1376 Salt Management Workshop Proceedings › arsuserfiles › 20360500 › pdf_pubs › P137… · 1994). In the coastal areas of the United States, the gradual intrusion of seawater

1 3 7 6

Salt Management Workshop Proceedings

Considerations forSalt Management inSoil and Water forProducingHorticultural Crops:Introduction to theWorkshop

Alexander A. Csizinszky

New Insights inPlant BreedingEfforts for ImprovedSalt Tolerance

Michael C. Shannon

Managing Salineand Sodic Soils forProducingHorticultural Crops

Charles A. Sanchez and Jeffrey C. Silvertooth

ApplyingPreferential FlowConcepts toHorticultural WaterManagement

John S. Selker

Considerations forSalt Management inSoil and Water forProducingHorticultural Crops:Introduction to theWorkshop

Alexander A. Csizinszky’

In nonhalophytic plants, saline condi-tions reduce vegetative growth andyields. Salt stress effects on plants fromsoil (24 dSrn_’ EC and <15% exchangeable Na+) or from saline (brackish) irrigation

water (24 dSW EC) may be divided into threebroad categories: 1) osmotic effects, 2) specific ioneffects, and 3) interference with the uptake of essen-tial nutrient ions (Hausenbuiller, 1972). Generallysalts, once introduced, accumulate in the upper soilprofile when they are transported by upward capil-lary movement of water, which evaporates, andwhen rainfall and irrigation are insufficient to leachsalts to lower soil depths (Suarez, 1992). Thus, soilsalinity is a particularly severe problem in the aridand semi-arid zones of the world or in areas whereonly poor-quality water is available for irrigation(Hamdy et al., 1993; Rowley, 1993; Satti et al.,1994). In the coastal areas of the United States, thegradual intrusion of seawater into the fresh-wateraquifers threatens water supplies for agricultural,industrial, and municipal users (Cole, 1993).

Crops grown on saline soils or irrigated withsalinewater requiredifferent management and irriga-tion methods than plants in nonsaline soils or irri-gated with good-quality water. For example, in soilsthat have >15% exchangeable Na+, plants requiresoluble Ca2+fertiIizers to remedy ionic imbalances,which adversely affect their growth and development.Selecting crops and cultivars that can tolerate salineconditions and may produce economically accept-able yields is very important. Plants in the samefamily, or even the same species, react differently tosaline conditions. For example, in Cruciferae, theorder from highest to lowest tolerance to balancedfertilizer salts was cabbage>cauliflower > broccoli >kohlrabi (Csizinszky, 1979), and, in Leguminosae,

‘Associate horticulturist, Gulf Coast Research andEducation Center, University of Florida, IFAS,Bradenton, FL 34203.

Florida Agricultural Experiment Station journal seriesno. R-04724.

the order was green pea > faba bean > soybean >green bean (Delgado et al., 1994). In Solanaceae,differences to salt stress among tomato cultivarswere reported by Perez-Alfocea et al. (1993) andamong eggplant cultivars by Zurayk et al. (1993).

Many plants have salt-sensitive stages duringtheirdevelopmentwhenirrigation bysainewatershouldbeavoided.Germinatingseedsandyoungseedlingsareparticularly sensitive to salt stress, and irrigation withsaline water during thesestages is harmful. Hamdyetal.(1993)recommended irrigationwith good-qualitywaterduring the salt-sensitive stages of growth, followed bysaline water during later growth stages, when good-quality water may be scarce for irrigation purposes.

Althoughconsumptiveuseofwaterbyagricultureand industry has declined since the mid-1980s thedemand for food, water, and land by an ever-growingpopulation has increased (Solley, 1993; Suarez, 1992).Under these conditions, the knowledgeofcropmanagement on marginal lands, an understanding of irrigationmanagement with brackish water, and crop selection forsaline environmental conditions will increase in impor-tance for horticulturists.

In the following articles, some soil and cropmanagement methods, breeding and selectingplants for salt tolerance, and aspects of water andsolute transport in soils are discussed.

Literature Cited

Co/e, J. 1993. Plenty of water but less to use. The Packer100(34):A1.

Csizinszky, A.A. 1979. Effect of salt on seed germinationand transplant survival of vegetable crops. Fla. Sci.42:43-51.

Delgado, M.J., F. Ligero, and C. Lluch. 1994. Effects ofsalt stress on growth and nitrogen fixation by pea, fababean, common bean and soybean plants. Soil Biol.Biochem. 26:371-376.

Hamdy, A., S. Abdel-Dayem, and M. Abu-Zeid, 1993.Saline water management for optimum crop production.Agr. Water Mgt. 24:189-203.

Hausenbuiller, R.L 1972. Soil science, principles andpractices. W.M. Brown and Co., Dubuque, Iowa.

Perez-Alfocea, F., M. T. Estan, M. Caro, and M. C. Bolarin.1993. Response of tomato cultivars to salinity. Plant Soil150:203-211.

Rowley, G. 7993. Multinational and national competitionfor water in the Middle East: Towards the deepeningcrisis. J. Environ. Mgt. 39:187-197.

Satti, S.M.E.. M. Lopez, and F.A. Al-Said. 1994. Salinityinduced changes in the vegetative and reproductivegrowth in tomato. Commun. Soil Sci. Plant Anal. 25:501-510.

So//ey, W.B. 7993. Water-use trends and distribution intheUnitedStates, 1950-1985. HortScience28:283-285.

Suarez, D. 1992. Perspective on irrigation managementand salinity. Outlook Agr. 21:287-291.

Zurayk, R., M. Nimah, and M. Hamze. 1993. The salttolerancepotential of localcultivarsofeggplant (Solanummelongena L.). Biol. Agr. Hort. 9:317-324.

New Insights inPlant BreedingEfforts for ImprovedSalt Tolerance

Michael C. Shannon’

Additional index words. salinity, manage-ment, irrigation, genetic selection, systemsresearch

Summary. The lack of improvement for salttolerance has been attributed to insufficientgenetic variation, a need for rapid and reliablegenetic markers for screening, and thecomplexities of salinity x environmentinteractions. Salt tolerance is a quantitativecharacteristic that has been defined in manyways subject to changes with plant developmentand differentiation; thus, assessing salttolerance among genotypes that differ in growthor development rate is difficult. Salt tolerancealso varies based on concentrations of majorand minor nutrients in the root zone. Plantgrowth models may provide a method tointegrate the complexities of plant responses tosalinity stress with the relevant environmentalvariables that interact with the measurement oftolerance. Mechanistic models have beendeveloped over the last few years that areresponsive to nitrogen or drought stress but notto salinity stress. Models responsive to salinitystress would provide insights for breeders andaid in developing more practical research on thephysiological mechanisms of plant salttolerance.

S alinity reduces yield and di-minishes quality in fruit andvegetables. Historically, effortsto manage high salinity in soilor water has been through crop

substitution and agromanagement. Crop substitu-tion, or replacing salt-sensitive crops with moretolerant ones, has been practiced from the dawn ofagriculture (Shannon, 1987) and is still probablyone of easiest and most practiced strategies for

‘U.S. Salinity Laboratory, USDA-ARS-PWA, 450 W.Big Springs Road, Riverside, CA 92507.

HortTechnology - April-June 1996 6(2)

dealinq with salinity. When high-quality water isnot available for irrigation and leaching, cropsubstitution has been the main approach to deal-ing with salinity. More salt-tolerant crops aresubstituted for salt-sensitive species. Thus, barleymay be substituted for wheat, cotton for corn,sugar beet for lettuce, etc. Unfortunately, mostvegetable crops are more sensitive to salt thanmost field crops. Harvest quality usually has amore significant impact on marketable yield ofhorticultural crops than field crops. The salt-toler-ance tables developed by the U.S. Salinity Labora-tory have been valuable guides for extension per-sonnel and growers in determining which cropscan be grown based on the expected soil salinity(Maas, 1990; Maas and Hoffman, 1977).

Agromanagement techniques that minimizeand avoid salinity effects include leaching, deepplowing, applying amendment and carefully choos-ing fertilizer sources, installing drainage, landleveling, and irrigation technologies. Other man-agement options include using drip or sprinklerirrigation to improve water application efficiencyand elaborations in bed formation and plantingdesign to facilitate removal of accumulated saltsfrom the areas in which roots are developing andextracting water (Rhoades, 1993). Somestrategiesthat have not been adequately researched involvemanipulating population densities to improve plantstand and applying nonsaline or more saline waterdependent on variable salt tolerance with growthstage.

More recently, crop breeding and geneticmanipulation, using tools such as tissue cultureand molecular biology, have been proposed asadjunct strategies to deal with salinity. Past workhas shown that different crop species vary withrespect to at least two of their parameters of salttolerance-the salt tolerance threshold and yielddecline beyond that threshold (Maas and Hoffman,1977). These parameters reflect the simplest wayto conceptualize the effects of salt on yield.

In the last 2 decades there has been greatinterest in breeding plants for improved salt toler-ance (Shannon, 1979; Shannon and Noble, 1990).Strategies that have been tried or suggested in-clude conventional screening, selecting and breed-ing with established cultivars, introducing highsalt tolerance in cultivated species through intro-gression with tolerant wild relatives, or domesti-cating salt-tolerant wild or halophytic speciesthrough the genetic improvement of agronomic orhorticultural characteristics. Some of these effortshave resulted in limited success but major ad-vances have not been noted.

Examples of screening or selection criteriathat have been tried include selection during ger-mination or emergence, resistance to salinity-induced reductions in plant height or weight, andmaintenance of high yield or quality under saltstress and plant survival (Shannon and Noble,1990). Extensive efforts have also been made toidentify reliable physiological or biochemical mark-

ers for salt tolerance. These markers include ex-clusion of ions (Na, Cl) from shoots or specifictissues, maintenance of nutrients (K, Ca, Mg, P,NO,), and ion selectivity (high K/Na,Ca/Na, orNO$CI ratios) (Cheeseman, 1988; Jeschke, 1984;Shannon, 1979, 1982). Accumulated metabolic-compatible solutes such as proline, glycinebetaine,and certain sugars and alcohols have also beenproposed as indices for high salt tolerance, but theevidence that accumulation of compatible solutesoffers a quantitatively measurable improvement insalt tolerance has not been universally accepted

Physiologists have offered even more com-plex criteria for salt tolerance. Some of theseinclude the ability of some species or varieties tomaintain membrane integrity due to componentstructure or the ability to sequester ionsviacompartmentation, excrete ions through saltglands, or dilute ions using succulence (Liittgeand Smith, 1984; Thomsom et al., 1988). Someproposed morphological adaptations for salt tol-erance include change in root-to-shoot ratio, ro-sette growth, succulence, reduction of staturethrough shoot or tiller abortion, and early initiationof reproductivegrowth (Wuchli and Epstein, 1990;Maasand Grieve, 1990; Maas and Nieman, 1978).Metabolic adaptations that have been associatedwith tolerance include increased respiration andimproved water use efficiency at low salinity lev-els, increased mesophyll resistance, decreasedCO, fixation, synthesis of compatible organic sol-utes, and the initiation of CAM metabolism (LOttgeand Smith, 1984;Shannon et al., 1994). It may notbe obvious, but reduced growth rate, in itself, maybe an adaptation to high salinity. A smaller plantrequires less water and does not concentrate ex-cluded salts in the root zone to the same extent asa plant that maintains a more rapid growth rate.

