome by shaking andcentrifuging samples in centrifuge tubes with successiveportions of the extraction and wash liquids. Neutralnormal ammonium acetate solution is used for the ex-traction of exchangeable plus soluble cations, becauseits interference in analytical procedures is easily elimi-nated. Of the common cations, sodium appears to bethe most suitable for determining cation-exchange-capacity. As mentioned previously, ammonium andpotassium are subject to fixation in difficultly exchange-able form and the usual presence of calcium and mag-nesium carbonates in saline and alkali soils precludesthe use of extractants containing calcium or magnesium.The fact that sodium is a prominent cation in mostsaline and alkali soils also favors its use in the determi-nation of cation-exchange-capacity (Method 19).

Gypsum

Gypsum is found in many soils of arid regions, inamounts ranging from traces to several percent. Insome soils, gypsum was present in the sedimentary de-posits from which the soil was derived; whereas, inother soils the gypsum was formed by the precipitationof calcium and sulfate during salinization. Owing toleaching, gypsum commonly occurs at some depth in theformer instanceusually greatest fn

while in the latter its content isthe surface layers of the soil.

Information regarding the gypsum content of alkalisoils is important, because it usually determines whetherthe application of chemical amendments will be re-quired for reclamation. Also, the presence of con-siderable amounts of gypsum in the soil might permitthe use of an irrigation water having an unfavorablyhigh sodium content.

The precise determination of gypsum in soils isdifficult, because of inherent errors involved in theextraction of this mineral by water. Studies by Reite-meier ( 194,6) and others show that at least three factorsother than the solution of gypsum may influence theamounts of calcium and sulfate extracted from gypsif-erous soils. They are: (1) The solution of calciumfrom sources other than gypsum ; (2) exchange reac-tions in which soluble calcium replaces other cations,such as sodium and magnesium; and (3) the solutionof sulfate from sources other than gypsum.

4 If the soil is known to contain carbonates of calcium andmagnesium, determination of these cations is omitted. Like-wise, if the soil is known to contain gypsum not complete!ysoluble in the saturation extract, the determination of calciumis omitted. In the absence of prior knowledge fegarding thecalcium and magnesium carbonate and gypsum contents of thesoil, the calcium and magnesium determinations are disregardedif upon completion of the exchangeable-cation analysis the sumof the values obtained for exchangeable-cation contents is foundto exceed the cation-exchange-capacity value.


Three methods are given in chapter 6 for the esti-mation of gypsum in soils. Methods 22a and 22b arebased on the low solubility of the salt in an aqueoussolution of acetone. Method 22a is essentially quali-tative, although a rough estimate of gypsum contentmay be obtained by visual observation of the amountof precipitate obtained. This method can be success-fully employed under field conditions. In Method 22bthe separated and washed gypsum precipitate is deter-mined quantitatively. The use of Method 22c is ad-vantageous when characterization of the soil includesthe determination of calcium plus magnesium in thesaturation extract. It is based on the increase insoluble-divalent-cation content as the moisture contentof the soil is increased from the saturation percentageto a moisture content sufficient to dissolve the gypsumpresent. It should be noted that this method can givenegative values for gypsum content as a result of thereplacement of exchangeable sodium and potassium bycalcium as the moisture content of the soil is increased.This is likely to occur only in alkali soils containinglittle or no gypsum.

Alkaline-Earth Carbonates (Lime)

The alkaline-earth carbonates that occur in signifi-cant amounts in soils consist of calcite, dolomite, andpossibly magnesite. Owing to low rainfall and limitedleaching, alkaline-earth carbonates are usually a con-stituent of soils of arid regions. The amounts presentvary from traces to more than 50 percent of the soilmass. Alkaline-earth carbonates influence the textureof the soil when present in appreciable amounts, forthe particles commonly occur in the silt-size fraction.The presence of fine alkaline-earth carbonate particlesis thought to improve the physical condition of soils.Conversely, when alkaline-earth carbonates occur ascaliche or as cementing agents in indurated layers, themovement of water and the development of root systemsis impeded. Alkaline-earth carbonates are importantconstituents of alkali soils, for they constitute a poten-tial source of soluble calcium and magnesium for thereplacement of exchangeable sodium. &4s discussed inanother section, the choice of chemical amendmentsfor the replacement of exchangeable sodium is directlyrelated to the presence or absence of alkaline-earthcarbonates.

Effervescence upon application of acid (Method 23a)can be used to detect as little as 0.5 percent of alkaline-earth carbonates in soils. This test suffices for mostpurposes. When a better estimate of the alkaline-earth-carbonate content of soils is desired, Methods 23b or23c may be used. A quantitative determination ofsmall amounts of alkaline-earth carbonates in soils issometimes desirable in connection with proposed appli-cations of acid-forming amendments. For precise de-terminations, the reader is referred to the methods ofWilliams (1949) and Schollenberger (1945).

Physical Determinations

The problem of evaluating soil physical conditionshas recently been separated into components by theAmerican Society of Agronomy (1952) ; and they arediscussed under the headings of mechanical impedance,aeration, soil water, and soil temperature. These arelogical ultimate aspects; but, for practical work onalkali soils, measuring methods are needed that yieldimmediate results having more or less direct diagnosticsignificance. Some progress is being made towardevaluating the physical status of soil in terms of physi-cal properties, i. e., intrinsic qualities of soil that canbe expressed in standard units and that have valueswhich are substantially independent of the method ofmeasurement. Infiltration rate, permeability, bulkdensity, pore-size distribution, aggregation, and modu-lus of rupture appear to be such properties. Experi-ence indicates that the physical status of any given soilis not static. There is a range of variation of physicalstatus that is related to productivity, and this is re-flected in corresponding ranges in the values of per-tinent physical properties.

Information on the existing physical status of aproblem soil is useful for purposes of diagnosis orimprovement, but it might also be useful to know howmuch better or worse the status can be made by chemi-cal and physical treatments simulating those applicableunder field conditions. Soils can be treated to increasethe exchangeable-sodium-percentage and then puddledto indicate how unfavorable the physical status can bemade. It should also be possible by use of soil amend-ments and chemical aggregants to get some indicationof how favorable the physical status can be made.Practical use of the concept that there is a range ofphysical states for any given soil may have to wait forrefinements in measuring methods, but the idea seemsto be pertinent to the improvement of alkali soils.

