Soil Test Analysis Methods for British Columbia Agricultural Crops C.G. Kowalenko, Editor

Proceedings of a workshop of the British Columbia Soil and Tissue Testing Council Held at the Langley Conference Centre 24 November, 1992

Distributed by: BC Ministry of Agriculture, Food and Fisheries Resource Management Branch Abbotsford, BC

First Published: 1993

ERRATA Soil Test Analysis Methods for British Columbia Agricultural Crops. e.G. Kowalenko, Editor Proceedings of a workshop for the B.e. Soil and Tissue Testing Council held at Langley, 24th November 1992. Two sentences in Appendix XI - "Selected pages of 1987-1988 and 1988-1989Annual reports to British Columbia Ministry and Food by N.A. Gough", were incorrectly written. The incorrect sentences were: • Page 136 -

"Table 8 indicates that maximum revenue occurred at 0 kg P20s/ha";

• Page 157 -

"Unfortunately, the F value for the 1st cut yield was significant at 7% probability".

The correct sentences are written below: • Page 136 -

"Table 9 indicates that there was no marginal contribution to revenue from phosphate application".

• Page 157 -

"Unfortunately, the F value for the 1st cut yield was not significant at the 5% level".

SOIL TEST ANALYSIS METHODS FOR BRITISH COLUMBIA AGRICULTURAL CROPS C. G. Kowalenko, Editor Proceedings ofa workshop ofthe British Columbia Soil and Tissue Testing Council Held at Langley Conference Centre 24 November 1992

Printed by British Columbia Ministry of Agriculture, Fisheries and Food Victoria

1993

·

ii

CONTENTS ACKNOWLEDGEMENTS ..................................................................................... v PREFACE ............................................................................................................... vi SUMMARY OF RECOMMENDED METHODS ................................................... vii INTRODUCTORY COMMENTS .......................................................................... 1 REVIEW OF BASIC CONCEPTS AND INGREDIENTS FOR A SOIL TEST SYSTEM ..................................................................................................... 3 SA.LvtPLE PREPARATION: MOISTURE CONTENT AND SUB SAMPLING METHOD ................................................................................. 6 MEASUREMENT OF pH AND DETERMINATION OF LIME REQUIREMENT .................................................................................................. 9 SALINITY AND SODICITY MEASUREMENTS ................................................ 16 NITRATE, TOTAL NITROGEN AND ORGANIC MATTER DETERMINATIONS ........................................................................................... 19 PHOSPHORUS ...................................................................................................... 24 POTASSIUM, MAGNESIUM AND CALCIUM .............................................: ..... 28 SULPHUR .............................................................................................................. 33 BORON ..............................................................................~:.................................. 38 ZINC, MANGANESE, COPPER AND IRON ....................................................... 40 APPENDICES I. Soil and Tissue Testing Council Technical Meeting, Nov. 24, 1992 .................... 43 II. REQUIREMENTS FOR A SAMPLE-BASED PLANT NUTRIENT MANAGEMENT SYSTEM FOR BRITISH COLUMBIA ........................... 44 III. Selected pages of proceedings of "Meeting No. 5 of the British Columbia Subcommittee on Soil Testing Procedure held 22 November 1966 at the University of British Columbia .................................................... 48 IV. NATURE OF SOIL PROPERTIES AND THEIR RELATION TO LIME REQUIREMENTS ........ ................ .............................. ................. ...... 52 V. BACKGROUND RESEARCH IN SUPPORT OF THE BRITISH COLUMBIA SOIL TESTING SERVICE ..................................................... 65 VI. INCUBATION LIME REQUIREMENT TRIAL ON SIX B.C. CENTRAL SOILS ........................................................................................ 75 VII. LIMING TRIALS IN BRITISH COLUMBIA'S CENTRAL INTERIOR .................................................................................................. 82 VIII. Liming Trials on Corn Production ................................................................ 98 IX. LIME REQUIREMENT DETERMINATION OF ACID MINERAL AND ORGANIC SOILS USING THE SMP BUFFER-pH METHOD ....... 106 X. THE RELATIONSHIP BETWEEN ELECTRICAL CONDUCTIVITY MEASURED ON A SATURATED PASTE EXTRACT AND ELECTRICAL CONDUCTIVITY MEASURED ON A 2: 1 EXTRACT ..... 120 XI. Reports on phosphorus and potassium soil test/yield correlation trials in interior British Columbia ........................................................................... 128 iii

XII. COMPARISON OF FOUR SULPHATE SULPHUR EXTRACTANTS . FOR PREDICTING AVAILABLE SOIL SULPHUR FOR BARLEY GROWTII IN A POT STUDY ................................................................... 164 XIII. SULPHUR CORRELATION PROJECT (D.A.T.E. Project #3) ................ 172 XIV. NEW SOIL SULFUR INTERPRETATIONS ............................................ 185

iv

ACKNOWLEDGEMENTS The organizing committee for this workshop consisted of R.A. Bertrand, C.G. Kowalenko, T.F. Guthrie and N .A. Gough. British Colwnbia Ministry of Agriculture, Fisheries and Food paid for the use of the conference facility and printed the proceedings. Western Canada Fertilizer, through Executive Secretary D.C. McLean, kindly provided the lunch which greatly facilitated both formal and informal discussions.

v

PREFACE On November 24, 1992 at Langley Conference Centre, the British Columbia Soil and Tissue Testing Council organized a workshop to discuss laboratory methodology (i.e. extraction and analysis of numents) suitable for soil testing for British Collllllbia soil and crop conditions. The format of the meeting was to have speakers initiate discussion by suggesting a recommended method for a specific analysis and briefly outline the data upon which this recommendation was based. After discussion, a final recommendation was made by consensus of those present. The intent of this publication is to provide a written record of the final recommendations, and a more thorough review of data upon which the recommendations were based than was possible at the workshop. The review of background data should be useful for assessing the suitability of laboratory methods recommended, information on which fertilizer recommendations can be formulated and show where laboratory methods research should be directed to enhance confidence in the soil test system.

vi

SUMMARY OF RECOMMENDED METHODS Purpose

Preparation

Reporting pH Lime Electrical conductivity Sodicity Nitrate

Total N Organic matler Phosphorus Potassium, calcium, magnesium Sulphur

Boron

Other micronutrients (Cu, Zn, Mn, Fe)

Recommendation Air dried. Moisture content of air dry sample should be determined to constant weight in oven at 110 C on a separate sample from extraction as oven drying may change element extractability Air dried organic samples may contain significant water. Weighed (preferred) or volume. Ifvolume used, should include a measure of bulk density of scooped sample. Special consideration should be given to organic samples regarding field bulk densities. Oven dry basis (preferred) i.e. corrected for water content at air dry state. Air dry basis acceptable, but should clearly documented. Preparation: 1:2 or 1:1 v/v soil to water or 1:2 v/v soil to 0.01 M CaCI2. Measurement: electrode. Shoemaker, McLean, Pratt single buffer. Saturated paste or saturated paste extract or 1:2 v/v soil to water (preferably with saturated paste equivalent adjustment i.e. multiply by 2). Extraction: IN ammonium acetate (= exchangeable) or Kelowna (= extractable). Measurement: flame emission, atomic absorption or ICAP-AES. Extraction: any salt solution preferably one with high ionic strength (b]lt no nitrate). Measurement: colorimetric (manual or automated) or specific ion electrode that is compatible with extract solution. ." Kjeldhal or with dry ash instrumentation. Loss-an-ignition, dry ash instrumentation or Walkley-Black. Extraction: Kelowna or Bray PI. Measurement: colorimetric (= inorganic P) or ICAP-AES (= total P). Extraction: IN ammonium acetate (= exchangeable cation) or Kelowna (= extractable cation). Measurement: as for sodicity above. Extraction: 0.1 M or 0.01 M CaCI2 (= solution sulphate S) or Kelowna (extractable S). Measurement: barium-based method (= inorganic sulphate S; must be compatible with extracting solution) or HI-reduction (= total sulphate S) or ICAP-AES (= total S). NOTE: choice based on very limited local data. Extraction: hot water. Measurement colorimetric (azomethine or modified curcumin) or ICAP-AES. Extraction: DTPA-TEA Measurement: ICAP-AES or atomic absorption. NOTE: choice based on very limited local data.

GENERAL NOTE: Extraction and measurement method should be clearly identified especially when alternate methods are used. For recommendations, preference was given to methods that have considerable local supporting data.

vii

viii

INTRODUCTORY COMMENTS T. Pringle Assistant Deputy Minister, B.C. Min. of Agric., Fisheries and Food, Victoria Soil testing and plant analysis are important tools in assisting fanners and the agricultural industry to use ferti1izers in a profitable and environmentally sound manner. We need to continue to develop and use new technology so that our industry is economically competitive. We need to ensure that our management practices will provide environmental sustainability. The Ministry will provide support but we expect industry to take the lead role with concerns and adopting procedures so that soil testing and fertilizer recommendations are top quality. The Soil and Tissue Testing Council is an important partnership to provide top quality services. We look forward to working closely with the COlmcil to meet the needs offanners and the industry.

J.M. Crepin Chief Executive Officer, Norwest Labs, Edmonton Have the rules of the game changed? The role of soil and plant analysis laboratories has always been to provide analytical infonnation for the purpose of soil and crop management. The role is the same whether a laboratory is privately owned or a public operation, The client is the same and the need for quality and service is the same, but private laboratories have more to gain by providing quality and service. Twenty five years ago when I started in soil testing, some tests were carried out using a spot plate. The lack of accuracy forced us to interpret the data as very low, low, medium, high and very high, Many thought that these were very crude systems of measuring soil quality and soil fertility, Research, therefore, focused at developing automated instnmaental analysis and testing various methods in relation to crop response to nutrient application levels. This makes soil testing an empirical method of measuring the nutrient supplying power of a soil. Empirical, because we are trying to compare a chemical extraction which occurs in 5 to 30 minutes, to nutrient availability to a crop over a period of 90 to 120 days. Research has shown that while there is a direct and wen defined relationship between a soil testing method and a yield, I feel that too much accuracy is implied by the user of the infonnation. I disagree with those who feel that more research is needed in soil test correlation to improve the correlation. Many soil tests are very accurate when tested in the controlled environment of a greenhouse but not as good when tested against crops in the field. This does not mean that the test is poor, but that many factors such as rainfall, management, weed, and pest control for example play a major part in detennining a crop yield and its nutrient use efficiency. Such factors are difficult to model to improve the soil test recommendation. In a way, soil testing is very similar to other testing methods used in biology, animal nutrition and medicine. However, some will argue that unlike medicine, soil testing laboratories do not use standardized methods. But, would soil science or the public be belter served if soil testing laboratories were using the same methods? The answer is both yes and no, However, it would not make any difference as to the value oftesting as long as the interpretation is related to the method, A soil testing laboratory does not really care what method it uses as long as the method is mgged and that it allows that laboratory to provide sound nutrient or lime application recommendations. After all, the recommendation has been the objective for carrying out a soil test in the past. But, I feel that soil testing as a service has a greater role to play especially in the field of environmental protection. The establishment of guidelines for nutrient management is needed for the purpose of soil and water conservation and there is no belter group than the B,C. Soil and TIssue Testing Council to draft these guidelines. Once established, these guidelines may be legislated and the methods used for testing will therefore become standardized, think that we have enough pertinent information to obtain a consensus on the fonowing:

i

- At what nutrient level is a soil deficient for most crops and what is an excessive level for the purpose of crop production. - What level can be considered an environmental risk and at what depth in relation to rooting depth and gr01Uldwater.

- What, if any, is the maximum level of application offertilizer, manure, sewage, sludge, or compost which should be allowed without soil analysis. We have a fiduciary role to play. A role that no one else is in a better position to assume, especially if we want to maintain a certain level of credibility with the public. We need to accept our responsibilities and to be pro-active in our relationship with the producers.

2

REVIEW OF BASIC CONCEPTS AND INGREDIENTS FOR A SOIL TEST SYSTEM C.G. Kowalenko Research Scientist, Agriculture Canada Research Station, Agassiz

The intent of this workshop was to examine the extraction and analysis component of soil testing in British Columbia and develop a consensus for those methods which are most acceptable for routine use. Although extraction and analysis is an important component of soil testing, it is by no means the only component and a review of the other components is important to set the proper context for this specific topic. At a presentation at a soil fertility meeting on 6 June 1989 (where the formation of the Soil and Tissue Testing Cowlcil was proposed), I outlined the what (activities), who (agencies) and how (assembly) of the soil test system in British Columbia (see Appendix II). The activities of the system includes development of laboratory methods as well as their interpretation, and then their implementation, promotion, utilization and monitoring. Although these components of the system are distinct, they haven't always progressed in a logical fashion (i.e. implementation has often been initisted before extensive background information had been generated). This has occurred because of the need to have something in place despite limited resources for research and development. A number of different agencies have been and continue to be involved to make the system evolve and function. Coordination of all aspects are needed. For the development' of laboratory analysis methods for the soil test system, research data generated in relation to basic principles, correlation and caheration should all be considered. Information on basic principles would include studies that enhance a general understanding of nutrient reactions and interactions in soils, developing instrumentation for sensitive, accurate, interference-free quantification of nutrient elements in extracts, etc. Correlation usually refers to those studies that derive relationships between nutrient extracted by a specific chemical solution and nutrient uptake, whereas calibration includes studies that extend the correlation to a desired agronomic output such as yield and usually to a field scale basis. General aspects of soil testing are discussion in more detail in various publications such as Brown (1987) and Westerman (1990).