At least threedogmas have been establishedwith respect to salinity stress. First, salinity is amulticomponent stress in a physical and chemicalsense and varies in intensity over the three dimen-sions of space, time, and concentration. Its effectsdepend on specific salt composition and are sub-ject to environmental interactions. The seconddogmaisthatwithintheplantkingdomthereexistsa vast array of adaptive responses to salinitystresses. The physiological significance or quan-titative importance of most of these has not beenexplicitly proven or quantitatively described andwill continue to be a major research objective. Athird well-known dogma is that salinity stress ishighly sensitive to interactions with other environ-mental variables, especially humidity and lightintensities (Magistad et al., 1943; Salim, 1989).

The complexities intertwined in the afore-mentioned dogmas highlight the need for a methodof conceptual integration. Within the last 50 yearsmuch information has been developed concerningsalt tolerance and effects of salinity on plants orplant components. However, the component ob-servations and facts constitute a hodgepodge thatresists comprehension. A new impetus is needed

to advance breeding for salt tolerance to a higherlevel. Two technologies that will be useful in thebreeding effort are crop modeling and systemsmethodology. Process-based plant growth mod-els help simulate complex crop responses to sa-Iinityinanintegrativemanneroverdifferentgrowthstages and soil-environment scenarios. Althoughmechanistic models responsive to salinity stressare just beginning to be developed, there is rea-sonable hope that such models will soon provideanew basisfortesting hypotheses and experimen-tal design. Most existing plant growth models treatthe soil simplistically. Integrated soil chemistryand physics models that describe water movementand chemical interactions are available (Simuneket al., 1992; Suarezand Simunek, 1994) but thesemodels treat the plant in a simplistic manner andincorporate the effects of salinity through empiri-cal rather than mechanistic relationships. Thesemodels oversimplify calculations for yield androot water uptake. Second-generation models thataddress and correct some of these problems arecurrently being developed. These models will bedeveloped to handle salinity stress in a moremechanistic and comprehensive manner, includ-ing differential salt sensitivity relative to growthstage and a better coupling of growth and rootwater uptake with above-ground plant growth anddevelopment.

Systems approaches also provide an oppor-tunity to coalesce and use efficiently availabletechnology and the accumulated knowledge con-cerning salt tolerance that has been obtained in thelast 2 decades (Shannon, 1994; Wymore, 1993).System approachesasapplied to the developmentof salt tolerance include the followino:1)2)3)

4)5)6)

-Defining the problem;Establishing inputs and outputs;Identifying requirements for performance andresource use;Assessing available technology;Developing a solution concept; andDeveloping screening and selection proce-dures that will lead to a plant ideotype to fit theprescribed requirements of the agronomic sys-tern.

The problem to be defined is the specificsalinity problem that is being addressed in thecontext of crop species, agricultural environment,and specific ion and toxicity thresholds duringcropping. This requires significant input from grow-ers and produce buyers. A description of thesalinity situation should includesourceof salinity(e.g., irrigation water, soil, water table), salt-spe-cific ion distribution and concentration informa-tion, and environmental factors such as range ofaffected crop land, climate, soil type, and drainageavailability.

Genetic variability should be assessed todetermine the extent to which cultivars express asalt tolerance that might prove useful for the prob-lem being considered. This assessment shouldincludethe known salinity threshold, yield decline

Hort Technology . April-June 1996 6(2)

parameters for the crop, and information onintervarietal and interspecific variability. Also, in-formation should be collected on known interac-tions among salinity responses and other environ-mental variables. Available physiological or mor-phological mechanisms with importance for salttolerance should be noted.

Accurately defining the problem is essentialfor establishing input and output require-ments. Assessments should be made with regardto the economic situation and genetic variability.The economic situation should be appraised withconsideration to costs to the grower, such as seed,water fertilizer, chemicals, field operations, fuel,labor, and overhead. Such costs should beweighedagainst market considerations (e.g., yield, quality,earliness), allotments and price supports, andindirect effects, such as long range sustainabilityor potential of using and generating saline drain-age or groundwater.

P e r f o r m a n c e a n d r e s o u r c e u s e r e -quirements must be identified at this juncture.Crop yield and quality requirements should bedefined and fixed costs for production should bespecified. Managementoptionsshould beweighedagainst breeding potential. If management is arelatively inexpensive input into the system, thenperhaps a full-scale breeding effort is not needed.If using saline water is a requirement, adequategenetic variability must be potentially available toproduce a crop that is economically successful.For efforts that require management and breeding,for example, tolerance at specific growth stagesshould be identified. The magnitude of increase insalt tolerance needed to achieve success shouldbe identified.

Available technology must be assessedto determine the types of germplasm and equip-ment that are available. Salt-tolerant rootstockshave been identified for a number of woody spe-cies, including citrus, grape, avocado, andstonefruits (Maas, 1986). Cultivars of lettuce, to-mato, melon, and several other crops that havegreater relative salt tolerances than others havebeen identified (Shannon and Francois, 1978;Shannon and Noble, 1990; Shannon et al., 1983).In some instances, a closely related salt-tolerantspecies might be useful in place of a more sensi-tive species; for instance, cantaloupe might beused instead of watermelon, or a salt-tolerantforage such as tall wheatgrass or subterraneanclover might replace crested wheatgrass or whiteclover(Shannon, 1978; Shannon and Noble, 1995).Salt-tolerant lines may be incorporated into theoverall strategy as substitutes for sensitive culti-vars or species or as potential parents in a crossingand selection program designed to achieve greatersalt tolerance.

The next step is to develop a solutionconcept. The essence of this is to develop afunctional conceptual model of the plant ideotypethat will fill the niche in a describedagromanagement system, a system that will fulfill

all of the established requirements. Crop growthsimulation models may someday be useful inhypothesis testing of such ideotypes before screen-ing and selection technologies are used in a full-scale breeding program.

The concept of the ideotype can be used tod e v e l o p s c r e e n i n g a n d s e l e c t i o n p r o c e -dures Thescreening program may be focused ona selected combination of physiological and mor-phological characters that are predicted to best fita proposed management system in the target en-vironment. Any combination of technologies fromconventional breeding to molecular approachesmight be used to achieve the desired ideotype.After the desired genotype is obtained through anyprocedure, it is important that procedures be es-tablished for germplasm maintenanceand protec-tion and future varietal improvement and develop-ment. Otherwise, the unique combinations thatinfer salt tolerance with the specific target environ-ment cannot be expected to endure growth stages(Shannon, 1985).

Many successful plant breeders are able toconceptualize intuitively the steps that have beenoutlined herein to produce successful cultivars innonsaline situations; however, salinity problemsassociated with irrigated agriculture are complexand require a more rigorous and structured effort.The lack of such a structured effort has contributedas much to the lack of progress in breeding for salttolerance as have previous gaps in informationconcerning salt tolerance mechanisms.

Literature Cited

Cheeseman, J.M. 1988. Mechanism of salinitytolerance in plants. Plant Physiol. 87:547-550.

Jeschke, W.D. 1984. K’-Na’exchangeat the cellu-lar membranes, intercellular compartmentation ofcations and salt tolerance, p. 37-66. In: R.C.Staples and G.H. Toenniessen. (eds.). Salinitytolerance in plants: Strategies for crop improve-ment. Wiley, New York.

L&ch/i, A. and E Epstein. 1990. Plant responses tosalineand sodicconditions, p, 113-137. In: K.K. Tanji.(ed.). Agricultural salinity assessment and management.Amer. Soc. Civil Eng. Manuals Rpt. Eng. PracticeNo.71,Amer. Soc Civil Eng., New York.

Liittge, U. and J.A.C. Smith. 7984. Structural,biophysical, and biochemical aspects of the role ofleaves in plant adaptation to salinity and waterstress, p. 125-150. In: R.C. Staples and G.H.Toenniessen. (eds.). Salinity tolerance in plants:Strategiesforcrop improvement. Wiley, NewYork.

Maas, E V. 1986. Salt tolerance of plants. Appl.Agr. Res. 1:12-26.

Maas, E. V. 1990. Crop salt tolerance, p. 262-304.

In: K.K. Tanji. (ed.). Agricultural salinity assess-ment and management. Amer. Soc. Civil Eng.Manuals Rpt. Eng. Practice No. 71. Amer. Soc.Civil Eng., New York.

Maas, E. V. and C.M. Grieve. 1990. Spike and leafdevelopment in salt-stressed wheat. Crop Sci.30:1309-1313.

Maas, E.V. and G.J. Hoffman. 1977. Crop salttolerance-Current assessment. J. lrr. DrainageDiv., Amer. Soc. Civil Eng. 103(lR2):115-134.

Maas, E. V. and R.H. Nieman. 1978. Physiology ofplant tolerance to salinity. In:G.A.Jung(ed.). Croptolerance to suboptimal land conditions.ASASpec.Publ. 32:277-299.

Magistad, O.C., A.D. Ayers, C.H. Wadleigh, andH.G. Gauch. 7943. Effect of salt concentration,kind of salt, and climate on plant growth in sandcultures. Plant Physiol. 18:151-166.

Rhoades,J.D. 1993,Practices to control salinity inirrigated soils, p. 379-387. In: H. Lieth and A. AlMassoum (eds.). Towards tational use of highsalinity tolerant plants. vol 2. Proc. 1st ASWASConf. 8-15 Dec. 1990, Al Ain Univ, United ArabEmirates, Kluwer Acad. Publ., Dordrecht.

Salim, M. 1989. Effects of salinity and relativehumidity on growth and ionic relations of plants.New Phytol. 113:13-20.

Shannon, M.C. 1978. Testing salt tolerance vari-ability among tall wheatgrass lines. Agron. J.70:719-722.