Infiltration Rate

Water-movement rates attainable in soil under fieldconditions relate directly to irrigation, leaching, anddrainage of saline and alkali soils. Infiltration refersto the downward entry of water into soils and the term“infiltration rate” has special technical significance insoils work. Definitions of soil-water terms adopted bythe Soil Science Society of America (1952) are fol-lowed, and are included in the Glossary.

The infiltration rate of soil is influenced by suchfactors as the condition of the soil surface, the chemicaland physical status and nature of the soil profile, andthe distribution of water in the profile. ,411 of thesefactors change more or less with time during infil-tration.

The infiltration rate is measured under field condi-tions. The principal methods used have involved flood-ing or impounding water on the soil surface, sprinklingto simulate rain, and measuring water entry from rillsor furrows. In addition to the multitude of localphysical conditions that are encountered in the field,


the availability of equipment, materials, and serviceswill largely decide what method to use in measuring in-filtration. Although many measurements have beenmade, as evidenced by the extensive bibliography ofDavidson (1940)) there does not seem to be a generallyaccepted procedure applicable to all situations. Manyof the infiltration measurements made by this Labora-tory have been in connection with basin irrigation ontest plots ranging from 10 to 20 feet square. Thewater-subsidence rate in a large plot is probably thebest indication of the infiltration rate as related toleaching operations, but this method is usually notfeasible for exploratory or diagnostic measurements innew areas. The cylinder method of Musgrave (1935) isprobably the most versatile of the various methodsavailable. A guard ring is needed if lateral spreadingis excessive. Procedures for making infiltration meas-urements are given in Method 28.

Water having the same quality as that which will beused for irrigation or leaching must be used for infil-tration tests in the field, otherwise the results may bemisleading. The length of time the tests should be con-ducted and the depth of water to be applied dependupon the purpose of the test and the kind of informationthat is sought. If it is a matter of appraising an irriga-tion problem, the depth corresponding to one irrigationmay be sufficient; but, if information on infiltration forplanning a leaching operation is needed, it may be de-sirable to apply the full depth of leaching water to atest plot. It often happens that subsurface drainage issufficiently restricted to cause the infiltration rate todecrease considerably with time. It should be kept inmind, therefore, that although small area tests will giveuseful information on soil changes during leaching, theinfiltration values thus obtained will apply to large areasonly if underdrainage is not limiting.

Experience indicates that the infiltration rate of agiven soil can be high or low, depending on physicalstatus and management history. Infiltration rate isoften critically influenced by surface soil conditions,but subsurface layers also are sometimes limiting.Water distribution in the profile and depth of waterapplied are modifying factors. The infiltration ratecan be undesirably high or undesirably low. It is thelow end of the range that may be a critical limitingfactor in the agricultural use of alkali soils. It is diffi-cult to specify a boundary limit between satisfactoryand unsatisfactory infiltration rates at the low end ofthe range, because so many factors are involved, in-cluding the patience and skill of the farmer. However,if the infiltration rate is less than 0.25 cm./hr. (0.1in./hr.) special water-management problems are in-volved that may make an irrigation enterpriseunprofitable for average operators.

Permeability and Hydraulic Conductivity

The permeability of soil, in a qualitative sense, refersto the readiness with which the soil conducts or trans-mits fluids. In a quantitative sense, when permeabilityis expressed with numbers, it seems desirable that per-

meability be defined as a property of the porous mediumalone and independent of the fluid used in its measure-ment. The term “hydraulic conductivity,” on the otherhand, is used to refer to the proportionality factor inthe Darcy flow equation. These distinctions representincreased specialization in the use of these terms as ap-proved by the Soil Science Society of America ( 1952).No change in the qualitative use of the word “permea-bility” is involved. In the quantitative sense, involvingnumerical values, the term “intrinsic permeability”will mostly be used and will refer to a length-squaredmeasurement that may be identified in a general wayto the cross-sectional area of some equivalent or effec-tive size of pore.

An immediate consequence of this clarifitiation ofnomenclature is a new method for evaluating pore-spacestability or structural stability of soil. For porousmedia with fixed structure, such as sandstone or firedceramic, measurements of intrinsic permeability withair, water, or organic liquids all give very nearly thesame numerical value. Gravity, density, and the vis-cosity of the liquid are taken into account in the flowequation. However, if the intrinsic permeability fora soil as measured with air is markedly greater than thepermeability of the same sample as subsequently meas-ured using water, then it may be concluded that theaction of water in the soil brings about a change instructure indicated by the change in permeability. Theratio of air to water permeability, therefore, is a meas-ure of the structural stability of soils, a high ratioindicating low stability.

Intrinsic-permeability measurements are based onthe equation v=k’dgi/~, where v is the flow velocity,k’ is the intrinsic permeability, d is the density of thefluid, g is the acceleration of gravity, i the hydraulicgradient, and 71 is the viscosity. Procedures for measur-ing intrinsic permeability with gases and liquids aregiven as Methods 37a and 37b. The air-water per-meability ratio increases greatly as the exchangeable-sodium content of the soil increases, indicating thatexchangeable sodium decreased the water stability ofthe soil structure.

It is seen from the Darcy equation, v=ki, that k ,the hydraulic conductivity, is the effective flow velocityor discharge velocity of water in soil at unit hydraulicgradient, i. e., when the driving force is equal to 1gravity. Methods 34a and 34b give procedures formeasuring hydraulic conductivity on undisturbed anddisturbed soil samples.

