The phytoavailability of potassium to small grains as influenced by edaphic and environmental factors by Robert Olson Miller A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Crop and Soil Science Montana State University © Copyright by Robert Olson Miller (1988) Abstract: Soil characteristics and rhizosphere conditions greatly influence plant available potassium (K). Research in Montana has shown crop response to K fertilizers on soils testing high in ammonium acetate extractable K (Kex). The objective of this research was to use an in-vitro technique to determine: (1) plant available K on Montana soils and its relation to soil properties; (2) the influence of temperature and water potential on K uptake; and (3) the influence of K additions on plant available K. Soil characteristics which were found to be strongly related to plant available K (R2 0.93) are: clay content, Kex diffusion gradient coefficient, and cation exchange capacity. Spring wheat in these studies utilized a portion of Kex, but quantities differed across soils. This data suggests that spring wheat utilizes significant quantities of soil K from nonexchangeable and K-mineral sources. Spring wheat K uptake increased exponentially with temperature with little or no K absorbed at 8° C. Water potentials alone had little influence on K uptake. The interaction with temperature suggests that K uptake was not limited by water potential (water content) until temperature exceeded 16° C. Thus until spring wheat K demand reached a critical K demand K uptake was not limited by water potential. Potassium additions increased spring wheat K uptake nonuniformly across soils. Greatest K uptake occurred on the Kevin soil and the least on the Beaverton. Data on soil rhizosphere Kex indicated K additions enhanced the availability of Kex supplies in the near rhizosphere (0 - 0.5 mm). Soil Kex flux data indicated that K additions did not necessarily increase Kex utilization. These results indicate that phytoavailable K is governed primarily by three soil factors and that environmental water potential play important roles as regulators. Spring wheat utilizes soil K forms other than that explained by Kex in the rhizosphere, and that the quantity utilized is highly soil dependent. THE EHYTOAVAIIABIHTY OF POTASSIUM TO SMALL GRAINS AS INFLUENCED BY EDAPHIC AND ENVIRONMENTAL FACTORS by Robert Olson Miller A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Crop and Soil Science MONTANA STATE UNIVERSITY Bozeman, Montana November 1988 APPROVAL of a thesis submitted by Robert Olson Miller This thesis has been read by each member of the thesis committee and has been found to be satisfactory regarding content, English usage, format, citations, bibliographic style, and consistency, and is ready for submission to the College of Graduate Studies. Date Chairperson, Graduate Committee Approved for the Major Department Date Approved for the College of Graduate Studies /2.//O - Z g ? Date Graduate Dean (c) COPYRIGHT by Robert Olson Miller 1988 All Rights Reserved iii STATEMENT OF PERMISSION TO USE In presenting this thesis in partial fulfillment of the requirements for a doctoral degree at Montana State University, I agree that the Library shall make it available to borrowers under rules of the Library. I further agree that copying of this thesis is allowable only for scholarly purposes, consistent with "fair use" as prescribed in the U.S. Copyright Law. Requests for extensive copying or reproduction of this thesis.should be referred to Ifiiiversity Microfilms International, 300 North Zeeb Road, Ann Arbor, Michigan 48106, to whom I have granted "the exclusive right to reproduce and distribute copies of the dissertation in and from microfilm and the right to reproduce and distribute by abstract in any format." Signature Date Q- C iv VITA The author, Robert Olson Miller, was b o m on August 12, 1955, in Lincoln, Nebraska, to Russell H. Miller and Beatrice Miller. He received his primary and secondary education in Springfield, Nebraska. He attended the University of Nebraska 1973-1981, where he completed his Bachelor of Science in 1978 and Master of Science in 1981 in agronomy. He married Annette J. May on June 20th 1981. From January 1982 until July 1984 he worked as a Research Associate in the Horticulture sciences Department of Texas ASM University. Bob and Annette have one son, Jason D. Miller. He began study and research toward the Fh.D. in the fall quarter of 1984 at Montana State University. v: ACKNOWLEDGMENTS The author wishes to extend his sincere appreciation to his advisor and friend. Dr. Earl 0. Skogley, without whose efforts, guidance, support, and humor this dissertation would not have been possible. I would also like to thank the members of my committee. Dr. Hayden Ferguson, Dr. Ralph Olsen, Dr. Rick Engel, and Dr. Richard Stout, for their advice, guidance and encouragement lending a great contribution to this dissertation. A special thanks to Gordon Williamson at Technical Services for cooperation in the fabrication of equipment and Carol Bittinger, Computer Services, for help with statistics. Thanks also to friends Jae Yang, Thomas Deluca, and Stuart Georgitis for their time and assistance through the course of my doctoral program. I would like to thank the Department of Plant and Soil Science for its truly appreciated financial support. My parents, Mr. Russell Miller and family deserve special thanks for their loving support. My deepest love and appreciation to my caring friend and loving wife, Annette May Miller, the inspiration in my life. A special thanks to my son Jason, whose humor helped complete this thesis. TABLE OF CONTENTS TABTE OF CONTENTS............. TTST OF TABLES.......................................... vi vii TTST OF FIGURES.................................................... xii ABSTRACT.... ...................................................... xiii INTRODUCTION.................................. I Definitions.................................................. 3 LITERATURE REVIEW.................................................. 5 Soil K Ehytoavailability.... ................................. Temperature and Phytoavailable K............................. Soil water Potential and K Availability............ Study of Ehytoavailable K .................................... 5 7 9 10 MATERIAIS AND METHODS....... 12 Methods Development.................... Calculations............................. Physical and Chemical Methods................................ 12 14 15 RESULTS AND DISCUSSION............................................ 17 Methods Developmait Study (MD).......... Fhytoavailability of K on Montana Agricultural Soils (KRTS)... Temperature and Moisture Influences on K Phytoavailability (TMS)........................................................ Effects of K Rate on Soil and Plant K (KRS).................. 17 20 39 52 SUMMARY AND CONCLUSIONS....... 65 LTTERATURE CITED......... 73 APPENDICES........................................................ 80 Appendix A: Appendix B: Location description, physical and chemical characteristics of twenty Montana soils......... 81 Correlation matrix for properties of twenty Montana soils................................. 98 vii U S T OF TABLES Table 1. Page The influence of spring wheat plant density on mean net K uptake from the Edgar soil series after 96 hours............ 18 The influence of spring wheat seedling age on net mean K uptake on the Edgar soil series after 96 hours.............. 18 Mean (n=3) net accumulation of K by spring wheat seedlings as influenced by time of soil contact on the Edgar soil series...................................................... 19 Selected physical properties of twenty Montana soils used in the ERTS study...................................... 21 Selected chemical properties of twenty Montana soils used in the PETS study....................................... 22 6 . Analysis of variance for net K uptake by spring wheat on twenty Montana soils maintained at a water potential of -33 Epa, after 96 hours.......................................... 22 2. 3. 4. 5. 7. Mean net (n=4) plant K uptake by spring wheat seedlings after 96 hours of growth on twenty Montana soils maintained at a water potential of -33 Fpa................... 23 8 . Analysis of variance for soil Eex concentration for five rhizosphere distances and twenty Montana soils............... 24 9. 10. 11. 12. 13. Mean soil Eex concentration at five rhizosphere distances for twenty Montana soils after 96 hours maintained at a soil water potential of - 33 Ipa, mean of four replications....... 25 Regression equations between Eex concentration (Y) and rhizosphere distance (D) on twenty Montana soils............. 26 Estimated Eex diffusion gradient coefficient d [EexJZdD diffusion distance (D^), soil Eex flux (SEex), and Eex effective diffusion coefficient (Dc) on twenty Montana soils........................................................ 29 Mean net plant K flux, soil K flux (SEex), difference (umol) and % Eex utilized by spring wheat seedlings after 96 hours growth on twenty Montana soils...................... 301 * 3 Linear regression model equations between the amount o K absorbed by spring wheat (umol pot-1) and physical and chemical characteristics for twenty Montana Soils. 32 viii IIST OF TARTBS—OCMTOMIED Table 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. Page Malticollinearity detection of regression model 8 with dependent variable plant K uptake and independent variables: clay, d[Kgx]/dD, and CEC..................................... 34 Changes in IID soil solution K concentration as related to rhizosphere distance for spring wheat for the Amsterdam soil after 96 hours.......................... 35 Regression of Kex concentration gradient (d[Kex]/dD), (umol cm-3 cm-1) as dependent variable (Y) and independent variables: Kex; CACO3 ; and saturated paste Ca................ 37 Analysis of variance for net plant K uptake as influenced by temperature (TEMP) and water potential (KPA) on four Montana soils (SOIL)................ 40 Mean neta K uptake (umol pot-1) by spring wheat as influenced by temperature (C) and soil water potential (Kpa) on four Montana soils................................ .. 41 Calculated QlO for K uptake by spring wheat on four Montana soils for temperatures 8 to 24 C, at a water potential of of -12 Kpa........................................ 42 Regression equations of dependent variable K uptake (umol) (Y) and independent variables: water potential (Kpa) and temperature (T) for four Montana soils....................... 44 Analysis of variance table for Kex (umol cm-3) at five distances (D) in the rhizosphere of spring wheat as influenced by temperature (TEMP) for four Montana soils (SOIL)....................................................... 47 Mean (n=3) soil Kex concentration (umol cm-3) in the rhizosphere of spring wheat as a function of distance (mm) and influenced by temperature (C) on four Montana soils maintained at a water potential of - 33 Kpa.................. 48 Regression equations of dependent variable soil Kex (Y) and independent variables: rhizosphere distance (D) and temperature (T) for four Montana soils at - 33 Kpa........... 492 * 4 Volumetric water content at three water potentials for four Montana Soils.......................................... ..... 50 ix EEST of TATOt Fsktmtinijed Table 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. Page Mean (n=3) plant K uptake by spring wheat as influenced by K additions as KCL on four Montana soils at 24 C and -33 Kpa.......................................................... 53 Analysis of variance table for plant K uptake (umol pot-1) by spring wheat as influenced by K addition (KRATE) for four Montana soils (SOIL), 24 C and water potential - 33 Kpa. I.................................... 53 Regression equations for dependent variable plant K uptake (Y) and independent variable potassium addition (KRATE) for four Montana soils, 24 C and water potential - 33 Kpa........ 54 Analysis of variance table for Kex concentration (umol cnf3) in the rhizosphere of spring wheat as influenced by distance (D) and K addition (KRATE) on four Montana soils (SOIL)................................................. 54 Mean (n=3) rhizosphere Kex concentration as influenced by rhizosphere distance (D) and K rate, for four Montana soils, temperature 24 C and water potential -33 Ipa.......... 56 Regression equations between rhizosphere distance (D) and Kex (Y) at five rates of K addition for four Montana soils, temperature 24 C and water potential - 33 Ipa................ 57 Initial soil Kex concentration, diffusion gradient (d[KexJZdD) and diffusion distance for four Montana soils, temperature 24 C andwaterpotential -33 Ipa................... 60 Mean Kex plant, soil arid difference flux values as influenced by five K additions on four Montana soils, 24 C and water potential of - 33 Ipa.............................. 62 Soil extractable Cl and electrical conductivity (EC) on four Montana soils............................... 63 Multiple linear regression model for the influence of temperature (T), water potential (KPA), and estimated phytoavailable K index (ERKE) on spring wheat K uptake....... 68 Soil series, texture and location description of twenty Montana soils............................ 823 * 6 Soil series name and classification and family of twenty Montana soils................................................ 83 X IZ ST OF TAHTFS-OCMTIMJED Table 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. Page Exchangeable cation concentrations for twenty Montana soils as determined by 1.0 N ammonium acetate........... •.... 84 Soil pH, organic matter, calcium carbonate and electrical conductivity of twenty Montana soils......................... 85 Soil particle size analysis as determined by pipette (Gee and Bauder, 1986) on twenty Montana soils.................... 86 Saturated paste pH, cation concentration and potassium activity ratio (AR) of twenty Montana soils.................. 87 Exchangeable and nonexchangeable soil potassium as determined by magnesium chloride and sodium tetraphenyl boron, and nitric acid respectively.......................... 88 Soil extractable phosphorus as determined by Bray and Olson extraction methodologies for twenty Montana soils............ 89 Soil solution cation concentration on a solution basis and potassium activity ratio (AR) of twenty Montana soils as determined by immiscible liquid displacement (Mubarak and Olson, 1976)............................................. 90 Soil solution cation concentration and potassium activity ratio (AR) of twenty Montana soils at 0.33 Epa on a soil mass basis as determined by immiscible liquid displacement (Mubarak and Olsen, 1976)........ ............ ............... 91 Moisture contents of twenty Montana soils at - 0.15 - 0.33,- 0.66 and - 1.00 Kpa water potential....... ......... 92 Moisture contents (gm Kg”1) of twenty Montana soils at - 2.00, - 5.00, and - 15.00 Epa water potential.............. 93 Soil, silt, and clay for twenty Montana soils as determined by hydrometer (Gee and Bauder, 1986).............. 94 Available water at four water potential ranges for twenty Montana soils................................................ 954 * 9 Cation exchange capacity and mineralogy percentages of the clay fraction of twenty Montana soils provided by Bernard Schaff, method of Jackson (1958)......... 96 xi EEST OF TftBEES-CXMFENUED Table 50. 51. Page Soil extractable NO3, Cl, and S for twenty Montana soils........................................................ 975 1 Correlation matrix for twenty soil characteristics on twenty Montana soils........................................ 99 xii LIST OF FIGURES Figure 1. 2. 3. 4. 5. Page Schematic of method for growing wheat seedlings on soil columns separated by a nylon screen..................... 13 Accumulation of K by twenty five spring wheat seedlings on the Edgar soil series at five time intervals. ......... ;. 19 Relationship of plant K uptake [dependent variable - Y] and soil extractable K by NH4OAC (Kex) on twenty Montana soils............................ 24 Changes in the concentration of Kex as a function of distance in the rhizosphere of spring wheat for four Montana soils........................ 28 Potassium uptake QlO for spring wheat seedlings on four Montana soils at a water potential of -12 Kpa................ 41 6. Response function of plant K uptake (umol pot-1) 1 temperature, and soil water potential; soil series Amsterdam....... *.......... ................................ 45 7. Response function of plant K uptake (umol pot-1), temperature, and soil water potential; soil series Beaverton.......... 8 . Response function of plant K uptake (umol pot-1), temperature, and soil water potential; soil series Edgar. 9. 10. 11. 12. 13. 14. 45 ... 46 ... 46 Response surface of mean Kex concentration as influenced by rhizosphere distance and K addition, soil series Amsterdam. . 58 Response surface of mean Kex concentration as influenced by rhizosphere distance and K addition, soil series Beaverton. . 58 Response function of plant K uptake (umol pot-1), temperature, and soil water potential; soil series Kevin. Response surface of mean Kex concentration as influenced by rhizosphere distance and K addition, soil series Edgar....... 59 Response surface of mean Kex concentration as influenced by rhizosphere distance and K rate, soil series Kevin........... 59 Model of phytoavailable K in the rhizosphere of spring wheat for Montana soils...... ..................................... 