Shannon, M.C. 1979. In quest of rapid screeningtechniques for plant salt tolerance. Hort Science14:587-589.

Shannon. M.C. 7982. Genetics of salt tolerance:New challenges, p, 271-282. In: A. San Pietro(ed.). Biosaline research: A look into the future.Plenum Publ. Corp.

Shannon, M. C., J. D. McCreight, and J. H. Draper.1983. Screening tests for salt tolerance in lettuce.J. Amer. Soc. Hort. Sci. 108:225-230.

Shannon. M.C. 1985. Principles and strategies inbreeding for higher salt tolerance. Plant Soil89:227-241.

Shannon, M.C. 1987. Salinity-An environmen-tal constraint on crop productivity, p. 9-18. 4thAustral. Agron. Soc. Conf., Responding to Change,La Trobe, Univ., Melbourne,Australia,24-27Aug.

Shannon, M.C. 1994. Development of salt stresstolerance-Screening and selection systems for


genetic improvement, p. 117-332.ln:V.C. Baligar,R.R. Duncan, and J. MaranviIle(eds.). Proc. Work-shop on Adaptation of Plants to Soil Stresses,Sorghum and Millet Collaborative Support Pro-gram (INTSORMIL), Univ. of Nebraska, l-4 Aug.1993.

Shannon, M.C. and L.E. Francois. 1978. Salt tol-erance of three muskmelon cultivars. J.Amer.Soc.Hort. Sci. 103:127-130.

Shannon, M. C., C. M . Grieve, and L.E Francois.1994. Whole plant response to salinity, p. 199-244. In: R. E. Wilkinson (ed.). Plant responsemechanisms to the environment, Marcel Dekker,New York.

Shannon, M.C. and C.L. Noble. 1990. Geneticapproaches for developing economic salt-tolerantcrops, p. 161-185. In: K.K. Tanji. (ed.). Agricul-tural salinityassessmentand management. Amer.Soc. Civil Eng. Manuals Rpt. Eng. Practice No. 71.Amer. Soc. Civil Eng., New York.

Shannon, M. C. and C. L. Noble. 1995. Variation insalt tolerance and ion accumulation among sub-terranean clover cultivars. Crop Sci. 35:798-804.

Simunek, J., T Vogel, and M. Th. van Genuchten.1992. The SWMS_2 code for simulating waterflow and solute transport in two-dimensional vari-ably saturated media. version 1.1. Res. Rpt. No.126, U.S. Salinity Lab., Riverside, Calif.

Suarez, D.L. and J. Simunek. 1994. Modelingequilibrium and kinetic major ion chemistry withCO, production/transport coupled to unsaturatedwater flow, p. 1215-1246. In: Gee and N.R. Wing(eds.). In situ remediation: Scientific basis forcurrent and future technologies. Part 2, G.W. 33rdHanford Symp. on Health and the Environment,Pasco, Wash.

Thomson, W. W . CD. Faraday, and J. W. Oross.1988. Salt glands, p. 498-537. In: D.A. Baker andJ.L. Hall (eds.). Solute transport in plant cells andtissues. Longman Scientific &Technical, Harlow,Essex, England.

Wymore, A. W. 1993. Model-based systems engi-neering. CRC Press, Boca Raton, Fla.

Managing Salineand Sodic Soils forProducingHorticultural Crops

Charles A. Sanchez’ andJeffrey C. Silver-tooth*

Additional index words. irrigation, waterquality, salt tolerance, soil amendments,soil analysis

Summary. About 33% of all irrigated landsworldwide are affected by varying degreesof salinity and sodicity. Soil with anelectrical conductivity (EC) of the saturatedextract >4 dSm is considered saline, butsome horticultural crops are negativelyaffected if salt concentrations in the rootingzone exceed 2 dS&. Salinity effects onplant growth are generally osmotic innature, but specific toxicities and nutritionalbalances are known to occur. In addition tothe direct toxic effects of Na salts, Na cannegatively impact soil structure. Soil withexchangeable sodium percentages (ESPs)or saturated extract sodium absorptionratios (SARs) >15 are considered sodic.Sodic soils tend to deflocculate, becomeimpermeable to water and air, and puddle.Many horticultural crops are sensitive to thedeterioration of soil physical propertiesassociated with Na in soil and irrigationwater. This review summarizes importantconsiderations in managing saline andsodic soils for producing horticulturalcrops. Economically viable managementpractices may simply involve a minor,inexpensive modification of culturalpractices under conditions of low tomoderate salinity or a more costlyreclamation under conditions of high Na.

‘Department of Soil, Water, and EnvironmentalSciences, Yuma Agricultural Center, University ofArizona.

2Department of Plant Sciences, University of Arizona,Tucson.

The cost of publishing this paper was defrayed in partby the payment of page charges. Under postalregulations, this paper therefore must be herebymarked advertisement solely to indicate this fact.

P roblems associated with saltand Na-affected soils have ex-

I

isted for thousands of years inlocations ranging from Asia,the Middle East, Africa, South

America, and North America. For example, one ofthecausesforthedeclineanddisappearanceofthenative American Hohokam civilization in centralArizonaisthoughtto be soil salinization of the GilaRiver Valley, which they irrigated (Bohrer, 1970).About 33% of all irrigated land worldwide is madeup of saline soils (Yenson and Bedell, 1993).About 6 million ha of irrigated landislosteachyeardue to drainage and salinization problems (Bohnet al., 1985).

In addition, about 405 million ha of land isassociated with saline aquifers. Problems of soilsalinity and sodicity occur primarily in arid areaswhere evapotranspiration exceeds rainfall. How-ever, under certain conditions, soil salinity prob-lems exist in humid regions as well.

In the United States, 5 million ha in the 17western states is classified as saline (Bohn et al.,1985). Many of these soils are used for producinghorticultural crops. Several horticultural crops aresensitive to saline conditions and are stronglyaffected by the deterioration of soil physical con-ditions associated with Na. The objective of thispaper is to review and summarize various consid-erations in managing saline and sodic soils usedfor producing horticultural crops.

S o i l c o n s i d e r a t i o n s

Clremical.The salt and Nastatus ofthesoilshould be chemically characterized. The salinitystatus of a soil usually is determined by measuringthe electrical conductivity (EC,) of the water extractfrom the saturated soil paste. Soils are classifiedas being saline if EC, exceeds 4 dSml (Table 1).However, many salt-sensitive horticultural crops,such as lettuce (Lactuca sativa L.), are harmed byEC, >2 dSm-‘.

Sodicity refers to an excess of exchangeableNa in the soil. In addition to direct toxic affects fromNa salts, Na negatively influences soil structure,thus rendering soils unsuitable for crop produc-tion. Sodic soils often are characterized by dis-persedsoilparticles, lowwaterpermeabilityduetopore clogging and colloid swelling, poor aeration,high soil pH (>9), and dispersed organic matter(black alkali).

Soils with exchangeable Na percentages(ESPs) >15 are considered sodic. However, ESPvalues of 5 to 10 may reduce infiltration in somesoils. Because determining ESP is tedious, thesodium absorption ratio (SARe) in water extractsfrom saturated soil pastes is often used.

SAR,= (Na)/((Ca + Mg)!2)Os [ 1 ]

where concentrations are expressed in meq/liter.Although SAR, has a chemical basis in the

Hort Technology * April-June 1996 6(2)

Table 1. Classification of salt- and sodium-affected soils.

Criterion Normal

EC, (dSm-‘) <4SAR, <15

Saline

>4<15

Sodic

<4>15

Saline-Sodic

>4>15

Table 2. A comparison of salt levels and moisture contents for three soils.

Salt level ormoisture content

Soil situationEC, (dSm’)Dry soil weight (kg/15 cm-ha)

At saturationWater content (%)Water (kg.ha-‘)Salt concentration (ppm)Total salts (kg.ha-‘)

At field capacityWater content (%)Water (kg.ha-‘)Salt concentration (ppm)

Sand Loam Muck

2.0 2.0 2.02,400,000 2,128,000 616,000

24 40 150576,000 851,200 924,000

1,280 1,280 1,280737 1,090 1,183

6 20 140144,000 425,600 862,400

5,118 2,560 1,370

theory of cation exchange (Sposito and Mattigod,1977) we typically measure what has been calledthe practical SAR, ratio, which is only empiricallyrelated to SAR, and ESP. Statistical relationshipsderived from several soils indicate that practicalSAR, values are about 12% < SAR, values andreasonably close to ESP values (Sposito andMattigod, 1977). Hence, soils with practical SAR,values>15 are generally considered sodic (Table 1).

Soil-water relationships. Interpretationsof soil salinity status and its potential impact onplant growth can be somewhat confounded bysoil-water relationships, which are largely influ-enced by soil texture and soil organic mattercontent. Forexample, consider a sand,a loam,anda muck soil all having an EC, of 2 dSm andcontaining 24%, 40%, and 150% moisture atsaturation, respectively(Table2). However, at fieldcapacity (about 0.1 bar tension), moisture con-tents are 6% for sand, 20% for loam,and 140% formuck. Assuming no leaching, precipitation orplant uptake of salts, at field capacity the saltconcentration in sand would be almost twice thatof loam and almost four times that of muck. Thesecomplications are especially important for salt-sensitive crops such as lettuce, for which ECBvalues 22 dSm adversely affect plant growth inloam, whereas EC,values>l.5dSm1 might nega-tively affect yield in sand.

Hammond(1966)suggestedmultiplyingECBvalues by one half the ratio, saturation percentagemoisture:fieldcapacitymoisture, tocorrect roughlyfor differences in water relationships among soils.When this type of soil moisture information is notavailable, a rule of thumb is to multiply values forsandy soils by 2 and values for muck soils by 0.5before interpretation (Hammond, 1966). More re-

cently, as part of a research program directedtoward developing technologies for measuring insitu bulk soil salinity (EC,), more accurate modelshave been developed that interrelate ECa and EC,based on selected soil properties (Rhoades 1981;Rhoades et al., 1989a, 1989b,1990).

Spatial variability The salt and Na statusof afield typically is determined from a compositesoil sample. However, such a protocol ignoresvariation in salinityacrossagivenfield. Managing10- to 20-ha units based on a composite soilsample mightaffectsalt-sensitivecropsadversely.Note the variation in EC, from across a field insouthwestern Arizona before being planted to let-tuce (Fig. 1). Lettuce yield and quality would becompromised severely in one corner of this fieldshould the entire field be managed based on theEC,of about 2.1 of a composite soil sample.