It should be remembered that the theoretical relationship of crop growth to soil nutrient extraction upon which soil test recommendations are based is not linear but rather curvilinear (i.e. a sigmoid curve) over the entire range that has to be considered (Figure I). At the most responsive portion of the curve, the relationship could be assumed to be linear. At some point, there is little or no growth response as the amount of nutrient extracted increases, but eventually a toxic effect will reduce growth as the nutrient becomes excessive. Various mathematical relationships have been proposed to represent the curvilinear relationship including quadratic and exponentialllogarithrnic equations. In British Columbia, the Mitscherlich-Bray equation has been used which in its generalized form is: log (A - y) = log A- c1b 1 where A = maximum yield, b I = nutrient soil extraction value when less than adequate, y = yield at b I and OJ = a proportionality constant. Fertilizer recommendations are based on this theoretical relationship and modified by philosophies that assume nutrient replacementlbuild up or sufficiency plus starter/pop-up considerations. The soil test value is usually accomplished by extraction with a specific chemical solution. The soil test value is usually not directly related quantitatively to crop response, but is usually more of an index of availability. Two extracting solutions may be well correlated with crop growth but extract different absolute amounts of the nutrient, as shown in the hypothetical cases in Figures 2a and 2b. In order to allow the use

3

- - .........

100 80

60

:?. e

'-'

,

40

"0

Q;

.~

20 0

'- -- --

--

........... -.......'- '..



Soil nutrient test value Figure 1. Theoretical yield response (dark line) and related fertilizer recommendations using buildup/replacement (short-dashed line) or SUfficiency plus starter/pop-up (long-dashed line) philosophies in relation to soil test values.

of more than one extraction solution for the same nutrient, a correlation between the two extractions for a wide range of soils is often done i.e. this is what was done in British Colwnbia in switching from Bray PI fur phosphorus and CaC12 for, sulphur to the Kelowna multiple element extraction. Various types of correlation may result includingalinear relationship that mayor may not go through the origin of the graph (Figure 2a), a curvilinear relationship that mayor may not go through the origin (Figure 2b), or they may not be correlated at all (Figure 2c). Many factors will influence the extraction of a nutrient from the soil even with the same chemical solution. The intensity of extraction can be changed by the time and temperature during extraction, the soil to solution ratio, the concentration to the extracting solution and even the vigor of shaking during extraction. The amount of nutrient extracted at various intensities of extraction will vary from soil to soil. Since the amount of nutrient extracted can be influenced by so many factors, it is essential that the procedure is well defined and consistent. Besides the extraction of the nutrient from the soil, the method of quantifYing that element in the extract must also be known and taken into consideration during interpretation of the vaiues. The quantification method must be accurate, repeatable, have suitable range, and free from interference from the extracting solution itself and anything that is extracted from the soil besides the element in question. Methods of quantification may measure different forms of the element, which may result in very different quantities. Some methods (e.g. atomic absorption, ICAl'-}\.ES) will measure all of the element in the solution. Some methods (e.g. colorimetry, ion chromatography) will measure only the inorganic andlor a specific ionic form of a given nutrient. Other methods may include a combination of forms (e.g. hydriodic acid reduction and subsequent sulphur quantification will include sulphate-sulphur of both inorganic and organic forms). It should not be expected that a particular soil test method will be well correlated with growth response since chemical solution extraction together with a specific element quantification method could not precisely simulate extraction of the nutrient by a plant root for a wide range of soil, plant and weather combinations. Compromises must be made from theoretical or ideal conditions to practical situations, particularly when applied to the entire province where soil, weather and crops are so regional and diverse.

4

a). ('l

-- 1~

~

()

ell

~

~

1m

0

:8trnd: 1 b).

.... lOOT

''l

5:t7:

()

~

~

20

0 -J

N

"0

~

'"'

100 50

I

0

0

40

80

60

•

100

Elitract 1

•

c).

•

•• •

20

•

• •

40

60

• • •• • •

•

•• • • • 80

100

Extract 1 Figure 2. Examples of (a) linear (b) curvilinear and (c) no correlations that may occur when a nutrient is extracted from soil by different extractants.

References. Brown, J.R. (ed.) 1987. Soil testing: Sampling, correlation, calibration and interpretation. Soil Sci. Soc. Am. Special Publication no. 21. Soil Sci. Soc. Am., Madison, Wisc. 144 pp. Westerman, R.L. (od.) 1990. Soil testing and plant analysis. Third edition. Soil Sci. Soc. Am. Book Series no. 3. Soil Sci. Soc. Am., Madison, Wisc. 784 pp.

5

SAMPLE PREPARATION: MOISTURE CONTENT AND SUB SAMPLING METHOD T.F. Guthrie Laboratory Manager, Norwest Labs, Langley RECOMMENDATION: Moisture content- Analyses sbould be done on soUs dried at a maximum temperature of 35 to 50 C or on field moist state sborily after receipt. Field moi.'t samples sbould be refrigerated. Corrections to oven dry soil weigbt (determined on a separate subsample) sbould be made, particularly for field moist samples. If no correction is made, basis of expression (fresb, air-dried) sbould be clearly sbown. Subsampling metbod - For extraction of nutrients, eitber a volume (scooped) or weigbed subsample of soil can be used. A volume subsample is more effident for laboratory operations tban weigbIng, but most soil test calibrations bave been derived wltb welgbed subsample extractions. Tbe measurement method (volume or welgbt) sbould be specified and an atljustment for field bulk densily, especially for organic samples, sbould be incorporated for nutrient availability Inde" reports. A relatlonsblp between organic matter by loss-on-Ignltlon and fleid buik densily is proposed. Moisture content oftbe sample used for analysis. Extraction of nuttients can be done more accurately on dried and ground than from moist soil samples since the.sample is easier to homogenize and weigh (James and Wells 1990; Jackson 1965). Air drying need not be complete "in order to facilitate mixing and subsampling, but should not be done at elevated temperatures (more than 35 to 50 C), since the extractability of certain nuttients can be altered at high temperatures (James and Wells 1990; Jackson 1965). Extraction of samples on a "fresh" or "field" moisture content may also be suitable, provided the sample is not stored very long since changes to the extractability of some nuttients may occur. Storage of fresh sanlples, even for short periods, should be under cool (refrigeration) temperatures. Since field moist and even air dried soils can contain variable water contents (Buckman and Brady 1961), correction to oven dried (done on a separate subsample) basis will enhance the accuracy affinal extraction results. Weight versus volume subsampling. To analyze soil for a particular parameter, such as nitrate or phosphate, a small portion of the dried and ground sample must be shaken with an extractant for a specified time, then filtered. The concentration of the parameter is then detennined in the filtrate. Most methods specify that a particular ratio, for example 1 to 10, of soil to extractant be used. Two basic techniques can be used for measuring out a soil sample for analysis: (I). weighing the sample using a balance (weight in grams), or (2). scooping out a volume of sample using a specially designed scoop (volume in ml). The method used is important for interPretation of lbe results, yet it is often ignored in many analytical procedures. For example, this subject is not directly addressed by James and Wells (1990) in a book on soil test methods nor by Jackson (1956) or Page et al (1982) in books on soil analyses. Some of the differences in results between laboratories can be accounted for by the fact that different laboratories use different methods for suhsarnpling (i.e., weigbt or volume). Regardless of whether a laboratory uses a weighed or scooped soil sample, the method used should be consistent with the technique used during the initial calibration studies for the parameter being analyzed, or at least a correlation between the two methods derived (van Lierop 1989). Mehlich (1973) showed that there is consistency of nuttient extraction by each of weigbt and volume methods, but a volume weigbt conection is needed to make two the methods uniform with each other. Assumptions of the volume (scoop) sllbsampling method are: (1). bulk density of soil is not affected by drying and grinding, and

6

(2). the weight of soil in a scoop varies in proportion to the bulk density of soil in the field. The advantages ofthis method are: (I). fasterthan weighing, and (2). conversion of results to kgIha is more straightforward (at least in theory). The dtsadvantages of this method are: (I). there is a greater potential for operator error due to different scooping techniques (ShakalI990), (2). variability can be caused by the degree to which a soil sample packs into the scoop making design, size, etc. of the scoop important (Grava 1975; Tucker 1984), and (3). drying and grinding a soil does change its bulk density, which can lead to error in interpretation of results. Proponents of the weight subsampling method make the assumption that the bulk density of all mineral soils is approximately 1.47 glcm. With this assumption the conversion from ppm to kg/ha is done using the tormula kglba = (2) (ppm). The advantage of this method is that the results are more consistent and reproducible, since there is not so much reliance on operator technique. The disadvantages of this method are: (I). slower than scooping, and (2). soil bulk density decreases as organic matter content increases. Thus the same conversion factor cannot be used to convert ppm to kgiha for all soils. A modified weight subsampJing and reporting method, as used at Norwest Labs, involves the following procedure: (I). use a weighed subsample for soil extractions, (2). determine the percentage organic matter (e.g., loss-on-ignition or Leco C analyzer value converted to organic matter), ,. (3). estimate soil bulk density from percentage organic matter using a regression relationship published by Curtis and Post (1964), then' < (4). from the bulk density, the nutrient value derived by the soil extraction of a weighed subsample is adjusted to a "field" volume basis. This method of extraction and reporting should more closely reilect fertilizer application conditions than simple scooped or weighed methods by accounting for field bulk density. The accuracy of this method is dependent on the precision of the loss-on-ignition and bulk density relationship. The relationship between organic matter content and field bulk density detennined by Curtis and Post (1964) involved undisturbed forest soils. Also, the relationship was curvilinear and the regression equation was: y = 2.09963 - 0.00064 x - 0.22302 x 2, where y = log (bulk density x 100) and x = log (% loss-on-ignition), therefore organic matter content has an increasing influence when it becomes greater than 10%. Harrison and Bocock (1981) showed that accuracy of bulk density prediction using loss-onignition can be improved by using regression relationships that are specific for certain soil types or depth layer.

References. Buckmall, H.O. and Brady, N.C. 1961. The nature and properties of soils. Sixth edition. The Macmillan Company, New York. pp. 162-190. Curtis, R.O. and Post, B.W. 1964. Estimating bulk density from organic matter content in some Vermont forest soils. Soil Sci. Soc. Proc. 28: 285-286. Grava, J. 1975. Causes for variation in phosphorus soil tests. Comm. Soil Sci. Plant Anal. 6: 129-138. Harrisoll, A.F. and Bocock, K.L. 1981. Estimation of soil bulk-density from loss-on-ignition values. J. Applied Ecology 18: 919-927. James, D.W. and Wells, K.L. 1990. Soil sample collection and handling: Technique based on source and degree offield variability. In R.L. Westerman (ed.) Soil testing and plant analysis. Third edition. Soil Sci Soc. Am. Book Series no. 3, Soil Sci. Soc. Am., Madison, Wisc. pp. 25-44. Jackson. M.L. 1965. Soil chemical analysis. Prentice-Hall, Inc., Englewood Cliffs, N.J. 498 pp. Mehlich, A. 1973. Unifonnity of soil test results as influenced by volume weight. Comm. Soil Sci. Plant Anal. 4: 475-486.

7

Page, A.L., Miller. R.H. and Keeney, D.R. 1982. Methods of soil analysis. Part 2. Chemical and microbiological properties. Second edition. Agronomy Series no. 9. Am. Soc. Agronomy, Madison, Wisc. 1159 pp. SbftkAl, A.H. 1990. Anttiy.io oflaborotory error. Comm. Soil Sci, Plant Anal, 21: 1633-1644, Tucker, M.R. 1984. Volumetric soil measures for routine soil testing, Comm, Soil Sci. Plant Anal, 15: 833-840, van Lierop, W. 1989. Effect of assumptions on accuracy of analytical results and liming recommendations when testing a volume or weight of soil, Comm, Soil Sci, Plant Anal, 20: 121-137.

8

MEASUREMENT OF pH AND DETERMINATION OF LIME REQUIREMENT R. Kline and e.G. Kowalenko Soil ConservatioIlfManagemem SpecialisT, Resource Management Branch, B.C. Min. of Agric., Fisheries and Food, Prince George, and Research Scientist, Agriculture Canada Research Station, Agassiz, respectively Recommendations: Measurement of pH: by potentiometric electrode In 1:2 (volume to volume or weight to weight) soil to solution mixture. The solution should be either water or 0.01 M CaCI2. Although the pH values obtained using the two different solutions are quite closely correlated, they are not eqnivalent. For this reason, reportiug of values should document the method used, especially the nafure of the solution. Determination of Ihne regulrement: by the Shoemaker-McLean-Pratt single buffer (SMPSB) method that Involves the measurement of pH of the soil in a buffer made from triethanolamine (TEA). calcium chloride dihydrate [CaCI2.2HZO]. calcium acetate [Ca(OAc)zl. potassium dichromate [K2Cr04]' and para-nitrophenol. A scooped soil sample should be used, particularly for organic samples where the bulk density Is hnportant for the Hme recommendations. For crops more sensitive to soluble aluminum and manganese than soil pH, recommendations from measurement of almninum and manganese in a 1:2 (wt/vol) soil to O.02M CaCI2 solution shaken for 5 to 15 minutes can also be used.