71 xiii ABSTRACT Soil characteristics and rhizosphere conditions greatly influence plant available potassium (K). Research in Montana has shown crop response to K fertilizers on soils testing high in ammonium acetate extractable K (Kex). The objective of this research was to use an in-vitro technique to determine: (I) plant available K on Montana soils and its relation to soil properties; (2) the influence of temperature and water potential on K uptake; and (3) the influence of K additions on plant available K. Soil characteristics which were found to be strongly related to plant available K (R2 0.93) are: clay content, Kex diffusion gradient coefficient, and cation exchange capacity. Spring wheat in these studies utilized a portion of Kex, but quantities differed across soils. This data suggests that spring wheat utilizes significant quantities of soil K from nonexchangeable and K-mineral sources. Spring wheat K uptake increased exponentially with temperature with little or no K absorbed at 8 C. Water potentials alone had little influence on K uptake. The interaction with temperature suggests that K uptake was not limited by water potential (water content) until temperature exceeded 16 C. Thus until spring wheat K demand reached a critical K demand K uptake was not limited by water potential. Potassium additions increased spring wheat K uptake nonuniformly across soils. Greatest K uptake occurred on the Kevin soil and the least on the Beaverton. Data on soil rhizosphere Kex indicated K additions enhanced the availability of Kex supplies in the near rhizosphere (0 - 0.5 mm). Soil Kex flux data indicated that K additions did not necessarily increase Kex utilization. These results indicate that phytoavailable K is governed primarily by three soil factors and that environmental water potential play important roles as regulators. Spring wheat utilizes soil K forms other than that explained by Kex in the rhizosphere, and that the quantity utilized is highly soil dependent. I INTRODUCTION Soil potassium (K) can be characterized by four major indices: (I) I Soil solution K (KsoI); (2) exchangeable K (Kex); (4) nonexchangeable K (Knex); and (4) mineral K (Km^n). Soils are relatively high in total K but only a small relative proportion is found in solution' form, which is available to plants. Replenishment of solution K at the soil-root interface is dependent on rates of release from other forms and ion transport in the rhizosphere. Research in Montana over the past twenty years has shown crop response to K fertilizers on soils testing "high" in NH4OAC-BXtractable K (Skogley, 1976). Similar results have been noted for many crops around the world (Talibudeen et al. 1978). This is due to the fact that the NH4OAC procedure does not measure factors relating to crop K demand, soil K release equilibria, and K transport in the rhizosphere. A method using intact plants to study soil K release and transport at the soil-root interface using in-vitro techniques would aid the development of a reliable K soil test that is sensitive to these factors which control K availability. Nutrient phytoavailability can be separated into two components, absorption characteristics of plant roots and the capacity of the soil to supply nutrients. Factors influencing soil nutrient capacity are (I) quantity and form of nutrient present, and (2) the chemical equilibria of the soil system affecting the replenishment (intensity) 2 of a given nutrient in the soil solution. Such relationships have been used extensively in describing solution cation activity in inorganic systems. Nutrient absorption and uptake is dependent on the nutrient status in the plant/ morphological characteristics of the root system, root age and influx characteristics, and rhizosphere environment. Nutrient acquisition occurs via three mechanisms, interception, mass flow, and diffusion. Interception results from direct contact between a root and a soil nutrient. Mass flow is the movement of a nutrient to the root via the convective flow of soil water, the result of transpiration. \ Diffusion, is the movement of a nutrient down the chemical potential gradient established by the plant at the soil-root interface. Work of Barber (1962) and Barber (1972) has established that diffusion, and mass flow to a lesser extent, are the major mechanisms by which nutrients are acquired by plant roots. In many instances the constraints of the rhizosphere environment limit soil nutrient transport and plant uptake. both plant biochemistry and soil chemistry. Temperature affects Soil water potential in the rhizosphere affects soil nutrient availability and transport, plant physiology, and soil-root contact. Several methods have been proposed to study the soil-root interface in an attempt to describe nutrient availability. However, physical limitations and the complexity of the processes involved have impeded research. Recently Kucheribuch and Jungk (1982) described a procedure for studying soil K concentration gradients in the rhizosphere of rape. 3 Plant roots are separated from the soil by a flat nylon screen which prohibits root penetration but permits root hairs to make contact with the soil surface. Such a system allows the rhizosphere to be treated as a plane and, with microsampling techniques, be analyzed according to traditional soil testing procedures. Accurate information can thus be obtained depicting plant nutrient availability and absorption in addition to the status of soil rhizosphere K. The objectives of this study were to: (I) determine those soil physical and chemical properties related to phytoavailable K; (2) research the effects of temperature and soil water potential on plant K uptake; and (3) evaluate the influence of K additions on K phytoavailability and K dynamics in the rhizosphere. The in vitro method of studying the soil-root interface provides detailed information on both phytoavailable K and K dynamics within the soil rhizosphere. Definitions A list of abbreviated terms used in describing phytoavailable K are as follows: %ex Soil extractable K, by I N Ammonium acetate, pH 7; kSoI - Soil solution K concentration determined by immiscible liquid displacement (IID); kUGX ~ Soil nonexchangeable K concentration, determined by sodium tetraphenyl boron; Kmin ~ Soil mineral K, associated with micas and feldspars; D - Rhizosphere distance perpendicular to the soil-root interface; 4 d[KexJZdD - Rhizosphere Kex diffusion gradient coefficient; Dc SKex - Effective diffusion coefficient for Kex in rhizosphere; Distance of estimated K depletion in the rhizosphere; Calculated soil flux of Kex removed by plants. 5 U rEERAIURE REVIEW Soil K Fhvtoavailabilitv Soil K can be divided into four major forms: K solution, that portion of K dissolved in the soil solution; K exchangeable, that portion which is readily exchangeable for other cations; K nonexchangeable, which is not readily exchangeable and is released slowly to the solution; and K bearing minerals (Mclean and Watson, 1985). Each of the solid phases tends to be in equilibrium with K in solution. Ehytoavailable K is described as that K fraction in the soil which is accessible to plant roots for absorption, and involves aspects of both soil characteristics and plant nutrient absorption (Bertsch and Thomas 1985). Soil solution K is a reflection of the parent material and the degree of weathering that has occurred. Feldspars and Micas are the dominant K minerals which contribute to and control the dynamics of soil solution K (Sparks and Haung, 1985). Interfacial reactions are the most significant means by which K is weathered from feldspars. Micas, such as muscovite, undergo zonal weathering which results in the release of K from interlayer positions. Thus, the stage of weathering determines the equilibrium each mineral phase has with the soil solution and with each other. The equilibrium between solution phase and solid phase K indices dictates the availability of soil potassium. Beckett (1964) described 6 K availability in terms of a quantity and intensity ratio (Q/I). He plotted the amounts of K adsorbed or desorbed against the activity ratio for K in solution, and was thus able to predict the capacity of a soil to maintain solution K. Such a method could be used to make K fertilizer recommendations. However, it was shown by Rasnake (1973) that Q/I curves change with fertilization, and thus have limited applications. Both physical processes, such as freezing and thawing and wetting and drying, and biological activity affect the release or fixation of K from the soil solution. Cook and Hutcheson (1960), reported that the drying of soil samples increased exchangeable K concentrations. Song and Haung (1983), in a review of the physical chemistry of soil K, reported several instances where rhizosphere organic acids increased K release from micas. Thus, soil solution K is affected by physical processes and amount of biological activity as well as chemical processes. Ultimately the phytoavailability of potassium is not only dependent on the amount of K in the soil and the capacity of the soil to replenish solution K. It is also dependent on the absorption characteristics of plant roots, which are determined by overall nutrient requirements, type of root system, physiology, and growing conditions (Olsen, 1968 and Barber 1985b). Nutrient acquisition occurs via three mechanisms, interception, mass flow, and diffusion. Interception results from direct contact between a root and a soil nutrient. Mass flow is the movement of a nutrient to the root via the convective flow of soil water, the result of transpiration. Diffusion, 7 is the movement of a nutrient down the chemical potential gradient established by the plant at the soil-root interface. Work of Barber (1962), Nye et al. (1975), and Nye and Tinker (1977) has established that diffusion and, to a lesser degree, mass flow are the major mechanisms by which nutrients are transported to plant roots. Temperature and Fhvtoavailable K Temperature effects on nutrient availability can be divided into two components: (I) Plant aspects and (2) soil aspects. Plant aspects can be further separated into direct and indirect effects. Temperature indirectly influences ion absorption by affecting growth rate, translocation, and transpiration. It directly influences the kinetics of physical and biological processes. Nobel (1974) estimated a QlO of I to 2 for physical reactions and 2 to 4 for bioltigical reactions. With respect to water absorption, a physical process, Markart et al. (1979) found the QlO for water absorption by soybeans increased from 1.3 to 13 between temperatures of 14 and 9 C. phase change in plasmaletnma membrane lipids. This was attributed to a Ion absorption was found by Carey and Berry (1978) using solution culture to be strongly inhibited at .10 C, resulting in a change in QlO from 15 to 8 below this temperature. Similar results were found for P by Barber (1985a), where P uptake by fescue cease! when temperature was lowered to 10 C . The influence of temperature on cereal root system growth and morphology has been reported by Abbas Al-Ani-ani and Hay (1983). Root systems of wheat increased in length and in number of seminal axes as temperature rose from 5 to 25 C . Nye and Tinker (1977), have found 8 that the mean angle of maize roots to the horizon was smallest at a temperature of 17 C. temperature. The angle increased with increases in Thus temperature determines the size and morphology, of the root system, in addition to its three dimensional orientation in the soil. Physical effects of temperature on soil aspects can easily be seen in terms of effects on ion diffusion and soil solution viscosity. As temperature decreases ion diffusion decreases according to the StokesEinstein equation. Viscosity also increases with decreasing temperature and the combined effect results in a 22 % decrease in diffusion of K for a temperature change from 25 to 15 C (Barber 1985a). Temperature also influences the vapor pressure and surface tension of water, therefore soil water movement is modified as temperature changes. Temperature affects soil chemical reactions by influencing cation ion solution activity, cation bonding energies and extent of the double layer (Barber, 1985a). These reactions ultimately determine solubility, solution speciation and the stability of the soil system to chemical change (Yang, 1987). Sparks and Leibhardt (1982) found K selectivity coefficients decreased with temperature, indicating decreased K sorption with higher temperatures. In many instances temperature changes affect physical and chemical processes as a unit. A decrease in temperature results in decreased solubility, but precipitation is inhibited by decreased diffusion. Thus the system can be considered a physiochemical process which is very sensitive to 9 temperature change and which influences nutrient availability (Sutton, 1969). Soil Water Potential and K Availability Water potential is defined as the chemical potential of water with respect to pure water at the same temperature, expressed in terms of energy. Equivalent water potential in different soils implies equal free energy levels, but different volumetric and /or gravimetric water contents (Nye and Tinker, 1977). As water potential decreases (more negative), resistance to water withdrawal increases. As with temperature, the influence of water potential (ie. water content) can be divided into plant and soil aspects. Danielson (1967) reviewed the effects of soil water content and found that root to shoot ratios increased with decreases in water content. This was attributed to an increased root system and decreased shoot growth associated with drier soils. Hsieh et al. (1972) found similar results for maize. Mackay and Barber (1987), evaluated the effects of water potential on root hair growth on c o m and found that increasing water potential, from - 175 to - 7.5 Kpa, resulted in cessation of root hair growth behind the root cap. Reversing water potential (decreasing it) resulted in new root hair formation. In another experiment, Mackay and Barber (1983) found the total root length with root hairs and root hair density increased as soil moisture content decreased from 32 to 22 percent. With soil drying, root hair growth increased, with respect to length and density, to maintain liquid continuity with the soil. Water potential (water content) has great significance with 10 respect to physical and chemical processes. Decreases in water potential result in the physical draining of soil pores of ever decreasing diameter, and a decrease in water film thickness over the surface of particles. This reduction in the continuity of water films decreases the mobility of ions subjected to chemical gradients (Nye and Tinker, 1977). Thus nutrient diffusion is impeded by increased tortousity as soil becomes drier (Wamcke and Barber, 1972). Bertsch and Thomas (1985), reviewed the influence of soil water content on soil chemical processes. As water content decreased, ion speciation in the solution changed because the activity ratio (AR) must be maintained, as demonstrated by results of Mengel and von Braunschweig (1972). With increased water potential solution ion concentration decreases, the result of dilution accompanied by a net adsorption divalent cations and desorption of monovalent cations (Moss, 1963). With decreased water potential the opposite reaction occurs, resulting in a decrease in solution monovalent ions such as K. Thus K availability decreases with reduced water potential due to slowed diffusion, a physical process and the AR mechanism, a chemical process. Study of Phvtoavailable K Soil tests of plant available K traditionally have relied on correlation between a K extractable index based on cation displacement and K response to K additions (Feigeribaum and Hagin, 1967). This 'Index1 value of soil then implies a level of phytoavailable soil K. Reitemeier et al. (1947) studied K availability based on greenhouse, Neubauer, and laboratory methods using fourteen soils. 11 Their results indicated a relatively high correlation between K uptake by clover and electrodialysis of K, as well as Neubauer extractable K. The Neubauer technique was developed by Thomtoii (1935) to study nutrient accumulation and involves growing plant seedlings in soil for one to to three weeks followed by nutrient uptake assessment. The technique has been modified numerous times for specific experiment purposes (Masses, 1973.). Research in methods of evaluating soil changes in nutrient concentration were first described by Brown et al. (1964), using ion exchange resins. This method involved mating a resin against a soil column and thin slicing the soil at micrometer distances from the resin at the end of the experiment. Several papers have since used this methodology to study K, P and N availability (Vaidyanathan and Nye, 1966). Kuchenbuch and Jungk (1982) described a method combining the Neubauer technique with that of Brown et al. (1964). seedlings are grown on nylon screen. In it, plant Root hairs grow through the screen and make contact with the soil but roots do not. Potassium depletion gradients are then studied to determine K diffusion in the rhizosphere. This technique has many advantages for the purpose of studying spring wheat phytoavailable K, and was adapted for use in the studies reported here. 12 MATERIALS AND METHODS To evaluate the influence of soil properties and environmental factors on K phytoavailability four experiments were initiated. These were: Methods Development (MD); Ehytoavailability of K on Montana agricultural soils (KEdS); Terrperature and moisture influences on plant and soil K (IMS); and Effects of K rate on soil and plant K (KERS). In the MD study, techniques were developed using spring wheat seedlings to evaluate plant available soil K. In the PKTS study plant available soil K dynamics were assessed on twenty Montana soils to identify soil characteristics important to K availability. The influence of temperature and water potential on plant K availability were investigated in the IMS study, and in the EERS- study the affect of K additions on plant K uptake and Kex concentration gradients were evaluated. Methods Develonment A modification of the method of Kudheribuch and Jungk (1982), was employed to determine plant K uptake. Nylon cloth, 325 mesh, 38 percent porosity, and 40 urn thickness (Nitex Co. Zurich, Switzerland) was cemented to one end of a plexiglass cylinder, 4.5 cm ID and 1.5 cm in height. The cylinders were washed with a 5 percent solution of sodium hypochlorite to reduce the possibility of contaminating molds CIRhizopus Sp.) The nylon covered end of the cylinder was then placed 13 on a agar media consisting of 0.065 % agar purified (nutrient free) and 2 uM CaSO4 . Various numbers of spring wheat seeds (Triticum aestivum L. cv. Pondera,) were placed in the cylinder, on the screen and allowed to germinate/grow for 5 days at 22 C under etiolated conditions. An identical cylinder, but with no screen was filled with soil wetted to a water potential of approximately - 33 Kpa to a bulk density of 1.25 gm cm-3. This cylinder containing soil was then placed on a high flow moisture plate maintained at a potential of - 33 Kpa. The cylinder containing the germinated seeds (referred to as a pot) was then placed on the soil column so that the nylon screen separated the seed from the soil (Figure I). Figure Root hairs but not roots penetrate the I. Schematic of method for growing wheat seedlings on soil columns separated by a nylon screen. 