Detailed mapping of soil salinity from EC,measurement is generally imprac-tical. However, field determinationof soil-paste electrical conductivi- @Jties (EC,) (Rhoades et al. 1989a)andfieldmeasurementsofbulksoilsalinity (EC,) using four electrode ro

sensors, electromagnetic measure-ments, or time domain reflec- t btometric sensors show promise asapractical meansofassessingfieldvariability (Lesch et al. 1992;Rhoades and Oster, 1986).

Because irrigation efficien-cies in the southwestern UnitedStates are generally low (largeleaching fractions and sometimestail water) for many horticulturalcrops, field variability in salinity is

notamajorproblematpresent. However,as irriga-tion efficiencies are enhanced, site-specific tech-nologies for salt and Na management might proveeconomical.

Crop considerations

Osmotic effects. The general effects ofbulk soil salinity on plants are osmotic. High saltconcentrations in the plant rooting zone hinder theplants ability to take up water from the soil solu-tion Plants often counter this osmotic gradient byosmotic adjustment, i.e., increasing their internalsolute concentration, by producing organic acidsor accumulating salts, usually K. However, thisprocess requires energy that would otherwise beused for plant growth and development.

Tolerances to soil salinity have been estab-lishedfor most important crop species (Bernstein,1964; 1974). Most economically important horti-cultural crops have been classified as moderatelysensitive and sensitive to salinity. Specific toler-ances, with respect to EC, values for O%, 10%,25%, and 50% yield reductions, are shown forseveral economically important horticultural cropsin Table 3. A more complete list can be found inMaas and Hoffman (1977) and Maas (1984). Re-cent publications establishing tolerances for addi-tional horticultural crops (Francois, 1995)attesttothe fact that this is an ongoing area of research.

Specific toxicities. In addition to gen-eral salinity effects, many horticultural crops areaffected negatively by high concentrations of spe-cific ions such as Cl, B, and Na (Fig. 2). Croptolerances to Cl and B have been established formany economically important horticultural crops(Bernstein, 1965; Eaton, 1944; Maas, 1984; Pear-son, 1960). For example, understanding the Cltolerance for various citrus rootstocks is of utmostimportance in selecting the appropriate rootstock

Fig. 1. Variability in soil EC, values in a field in south-western Arizona (Yuma Valley) before lettuce seeding.(Unpublished data provided by Agro Phosphate Inc.)

Croo 1 0 0

Yield potential (%)

90 75 50

VegetableBroccoli (Brassica oleracea L. Italica)Tomato (lycopersicon esculentum Mill.)Cucumber (Cucumis sativus L.)Celery [Apium graveolens L. var. dulce (Mill) Pers.]Cabbage (Brassica oleracea L. Capitata group)Pepper (Capsicum annuum L. var. annuum)Lettuce (lactuca sativa L.)Radish (Raphanus sativus L.)Onion (Allium cepa L. cepa group)Carrot (Daucas carota L.)Bean (Phaseolus vulgaris L.)

FruitDate palms (Phoenix dactylifera L.)Grapefruit (Citrus paradisi Macf.)Orange (Citrus sinensis L. Osb.)Peach [Prunus persica (L.) BatschlGrape (Vitas sp.)Plum (Prunus domestica L.)Blackberry (Rubus sp.)Strawberry (fragaria sp.)

2.8 4.9 5.5 8.22.5 3.5 5.0 7.62.5 3.3 4.4 6.31.8 3.4 5.8 9.91.8 2.8 4.4 7.01.5 2.2 3.3 5.11.3 2.1 3.2 5.11.2 2.0 3.1 5.01.2 1.8 2.8 4.31.0 1.7 2.8 4.61.0 1.5 2.3 3.6

4.0 6.8 11.0 18.01.8 2.4 3.4 4.91.7 2.3 3.3 4.81.7 2.2 2.9 4.11.5 2.5 4.1 6.71.5 2.1 2.9 4.31.5 2.0 2.6 3.81.0 1.3 1.8 2.5

EC, dSm’

‘Adapted from Maas and Hoffman (1977) and Maas (1984).

for specific conditions of soil and irrigation water(Table 4).

Many horticultural plant species are verysensitive to Na. Most deciduous fruit crops, citrus,and nut crops are affected by soil ESP levels as lowas 5 (Pearson, 1960). Some vegetable crops areaffected by ESP values as low as 10. These effectsare observed even where soil physical conditionsremain favorable, suggesting the effects are directNa toxicity.

Nutritional imbalances. Salinity alsoaffects plant growth by inducing nutritional imbal-ances. For example, high concentrations ofmonovalent salts in the rooting zone can reduceplant uptake of Ca (Fig. 3). Another example isbicarbonate-induced deficiencies of the micronu-trients Fe, Zn, and Mn.

Cropping systems. Information of salin-ity tolerance also should be used in the context ofcropping strategies. For example, a common crop-ping sequence in the low desert region of thesouthwestern United States is cotton and lettuce.While cotton is generally tolerant of saline condi-tions, lettuce is not. Hence, during cotton produc-tion, soils should be managed to preclude saltaccumulation, which might adversely affect a sub-sequent crop of lettuce.

Water considerations

Water quality. The quality of water usedfor irrigation is a major consideration. The salinitystatus of water is assessed using EC measure-

ments. Based on statistical relationships devel-oped from several waters, total dissolved solids(TDS mg.liter’) can be estimated by multiplyingEC, (dSm-I) values by 640.

Evapotranspirationconcentratessaltsaddedin irrigation water. Hence, irrigation water withmodest amounts of salt can result in saline soilconditions. An assessment of irrigation water suit-ability must consider the salinity tolerance of the

crop, leaching volume achieved, and the predictedlevel of salinity in soil water from using thisirrigation water (Rhoades, 1972).

In addition to total salts, the Na concentra-tion of irrigation water is important. The Na hazardof irrigation water often is assessed using theSAR,+,, the same ratio defined previously for soilextracts (SAR,). If SARw values of irrigation waterare >7 to 10, low water infiltration and conversionof the soil to sodic conditions may occur. How-ever, the SARwalone neglects the potential sodiumhazardassociated with water high in bicarbonates.The carbonate (CO;-) and bicarbonate (HCO,) inirrigation water can precipitate Ca, thereby in-creasing the SAR of irrigation water.

Over the years, several methods have beenproposed for predicting the potential Na hazardassociated with carbonateand bicarbonate ions inirrigation water. The residual sodium carbonate(RSC) method was used commonly at one time(Eaton, 1950):

RSC = [HCO,- +CO,2-l- Ka2+ + Mg2+l [2 ]

Waters with RSC >2.5 were considered un-suitable for irrigation purposes; waters with RSCof 1.25 to 2.5 were considered marginal; andwaters <l.25 were considered safe (Wilcox et al.,1954). The RSC method assumes all HCO, pre-cipitatesand generally overpredicts the Na+ hazardof water. This method has been abandoned largelyin favor of adjusted SAR methods proposed byothers (Bower et al., 1968; Rhoades, 1968; Suarez,1981). Presently, the adjusted SAR method pro-posed by Suarez (1981) generally is favored, be-

Fig. 2. Chloride toxicity to furrow-irrigated grapes nearSafford, Ariz. (Photograph courtesy of T. Doerge.)

HortTechology - April-June 1996 6(2)

Table 4. Tolerance of citurs rootstocks to chloride in soil (C/J and irrigation water:

Rootstock

Sunki mandarinGrapefruitCleopatra mandarinRangpur limeSampson tangeloRough lemonSour orangeRonkan mandarinCitrumeloTrifoliate orangeCuban shaddockCalamondinSweet orangeSavage citrangeRusk citrangeTroyer citrange

Soil (Cl,)

25.0

15.0

10.0

meq/liter

Irrigation water (Cl,)

16.6

10.0

6.7

‘Adapted from Maas (1984).

cause it issimpler and moreaccurate than alterna-tive procedures. This equation is discussed inmoredetail belowwith respectto leaching require-ments.

Irrigation method. Irrigation method andvolume of water applied have a pronounced influ-ence on salt accumulation and distribution (Ayersand Westcot, 1989; Bernstein and Francois, 1973).Data collected from a citrus experiment conductedin southwestern Arizona show various salt distri-bution patterns as a result of different irrigationmethods(Fig.4). Flood irrigation and an appropri-ate leaching fraction generally move salts belowthe root zone. Similar results can be obtained witha properly managed sprinkler irrigation system.However, in hot, arid climates, evaporation lossduring sprinkling might increase the salinity ofwater moving into the soil (Ayers and Westcot,1989). Robinson’s (1973) study suggests that theconcentration of salts in sprinkler water for salt-sensitive crops is generally not a problem in thecenter of the field but may be a problem near theedge of the field, where evaporation losses arehigher. Another major concern with sprinkler irri-gation is the potential for leaf burn. Harding et al.(1958) reported that water normally suitable forsurface irrigation may be harmful to citrus. Theyfound that leaf burn and defoliation of the lowerpart of citrus trees can result from sprinkling waterhaving E&values of 0.8 to 1.5 dS.mP. Tolerancesof crops to foliar injury using sprinkler irrigationcan be found in Maas (1984) and Ayers andWestcot (1989).

With furrow and pressurized irrigation,soluble salts in the soil move with the wetting front,concentrating at its termination or at its conver-gence with another wetting front. In drip-irrigatedplots, water moves away from the emitter and saltsconcentrate where water evaporates. In furrow-irrigated crops, water movement is from the furrow

into the bed via capillary flow. When adjacentfurrows are irrigated, salts concentrate in the cen-ter of the intervening bed (Fig. 5). Manipulatingbedshapeandplantingarrangementarestrategiesoften used to avoid salt damage in furrow-irrigatedrow crops. This is discussed in more detail below.Because drip irrigation maintains more constantfavorable conditions of soil moisture,plants tolerate higher levels of salinitythan with furrow irrigation (Bernsteinand Francois, 1973).

Management strategies

Leaching. Leaching excesssalts and maintaining favorable saltbalance remains the best strategy toprevent detrimental salt accumulationin the soil profile. Leaching also can beused to reclaim nonsodic saline soils.Most leaching strategies are based onasteady-stateassumption of mass bal-ance and are calculated by

LF = D,,&= E&/E& [3]

where LF(leaching fraction) is the frac-tion of water that leaves the root zone asdrainage water, Ddw and Dlw are thevolumes of drainage and irrigation wa-ter, respectively, and EC,, and EC!, arethe EC of irrigation water and drainagewater, respectively. Typically ECdw isselected from crop tolerance data (suchas that presented in Table 4) to calcu-late the desired leaching fraction. Be-cause salt concentrations of saturationextracts (ECevalues) are typically moredilute than the salt concentration of theactual drainage water under field con-

ditions, using EC,values in the denominator of Eq.[3] (EC,J traditionally has exaggerated the calcu-lated leaching fraction. As noted above, informa-tion on soil texture and soil moisture can be usedto estimate bulk salinity in the rooting zone (EC,values).