Under some circumstances, especially when the soilsurface has been subject to submergence by water fora considerable period and when the hydraulic con-ductivity is nearly uniform with depth, the hydraulicgradient beneart the soil surface may approach unity,i. e., the downward driving force is composed entirelyof the gravity force with no pressure gradient. Underthis condition the infiltration rate is equal to the hy-draulic conductivity, but this is probably the exceptionrather than the rule under field conditions. Conse-quently, the relation between infiltration rate andhydraulic conductivity is not a simple one. For ex-


ample, at the Malheur Experimental Area in Oregon,very low hydraulic-conductivity values were obtainedand yet infiltration was adequate to support good cropswith sprinkler irrigation. It was found by use oftensiometers that values for the hydraulic gradientduring infiltration ranged up to 10 in some cases. Thiss.oil was deep and silty and the suction gradient in thesoil added significantly to the rate of downward move-ment of water. If the downward flow is interruptedby a layer of very low conductivity, then the hydraulicgradient may approach zero as the soil pores becomefilled and the condition of static equilibrium undergravity is approached.

It is to be expected that if the hydraulic conductivityof surface soil is as low as 0.1 cm./hr. (0.04 in./hr.)leaching and irrigation may present serious difficulties.Irrigation agriculture under average conditions ofmanagement skill, water quality, and drainage condi-tions would have doubtful success unless the hydraulicconductivity could be increased appreciably by soil-improvement measures.

Moisture Retention by Soil

The effect of soil salinity on crops is related to therange over which the moisture content of the soil varies,because the concentration of the soil solution dependsboth on the amount of soluble salt and the amount ofwater present. The permanent-wilting percentage, asindicated in the review by Veihmeyer and Hendrickson(1948)) is generally accepted as being the lower limitof water available for plant growth in nonsaline soil.For all practical purposes, the 15-atmosphere percent-age (Method 31) can be used as an index of thepermanent-wilting percentage and, therefore, also asan acceptable index of the lower limit of the availablerange of soil moisture. This lower limit appears to bean intrinsic property of the soil that is largely deter-mined by soil texture and appears to be substantiallyindependent of the kind of plant grown on the soil.

It is much more difficult to set an upper limit forthe range of water content available to plants in thefield. In addition to dependence upon soil texture atthe point in question, the upper limit depends also onthe variation throughout the profile of such factors aspore-size distribution and water conductivity. The dis-tribution of water with depth influences the hydraulicgradient, and, therefore, also the rate of downwardmovement of water. For example, with or withoutactive roots, the moisture content in the surface layersof a deep permeable soil will decrease more slowly ifthe profile is deeply wetted than if only a shallow depthis wetted and the underlying soil is dry. Also, the totalamount of water actually available from any given layerof surface soil depends on the rooting depth and trans-piration rate of the crop. The hydraulic boundary con-ditions that characterize the field situation would beextremely difficult to reproduce for a soil sample re-moved from the profile, and it is not surprising that nogenerally satisfactory laboratory method has beenfound for estimating the upper limit of water available

for crop growth under field conditions. A field de-termination under representative field conditions is thebest method for obtaining the upper limit of the field-moisture range.

For most medium- to fine-textured soils, the upperlimit of available water is approximately twice themoisture percentage of the lower limit. This does nothold true for the coarse-textured soils. It has beenfound by the United States Bureau of Reclamation(1948) that for the sandy soils occurring on the YumaMesa, Arizona, the water retained in a sample of soilat the yro-atmosphere percentage (Method 29) satis-factorily approximates the upper limit of availablewater under field conditions.

Density and Porosity

The bulk density (apparent density) of soil is themass of soil per unit volume, and the porosity of soilis the fraction of the soil volume not occupied by soilparticles. Bodman (1942) has discussed soil densityin connection with water content and porosity relation-ships and has prepared useful nomograms (fig. 8).

The bulk density of soil can be measured by severalmethods. For a certain range of moisture contents withsoils that are comparatively free of gravel and stones,it is possible to press into the soil a thin-walled tubehaving a suitable cutting edge. The soil is thensmoothed at each end of the tube and oven-dried at105” C. The bulk density is the mass of soil containedin the tube divided by the volume of the tube, as indi-cated in Method 38.

The porosity of soil (n) may be obtained directlyfrom air-pycnometer measurements or can be calcu-lated from the relation II= (d, - &) /a&, where d, is theaverage density of the soil particles and & is the bulkdensity.

The particle density of many soils averages around2.65 gm. cm.?. The average particle density for peatsoils or for pumice soils is much lower. Direct meas-urements of particle density can be made with pycnom-eter bottles (Method 39).

The bulk density of most soils ranges from 1.0 gm.cm.? for clays, to 1.8 gm. cm.? for sands. This corre-sponds to the range of 62.4 to 112 lb. ft.?. The corre-sponding porosity range will be from about 0.60 to 0.30.Bulk density may become a critical factor in the pro-ductivity of soil. Veihmeyer and Hendrickson (1946)found that plant roots were unable to penetrate agravelly loam soil when the bulk density exceeded avalue of around 1.8 gm. cm.?. Also, when the bulkdensity of medium- to fine-textured subsoils exceedsabout 1.7 gm. cm.-3, hydraulic conductivity values willbe so low that drainage difficulties can be anticipated.

Aggregation and Stability of Structure

The arrangement of soil particles into crumbs or ag-gregates that are more or less water stable is an im-portant aspect of soil structure. Alkali soils oftenhave a dense, blocky, single-grain structure, are hard totill when dry, and have low hydraulic conductivity when

WATERCONTENTPERCENT

so

40

i

30i-

i*ot

10

9

81

7 -

5 --

5

4

3 --

2 --

I -

A

DEPTH OF WATERPER F00T0FSolL

‘F FEET

1

100I I .9010 .SO9

70S

7 6 0

6 SO

in .10

.09

.OS0 9

0 70.8

0.7 .OG

0.6 0 5

a5 1 .04

04.03

0.3

t .02

I-9014 y-

1.3 -_-- 80

1.2 --

- - 7 0I., - -

1 .0

?i50

TOTALPORE SFWE

POROSITY 1 \IWWiATIO0 0

0.1 0.1a2

0 . 20.3

0.3 0.4

a50.6

0.4

0.8

0.5I1.0

0.6 IS

2.0

0.72.5

3 0

55

0.8 4.0

D E N S I T Y A N D S O I L W A T E R D E N S I T Y A N D S O I L A I R S P A C E

FIGURE 8.-Nomograms giving soil density, soil water, and soil air space relationships (Bodman, 1942).