14 nYlon screen into the upper layer of the soil. The system including the high flow plate was covered with a plexiglass chamber. Relative humidity in the chamber was regulated to near 100 percent to minimize mass flow of water from the soil to the plants. The photoperiod was 14 hr and light intensity was 12 Wm-2. v The plants were left on the soil column from one to eight days. At the end of each uptake period soil columns were thin sectioned, using a modified micrometer and thin slicing knife, at five planar distances representing mean distances of 0.25, 0.75, 1.5, 2.5, and 4.0 mm away from the nylon screen. Moisture content and exchangeable cations (as extracted by 1.0 N NH4OAC) were determined on each soil thin section sample. Whole spring wheat plants were removed from the upper cylinder/pot and placed in digestion tubes for total K analysis by the method of Havlin and Soltanpour (1980). Control pots containing spring wheat plants were grown parallel without soil and used to calculate net K uptake. Calculations The quantity of a substance diffusing from a source across a plane to a sink of constant concentration is given by Crank (1956) and Vaidyanathan and Nye (1966): M t = 2 (C1-C2) ((Dc)Vrr)0 *5 [1] where M t = amount diffusing across the soil surface from zero to time t. C1 = initial concentration of ions in the soil. C2 = constant concentration at the soil-sink interface. 15 (Dc) = the effective diffusion coefficient for the system within the limits of C1 and C2 . If the ion sink maintains C2 at or near zero concentration, then the effective diffusion coefficient can by calculated : Dc = (Mt)2Y /4 (C1)2 t [2] Physical and Chemical Methods Physical and chemical analyses were performed as follows: (1) particle size analysis by hydrometer method (Gee and Bauder, 1986) ; (2) particle size distribution by pipette (Gee and Bauder, 1986); (3) moisture retention curve by pressure plate (Peters, 1965); (4) percent CaCO3 by gravimetric loss (U.S. Salinity Lab, 1954); (5) percent organic matter (Sims and Haby, 1970); (6) soil pH and conductivity by 2:1 dilution (U.S. Salinity Lab, 1954) ; (7) extractable K, Na, Ca, and Mg, by 1.0 N NH4OAc extraction and atomic absorption analysis (Bower et al., 1952); (8) extractable K by 0.5 N MgCl and 0.5 N NH4OAc (Rich, 1964); (9) soil extractable K by sodium tetraphenyl boron (Scott and Reed, 1960); (10) soil extractable K by 1.0 N HNO3 (Mclean and Simon, 1958) ; (11) saturated paste pH, K, Ca, and Mg extraction and analysis by atomic absorption (U.S. Salinity Lab, 1954); (12) soil solution K, Ca, Mg, NO3 , and Cl by extraction by immiscible liquid displacement (Mubarak and Olson, 1979) and determination of K , Ca, and Mg by atomic absorption, NO3 by Cd reduction (Willis, 1980) and Cl by ferric cyanide reduction; 16 (13) soil sulfate by acetate soluble SO4 (Bardsley and Lancaster, i960) (14) available phosphorus by NaHOO3 (Olsen et al., 1954) and modified Bray #1 (Smith et al., 1957); (15) extraction of NO3-N in soils (Sims and Jackson, 1971) and analysis by Cd reduction (Willis, 1980); (16) plant tissue K, Ca, and Mg by 15:1 nitric-perchloric digest (Havlin and Soltanpour, 1980), analysis performed by atomic absorption; (17) clay mineralology, illite, vermiculite, and montmorillinite (Jackson, 1958). 17 RESULTS AND DISCUSSION Methods Development Study Fhytoavailable K can be best described by a plant model which simulates K uptake and transport in the rhizosphere. Techniques for studying the soil-root interface have, in the past, proved quite difficult. Kudheribuch and Jungk (1982) have described a method for studying fhizospheric soil. Using modifications of this technique a simple, accurate, and reproducible method is described for studying both soil Kex concentration gradients and plant K uptake. Spring wheat was chosen for this study, due to its seedling growth characteristics and importance as a major crop on the Northern Great Plains. To study phytoavailable K, plant seedlings must exhibit high demand for K, and growth conditions must be optimal with respect to root/soil contact. The objectives of the method development study (MD) were: (I) determine the plant density required to cover the soil-root interface (ie. nylon cloth) with roots; (2) determine seedling age (hours after germination) for maximum K uptake; and (3) assess the rate of plant K accumulation by spring wheat from soil columns to determine the time frame for subsequent studies. Results of studies to determine the influence of plant density on I net K accumulation are presented in Table I. increased with increasing plant density. Accumulation of K Although high plant densities favored greater uptake, low populations are desirable to minimize plant root competition. Twenty-five seeds provide sufficient root mass to 18 cover the nylon screen. Seedlings germinated for 120 hours had a mean root length of 30 cm and produced sufficient root mass to cover greater than 75 percent of the screen area. It was observed that root hairs protruded through the nylon screen less than 150 um, and thus soil penetration was limited. The effect of seedling age on net plant K accumulation is presented in Table 2. Young seedlings exhibited lower K uptake than Table I. The influence of spring wheat plant density on mean net K uptake from the Edgar soil series after 96 days. Plant Population (plants pot-1) 25 35 45 15 ------------- K uptake umol pot-1 ------------22.Ia 35.6 46.4 52.3 a Means of 2 replications. older ones, with maximum accumulation occurring at approximately 150 hours. This suggests younger seedlings (< 48 hours) have near sufficient K reserves to meet demand while older seedlings (>192 hours) have decreased net accumulation, probably the result of depleted endosperm reserves and slowed metabolism. Table 2. The influence of spring wheat seedling age on net K uptake on the Edgar soil series after 96 hours. 48 Seedling Age, Hours After Germination 96 150 192 -------------K uptake umol pot- 1 -------------10.5 28.5 34.5 18.2 19 The net accumulation of K by spring wheat as influenced by time of soil contact is presented in Table 3. Accumulation from zero to 96 hours was linear and averaged 0.44 umol K hour-1 pot-1 (Figure 2). The rate of K accumulation increased at a decreasing rate for periods longer Table 3. Mean (n=3) net accumulation of K by spring wheat seedlings as influenced by time of soil contact on the Edgar soils series. Oa 24 Time 48 (hours) 96 150 192 ---------------- K uptake umol pot-1 ----------------0.0 12.1 21.0 42.1 52.9 55.7 a Initial K content of 25 wheat seedlings germinated for 150 hours, 42.1 umoles. Time (days) Figure 2. Accumulation of K by 25 spring wheat seedlings on the Edgar soil series at five time intervals. 20 than 96 hours. Thus K accumulation by spring wheat from 0 to 96 hours has reached a steady state condition. These results indicate 25 spring wheat seedlings which have been germinated for 120 hours provide sufficient K demand for use as a biological sink for soil K during growth periods of 96 hours. Soil thin sectioning of the rhizosphere, data not shown, indicated Kex concentration profiles extending from the soil-root interface to a distance of 0.5 mm after 96 hours. Thus spring wheat plants provide a good model for studying both phytoavailable K and soil movement of Ke x . Fhvtoavailable of K on Montana Agricultural Soils (FKTS) Twenty soils were collected from majors agricultural areas from throughout Montana where previous field fertility research had been conducted. Soils were air dried and physically and chemically characterized according to established methods (see page 15). In the FKTS study, soil columns 4.5 x 1.5 cm were prepared with each of the twenty soils. Columns were replicated four times and placed on a one bar moisture plate with water potential maintained at - 33 Kpa. Twenty-five spring wheat seedlings germinated for 120 hours, prepared as previously described, were placed on soil columns for 96 hours. Plant K uptake and soil rhizosphere Kex concentration profiles were determined as described previously. Selected physical and chemical properties of those soils used in the FKTS study are presented in Table 4 and Table 5 respectively. Soil organic matter (CM) ranged from ranged from H O to 477 gm Kg-1 of soil. 1.27 to 5.41 % and clay content Soil pH ranged from 5.62 to 8.34 and Kex concentration (mass basis) ranged from 0.619 to 2.115 cmol 21 kg-1 on the twenty soils. In general these soils represent the broad spectrum of characteristics upon which spring wheat is produced in Montana and the Northern Great Plains. Analysis of variance results. Table 6, indicated significant (p>0.0001) differences between soils in plant K uptake. Table Mean plant K 4. Selected physical properties of twenty Montana soils used in the EKTS study. Soil Series CE % Sand - Silt g Kg 1 Clay - - 33 Kpaa % Amsterdam Bear Paw Beaverton Bozeman Chanta 2.37 3.22 5.37 3.19 1.78 182 246 196 199 342 536 303 463 503 473 281 451 341 297 184 24.61 26.78 27.07 26.11 18.26 Cherry Chinook Creston Edgar Evanston 3.13 1.97 5.41 1.40 2.05 232 572 446 409 352 453 256 420 350 356 314 170 134 240 290 24.78 14.97 21.06 16.17 20.83 Flathead Fort Collins Gerber Kevin Manhatten 4.80 1.27 2.62 3.13 2.40 576 506 206 269 202 313 333 316 386 576 HO 160 477 344 220 22.79 14.65 27.38 22.37 23.34 Rothiemay Round Butte Stryker Tanna Vamey 4.26 2.39 3.58 1.92 1.68 342 139 149 439 596 353 640 570 340 203 304 220 280 220 200 23.21 31.40 33.36 18.11 15.18 2.89 0.11 2.95 0.12 330 19 5.7 31 407 21 5.2 35 262 17 6.4 28 22.63 0.58 2.58 0.97 N=4 Mean STD DEV CV % ISD (0.05) a Soil water content by pressure plate, -33 Kpa and 1.3 gm cm 3 . 22 Table 5. Selected chemical properties of twenty Montana soils used in the EKTS study. Soil Series PH (Isl) CEC cmol kg-1 CACO3 % cmol Kg ■*- 5.47 0.72 0.49 0.66 0.17 1.43 1.92 0.823 1.74 0.955 0.57 0.96 0.24 1.16 0.81 1.36 1.40 0.619 0.887 1.32 1.08 1.85 1.31 0.73 0.51 1W , Xsol^ uM Amsterdam Bear Paw Beaverton Bozeman Chanta 8.24 7.93 5.62 7.34 6.62 22.5 22.6 25.1 23.9 9.3 Cherry Chinook Creston Edgar Evanston 7.35 8.34 5.69 6.81 6.35 28.8 10.4 13.5 10.9 13.9 Flathead Fort Collins Gerber Kevin Manhatten 7.18 6.13 6.78 6.92 8.15 14.1 7.6 30.7 13.7 20.1 0.25 0.26 0.72 0.42 4.04 0.877 0.988 2.05 2.11 1.89 1.31 1.17 0.95 1.88 0.84 , Rothiemay Round Butte Stryker Tanna Vamey 7.54 6.05 7.88 8.24 8.13 20.9 10.6 14.4 13.9 16.8 1.60 0.28 2.06 3.70 5.03 1.48 0.778 0.822 0.942 1.11 1.17 0.57 0.86 2.28 0.61 — 1.92 0.17 9.21 0.25 1.28 0.014 1.14 0.021 1.22 0.031 2.54 0.044 N=4 Mean STD DEV C.V. % LSD (0.05) 7.17 0.087 1.28 0.13 0.44 11.33 0.13 0.14 ' 0.34 a KsoI, determined by IID on soil, water content - 33 Kpa. Table 6 . Analysis of variance for net K uptake by spring wheat on twenty Montana soils maintained at a water potential of -33 Kpa, after 96 hours. Source DF Sum of Squares Mean Square Soil Error Corrected Total 19 60 79 9564.31 525.98 10090.30 503.38 8.77 F Value Pr > F 57.42 0.0001 23 Table 7. Mean net (n=4) plant K uptake by spring wheat seedlings after 96 hours of growth on twenty Montana soils maintained at a water potential of -33 Kpa. Soil Series K umol Soil Series K umol Amsterdam Bear Paw Beaverton Bozeman Chanta 32.25 56.34 27.12 42.16 31.95 Flathead Fort Collins Gerber Kevin Marihatten 20.87 35.94 52.52 52.21 38.64 Cherry Chinook Creston Edgar Evanston 32.84 17.28 22.62 40.41 44.46 Rothiemay Round Butte Stryker Tarma Vamey 49.40 35.32 44.11 34.65 22.24 Mean Std Dev CV % ISD (0.05) 36.67 2.96 8.07 4.19 uptake for each soil is presented in Table 7. Potassium uptake ranged from 17.24 to 56.34 umoles pot-1 after four days for spring wheat. Means separation (ISD) indicated that the twenty soils could be readily separated, and according to their ability to provide. phytoavailable K. A regression of the standard K soil test (I N NH4OAc) and plant K uptake indicated a low correlation (Figure 3). These results are in accordance with Skogley and Haby (1981), who found regression coefficient in the field for K response of R2 = 0.32 . Results of ANOVA on rhizosphere soil Fex concentrations at five planar distances indicated high significance in SOIL, DIST and the interaction term (Table 8). Thus spring wheat seedlings, grown on these soils induced significant differences in Kex between rhizosphere distances and between soils. 24 Soil mean rhizosphere concentrations (volume basis), Table 9, indicated that Kex concentration varied greatly between soils at any given distance. Cn all soils there was a significant reduction in Kex concentration in the near rhizosphere (0.25 mm) as compared to both greater distances and initial soil Kex concentration. K6X Figure Table Soils high ( c m o l k g ' 1) 3. Relationship of plant K uptake [dependent variable - Y) and soil extractable K by NH4CAc (Kex) on twenty Montana soils. 8 . Analysis of variance for soil Kex concentration for five rhizosphere distances and twenty Montana soils Source DF Sum of Squares Mean Square SOIL DIST S0IL*DIST Error Corrected Total 19 4 76 330 439 8226.62 865.60 114.10 59.40 9265.74 432.98 216.40 1.51 0.180 F Value Pr > F 2405.41 1202.12 8.33 0.0001 0.0001 0.0001 25 Table 9. Mean soil Kex concentration at five rhizosphere distances for twenty Montana soils after 96 hours maintained at a soil water potential of -33 Kpa, mean of four replications. Rhizosphere Distance mm Amsterdam Kex (umol cm 3) Bear Raw Beaverton Bozeman Chanta 0.25 0.75 1.50 2.50 4.00 10.41 11.70 12.55 13.75 14.36 13.60 15.46 16.69 18.01 19.77 6.68 7.53 8.01 8.38 8.77 13.63 14.63 16.01 17.19 18.31 8.68 9.23 9.60 10.19 11.10 std Initial3 0.14 15.11 0.12 22.63 0.09 8.68 0.21 19.83 0.33 11.35 Chinook Creston Edgar Cherry Evanston 0.25 0.75 1.50 2.50 4.00 8.82 11.30 12.44 13.84 14.92 11.30 12.50 12.66 12.98 13.12 3.83 4.32 5.08 5.89 6.48 5.32 6.31 7.16 8.10 8.99 9.34 11.11 11.70 12.74 13.55 std Initial 0.38 16.68 0.14 14.57 0.12 8.10 0.12 9.95 0.19 16.37 Gerber Kevin Flathead 0.25 0.75 1.50 2.50 4.00 std Initial Ft. Collins Marihatten 5.08 6.65 8.24 8.89 9.86 6.92 7.79 8.42 9.19 9.76 17.26 18.64 20.27 21.24 21.98 18.14 20.08 20.84 22.23 23.42 12,84 15.10 16.34 17.79 18.49 0.41 10.70 0.35 10.80 0.21 24.05 0.15 25.65 0.17 21.82 Rothiemay Round Butte Stryker Tanna Vamey 0.25 0.75 1.50 2.50 4.00 9.61 12.06 13.63 14.73 16.08 5.02 5.80 6.46 7.20 7.81 4.85 5.99 6.84 ' 8.04 8.91 7.57 8.84 9.25 10.14 10.87 8.95 10.34 11.22 12.20 12.68 std Initial 0.31 18.10 0.14 9.44 0.24 10.42 0.25 11.70 0.58 13.22 a Initial concentration of Kex at time=0 26 Table 10. Regression equations between Kex concentration (Y) and rhizosphere distance (D) on twenty Montana soils. Soil Series R 2 ** Equation 8.4 + 0.137(D0 *5) Amsterdam Y = Bear Paw Y = 11.4 + 0.145(D0-5) Beaverton Y + F P>F - 6 .68E—4 (D) 0.99 617 .0001 - 2.38E-4 (D) 0.99 1049 .0001 0.088(D0 *5 ) - 5.77E-4 (D) 0.98 345 .0001 Bozeman Y = 11.8 + 0.113(D0,5) - 1.61E-4 (D) 0.99 436 .0001 Chanta Y = 8.3 + 0.017(D0 *5 ) - 4.07E—4 (D) 0.87 54 .0001 Cherry Y = 5.5 + 0.240(D0 *5) - 1.46E-3 (D) 0.96 211 .0001 Chinook Y = 10.8 + 0.073(D0 -5 ) - 5.77E—4 (D) 0.92 96 .0001 Creston Y = 2.8 + 0.058(D0 *5 ) + 5.10E-6 (D) 0.98 789' .0001 Edgar Y = 3.9 + 0.091(D0 *5) - 1.75E-4(D) 0.99 1108 .0001 Evanston Y = 7.2 + 0.153(D0 *5) - 8.65E-4(D) 0.97 283 .0001 Flathead Y = 2.3 + 0.196(DO'S) - 1.23E—3 (D) 0.91 81 .0001 Fort Collins Y = 5.6 + 0.087(D0 '5) - 3.34E—4 (D) 0.94 165 .0001 Gerber Y = 14.4 + 0.193(D0 '5) - 1.15E-3(D) 0.96 267 .0001 Kevin Y = 16.0 + 0.151(D0 '5) - 5.56E-4(D) 0.98 370 .0001 Manhatten Y = 9.3 + 0.251(DO'S) - 1.68E-3(D) 0.99 929 .0001 Rothiemay Y = 6.1 + 0.253(D0 -S) - 1.53E-3(D) 0.97 . 256 .0001 Round Butte Y = 3.8 + 0.077(D0 -S) - 2.35E-4(D) 0.97 316 .0001 Stryker Y = 3.2 + 0.105(D0-S) - 2.47E-4(D) 0.98 256 .0001 Tanna Y = 6.2 + 0.100(D0 -5) - 4.14E-4 (D) 0.94 152 .0001 Vamey Y = 6.4 + 0.189(D0 -5) - 1.72E-3 (D) 0.92 98 .0001 = 5.5 a. Units for Kex (Y) are umol cm-3; Distance D, mm(10~3). ** Regression values based on mean of four observation at each distance. 27 in Fexz in general, exhibited the greatest change in Kex between distances of 0.25 and 4.00 ram. Multiple regression analysis (SAS Institute, 1987) of rhizosphere Kex concentration (Y), with respect to distance (D), indicated that Kex was best described as a two variable equation using distance and square root of distance, equation [3] (Table 10). Y = C1 + C2 (D) + C3 (D0-5) The equation was: [3] On all twenty soils highly significant regressions were found (P>0.0001). Regression.coefficients for the square root term (D®*^ are approximately 1000 times greater than the term for distance alone. This indicates the diffusive movement of Kex closely follows Pick's first and second laws of diffusion (Nye and Tinker, 1977). The distance term (D) reflects correction term for Kex concentration for Kex at greater rhizosphere distances. Using the regression equations for Kex, an overall Fex diffusion gradient coefficient (U(Kex)ZdD) can be determined from the second derivative of each equation in Table 10. Thus d[Kex]/dD represents the change in Kex concentration at any rhizosphere distance from the soildistance (equation 3), the extent of depletion in rhizosphere Kex was calculated. Table 11. Thus, the maximum distance of Kex diffusion could be compared between soils. This distance (Dl) ranged from 3.3 ram to 10.1 ram and averaged 5.2 ram on the twenty soils. Generally soils which exhibited high plant K uptake had the greatest rhizosphere diffusion distances. Kudhenbuch and Jungk (1982), using rape, found that the distance of K depletion increased with fertilizer additions. 