This above relationship also ignores ionuptake by plants and precipitation and dissolutionreactions of carbonates in the rooting zone. Errorsassociated with the latter could have disastrouseffects when water high in sodium or bicarbonateare used for irrigation. Several approaches havebeen used to estimate leaching fractions in whichcalcium carbonate tends to precipitate or dissolvein soils. Bower et al. (1968) proposed using amodified saturation index based on Langelier’sindex. However, this equation overpredicted thesodium hazard associated with drainage waters(Oster and Rhoades, 1975). Rhoades (1968) modi-fied Langliers’s index to include empirically de-rivedparametersaimedatpredicting mineral weath-ering. It worked well for many western soils. How-ever, the value of the weathering coefficient variessomewhat with soil-water combination (O'Connor,1971; Osterand Rhoades, 1975), thus limiting its

Fig. 3. Celery blackheart, a calcium-related physiologi-caldisorder. In this case blackheart was caused by a highconcentration of monovalent salts in the rooting zone.


following:

Adj SAR,= (Na,ylLF)/I(Mg,JLF + X(P,,)‘V [4 ]

where Nalw and Mgiw are the concentrations of Naand Mg in irrigation water as mmolJiterr’, P,,, isthe partial pressure of CO, in the soil, and X isselected from a table (Suarez, 1981) based on theHCO,-/Caratioand the ionic strength of the irriga-tion water (which is estimated by ECJ Becausethisrelationship isnotverysensitivetoP,,,, roughestimates are usually satisfactory. Equation [4]attempts to address shortcomings in previousequations by calculating equilibrium pH ratherthan assuming one. This equation is simpler andgenerally more accurate than others used and hasfound widespread application in the western UnitedStates.

Drainage. A prerequisite to using leachingas a management tool is good internal and externaldrainage. Poor internal soil drainage caused bysurface crusting, hardpans, and sodic conditionsoften is managed by tillage and soil amendments.Deepchiseling isperformedat leastonceayearon

predictive potential. Suarez (1981) proposed the most cropland in the low desert of southwesternUnited States (Fig. 6). When sodic conditionsexist, an aggressivesoil amendment program usu-ally is required.

External drainage on most farmland in thewestern United States is maintained through anintricate network of open drains and tile drains. Insome situations, drainage wells are required.

Planting configuration, bed arrange-ment, and irrigation. Water movement in fur-row irrigation is from the furrow into the bed. Thiscan result in localized elevated salinity, which canharm seedling plants (Bernstein and Fireman,1957), even under low to moderately saline condi-tions. Many crops, such as lettuce and broccoli,are seeded in two rows near the outer edge of thebed with a ridge or corrugation in the center. Thisarrangement places young seedlings out of thezone of highest salt accumulation. For crops suchas cauliflower, which typically are seeded one rowper bed, plants often are offset from the bed center,or only every other furrow is irrigated during standestablishment.

stand establishment has become a widespread 1practice for vegetables produced in the low desert(Hartz, 1994). Sprinkling moves salts with waterdownward below the seed zone. Crops such assweet corn can be planted in the water-furrow orplanted in mulched moist beds to avoid salt accu-mulation associated with furrow irrigation. Moredetails on options for bed and planting configura-tions with regard to salinity management can befound in Ayers and Westcot (1989).

Otherstrategiesalsoavoid gradientscausedby furrow irrigation. Sprinkler irrigation during

FLOOD

Fertilizer management. Manyfertilizerscontainsolublesaltsin high concentrations. There-fore, nutrient source, rate, timing, and placementare important considerations in the production ofhorticultural crops. Salt indices for most commer-cial fertilizer products have been reported by Rader(1943).For example, KCI has a salt index 2.5 timesthat of K,SO,. Generally, band application of fertil-izers with high salt indices near seedlings shouldbe avoided. An interesting example where source

Fig. 4. Salt distribution patterns below a ‘Valencia’ or-ange grove in southwestern Arizona (Wellton Mesa) asinfluenced bv irrigation system. (Unpublished data pro-

1 vided by R. Roth.)

n ’ _ 2’3 4.6 - 6.9K E Y : dSm-1

2.3 - 4.6 m 6.9 - 9.2


Fig. 5. Salt distribution in furrow-irrigated melons. Notemelonsareseededon thesouthsideofthebedawayfromthe high concentrations of salts in the bed center. (Pho-tograph courtesy of T. Doerge.)

may be important, even with broadcast applica-tion, is for lettuce production in southern Florida.During prolonged periods of no rainfall, K andother salts have the potential to accumulate nearthe soil surface as a result of upward movementwith water from the subsurface irrigation systems.Field studies have shown that under these condi-tions lettuce yield reduction is more pronouncedwhenKCl isusedcompared toKZS04(C.A.Sanchez,unpublished data).

The salt content of amendments such asgypsum and manures also should be considered.For many soils in the western United States, gyp-sum application is a useful management practicefor precluding sodium accumulation on the soil’sexchangecomplex, maintaining soil structure, andimproving water infiltration. However, in a recentfield study conducted near Yuma, Ariz., gypsumreduced growth and marketable yield of lettucewhen applied immediately before seeding (C.A.Sanchez and J.G. Silvertooth, unpublished data).Hence, for salt-sensitive crops such as lettuce,gypsum should be applied so that soluble saltsreleased during dissolution do not negatively af-fect lettuce production. This probably can be ac-complished by lengthening the time interval be-tween application and planting.

Another interesting phenomenon concern-ing fertilizer management involves applying am-monia in irrigation water. Anhydrous ammoniaraises the pH of irrigation water, causing the pre-cipitation of CaCO,, volatilization of NH,, andincreased sodium hazard (Fig. 7). The simulta-neous use of acid with anhydrous ammonia elimi-nates this problem.

S o i l a m e n d m e n t s a n d w a t e r t r e a t -ments. Soil amendments and water treatmentsoften offer a practical and economical means formanaging many problems common to saline andsodic soils. Soil applications of amendments areused for initial reclamation and long-term mainte-nanceofsoil quality. Ingeneral, waterapplicationsare intended to alter the chemistry of irrigationwater such that no further degradation in soilquality will occur. Rates of amendments used forsoil application are typically large and primarilybased oneconomics. Forwatertreatments, rates ofamendments are typically much smaller and arenearly always based on solubilities and stoichi-ometry.

Amendments such as gypsum and elemen-

tal Shave been used for years. Gypsum is primarilyused on Na-affected soils as a source of Ca2+ todisplace Na+ from the soil’s colloidal exchangecomplex. The exchange flocculates soil particles(bunching of particles into larger aggregates). TheCa*+ ions reverse the effect of Na+ ions, which tendto disperse soil particlesand restrict water infiltra-tion The resulting displaced Na+ ions are leachedreadily from the soil profile. Gypsum is a neutralsalt that does not directly reduce pH. However, itcan indirectly lower the pH of sodic soils byreducing the hydrolysis reactions associated withNa ions on the exchange complex. Gypsum oftencan reduce surface sealing and improve waterinfiltration in nonsodic soils by releasing electro-lytes into percolating water (Ben-Hur et al., 1992;Shainberg et al., 1990; Warrington et al., 1989).

Elemental S is also a common soil amend-ment. Sulfur added to soil undergoes a slow bio-chemical oxidation to sulfuric acid. This affects thesoil first by neutralizing soil bases and loweringpH directly, and second by dissolving native soilCaCO, to form gypsum, which then acts as de-scribed above. Iron pyrites also have been sug-gested as soil amendments (McGeorge andBreazeale, 1955). Like elemental S, iron pyritesoxidize to sulfuric acid (Banath and Holland, 1976).

Sulfuric acid often is applied to produce aneffect more immediate than that produced by el-emental S (Miyamoto et al., 1974; 1975a). Sulfuricacid can be added directly to the soil by specializedequipment or used as awatertreatment. Studies inArizona showed that water infiltration and seedlingemergence were increased by gypsum, elementalS, or acid applied to the soil (McGeorge et al.,

Fig. 6. Chisel used routinely in the southwestern UnitedStates.

HortTechnology * April-June 1996 6(2)

Fig. 7. Precipitation of carbonates in irrigation ditchesafter application of ammoniacal fertilizer. (Photographcourtesy of J. Stroehlein.)

1956). Because of the corrosive nature of sulfuricacid, other acid materials or acid-forming prod-ucts have been used successfully as less hazard-ous substitutes. Acid or acid-forming amend-ments often are used to obtain the added benefit ofenhanced P and micronutrient availability(Stroehlein and Pennington, 1986). Sulfuric acidand other acid-forming liquid materials have beenhighly effective when added to irrigation water asa water treatment (Stroehlein and Pennington,1986; Yahia et al., 1976). Acid added to irrigationwater not only improves infiltration by increasingthe electrolyte concentration, but it reduces thecarbonate and bicarbonate hazard of irrigationwaters (Miyamoto et al., 1975b).

Calcium polysulfide and ammonium polysul-fide also are used as water treatments to improveinfiltration (Stroehlein and Pennington, 1986).Thesematerialsare initially basic and reacttoformcolloidal S. The Ca2+ or ammonium (NH,‘) ion canreplace Na+ on the exchange complex. Ammoniumand S are then oxidized to form acid (H’). Studieshave shown improved infiltration with these prod-ucts (El-Tayib et al., 1979; Cairns and Beaton,1976). Not all of the beneficial effects can beattributed to S and ammonium oxidation. One fieldstudy in Arizonashowed an immediate increase ininfiltration before oxidation could occur (Stroehleinand Pennington, 1986), probably due to electro-lyte effects.

Syntheticpolymerswerefirst investigated inthe 1950s and 1960s. An extensive series of stud-ies conducted in Arizona with polymers availableat the time resulted in reduced soil compaction,improved stands, and increased yields compared

to untreated controls (Fuller et al., 1953) Althougheffective, most of these products evaluated in the1950s and 1960s could not be used economicallyunder field conditions.