REALDENSITYGKJCM

9! 2.9 2.7 2.6 2.3 2.4 2.5 26 2. 2.2 2.0 I

u

!+

p $ 2 g 2 $

z

3.0B


wet. This is generally because the aggregates and alsothe pores of such soils are not stable. The aggregatesslake down in water, and the pores become filled withfine particles.

Several methods have been proposed for measuringthe water stability of soil aggregates, the most commonbeing the wet-sieving method proposed by Yoder(1936). A modification of the Yoder procedure isgiven as Method 42a. Soils that are low in organicmatter and contain appreciable amounts of exchange-able sodium seldom contain aggregates of larger sizesand for that reason measuring procedures adapted forthe smaller aggregates are included as Method 42b.This determination is related to Middleton’s (1930)“dispersion-ratio,” but Method 42a gives the percentageby weight of particles smaller than 50~ that are boundinto water-stable aggregates greater than 50,. Insuf-ficient data are available at the present time to specifylimits that will help to distinguish between problem andnonproblem soils as far as aggregate-size distribution isconcerned.

Childs (1940) followed the change in moisture-reten-sion curves with successive wettings to get an index ofthe stability of structure, or, more precisely, the stabil-ity of the pore-space arrangement. Reeve and co-workers (fig. 1) have shown that the ratio of the airpermeability to the water permeability for soils is alsoa useful index of the stability of soil structure (Method37).

Recent studies by Allison (1952) and by Martinand associates (2952) indicate that dispersed soils maybe rapidly and effectively improved by application ofaggregating agents of the polyelectrolyte type. Ap-plied at the rate of 0.1 percent on the dry-soil basis,this material has effectively improved the physicalcondition of alkali soils on which it has been tried.Salinity appears to have little or no effect on the process.A higher degree of aggregation was obtained wherethe aggregating agent in solution was sprayed on drysoil and mixed in than when it was applied dry to amoist soil followed by mixing. Regardless of the man-ner of application, large increases in infiltration rateand hydraulic conductivity resulted from its use.

Although not yet economically feasible for generalagricultural use, aggregating agents can be an effectiveresearch tool for investigational work with saline andalkali soils. By their use, for instance, plant responseto different levels of exchangeable sodium or differentCa : Na ratios may be studied on “conditioned” soilsin the absence of poor structure and accompanyingconditions of deficient aeration and low water-move-ment rates ordinarily present in alkali soils,

It seems likely, also, that soil-aggregating chemicalsmay provide a rapid method for appraising the struc-tural improvement potentially attainable from organic-matter additions. Organic-matter additions, whileslower to give results, have long been used in agricul-ture. There may be soils, such as those high in siltand low in clay, in which coarse organic matter maygive improvements in physical condition that are unat-tainable with chemical aggregants.

259525 0 - 54 - 3

Crust Formation

Soils that have low stability of structure disperseand slake when they are wetted by rain or irrigationwater and may develop a hard crust as the soil surfacedries. This crust presents a serious barrier for emerg-ing seedlings, and with some crops often is the maincause of a poor stand. Alkali soils are a special prob-lem in this regard, but the phenomenon is by no meanslimited to these soils.

Factors influencing development of hard surfacecrusts appear to be high exchangeable sodium, loworganic matter, puddling, and wetting the soil to zerotension, which occurs in the field with rain or irriga-tion. Crust prevention would, therefore, involve re-moval of exchangeable sodium, addition of organicmatter, and care to avoid puddling during tillage andother operations. Where possible, the placement ofthe seed line somewhat above the water level in a fur-row is desirable so that the soil above the seed will bewetted with water at appreciable tensions, thus lessen-ing the tendency for soil aggregates at the surface todisintegrate.

The procedure for measuring the modulus of ruptureof soil (Method 43) was developed for appraising thehardness of soil crusts, since a satisfactory measuringmethod is essential in developing and testing soiltreatments for lessening soil crusting.

Choice of Determinations andInterpretation of Data

Equilibrium Relations Between Soluble andExchangeable Cations

Cation exchange can be represented by equationssimilar to those employed for chemical reactions insolutions. For example, the reaction between calcium-saturated soil and sodium chloride solution may bewritten : CaX, + 2NaClti2NaX + CaCl,, where X desig-nates the soil exchange complex. As shown by theequation, the reaction does not go to completion, be-cause as long as soluble calcium exists in the solutionphase there will be adsorbed calcium on the exchangecomplex and vice versa. Equations have been pro-posed by various workers for expressing the equilibriumdistribution of pairs of cations between the exchange-able and soluble forms. For metallic cation pairs ofequal valence, many of the equations assume the sameform and give satis’factory equilibrium constants, butvariable results are obtained with the different equationswhen cations of unequal valence are involved. Accord-ing to the work of Krishnamoorthy and Overstreet(1950) 3 an equation based on the statistical thermo-dynamics of Guggenheim (1945) is most satisfactoryfor cation pairs of unequal valence. All of the equa-tions become less satisfactory when applied to mixturesof cation-exchange materials having different equi-librium constants.

The use of cation-exchange equations for expressingthe relationship between soluble and exchangeable

26 AGRICULTURE HANDBOOK 6 0,

cations in soils of arid regions involves inherent diffi-culties. The difficulties arise from the presence ofmixtures of different kinds of cation-exchange mate-rials in soils and from the fact that usually four cationspecies must be dealt with. Moreover, there are noaccurate methods available for determining exchange-able calcium and magnesium in soils containingalkaline-earth carbonates and gypsum. Despite thesedifficulties, some degree of success has been attainedin relating the relative and total concentrations ofsoluble cations in the saturation extract of soils to theexchangeable-cation composition, using a somewhatempirical approach. Direct determinations show that,when soils are leached with salt solutions containinga mixture of a monovalent cation and a divalent cationuntil equilibrium between the soil and solution ises,tablished, the proportions of exchangeable mono-valent and divalent cations present on the soil-exchangecomplex vary with the total-cation concentration aswell as with the monovalent : divalent cation ratio ofthe salt solutions. Gapon (1933)) Mattson and Wik-lander (1940), Davis (1945)) and Schofield (1947)have proposed, in effect, that the influence of total-cation concentration is taken into account and a linearrelation with the exchangeable monovalent: divalentcation ratio is obtained when the molar concentrationof the soluble monovalent cation is divided by thesquare root of the molar concentration of the solubledivalent cation.