28 V Amsterdam O Tanna X Beaverton X Creston Rhizosphere Distance (mm) Figure 4. Changes in the concentration of Kex as a function of distance in the rhizosphere of spring wheat for four Montana soils. Symbols represent observed data points and lines represent estimate d[Kex]/dD. Intergrating equation [3] for each soil front the soil-root interface to the estimated distance limit of diffusion (D^), and subtracting the initial concentration of Kex an estimate of the amount of soil Kex flux (SKex) was calculated (based on a soil column cross sectional area of 15.5 cm and 96 hours). Soil Kex flux ranged from 3.6 to 82.6 and averaged 45.2 nmol m-2 s-1 pot-1 over all soils. The SKex estimate represents the amount of Kex which was removed from each soil by spring wheat seedlings in 96 hours. A poor relationship was found between SKex and plant K uptake (R2 = 0.28). The final term Dc , Table 11, represents the estimated effective diffusion coefficient for Kex calculated by equation [2] (Vaidyanathan and Nye, 1968). Diffusion terms reported in the literature are for K 29 alone, while these values represent Kex, based on the estimated values d [KexJZdD and SKex. Among soils, Dc varied by two orders of magnitude. Dc was the lowest on the Beaverton soil and greatest on the Creston soil series. Thus, on the Creston soil Kex diffusion was 146 times faster Table 11. Estimated K^x diffusion gradient coefficient d[Kex]/dD diffusion distance (D1), soil Kex flux (SKex), and Kex effective diffusion coefficient (Dc) on twenty Montana soils. Soil Series CltKexJZdD umol cm 2 cm I Dl mm Flux SKex nmol m 2 s Dc cm2 s Amsterdam Bear Paw Beaverton Bozeman Chanta O .0864a 0.1390 0.0411 0.1112 0.0600 6.4b 8.1 3.3 6.3 4.5 28.9C 82.6 3.6 46.7 17.0 2.27e-6d 8.24e—6 1.19e-7 3.43e—6 1.41e-6 Cherry Chinook Creston Edgar Evanston 0.1227 0.0240 0.0644 0.0847 0.0854 6.8 3.8 8.3 6.0 7.9 59.0 33.7 42.9 32.4 80.3 7.74e-6 2.52e—6 1.74e-5 6.62e—6 1.49e—5 Flathead Fort Collins Gerber Kevin Marihatten 0.0975 0.0907 0.1005 0.1135 0.1128 6.4 5.1 7.0 10.0 5.5 35.3 33.5 70.9 70.0 68.4 6.71e—6 5.OOe-6 5.lie—6 4.48e-6 5.40e—6 Rothiemay Round Butte Stryker Tanna Vamey 0.1301 0.0637 0.0936 0.0703 0.0398 6.8 10.1 7.1 3.6 6.4 66.7 65.3 44.4 13.6 28.0 8.46e—6 1.74e-6 1.21e-5 9.89e-7 2.59e-6 0.0866 6.5 45.2 5.86e-6 Mean a. Estimate of d [KexJZdD based second derivative of regression equation of rhizosphere distance and Kex. b. Estimates for Dl based on Kex depletion gradients and initial Kex concentrations. d. Estimates of SKex flux based on integration Kex depletion gradients, a cross sectional area of 15.5 cm2 and 96 hours. c. Dc estimate determined from equation.I (Vaidyanathan and Nye, 1966). 30 than on the Beaverton soil. These diffusion coefficient values are considerably larger than those reported by Rowell et al. (1967) and Barber (1985a). This is due to (I) the fact this system used intact plants rather than resins which could result in differing effective diffusion gradients and (2) these results are for Dc of Kex and not K ions alone. Comparisons of net plant K uptake and calculated SKex are presented in Table 12. Differences between plant and soil K flux Table 12. Mean net plant K flux, soil K flux (SKex), difference (umol) and % Kex utilized by spring wheat seedlings after 96 hours growth on twenty Montana soils. Soil Series Plant K Flux SKex Flux Delta K Flux Kex Utilized % ------ nmol m 2 s 1 pot -I Amsterdam Bear Paw Beaverton Bozeman Chanta 61.7 107.8 51.9 80.7 61.1 28.9 82.6 3.6 46.7 17.0 32.6 25.3 48.3 33.9 44.1 46.9 76.6 7.0 58.0 27.9 Cherry Chinook Creston Edgar Evanston 62.8 32.8 43.1 77.4 85.1 59.0 33.7 42.9 32.4 80.3 3.8 0.6 0.4 44.8 4.6 93.9 102.3 99.1 41.8 90.1 40.0 68.8 100.6 100.0 74.0 35.3 33.5 70.9 70.0 68.4 4.6 35.2 29,5 30.8 5.6 88.5 48.7 70.5 69.0 92.5 94.7 67.6 84.5 66.3 42.5 66.7 65.3 44.4 13.6 28.0 27.8 1.3 40.0 52.7 15.5 70.4 96.3 52.6 20.5 63.5 70.1 45.2 23.9 65.6 Flathead Fort Collins Gerber Kevin Manhattan Rothiemay Round Butte Stryker Tanna Varney Mean 31 are represented by Delta K flux; that portion of K taken up by spring wheat that is not accounted for by changes in the Kex. This is better shown by the term Kex utilized (%) which represents SKex flux divided by Plant K flux. Only on six of the twenty soils did SKex account for more than 70 % of the total K taken up by spring wheat seedlings. Qn six of the soils, SKex flux accounted for less than 50 % of the plant K uptake. For the Beaverton soil only 7 % of the K absorbed by the plant was due to SKex flux. These differences between plant K SKex flux clearly demonstrate that Kex does not adequately predict phytoavailable K on these soils. Stepwise linear regression (SAS Institute, 1987) was employed to select an optimum subset of soil physical and chemical characteristics which best explain variation in K uptake by spring wheat. Soil physical characteristics which were used in model selection were: % sand, %silt, % clay, organic matter (CM), % CAG03, % coarse silt, % medium silt, % fine silt, % clay by pipette, available water capacity, . % illitic clay, % vermiculitic clay, and % montmorillinitic clay. Soil chemical characteristics were: pH, EC, NO3 , Cl, Bray-P, Olsen-P, S, K extractable, Ca extractable, Mg extractable, Na extractable, potassium absorption ratio, IID solution K Ca, Mg, K activity ratio, nonexchangeable K (sodium tetraphenyl boron and nitric acid), extractable K (MgCl), paste pH, paste K, Ca and Mg concentration, paste K activity ratio and ratio of KexZto solution K. Additional calculated variables included were: ^(Kex)ZdD, distance of diffusion (D), and effective diffusion coefficient for Kex (Dc). 32 Variable selection was carried out using forward and maximum R selection methods (Freund and Littell, 1986). Models were selected on the basis of significance, coefficient of determination and of regression coefficients. Models explaining plant K uptake using one to five independent variables are listel in Table 13. Models 1 - 5 represent those soil properties which individually indicate high Table 13. Linear regression model equations between the amount of K absorbed by spring wheat (umol pot-1) and physical and chemical characteristics for twenty Montana soils. Equations3 Model R2 _ 16.4 + 2.62 (ILL) 0.72 0.59 0.55 0.42 0.38 F P>F .0001 .0001 .0001 .002 .003 I 2 3 4 5 Y Y Y Y Y 6 7 Y 21.2 — 0.58(CEC) + 3.28(ILL) Y = 11.5 + IlSfd(KexVdX) + 1.95 (ILL) 0.79 0.78 29 .0001 .0001 8 Y = 8.53 4* 1.08(CIAY) - 0.98(CEC) + 194 Cd(KexVdX) 0.93 71 .0001 Y = 16.3 + 2.61(ILL) - 0.59(CEC) + IlOfd(KexVdX) 0.85 31 .0001 0.94 55 .0001 0.90 35 .0001 0.96 68 .0001 9 10 11 12 a @ Y Y = = = = 13.2 13.5 23.1 17.5 + + + + 0.90 (CLAY) 267 Cd(KexV dX) @ 175(FINE SILT) 15.I (Kex) 5.74 + I.09(CIAY) - I.00(CEC) + 189 Cd(KexVdX) + 0.OS(SILT) ' 11.4 + 0.93 (CLAYPIP) - 1.09(CEC) + 0.77(FINE SILT) + 208Cd(KexVdX) Y = 2.82 + 0.91(CLAY) - I.15(CEC) + 205 (d[Kex]/dX) + 0.14 (SILT) + 0.69 (MONT) 46 25 22 13 11 32 Independent variables: ILL, % illitic clay; CLAY, % clay hydrometer; d[Kex]/dD, diffusion gradient coefficient; FINE SILT, fine silt particle size fraction; CEC, cation exchange capacity; SILT, % silt by hydrometer; and CIAYPIP, clay particle size by pipette, (![KexVdD, Kex Diffusion Gradient Coefficient, estimated from Kex depletion gradients. 33 correlations with plant K uptake. Models 6 through 12 represent those equations with multiple independent variables. Spring wheat. K uptake was highly correlated to % illitic clay (ILL) Viiiich accounted for 72 percent of the variation in K uptake. Individually clay content, d[Kg^]/dD and fine silt particle size accounted for 59, 55, and 42 percent, respectively, of the variation in plant K uptake. illitic clay. The best two variable models both included percent A model including percent clay, d[Eg^]/dD and CEC independent variables accounted for 93 % of the variation in plant K uptake. Model 12 indicated that 96 percent of the variation in plant K uptake was accounted for by percent clay, dfE^j/dD, CEC, percent silt and percent montmorillinitic clay. However, not all regression coefficients were significant for this model. The best three-variable model, 8 , was selected as the best model and was tested for multicoll inearity. The existence of multicoll inearity results from intercorrelation between independent variables and it would lead to inflated variances in predicted values and coefficient estimates that are nonsignificant or have incorrect sign. The statistics used to test for multicollinearity were: (I) relative significance of coefficient estimates, (2) test for variance inflation (VIE) and (3) analysis of structure matrix (Table 14). All regression coefficients estimates were highly significant (E>0 .0001) . The VIF test indicated that significant multicollinearity existed for variables with VIF greater than 14.36, of which none were found among the variables. Eigenvalues values indicate little linear dependence between variables (bound of condition numbers less than 10). Thus 34 Table 14. Multicollinearity detection of regression model 8 with dependent variable plant K uptake and independent variables: clay, d[Kex]/dD, and CEC Dependent Variable: K Uptake Source Model Error C Total Sum of Squares 2224.7 166.3 2391.1 DF 3 16 19 Root MSE Dep Mean Variable INTERCEPT CIAY d[Kex]/dD CEC 3.22 36.66 R-square C.V. Parameter Estimate DF I I I I T for HO: Parameter=O 2.51 0.116 27.46 0.160 Variance Inflation I 0.00 2.27 1.34 2.11 F Value 71.327 Prob>F 0.0001 0.9304 8.79 Standard Error 8.53 1.07 . 194.15 -0.98 Variable I)F INTERCEPT CLAY I I d[Kex]/dD I CEC Mean Square 741.5 10.3 Prob > |T| 3.396 9.281 7.068 -6.147 0.0037 0.0001 0.0001 0.0001 Test 1/(1-0.9304) 14.36 Collinearity Diagnostics Number Eigenvalue Condition Number 1 2 3 2.10623 0.61890 0.27487 1.00000 1.84477 2.76815 Var Prop Var Prop Var Prop PdAY SMPE3QD CEC 0.0796 0.0535 0.8669 0.0915 0.8767 0.0318 0.0810 0.1498 0.7693 multicollinearity test indicates there were no significant interdependence between independent variables with regard to plant K uptake for model 8 and suggest this is a good regression model. For models 10 - 12 some degree of multicollinearity was found between independent variables. 35 To investigate the contribution of Ksol to plant K uptake, an experiment was conducted to determine Ksol flux in the soil rhizosphere (Table 15). Results, using the Amsterdam soil and the same experimental conditions as used in the twenty soil study indicated that Ksol flux accounted for 0.77 m o l e s K pot-1, or less than 4 percent of Table 15. Changes in IID soil solution K concentration as related to rhizosphere distance for spring wheat for the Amsterdam soil after 96 hours. 0.25 0.75 Rhizosphere distance (mm) 1.50 2.50 4.00 6.00 ---------------------- ItM---------------------0.13 0.210 total K uptake. 0.24 0.28 0.39 0.47 Thus, Ksol flux alone is relatively unimportant in overall spring wheat K uptake, at least in the early stages of plant growth and when mass flow is minimal. Results of this study indicate spring wheat plants utilize other than that extracted by I N RH4OAC. soil K Furthenirare, the amount of plant K obtained from sources other than Kex was highly soil dependent. It is hypothesized that plants are capable of utilizing K from Knex and Kmln soil forms in the near rhizosphere (0.25 mm). Evidence to support this hypothesis includes the extreme depletion of Ksol in the near rhizosphere (Table 15) and slow transport of Kex from greater distances (Table 11). In addition, spring wheat plants are capable of utilizing K in these forms in time periods as short as 96 hours. Reitemeier et al. (1947) reported Ladino clover utilized Knex on A l a b a m soils. Recently Richards, Bates and Sheppard (1988) reported that K uptake by 36 alfalfa was related to Knex on Ontario soils. Kudhenbuch and Jungk (1984) suggested that as much as 60% of the K taken up by rape was due to Knex. The results of this experiment further support this hypothesis. A significant relationship is evident by the coefficient of determination, between plant K uptake and percent illitic clay, and also fine silt, 5 - 2 um (Table 12). Illitic clay (hydrous mica) which contains interlayer or nonexchangeable K, ranged from 2.7 to 14.7 percent for the twenty soils. Chute and Quirk (1967) have reported that K diffusion from illitic clay (<2 um) was ten times slower than that from silt size particles. von Reicheribach and Rich (1969) indicated that K could be removed more easily from coarser particles (fine silt) than finer particles. This information further supports the hypothesis that nonexchangeable K is available to spring wheat seedlings and that it probably released from predominately fine silt and clay sized particles. In describing phytoavailable K according to a mechnistic approach, Barber (1985a) and Cushman (1982) have presented results showing the effective diffusion coefficient (Dc) for K is often most limiting. However K diffusion alone is not necessarily the most limiting factor in this model. Results of this study are based on Correlation of soil variables with phytoavailable K and not direct cause and effect relationships. However, if one uses strictly a quantity and intensity approach, transport processes do not play a major role in K availability. Instead the capacity of a soil to replenish Kso^ is of major importance. 37 These results suggest quantity and intensity relationships account for a majority of K availability. Regression model 8 (Table 13) involving clay, (^[KexVdD, and GEC accounted for 93 % of the variability in plant K uptake and had no significant multicollinearity (Table 14). Each of these variables influences Dc, but they also have a major role in quantity and intensity relations, as follows: (1) Clay represents a potential capacity of a soil to supply K through increased CEC and clay-sized K bearing minerals (both primary and secondary). Further, clay highly influences the soil volumetric water content (VWC). to have high VWC at -33 Kpa. Soils high in clay tend Increased VWC will generally increase KsoI as well as K diffusion (Barber 1985a). (2) The Kex diffusion gradient coefficient (^[KexVdD) is positively related with plant K uptake, and appears to represent K diffusion in the rhizosphere since fit changes with the square of the distance. In addition, it accounts for factors that limit Kex movement in the rhizosphere such as CaOO3, as indicated by the negative coefficient for CaCO3 in Table 16. Table 16. Regression of Kex concentration gradient (d[KexIZdD), (umol I cm“3 cm--*-) as dependent variable (Y) and independent variables: Kex; CACO3 ; and saturated paste Ca. Model I Y = Equation 2.6E-2 + 4.9E-2(Kex) - 5.8E-3(CaCO3) l.lE-4(pasCa) ■ R2 0.70 F P>T 12.2 .0002 38 (3) The importance of CEC in the K phytoavailability model appears contradictory due to the negative coefficient. However, CEC accounts for the resistance within the soil to supply K through diffusion. High CEC implies high K exchange capacity, increased K affinity and thus restricted K movement (trans­ port) . Furthermore, CEC is highly regulated by organic matter (CM) content, and exchange sites on CM have been found to have low affinity for K. This translocates into relatively small increases in K capacity with increased CEC due to CM. These results are similar to those found by Jaworski and Barber (1958) who found a negative regression coefficient for CEC in a linear regression equation for K uptake by alfalfa. The interrelation of these variables for various soils is apparent. The Beaverton has relatively high clay content (341 g Kg-1) suggesting a high potential to supply K. capacity Cation Exchange and d[Eg^]/dD on this soil, however, both indicate low phytoavailable K, as was indicated by low spring TAheat K uptake (27.12 umol pot-1). The Creston soil has a clay content of 134 g Kg-1, a value of 0.0644 for d[Kex]/dD, and a CEC of 13.5 cmol(-) Kg-1. For the Creston soil these values represent low K potential, a low to moderate d[Kex]/dD, and low CEC, and thus little resistance to K movement relative to K uptake. The Stryker soil series, with a clay content of 280 g Kg-1, d[Eg^]/dD of .0936, and CEC of 14.4 cmol(-)Kg-1 has moderately high K potential, moderate d[Kex]/dD and low CEC, resulting 39 in moderately high K uptake (44.11 umoles pot-1). Overall, these results show that the amount of Kex used by spring wheat seedlings is soil dependent and poorly related to total K uptake, ViMle % illitic clay was hic^ily correlated with K uptake. Spring wheat plants are capable of utilizing K from soil sources other than Kex. Fhytoavailability was best described by a multiple linear regression equation including the variables percent clay, Kex diffusion gradient coefficient and CEC. Apparently these variables were most highly related to plant K uptake, and thus most influential in soil K supply and transport in the rhizosphere. Temperature and Moisture Influences on K Fhytoavail ability (TMS) In the TMS study, plant K uptake was determined at five temperatures and three soil water potentials on four soils in a split plot design. Temperature and water potential were main plots and soils as subplots. The factor combinations used are described below, with three replications. a. Temperature: 8 , 12, 16, 20, and 24 C b. Water Potential: - 12 - 33 - 70 KPa c. Soils Series: Amsterdam, Beaverton, Edgar, and Kevin A high flow ceramic moisture plate with attached water column was employed in this study to regulate water potential. An isotemperature plate was attached to the upper surface of the ceramic moisture plate. It consisted of an aluminum plate containing access holes 4.8 cm ID and 2.0 cm in height in which soil columns were to be placed. Within the aluminum plate were conduit tubes, which when connected to a constant 40 temperature water bath would maintain soil columns at a constant temperature +-1.0 C. Soil columns, 4.5 x 2.0 cm (BD 1.25 gm cm--5) were prepared for each of the soil series and placed into the isotemperature plate access holes. Pondera wheat seedlings, (0.75 gm) five days old, were placed on soil columns for four days. Plant K uptake, and soil rhizosphere Kex concentration profiles were determined as described previously. Analysis of variance of the TMS study for net plant K uptake is presented in Table 17. Main plot effects, temperature, water potential plant K uptake were affected by these treatments. Sub plot effects and Table 17. Analysis of variance for net plant K uptake as influenced by temperature (TEMP) and water potential (KPA) on four Montana soils (SOIL). Dependent Variable: K uptake (umol pot-1) Source Model Error Corrected Total DF 59 120 179 Sum of Squares 25652.0 128.9 25780.9 Mean Square 434.78 1.04 C.V. 8.11 Source DF Anova SS Mean S 4 2 8 Error a 16 24.7 1.5 3 12 6 24 3971.2 1344.7 70.5 272.3 1323.5 112.2 11.5 11.1 56.3 1.2 Error b 48 . P>F 404.66 .0001 Root MSE 1.036 TEMP KPA KPA*TEMP SOIL S0IL*Tenp S0IL*KPA SOIL*KPA*TEMP ■ 18880.0 150.6 1112.9 F Value 4720.0 75.3 139.1 K uptake Mean 12.77 •F Value 3146 48.9 90.1 1093.8 93.5 10.3 9.2 P>T .0001 .0014 .0001 .0001 .0001 .0031 .0001 41 - O- Kevin -S - Beaverton Amsterdam -Q- Edgar Temperature ( C ) Figure 5. Potassium uptake QlO for spring wheat seedling on four Montana soils at a water potential of -12 Kpa. Table 18. Mean neta K uptake (umol pot-1) by spring wheat as influenced by temperature (C) and soil water potential (Ipa) on four Montana soils. Soil Series Water Potential (Ipa) Temperature 8 12 16 (C) 20 24 umol pot 1 Amsterdam 12 33 70 0.5 1.1 0.6 3.8 3.9 3.9 8.7 7.5 11.4 17.9 18.3 12.7 31.6 33.2 18.9 12 33 70 - 2.2 - 1.1 - 0.4 2.3 3.2 3.5 3.7 6.2 5.4 11.4 11.5 12.3 22.7 20.4 15.8 12 33 70 1.4 1.1 0.7 2.1 2.6 3.1 9.6 9.0 11.5 14.3 13.8 16.2 33.9 29.2 21.7 12 33 70 2.1 2.0 1.8 7.1 10.6 10.4 15.3 20.2 21.7 31.5 30.9 27.5 52.1 43.6 31.2 Beaverton Edgar Kevin 42 their interaction were all highly significant (P=O-Ol) indicating interactions were also highly significant. Thus, soil and these environmental factors had a major impact on plant potassium uptake. The dynamic effects of temperature are noted by the large Mean square term associated with this variable. It is further depicted in the mean plant K uptake values (Table 18). Temperature increases resulted in an exponential increase in K uptake by spring wheat across all soils (Table 18, Figure 5). At 8 C, statistically there was no net K uptake on any soil over that of the control plants (grown in the absence of soil). However, trends indicated greater K uptake by spring wheat on the Kevin soil than the other three, and an actual loss of plant K from plants grown on the Beaverton soil. The influence of low temperature on plant K uptake is also illustrated by data presented in Table 19 (Figure 5). Clearly, Table 19. Calculated QlO for K uptake by spring wheat on four Montana soils for temperatures 8 to 24 C, at a water potential of of -12 Kpa. Temperature Range C Amsterdam Soil Series Beaverton Edgar 8 - 12 43 498 26 63 12 - 24 6.1 8.0 7.0 3.8 Kevin QlO there is major change in the QlO for K uptake between 12 and 8 C on three of four soils in this study. Similar results have been presented by Carey and Berry (1978) with barley. They observed a sharp discontinuity in the QlO for Rb uptake occurring at 10 C. These results for K using a soil system and spring wheat support their findings. 