Synthetic polymers such as polyacrylamide(PAM), polyvinyl alcohol (PVA), and polysaccha-rides have more recently shown promising resultsin research studies. Reduced soil surface crusting(Helaliaand Letey, 1989; Terry and Nelson, 1986;Wood and Oster, 1985) improved aggregationand reduced clay dispersion (Aly and Lety, 1988)enhanced stand establishment (Cook and Nelson,1986; Wallace and Wallace, 1986) reduced soilerosion with furrow and sprinkler irrigation (BenHur et al., 1990; Lentz et al., 1992; Levy et al.,1992) and improved reclamation of saline andsodic soils (Wallace et al., 1986) have been re-ported with the use of synthetic polymers.

Severalstudiesin Israel and the United Statesshowed improved infiltration with PAM and otherpolymers (Ben-Huret al., 1989; Levyetal., 1992).In most of these studies slaking and dispersionwere the dominant mechanisms limiting waterintake. However, onestudyshowed polymersto beeffective in reducing clay swelIing (Emerson, 1963).Shainberg et al. (1990) found that polymers in-creased water infiltration 3-fold over untreatedsoils. When gypsum was added with the polymer,infiltration increased another 3-fold over the poly-mer treatment alone.

Some researchers reported that syntheticpolymers may be useful tools for managing andreclaiming sodic soils. Allison (1952) reported anincrease in aggregate stability in several saline-sodic soils that were treated with polymers.Wallaceet al. (1986) found that PAM added to a sodic soilimproved soil aggregation, increased water pen-etration, and improved crop seedling emergence.Other researchers, however, found that polymersused alone were less effective in sodic soils. Aly

and Lety (1990) found that the stability of aggre-gates treated with polymers decreased as the soilSAR increased from 1 to 15. Zahow and Amrheim(1992)foundincreased hydraulicconductivityfrompolymer addition when the soil had an ESP <15,but found little or no increase when the soil ESPexceeded 15. However, these workers did find thatsynthetic polymers used in combination with gyp-sum increased hydraulic conductivity 4-fold overgypsum alone. These results underscore the needfor a Ca source when reclaiming sodic soils andillustrate the apparent synergistic effects of gyp-sum and synthetic polymers.

Conclusions

Many economically important horticulturalcrops are sensitive to soil and water salinity and tothe deterioration of soil physical properties asso-ciated with Na in soil and irrigation water. There-fore, soil chemical and physical properties, croptolerance, water quality, fertilization, and irrigationmethod are important considerations for the pro-duction of horticultural crops. This is especiallytrue in the southwestern United States, wherewarmwintertemperaturesfavorproductionofmanyfruit and vegetable crops, but problems of salinityand sodicity are commonplace.

For purposes of organization we have out-lined important considerations individually, but inreality they must be considered concurrently. Overthe years, researchers have developed and cali-bratedanumber of useful diagnosticand prognos-tic tools for dealing with salt and Na-related prob-lems. Once problems (or the potential for prob-lems) are identified, profitable production of hor-ticultural crops depends on the selection of aneconomically viable management strategy. Thismay simply involve a minor, inexpensive modifi-cation of cultural practices underconditionsof lowto moderate salinity, or more costly reclamationunder conditions of high Na.

Literature Cited

Allison, L.E. 1952. Effect of synthetic polyelectro-lyteson thestructureofsalineandalkali soils. SoilSci. 73:443454.

Aly, SM. and J. Letey. 7988. Polymer and waterquality effects on the flocculation of montmorillo-nite. Soil Sci. Soc. Amer. J. 52:1453-1458.

A/y, SM. and J. Letey. 1990. Physical propertiesof sodium treated soil as affected by two polymers.Soil Sci. Soc. Amer. J. 54:501-504.

Ayers, RS. and D. W. Westcot. 1989. Water qualityfor agriculture. FAO lrr. Drainage Paper 29.

Banath, CL. and JF Holland. 1976. Iron pyrites


as a sulfur fertilizer in an alkaline soil. Austral. J.Agr. Animal Husb. 16:376-381.

Ben-Hur, M., J. Faris, M. Malik, and J. Lety. 1989.Polymers as soil conditioners under consecutiveirrigations and rainfall. Soil Sci. Soc. Amer. J.53:1173-1177.

Ben Hur, M., J. Lety, and I . Shainberg. 1990.Polymer effects on erosion under laboratory rain-fall simulator conditions. Soil Sci. Soc. Amer. J.54:1092-1095.

Ben-Hur, M., R. Stern, A.J. van der Merwe, and I .Shainberg. 1992. Slope and gypsum effects oninfiltration and erodability of dispersive andnondispersive soils. Soil Sci. Soc. Amer. J.56:1571-1576.

Bernstein, L. 1964. Salt tolerance of plants. U.S.Dept. of Agr. Info. Bul. 283.

Bernstein, L. 1965. Salt tolerance of fruit crops.U.S. Dept. of Agr. Info. Bul. 292.

Bernstein, L. 1974. Crop growthandsalinity. In: J.van Schilfgaarde (ed.). Drainage for agriculture.Agronomy 17:39-56..

Bernstein, L. and M. Fireman. 1957. Laboratorystudies on salt distribution in furrow irrigated soilwith special reference to pre-emergence period.Soil Sci. 83:249-263.

Bernstein L. and L.E. Francois. 1973. Compari-sons of drip, furrow, and sprinkler irrigation. SoilSci. 115:73-86.

Bohn, H.L., B.L. McNeal, and G.A. O’Connor.1985. Soil chemistry. 2nd ed. Wiley, New York.

Bohrer, V.L. 1970. Ethnobotanical aspects ofSnaketown, a Hohokam village in southern Ari-zona. Amer. Antiquity 35:413-430.

Bower, C.A., G. Ogata, and J.M. Tucker. 1968.Sodium hazard of irrigation waters as influencedby leaching fraction and by precipitation or solu-tion of calcium carbonate. Soil Sci. 106:29-34.

Cairns, R.R., and J.D. Beaton. 1976. Improving asolonetzic soil by nitrogen-sulfur materials. Sul-fur Inst. J. 12:10-12.

Cook, D.F. and S.D. Nelson. 1986. Effect of poly-acrylamide on seedling emergence in crust-form-ing soils. Soil Sci. 141:328-333.

El-Tayib, YM,, J. Lety, and R. Branson. 1979.Sulfur compounds in water treatment-effect oninfiltration rate. Sulfur 3:7-11.

Eaton, F.M. 1944. Deficiency, toxicity, and accumu-lation of boron in plants. J. Agr. Res. 69:237-277.

Eaton, F.M. 1950. Significance of carbonates inirrigation waters. Soil Sci. 69:123-133.

Emerson, W. W. 1963. The effect of polymers on theswelling of montmorillonite. J. Soil Sci. 14:52-63.

Francois, L.E. 1995. Salinity effects on bud yieldand vegetative growth of artichoke (Cynarascolymus L.). HortScience 30:69-71.

Fuller, W. H., N. I/. Gomness, and L. V. Sherwood.1953. The influence of soil aggregate stabilizerson stand, composition, and yield of crops oncalcareous soils of southern Arizona. Univ. Ari-zona Tech. Bul. 129.

Hammond, L.C. 1966. Soil-water-salinity-plantrelationships, Proc. Soil and Crop Sci. Soc. Fla.26:257-264.

Harding, R. B., M. P. Miller, and M. Fireman. 1958.Absorption of salts by citrus leaves during sprin-kling with water suitable for surface irrigation.Proc. Amer. Soc. Hort. Sci. 71:248-256.

Hartz, TK 1994 Minimizing environmental stress infield establishment ofvegetablecrops in the southwest-ern United States. HortTechnology 4:29-34.

Helalia, A. M. and J. Letey. 1989. Effects of differentpolymers on seedling emergence, aggregate sta-bility, and crust hardness. Soil Sci. 148:199-203.

Lentz, R.D., I . Shainberg, R.E. Sojka, and D.L.Carter. 1992 Preventing irrigation furrow erosionwith small applications of polymers. Soil Sci. Soc.Amer. J. 56:1926-1932.

Lesch, SM., J.D. Rhoades, L.J. Lund, and D.L.Corwin. 1992. Mapping salinity using calibratedelectromagnetic measurements.. Soil Sci. Soc.Amer. J. 56:540-548.

Levy, G.J, J. Levin, M Gal, M Ben-Hur, and I .Shainberg. 7992. Polymers’ effects on infiltrationand soil erosion during consecutive simulated sprin-kler irrigations. Soil Sci. Soc. Amer. J. 56:902-907.

Maas, E.V. 1984. Salt tolerance of plants. In: B.R.Christie (ed.). The handbook of plant science inagriculture. CRC Press, Boca Raton, Fla.

Maas, EV: and G.J. Hoffman. 1972 Crop salttolerance-Current assessment. Amer. Soc. CivilEng. Proc. J. lrr. Drainage. Div. 103:115-131.

McGeorge, W.T. and E.L. Breazeale. 1955. Thevalue of pyrite and Pyrhotite as soil conditioners.

Univ. Arizona Agr. Expt. Sta. Rpt. 124.

McGeorge, W. T., EL. Breazeale, and J. T. Abbott.1956. Polysulfides as soil conditioners. Univ.Arizona. Bul. 131.

Miyamoto, S., H. L. Bohn, J. Ryan, and M.S. Yee.1974. Effectsofsulfuricacidandsulfurdioxideonaggregate stability of calcareous soils. Soil Sci.118:299-303.

Miyamoto, S, R.J. Prather, and J. L. Stroehlein. 1975a.Sulfuric acid and leaching requirement for reclaimingsodium-affected soils. Plant Soil. 43:773-585.

Miyamoto, S., R.J. Prather, and J.L. Stroehlein.1975b. Sulfuric acid for controlling calcite pre-cipitation. Soil Sci. 120:264-271.

O’Connor, G.A. 1971. Testing sodium hazard pre-dictions. Soil Sci. Soc. Amer. Proc. 35:510-512.

Oster, J.D. and J.D. Rhoades. 1975. Calculateddrainage compositions and salt burdens resultingfrom irrigation with river waters in the westernUnited States. J. Environ. Qual. 4:73-79.

Pearson. G.A. 1960. Tolerance of crops to ex-changeable sodium. USDA Info. Bul 216.

Rader, L.F. 7943. Salt index for fertilizers. Soil Sci.55:201-208.

Rhoades, J.D. 1968. Leaching requirement forexchangeable sodium control. Soil Sci. Soc. Amer.Proc. 32:652-656.

Rhoades, J,D. 1972. Quality of water for irrigation.Soil Sci. 113:277-284.