The acidity or alkalinity of the soil has a profound influence on many soil chemical, biochemical and biological processes and ultimately on the nutritional and toxic status of soil elements for plant growth (McLean 1982). Various plants respond differently to the relative acidity or alkalinity of the soil. For example, blueberries are quite tolerant of soil acidity whereas alfalfa prefers a more alkaline soil. The acidity or alkalinity of a soil is largely dependent on the nature of the material from which the soil was derived and the various factors (natural or applied) that have been imposed on that material. Natural weathering processes tend to acidify soils and many agricultural practices (such as fertilizer application, drainage, irrigation) accelerate the process. The acidity of the soil can be purposely reduced by management practices, the most common being the application of limestone. Alkalinity of soils can also be reduced by the addition of acid-forming components such as elemental sulphur. Knowing the relative acidity/alkalinity of a soil is important for general management decisions such as choosing the best crop type for a particular soil. In those instances where the acidity level is unacceptable, a determination of the amOlmt of amendment required to make the adjustment can pro\~de information for specific management practices. The generally used method of expressing the relative acidity/alkalinity of a soil is pH. "Lime requirement" is the term used when a specific amendment recommendation is determined to decrease the acidity of the SOll. The term time is used because timestone is the most common (but not only) product used to reduce soil acidity. Liming of soils to reduce acidity is quite common in British Columbia. There are a few instances where the alkalinity of British Columbia soils has been considered to be too high for specific crops. There has been some recent research on the amount of amendment needed to acidify calcareous soils(Neilsen et al1993), but acidulation of soils has not been widespread. Since pH is such a fundamental aspect of soils, many measurements have been made in British Columbia, for example, for soil survey reports. The focus of this report., however, will be the measurement of soil pH in relation to recommendations for agricultural management practices. MI'3surement of pH Theoretical background The chemlcai definition of pH is the negative logarithm of the hydrogen ion activity jn a solution or pH = -log! 0 [H+] (Jackson 1965). The value for pH is based on the dissociation product (Kw) of water into H+ and OW ions. An equation of this relationship is Kw = [W][OW], where Kw is the dissociation product or constant and [H+] and [OW] are the activities (which is essentially equivalermt to concentrations) of the two ions. In pure water at DOC, the activities of H+ and OW ions are the same and equal to 10-7, and Kw = 10- 14 By rearranging the dissociation equation and converting to a logarithmic scale_ pH of pure water is 7.0. When [H+] is in excess of [OW J the pH would'be < ~.O and the solution is acidic. The solution is basic when [OW] exceeds [WI and the pH is> 70.

9

Solution temperature influences the ionic dissociation of pure water, and the pH is lower than 7.0 when abQve 23°C, and higher than 7.0 when below the 13°C level. While the differences may be in the order of a few tenths of a pH unit, it is important that pH should be measured at standard room temperatures or suitable adjustments should be made if not at this temperature. The pH ot'soil is a measurement of the hydrogen ion activity [H+] in the soil's solution at equilibnum with soil particles (Jackson 1965; van Lierop 1990). Undissociated IV ions remaining on soil exchange sites are not detected by the pH measurement, but are a component of the soil's exchange acidity. The soil's exchange acidity influences the amount oflline or acidifying agents that are required to produce a lasting soil pH change. ~easurementprocedures

Colorimetric and potentioruetric/electrometric procedures have been used most commonly for soil pH measurement (Jackson 1965; ~cLean 1982; van Lierop 1990). Colorimetric procedures are less precise md em be interfered wiili by ilie color derived from the soil that has no direct relationship to the pH, or there can be or absorption of ilie dye of the pH indicator. Commercial litmus paper strips are available iliat can detennine the approximate pH when placed in contact wiili moist soil. This procedure is best suited when accuracy is not essential but quick results are needed, or when access to a standard laboratory pH meter is limited. ~ost laboratories use the potentiometric meiliod of pH detennination over ilie colorimetric procedure. This procedure uses a glass electrode having cation exchange properties with a high degree of sensitivity to H+ ions, paired wiili a reference electrode (either AgCl or Hg-Hg2ClZ) attached to an electromotive force (emf) meter (Jackson 1965; ~cLean 1982; van Lierop 1990). In order to function properly, the electrodes must be immersed in a solution ilierelore water is usually added lor soil pH measurement. The amount of water that ~ added varies considerably, from ilie point of saturated paste to soil:solution ratios as high as I :5. The choice of ilie amount of water has involved a compromise between practical and ilieoretical considerations. Practical considerations may include the relative difficulty of preparing the sample (e.g., it is faster md easier to prepare a specific soil:solution ratio ilian a saturated paste), ease Qf insertion into the soil/solution ruixture, electrode cleansing between sample analysis and consistence wiili hist3ricmeasurements. The interpretation of ilie measurement should take 1his into consideration. Sources of variation Boili field and laboratOlY factors can influence ilie pH of a soil and its measurement. The pH may vary from soil type to soil type due to ilie constituents ofits make-up, weailiering, amendments and management. Wi1hin a soil type, variation may be due to heterogeneity caused by natural processes or specific management practices (e.g., banding fertilizers) and depili of the horizon or profile considered. A number of studies, boili in British Columbia (Kline 1987; Kowalenko 1991) and elsewhere (1vlcLean 1982), have shown iliat soil pH changes from fall to spring. The changes have been boili increases and decreases. The interpretation of a soil pH measurement, ilien, is dependent on when and how ilie sample was taken. Sample handling (drying, temperature, grinding, etc) will also have an influence. Some laboratory sources for pH variations (Jackson 1965; ~cLean 1982; van Lierop 1990).are: I. Salt or liquid junction potential effect: In order for ilie glass electrode meiliod of pH measurement to work, iliere must be an electrical "current" from one electrode (or portion of an electrode in ilie case of single electrode types) to anoilier. If the salt content ofilie solution in which the electrodes are place is vety small. the current is impeded md the measurement of pH may not be accurate. The salts present in ilie soil could vary naturally or from amendments such as fertilizer. The salts could influence ilie reaction ofilie electrode but not the pH ofilie soil. To overcome 1his potential problem, salt solutions sl1ch as 0.01 ~ CaCh or IN KCl have been used instead of water. 2. Suspension effect: ~easurements cm be taken ill a solution wiili or without the soil sediments present. The sediments can be separated by filtration or settling (by centrifugation or time). Settling meiliods can only be done where iliere is a wide soil to solution ratio. Usually readings in ilie presence of sediment are lower than in clear solutions. Some of 1his may be related to ilie liquid junction potential effect since it is thought iliat iliere is greater ionic activity near ilie soil particles than further away. 3. Dilution effect: Soil pH measurements increase wiili increasing soil to solution ratios (e.g., from 1:1 to 1:5) since the hydrogen ion from ilie soil simply get diluted. 4. Temperature effect: Temperature influences the dissociation of W and OW ions of water and therefore pH measurements. Procedures are available for compel1Sating lor temperatures during measurement, usually to 23°e. Although this compensation will mak" laboratory measurements comparable, soils in the field are rarely at room temperature. 5. Carbon dioxide effect: Carbon dioxide from soil carbonates or absorbed from the atmosphere can react with ilie

10

H+ ions in the added solution thus altering the pH of the soil. Only small variations are expected by this factor. Research and mea$uremenls in British Columbia 111e measurement of soil pH in British Columbia has changed somewhat over the period that soils have been analyzed lor soil test pwposes. The common test during the 1960's used a l:l soil water ratio (Nelson 1967). Considerable discussion on the method tor measuring pH took place at the 5th meeting of the British Columbia Subcommittee on Soil Testing Procedures (Anonymous 1967). At that meeting, several people indicated a preference to change to the calcium chloride procedure to reduce the potential variations that occur with the water procedure. Some concerns were expressed over making a change when producers had become used to the interpretations of a water based pH. Since the discussion was mainly about lime requirement recommendations in the Lower Fraser Valley, the method of soil pH measurement was not changed. Between the mid 1960's and mid 1970's a decision was apparently made to use a 1:2 soil:water pH procedure (Neufeld 1980). It would appear that the change was based on expediency since electrical conductivity measurements were done on the same system. Although the accuracy and precision of the pH measurement could be improved by adopting a different ratio or use of a salt solution instead of water (van Lierop 1990), pH values with 1:2 ratio and with water are finnly established in many producer's and extentionisfs minds. Since the difference in pH by the different methods are not large, at least relative to decisions for making management changes, and closely correlated, a change may not be warranted. However, the method of measurement (especially the soil to solution ratio and type of solution) should be docurnented or considered when comparing to literature reports. Determination of lime requirement General background information and research The lime requirement of acid soils is the amount of basic material (liming agents) needed to neutralize the soil acidity from the initial level to a less acid level, or from a low pH to a higher pH (McLean 1982). Originally, soil pH was thought to represent the to¥ acidity found in soils, but it became evident that soils usually have hydrogen ions in excess of that measured in soil solutions. These additional hydrogen ions were assumed to be on exchange sites and became known as exchange acidity. Various soil components or component sites can influence the actual "acidity" of the soil and the activity of some of these are themselves dependent on pH. These include: 1. weak acidic groups on the surfuce of organic matter particles, and from the hydrolysis of non-exchangeable aluminum hydroxides (at pH > 5.5); 2. hydrolysis of exchangeable aluminum in the Al3+ form (at pH < 5.5); 3. dissociated H+ ions from H20 and other sources (at pH < 4.0). The acidity of each soil, then, varies witll organic matter (content and stage of decomposition), clay content and types (e.g., 2:1 or l:llayer configuration), elements (Co, No, AI, Fe, etc.) present, degree of weathering, etc. Some researchers have found that soil clay content has less influence on the lime requirement than other soil factors such as organic matter (Keeney and Corey 1963) or AI-organic complexes (Pionke and Corey 1967). Webber et al (1977), in work on soils in the Peace River region, concluded that lime requirement was best related to "measurements of pH, AI, exchange acidity and organic matter content but not to clay content". Various methods, such as titration, have been used to measure the actual acidity of the soil (McLean 1982). However, titration has not been found to be practical for detemtining the lime requirement of the soil. Witl, the objective of having a laboratory method that is simple, reasonably reliable and quick, buffer procedures (such as by Mehlich) have been proposed as early as 1939. The basis of this approach is to measure pH in a chemical solution that would reflect both solution and bound hydrogen ions in the soil. The use of percentage base saturation of soils has been proposed (peech 1965), but this lime recommendation method has been questioned and may have resulted in enoneous ideas about pH buffering of soils lMagdoff and Bartlett, 1985). Research and methods in British Columbia A sunuuary of lime response trials conducted in the lower Fraser Valley from the 1920's to the mid 1960's showed few had positively influenced crop growth (Fletcher 1965). Data collection appeared spotty, and some of the responses may have been attributed to "increased nitrogen mineralization of the organic matter" indicating that not much nitrogen was added as fertilizer to some of the lime trials. The variable results of liming trials, observations of agricultural extension workers, plus a provincial Lime Subsidy Policy encouraged activity to develop procedures that would identifY soils that would benefit from liming. A recommendation system based on soil properties (nature of surficial deposits, texture and organic matter contents) and crop groupings was developed for the lower Fraser Valley and Vancouver Island (John 1965, 1966, 1967). Dr. John concluded that there were acidic soils in the lower Fraser Valley that were receiving lime but did not require it and tabular data indicated that "soluble aluminum" (extracted in