43 The influence of water potential on K uptake is less dramatic as illustrated by the data in Table 18. Below 16 C water potential differences had little or no influence on plant K uptake on any of the four soils. However, at temperatures of 20 and 24 C, decreasing water potentials resulted in reduced K uptake. Thus, it was not until higher temperatures had stimulated plant growth and K demand that water potential had a significant influence on K uptake. This effect was most pronounced on the Kevin and least on the Beaverton soils. Generalized linear regression models for plant K uptake as influenced by temperature and water potential are presented in Table 20. The generalized form of the equation was: Y = C1 + C2 (a(T)b ) - C3 (a(T)b ) (Ipa)2 + C3 (T) (Epa) [4] Where a and b are coefficients associated with the power equation for temperature and Epa is water potential. Assumptions were made in setting the boundary conditions for the equations. The most important of these was that K uptake at any one temperature is a theoretical upper limit at the highest water potential, and decreased with decreases in water potential. The derived model would be a linear type, containing the exponential effects of temperature and the quadratic effects of water potential. Coefficients of multiple determination (R2) indicate more than 95 % of the variability in K uptake could be explained by independent variables temperature and water potential. Negative coefficients for the linear temperature by quadratic water potential term indicate the reduction in K uptake with this variable interaction. 44 Table 20. Regression equations of dependent variable K uptake (umol) and independent variables: water potential (Kpa) and temperature (T) for four Montana soils. Soil Series Equation R2 F P>T Amsterdam Y = -1.4 + 1.0(7.4E-4)(T3 *4) I .4E—4(7.4E—4)(T3 *4) (Kpa)2 + 6 .4E-3(T)(Kpa) 0.98 704 .0001 0.97 446 .0001 + I.2 (2 .OE-3)(T3 -O) I.IE-4(2.OE-3)(T3 -0) (Kpa)2 + 4.5E—3 (T) (Ipa) 0.95 287 .0001 Y = -1.2 + 0.91(5.OE-3)(T2 -9) 1.4E-4 (5.0E-3) (T2 -9) (Ipa)2 + 1.4E-2 (T) (Ipa) 0.96 326 .0001 Beaverton Y = -2.4 + 1.0(8.2E-4)(T3 *2) I.2E-4(8 .2E-4)(T3 -2) (Kpa)2 + 5.6E-3 (T) (Kpa) Edgar Y = -1.6 Kevin number of observations per soil, 45. In addition, differences in K uptake across soils can be seen by comparing variable coefficients for the Kevin and Edgar soils. Surface response models of plant K uptake on these four soils as influenced by temperature and water potential are shown in Figures 6 - 9. These figures indicate that the model is not exact, as K uptake at 8 C does increase with decreased water potential. This, however, is the result of mathematical computation and is relatively unimportant, since most plant growth occurs at higher temperatures. Rhizosphere Kex concentrations were determined on a reduced data set which consisted of main effects, soil, temperature and distance, 45 Figure 6 . Response function of plant K uptake (umol pot-1), temperature, and soil water potential; soil series Amsterdam. Figure 7. Response function of plant K uptake (umol pot 1J, temperature, and soil water potential; soil series Beaverton. „ K uPlato. P°‘' 46 8 . Response function of plant K Lptake (umol pot”1), temperature and soil water potential; soil series Edgar. K uptake, umoi pot1 'igure vtelBr Potential Figure 9. Response function of plant K uptake (umol pot 1), temperature and soil water potential; soil series Kevin. 47 at a single water potential of - 33 Kpa. Analysis of variance indicated that all main effects were highly significant (E>0 .0001). Interaction effects were all highly significant (E>0.0001) with the exception of SOIL*TEMP*D (Table 21). Mean rhizosphere Kex concentration indicated little change in concentration with distance at low temperatures (Table 22). As temperatures increased from 12 to 24 C there was a rapid reduction in near rhizosphere (0.25 mm) Kex. In general this is seen across all soils, but the amount of change in Kex was soil dependent. Generalized regression models were developed and indicated that Kex concentration was influenced primarily by the interaction square Table 21. Analysis of variance table for Kex (umol cm-3) at five distances (D) in the rhizosphere of spring wheat as influenced by temperature (TEMP) for four Montana soils (SOIL). Dependent Variable: Kex Source Model Error Corrected Total Source HEP TEMP EEP*TEMP SOIL D S0IL*D SOIDfcTEMP TRMPfcD SOTTl fcTRMPfcD DF Sum of Squares Mean Square 111 188 299 7537.18 149.23 7686.42 67.94 0.79 F Value 85.04 C.V. 7.21 Root MSE 0.886 P > F 0.0001 KVOL Mean 12.28 DF Anova SS Mean SS F Value P > F 2 4 8 3.2 220.21 1.61 1.6 55.05 0.20 8.0 272.50 0.4520 0.0001 3 4 12 12 16 48 6858.98 242.97 19.62 109.78 66.01 17.83 2286.32 60.74 1.63 9.14 4.12 .0.37 2910.83 77.33 2.08 11.65 5.25 0.47 0.0001 0.0001 0.0198 0.0001 0.0001 0.9986 48 root of distance and temperature (Table 23). The generalized form of the equation was: Y = C1 - C2 (D). - C3 (T) + C4 (T) (D)0 *5) - C5 (T)2 (D)0 -5 [5] The effects of temperature were most pronounced for the Kevin soil and Table 22. Mean (n=3) soil Kex concentration (umol cm-3) in the ihizosphere of spring wheat as a function of distance (mm) and influenced by temperature (C) on four Montana soils maintained at a water potential of - 33 Kpa. Soil Series Distance (mm) Temperature (C) 8 12 16 ■ un>o_L CIU 20 24 ■■ Amsterdam 0.25 0.75 1.50 2.50 4.00 13.34a 13.86 14.00 13.98 14.11 0.25 0.75 1.50 2.50 4.00 7.99 8.19 8.26 8.28 8.29 0.25 0.75 1.50 2.50 4.00 0.25 0.75 1.50 2.50 4.00 13.01 13.47 13.63 14.08 14.28 11.22 12.70 13.26 13.58 14.01 10.47 12.27 12.91 13.55 13.93 9.55 11.15 11.85 12.50 14.11 7.94 8.07 8.28 8.25 8.36 6.92 7.18 7.40 7.64 8.05 6.51 6.90 7.24 7.69 8.03 5.78 6.37 6.91 7.56 8.24 9.18 9.30 9.40 9.55 9.70 8.57 8.88 9,30 9.54 9.68 7.07 7.92 8.35 9.08 9.49 6.59 7.61 8.13 8.86 9.45 6.18 7.25 7.88 8.39 9.52 20.92 21.57 22.08 22.43 22.80 20.27 20.81 21.93 22.16 22.80 17.30 18.15 19.53 20.41 21,97 15.50 17.11 18.04 20.55 22.44 13.52 16.44 17.67 . 19.31 „ 21.87 Beaverton Edgar Kevin 49 least for the Beaverton soil. For all soils the coefficient of multiple determination and F value indicated high significance. This model type for rhizosphere Kex is reasonable when these results are compared to plant K uptake. Table 23. Regression equations of dependent variable soil Kex (Y) and independent variables: rhizosphere distance (D) and temperature, (T) for four Montana soils at - 33 Kpa. Soil Series Equation P>T Amsterdam Ya = 15.6 - 4.7E-4(D) - 0.33(T) + 6.3E-3(D0 *5) (T) - 6.3E-7(D0 *5) (T2) .97 150 .0001 .97 150 .0001 .97 151 .0001 .95 90 .0001 Beaverton Y = 9.3 - 1.3E-3 (D) - 0.18(T) + 2.0E-3(D0 *5) (T) + 2.8E-7(D°*5) (T2) Edgar Y = 10.6 - 2.IE—4 (D) - 0.25(T) + 4.1E—3 (D0 *5) (T) - 3.2E-7(D0 *5) (T2) Kevin Y = 23.6 - 1.0E-4(D) - 0.54(T) + 8 .8E-3(D0 '5) (T) - 1.2E-6(D0 *5) (T2) a units: Kex of umol cm 3 ; D, distance mm; and T, temperature C. Plant K uptake was influenced by temperature in accordance with published nutrient research conducted under both field and greenhouse conditions (Nielsen et al. 1960; Nielsen et al., 1961; and Smith, 1971). Differences across soils in plant K uptake reflected inherent differences in soil capacities to replenish solution K in the rhizosphere. Decreases in temperature have been shown by Yang 50 (1987) to increase the selectivity of exchange sites for K. Although decreases in temperature would have pronounced effects on both K release reactions and transport, the effects of temperature on the biological system are much more significant (Rovira and Bowen, 1973). Thus, at low temperature there is a large reduction in biological K demand, and less of an effect on chemical and physical processes associated with K availability in the rhizosphere. Similar effects of temperature on ion absorption have been demonstrated by Ching and Barber (1979) for c o m and by Nye and Tinker (1977). The effects of water potential, and associated volumetric water content (VWC), are less distinct (Table 24). decrease in K uptake with decreases in VWC. Clearly there is a Even though the number of data points are limited and skewed to the moist end of the plant available VWC range, these results indicate a major impact of decreased Table 24. Volumetric water content at three water potentials for four Montana soils. Water Potential (Kpa) Amsterdam - 12 - 33 - 70 0.4OOa 0.301 0.242 Soil Series Beaverton Edgar 0.425 0.347 0.279 — 3 <~m—3 dll 0.250 0.206 0.141 , Kevin 0.313 0.275 0.236 a Determined at a BD. 1.25 gm cm VWC on plant available K. Kuchenbuch, Claassen and Jungk (1986), evaluated the effects of VWC on K uptake by onions and concluded that decreasing VWC from 0.4 to 0.1 cm3 cm-3 decreased the effective diffusion coefficient for K one order of magnitude. In their study they found K uptake by onions decreased according to a curvilinear 51 response function. Similar results for K transport have been presented by Barber (1985a), Nye and Tinker (1977), and Claassen and Jungk (1984). Reductions in VWC result in a more tortuous diffusion path, increased ionic strength, and decreased soil-root contact. The present study was not conducted in a manner to allow determination of the contribution of each component. However, the significant shifts in K uptake that are evident between soils at the same water potential, but dissimilar VWC, does indicate that other soil properties (i.e. clay type, CEC) play an important role in influencing phytoavailable K. This is best seen in comparing the Beaverton and Kevin soil series, which differed much more in other soil properties and K uptake by plants grown on them than they do in VWC. It should also be pointed out that the current study was conducted under conditions of minimal mass flow, which may have reduced the magnitude of influence of water potential on K movement and uptake. The interaction of temperature and moisture on K diffusion to ion exchange resins have been documented by Schaff and Skogley (1982). These results from experiments in which intact plants were used under carefully controlled conditions provide further evidence of such interactions. Overall K uptake was not limited by VWC (within the water potential range of -12 to -70 Kpa) until temperature exceeded approximately 16 C on all soils (Table 22). At greater temperatures plant demand for K increased, and transport in the rhizosphere (i.e. K diffusion) then became limiting (Figures 6 - 9). Rhizosphere soil Kex concentrations closely reflect the influences of temperature on plant K uptake. The effects of temperature on K transport in the rhizosphere have been discussed by Barber (1985a). 52 But in regard to changes in Kex the greatest influence is that directly related to plant K demand. Thus temperature governs the magnitude of the chemical potential gradient induced by the plant which affects rhizosphere soil Kex depletion gradients. Effects of K Rate on Soil and Plant K (KRS) In the KRS study, 200 grams of four soils were equilibrated with one of five rates of K as KCL, in a complete factorial design. Montana soils Amsterdam, Beaverton, Edgar, and Kevin were used with K added at equivalent rates of 0, 50, 100, 200, and 400 Kg K Ha-1 on a field basis, applied in a 20 ml aliquot. Each soil was brought to a moisture content equivalent to a water potential of - 33 Kpa and allowed to equilibrate for 24 hours. Prepared soils were packed into cylinders (4.5 X 1.5 cm, BD 1.25 gm cm-3) and placed on a high flow ceramic moisture plate. - 33 Kpa. Water potential was adjusted and maintained at Pondera wheat seedlings five days old (0.75 gm), were placed on the columns for 96 hours. Plant K uptake and Kex concentration profiles were determined as described previously. Spring wheat seedlings generally increased their K uptake (umol pot-1) with greater soil K application on all four soils (Table 25) . However, soils did not respond equally to K additions as indicated by the significant interaction between SOIL*KRATE for K uptake (Table 26) . As K rate increased from 0 to 400 Kg ha-1, K uptake increased 24, 13, 14, and 7 umol pot-1 on the Amsterdam, Beaverton, Edgar, and Kevin soil series respectively. This is a 74 percent increase on the Amsterdam soil, while only a 14 percent increase in K uptake on the Kevin soil. 53 Table 25. Mean (n=3) plant K uptake by spring wheat as influenced by K additions as KCL on four Montana soils at 24 C and -33 Kpa. K Rate Kg ha-1 Soil Series Beaverton Edgar Amsterdam Kevin ------- umol pot -I 0 50 100 200 400 32.24 36.30 42.21 46.55 56.21 32.30 35.32 38.58 43.01 48.92 38.21 43.95 49.92 56.32 52.14 54.54 53.91 55.62 58.41 62.04 4.15 4.62 3.85 5.21 LSD(0.05) Table 26. Analysis of variance table for plant K uptake (umol pot-1) by spring wheat as influenced by K addition (KRATE) for four Montana soils (SOIL), 24 C and water potential - 33 Kpa. Dependent Variable: K uptake Source Model Error Corrected Total DF Sum of Squares 21 100 119 10906.9 84.9 10991.8 Mean Square 519.8 0.87 Root MSE 1.49 F Value 599.30 C.V. 3.19 P > F 0.0001 K uptake Mean 46.60 Source DF Anova SS Mean S F Value P > F SOIL KRATE SOIL*KRATE 3 4 12 5614.3 4280.5 875.1 1871.4 1070.1 72.9 843.56 482.36 32.87 0.0001 0.0001 0.0001 Linear regression equations (SAS Institute, 1987) between plant K uptake and K addition on four Montana soils are presented in Table 27. All regression models indicate that K uptake was best explained by quadratic function of K rate on the Amsterdam, Beaverton, and Edgar soil series. The generalized form was: Y = C1 + C2 (X) + C3 (X2) [6] 54 Table 27. Regression equations for dependent variable plant K uptake (Y) and independent variable potassium addition (KRATE) for four Montana soils 24 C and water potential - 33 Kpa. Soil Series Equation R2 F P>F 0.97 447 .0001 Y = 28.7 + I.IE-I(KRATE) - 1.4E-4(KRATE2) 0.96 288 .0001- Y = 37.8 + I.SE-I(KRATE) - 2.8E-4(KRATE2) 0.94 188 .0001 Y = 53.8 + 2.IE-2 (KRATE) 0.78 95 .0001 ' Amsterdam Ya = 32.2 + 9.OE-2 (KRATE) - 7.8E-5(KRATE2) Beaverton Edgar Kevin a Units: plant K uptake (Y), umol pot-1; KRATE, Kg ha-1. Table 28. Analysis of variance table for Kex concentration (umol cm-3) in the rhizosphere of spring wheat as influenced by distance (D) and K addition (KRATE) on four Montana soils (SOIL). Dependent variable: Kex Source Model Error Corrected Total DF Sum Squares Mean Square 101 200 299 9329.58 18.20 9347.79 94.23 0.09 Source DF Anova SS SOIL KRATE D S0IL*KRATE S0IL*D KRATE*D SOIL*KRATE*D 3 4 4 12 12 16 48 8204.67 341.38 664.55 41.76 50.78 11.71 14.70 F Value Pr > F 1026.20 0.0001 Root MSE 0.301 C.V. 2.24 Kpx Mean 13.44 Mean S . F Value Er > F 2734.89 85.34 166.13 3.48 4.23 0.73 0.31 30042.65 937.52 1825.03 38.23 46.49 8.04 3.37 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0005 55 Plant K uptake on the Kevin soil series indicated a linear response function. These results for the first three soils are comparable with those of Nemeth (1975) using alfalfa which indicated K uptake generally followed a quadratic response. Statistical analysis of rhizosphere soil Kex concentration are presented in Table 28. Main effects of SOIL are highly significant (pcO.OOOl) as are KRATE and D. The dramatic effects of SOIL and D are evident by the large Mean Squares for both terms. Interaction combinations also show high significance (P< 0.001). Rhizosphere Kex concentration decreased with distance from the bulk soil to the soil-root interface for all soils and K rates (Table 29). As mean planar D decreased from 4.00 to 0.25 mm Kex concentration on the Amsterdam decreased 3.85 umol cm-3, 26 percent, at the zero K rate. Differences between 0.25 and 4.00 mm Kex concentrations increased with increasing K rate. The relationship of Kex concentration with respect to distance (D) could best be described by a square root function (Table 30), and was highly significant across K rates and soils (P<0.006). The equation had the general form: Y = C1 + C2 (D0 -5) [7] For the Amsterxfem soil, as K rate increased the slope of the diffusion gradient coefficient (d[Kex]/dD), representing change in Kex concentration with distance, increased and was significant at the 0.05 level. Kex. Higher K rates, therefore increased the diffusion gradient of This effect was noted on the Amsterdam and Beaverton soils. 