Rhoades, J.D. 1981. Predicting bulk soil electricalconductivity versus saturation paste extract elec-trical conductivity calibrations from soil proper-ties. Soil Sci. Soc. Amer. J. 45:42-44.

Rhoades, J.D. and J.D. Oster. 7986. Solute con-tent, p, 985-1006. In: A. Klute (ed.). Methods ofsoil analysis. Part 1. 2nd ed. Monog 9. ASA andSoil Sci. Soc. Amer., Madison, Wis.

Rhoades, J. D., N.A. Manteghi, P. J. Shouse, andA.J. Alves. 1989a. Estimating soil salinity fromsaturated soil-paste electrical conductivity. SoilSci. Soc. Amer. J. 53:428-433.

Rhoades, J. D., N.A. Manteghi, P.J. Shouse, andWJ. Alves. 1989b. Soil electrical conductivity andsoil salinity: New formulations and calibrations.Soil Sci. Soc. Amer. J. 53:433-439.

Rhoades, J. D., P.J. Shouse, WJ. Alves, N.A.


Manteghi, and S.M. Lesch. 1990. Determiningsoil salinity from soil electrical conductivity usingdifferent models and estimates. Soil Sci. Soc.Amer. J. 54:46-54.

Robinson, L.A. 1973. Increase in conductivity ofirrigation water during sprinkling. Agron. J. 65:130.

Shainberg, I., D. N. Warrington, and P. Rengasanmy.1990. Water quality and PAM interactions in re-ducing surface sealing. Soil Sci. 149:301-307.

Sposito, G. and S. V. Mattigod. 1977. On the chemi-cal foundation of the sodium absorption ratio. SoilSci. Soc. Amer. J. 41:323-329.

Stroehlein, J, L. andD.A. Pennington. 1986. Use ofsulfur compounds for soil and irrigation watertreatments. Agronomy 27:435-454.

Suarez, D.L. 1987. Relationship between pHcandsodium adsorption ratio (SAR) and an alternativemethod of estimating SAR of soil or drainagewaters. Soil Sci. Soc. Amer. J. 45:469-475.

Terry, R.E and S.S. Nelson. 1986. Effects of poly-acrylamide and irrigation method on soil physicalproperties. Soil Sci. 141:317-320.

Wallace, A. and G.A. Wallace. 1986 Effects of soilconditioners on emergence and growth of tomato, cot-ton, and lettuceseedlings. Soil Sci. 141:313-316.

Wallace, A, G.A. Wallace, and A.M. Abouzamam.7986. Amelioration of sodic soils with polymers.Soil Sci. 141:359-362.

Warrington, D., I. Shainberg, M. Agassi, and J.Morin. 1989, Slope and phosphogypsum’s effecton runoff and erosion. Soil Sci. Soc. Amer. J.53:1201-1205.

Wilcox, L. I/., G. Y Blair, and CA. Bower. 1954.Effect of bicarbonate on suitability of water forirrigation. Soil Sci. 77:259-266.

Wood, J.D. and J.D. Oster. 1985. The effect ofcellulose xanthate and polyvinyl alcohol on infil-tration, erosion, and crusting at different sodiumlevels. Soil Sci. 139:243-249.

Yahia, TA., S. Miyamoto, and J. L. Stroehlein.7975. Effect of surface applied sulfuric acid onwater penetration into dry calcareous and sodicsoils. Soil Sci. Soc. Amer. Proc. 39:1201-1204.

Yenson, N.P. and J.L. Bedell. 7993. Consider-ations for the selection, adaptation, and applica-tion of halophyte crops to highly saline desertenvironments as exemplified by the long-termdevelopment of cereal and forage cultivars ofDistichlis spp. (Poaceae), p. 305-313. In: H. Leithand A. Al Masoom (ed.). Toward the rational use ofhigh salinity tolerant plants. vol. 2. Kluwer Aca-demic Publishers, The Netherlands.

Zahow, M.F. andC. Amrheim. 1992 Reclamationofasalinesodicsoil usingsyntheticpolymersandgypsum. Soil Sci. Soc. Amer. J. 56:1257-1260.

Preferential FlowConcepts toHorticultural WaterManagement

John S. Selkerl

Additional index words. macropore,finger, funnel, contamination, groundwater

Summary. Avoiding groundwatercontamination from agricultural activitiesis possible only if the processes thatcontrol deep percolation are understood.The source of contaminant movement togroundwater is typically through preferen-tial flow, processes by which the bulk soilis bypassed by some part of the infiltratingwater. Three mechanisms give rise topreferential flow: fingered flow, funnelflow, and macropore flow. Fingered flowoccurs in coarse-textured soils and can beminimized by starting with an initially well-wetted profile. Funnel flow is likely inlayered soil profiles of silt or coarser-textured soil,.in which avoiding slow over-irrigation is critical. Macropore flow isobserved in all structured soils in whichmaintaining irrigation rates well below thesaturated conductivity of the soil isessential. These prescriptions are quitedifferent than conventional recommenda-tions, which fail to consider groundwaterprotection.

T he quantitative study of the move-ment of water through soils iswell over 100 years old (e.g.,Lawes et al. 1882) and contin-ues to be an area of active research.

This interest generally stems from two concerns:

lDepartment of Bioresource Engineering, Gilmore Hall,Oregon State University, Corvallis, OR 97331-3906.The cost of publishing this paper was defrayed in partby the payment of page charges. Under postalregulations, this paper therefore must be herebymarked advertisement so/e/y to indicate this fact.

1) how to manage soil water to obtain the greatestcrop response and production; and 2) how tominimize the impact of cropping activities on theenvironment.

Until the 1970s most of the research con-cerning soil water movement concentrated on at-temptstopredictflowin homogeneous soils(Gard-ner, 1958; Green and Ampt, 1911; Philip, 1969).Thisapproach was used not because it was thoughtto be the most realistic representation of soils, butbecause it allowed the application of powerfulmathematical tools to solve some fundamentalproblems. In the 1970s the first widespread ob-servations of groundwater contamination werenoted. Although the most famous of these cases(e.g., Love Canal) were due to industrial sources,agriculturalactivitiesalso have been implicated. Inmanyofthesecasescontamination would not havebeen predicted to reach the aquifer using modelsthat assumed homogeneous soil and uniform wa-ter content. These realizations led to the develop-ment of more realistic and complete descriptionsof solute movement through soils. The unac-counted for and widespread behavior of watermoving unevenly through soil is described aspreferential flow. These processes allow some oftheinfiltratingwatertopercolatemorequicklythanif the soil were uniformly carrying flow.

This article summarizes the research find-ings that explain these transport processes, pro-viding horticulturists guidelines to enhance effi-cient use of water, increase efficiency of saliniza-tion control, and minimize groundwater contami-nation

Fingered flow

Hill and Parlange (1972) noted that in coarse-textured soils water tends to move in isolatedregions, or fingers of flow. This process is recog-nized as an instability in the wetting front, analo-gous to the instability in the darting tongues of aflame, or the flapping of a flag. This process wassoon put into the mathematical frameworkof linearinstability theory (Parlange and Hill, 1976) whichhas since been shown to provide reasonable esti-mates of the physical dimensions (i.e., width) ofthese fingers of flow (Glass et al., 1989; Selker etal., 1992).

From a practical point of view, the fact thatthe width of fingers can be predicted is useful. Itconfirms that our conceptual model for fingeredflow is capturing the physical basis of the process.From this position of confidence, we then can usethis result to see where fingers are likely to beprominent. The size of fingers is related inverselytothecharacteristicgrainsizeofamedium(Fig.1):when the soil hasatextureofsiltorfiner, thefingerdimensions are predicted to be >l m. Fingeredflow occurs only in unstructured soils. Hence,fingered flow is expected only in soils that arepredominantly sand.

Finger width is not strongly affected by the


flux through the system (Fig. 1). Rather, the fluxthrough the system typically affects the number offingers that form. When the flux is increased up tothe rate of the saturated conductivity of the soil, thefingers will grow in width and frequency to thepoint at which they finally merge to yield a flatwetting front without fingered properties. Sincemostsandysoils haveextremelyhighconductivitycompared to naturally occurring infiltration rates,this rarely occurs in natural conditions.

Understanding effects of prior moisture con-tent on finger formation and finger persistence isimportant in managing production on sandy soilsAs shown by Lui et al. (1993), when soil is at fieldcapacity, fingers will be about IO-fold wider thanthose developing in dry conditions. Thus, if asandy soil is not allowed to dry completely, thebypassing effect will be reduced, if not entirelyeliminated. I saw a pertinent example of the impor-tance of this observation in Florida. Using a dripsystem, when irrigation was initiated before thefield had significantly dried in the spring, the year'scrop did well. In years when irrigation was delayeduntil the soil was dried, narrow fingers formedthrough the root zone, leaving most of the soil dry,thus damaging the crop.

Once a finger has formed in a particularlocation, it will remain (persist) until the soil haseither dried entirely or has been completely satu-rated (Glass et al., 1989). Given the very highconductivity of sandy soils, eliminating persistentfingers requires either drying from the surface,which is effective to a maximum depth of about 1m, or raising the water table. A striking example ofthe impact of persistence was seen in Lafayette,Ind. (Fig. 2). Distinct fingers of flow were madevisible in the coarse material due to the chemicalweathering of the calcareous material. Within thefingers, pebbles were readily crushed manually,while outside of the fingers the gravel strength wastypical of unweathered material. Clearly these fin-gers had persisted for many decades.

It is crucial when initiating irrigation on dry,sandy soils that the rate of irrigation is nearly thatof the saturated conductivity of the soil, so theprofile starts off from an entirely wetted condition.From this initial state, any fingers will be quitelargeandlikelynotaproblemfromaproductionorcontamination point of view. If problematic fin-gered flow is found, it can be eliminated onlythrough complete soil drying or saturation. Giventhe length of time required for total drying, achiev-ing a brief period of saturation is the most practicalapproach to eliminate fingered-flow pathways.

Macropore flow

Aside from very sandy soils, almost all othersoils have macroscopic structure. In addition togrouping soil particles into larger units, there arestructures that arise from plant and fauna activi-ties, leaving root channels, worm holes, and ani-mal burrows. These structural elements consist of

pores that are generally several orders of magni-tude larger than the characteristic grain size of thesoil. Forinstance, inasiltysoilwithamedian poresize of 1 x~O-~ m, the interpedfacespacing wouldbe of the order of 1 x 10” m, and it would havedeep root channels 1 x 1O-3 m in diameter. Suchconnected systems of large pores are referred to asmacropores.