Two ratios of the latter type, designated as thesodium-adsorption-ratio (SAR) and potassium-adsorp-tion-ratio (PAR) , are employed for discussing theequilibrium relation between soluble and exchangeablecations. The sodium-adsorption-ratio and potassium-adsorption-ratio are defined as Na+/l/( Ca++ + Mg”) /2and K+/ d( Ca++ + Mg”) /2, respectively, where Na+, K’,Cat+, and Mg” refer to the concentrations of the desig-nated soluble cations expressed in milliequivalents perliter.

The relationship between the sodium-adsorption-ratio and the ratio exchangeable sodium : (exchangecapacity minus exchangeable sodium) at the saturationmoisture percentage for 59 soil samples representing 12sections in 9 Western States is shown in figure 9. Asimilar relationship involving the potassium-adsorptionratio, exchange capacity, and exchangeable potassiumis given in figure 10. The correlation coefficients forthe two sets of values are sufficiently good to permitpractical use of the relations. Data for soils havingexchangeable sodium/ (exchange capacity minus ex-changeable sodium) and exchangeable potassium/(exchange capacity minus exchangeable potassium)ratios greater than 1, which correspond to exchange-able-cation-percentages of more than 50, are not in-cluded in the graphs. Limited data indicate that forthese soils the relations shown in the graphs are some-what less precise. Using the data presented in figure 9,the relation between the exchangeable-sodium-percent-age (ESP), and the sodium-adsorption-ratio (SAR) isgiven by the equation :

U. S. DEPT. OF AGRICULTURE

ESP= 100 ( - 0.0126+ 0.01475 SAR)1 + ( - 0.0126 + 0.01475 SAR)

Similarly, the relation between the exchangeable-potassium-percentage (EPP) and the potassium-adsorption-ratio (PAR) is given by the equation :

EPP= ~~~~_(~z0360+0.1051 PAR)1 + (0.0360+ 0.1051 PAR)

The former equation was employed to obtain the aver-age relation between exchangeable-sodium-percentageand the sodium-adsorption-ratio, which is shown by thenomogram given in figure 27, chapter 6.

Chemical Analyses of Representative SoilSamples

Data of typical chemical analyses of saline, non-saline-alkali, and saline-alkali soil samples are givenin table 5. Similar analyses of samples of normal soilsfrom arid regions are also given for comparative pur-poses. These analyses are presented to show thedifferences in the chemical characteristics of the fourclasses of soils and to illustrate how the analyses may beinterpreted and cross-checked for reliability.

Nonsaline-Nonalkali Soils

Samples numbered 2741, 2744, and R-2867 areclassed as normal with respect to salinity and alkali,because the electrical conductivity of their saturationextracts is less than 4 mmhos/cm. and their exchange-able-sodium-percentage is less than 15. The reactionof the samples ranges from slightly acid to slightlyalkaline. While the composition of the soluble ionsvaries somewhat, the amounts present are small, andall of the saturation extracts have low sodium-adsorp-tion-ratios. Alkaline-earth carbonates may or may notbe present. Also, gypsum may or may not be present,although none of the samples selected contains thisconstituent.

Saline Soils

The electrical conductivity of the saturation extractsof these samples is in excess of 4 mmhos/cm., but theexchangeable-sodium-percentage is less than 15. Inno case does the pH reading exceed 8.5. Chloride andsulfate are the principal soluble anions present in thesesamples, the bicarbonate content is relatively low, andcarbonate is absent. The soluble-sodium contents ex-ceed those of calcium plus magnesium somewhat, butthe sodium-adsorption-ratios are not high. Gypsumand alkaline-earth carbonates are common constituentsof saline soils. As shown by the values for the electri-cal conductivity of the saturation extracts, the salinitylevels are sufficiently high to affect adversely the growthof most plants. Reclamation of the soils will requireleaching only, providing drainage is adequate.

Nonsaline-Alkali Soils

The exchangeable-sodium-percentages of these soilsamples exceed 15, but the soluble-salt contents are low.


IO 20 3 0 4 0 5 0 6 0

S O D I U M - A D S O R P T I O N - R A T I O - S A R

FIGURE 9.-Exchangeable-sodium ratio (ES/[CEC-ES1 ) as related to the sodium-adsorption-ratio (SAR) of the saturation extract.ES, exchangeable sodium; CEC, cation-exchange-capacity.


Y = 0 . 0 3 6 0 + 0 . 1 0 5 1 x

I = 0 . 9 7 2 r2 t 0 . 9 4 5

I 2 3 4 5 6 7

P O T A S S I U M - A O S O R P T I O N - R A T I O - PAR

F IGURE IO.-Exchangeable-potassium ratio (EP/[CITC-EP]) as related to the potassium-adsorption-ratio (PAR) of the saturationextract. EP, exchangeable potassium; CEC, cation-exchange-capacity.


T-mm 5.-Chemicd analyses of soil samples *from arid regions

SOIL DETERMINITIONS

29

-

ISatura- pH oftion satn-

percent- ratedage soil

Exchangeable-cation-percentagesCation-exchange-capactty

Alkaline-earth car-bonates *

Soil and sample No. GypsunNa K Ca Mg H

_

Meg./100 gm.

z0

7. 1

::

:0

4: 20’

Normal soils:2741. .2744..............R - 2 8 6 7 .

Saline soils:

!feq./lOO gm.20. 3

n35.6 6. 432.4 7. 840.4 7. 9

29.417.4

52.046. 540.0

:z8: 0

14.417.018.6

58. 8 8. 3 33.461.2 7. 3 34.238. 7 9.6 21.9

61.559. 735. 8

7.3

;::

35. 740. 326. 2

210

3

13

1:

81 . . ...541 .