56 Table 29. Mean (n=3) rhizosphere Kex concentration as influenced by rhizosphere distance (D) and K rate, for four Montana soils, temperature 24 C and water potential -33 Kpa. Soil Series Distance (mm) 0 50 K rate (Kg ha"-1) 100 200 400 - umol cm ^ Amsterdam 0.25 0.75 1.50 2.50 4.00 10.61 11.57 12.29 13.40 14.46 10.27 11.48 12.73 13.83 14.94 10.18 11.81 12.82 14.46 15.61 10.94 12.39 13.79 15.18 16.77 12.44 14.08 15.76 17.35 19.27 0.25 0.75 1.50 2.50 4.00 6.46 7.44 8.06 8.44 8.89 6.37 7.47 8.36 8.95 9.39 6.86 7.88 8.86 9.69 10.20 7.88 8.79 9.33 10.19 10.88 9.34 10.24 10.97 12.11 13.23 0.25 0.75 1.50 2.50 4.00 7.30 8.25 8.67 9.10 9.44 6.58 7.58 8.32 9.10 9.73 6.48 7.63 8.45 9.29 10.10 7.59 8.48 9.20 10.20 11.00 10.14 10.91 11.52 12.18 13.08 0.25 0.75 1.50 2.50 4.00 17.99 19.57 20.52 21.53 23.76 17.56 20.98 21.78 22.90 24.09 19.35 21.88 22.58 24.13 25.26 19.91 21.56 22.88 24.01 25.80 20.33 21.88 22.65 24.41 26.04 Beaverton Edgar Kevin The diffusion gradient coefficient for the Edgar soil increased with low K rates, but decreased at higher K rates. For the Edgar soil this decrease was highly significant between coefficients (P< 0.0001). For the Kevin soil d [KexJZdD showed no significant changes 57 Table 30. Regression equations between rhizosphere distance (D) and Kex (Y) at five rates of K addition for four Montana soils, temperature 24 C and water potential - 33 Kpa. Soil Series Ka Rate Equation R2. F P>T Amsterdam 0 50 100 200 400 Y Y Y Y Y = = = = = 9.30 + 0.0811(D0 *5) 8.77 + 0.0994(D0 *5) 8.49 + 0.1149(D0 *5) 9.02 + 0.1229(D0-5I 10.15 + 0.1441(D0 *5) 0.997 0.997 0.993 0.999 1.000 974 1059 416 55734 61913 .0001 .0001 .0003 .0001 .0001 0 50 100 200 400 Y Y Y Y Y = = = = = 5.87 5.62 5.89 6,97 7.97 + + + + + 0.0513(D0 -5) 0.0637(D0 -5) 0.0720(D0 *5) 0.0698 (D0 •5) 0.0822(D0*5) 0.964 0.960 0.971 0.994 0.995 80 72 133 489 628 .0029 .0035 .0014 .0002 .0001 0 50 100 200 400 Y Y Y Y Y = = = = = 6.86 5.67 5.41 6.46 9.19 + + + + + 0.0435(D0 *5) 0.0664(D0 -5) 0.0757(D0 -5) 0.0726(D0 *5) 0.0609(D0 *5) 0.945 0.989 0.992 0.997 0.998 51 274 379 1058 1911 .0055 .0005 .0003 .0001 .0001 0 50 100 200 400 Y Y Y Y Y = = = = = 16.18 16.50 17.98 18.28 18.43 169 60 76 271 305 .0010 .0058 .0032 .0005 .0004 Beaverton Edgar Kevin + + + + + 0.1153(D0 -5) 0.1271(D0 -5) 0.1196(D0-5) 0.1190(D0-5) 0.1188(D0-5) 0.983 0.949 0.962 ' 0.989 0.990 a K rates applied as KCl, Kg ha"1. b units: distance (D) mm x(10"3) and Kex (Y) umol cm"3 . with increasing K rates. ' Thus on the Kevin soil, K rate did not influence the Kex diffusion gradient coefficient. A graphic representation of the influence of K rate on Kex concentration with respect to distance is presented for these four soils in Figures 10 - 12. For the Edgar, and to a lesser extent on the Amsterdam and Beaverton soils, an important feature is indicated in the 58 Figure 10. Response surface of mean Kex concentration as influenced by rhizosphere distance and K addition, soil series Amsterdam. Figure 11. Response surface of mean Kex concentration as influenced by rhizosphere distance and K addition, soil series Beaverton. 59 Figure 12. Response surface of mean Kex concentration as influenced by rhizosphere distance and K addition, soil series Edgar. Figure 13. Response surface of mean Kex concentration as influenced by rhizosphere distance and K rate, soil series Kevin. 60 mean surface response figures. With the first K addition the Kex concentration of the first rhizosphere distance decreased. The effect of increased Kex depletion as a function of K rate was significant for two of the three soils and represents a priming effect that low K additions has on potential rhizosphere K depletion. Initial soil Kex concentration and d[Kex]/dD values for each soil and K rate are listed in Table 31. By substituting the initial Kex concentration (time=zero) into the regression equation an estimate was Table 31. Initial soil Kex concentration (Kex), diffusion gradient (d[Kex]/dD) and diffusion distance for four Montana soils, temperature 24 C and water potential -33 Kpa. Soil Series K rate (Kg Ha-1) Kex (umol cm-3) ^[KexV d D Diffusion3 distance (mm) Amsterdam 0 50 100 200 400 11.52 15.27 16.32 17.42 20.89 0.0811 0.0994 0.1149 0.1229 0.1441 4.2 4.3 4.6 4.7 5.4 0 50 100 200 400 8.93 9.72 10.89 11.77 12.84 0.0513 0.0637 0.0720, 0.0688 0.0822 3.5 4.1 4.8 4.9 5.5 0 50 100 200 400 9.81 10.41 10.92 11.74 13.95 0.0435 0.0664 0.0757 0.0757 0.0609 4.6 5.1 5.2 5.3 6.1 0 50 100 200 400 22.94 23.64 24.27 25.53 27.15 0.1153 0.1271 0.1196 0.1190 0.1188 3.4 3.2 2.9 3.6 5.1 Beaverton • Edgar Kevin a Diffusion distance (mm) calculated from equation of d[Kex]/dD and initial Kex. 61 made on the extent into the rhizosphere to vSiich Kex was depleted. These results indicate that increased K rates, in general, increased the diffusion distance of Kex. With no added K, Kex diffusion distances ranged from 3.4 to 4.2 mm, while at the highest K rate it ranged up to 5.1 to 6.1 mm across soils. This effect suggests that recently added K is more mobile in soils than native K. By intergrating the regression equation representing Kex as a function of distance, from the soil-root interface to the estimated distance of diffusion, the estimated soil Kex flux can be calculated. SKex flux in general increased with increasing K rate for all soils (Table 32). Only on the Beaverton soil at the highest K rate did SKex flux actually decrease. A comparison of the difference between plant and soil flux (Delta K Flux) indicates that K additions resulted in increased discrepancies between the two values. However, examining the values for % Kex utilized (Table 32), there is a trend toward increased utilization of Kex as K rate increased. Amsterdam and Kevin soils. This trend was clearly evident on the Qn the Beaverton and Edgar soils the % Kex utilized initially increased with K additions, but decreased at higher K levels. These results indicate that adding moderate amounts of K to soils generally increased the availability (plant utilization) of Kex in the rhizosphere of spring wheat. In summary, four points can be made from the results of the KRS study. They are: (I) plant K uptake as influenced by K additions responded according to quadratic functions; (2) diffusion gradients of Kex (d[Kex]/dD) were highly influenced by K rate and there were no 62 Table 32. Mean Kex plant, soil and difference flux values as influenced by five K additions on four Montana soils, 24 C and water potential of - 33 Kpa. Soil Series K Rate Plant Flux Kg ha-1 SKex Flux '■ Delta K Flux ^ex Utilized nmol ItT 2 s-1 - (%) Amsterdam 0 50 100 200 400 61.9 69.6 80.9 89.2 107.7 21.8 27.5 36.0 38.8 56.9 40.0 42.0 45.1 50.3 50.7 35.2 39.6 44.4 43.6 52.9 0 50 100 200 400 52.1 67.7 73.9 82.4 93.7 10.7 16.8 23.8 24.1 20.5 41.8 50.8 50.1 58.2 73.2 20.4 24.8 32.3 29.3 21.9 0 50 100 200 400 73.2 84.1 95.6 107.9 99.9 13.5 23.8 28.4 27.3 28.6 • 59.7 60.3 67.3 80.5 78.0 18.4 28.4 30.3 25.3 21.9 0 50 100 200 400 104.5 103.3 106.6 111.9 118.9 22.3 22.3 23.2 26.6 45.6 82.0 81.0 83.3 85.3 73.2 21.4 21.6 21.8 23.8 38.4 Beaverton Edgar Kevin consistent trends across soils; (3) Kex depletion in the near rhizosphere (0.25 ram) was enhanced by K additions; and (4) small K additions increased Kex utilization on soils initially, but higher levels affected soils differently. Spring wheat K phytoavailability responded to K fertilizers according to accepted theories. However, of the total plant K flux, only 18 - 50 % was explained by fluxes of soil Kex. These results are consistent with those of Kuckeribuch et al. (1986) who have shown a major portion of K uptake by rape was not explained by soil fluxes of 63 Kex. Therefore, a majority of plant available K came from soil K forms other than Kex, most probably from Knex and Km^n . Although no direct evidence for this was obtained in this experiment, these results strongly suggest these K forms are plant available. Further, even with the addition of K fertilizer, and concurrent enhancement of the Kex, plants continue to utilize K other than Kex on a majority of soils. The depression effect of low K rates on soil Kex in the near rhizosphere (< 0.5 mm), and related influence on d[Kex]/dD for the Amsterdam, Beaverton and Edgar soils , are not easily explained. are three hypothesis for this phenomenon. There They are: (I) plant deficiency of chloride on these soils, thus a subsequent stimulation by KCL additions; (2) a priming effect of K additions on the Knex and Km^n in the near rhizosphere; and (3) an increase in ionic strength in the soil solution due to Cl which facilitates K dissolution from Knex and Kmin forms. Although chloride concentrations are lower on the Edgar and Beaverton soils (Table 33), these results do not seem completely plausible, due to the low requirement of Cl by spring wheat (Fixen 1986). The priming effect is a possibility, but this research provides Table 33. Soil extractable chloride and electrical conductivity (EC) on four Montana soils. Soil Series Amsterdam Beaverton Edgar Kevin Chloride EC mg kg-1 S Hf' 49.7 33.6 32.6 48.8 15.5 8.5 6.2 12.5 64 no tangible evidence for support. The influence of added K on ionic strength may, however, be an important relationship. Mclean and Watson (1985), postulated that 'mass flow triggered' diffusion flew may increase K availability. In this process, mass flow of 'ions' to the near rhizosphere leads to an excess of major Cations in the soil solution. plant absorption. (ie. Ca, Mg, Na) Potassium is removed from the solution because of The result is stimulated removal of K from the rhizosphere due to increased ionic strength. In the PRS study, mass flow was near zero, therefore KCL additions were the main mechanism enhancing soil K dissolution. Supporting evidence for this hypothesis is the low EC values reported for the Edgar and Beaverton soils relative to the other two soils (Table 33). Kuchenbuch and Jungk (1984) have reported similar results with the application of other fertilizer salts. Additional research is needed to elucidate this ■ process as a mechanism of stimulating K release in the rhizosphere. 65 SUMMARY AND CONCLUSIONS Soil characteristics and rhizosphere environment conditions greatly influence the phytoavailability of K. Soil characteristics which are strongly related to plant available K are: clay content, Kex diffusion gradient coefficient, and cation exchange capacity. Spring wheat in these studies utilized part of Kex, but quantities differed across soils. Potassium was obtained from other K forms, presumably Knex and KmJjn in the near rhizosphere over Kex at greater distances. Potassium from these soil forms is thus apparently plant available, within time periods of hours, to spring wheat seedlings. Release of Knex appears to depend on the characteristics of the soil solution and the experimental conditions. Temperature had a large dynamic influence on spring wheat K uptake. 8 C. Little or no K was absorbed by wheat seedlings across soils at Increasing temperature caused an exponential increase in K uptake to the highest experimental treatment (24 C). Temperature affected plant metabolism and nutrient absorption more than K equilibria and transport in the soil. Water potential had little influence on K uptake until temperature reached 16 C on any soil. This interaction is highly significant and indicates that K absorption is not limited by transport until K demand exceeds a certain threshold (when mass flow is near zero). Until that critical demand is reached, all phytoavailable K is obtained from the near rhizosphere. With increased temperature there is increased demand 66 on K at greater rhizosphere distances. Potassium additions increased spring wheat K uptake nonuniformly across soils (Table 25). Soils with low plant K uptake (< 40 umol pot-1) at the zero K rate level never reached maximum K absorption with K additions, as did soils with higher zero rate levels. Data on soil Kex rhizosphere K concentrations indicated K additions enhanced the availability of Kex supplies in the near rhizosphere. occurred on two of four soils. This effect Generally as K rate increased, wheat utilized a greater portion of Kex over other soil K forms, but was soil dependent. Spring wheat phytoavailable K is very sensitive to K soil equilibria, as governed by several soil parameters. But the role of environmental factors is critical in regulating both plant K demand and soil K supply processes. Thus to develop a general overall model for predicting K requirement and uptake by spring wheat seedlings it would be necessary to combine the effects of these two major systems. The present study was not designed to be a mechanistic approach which accounts for individual components of both systems, such as K selectivity coefficients and the effects of temperature on MichaelisMenten kinetics. Such an approach requires multiple formulas for both soil and plant parameters, each with several variables. The approach used in this study was to consider the system as a Vihole, identify those soil components which are relatively strongly influential in plant K availability and uptake, and determine or calculate the strength of the relationship. Although correlation does not imply cause and effect, it does suggest components which in the 67 overall system influences plant available K. The use of intact spring wheat seedlings provides a stronger basis that other less direct methods for correlating plant K uptake directly to soil components. Soils can be compared based on differences in soil K quantity and intensity characteristics and soil-root interface effects. With this approach, multiple regression equations from the ERTS and the TMS studies were combined into an overall equation which predicts spring wheat phytoavailable K as influenced by soil and environmental factors (Table 34). Such a model is limited, since it is restricted to soil factor limits inherent in the twenty soils and within the experimental conditions. Furthermore, only four soils were used in this model, so it can be considered as a "first approximation" only. Multiple linear regression models were prepared by calculating the predicted plant K uptake (EPKI) and combining it to the original model developed in the TMS study. Since there was significant interaction between temperature and water potential with soils, cross variables containing EEKE were added to the model. Results of forward stepwise regression resulted in a highly significant model containing four variables (Table 34). The equation took the generalized form: Y = -C0 + C1 (EPKI) + C2 (EPKE) (Kpa) - C3 (Kpa2) (3.0E-4 (T2 *9)) + C4 (EPKE)(3.OE-4(T2 •9)) [9] As in the TMS study the temperature factor (T) was incorporated in cross effects both as linear and exponential components. Water potential (Kpa) occurs in two variables as a linear and as a quadratic term. The term with water potential squared and 68 temperature exponential retained its negative coefficient from the original model in the TMg study. All variables in the model had significant regression coefficients. This model predicts 94 % of the variation in plant K uptake by spring wheat seedlings on these four soils. Such an empirical model within the experiment boundary conditions of soil and environment relates soil and environmental factors to plant K availability. Such a model has the potential to help one predict K uptake by spring wheat seedlings on numerous soils across a wide range of growing conditions. Table 34. Multiple linear regression model for the influence of temperature (T), water potential (KPA), and estimated phytoavailable K index (EEKI) on spring wheat K uptake. Dependent Variable: Plant K uptake umol pot--*Analysis of Variance Source DF Model Error C Total 4 175 179 Root MSE Dep Mean C.V. Variable3 INTERCEP EPKE *TT KPA*KPA*TT KPA*T EEKI Sum of Squares 24465.62 1464.67 25930.30 2.89 12.81 22.58 Parameter Estimate DF I I I I I -5.88475 1.65e-2 —8.31e—5 7.49e-3 0.10218 Mean Square 6116.40 8.36 F Value Brob>F 730.790 0.0001 R-square Adj R-sq 0.9435 0.9422 Standard T for HO: Error Barameter=O 0.982695 0.000352 0.000008 0.001241 0.023184 -5.9 46.7 -9.4 6.0 4.4 Brob > |T| 0.0001 0.0001 0.0001 0.0001 0.0001 a Variable description: REST, estimated phytoavailable K index; KPA, soil water potential Kpa; T, temperature C; and TT, exponential temperature component TD= 5.OE-S(T)2 *91. 