II

an unstated concentration ofKCn was not a problem in the lower Fraser Valley soils. Base saturationofsoils from the lower Fraser Valley showed that many soils had sufficient calcium and magnesium levels, and that some of these soils were "lime saturated to a greater degree than previously believed." At about the same time, Clark (1965) presented deta on baae saturation of British Columbia soils and proposed that that measurement be developed for lime recommendations. The procedure has been used in Washingron State ,Tumer et a11975) but was never adopted in British Columbia. The soil and crop grouping approach to predicting soil lime requirements for south coast soil and crop combinations proposed by John (1966) was adopted and used by the British Columbia Ministry of Agriculture for many years (Neufeld 1980). Numerous research studies that were conducted during and after this period have contributed information to an understanding of pH and liming even though they were not specifically directed to the development of recommendations (Beaton ot aI 1968; Eaton and John 1971; Heal 1948; Herath and Eaton 1968; Jolm et aI I 972o,b; John and van Laerhoven 1972, 1976; Kowalenko 19800, b; Kowalenko and Maas 1981; Kowalenko et a11980; Kowalenko and van Laerhoven 1980; Maas 1975). Most of the studies were on south coast soils and crops. Neufeld's (1980) outline of methods and interpretations also included a separate table of recommendations for "orchard" soils using the soil and crop grouping approach, but data on which these recommendetions were based have not been documented. More recently there has been considerable research on acid soils that have developed in the Okanagan Valley but the studies have tended to examine the effect on fruit quality and soil response, and indirectly on the development of lime recommendations and (Fisher et a11977; Hogue et a11983; Hogue 1988; Hoyt and Drought 1990; Hoyt and Neilsen 1985; Lidster et a11975; Mason and McDougald 1974; Neilsen et a11981, 1982, 1990; Parchomchuk et al1993; Ross et a11985). The 1980 publication on soil test methods included a procedure for extraction (0.02M CaCli) and analysis of aluminum and manganese, but these analyses were not included as the basis for any of the recoIIL'TIendations (Neufeld 1980). The analysis of these elements was likely included because of extensive research on an acid problem identified in Peace River area soils (Anonymous 1988; Hoyt 1977; Hoyt et aI 1967; Hoyt and Nyborg 1971, 1972,1987; Hoyt and Webber'1974\Nyborg and Hoyt 1978; Webber 1976; Webber et a11977, 1982). Not all acidic soils in British Columbia have high levels of soluble ahmtinum or manganese and plants vary in their sensitivity to these elements, making this approach to lime requirement quite specific to aluminum or manganese toxicity. In 1983, Dr. William van Lierop introduced a modified version of Shoemaker-Mclean-Pratt single buffer procedure (SMP-SB) for determining lime requirement of British Columbia soils (van Lierop and Tran 1983). This procedure was probably based on research in Quebec and included a bulk density adjustment of the scooped soil samples to a weight basis (fran and van Lierop 19810, b; 1982, van Lierop 1989). It is uncertain whether the study used any soil samples from British Columbia. The lime requirement regression equations for scooped soils (volume basis) were adopted by the British Columbia Ministry of Agriculture anf Food Soil and Tissue Testing Laboratory (van Lierop and Tran 1983; Gough 1992). A laboratory trial has shown good correlations between the adjusted SMP-SB and incubation lime requirements for six British Columbia central interior soils (Kline 1984) and a follow-up field trial found that the pre1983 prediction methods proposed by John (1966) were almost equivalent to the SMP-SB procedure when predicting the lime requirements to increase soil to pH of6.0 on coarse to medium texhtred soils (sandy loams to silt loams), but fell short of the target of6.5 for fine textured acid clay soils (Kline 1987). Previous research involving 24 Peace River (from both Alberta and British Columbia) soils, one from the Fraser Valley (Hazelwood soil) and 14 from elsewhere in Canada showed a good correlation between SMP-SB and incubation derived lime requirements (Webber et a11977). The incubation method is assumed to provide good information on lime requirement but is not convenient or suitable for routine testing for recommnedation purposes. The SMP-SB procedure uses triethanolamine (TEA), calcium chloride dihydrate [CaCIZ.2HZO]' calcium acetate [Ca(OAc)2j, potassium dichromate [K2Cr04j, and para-nitrophenol to create a buffer that reacts with the soil's exchange and solution acidity. The change in the pH of a buffer solution resulting from the reaction with the soil is used an indicator of the soil's acidity that requires neutralization (McLean 1982). The soil-buffer pH values are measured and calibrated against CaC03 incubation lime requirements for a region's agricultural soils to specified pH levels, usually 5.5, 6.0 and 6.5 for mineral soils, and 5.4 for organic soils (Gough 1992). The equations relating .lime requirement (LR) with buffer pH measurement., (x) for scooped mineral and organic soils for a plow layer of 20 em (2 mi1lion literslhectare) were: Mineral soils: LR 15.5) = 3.988x2 - 54.54x + 187 LR(6.0)=3129x2 -45.17x+ 164

12

LR (6.5) = 1.I89x2 - 23.55x + 107. Organic soils: LR (5.4) = 69.3 - 11.56x The advantages of the SMP-SB procedure are that it is simple and economical to do in tlle laboratory, is relauvely sensitive and accurare, and is responsive ro soils high in soluble aluminum that are below pH 5.8 and have less than 10% organic matter contents (McLean 1982; van Lierop 1990). The procedure is weak for soils having low lime requirements and oflow exchange capacities. References: Anonymous 1967. Meeting No.5. of the British Columbia Subcommittee on Soil Testing Procedures held 22 November 1966 at the University of British Columbia. Printed at Agassiz, British Columbia. pp. 24-26. (see Appendix III). Anonymous 1988. Recommended methods of soil analysis tor Canadian prairie agricultural soils. Alberta Agriculture, Edmonton. pp. 1-38. Beaton, J.D., Speer, R.C. and Harapiak, J.T. 1968. Response of red clover to Kimberley electric furnace iron slag and other liming materials. Can. J. Plant Sci. 48: 455-466. Clark, J. S. 1965. Base saturation properties of some British Columbia soils. In Report of the Third Meeting of the British Columbia Soil Science Workshop at the University of British Columbia, 14-15 October 1965. pp. 67-76. Eatoll, G. W. and John, M. K. 1971. Effect of lime and manganese upon growth and mineral composition of pea cv.. Dark Skin Perfection. Agron. J. 63: 219-221. Fisher, A.G., Eaton, G.W. and Poritt, S.W. 1977. Internal bark necrosis of Delicious apple in relation to soil pH and leaf manganese. Can. J. Plant Sci. 57: 297-299. Fletcher, H.F. 1965. Response of crops to liming in the Lower Mainland of British Columbia. In Report of the Third Meeting of the British Columbia Soil Science Workshop at the University of British Columbia.. 14-15 October 1965, pp. 57-66.' ',"'" . Gough, N.A. i992. Soil and plant tissue testing methods and interpretations of their results for British Columbia agricultural soils. Draft - British Columbia Ministry of Agriculture, Fisheries and Food. 100 pp. Hea~ G.H. 1948. The effect ofliming on boron availability in Ladner clay and boron fixation in ground tourmaline. M. Sc. Thesis. The University of British Columbia, Vancouver. Hel'atli, H.M.E. and Eaton, G.W. 1968. Some effects of water table, pH, and nitrogen fertilization upon growth and nutrient-element content of highbush blueberry plants. Proc. Am. Soc. Hort. Sci. 92: 274-283. Hogue, E.J. 1988. The relationship of internal bark necrosis in "Delicious" apples to tree characteristics and soil properties. Comm. Soil Sci. Plant Anal. 19:1041-1048. Hogue, E.J., Neilsen, G.B., Mason, J.L. and Drought, B.G. 1983. The effect of different calcium levels on cation concentration in leaves and fruit of apple trees. Can. J. Plan! Sci. 63: 473-479. Hoyt, P.B. 1977. Effects of organic mater content on exchangeable AI and pH-dependent acidity of very acid soils. Can. J. Soil Sci. 57: 221-222. Hoyt, P.B. and Drought, B.G. 1990. Techniques for speeding the movement of lime into an orchard soil. Can. J. Soil Sci. 70: 149-156. Hoyt, P.B. and Neilsen, G.H. 1985. Effects of soil pH and associated cations on growth of apple trees planted in an old orchard soil. Plant and Soil 86: 395-401. Hoyt, P.B. and Nyborg, M. 1971. Toxic metals in acid soil I: Estimation of plant-available alwninum. Soil Sci. Soc. Amer. Proc. 35: 236-240. Hoyt, P.B. and Nyborg, M. 1972. Use of dilute calcium chloride for the extraction of plant available alwninum and manganese from acid soil. Can. J. Soil Sci. 52: 163-167. Hoyt, P.B. and Nyborg, M. 1987. Field calibration ofliming responses offour crops using pH, AI and Mn. Plant md Soil I 02: 21-25. Hoyt, P.B. and Webber, M.D. 1974. Rapid measurement of plant-available alwninum and manganese in acid Canadian soils. Can. J. Soil Sci. 54: 53-61. Hoyt, P.B., Hennig, A.M.F. and Dobb, J.L. 1967. Response of barley and alfalfa to liming ofsolonetzic, podzolic and g1eysolic soils of the Peace River region. Can. J. Soil Sci. 47: 15-21. . Jackson. M.L 1965. Soil chemical analysis. Prentice-Hall, Inc., Englewood Cliffs, N.J. 498 pp. John, M.K. 1965. Principles involved in liming of Bntish Columbia soils. In Report of the Third Meeting of the

13

British Colmnbia Soil Science Workshop at the University of British Colmnbia, 14-15 October 1965. pp.49-57. John, M.K. 1966. Nature of soil properties and their relation to lime requirements. In Proceedings of the British Colmnbia Subcommittee on Soil Testing, 22 November 1966, at the University of British Colmnbia. pp. 27-33. (see Appendix IV). John, !\I.K. 1967. Background research in support of the British Colmnbia soil testing service. In Report of the meetirig of the Western Section of the National Soil Fertility Committee, 8-9 February 1967, Saskatoon, Saskatchewan. pp. -1-12. (see Appendi.'[ V). John, M. K., and van Laerhoven, C. 1972. Lead uptake by lettuce and oats as affected by lime, nitrogen and source oflead. J. Environ. Quai. I: 169-171. John, M. K., and van Laerhoven, C. 1976. Effects of sewage sludge composition, application rate, and lime regime on plant availability of heavy metals. J. Environ. Qual. 5: 246-251. John, M. K., Case, V.W. and van Laerhoven, C. 1972a. Liming of a1fulfu (Medicago sativa L.) l. Effect on plant growth and soil properties. Plant and Soil 37: 353-361. John, M.K., Eaton, G. W., Case, V.W., and Chuah, H.H. 1972b. Liming of alfalfa C~fedicago sativa L.) II. Effect on mineral composition. Plant and Soil 37: 363-374. Keeney, D.R. and Corey, R.B. 1963. Factors affecting the lime requirement of Wisconsin soils. Soil Sci. Soc. Am. Proc. 27: 277-280. KlIne, R. 1984. Incubation lime requirement trial on six British Columbia Central Interior soils. Unpublished report - British Colwnbia Ministry of Agriculture and Food. pp. 1-10. (see Appendix VI). KlIne, R. 1987. Liming trials in British Columbia's Central Interior. Unpublished report - British Colwnbia Ministry of Agriculture, Fisheries and Food. pp. 1-14. (see Appendix VII). Kowalenko, C.G. 1980a. Response of cauliflower to soil lime and foliar manganese and zinc applications. Res. Review (Agassiz), Feb. pp.l1-12. Kowalenko, C.G. 19~Ob. Updat~.911 soil lime and foliar manganese and zinc application trial. Res. Review (J\.gassiz), Nov. p.8. . Kowalenko, C.G. 1991. Fall vs spring soil sampling for calibratirig nutrient applications on individual fields. J. Production Agric. 4: 322-329. Kowalenko, C.G., and Maas, E.F. 1981. Some effects of fertilizer and lime application to filbert orchards in the Fraser Valley of British Columbia. Can. J. Soil Sci. 62: 71-77. Kowalenko, C.G. and van Laerhoven, C. 1980. Liming trials on com production. Technical report - Agriculture Canada, Agassiz Research Station, Agassiz, British Colmnbia. (see Appendix VIII). Kowalenko, C.G., Maas, E.F., and van Laerhovcn, C.1. 1980. Residual effects of high rates of limestone, P, K and Mg applications: Evidence of induced Mri and Zn deficiency in oats. Can. J. Soil Sci. 60: 757-761. Lidster, P.D., Porritt, S.W., Eaton, G.W. and Mason, J. 1975. Spartan apple breakdown as affected by orchard factors, nutrient content and fruit quality. Can. J. Plant Sci. 55: 443-446. Ma.s, E.F. 1975. The organic soils of Vancouver Island. Agric. Can. pp. 1-7. Magdoff, F.R., and Bartlett, R.J. 1985. Soil pH buffering revisited. Soil Sci. Soc. Am. J. 49: 145-148. Mason, J.L. and McDougald, J.M. 1974. Influence of calcimn concentration in nutrient solutions on breakdown and nutrient uptake in "Spartan" apple. J. Am. Soc. Hort. Sci. 99: 318-321. McLean, E. D. 1982. Soil pH and lime requirement. In A.L. Page, R. H. Miller, and D.R. Keeney (ed.) Methods of Soil Analysis Part 2, Chemical and Microbiological Properties. 2nd Edition. Agronomy Book series no. 9, Am. Soc. ofAgron., Inc. Madison, Wis. pp. 199-224. Neilsen, G.H., Hogue, E. and Drought, B.G. 1981. The effects of surface-applied calcium on soil and mature Spartan apple trees. Can. J. Soil Sci. 61: 295-302. Neilsen, G.H., Hoyt, P.B. and Lao, D.L. 1982. EffeclS of surface soil pH on soil cation content, leaf nutrient levels and quality of apples in British Columbia. Can. J. Plant Sci. 62: 695-702. Neilsen, G.H., Neilsen, D. and Atkinson, D. 1990. Top and root growth and nutrient absorption of Pnmus avium at two soil pH and P levels. Plant and Soil 121: 137-144. Neilsen, D., Hogue, E.J., Hoyt, P.B. and Drought, B.G. 1993. Oxidation of elemental sulphur and acidulation of calcareous orchard soils in southern British Columbia. Can. J. Soil Sci. 73: 103- 114. Nelson, C.H. 1967. The British Colmnbia soil testing laboratory: Services offered. In Report of the meeting of the Western Section ofthe National Soil Fertility Committee, 8-9 February 1967, Saskatoon, Saskatchewan. pp.I-4. Neufeld. J. H. 1980. Soil testing methods and interpretation. B. C. Min. of Agric. 80-2. Victoria. 29 pp.