3 .2

17 .

18 3

:: 3: . . . ..“f

26

%S: 2:8 . . . . . . .

29................

........

........

........

........30

........

35................

0’0

z0

1;0

10

8

574. ..............756. ..............575. ..............

Nonsaline-alkali soils:2747 . . . . . . . . . . . . . .2738, .............535 . . . . . . . . . . . . . . .

Saline-alkali soils:2739. .............2740. .............536. ..............

-

SATURATION EXTRACT DETERMINATIONS

AnionsCations Sodium-adsorp-

tion-ratio

(SARI

Soil Elec-and trical

sample conduc-No. tivity Ca++ Mg++ Na+ K+ Total COa= HCO,- so4= Cl- Total

N o r m a lsoils:2741. ..__2 7 4 4 . .R-2867.

Saline soils:5 7 4 . . .756.....575.....

Nonsaline-alkalisoils:2 7 4 7 . .2 7 3 8 .535.....

Saline-alkalisoils:2739.....2 7 4 0 . .536.....

Mmhos/cm.0.601.68.84

Meq.11. Meq.11. Meq.11.2. 71 2.26 1.203.33 1.94 12.22. 76 1. 69 5.22

Meq.11.0.91

.::

Meq.11.7. 08

18.179.85

13.9 31. 5 37. 2 102.012.0 37. 0 34. 0 79. 08.8 28.4 22.8 53.0

.2140

1: 10

170.91150.40105.30

1. 74 1. 10 1.42 15.6 .42 18. 54~2. 53 1. 41 1.01 21. 5 .28 24.203. 16 1. 10 .30 29.2 4. 10 34.70

9. 19 6. 73 9. 85 79. 516. 7 32.4 38. 3 145.05.6 .60 .90 58. 5

1;;1. 6

96.56 0 2. 35 20. 1 72.0 94. 45 27.6216.21 0 3.29 105.0 105.0 213.29 24. 461.60 5. 00 19.9 21. 5 16.3 62.70 67.6

Meg& Meg.11 Meq.11. Meg./.2.60 2.09 0.87 5. 566. 14 4.28 4.93 15.356.63 2.67 .44 9. 74

0. 8

37::

4. 50 90.0 78.0 172.50 17. 47. 20 62.2 47. 0 148.40 13.35.20 74. 0 29.0 108.20 10. 5

6.51 8.48 2.86 17.85 13. 93.29 3.80 16. 7 23.79 19.6

18.70 4. 60 7. 50 39.20 35.0

Meg&

ii0

0

i

88. 40

t +, Present; -, absent. 2 Includes 32.0 meq./l. of NOS.

Usually the pH readings are greater than 8.5, but they is an example of a nonsaline-alkali soil that is freemay be lower if the exchangeable-sodium-percentage of alkaline-earth carbonates and contains some ex-does not greatly exceed 15 (sample No. 2747) or if changeable hydrogen. Replacement of exchangeablealkaline-earth carbonates are absent (sample No. sodium will be required for its reclamation. Gypsum2738). Gypsum seldom occurs in these soils. The is a suitable amendment, but the application of acidchief soluble cation is sodium, and appreciable amounts or acid-forming amendments may cause excessive soilof this cation may be present as the bicarbonate and acidity unless limestone is also applied. The applica-carbonate salts. The sodium-adsorption-ratio of the tion of limestone alone will tend to replace the ex-saturation extract may be quite high. Sample No. 2738 changeable sodium. Sample Nos. 2747 and 535 will


also require replacement of exchanieable sodium forreclamation; but, owing to the presence of alkaline-earth carbonates, acid, any acid-forming amendment,or gypsum may be applied. The application of lime-stone alone will obviously be of no value.

Saline-Alkali Soils

Soils of this class are characterized by their appre-ciable contents of soluble salts and exchangeablesodium. The electrical conductivity of the saturationextract is greater than 4 mmhos/cm., and the exchange-able-sodium-percentage exceeds 15. The pH readingmay vary considerably but is commonly less than 8.5.Except that a higher proportion of the soluble cationsconsists of sodium, the composition of the soluble saltsusually is similar to that of saline soils. Althoughonly the most salt-sensitive plants will be affected bythe salinity level of sample No. 536, the exchangeable-sodium-percentage is too high to permit the growth ofmost crops. Both replacement of exchangeable sodiumand leaching are required for reclamation of these soils.With respect to the suitability of various amendmentsfor the replacement of exchangeable sodium, sampleNo. 2739, like No. 2738, will require the application ofsoluble calcium, whereas sample No. 536, like samples2747 and 535, can be treated with soluble calcium,acid, or acid-forming amendments. Owing to its highcontent of gypsum, sample No. 2740 will not require theapplication of amendments for the replacement ofexchangeable sodium.

Cross-Checking Chemical Analyses forConsistency and Reliability

A means of locating gross errors in the chemicalanalyses of soils is provided by the considerable num-ber of interrelations that exist among the values ob-tained for various determinations. An understandingof the principles involved in these interrelations aidsin the interpretation of the analyses.

E LECTRICAL CONDUCTIVITY AND TOTAL CATION CON-CENTRATION.-The EC of soil solutions and saturationextracts when expressed in millimhos per centimeter at25” C. and multiplied by 10 is approximately equal tothe total soluble-cation concentration in milliequiva-lents per liter.

CATION AND ANION CONCENTRATION.-The total solu-ble-anion concentration or content and the total soluble-cation concentration or content, expressed on anequivalent basis, are nearly equal.

PH AND CARBONATE AND BICARBONATE CONCENTRA-TIONS.-If carbonate ions are present in a s&oil extractin titratable quantities, the pH reading of the extractmust exceed 9. The bicarbonate concentration seldomexceeds 10 meq./l. in the absence of carbonate ions,and at pH readings of about 7 or less seldom exceeds3 or 4 meq./l.