69 Further refinements can be added as more soils and conditions are studied. Further research is needed to correlate this method of assessing soil fertility (using intact seedlings under controlled growth conditions) with K uptake of plants grown under greenhouse of field conditions. Such results could be employed to help predict plant K requirements, which on the Northern Great Plains appear to be highly seasonally dependent (Skogley and Haby, 1981 and Veeh and Skogley, 1986). Although determining these soil characteristics would be impractical on a yearly basis, it would prove useful in establishing baseline data on major soil series cropped to spring wheat. Further, with refinement of proper techniques, other plant species could be included, as has been done here with spring wheat and previously with rape. Results of the KRS study provide an estimate of the upper limit of spring wheat K demand under these growing conditions. The results further elucidated the soil phenomenon of Kex release in the near rhizosphere, triggered or primed by addition of KCL. Although the process is yet unclear, the hypothesis of diffusion aided mass flow may provide an explanation for this effect. Since the plasma membrane of a root hair acts as an ion selective membrane, ion absorption is specific. Potassium and nitrate are absorbed while calcium and other multivalent cations are excluded. These ions remain in the soil solution and accumulate if mass flow becomes significantly great. With the experimental conditions in this study, mass flow was minimal and yet calcium and magnesium did 70 accumulate. At the same time as K was being removed from the soil-root interface soil solution a counter ion gradient of protons entered the rhizosphere. Increased proton concentration would be expected to influence K exchange and the weathering of soil minerals. These effects were not directly accounted for in the present studies. The two processes, divalent cation accumulation and proton efflux, radically alter the ionic strength and ion speciation in the near rhizosphere soil solution. Other researchers (Sparks ,1980; Yang, 1987; and Jardine and Sparks, 1984) have described K release thermodynamically using soil solution analysis. However, none of these researchers have examined the soil system under conditions where: (I) K in solution was undergoing continuous depletion; (2) solution divalent cation concentration is static or increasing; and (3) proton concentration is increasing. Such an experiment, although extremely difficult to design, would yield a wealth of information about soil K equilibria and K phytoavailability. An overall model of phytoavailable K for Montana soils, as summarized by this research is presented in Figure 14. It suggests that as much as 80 percent of plant K uptake is the result of K release in the near rhizosphere (<0.5 mm) from Knex and Kmin forms. The quantity of K released (Twenty Montana soil study) was highly soil dependent indicating either inherent differences in the quantity of K in these forms for a specific soil and/or the .rate of release form these forms. Other researchers (Reitemeier et al., 1988; Bertsch and Thomas, 1985; Barber and Mathews (1962); and Talibudeen et al., 1978) have 71 PHYTOAVAILABLE POTASSIUM „ 2 10 Log Distance (um) 3 10 10 ' 10 10 H----- 1----- h 5 t Root Surface Root Hair 0 - 8 0 % of K uptake Zone of K release from nonexchangable forms and K - minerals 20 - 90 % of K uptake Zone of K release from exchangeable forms < 5 % of K uptake Zone of K depletion of solution K Figure 14. Model of phytoavailable K in the rhizosphere of spring wheat for Montana soils. suggested K release from illite and K minerals is plant available, however none have actually quantified plant and soil K fluxes. It is my hypothesis that K is released from micas, hydrous micas, and the surfaces of feldspars of particle sizes ranging from 10 to 0.5 um in the near rhizosphere and accounts for a significant portion of plant K uptake. Potassium release is due to (I) intense biological depletion of soil solution K; (2) root exudation of protons and organic acids; and (3) accumulation of divalent cations in the soil solution in the near rhizosphere. The total quantity of K and more importantly rate of K release is governed by the quantity of K-minerals and extent of chemical weathering which has occurred. 72 Evidence for support of this hypothesis can be divided into plant and soil aspects. Plant roots are capable of depleting K in the soil solution (Table 15). Nye and Tinker (1977) reviewed the literature and reported that in the rhizosphere plant roots of many species exude organic compounds and that roots are capable of reducing the pH at the soil root interface to a value of 4.0. In terms of soil. Song and Haung (1983) found that in the presence of citric and oxalic acids, K release from K-bearing minerals and K dissolution increase and the weathering sequence was biotite > microcline = othoclase > muscovite. von Reichenbach and Rich (1969) have found that the rate of K release was faster from soil particles 2 - 5 urn, intermediate for particles 5 - 20 urn and slowest for particles less than 0.2 urn. In general Montana soils, in comparison to soils of the midwest and southern United States, have been weathered less. This is due to the ustic moisture and frigid temperature regimes which results in decreased chemical weathering of silt and coarse clay sized K-bearing minerals. Thus in the presence of a plant root, K is more likely on these Montana soils to be released from micaceous and feldspar materials in the near rhizosphere. Any soil test which measures phytoavailable K must be able to assess K release rates from these soil K-bearing minerals in order to provide a useful means of predicting small grain K requirements. 73 LL'I'EHMURE CITED 74 UTERAIURE CITED Abbas Al-Ani, M. K., R. K. Hay. 1983. The Influence of Growing temperatures on the growth and morphology of cereal seedling root systems. J. of Expt. Bot. 34:1720-1730. Barber, S. A. 1962. A diffusion and mass flow concept of soil nutrient availability. Soil Sci. 93:39-49. Barber, S. A. 1972. Influence of the plant root on ion movement in Soil. Sci. 18:522-564. Barber, S. A. 1985a. Nutrient absorption by plant roots. In Soil Nutrient Bioavailability, A Mechanistic approach. John Wiley and Sons, New York. Barber, S. A. 1985b. Potassium availability at the soil-root interface and factors influencing potassium uptake. In Potassium in Agriculture, R. 0. Munson (ed.) Chap. 11:301-324. ASA-CSSASSSA, 677 S. Segoe Road, Madison W I . Barber, T. E., B. C. Mathews. 1962. Release of non-exchangeable soil potassium by resin-equilibration and its significance for crop growth. Can. J. of Soil Sci. 42:266-272. Bardsley, C. E. and J. D. Lancaster. 1960. Determination of reserve sulfur and sulfate in soils. Soil Sci. Am. Proc. 24:265-268. Beckett, P. 1964. Potassium-calcium exchange equilibrium in soils: specific adsorption sites for potassium. Soil Sci. 97:396-383. Bertsch, P. M. and G. W. Thomas. 1985. Potassium Status of temperate region soils. In R. D. Manson (ed.), Potassium in Agriculture, p p . 131-159. ASA, CSSA, SSSA, Madison, W l . Bower, C. A., R. F. Reitemeier, and M. Fineman. 1952. Exchangeable cation analysis on saline and alkaline soils. Soil Sci. 73:251261. Brown, D. A., B. E. Fulton B. E. and R. E-, Phillips. 1964. Ion diffusion: A quick-freeze method for the measurement of ion diffusion in soil and clay systems. Soil Sci. Soc. PSroc. 38:628-631. Carey, R. W. and J. A. Berry. 1978. Effects of low temperature on respiration and uptake of rubidium ions by excised barley and c o m roots. Plant Physiol. 61:858-860. 75 Ching P. C. and S. A. Barber. 1979. Evaluation of temperature effects on K uptake by c o m . Agron. J. 71:1040-1044. Chute, J. H. and J. P. Quirk. 1967. Diffusion of potassium from mica-like materials. Nature (London) 213:1156-1157. Claassen, N. and A. Jungk. 1984. Effect of K uptake rate, root growth and root hairs on potassium uptake efficiency of several plant species. Z. Pflanzenemadhr. Bodenk. 147:276-289. Cook, M. G. and Hutcheson Jr. 1960. Soil potassium reactions as related to clay mineralogy of selected Kentucky soils. Soil Sci. Soc. Am. Proc. 24:252-256. Crank, J. 1956. The mathematics of diffusion. Oxford, 30-31, and 245-256. Claredon Press, Cushman, J. C. 1982. Nutrient transport inside and outside the rhizosphere: Theory. Soil Sci. Soc. Am. J. 46:704-709. Danielson, R. E. 1967. Root systems in relation to irrigation. In Irrigation of Agricultural lands, Hogan R. M . , Haise H. R. and Edminster T. W. , pp. 390-424. Agronomy monograph No. 11, Am. Soc. Agron. Madison, NI. Feigenbaum, S. and J. Hagin. 1967. Evaluation of methods for determining plant available K based on K uptake. J. of Soil Sci. 18:200-203. Fixen, P. E., R. H. Gelderman, J. R. Gerwing, and F. A. Cholick 1986. Response of spring wheat, barley, and oats to Cl in KCL fertilizer. Agron. J. (Submitted). Freund, R. J. and R. C. Littell. 1986. Multicollinearity: Detection and remedial measures. In SAS system for regression, pp 75-103. SAS Institute Inc. Cary, NC. Gee, G. W. and J. W. Bauder. 1986. In Methods of Soil Analysis Part I. Hiysical and Mineralogical Methods. Agronorrry Monogram no. 9. pages 383-410. Havlin, J. , Soltanpour, P. N. 1980. A nitric acid plant tissue digest method for use with inductively coupled plasma spectrometry. Commun. in Soil Sci. and Plant Anal. 11:969-980. Hsieh, J. J. C., Gardner, W. H., Campbell, G. S. 1972. Experimental control of soil water content in the vicinity of root hairs. Soil Soc. Am. Proc. 36:418-421. Jackson, M. L. 1958. Soil chemical analysis. Prentice-Hall. ppl51-154. Engelwood Cliffs N. J. 76 Jardine, P. M . , Sparks, D. L.. 1984. Potassium-calcium exchange in a multireactive soil system: I. Kinetics. Soil Sci. Am. J. 48:39-45. Jaworski, C. A. and S. A. Barber. 1958. Soil properties in relation to potassium uptake by alfalfa. Soil Sci. 87:37-41. Kudheribuchz R., Jungkz A. 1982. A method for determining concentration profiles at the soil root interface by thin slicing rhizospheric soil. Plant and Soil. 68:391-394. Kucheribudh R., Jungkz A. 1984. Influence of potassium supply on the availability of potassium in the rhizosphere of rape (Brassica napus). Z. Pflanzenenadhr. Bodenk. 147:435-448. Kudheribudhz R . , N. Claassenz and A. Jungk. 1986. Potassium availability in relation to soil moisture. II Calculations by means of a mathematical simulation model. Plant and Soil. 95:233-243. MacKayz A. D. and S. A. Barber. 1983. Interaction of soil temperature and phosphate status on root growth and phosphorus uptake during early c o m growth. Plant and Soil. 86:321-331. MacKayz D. C. and S. A. Barber. 1987. Effect of cyclic wetting and drying of a soil on root hair growth of maize roots. Plant and Soil. 104:291-293. Markartz A. H., E. L. Ficus, A. W. Naylor, and P. J. Kramer. 1979. Effect of temperature on water and ion transport in soybean and broccoli systems. Plant Fhysiol. 64:83-87. Masseez T. W. 1973. Soil characterization by diffusion measurements. Ph.D. Thesis. Montana State University. Bozeman, MT. Mcleanz 0. E. and M. E. Watson. 1985. Soil Measurements of Plant available potassium. In R. D. Munson (ed.), Potassium in agriculture, pp. 277-324. ASA, CSSAz SSSAz Madison, W I . Mcleanz E. 0. and R. H. Simon. 1958. Potassium status of some Ohio soils as revealed by greenhouse and laboratory studies. Soil Sci. 85:324-332. Mengelz K. and L. C. von Braunschweig. 1972. The effect of soil moisture upon the availability of potassium and its influence on the growth of young maize plants. Soil Sci. 114:142-148. Moss, P. 1963. Some aspects of the cation status of soil moisture. I. The ratio law and soil moisture content. Plant and Soil. 18:99-113. 77 Mubarak, A., and R. A. Olsen. 1976. soil solution by centrifugation. 40:329-331. Immiscible displacement of the Soil Sci. Soc. Am. J. Nemeth, K. 1975. The effect of K fertilization and K removal by ryegrass in pot experiments on the K concentration of the soil solution of various soils. Plant and Soil. 42:97-107. Nielsen, K. F., R. L. Halstead, and A. F. Maclean. 1960. The influence of soil temperature on the growth and mineral composition of oats. Can. J. Soil Sci. 40:255-263. Nielsen, K. F., R. L. Halstead, A. F. Mclean, S. J. Bourget, and R. M. Holmes. 1961. The influence of soil temperature on the growth and mineral composition of c o m , bromegrass and potatoes. Soil Sci. Soc. Am. Proc. 25:369-371. Nobel, P. S. 1974. Introduction to biophysical plant physiology. Freeman, San Francisco. Nye, P. H. and P. B. Tinker. 1977. Solute transport in the soil near root surfaces. In Solute movement in the soil-root system, pp 127-187. Univ. California press. Nye, P. H . , Brewster, J. P., Ehat, K. K. S. 1975. The possibility of predicting solute uptake and plant growth response from independently measured soil and plant characteristics. I The therodical basis of the experiments. Plant and Soil. 42:161-170. Olsen, R. A. 1968. The driving force on an ion in the absorption process. Soil Sci. Soc. Am. Proc. 32:660-664. Olsen, S.R., F.S. Watanabe, and L. A. Dean. 1954. Estimation of available phosphorus in soils by extraction with sodium bicarbonate. Circular No. 939. U.S. Govt. Printing Ofc. Washington, D. C. Peters, D. B. 1965. Water availability. In, C. A. Black (ed.). Methods of soil analysis, part I. Agron. Mongr. 9. Am. Soc. of Agron., Madison, W I . Chpt. 19. Rasnake, M. and G. W. Thomas. 1976. Potassium status of some Alluvial soils in Kentucky. Soil Sci. Soc. Am. Proc. 40:883-886. Reitemeier, R. F., R. S. Holmes, I. C. Brown, L. W. Klipp, and R. Q. 'Parks. 1947. Release of nonexchangeable potassium by greenhouse Neubauer and laboratory methods. Soil Sci. Soc. Proc. 14:101-104. Rich, C. I. 1964. Effects of cation size and pH on potassium exchange in Nason soil. Soil Sci. 98:100-106. 78 Richards, J. E., Bates, T. E. and Sheppard, S. C. 1988. Studies on the potassium-supplying capacities of southern Ontario soils. I Field and greenhouse experiments. Can J. Soil Sci. 68:183-197. Rovira, A. D., Bowen, G. D. 1973. The influence of root temperature on 14C assimilate profiles in wheat roots. Planta. 114:101-107. Rowell, D. L., Martin, M. W., Nye, P. H. 1967. The measurement and mechanism of ion diffusion in soils. III. The effect of moisture content and soil solution concentration on the self-diffusion of ions in soils. J. of Soil Sci. 18:205-222. SAS Institute. 1987. SAS/STAT Users guide for personal computers. Version 6 ed. SAS Institute Inc., Cary, NC. Schaff, B. E. and E. 0. Skogley. 1982. Diffusion of potassium, calcium and magnesium in bozeman silt loam as influenced by temperature and moisture. Soil Sci. Soc. Am. J. 46:521-524. Scott, A. D. and M. G. Reed. 1960. Determination of the precipitated potassium in sodium tetraphenylboron-micaceous mineral systems. Soil Sci. Soc. Am. Proc. 24:326-327. Sims, J. R. and G.D. Jackson. 1971. Rapid analysis of soil nitrate with chromotropic acid. Soil Sci. Soc. Am. Proc. 35:603-606. Sims, J. R. and V. A. Haby. 1970. Simplified colormetric determination of organic matter. Soil Sci. 112:137-141. Skogley, E. 0. and V. A. Haby. 1981. Predicting crop responses on high-potassium soils of frigid temperature and ustic moisture regimes. Soil Sci. Soc. Am. J. 45:533-536. Skogley, E. 0. 1976. Potassium in Montana soils and crop requirements. Montana Agric. Exp. Stn. Res. Report no. 88. Smith, D. 1971. Levels and sources of potassium for alfalfa as influenced by temperature. Agron. J. 63:497-500. Smith, F. W., B. G. Ellis, and J. Grava. 1957. Use of acid fluoride solutions for the extraction of available phosphorus in calcareous soils and to which rock phosphate was added. Soil Sci. Soc. Am. Proc. 21:400-404. Song, S. K. and P. M. Haung. 1983. Dynamics of potassium release from potassium-bearing minerals as influenced by oxalic and citric acids. Agron. Abstr. American Society of Agronomy, Madison, WI, p 222. Sparks, D. L. 1980. Chemistry of soil potassium in Atlantic soils. Commun. is Soil Sci. and Plant Anal. 11:435-449. 79 Sparks, D. L., Liebhardt, W. C. 1982. Temperature effects on potassium exchange and selectivity in Delaware soils. Soil Sci. 133:10-17. Sparks, D. L. and P. M. Haung. 1985. Riysical chemistry of soil. R. D. M m s o n (ed.), Potassium in agriculture, pp. 277-324. ASA, CSSA7 SSSA, Madison, W I . In Sutton, C. D. 1969. Effect of low soil temperature on phosphate nutrition of plants-a review. J. of Sci. of food and Agr. 20:1-3. Talibudeen, 0., Beasley, J. D., Lane, P., Rajendran, N. 1978. Assessment of soil potassium reserves available to plant roots. J. of Soil Sci. 29:207-218. Thornton, S. F. 1935. Soil and fertilizer studies by means of the Neubauer method. Purdue Univ. Agr. Exp. Sta. Bull. 309:1-38. U.S. Salinity Laboratory. 1954. Diagnosis and improvement of saline and alkali soils. Ag. Handbook N. 60. USDA. U.S. Govt. Printing. Ofc. Washington, D. C. Vaidyanathan, L. V . , Nye, P. H. 1966. The measurement and mechanism of ion diffusion in soils. II An exchange resin paper method for measurement of the diffusive flux and diffusion coefficient of nutrient ions in soils. J. of Soil Sci. 17:177-203. Vaidyanathan, M. C., Drew, M. C., Nye, P. H 1968. The measurement and mechanism of ion diffusion in soils. TV. The concentration dependence of diffusion coefficients of potassium in soils at a range of moisture levels and a method for the estimation of the differential diffusion coefficient at any concentration. J. of Soil Sci. 19:94-107. Veeh, R. H. and E. 0 Skogley. 1986. Small grain response to potassium fertilizers as related to soil and site characteristics. Soil Sci. Soc. Am. J. 50:373-378, von Reichenbach, H. G. and C. I. Rich. 1969. muscovite as influenced by particle size. 17:23-29. Potassium release from Clays. Clay Miner. W a m c k e , D. D. and S. A. Barber. 1973. Diffusion of Zn is soil. III. Relation to zinc adsorption isotherms. Soil Sci. Soc. Amer. Proc. 37:355-358. Willis, R. D. 1980. Reduction column for automated determination of nitrate and nitrite in water. Anal. Chem. 52:1376-1377. Yang, J., 1987. Characterization of soil K availability by chemical and thermodynamic parameters. Ph.D. Thesis. Montana State University. Bozeman, MT. 80 APPENDICES APPENDIX A location Description, Ehysical and Chemical Characteristics of Twenty Montana Soils 82 Table 35. Soil series, texture and location description of twenty Montana soils. Soil Series Texture Amsterdam silt loam Gallatin Bear Paw clay Chouteau Beaverton silty clay loam Gallatin Bozeman silty clay loam Gallatin Chanta loam Prairie Cherry clay loam Garfield Chinook sandy loam Broadwater Creston loam Flathead Edgar loam Rosebud Evanston clay loam Chouteau Flathead sandy loam Flathead County Fort Collins loam Rosebud Gerber clay Cascade Kevin clay loam Pondera Manhatten silt loam Gallatin Rothiemay clay loam Teton Round Butte silt loam Lake Stryker silty clay loam Flathead Tanna loam Stillwater Vamey sandy clay loam Madison location Description NE 1/4, Sec 7, T2N, R5E SW 1/4, Sec 12, T28N, R15W SE 1/4, NE 1/4, Sec T2S, R 6E SW 1/4, Sec 31, T2S, R5E SW 1/4, Sec 20 r T21N, R52E NW 1/4, Sec 4, T19N, R34E SW 1/4, SW 1/4, Sec T5N, R2E SE 1/4, Sec 33, T29N, R20W NE 1/4, Sec 20, T6N, R24E SW 1/4, Sec 34, T29N, R5E SW 1/4, SW 1/4, Sec T29N, R22W SE 1/4, Sec 15, T6N, R41E NW 1/4, Sec 10, T22N, R8E NW 1/4, Sec 8 , T29N, R2W SW 1/4, Sec 7, TIN, R3E SE 1/4, Sec 26, T22N, R2W SE 1/4, NE 1/4, Sec T22N, R21W NW 1/4, SW 1/4, Sec T30N, R22W NW 1/4, Sec 21, T3N, R20E NE 1/4, Sec 9, T3S, RlW 22, 35 15, 32, 17, 83 Table 36. Soil series name and classification and family of twenty Montana soils. Soil Series Soil Taxonomic Classification Amsterdam Typic Cryoborolls Fine-silty Bear Baw Typic Argiborolls Fine, Montmorillinitic Beaverton Typic Argiborolls Loamy-skeletal, mixed Bozeman Badhic Argiborolls Fine silty Chanta Aridic Haploborolls Fine-loamy over sand, mixed Cherry Typic Ustpchrepts Fine-silty, mixed, frigid Chinook Aridic Haploborolls Coarse, loamy, mixed Creston Udic Baploborolls Coarse-silty, mixed Edgar Ustollic Cairiborthids Fine-loamy Evanston Aridic Argiborolls Fine-loamy, mixed Flathead Bachic Udic Haploborolls Coarse-loamy, mixed Fort Collins Ustollic Haplargids Fine-loamy, mixed, mesic Gerber Udorthentic Chramusterts Fine, Montmorillinitic, frigid Kevin Aridic Argiborolls Fine, Montmorillinitic Manhatten Typic Calciborolls Coarse loamy Rothiemay Aridic Calciborolls Fine-loamy, mixed Round Butte Xerollic Natragids Fine, mixed frigid Stryker Aquic Eutrdboralfs Fine-silty, mixed Tanna Aridic Argiborolls Fine-montmorillinitic Vamey Aridic Argiborolls Fine-loamy, mixed 84 Table 37. Exchangeable cation concentrations for twenty Montana soils as determined by 1.0 N ammonium acetate. Soil Series K Ca Mg Na cmol Kg 1 ---Amsterdam Bear Paw Beaverton Bozeman Chanta 1.426 1.919 0.825 1.742 0.955 4.812 2.319 2.062 2.129 0.757 2.756 3.919 4.090 3.763 2.890 0.039 0.061 0.076 0.052 0.067 Cherry Chinook Creston Edgar Evanston 1.364 1.398 0.619 0.887 1.323 1.337 4.268 1.434 0.859 0.925 3.412 2.178 1.372 3.317 3.702 0.049 0.062 0.056 0.028 0.061 Flathead Fort Collins Gerber Kevin Marihatten 0.877 0.988 2.049 2.115 1.890 1.722 0.431 2.380 1.208 4.707 2.140 2.640 6.730 3.407 2.707 0.128 0.038 0.061 0.049 0.056 Rothiemay Round Butte Stryker Tanna Vamey 1.483 0.778 0.822 0.942 1.110 3.125 1.096 2.948 3.547 4.448 3.057 1.137 6.603 1.913 2.214 0.081 0.103 0.708 0.038 0.029 1.276 0.014 1.14 0.021 2.326 0.031 1.35 0.044 3.197 0.036 1.14 0.052 0.194 0.094 4.78 0.006 N=4 Mean STD DEV CV % ISD (0.05) 85 Table 38. Soil pH, organic matter, calcium carbonate and electrical conductivity of twenty Montana soils. Soil Series PH (1:1) CM CACO3 EC % % S nfl Amsterdam Bear Baw Beaverton Bozeman Chanta 8.24 7.93 5.62 7.34 6.62 2.37 3.22 5.37 3.19 1.78 5.47 0.72 0.49 0.66 0.17 15.5 13.4 8.5 13.8 8.4 Cherry Chinook Creston Edgar Evanston 7.35 8.34 5.69 6.81 6.35 3.13 1.97 5.41 1.40 2.05 0.44 11.33 0.13 0.14 0.34 10.0 14.1 35.6 6.2 11.5 Flathead Fort Collins Gerber Kevin Manhatten 7.18 6.13 6.78 6.92 8.15 4.80 1.27 2.62 3.13 2.40 0.25 0.26 0.72 0.42 4.04 16.5 8.0 26.6 12.4 14.0 7.54 6.05 7.88 8.24 8.13 4.26 2.39 3.58 1.92 1.68 1.60 0.28 2.06 3.70 5.03 35.0 11.8 64.5 12.4 15.9 2.89 0.11 2.95 0.12 1.92 0.17 9.21 0.25 Rothiemay Round Butte Stryker Tarma Vamey N=4 Mean STD DEV CV % ISD (0.05) 7.17 0.087 1.28 0.13 17.7 4.20 23.73 5.95 - 86 Table 39. Soil particle size analysis as determined by pipette (Gee and Bauder, 1986) on twenty Montana soils. Soil Series < 50 Particle Size Fraction (um) 50 - 20 20 - 5 5-2 ■ gm Jxy > 2 —— — Amsterdam Bear Baw Beaverton Bozeman Chanta 161 202 140 171 356 338 99 244 277 290 194 175 225 195 154 80 162 58 69 37 224 359 328 286 162 Cherry Chinook Creston Edgar Evanston 216 700 493 527 360 212 90 194 170 169 218 58 161 91 147 119 28 46 27 86 233 122 104 182 234 Flathead Fort Collins Gerber Kevin Marihatten 565 552 182 306 210 121 178 130 133 415 163 104 120 169 151 54 27 98. 