14

Nyborg. M. and Hoyt, P.B. 1978. Effects of soil acidity and liming on mineralization of soil nitrogen. Can. J. Soil Sci. 58: 331-338. Parchomchuk, P., Neilsen, G.H. and Hogue, EJ. 1993. Effects of drip irrigation ofNH4-N and P on soil pH and cation leaching. Can. J. Soil Sci. 73: 157-164. PeilCh, M. 1965. Lime requtrement In C.A. Black, D.D. Evans. J.L. White, L.E. Ensminger, F. E. Clark (ed.) Methods of Soil Analysis Part 2, Chemical and Microbiological Properties. Agronomy 9. Am. Soc. of Agron., Madison, Wis. pp.927-932. Penny, D.C., Nyborg. M., Hoyt, P.B., Rice, W.A., Siemens, B., and Laverty, D.H. 1977. An assessment of the soil acidity problem in Alberta and northeastern British Columbia. Can. J. Soil Sci. 57: 157-164. Pionke, H.B. and Corey, R.B. 1967.' Relations between acidic aluminum and soil pH, clay and organic matter. Soil Sci. Am. Proc. 31: 749-752. Ross, G.J., Hoyt, P.B and Neilsen, G.H. 1985. Soil chemical and mineralogical changes due to acidification in Okanagan apple orchards. Can. J. Soil Sci. 65: 347-355. Tran, T.S. and van Lierop, W. 1981a. Evaluation and improvement of buffer-pH lime requtrement methods. Soil Sci. 131: 178-187. Tran, T.S. and van Lierop, W. 1981b. Evaluation des methodes de determination du besoin en chaux en relation avec les proprietes physiques et chimiques des sols acides. Science du Sol 3: 253-267. Tran, T.S. and van Lierop, W. 1982. Lime requtrement determination for attaining pH 5.5 and 6.0 of coarsetextured soils using buffer-pH methods. Soil Sci. Soc. Am. J. 46: 1008-1014. Turner, D.O., Halvorson A.R., Mortensen W. P., Baker A. S., and Fanning C. D. 1975 White clover-grass pasture for western Washington - Fertilizer Guide. Washington State University, Pullman. van Lierop, W. 1983. Lime requtrement determination of acid organic soils using buffer-pH methods. Can. J. Soil Sci. 63: 411-423. van Lierop, W. 1989, Effect of asSumptions on accuracy of analytical results and liming recommendations when testing a volume or weight of soil. Comm. Soil Sci. Plant Anal. 20: 121-137. van Lierop, W. 1990. Soil pH and lime requtrement determination. In R.L. Westerman (ed.) Soil testing and plant analysis. Third edition. Book series no. 3. Soil Sci. Soc. Am., Madison, Wisc. pp.73-126. van Lierop, W. and Tran, T. S. 1983. Lime requtrement determination of acid mineral and organic soils using fue SMP buffer-pH method. Internal report to British Columbia Min. of Agric. and Food. 14 pp. (see Appendix IX). Webber, M.D. 1976. Distribution constant for calcium plus magnesium and manganese exchange in acid Canadian soils. Can. J. Soil Sci. 56: 115-118. Webber, M.D., Hoyt, P.B. and Cornean, D. 1982. Soluble AI, exchangeable AI, base saturation and pH in relation to barley yield on Canadian acid soils. Can. J. Soil Sci. 62: 397-405. Webber, M.D., Hoyt, P.B., Nyborg M. and Comeau, D. 1977. A comparison of lime requirement methods for acid Canadian soils. Can. J. Soil Sci. 57: 361-370.

15

SALINITY AND SODICITY MEASUREMENTS R. Kline and C.G. Kowalenko Soil Conservation/Management Specialist, Resource Management Br., B. C. Min. of Agric., Fisheries and Food, Prince George, and Research Scientist, Agriculurre Canada Research Station, Agassiz, respectively

RECOMMENDATION: Salinitv - Under laboratory conditions, measure with electrical condncttvlty meter on 1:2 soll:water suspension or In a saturated paste. Since the values will differ with the proporllon of water used, the method of sample preparation should be documented. The meter. should be properly calibrated, and vaIues compensated for temperatu"e if not dOlle at 2S oC. Values are preferably expressed in deciSiemells per meter (dS/m) or mlIUmhosicentimeter (mmhos/em). Measurements can also be In the fteld using porous salinity sensors, electromagnellc (EM) or time domain refiectometry (TDR) instruments. Sodicity - Sodium absorption ratio: Measure sodium, calcium and magnesium in the solution of a saturated paste and calculate the ratio of sodium concentration to the square root of calcium-plus-magnesium concentrations. Extractable sodium: Sodlmn can be measured in extractions with 1.0 N annnonium acetate (exchangeable sodium) or by the Kelowna multiple element extractant consisting of a mixture of acetic acid and ammoulum ftuorlde (extractable sodium). Since the values by the two extraction solution are not precisely eqUivalent, the solution used should be clearly stated.

The total concentration allifthe types of salt present in the soil solution can influence absorption of nutrients and water by plant roots. Various plants have different tolerances to the amount of salt present (Rhoades 1982; Rhoades and Miyamoto 1990). Carrot and strawberry are sensitive whereas barley and tall wheatgrasss are tolerant to the salt content of the soil solution. The presence of a relatively large proportion of sodium in the soil solution enhances the detrimental effect of salts on crop grOV'lth since sodium compounds tend to be relatively soluble, sodium is not an essential nutrient for crop growth and can adversely affect soil strucnrre. A few crops, however, can substitute sodium for potassium in metabolic functions (Knudsen et alI982). Salinity is the term used in reterence to the amount of soluble salt in the soil, and sodicity relates to the relative proportion of sodium in the soluble salts of the soil solution. Soluble salts and the proportion of sodium in the solution are problems in arid climates, where evaporation can concentrate salts at the soil surtace from subsUliace horizons or from irrigation, and also in areas that are influenced by sea water. Soil salinity and sodicity measurements, theretore, are important for the management of many crops grown in British Columbia. Salinity

The standard approach to detennine the degree of soil salil1ity is to measure the electrical conductivity of soil solution suspensions or extracts with a calibrated meter in a laboratory at a standard temperanrre of 25°C, or with temperanrres compensation (Janzen 1993; Rhoades 1982; Rhoades and Miyamoto 19QO)' The standard unit of measurement currently used is deciSiemens per meter (dS/m), and is equivalent to milliSiemens per centimeter (mS/em) or millimhos per centimeter (mmhosicm) used previously. Field (in-situ) procedures involving porous salinity sensors, electromagnetic (EM) or time domain reflectometry (WR) are possible !Dasberg and :-ladler 1988; Herrry et al1987; Rhoades 1982; Rhoades and Miyamoto 1990), but are not consider in this discussion oflaboratory' methods. Electrical conductivity measurements in the laboratory are done either at saturation or fixed soil to water (e.g., 1:1 or 1:2) ratio (Janzen 1993; Rhoades 1982; Rhoades and Miyamoto 1990). The saturated paste procedure involves the addition of deionized water until a characteristic sanrration end-point is reached and the solution is collected by suction. The sanrration paste extract procedure requires more time than lixed ratio procedures requiring recognition of the saturation point is quite operator dependent, but the· value is directly related to tield soil moisnrre conditions and is useful for determining soil salinity impacts on plant growth over a range of soil moisnrre and texnrral conditions (Richards 1954). The fixed ratio methods are more convenient in the laboratory, since they can

16

be further simplified by measuring electrical conductivity in the supernatant solution and then used for other me"urements such as pH (Neufeld 1980). . Although there is a relatively good relationship between the fixed ratio measurements and saturation paste extracts at low salinity, the relationship is not good lor soils of high salinity. For example in British Columbia, an "apparent" soil saJinity reading (ECa) measured in 1:2 soil:water suspension was adjusted to saturated paste extract IEC) values by multiplying by 2 for ECa values up to I dSfm (0.5 dSfm unadjusted), but the procedure is s,,~tched to saturation paste extract when soils have ECu greater than I. Relationships between saturation paste extracts and fixed ratio procedures can be affected by the types of salts present in the soil. Carbonate and sulphate salts are less soluble than chloride salts, and the relative cation balance in soils may affect the performance of the suspension techniques. Addition of 0.1% (NaP0 3)6 at I drop/25 m1 extract will reduce the potential for CaC03 precipitation during equilibration periods (Rhoades and Miyamoto 1990). Research in Saskatchewan has verified the use of 1:2 soil water suspensions for diagnostic use compared to 1:2 soil:water and saturated paste extracts (Hogg and Henry, 1984). Alberta and Saskatchewan use fixed ratio extracts (1:2 and 1:1 soil:water mixture, respectively) and adjust values for soil textural groupings or derived relationships to the saturation paste extract (Soil Test Technical Advisoty Group 1988~ Hogg and Henry 1984). Texture has been showll to have an influence on the interpretation of EC values (Richards 1954). Although the EC will vary at diffemt field moisture contents, the saturated paste measurement still gives a good relative measure of the soluble salt content of soils. As indicated previously, a combination of a fixed ratio and saturated paste methods have been used in British Columbia to evaluate the salinity status of soils (Neufeld 1980). There is negligible documentation on the basis on which this was derived and appears that data from elsewhere was accepted. Wolterson (1983) indicated that there was a strong correlation between EC sat and ECl: I for one soil near Mud Bat and another on Westham Island flooded by sea water in 1982. Some further work under British Colmnbia soil, weather and crop conditions is advised lor verification. Sodlclty

., '

The sodium adsorption ratio (SAR) is an empirical calculation procedure relating mono-valent sodium (Na) cation concentration to the squore root sum of the divalent calcium (Ca) and magnesium (Mg) cation concentrations in the saturation paste extract, and has been used to indicate potential sodicity impacts on plant growth (Rhoades and Miyamoto 1990~ Russe1l1973~ Richards 1954). The relationship is an approximation in that concentration of the cations is used instead of activities, but the difference is very small. As for salinity, it appears that the sodium adsorption ratio was adopted in British Columbia from research in other areas Mth little or negligible local research (Neufeld 1980). Because of the relative difficulty preparing a saturated paste of the soil, the small amount of liquid that can be extracted from a saturated paste, and adoption of a fixed ratio method for measuring salinity, alternate methods for measuring the sodicity of soils have been proposed. Henry et aI (1987) have shown that exchangeable sodium percentage \ESP) is linearly related to SAR (i.e., ESP = (0.0147 x SAR)+O.99) suggesting that soluble and exchangeable sodium are quite closely related. Exchangeable sodium percentage is determined from cation exchange capacity (CEC) measurements. Exchangeable sodium extraction procedures remove both soluble and exchangeable cations, therefore adjustment is needed to make exchangeable values comparable to measurements by the saturated paste extraction procedure. The correction for water soluble sodium is important in dry region soils (Knudsen et aI 1982). A commonly used procedure for determining exchangeable sodium has been 1.0 N ammonium acetate (NH40AC) in a 1:5 wtfvol (soil:solution) extraction (Jackson 1965; Hendershot et alI993). The Kelowna multiple element solution correlated well Mth ammonium acetate for detarmining extractable sodium levels in British Columbia soils (van Lierop and Gough 1989). The authors noted that sodium, which is not difficult to determine by flame procedures, is a common contaminant therefore precautions should be taken for deriving meaningful values. This study did not examine the nutritional or toxicity implications of extractable sodium for agricultural crops in British Columbia. However, since this extraction of sodium is closely correlated Mth exchangeable sodium (which in tum is correlated Mth soluble sodium), it should provide initial detection of potentially damaging sodium levels in an ""tract used for other soil test purposes. .'ill estimated sodium adsorption ratio (SAR) can be calculated from the calcium, magnesium and sodium extracted by the Kelowna solution to determine if there are high sodium contents thal could cause crop growth problems. It is possible, however, to have a soil highly saturated with sodium but low exchangeabJe-sodium, therefore EC and SAR in saturated paste extracts would give more accurate results.

17

References: Dasberg, S. and Nadler, A. 1988. Soil salinity measurementq. Soil Use and Management 4: 127·133. Hendershot, WJI., Lalande, Land Dnqnette, M. 1993. Ion exchange and e.'{changeable cation:;. In 1!.R. Cartor (Ed.) Soil.ampling fUldmethod. of Molysis. Lewis Publishers, Boca Raton, Florida. pp.167·176 Henry, J.L., Harron, W., Flaten, D. 1987. The nature and management of salt affected land in Saskatchewan. Soils and Crops Branch, Saskatchewan Agriculture. Agdex 518. 23 pp. Hogg, T..1. and Henry, J.L. 1984. Comparison of I:I and 1:2 suspen:;ion:; and extracts with tile saturation paste extract in estimating salinity in Saskatchewan soils. Can]. Soil Sci. 64: 699· 704. Janzen, H.H. 1993. Soluble salts. In M.R. Carter (ed.) Soil sampling and methods of analysis. Lewis Publishers. Boca Raton, Florida. pp. 161·166. Jackson. M.L. 1965. Soil chemical analysis. Prentice·Hall, Inc., Englewood ClifTh, N.J. 498 pp. Knudsen, D., Peterson, G.A., and Pratt, P.F. 1982. Lithium. sodium and potassium. In A.L. Page, R.H. Miller. and D.R. Keeney (eds.) Methods of soil analysis. Part 2, Chemical and Microbiological Properties. 2nd Edition. Agronomy Book sereis no. 9. Am. Soc. of Agron., Madison, Wis. pp.225-246. Neufeld, J. H. 1980. Soil testing methods and interpretation. B. C. Min. of Agric. 80·2. Victoria. 29 pp. Rhoades, J.D. 1982. Soluble salts. In A.L. Page. R.H. Miller. and D.R. Keeny (eds.) Methods of soil analysis. Part 2, Chemical and Microbiological Properties. 2nd Edition. Agronomy Book series no. 9. Am. Soc. of Agron., Madison, Wis. pp. 167-179. Rhoades, J.D. and Miyamoto, S. 1990. Testing soils lor salinity and sodicity. In R.L. Westerman (ed.) Soil testmg and plant analysis. Third edition. Book series no. 3. Soil Sci. Soc. Am., Madison, Wisc. pp. 299-336. , Russell, E.W.1973. Soil condition:; and plant growth. lOth Ed. Longman, ;-.Jew York. Soil Test Technical AdvisOl'Y Group. 1988. Soil Test Recommendations tor Alberta: Technical Manual. Alberta Agnculture, Edmonton. pp. 7·8 Richards, L.A. (ed.) 1954. Diagnosis and improvement of saline and alkali solls. United States Salinity Laboratory Swr: United States Department of Agriculture Handbook No. 60. van Lierop, W. and Gough, N.A. 1989. Extraction of potassium and sodium from acid and calcareous soils with the Kelowna multiple element extractant. Can. J. Soil Sci. 69: 235·242. Wolterson, E. 1983. The relationship between electrical conducti,ity measured on a saturated paste and electrical conductivity measured on a 2:1 extract. Soil Science 315 Term Project, University of British Columbia, Vancouver. 16 pp. (see Appendix X).