PH ANDCALCIUMANDMAGNESIUMCONCENTRATIONS.-The concentration of calcium and magnesium in a sat-uration extract seldom exceeds 2 meq./l. at pH readingsabove 9. Therefore, calcium plus magnesium is low

if carbonate ions are present in titratable amounts, andcalcium plus magnesium is never high in the presenceof a high concentration of bicarbonate ions.

C ALCIUM ANDSULFATEINASOIL-WATEREXTRACTANDGYPSUM CONTENT OF THE SOIL.-The solubility of gyp-sum at ordinary temperatures is approximately 30meq./l. in distilled water and 50 meq./l. or more inhighly saline s#olutions. However, owing to the com-mon ion effect, an excess of either calcium or sulfatemay depress the solubility of gypsum to a value as lowas 20 meq./l. Hence, the saturation extract of a non-gypsiferous soil may contain more than 30 meq./l. ofboth calcium and sulfate (i. e. saline soil No. 756)) andthat of a gypsiferous soil may have a calcium concen-tration as low as 20 meq./l. As a general rule, soilswith saturation extracts that have a calcium concentra-tion of more than 20 meq./l. should be checked forthe presence of gypsum.

PH AND ALKALINE-EARTH CARBONATES.-The pHreading of a calcareous soil at the saturation percentageis invariably in excess of 7.0 and generally in excess of7.5; a noncalcareous soil may have a pH reading ashigh as 7.3 or 7.4.

PH AND GYPSUM.-The pH reading of gypsiferoussoils at the saturation percentage is seldom in excess of8.2 regardless of the ESP.

PH AND ESP.-A pH reading at the saturation per-centage in excess of 8.5 almost invariably indicates anESP of 15 or more.

ESP AND SAR.-In general, ESP increases with SAR.There are occasional deviations, but generally low SARvalues of the saturation extract are associated with lowESP values in the soil, and high SAR values denote highESP values.

CEC AND SP.-Because both cation-exchange-capa-city and moisture-retention properties are related tothe texture of soils, there generally exists a fair corre-lation among these properties, particularly in soils withsimilar parent materials and mode of origin.

Factors That Modify the Effect of ExchangeableSodium on Soils

As might be expected, alkali soils having similar ex-changeable-sodium-percentages may vary considerablywith respect to their physical properties, their ability toproduce crops, and their response to management prac-tices, including the application of amendments. Al-though the reasons for the variable behavior of alkalisoils are imperfectly understood, experience and limiteddata indicate that the effect of exchangeable sodiummay be modified by several soil characteristics. Deter-minations of some or all of these characteristics areoften of value in the investigation of alkali soils.

Texture

It is well known that the distribution of particle sizesinfluences the moisture retention and transmissionproperties of soils. Particle-size analysis may be made,using Method 41. As a rule, coarse-textured soils havelow-moisture retention and high permeability, whereas


fine-textured soils have high-moisture retention andgenerally have lower permeability. However, owingto a high degree of aggregation of the particles, thereare notable examples of fine-textured soils that aremoderately permeable. The presence of a high per-centage (50 or more) of silt-size particles (effectivediameter 2~ to 50~) often causes soils to have relativelylow permeability. There is also evidence that somesilt-size particles, presumably those having a platyshape, are more effective in reducing permeability thanothers. In general, the physical properties of fine-textured soils are affected more adversely at a givenexchangeable-sodium-percentage than coarse-texturedsoils. For example, the hydraulic conductivity of acoarse-textured soil having an exchangeable-sodium-percentage of 50 may be as great as that of a fine-textured soil having an exchangeable-sodium-percent-age of only 15 or 20. Inasmuch as fine-textured soilsgenerally have higher cation-exchange-capacities thancoarse-textured soils, expressing the critical levels ofsodium in milliequivalents per 100 gm. tends to elimi-nate the texture factor in evaluating the effect ofexchangeable sodium.

Surface Area and Type of Clay Mineral

Soil particles may be considered to have two typesof surfaces: extel’nal and internal. Primary mineralssuch as quartz and feldspars and the clay minerals kao-iinite and illite have external surfaces only. Clay min-erals of the expanding lattice type such as montmorillo-nite, which exhibits interlayer swelling, have’internal aswell as external surfaces. The external surface areaof soils is directly related to texture, whereas internalsurface area is related to the content of minerals thatexhibit interlayer swelling. Determinations of theamounts of ethylene glycol retained as a monomolec-ular layer by heated and unheated samples of soil(Method 25) permit estimation of the external and theinternal surface areas, provided appreciable amountsof vermiculite and endellite minerals are not present.In any case, the ethylene glycol retained bv unheatedsoil in excess of that retained by a corresponding heatedsample is an index of interlayer swelling.

As determined by Method 25, the external surfaceareas of most soils lie in the range 10 to 50 m.*/gm.(square meters per gram), whereas the internal surfacearea varies to a greater extent, being nil in soils thatcontain no interlayer swelling minerals and as highas 150 m.‘/gm. or more in soils with a high content ofexpanding lattice-type minerals. X-ray diffractionpatterns indicate that the clay fraction (particles <2peffective diam.) of many soils of arid regions are pre-dominantly interstratified mixtures of various propor-tions of montmorillonite and illite, although sometimesindividual crystals of these minerals occur. Theamount of kaolinite present is usually small.

It is generally recognized that soils containing clayof the expanding lattice (montmorillonitic) type exhibitsuch properties as swelling, plasticity, and dispersionto a greater extent than soils containing equivalent

amounts of nonexpanding lattice (illitic and kaolinitic)clays, especially when appreciable amounts of ex-changeable sodium are present. Whether the more ad-verse physical properties imparted by the former typeof clays are caused by their greater total surface areaor to the fact that they exhibit interlayer swelling isnot definitely known. Further studies may show thatthe susceptibility of soil to injury by exchangeablesodium is related to total surface area as measured byethylene glycol retention.