Before the 198Os, these soil features werenot considered in the prediction of movementthrough soils. Under most natural rainfall condi-tions the water would go through the bulk soil,never reaching a point of saturation at which thesepores would fill. However, during intense rainfalland typical irrigation conditions, the soil is takentoastateof nearsaturation, and these poresdofill.This results in a dramatic increase in conductivity,as the resistance to flow decreases with the squareof the aperture size. Thus, such macroscopic fea-tures can dominate flow during periods of intenseinfiltration (a single l-mm pore will carry as muchwater as about 1 m2 of uniformly packed silttextured media). Beyond the ability to allow greaterinfiltration, these poresallow chemicals to bypassthe bulk of the soil. In this way, chemicals thatwould be predicted to move slowly through the soilprofile can move quickly to shallow aquifers.Broadlyreferred toasmacroporeflow, this mecha-nism has been studied widely in the last 2 decades,and has been shown to have a major impact on thetransport properties of most soils (German andBeven, 1981; Gish and Jury, 1983).

Funnel flow

Many soils are generated by deposition ofmaterial by either processes of wind, water, orvolcanic activity, yielding a layered character withalternating series of finer and coarser texture duetovariation inthesourceofthematerialandenergyof the flow at the time of deposition. These layershave a strong effect on infiltration of water due tothe capillary properties that depend on the pore

size distribution in each layer. Although soils withcoarser texture will typically have higher saturatedconductivity, this rule breaks down in the contextof unsaturated flow.

To understand this, it is useful to visualize apair of stacked sponges, where the bottom spongeis of very poor quality with large pores, and the topsponge is of excellent quality with very fine pores.Ifthestackofspongesisataslightangleandwateris poured slowly onto the stack, the high-qualitysponge will retain the water and let it flow over thelower-quality sponge. Similarly, in the layered soilsystem, the finer-textured pores will retain infil-trating water and divert the water over the coarserlayer. This process, known as funnel flow, canintercept infiltration from broad areas into focusedstreams of infiltration (Kung, 1990). In a potatofield studied in Wisconsin, water landing on anarea of 2500 m2 was focused to an area <0.25 m2by the time it reached the water table at a depth of6 m (Kung, 1990). This 10,000-fold reduction ofareaofflowgivesrisetoaproportionallyincreasedvertical transport velocity, dramatically reducingthe time of travel of water and solutes from the soilsurface to unconfined aquifers.

Management guidelines

As shown above, there are processes thatcan cause a portion of the water to leave the rootzone rapidly through a small portion of the profilein soils of all textures. The key to preventinggroundwater contamination is to use our under-standing of these flow processes as a parameter inour irrigation management.

Traditionally the irrigation system design-ers’ concerns were dominated by the following:

Fig. 1. Dependence of finger width on the characteristic

1grain size of a soil and the relative flux (q/Ks) into thesystem (using the results of Parlange et al., 1990)

1000 j

Flux Ratio = 0.4 (Very High) FI ux)Flux Ratio= 0.01 (Typical Value)

10

Ll

SIR

1

Coarse

.OlCharacteristic Part&e Size (mm)


1. Selecting a system that will work with thetopography of the land (e.g., can’t use floodirrigation on rolling hills).

2. Selecting a system that suits the crop-machinery system the farmer uses.

3. Choosing application rates that suit theinfiltration capacity of the soil (in floodirrigation, this means that the applicationrate should be much higher than the infiltra-tion capacity).Irrigation efficiency was typically of para-

mount interest, as water is typically a scarce com-modity, either due to pumping costs, or overallsupply.

In the past 3 decades, preventing environ-mental degradation has joined the list of decisiveconcerns to agriculturists. Irrigation system de-signers have aided in environmental protectionthrough improved irrigation uniformity: If afield isnot being irrigated uniformly, and yet all points

Fig. 2. R. Bryant stands next to glossic features in a soilbelieved to have been formed as a result of fingered flowthat persisted for periods of hundreds, if not thousands,of years. The apparent difference in coloration in thefingers is not due to moisture content, but results fromthe extensive chemical weathering of the m a t e r i a l withinthe fingers.

become fully irrigated, there will be regions ofexcess irrigation that will contribute percolation tothe groundwater. This line of argument still onlyconsiders movement in theabsence of preferentialflow, and few have addressed the issue of howunderstanding preferential flow should influenceirrigation operation.

To address the impact of preferential flow onirrigation, consider the case of irrigating a coarsesoil. There are two forms of preferential flow thatshould be expected: fingered flowandfunnel flow.Irrigation on coarse-textured soils typically usesoverhead sprinkler or drip irrigation, as surfaceirrigation methods (e.g., flood and furrow irriga-tion) are not amenable to the high infiltrationcapacities of these soils. At the start of the irriga-tionseasonthefirstconsiderationmust betoavoidfinger formation in the root zone by ensuring thatthe root zone starts in a moist condition. If the soilis starting from a dry condition, a burst of irrigationat a rate close to saturated conductivity of the soilof sufficient amount to wet the root zone will eraseany residual fingers and render the profile far lesssusceptible to new finger development. From thispoint, the profile should not be allowed to drycompletely during the irrigation season.

Using a drip irrigation system, such a pro-cess can involve great additional expense and,

therefore, be impractical. In fingered flow, waterfrom drippers is likely to go vertically downward ina path ofareaequal to thesaturated conductivity ofthe soil divided by the application rate. For in-stance, a 5-liter.hh (0.005 m3.h-‘) emitter in acoarse soil with conductivity of 1 m.h-’ will keep acolumn of soil wet with a cross-sectional area ofabout 50 cm*. If the plant roots go to a depth of 0.5m, the irrigated volume of soil is just 2.5 liters (todetermine finger size on a site, simply drip about20 liters of dyed water onto an area of completelydry sand and measure the diameter of the wettedarea through excavation). If this soil had a fieldcapacityofabout 10%, anyappIicationabove0.25liter would be lost below the root zone. In such acase the system must be operated in bursts of 3min at a time; long enough to refill 0.25 liter ofholding capacity. The next irrigation would berequired when the plant had transpired this vol-ume. This high-frequency, low-volume approachis well adapted to drip irrigation, but, if ignored,water waste and aquifer contamination, particu-larly if chemigation is used, is likely.

Still considering coarse soils, what are theprescriptions to handle the possibility of funnelflow? Two management practices can be used tominimize the effects of funnel flow. The first prac-tice is applying high rates of water. The funnelingonly occurs when the flow is slow enough for thefine layers to carry water laterally under unsatur-ated conditions, and the high rate of applicationlets water cross through the fine layers into thecoarse. The second practice is applying water fora short duration with high frequency. Irrigating tokeep the top 30 cm well wetted would be suitablefor most annual crops. If a longer period of irriga-tion is used, the water will end up being restrictedby an upper, fine layer, then funneled by a lower,fine layer. The worst prescription for sites withlayered soils is slow over-irrigation.

In the case of finer, structured soils,macropore flow must be avoided. Macropore flowoccurs when water isappliedat rates thatapproachthe infiltration capacity of the soil in a givenhorizon. In this case, the prescription is exactlyopposite of that for coarse soils: apply the water asslowly as required to avoid saturating the soil. Inthe case of a clay soil, macropore flow was elimi-nated by reducing the irrigation rate to 0.0003m.h-’ (Selker et al., 1995). These extremely lowrates of irrigation may be achieved by pulsing theirrigation system or by using emitters designed toprovide continuous low-rate irrigation,

Conclusions

Many aquifers have been contaminatedthrough ignorance of the possibility of rapid flowof water through soils. Such preferential flow canbe avoided by following a few simple rules thatdepend on the nature of the soil at the site. Theserules are different from what one might expectusing native intuition, and thus have long been

Hort Technology - April-June 1996 6(2)

violated by irrigation system designersand opera-tors When understood, preferential flow can beavoided to save water and protect aquifers under-lying agricultural lands.

Literature Cited

Gardner, W.R. 1958. Some steady state solutionsof the unsaturated moisture flow equation withapplication to evaporation from a water table. SoilSci. 85:228-232.

Gish, T.J. and WA. Jury. 1983. Effect of plant rootsand root channels on solute transport. Trans.ASAE 26(2):440-444, 451.

German, P. and K. Beven. 7981. Water flow in soilmacropores. I. An experimental approach. J. SoilSci. Soc. 32:1-13.

G/ass, R.J, T.S. Steenhuis, and J.-Y Parlange.1989. Mechanism for finger persistence in homo-geneous, unsaturated, porous media: Theory andverification. Soil Sci. 148(1):60-70.

Green, W.H. and G.A. Ampt. 1911. Studies in soilphysics, I, The flow of water and air through soils.J. Agr. Sci. 4:1-24.

Hill, DE. andJ.-Y. Parlange. 1972 Wetting frontinstability in layered soils. Soil Sci. Soc. Amer.Proc. 36:697-702.

Kung, K-J.S. 1990 Preferential flow in a sandyvadose soil: I. Field observations. Geoderma46:51-58.

Lawes, J.B., J. H. Gilbert, and R. Warrington. 1882.On the amount and composition of rain and drain-agewaterscollectedat Rothamsted. William Clowesand Sons, London.

Lui, Y., B.R. Bierck, J.S. Selker, T.S. Steenhuis,andJ-Y. Parlange. 1993. High intensity x-ray andtensiometer measurements in rapidly changingpreferential flow fields. Soil Sci. Soc. Amer. J.57:1188-1192.

Parlange, J.-Y. and DE. Hill. 1976. Theoretical

analysis of wetting front instability in soils. SoilSci. 122:236-239.

Parlange, J.-Y, R. J. G/ass, and TS. Steenuis.1990. Application of scaling to the analysis ofunstableflow phenomena. Scaling in soil physics.Spec. Publ. 25, Soil Sci. Soc. Amer., Madison,Wis.

Philip, J.R. 1969. Theory of infiltration. Adv.Hydrosci. 5:215-296.

Selker, J.S., TS. Steenhuis, and Y.-J. Parlange.7992. Fingered flow in unlayered media underuniform irrigation. Soil Sci. Soc. Amer. J. 56:1346-1350.

Selker, JS., W Cao, and R. Roseberg. 1995. Useof ultralow rate application devices to eliminatemacropore flow during irrigation. Proc. 5th Intl.Microirr. Congr., 2-6 Apr. 1995, Orlando, Fla.ASAE Publ. 4-95. p, 54-59.


Documents

1376 Salt Management Workshop Proceedings › arsuserfiles › 20360500 › pdf_pubs › P137… · 1994). In the coastal areas of the United States, the gradual intrusion of seawater