18

NITRATE, TOTAL NITROGEN AND ORGANIC MATTER DETERMINATIONS C.G. Kowalenko Research Scientist. Agricultme Carlada Research Station, Agassiz

RECOMMENDA nON: Nitrate nitrogen - Any extractant/analysis comblnatton that Is compatible and where the extractant contains a moderately high anion concentration, e.g. 2 N KCl combined with automated colorimetry such as continuous or segmented flow based on uttrite after nitrate reduction. The high anion concentratton should ensure complele extraction from anion-adsorbing solis, and reduce the potential for microbial or enzyme alteration of nitrate during or after extraction. Precautions should be taken to minimize contamination particularly from sample containers, analysis reagents and fIlter paper. Total uttrogen - Any Kjeldabl or dry-asb-Instrumental method (e.g. Loco N analyser) that has been shown to be suitable for soli analyses. Oroanic matter content - Any loss-on-Ignition, wei-ash or specific-Instrmuentation methods that have been tested on soil materials; reports should clearly but briefly state method used (e.g. loss-on-Ignitlon, Walkley-Black, Loco Instrument) and basis of expression (e.g. % organic matter or % organic C, etc.). Uses of organic matter, total nitrogen and nitrate measuremeuts for soli testing. Organic matter and mttate concentrations have been routinely determined on soils samples submitted for soil test analysis, but neither measurement has been used directly for fertilizer recOlmnendations .except for nitrate for Peace River area nitrogen recorrunendations (Neufeld 1980). The organic matter content of soils is useful as general characterization infonnation, such as for distinguishing OrganiC from mineral soils (van Lierop 1989). Schreier (1983) speculated tlJat organic matter was an important parameter for raspbeny production because of its influence on water-holding capacity and cation retention. Goldin and Lavkulich (1990) showed that the organic matter and nitrogen contents decreased upon clearing and cultivating land tor various lengths of time in the Fraser Lowland, which would have an influence on soil productiviry. Nitrate has been used to repon possible nitrogen excesses, but not to make site-specific fertilizer recommendations. Currently only general nitrogen recorrunendations are available for British Columbia crop production. Research on nitrogen for crop production to the mid 1980's was largely crop oriented, with rew studies that reported soil analyses (Kowalenko 1987a), theretore development of recommendations based on a soil analysis have not been possible. Recent field research in British Columbia, particularly in the south coast (Kowalenko 1987b, 1989; Kowalenko and Hall 1987a, b; Kowalenko et al 1989; Zebarth et aI 1991), has included soil nitrogen measurements. Stevenson and Neilsen (1990) measured nitrate in drainage water from 1ysimeters growing apple trees, but did not report soil nitrate measurements. As more research data that includes soil nitrogen measurements accumulates, soil-test based recommendations may be possible for areas of British Columbia besides the Peace River area. Some soil-test oriented approaches that show potential for site-specific nitrogen recommendations include: 1. spring soil nitrate for inigated com in the interior (van Ryswyk 1985), 2. soil nittate at sidedress time lor com production grown at the coast (Weinberg 1987), and 3. thll ,oil nitrate as a "feed-back" approach for various south coastal crops (Kowalenko 1991). In each of these cases, soil samples deeper than currently recorrunended for soil test analysis (Plough layer) will likely be required. A nittogen simulation model developed in British Columbia (Bulley and Cappalaere 1978) uses total nitrogen (or organic matter with a suitable conversion to total nitrogen) as an input factor from which nitrogen lnineralization is calculated. AltllOUgh total nitrogen is probably the most reasonable !hctor to use tor the simulation of mineralization given available intormation, it is uot a suitably sensitive value to distingmsh soils of the same total nitrogen content but different nitrogen supplying capability. Wilcox and Walker (1946), for example, did not detect a relationship betweeri tree growth or tree yield with organic matter contents in the Okanagan where smface (0-8 inch) organic matter ranged from 0.6 to 3.9 %.

19

Nitrate measurement Nitrate is usually assumed to be present in soil solution, unadsorbed by the solid (organic and inorganic) components of the soil. The classical method of extracting nitrate and anunoniurn has been with 2 M KC1 (Keeney and Nelson 1982). Since nitrate is readily extracted from the soil, the high concentration ofKCI was required to extract ammonium. Initially, the British Columbia Soil Test Laboratory used 0.02 N CuS04 and 0.007 N Ag2S04 for nitrate extraction (Neufeld 1980), but this extractant was necessary to reduce chloride and organic matter intreferences that occur for the phenol 2:4-disulfonic colorimetric that was used for quantification. Subsequently, van Lierop (1986) found that nitrate could be successfully determined in the Kelowna multiple element extraction provided that the quantification method is compatible with the extraction solution and constituents extracted from the soil by the extractant. There was an excellent correlation (r = 0.998) between soil nitrate extracted with 2 N KCI and Kelowna extracts and measured by an automated copper-cadmium reduction procedure for 31 British Columbia soils. Since nitrate is readily extracted from soils, water or almost any other extractant could be used for extracting all the nitrate present in the soil. The choice of extraction solution should be compatible with the subsequent quantification of the extracted nitrate i.e., the quantification should not be subject to interference, must be reproducible, precise, and converrient to use. Kowalenko (1989) speculated that south coast British Columbia soils may adsorb nitrate since sulphate adsorption had been documented. Recently, column leaching and equilibration studies have confirmed that nitrate adsorption does occur in soils of that area (Kowalenko unpublished data). Further, it was observed that measurements of nitrate extracted by KCI extractant ranging from 0.1 to 2.0 M differed and it was speculated that the difference' was due differential microbial or enzyme activity in the solutions of different concentrations andlor different soil constituents extracted by KCI solutions of different concentrations. This speculation was supported by differences in visual ratings of microbial growth that occurred during sto;age of the extracts (Table I). As would be ' .' .,,-.

Table I Microbial growth (visually rated from negligible KCI concentrations from five Fraser Valley soils. Soil # 2b 6 9 lOa Mean

0.1 MKCI 0.88 0.88 1.00 1.50 1.05

= a to maximum = 3) in soil extracts of various

LOMKCI 1.75 2.25 1.88 1.62 1.50

2.0MKCI 0.50 0.25 0.13 0.00 0.21

expected, lowest microbial growth occurred in the extract that had the highest KCI concentration (2 M). From these preliminary observations on rritrate adsorption in the soil and possible microbial/enzyme activity in soil extracts, it is recommended that for British Columbia soils (especially those of the south coast or soils that are acidic) the extracting solution should contain a significant anion concentration to displace adsorbed nitrate. Also, to minimize potential for enzyme or microbial changes of the nitrate extracted, a high salt concentration, or elements that are known to reduce enzyme activity (e.g. copper, mercury, etc.) should be included in the extracting solution. However, it should be ensured that the quantification step (colorimetry, specific ion electrode, etc.) is compatible with the extracting solution. Precautions should be taken to minimize nitrate contamination from all potential sources such as from sample containers (Kitchen et al 1989), in extraction and analysis reagents (Qasim and Flowers 1989b), and filter paper (Qasim and Flowers 1989a, Sparrow and Masiak 1987). Tests at Agassiz Research Station (Kowalenko unpublished data) showed that washing filter paper with extracting solution just prior to filtering soil extraction solution can effectively minimize contmnination from filter paper. Appropriate reagent blanks, preparing standards in extraction solutions, etc., should always be done.

20

Toml nitrogen and organic matter measnrements. Initially, the provincial test laboratory used the Walkley-Black method to determine the organic matter content of soils (Neufeld 1980). Lass-an-ignitiou (LOI) was subsequently adapted (van Lierop 1986) in order to simplifY the measurement. Goldin (1987) fOlUld there was an excellent correlation between loss-an-ignition (600 C) and Leco Induction Furnace (Model 521) organic carbon (Leco-C) measurements for noncalcareous soils from northwestern Washington and British Columbia. The regression between the two methods varied for mineral and litter la)'er samples and were: Leco-C = - 0.710 + 0.405 (LOI) (r2 = 0.86) for mineral samples and Leco-C = - 2.492 + 0.417 (LOI) (r2 = 0.89) for litter layer samples. A comparison of loss-an-ignition (450 C) with Walkley Black organic C (W.B.-C) of 17 samples representing 11 different Fraser Valley soils also resulted an excellent correlation but a slightly different regresSIOn: W.B.-C = -0.79 + 0.56 (LOI) (r2 = 0.996) (Figure I).

~

::f!.

•C)

~

20.00 15.00

I

ejl

q

cO

:;i

'"00] 5.00 0.00 0.00

•

. •

•

•

"..

,.

5.110 . 10.00

15.00

20.00

25.00

30.00

Loss on ign.(%) Figure 1. Comparison of two methods of organic matter content measurement on 17 samples representing II different Fraser Valley soils.

The slopes of the regression equations are considerably less than one since the Leco-C instrument and Walkley-Black method measure carbon whereas loss-on-ignition measures organic matter. To convert % Walkley-Black organic C to % organic matter, Jackson (1965) recommended that the value be multiplied by 1.11 since he assumed that the Walkley-Black method oxidizes 90% of the organic matter and by 1.724 because soil organic matter averages approximately 58% carbon. The intercept of the above Walkley Black and loss on ignition (-0.79) suggests that the Walkley Black method was only 80% efficient for Fraser Valley soils and the organic carbon content was 56% (i.e. slope of 0.56). Since organic matter is used in only a general way for soil test recommendations, it is recommended that any well recognized organic matter or organic C method can be used for determination. However, the method of analysis should be clearly stated and the method of expression (organic matter or organic C) must be accurate. General assumptions and specific comparisons should be used or taken into account when interpreting values derived by different methods. Numerous methods are available to determine total nitrogen, but wet oxidation and dIy ash methods are most generally used (Bremner and Mulvaney 1982). The most common wet oxidation is some adaptation or modification of a method initiated by Kjeldahl and dIy ashing is by various commercial instrumentation based a method initiated by Dumas. A recent examination of 17 samples of 11 different Fraser Valley soils (Kowalenko unpublished data) showed that dIy ash by Leco N analyzer (Leco N) and a Kjeldshl (Kjel.N) compared very well: Kjel. N = -0.02 + 0.93 (Leco N) (rZ = 0.99) (Figure 2).

21

1.2 ,

Z

1 0.8 .;l 0.6

c

j

••

'" 0.4 1 C ,

Cl

0.2

I

1

••

O~I~~------------------------------~ o 0.8 1.2 0.2 0.4 0.6

Kjeldahl (%N) Figure 2. Comparison of two nitrogen analysis methods on 17 samples of 11 differenet Fraser Valley soils.

AB for organic matter/carbon analysis, it is recommended that any well recognized total soil nitrogen analysis method can be used., preferably with documentation of the analysis method (e.g. Kjeldahl, Leco, etc.). Although total nitrogen has not been routinely determined for soil test pmposes, the use of this value in the Nitrogen Simulation Model may make it a useful measurement to have. It is possible to use

organic carbon or organic matter values, but an assumed ratio or regression equation is needed to make the conversion to total nitrogen. Since total nitrogen can be used in a more direct way than organic matter, it may be more suitable to detefII)ine total nitrogen of soils and then derive the organic matter content from that value rather than the othet way around. Correlations between Kjeldahl nitrogen and loss-on-ignition organic matter or Walkley-Black organic carbon (Kowalenko unpublished data) were excellent for 17 samples of II different Fraser Valley soils: LOI = -1.71 + 26.6 (Kjel. N) (r2 = 0.96); LOl/KjeI. N = 23.5 and W.B.-C = -1.59 + 15.1 (Kjel. N) (r2 = 0.96); W.B.-C/Kjel. N = 11.2 (Figure 3).

e'?Z 1.50 100 . ~

0.50

I .r1I •

• •

'? 1.50 ~ 1.00

~ 0.50

0.00 ~--'.=-----------~--~ 0.00 10.00 20.00 30.00

I

• • ...

•

0.00 ~.=-------~-~ 0.00 5.00 10.00 15.00 20.00 W.B. org. C(%)

Loss on ign.(%)

Figure 3. Comparison of Kjeldahl N with two methods of organic matter analyses for 17 samples of 11 different Fraser Valley soils.

Reference•. Bremner, J.M. and Mulvaney, C.S. 1982. Nitrogen -- Total. In A.L. Page et al (eds.) Methods of soil analysis. Part 2, Chemical and microbiological properties. Second edition. Agronomy Series no. 9. Am. Soc. Agron., Madison, Wisc. pp. 595-624. Bulley, N.R. and Cappalaere, B. 1978. A dynamic simulation model of nitrogen movement on livestock farms. Paper 7-211. Canadian Society of Agricultural Engineering, annual meeting, Regina, Sask. 20 pp.