108 74 95 136 467 282 146 Rothiemay Round Butte Stryker Tanna Vamey 338 113 106 476 618 179 252 141 131 61 140 389 459 128 66 87 125 156 71 26 253 119 136 193 226 34.0 0.75 6.5 3.8 19.8 1.07 5.4 2.2 17.6 1.01 4.8 1.9 N=3 Mean STD DEV CV % ISD (0.05) 7.7 0.40 5.2 0.9 21.5 1.01 4.7 1.8 87 Table 40. Saturated paste pH, cation concentration and potassium activity ratio (AR) of twenty Montana soils. Soil Series PH Amsterdam Bear Paw Beaverton Bozeman Chanta 8.41 8.17 6.12 8.70 7.32 10.6 29.5 4.1 18.8 19.5 Cherry Chinook Creston Edgar Evanston 8.26 8.59 5.42 7.71 6.52 Flathead Fort Collins Gerber Kevin Marihatten Rothiemay Round Butte Stryker Tanna Vamey N=4 Mean STD DEV CV % ISD (0.05) K Ca ----mg L--*- Mg AR 48.5 64.8 23.9 57.5 23.6 4.3 11.0 6.0 8.4 11.1 0.104 0.200 0.039 0.143 0.142 18.8 39.4 35.2 14.0 23.9 21.7 47.2 284.6 29.4 38.5 6.9 7.1 33.0 10.7 12.9 0.170 0.326 0.131 0.102 0.158 8.69 6.91 8.19 7.73 8.29 28.4 24.5 21.3 39.9 22.5 149.9 18.0 56.1 38.7 56.3 24.8 10.3 14.3 10.3 4.6 0.128 0.187 0.131 0.291 0.210 7.56 6.63 8.45 8.28 8.94 24.8 19.7 23.8 13.6 11.3 207.8 32.9 306.2 65.1 61.4 21.0 4.6 49.5 6.9 3.3 0.113 0.201 0.075 0.109 0.114 7.74 0.116 1.50 0.16 22.2 1.48 6.68 2.10 13.7 1.25 9.53 1.77 0.174 0.058 4.41 0.012 81.6 5.58 6.83 7.89 88 sI I Table 41. Exchangeable and nonexchangeable soil potassium as determined by magnesium chloride and sodium tetraphenyl boron, and nitric acid respectively. HNO3 Amsterdam Bear Paw Beaverton Bozeman Chanta 4.154 7.002 2.219 6.232 3.684 3.68 3.90 2.34 4.88 3.10 14.40 15.12 13.81 16.68 14.70 Cherry Chinook Creston Edgar Evanston 5.510 6.409 2.559 3.704 4.215 3.29 1.94 0.95 3.30 5.69 16.78 13.81 12.56 15.09 15.95 Flathead Fort Collins Gerber Kevin Marihatten 4.072 4.488 6.838 8.855 6.389 0.55 3.15 4.15 4.88 3.38 14.61 14.83 17.23 15.62 16.48 Rothiemay Round Butte Stryker Tanna Vamey 3.894 6.041 2.805 3.411 3.486 2.05 2.30 0.88 3.11 1.37 15.09 12.74 12.74 14.85 13.81 4.798 0.307 6.42 0.255 2.95 0.15 5.10 0.32 14.02 1.57 11.2 0.62 M 3Cl2 I Soil Series N=4 Mean STD DEV CV % ISD (0.05) 89 Table 42. Soil extractable phosphorus as determined by Bray and Olson extraction methodologies for twenty Montana soils. Soil Series BRAY P OISON P ---------mg Kg-1 — -------- Amsterdam Bear Paw Beaverton Bozeman Chanta 37.0 136.3 118.7 80.5 43.8 10.6 53.4 40.5 31.2 17.4 Cherry Chinook Creston Edgar Evanston 41.1 28.2 113.1 37.5 54.6 20.7 30.0 40.2 10.9 29.9 Flathead Fort Collins Gerber Kevin Manhatten 117.5 61.5 72.5 92.6 96.6 21.4 25.2 33.9 43.5 11.7 Rothiemay Round Butte Stryker Tanna Vamey 101.8 78.8 70.7 48.7 96.3 41.5 31.1 28.6 14.7 31.0 N=4 Mean STD DEV 1 CV % ISD (0.05) 76.4 2.84 3.72 18.3 28.4 0.37 1.31 1.44 90 Table 43. Soil solution cation concentration on a solution basis and potassium activity ratio (AR) of twenty Montana soils as determined by immiscible liquid displacement (Mubarak and Olson, 1976). Soil Series K Ca Mg AR ---- m m o l ----Amsterdam Bear Paw Beaverton Bozeman Chanta 0.57 0.96 0.24 1.16 0.81 11.06 8.16 26.94 20.14 3.13 1.56 2.10 17.25 0.88 2.33 0.160 0.300 0.037 0.254 0.349 Cherry Chinook Creston Edgar Evanston 1.08 1.85 1.31 0.73 0.51 16.16 10.26 35.77 ' 5.55 8.83 4.15 2.40 10.12 3.69 1.33 0.239 0.520 0.194 0.242 0.160 Flathead Fort Collins Gerber Kevin Manhatten 1.31 1.17 0.95 1.88 0.84 16.87 4.05 10.24 8.28 9.02 5.28 3.76 4.22 4.08 1.26 0.278 0.421 0.251 0.536 0.263 Rothiemay Round Butte Stryker Tarma Vamey 1.17 0.57 0.86 , 2.28 0.61 40.42 5.18 30.95 5.20 12.82 9.09 1.30 23.97 2.09 1.45 0.167 0.224 0.116 0.844 0.162 1.05 0.023 2.23 0.21 16.91 6.30 3.73 3.89 N =4 Mean STD DEV CV % LSD (0.05) 4.97 0.181 3.63 1.14 0.285 0.008 3.41 0.051 91 Table 44. Soil solution cation concentration and potassium activity ratio (AR) of twenty Montana soils at 0.33 Bpa on a soil mass basis as determined by immiscible liquid displacement (Mjbarak and Olsenz 1976). Soil Series K Ca Mg AR enrol Kg-1 — .Amsterdam Bear Paw _ Beaverton Bozeman Chanta 0.0140 0.0258 0.0066 0.0304 0.0149 0.272 0.218 0.729 0.526 0.057 0.0384 0.0565 0.4675 0.0230 0.0426 0.025 0.049 0.006 0.041 0.047 Cherry Chinook Creston Edgar Evanston 0.0267 0.0277 0.0277 0.0119 0.0106 0.400 0.153 0.753 0.089 0.183 0.1031 0.0360 0.2134 0.0598 0.0278 0.037 0.063 0.028 0.030 0.023 Flathead Fort Collins Gerber Kevin Marihatten 0.0298 0.0172 0.0261 0.0422 0.0197 0.384 0.059 0.280 0.185 0.210 0.1205 0.0552 0.1156 0.0915 0.0295 0.042 0.051 0.041 0.080 0.040 Rothiemay Round Butte Stryker Tanna Vamey 0.0273 0.0179 0.0287 0.0413 0.0093 0.938 0.162 1.032 0.094 0.194 0.2114 0.0410 0.8007 0.0379 0.0221 0.025 0.039 0.021 0.113 0.020 0.0228 0.0005 2.19 0.004 0.346 0.012 3.54 0.022 0.129 0.004 3.29 0.029 0.038 0.0012 1.94 0.0071 N=4 Mean STD DEV CV % ISD (0.05) 92 Table 45. Moisture contents of twenty Montana soils at - 0.15 - 0.33,- 0.66 and - 1.00 Kpa water potential. Soil Series -0.15 Water Potential (KPa) —0.33 —0. 66 -1.00 'O Amsterdam Bear Paw Beaverton Bozeman Chanta 32.33 29.36 34.69 33.83 23.16 24.61 26.78 27.07 26.11 18.26 21.01 22.89 24.03 22.85 14.20 18.48 20.69 18.64 19.29 12.71 Cherry Chinook Creston Edgar Evanston 29.43 20.80 29.43 20.20 23.13 24.78 14.97 21.06 16.17 20.83 22.25 13.40 18.06 17.32 15.54 18.40 11.98 15.29 12.45 15.60 Flathead Fort Collins Gerber Kevin Marihatten 30.70 17.83 29.93 25.26 34.03 22.79 14.65 27.38 22.37 23.34 19.13 11.68 24.95 20.01 19.74 16.26 10.39 23.76 16.03 18.96 Rothiemay Round Butte Stryker Tanna Varriey 25.76 37.43 37.20 21.26 19.40 23.21 31.40 33.36 18.11 15.18 20.72 26.51 28.96 15.33 14.31 19.26 22.41 24.89 15.17 12.84 27.76 1.03 3.72 1.71 22.62 0.58 2.58 0.97 19.25 0.59 3.05 3.54 17.48 0.76 4.32 2.96 N=3 Mean std dev CV % ISD (0.05) 93 Table 46. Moisture contents (gm Kg 1J of twenty Montana soils at - 2.00, - 5.00, and - 15.00 Kpa water potential. Soil Series - 2.00 Water Potential (I^a) - 5.00 - 15.00 % Amsterdam Bear Paw Beaverton Bozeman Chanta 17.38 21.00 21.63 19.61 10.68 14.32 17.55 17.19 15.92 8.18 12.76 15.88 16.62 15.36 8.06 Cherry Chinook Creston Edgar Evanston 16.71 11.17 14.06 11.29 15.45 12.54 9.62 10.31 9.55 11.54 11.16 8.81 9.60 8.37 10.88 Flathead Fort Collins Gerber Kevin Manhatten 15.17 9.21 24.36 17.57 15.56 11.13 7.83 20.65 13.57 12.66 10.92 6.59 18.91 11.89 12.15 19.31 15.65 . 21.31 13.11 11.63 15.43 10.21 14.18 10.65 11.09 14.47 8.68 11.34 9.44 11.04 16.09 0.36 2.22 0.59 12.70 0.57 4.45 0.93 11.60 0.72 6.19 1.19 Rothiemay Round Butte Stryker Tanna Vamey N=4 Mean STD DEV CV % ISD (0.05) 94 Table 47. Soil, silt, and clay and for twenty Montana soils as determined by hydrometer (Gee and Bander, 1986). Soil Series Sand Silt Clay gm Kg 1 Amsterdam Bear Paw Beaverton Bozeman Chanta 182 246 196 199 342 536 303 463 503 473 281 451 341 297 184 Cherry Chinook Creston Edgar Evanston 232 572 446 409 352 453 256 420 350 356 314 170 134 240 290 Flathead Fort Collins Gerber Kevin Marihatten 576 506 206 269 202 313 333 316 386 576 HO 160 477 344 220 Rothiemay Round Butte Stryker Tanna Vamey 342 139 149 439 596 353 640 570 340 203 304 220 280 220 200 330 19 5.7 31 407 21 5.2 35 262 17 6.4 N=3 Mean STD DEV CV % ISD (0.05) 28 95 Table 48. Available water at four water potential ranges for twenty Montana soils. Soil Series 0.33-15.00 0.33-5.00 0.33-2.00 0.33-1.00 Amsterdam Bear Paw Beaverton Bozeman Chanta 11.85 10.90 10.45 10.74 10.20 10.29 9.23 9.88 10.18 10.08 7.23 5.77 5.43 6.50 7.58 6.13 5.09 4.09 6.82 5.55 Cherry Chinook Creston Edgar Evanston 13.61 6.15 11.46 7.80 9.95 12.23 5.35 10.75 6.61 9.28 8.06 3.80 7.00 4.88 5.38 6.37 2.99 5.77 3.71 5.23 Flathead Fort Collins Gerber Kevin Manhatten 11.86 8.06 8.47 10.48 11.18 11.66 6.82 6.73 8.80 10.67 7.61 5.44 3.02 4.80 7.77 6.53 4.26 3.62 4.34 6.14 Rothiemay Round Butte Stryker Tanna Vamey 8.74 22.72 22.02 8.67 4.14 7.78 21.19 19.18 7.45 4.09 3.90 15.74 12.05 5.00 3.55 3.95 8.32 8.47 2.93 2.33 10.97 4.42 7.48 1.35 9.91 4.09 7.98 1.31 6.52 2.99 10.05 1.08 5.131 1.881 20.03 1.69 N=3 Mean STD DEV CV % ISD (0.05) \ 96 Table 49. Cation exchange capacity and mineralogy percentages of the clay fraction of twenty Montana soils, provided by Bernard Schaff, method of Jackson (1958). Soil Series Illite Vermiculite gm Kg 1 Montmorillinite — CEC cmol Kg 1 Amsterdam Bear Paw Beaverton Bozeman Chanta 9.7 14.7 6.1 8.5 5.8 5.0 10.4 5.2 7.2 4.7 12.8 11.7 12.7 12.1 4.8 22.5 22.6 25.1 23.9 9.3 Cherry Chinook Creston Edgar Evanston 11.1 3.4 3.5 7.1 9.2 7.7 1.8 2.5 4.2 7.5 5.6 6.9 2.9 10.9 7.6 28.8 10.4 13.5 10.9 13.9 Flathead Fort Collins Gerber Kevin Manhattan 2.5 5.0 15.0 12.3 5.7 1.3 2.6 7.7 5.7 4.2 3.2 6.5 20.1 10.4 9.4 14.1 7.6 30.7 13.7 20.1 Rothiemay Round Butte Stryker Tanna Vamey 9.9 4.6 9.7 7.1 4.3 5.0 5.2 5.8 4.6 2.6 13.1 3.2 5.6 6.9 12.2 20.9 10.6 14.4 13.9 16.8 97 Table 50. Soil extractable NO3 , Cl, and S for twenty Montana Soil Series NO3 Cl — S mg Kg- 1 ---- Amsterdam Bear Baw Beaverton Bozeman Chanta 3.7 19.9 8.7 9.8 8.1 49.7 47.3 33.6 41.0 26.9 8.0 12.3 7.5 6.5 2.5 Cherry Chinook Creston Edgar Evanston 5.5 6.1 72.6 8.8 24.1 35.4 25.1 33.5 32.6 41.0 3.5 13.3 8.2 2.5 4.5 Flathead Fort Collins Gerber Kevin Manhatten 49.9 9.9 22.4 15.7 6.4 28.7 27.2 23.2 48.8 19.8 8.0 2.7 4.2 5.7 9.2 Rothiemay Round Butte Stryker Tanna Vamey 103.7 7.1 220.3 5.7 7.8 122.3 31.5 87.0 33.9 34.0 15.0 5.5 34.2 6.0 8.5 19.4 0.5 2.3 2.6 41.7 2.0 4.8 4.5 7.5 0.5 6.5 1.9 N=4 Mean STD DEV CV % ISD (0.05) soils. 98 APPENDIX B Correlation Matrix for Properties of Twenty Montana Soils 99 Table 51. Correlation matrix for 28 soil properties on 20 Montana soils. CORR K Uptake ^ex caO x MGex N^ex PAR CM CaOO3 pH EC NO3 Cl Bray-P Olsen-P Sulfur ^sol caSOl Gs o I m -33 Ypa H2O Sand % i Silt % Clay % CEC Illite % Verm % Mont % d[Kex]/dD K Uptake %ex caC X 1.0000 0.6223 0.6223 -0.1914 0.5862 0.1315 0.3502 -0.0987 -0.4252 0.0340 0.1650 0.1118 0.4187 0.1471 0.3378 0.1125 0.2468 0.0552 0.0577 0.4077 -0.5651 0.0979 0.7677 0.3084 0.8480 0.7689 0.4917 0.7455 1.0000 -0.1914 0.2595 0.2595 0.3660 -0.2558 0.4684 -0.1427 0.1235 0.3863 -0.1295 -0.2965 0.0558 0.1053 0.2562 -0.0802 0.3155 -0.1826 -0.3337 0.1364 -0.3300 -0.1176 0.6593 0.5225 0.6864 0.5365 0.6341 0.5764 -0.0270 0.0701 -0.4744 -0.0697 0.7948 0.7964 0.2173 0.0217 0.1477 -0.0088 -0.1791 0.4018 0.0464 0.1034 0.0116 -0.0099 0.0240 -0.0431 0.0141 0.2814 -0.0315 -0.1652 0.3334 -0.1411 1.0000 m Gg x 0.5862 0.3660 -0.0270 Naex 1.0000 0.1315 -0.2558 0.0701 0.5315 0.5315 -0.2948 0.0838 -0.2067 0.0445 0.4591 0.3635 0.2707 0.0026 0.1964 0.4098 0.0360 0.3532 0.5522 0.4886 -0.4919 0.0510 ' 0.7096 0.4971 0.6947 0.5685 0.5432 0.3413 -0.5260 0.2141 -0.0243 0.1441 0.8136 0.8272 0.4609 0.0275 0.0405 0.8752 0.1514 0.5825 0.8505 0.5390 -0.2881 0.3556 0.0218 -0.1014 0.0867 0.0576 -0.2290 0.0588 1.0000 100 Table 51— Continued. CORR K Uptake 1Nex caOsX M9ex Naex PAR CM CaCO3 pH EC NO3 Cl Bray-P Olsen-P Sulfur Ksol caSol Mgsoi -33 Epa H2O Sand % Silt % Clay % CEC Illite % Verm % Mont % d[Kex]/dD PAR 0.3502 0.4684 -0.4744 -0.2948 -0.5260 1.0000 -0.2006 -0.3116 -0.2491 -0.5121 -0.4729 -0.1920 0.0544 0.2281 -0.5573 0.2346 -0.4770 -0.6329 -0.1050 -0.0657 -0.0036 0.1074 -0.1208 0.1911 0.2337 -0.0810 0.4028 CM -0.0987 -0.1427 -0.0697 0.0838 0.2141 -0.2006 1.0000 -0.3056 -0.3025 0.4095 0.4867 0.3185 0.6827 0.4991 0.2991 0.2139 0.7764 0.5327 0.4806 -0.1680 0.1490 0.0834 0.3480 -0.0157 0.0010 -0,0743 0.1459 CaCO3 -0.4252 0.1235 0.7948 -0.2067 -0.0243 -0.3116 -0.3056 1.0000 0.6847 0.0157 -0.1292 —0.0661 -0.3549 -0.2093 0.2717 0.0367 -0.1621 -0.1409 -Oi3319 0.3141 -0.2222 -0.2242 -0.0936 -0.2536 -0.3718 0.0713 -0.4541 pH . 0.0340 0.3863 0.7964 0.0445 0.1441 -0.2491 -0.3025 0.6847 1.0000 0.1459 -0.0038 0.2114 -0.1911 -0.2650 0.4070 0.3070 -0.0885 -0.1180 -0.0853 0.0863 -0.1625 0.0608 0.1664 0.1733 0.0288 0.1944 0.1725 EC 0.1650 -0.1295 0.2173 0.4591 0.8136 -0.5121 0.4095 0.0157 0.1459 1.0000 0.9536 0.6292 0.1905 0.2346 0.8485 0.2840 0.7731 0.7589 0.4695 -0.1859 0.1857 0.0671 0.0674 0.1561 0.0039 -0.0383 0.1406 101 Table 51— Continued. CORR K Uptake kGX caGX MSex N^ex PAR CM CaOO3 PH EC NO3 Cl Bray-P Olsen-P Sulfur Hsol ^aSol Mg50I -33 Kpa H2O Sand % Silt % Clay % CEC Illite % Verm % Mont % d[K^]/dD NO3 0.1118 -0.2965 0.0217 0.3635 0.8272 -0.4729 0.4867 -0.1292 -0.0038 0.9536 1.0000 0.6435 0.2324 0.2384 0.8199 0.2420 0.8001 0.8034 0.4090 -0.1079 0.1932 -0.0642 -0.0894 0.0414 -0.0602 -0.2268 0.1406 Cl 0.4187 0.0558 0.1477 0.2707 0.4609 -0.1920 0.3185 -0.0661 0.2114 0.6292 0.6435 1.0000 0.1728 0.3285 0.6278 0.2047 0.6876 0.4886 0.3292 -0.2188 0.0844 0.2409 0.1197 0.3406 .0.1981 0.1513 0.4056 Bray-P 0.1471 0.1053 -0.0088 • 0.0026 0.0275 0.0544 0.6827 -0.3549 -0.1911 0.1905 0.2324 0.1728 1.0000 0.6607 0.1820 0.0867 0.4069 0.2262 0.3440 -0.0639 -0.0643 0.1782 0.2401 0.0340 ' 0.0882 0.1092 0.2662 Olsen-P 0.3378 0.2562 -0.1791 0.1964 0.0405 0.2281 0.4991 -0.2093 -0.2650 0.2346 0.2384 0.3285 0.6607 1.0000 0.2221 0.1968 0.3796 0.2320 0.2938 -0.0993 -0.2510 0.4593 0.1994 0.3367 0.3503 0.2155 0.1535 Sulfur 0.1125 -0.0802 0.4018 0.4098 0.8752 -0.5573 0.2991 0.2717 0.4070 0.8485 0.8199 0.6278 0.1820 0.2221 1.0000 0.2197 0.6683 0.7703 0.4499 -0.1808 0.1798 0.0660 -0.0035 0.1013 0.0242 -0.0780 0.0797 102 Table 51— Continued. CORR K Uptake ^ex caB X ^ex Naex PAR CM CaCO3 PH EC NO3 Cl Bray-P Olsen-P Sulfur Kgol caSOl M g 5Oi -33 Kpa H2O Sand % Silt % Clay % CEC Illite % Verm % Mont % dEIW/dD Ksol 0.2468 O .3155 0.0464 0.0360 0.1514 0.2346 0.2139 0.0367 0.3070 0.2840 0.2420 0.2047 0.0867 0.1968 0.2197 1.0000 0.1381 0.0344 0.1331 0.0092 -0.0813 0.0841 0.0276 0.2540 0.0790 -0.1269 0.3854 c a Sol M g 5Oi -33 Kpa SAND % 0.0552 -0.1826 0.1034 0.3532 0.5825 -0.4770 0.7764 -0.1621 -0.0885 0.7731 0.8001 0.6876 0.4069 0.3796 0.6683 0.1381 1.0000 0.7939 0.5397 -0.2850 0.2636 0.1283 0.3290 0.0902 0.0589 0.0296 0.1719 0.0577 -0.3337 0.0116 0.5522 0.8505 -0.6329 0.5327 -0.1409 -0.1180 0.7589 0.8034 0.4886 0.2262 0.2320 0.7703 0.0344 0.7939 1.0000 0.5404 -0.3167 0.3006 0.1332 0.0969 0.0838 0.0396 -0.0679 -0.0320 0.4077 0.1364 -0.0099 0.4886 0.5390 -0.1050 0.4806 -0.3319 -0.0853 0.4695 0.4090 0.3292 0.3440 0.2938 0.4499 0.1331 0.5397 0.5404 1.0000 -0.8309 0.6524 0.5141 0.4888 0.4399 0.5533 0.1024 0.3781 -0.5651 -0.3300 0.0240 -0.4919 -0.2881 -0.0657 -0.1680 0.3141 , 0.0863 -0.1859 -0.1079 -0.2188 -0.0639 -0.0993 -0.1808 0.0092 -0.2850 -0.3167 -0.8309 1.0000 -0.7702 —0.6367 -0.5213 -0.5821 -0.7003 -0.2791 -0.4161 . 103 Table 51— Continued. CORR K Uptake %ex Caex M-fex Naex PAR CM CaCO3 PH EC NO3 Cl Bray-P Olsen-P Sulfur ^sol c a SOl M=Tsol -33 Kpa H2O Sarxi % Silt % Clay % CEC Illite % Verm % Mont % dCNexl/dD Silt % Clay % 0.0979 -0.1176 -0.0431 0.0510 0.3556 -0.0036 0.1490 -0.2222 . -0.1625 0.1857 0.1932 0.0844 -0.0643 -0.2510 0.1798 -0.0813 0.2636 0.3006 0.6524 -0.7702 1.0000 -0.0015 0.0805 -0.0100 0.1900 -0.2477 0.1301 0.7677 0.6593 0.0141 0.7096 0.0218 0.1074 0.0834 -0.2242 0.0608 0.0671 -0.0642 0.2409 0.1782 0.4593 0.0660 0.0841 0.1283 0.1332 0.5141 -0.6367 -0.0015 1.0000 0.7200 0.9247 0.8682 0.7369 0.4951 CEC Illite 0.3084 0.5225 0.2814 0.4971 -0.1014 -0.1208 0.3480 -0,0936 0.1664 0.0674 -0.0894 0.1197 0.2401 0.1994 -0.0035 0.0276 0.3290 0.0969 0.4888 -0.5213 0.0805 0.7200 1.0000 0.6169 0.5908 0.6550 0.4288, 0.8480 0.6864 -0.0315 0.6947 0.0867 0.1911 -0.0157 -0.2536 0.1733 0.1561 0.0414 0.3406 0.0340 0.3367 0.1013 0.2540 0.0902 0.0838 0.4399 -0.5821 -0.0100 0.9247 0.6169 1.0000 0.8680 0.6123 0.6693 VERM 0.7689 0.5365 -0.1652 0.5685 0.0576 0.2337 0.0010 -0.3718 0.0288 0.0039 -0.0602 0.1981 0.0882 0.3503 0.0242 0.0790 0.0589 0.0396 0.5533 -0.7003 0.1900 0.8682 0.5908 0.8680 1.0000 0.4140 0.5854 104 Table 51— Continued. CORR K Uptake ^ex caGX m Zb x Naex PAR CM CaCO3 PH EC NO3 Cl Bray-P Olsen-P Sulfur 1Ssol caSol Mgsol -33 Kpa H2O Sand % Silt % Clay % CEC Illite % Vem % Mont % d[Kex]/dD MONT 0.4917 0.6341 0.3334 0.5432 -0.2290 -0.0810 -0.0743 0.0713 0.1944 -0.0383 -0.2268 0.1513 0.1092 0.2155 -0.0780 -0.1269 0.0296 -0.0679 0.1024 -0.2791 -0.2477 0.7369 0.6550 0.6123 0.4140 1.0000 0.2329 d[%ex]/dD 0.7455 0.5764 -0.1411 0.3413 0.0588 0.4028 0.1459 -0.4541 0.1725 0.1406 0.1406 0.4056 0.2662 0.1535 0.0797 0.3854 0.1719 -0.0320 0.3781 -0.4161 0.1301 0.4951 0.4288 0.6693 0.5854 0.2329 1.0000