Potassium Status and Soluble Silicate

Several medium- to fine-textured alkali soils havebeen examined at the Laboratory and have been found tobe much more permeable than would ordinarily be ex-pected on the basis of their high exchangeable-sodium-percentages. In some cases, the permeability is suchthat the soils can be leached readily with large quanti-ties of irrigation water and the excess exchangeablesodium removed without the use of chemical amend-ments. The soils have several characteristics in com-mon, which .include a high pH value (9.0 or higher),a high exchangeable-potassium-percentage (25 to 40))and an appreciable content of soluble silicate. Thesilicate concentration of the saturation extracts of thesesoils has been found to vary from 5 to 40 meq./l., andadditional quantities of this anion as well as sodiumare removed upon leaching. As shown by ethyleneglycol retention, Dyal and Hendricks (1952) and Bowerand Gschwend (1952)) saturation of montmorilloniteclays and soils with potassium followed by drying de-creases interlayer swelling. Moreover, Mortland andGieseking (1951) have shown by means of X-raydiffraction studies that montmorillonite clays, whendried in the presence of potassium silicate, are changedto micalike clays that would have less tendency to swelland disperse under the influence of exchangeablesodium. Ethylene glycol retention determinationsmade on some of the alkali soils having high exchange-able-potassium-percentages and containing appreciablesoluble silicate give relatively low values for interlayerswelling. While further research is needed to clarifythe role of exchangeable potassium and soluble silicate,there is a distinct indication that alkali soils containingunusually high amounts of these constituents are lesssusceptible to the development of adverse physicalconditions.

Organic Matter

While the organic-matter content of soils of aridregions is usually low under virgin conditions, it com-monly increases with the application of irrigation waterand cultivation, especially when crop management isgood. Aside from its value as a source of plantnutrients, organic matter has a favorable effect upons’oil physical properties.

There is considerable evidence that organic mattertends to counteract the unfavorable effects of exchange-able sodium on soils. Campbell and Richards (1950)


HYDRAULICi CONDUCTIVITY

L H

SOIL SAMPLE4

SATURATED PASTE8

SATURATION

FL ( SODlUg jLi- - - - SATURATIONI H EXTRACT I

I

H

-----P-./ \‘/POSSlEILE’\

‘;NFAVORABL$\I1 P H Y S I C A L ’‘\,CONDITIO N ,,’

\-,A

IALKALINE

H EARTH L

CARBONATES

AMENDMENT

F IGURE Il.-Sequence of determinations for the diagnosis and treatment of saline and alkali soils: H, High; L, low; Rs,electrical resistance of soil paste; SAR, sodium-adsorption-ratio; ESP, exchangeable-sodium-percentage; CEC, cation-exchange-capacity.


and Fireman and Blair 5 found that peat and muck soilscontaining appreciable quantities of exchangeablesodium had good physical properties, and numerousinvestigators have demonstrated a beneficial effect oforganic matter additions upon alkali soils. For ex-ample, Bower and associates (1951) found that theapplication of manure at the rate of 50 tons per acreto an alkali soil of the “slick spot” type increased thedegree of aggregation of the surface soil significantlyand the infiltration rate approximately threefold. Theavailable data indicate that organic matter improvesand prevents deterioration of the physical conditionof the soil by its interaction with the inorganic cation-exchange material, by serving as energy material formicro-organisms which promote the stable aggregationof soil particles, and by decreasing the bulk density ofsoils.

The organic-matter content of soils is ordinarily ob-tained by multiplying the organic-carbon content by1.72. The dry-combustion method is most accuratefor the determination of organic carbon, but it is time-consuming and cannot be applied to soils containingcarbonates. Wet-combustion methods such as the onegiven in Method 24 are suitable for use on soil contain-ing carbonates, but the application of a correction factoris required to compensate for the incomplete oxidationof the organic matter.

Sequence of Determinations for Soil Diagnosis

The salinity status and the hydraulic conductivityare measured for all samples. The sequence of furtherdeterminations depends on whether the result obtainedfrom a previous determination (fig. 11) is consideredto be highorlow. Criteria for distinguishing high andlow values are discussed in chapter 6.

The determinations are ordinarily discontinued whenthe guide lines of the two main branches of the diagramlead to a heavy-walled box, except in the case of an

' FIREMAN, M., and BL A I R, G. Y. CHEMICAL AND PHYSICAL

ANALYSES OF SOILS FROM THE HUMBOLDT PROJECT, NEVADA.1Unpublished.l January 1949.

alkali problem where alkaline-earth carbonates shouldalso be determined if the use of acid or acid-formingamendments is contemplated. At two places in the dia-gram, dotted lines indicate where optional alternate de-terminations can be made. The alternate determina-tions cost somewhat less but have lower reliability.

Hydraulic-conductivity measurements on disturbedsamples provide an indication of the moisture-transmis-sion rate of the soil. It has been found for most soilsthat exchangeable sodium is not excessive if this rate ishigh. However, coarse soils such as sands and peatsmay contain sufftcient amounts of exchangeable sodiumto be toxic to plants and yet have high permeability.If the hydraulic conductivity is low, the total extract-able sodium or the sodium-adsorption-ratio (SAR)should be determined. If either of these is low, thelow hydraulic-conductivity value previously obtainedmay be the result of an inherently unfavorable physicalcondition related to texture, low content of organicmatter, or high-swelling type clay rather than the pres-ence of exchangeable sodium. For these samples, or-ganic matter, ethylene glycol retention, and particle-sizeanalyses may yield useful information.

If the total extractable-sodium content or the SARvalue is high, the exchangeable sodium should be de-termined or, alternatively, the exchangeable-sodium-percentage can be estimated from the SAR value. Ifthe exchangeable-sodium content or exchangeable-sodium-percentage is high, a gypsum determinationshould be made. A high-gypsum value indicates thatleaching only is required, while a low-gypsum valueindicates need for amendments,. When there is a low-gypsum value, the presence or absence of alkaline-earthcarbonates is ascertained to indicate the type of chemi-cal amendment that can be used for the replacement ofexchangeable sodium. The addition of amendmentsshould be followed by leaching. Other determinations,such as pH, saturation percentage, cation-exchange-capacity, exchangeable potassium, toxic ions, and tex-ture, provide additional information and are made ifcircumstances warrant.

determination of the properties of saline and alkali soils€¦ · chapter 2 determination of the...

Documents