22

Golden, A. 1987. Reassessing the use of loss·on-ignition for estimating organic matter content in noncalcareous soils. Comm. Soil Sci. Plant Anal. 18: 1111-1116. Golden, A. and Lavkullch, L.M. 1990. Effects of historical land clearing on organic matter and nitrogen leve!. in soils of the Fmser Lowland of British Columbia, Canada and Washington, U.S.A. Can. J. Soil Sci. 70: '83-'92. Jackson. M.L. 1965. Soil chemical analysis. Prentice-Hall, Inc., Englewood Cliffs, N.J. 498 pp. Keeney, D.R. and D.W. Nelson. 1982. Nitrogen -- Inorganic forms. In A.L. Page et al (eds.) Methods of soil analysis Part 2 Chemical and microbiological properties. Second edition. Agronomy Series no. 9. Am. Soc. Agron., Madison, Wisc. pp. 643-69R Kitchen, N.R., Sherrod, L.A., Wood, C.W., Peterson, G.A. and Westfall, D.G. 1990. Nitrogen contamination of soils from sampling bags. Agron J. 82: 354-356. Kowalenko, C.G. 1987a. An evaluation of nitrogen use in British Columbia agriculture. Agriculture Canada Research Branch Technical Bulletin 1987-3E. Kowalenko, C.G. 1987h. The dynamics of inorganic nitrogen in a Fraser Valley soil with and without spring and fall ammonium nitrate applications. Can. J. Soil Sci. 67: 367-382. Kowalenko, C.G. 1989. The fate of applied nitrogen in a Fraser Valley soil using I~ in field microplots. Can. J. Soil Sci. 69: 825-833. Kowalenko, C.G. 1991. Fall vs spring soil sampling for calibrating nutrient applications on individual fields. J. Production Agric. 4: 322-329. Kowalenko, C.G., Freyman, S., Bates, D.L. and Holbeck, N.E. 1989. An evaluation of th.e T-Sum method for efficient timing of spring nitrogen applications on forage production in south coastal British Columbia. Can. J. Plant Sci. 69: 1179-1192. Kowalenko, C.G. and Hall, J.W. 1987a. Nitrogen recovery in single.- and multiple-harvested directseeded broccoli trials. J. Am. Soc. Hort. Sci. 112: 4-R Kowalenko, C.G. and Hall, J.W. 1987b. Effects of nitrogen applications on direct-seeded broccoli from a single harvest adjusted for maturity. J. Am. Soc. Hart. Sci. 112: 9-13. Neufeld, J.lL 1980. Soil testing methods and interpretations. Publication no. 80-2, B. C. Min. of Agric., Victoria. 29 pp. Qasim, M. and Flowers, T.H. 1989a. Errors in the measurement of extractable soil inorganic nitrogen caused by the impurities in filter papers. Comm. Soil Sci. Plant Anal. 20: 747-75R Qasim, M. and Flowers, T.H. 1989b. Errors in the measurement of extractable soil inorganic nitrogen caused by the impurities in the extracting solution. Comm. Soil Sci. Plant Anal. 20: 1745-1752. Schreier, H. 1988. Soil survey data for land use planning: A case srudy of raspberry cultivation in British Columbia. J. Soil Water Conserv. 38: 499-503. Sparrow, S.D. and Maslak, D.T. 1987. Errors in analyses for ammonium and nitrate caused by contamination from filter papers. Soil Sci. Soil Am. J. 51: 107-110. Stevenson, D.S. and Neilsen, G.H. 1990. Nitrogen additions and losses to drainage in orchard-type irrigatedlysimeters. Can. J. Soil Sci. 70: 11-19. van Lierop, W. 1986. Soil nitrate determination using the Kelowna multiple element extractant. Comm. Soil Sci. Plant Anal. 17: 1311-1329. van Lierop, W.1989. Effect of assumptions on accuracy of analytical results and liming recommendations when testing a volume or weight of soil. Comm. Soil Sci. Plant Anal. 20: 121-137. van Ryswyk, A.L. 1985. Nitrogen: Msin nutrient for irrigated silage com. Research Highlights of Kamloops Research Station and Prince George Experimental Farm. Agriculture Canada Research Branch. pp.24-27. Weinberg, N. 1987. Improving nitrogen fertilizer recommendations for arable crops in the Lower Fraser Valley. M. Se. Thesis, The University of British Columbia, Vancouver. Wilcox, J.e. and Walker, J. 1946. Some metors affecting apple yields in the Okanagan Valley. IV. Organic matter content of soil. Sci. Agric. 26: 460-467. Zebarth, B.J., Freyman, S. and Kowalenko, C.G. 1991. Influence of nitrogen fertilization on cabbage yield, head nitrogen content and extractable soil inorganic nitrogen at harvest. Can. J. Plant Sci. 71: 12751280.

23

PHOSPHORUS N.A. Gough and C.G. Kowalenko Soils Specialist, Resource Management Branch, B.C. Min. Agric., Fisheries and Food, Kelowna, and Research Scientist, Agriculture Canada Research Station, Agassiz, respectively

RECOMMENDATION: Kelowna (a mixture ofacctic acid andNH4F) or Bray PI (a mixture of HCl and NH4F) extractions preferred because of more extensive Idstorical use for British Columbia fertlllzer recommendations. The Bmy PI extractant does not work well for calcareous soil containing free carbonate, whereas the Kelowna can be used on acidic and calcareous soils. The Kelowna extractant can also be used for extracting other nulrients. Olsen (bIcarbonate) extraction has shown potential. can be used on calcareous and acidic soils, buJ has had little use for British Columbia fertlllzer recommendations. Inductively coupled argon plasma atomic emission spectrophotometry (ICAP-AES) for phosphorus quantification In the extracts preferred because of multi-element capabutty buJ colorimetric methods are satisfactory if compatible with the extractant. Colorhnetrlc methods are assumed to measure only Inorganic phosphorus whereas ICAP will measure total phosphorus. When reporting results, the extractant type and quantification method should be specified.

Basis for recommendation. Bertrand (1981) swnmarized the research that led to the adoption of the Bray PI extraction (Neufeld 1980) foi soil testing 6fphosphorus. The initial work (laboratory, greenhouse and field) was done on soils of the Lower Fraser Valley supplemented with extrapolation of research conducted in other areas of the world and supplemented by general observations. John et al (1967) reported results of a survey of 192 alfalfa fields in the southem interior of the province that compared the suitability of nine extractants to predict P aVailability. The Bray PI (0.03 N NH4F in 0.025 N HCI) and Olsen (0.5 M NaHC03l extractants were equally suitable and belter than the other extractants tested for predicting P availability. Linear correlation coefficients (r) for the soil test P values versus plant P concentration were 0.63 and 0.66, respectively. It should be noted that this was a field survey, therefore climatic factors were not controlled as would be the case for a greenhouse study. In addition, yield response (the main objective for optimum fertilizer recommendations) was not measured. The pH of the soils examined in the survey study by John et al (1967) ranged from 5.35 to 8.15. Multiple correlation analyses found that the Olsen method was influenced less in a negative way by soil pH for predicting available P than the other methods tested. Due to the interference by free carbonate in calcareous soils on the Bray PI (Yee and Broersma 1987), the Soil Test laboratory briefly used the Olsen extraction on soils having a pH 7.5 or greater, until the Kelowna (0.25 N acetic acid plus 0.01 N NH4F) extractant was adopted (van Lierop 1985). The change from Bray PI (and Olsen method for soils of pH >7.5) to Kelowna extraction occurred at about the same time that an ICAP-AES was acquired for the provincial Soil Test laboratory (van Lierop 1985). The use of Kelowna extractant together with the ICAP-AES quantification simplified laboratory activities by providing multiple element extraction and detennination. The adoption of the Kelowna extractant for P soil testing was based on a relatively thorough study where relationships were established among P extracted by various solutions from numerous samples representing British Columbia soils (van Lierop 1985, 1988, 1989). Regression equations among the three methods were reported as follows: Bray PI = 22.8 + 1.92 (Olsen) (r = 0.91; n = 60; pH = 3.6 - 8.5; van Lierop 1985) Bray PI = 1.95 + 2.44 (Olsen) (r = 0.98; n = 40; pH = 4.2 - 6.9; van Lierop 1988) Bray PI = 17 + 0.92 (Kelowna) (r = 0.93; n = 300; pH = 3.6 - 7.0; van Lierop 1985) Bray PI = -5.4 + 1.03 (Kelowna) (r = 0.99; n = 40; pH = 4.2 - 6.9; van Lierop 1988) Olsen = 0.42 (Kelowna) (r = 0.96; n = 60; pH ~ 3.6 - 8.5; van Lierop 1985) Olsen = 4.6 (or 1.6 in text) + 0.39 (Kelowna) (r = 0.98; n = 78; pH = 4.2 - 9.2; van Lierop 1988).

24

These regressions show that, although there was excellent correlation between various extractants, they do not extract the same amount ofP. This shows that the values measured should be regarded as indexes of available P and do not provide quantitative estimates of the amount of available P. It should be noted also that the above oomparisons were all made with P quantification by ICAP-AES (van Lierop 1985, 1988) which measures lotal P, whereas the laboratory had previously used a colorimetric method (Neufeld 1980) which would measure inorganic P in the extract. There do not appear to be reports comparing ICAP-AES with colorimetric quantifications in each of the extracts of British Columbia soils. It was apparently concluded that the Kelowna extractant with ICAP-AES measurement generally extracts slightly more P from soils than does the Bray PI extractant with colorimetric measurement when soil test ratings by the two methods are compared (Table I).

Table 1. Comparison of P values by Bray PI extractant/colorimetric measurement (Neufeld 1980) and Kelowna extractantlICAP-AES (van Lierop unpublished data) to general soil test rating of availability for Lower Mainland soils and growth of barley.

Rating LL MM My He H H+

BrayPII colorimetric UlIl m )

Kelownal ICAP-AES

5

10 15 25 30 40 50 75 100

10 IS 20 30 40 50 70

CW:ml)

In the late 1980's, a greenhouse study using British Columbia and Alberta soils evaluated various P extractants for their suitability for measuring available P (Yee and Broersma 1989). The study included 48 soils, with pH ranging from 5.2 to 8.4 and barley as the test crop. Linear correlations of barley dry matter yield after 59 days growth when 50% of the plants Were at early heading, P concentration and accumulation in the plant with soil P extraction values (Table 2), showed that Kelowna and Olsen were equally able to predict P availability, and superior to Bray PI for most soil types. Probably because of low buffering capability by the HCI acid, Bray PI extractant was not suitable for calcareous soils.

Table 2. Simple linear correlation coefficients (r) of barley yield, P concentrations and P accumulations with P extracted from soil by three extractants in • greenhouse trial on British Columbia and Albert. soils (Yee and Broersm. 1989)

Plant measurement Dry matter yield P concentration P accumulation

Kelowna 0.10 0.15 0.77

Soil extractant Olsen 0.10 0.75 0.78

Bray PI 0.65 0.65 0.68

About the same time as the study by Yee and Broersma (1989), van Lierop and Tran (1990) conducted a similar greenhouse P extraction correlation study on 41 Quebec soils having a pH range of 6.4 to 7.9. The crop used in that study was ryegrass, and yield and P measurements of three cuts (3, 5 and 8 weeks after seeding) were combined for comparison to P extracted by a variety of solutions. The objective

25

of the study was to evaluate Kelowna and EDTA- and DTPA-modified multiple-element extractants, for comparisons with Bray PI, Olsen and mixed acids-salts (Mehlich II and III) extractions. Phosphorus was quantified by ICAP-AES in Kelowna extracts but colorimetrically in the other extracts. In most cases, curvilinear (logarithmic, cubic) mther than linear relationships between plant measurements and soil extractable P were observed. Each of the extractants differed in their correlations with plant measurements, depending on whether relative yield, or tissue P concentration or uptake was considered. By considering all of the relationships with plant parameters, it was concluded that Kelowna extractant was most precise followed, in order, by EDTA-modified Kelowna, Bray PI and Mehlich 11& llI. The Olsen extractant had the least precise fit. Soon (1990) conducted a greenhouse P-extractionlbarley correlation study and supplemented it with data for 1971-1972 field NPK fer1ilizer trials on soils of the Peace River area. Available soil P measurements included Kelowna, Olsen and Miller-Axley (NH4F and H2S04l extractions. The greenhouse part of the study involved 17 samples representing major sub-groups having a pH range from 5.1 to 7.3. Phosphorus was quantified calorimetrically in each of the extracts. These three extractants measured available P in the soils reasonably well, but Kelowna extractant was preferred because it had a slightly better correlation and potential for other nutrient testing. Also, the Kelowna solution extracted "somewhat more P than the Olsen and modified Miller-Axley methods within a shorter extraction period". Field research reports in British Columbia that have directly or indirectly included an evaluation of P extractants in relation to crop response in the last 10 to 20 years are relatively few. Bray PI and P2 were used as extractants to document available P in different soil types in a soil survey of the lower portion of the Fraser Valley (Luttrnerding 1981). Bray PI appeared to have been a good extractant for detecting available P accumulation in soils of the south coast that had histories of heavy manure applications (Bomke and Lavkulich 1975) and high rates of P fer1ilizer application (Kowalenko et alI980), but variable effectiveness for filberts as judged by corre~1ions with leaf tissue P concentrations (Kowalenko 1984, Kowalenko and Maas 1982). Mehlich II extractable P appeared to differentiate soils of different parent material and management histories in the Fraser Lowland of British Columbia and Washington State (Goldin and Lavkulich 1988a), but further differentiation of management effects was probably limited by the relatively high variability of the P values for the samples of the study (Goldin and Lavkulich 1988b). Gough (see Appendix XI) has conducted a series of soil test calibration trials between 1984 and 1989 with alfalfa in the southern interior of British Columbia. From the data of those trials, a critical range was set at 31 - 45 )Jgirul with Kelowna extactant since yield responses to fertilizer P were obtained at soil P concentrations of7.7, 15.0 and 29.0)Jgirnl, but no suitable trials were conducted on soil having greater than 29 ~tg Plm!. The Kelowna extractant was also tested by van Ryswyk (1985) for the development of fer1ilizer P recommendations for irrigated corn. As there was no significant response to applied P at the lowest soil P concentration (10 )Jgirul) and tissue P concentrations at the various soil test P concentrations were not given, the suitability of the extractant for predicting soil P availability could not be established.

References. Bomke, A.A. and LavkuHch, L.M. 1975. Composition of poultry manure and effect of heavy applications on soil chemical properties and plant nutrition, British Columbia, Canada. In Managing livestock wastes. PROC-275, Am. Soc. Agric. Engineers, St. Joseph, Michigan. pp.614-617. Bertrand, R.A. 1981. Development of soil and tissue testing in B.C. In Soil fertility evaluation and research. 7th British Columbia Soil Science Workshop Report. B.C. Depart. ofAgric., Victoria. pp.3-28. Goldin, A. and Lavkullch, L.M. 1988a. Historical land clearing in the Fraser Lowland of British Columbia and Washiagton State: I. Effects on soil genesis. Soil Sci. Soc. Am. J. 52: 467-473. Goldin, A. and Lavkullch, L.M. 1988b. Historical land clearing in the Fraser Lowland of Blitish Columbia and Washington State: II. Effects on soil variability. Soil Sci. Soc. Am. J. 52: 473-477. John, M.K., van Ryswyk, A.L. and Mason, J.L. 1967. Effect of soil order, pH, texture, and organic matter on the correlation between phosphorus in alfhlfa and soil-test values. Can. J. Soil Sci. 47: 157-161. Kowalenko, C.G. 1984. Derivation of nutrient requirements of filberts using orchard surveys. Can. J. Soil Sci. 64: 115-123. Kowalenko, C.G. and Mass, E.F. 1982. Some effects offer1ilizer and lime application to filbert orchards in the Fraser Valley of British Columbia. Can. J. Soil Sci. 62: 71-77.

26

Kowalenko, C.G., Maa., E.F. and van Laerhoven, C.l. 1980. Residual effects oflugh rates oflirnestone, P, K, and Mg applications: Evidence of induced Mn and Zn deficiency in oats. Can. J. Soil Sci. 60: 757-761. Lnttmerdlng, H.A. 1981. Soils of the Langley - Vancouver map area. Volume 6 Technical Data -- Soil profile descriptions and analytical data. RAB Bulletin 18. Resource Analysis Branch, B.C. Min. Environ., Kelowna. 34' pp. Neufeld, J.H. 1980. Soil testing methods and interpretations. Publication no. 80-2, B. C. Min. of Agric., Kelowna. 29 pp. Soon, Y.K. 1990. Comparison of parameters of soil phosphate availability for the northwestern Canadian prairie. Can. J. Soil Sci. 70: 227-237. van Lierop, W. 1985. Comparison oflaboratory methods for evaluating plant-available soil phosphorus. In The role of soil analysis in resource management. Proceedings of the Ninth British Columbia Soil Sci. Workshop. MOE Technical Report 16. B.C. Min. of Environ., Victoria. pp.90-95. van Lierop, W. 1988. Determination of available phosphorus in acid and calcareous soils with the Kelowna multiple-element extractant Soil Sci. 146: 284-291. van Lierop, W. 1989. Effect of EDTA and DTPA on available-P extraction with the Kelowna multiple element extractant. Can. J. Soil Sci. 69: 191-197. van Lierop, W. and TraD, T.S. 1990. Relationship between crop response and available phosphorus by the Kelowna and EDTA- and DTPA-modified multiple-element extractants. Soil Sci. 149: 331-338. van Ryswyk, A.L. 1985. Tentative fertilizer recommendations for irrigated com in B.C. southern interior. In Proceedings of North Okanagan - Shuswap Soil Seminar, 28 March 1985, Enderby. Yee, A. and Broer.ma, K. 1989. An evaluation of multi-element soil tests. Progress Report. Agriculture . Canada, Prince George. 15 pp. Y 00, A.R. and Broersma, K. 1987. The Bray, Mehlich and Kelownll. soil P tests as affected by soil carbonates. Can. J. Soil Sci. 6~:·329-404.

27

POTASSIUM, MAGNESIUM AND CALCIUM N.A. Gough and C.G. Kowalenko Soils Specialist, Resource Management Branch, B.C. Min. Agric., Fisheries and Food, Kelowna, and Research Scientist, Agriculture Canada Research Station, Agassiz, respectively

RECOMMENDATION: Neutral nonnal ammonium acetate or Kclowna (mix of acetic acid and ammonimu fluoride) extractions are acceptable for K, Mg and Ca. Comparisons on British Columbia soils show that potasslmn extracted by tbe two extractions are well correlated bnt tbe Kelowna solutton extracts abont 20% less tban tbe ammonlmn acetate solution. Recommendations must take this Into consideration. Research data on British Colmnbia soils for Mg and Ca extractant Is Ihnlted. Calcium extraction values are used only as background Information and not directly for ferlillzer Ca recommendations. Magnesium values are used directly for Mg recommendations, sometimes adjusted by ratios with Ca or K soil extractable values. Potassium, magneslmn and calcimn In tbe extracts can be satisfactorily quantified by flame emission, atomic absorption or plasma emission metbods.

Basis for potassimn metbod recommeudatlon. Local research on which original potassium fertilizer recommendations based on soil analyses for British Columbia crops was conducted at about the same time as the work done on phosphorus, but does not appear to have been as exten~ive as for P (Bertrand 1981). Neutral normal ammonium was adopted for routine K soil testing for British' Columbia (Neufeld 1980). This extraction is used extensively for potassium soil testing (Habey et al 1990). It is assumed that ammonium acetate extracts all solution and a large proportion of exchangeable K from soils, and gives a good index of crop available K. Ammonium acetate has been used extensively on British Columbia soils to measure exchangeable K in relation to soil survey (e.g. Luttrnerding 1981), and probably while determining exchangeable cations of soils (e.g. John 1971b, 1972, 1974, John and Gardner 1971, John et al 1972, 1977), but this data was not interpreted in relation to plant response and often the values for K were not fully documented. With the introduction of an inductively coupled argon plasma atomic emission spectrograph (ICAP-AES) at the British Columbia Soil Test laboratory and its multi-element analysis capability (van Lierop 1985), the Kelowna multiple nutrient extractant (a mix of acetic acid and ammonium fluoride) replaced ammonium acetate in the 1980's. A study that included 60 British Columbia, 23 Alberta, 12 Quebec and 5 Ontario soil samples showed that Kelowna and ammor6um acetate extractions were closely correlated but that the Kelowna extraction extracted about 20% less K than did ammonium acetate (van Lierop and Gough 1989). Regression equations were similar for soils with pH ranging from 4.1 and 6.9, i.e. ammor6um acetate = 8 + 1.17 (Kelowna) (r = 0.98), and from pH 7.0 to 9.6, i.e. ammonium acetate = -I + 1.19 (Kelowna) (r = 0.98). Subsequent to the adoption of the Kelowna extractant for measuring the K status of British Columbia soils, Yee and Broersma (1989) conducted a greenhouse study with 48 soil samples from British Columbia and Alberta tbat compared Kelowna and ammonium acetate solutions as extractants for available K. They found that Kelowna and ammor6um acetate extractable K was linearly correlated with barley K concentrations (r = O.77and 0.66, respectively) and barley K accumulation (r = 0.66 and 0.57, respectively), but poorly correlated with plant dry matter production (r = 0.11 and 0.12, respectively). There have been a limited number of research studies that have reported soil potassium data in relation to crop availability, and most of them have involved the use of ammonium acetate. Bomke and Lavkulich (1975) observed that ammor6um acetate extractable K from soils of fields having histories of abundant manure applications was considerably higher than similar fields that did not. Ammonium acetate extractable K was reasonably well correlated with raspberry leaf K concentration (Kowalenko 1981), but filbert leaf K concentration correlation to this soil extraction was not consistent from year to year

28

(Kowalenko 1984, Kowalenko and Maas 1982). Goldin and Lavkulich (1988a) found that Mehlich II (a combination of ammoniwn fluoride, ammoniwn chloride, and hydrochloric and acetic acids) extractable K varied according to the type of parent material and management histories in soils cleared for different lengths for agricultural use, but variability was fairly high for the samples used for the study (Goldin and Lavkulich 1988b). All tbese studies were conducted under soutb coastal soil and crop conditions, and tbere are fewer docwnented studies for interior conditions. Broersma and van Ryswyk (1979) suggested that the high available K extracted by ammoniwn acetate from a Kamloops Station soil together with potassiwn fertilizer could have been a major factor in contributing to magnesium deficiency in irrigated forage com. Altbough Neilsen and Edwards (1982) did not find "direct positive plant-soil relationships" for K in an Okanagan study, the ammoniwn acetate extractant seemed to be useful to show that these soils contained more Ca and Mg than K and may have caused a poor balance of cations for apple production. Ammoniwn acetate extraction did not detect soil K effects to vegetation cover and nitrogen fertilizer rate treatments in an apple tree Iysimeter trial (Neilsen and Stevenson 1983). Neilsen et al (1989) found that both ammoniwn acetate and Kelowna extractants were useful for determining K deficiency for Okanagan and Creston valley soils, but the data for this study was quite limited. The Kelowna extractant was shown to be suitable for predicitng the availablity of K to alfalfa in a field trial tbat was inititated in 1985 and terminated in 1988 (Gough, see Appendix XI). Residual soil test K in the spring of 1986 was directly related to yield and tissue K concentration of 1985. In a study to develop fertilizer recommendations for irrigated silage com for the interior of the province, van Ryswyk (1985) did not present soil K data that would have evaluated the suitability of the Kelowna extractant to predict K availability. Recently, Parchomchuck et al (1993) found that the Kelowna extractant was able to detect leaching of K by drip fertigation of a gravelly sandy loam planted to McIntosh apples in the Okanagan Valley. Ratios of ammoniwn acetate extractable K and Mg have been used to modify Mg fertilizer recommendations (Neufeld 1980), but local research on which this was adopted has not been well docwnented (Bertrand 1981). -iM()st of the local data for incorporation of Mg:K ratios seems to be from studies in the Okanagan. Woodbridge (1955) speculated that a high amount of exchangeable K (the method of measurement was not stated) in an Okanagan soil made treatment of Mg deficiency of apple trees difficult. Mason (1964) showed, in a pot study, that seedling leaf Mg concentration but not dry weights were influenced by Mg:K ratios when a coarse texture Okanagan soil was acid leached and then reconstituted with various cations. Neilsen and Edwards (1982) concluded, from leaf Ca and K correlations, that soil cations should be suitably balance for proper apple production. Basis for magnesium method recommendation. By 1980 neutral normal ammoniwn acetate was used as a soil test for magnesiwn (Neufeld 1980), but the data upon which this was adopted has not been well docwnented (Bertrand 1981). Since mid-1980, witb the multi-element analysis capability of tbe ICAP-AES, tbe Kelowna extractant replaced the ammoniwn acetate extractant for magnesiwn, but there is no docwnentation upon which this change was made (van Lierop 1985). In a greenhouse soil test evaluation study by Yee and Broersma (1989), soil Mg was measured in several extractants including anrrnoniwn acetate and Kelowna. Significant simple linear correlations were obtained between extractable Mg by tbe Kelowna and ammoniwn acetate, and barley Mg concentrations (r = 0.66 and 0.62, respectively). It should be noted that the Mg measurements were incidental to the study which included treatments ofP, K, and S, therefore the Mg status of the soils was not evaluated. It is asswned that the ammoniwn acetate and Kelowna extractions of Mg would be closely related and, as for K, would extract all of tbe soluble and most of the exchangeable Mg present in soils (Rabey et alI990). A few field studies have been reported for British Colwnbia soil and crop conditions that include some data in regard to measurements of available soil Mg, most of which have used ammoniwn acetate extraction. Bomke and Lavkulich (1975) showed that at least one of four Fraser Valley fields tbat received high application rates of manure contained significantly more ammoniwn acetate extractable Mg than fields that had not received high rates of manure, and Kowalenko et al (1980) found that the same extractant detected residual Mg in one Fraser Valley field five growing seasons after a high Mg fertilizer application. Ammonium acetate extractable soil Mg and leaf tissue Mg was found to be correlated in a raspberry study all one field near Abbotsford (Kowalenko 1981), but variable (Significant and non-significant) correlations oftlris type were fOlmd in fertilizer trials (Kowalenko and Maas 1982) and an orchard survey (Kowalenko

29

1984) on filberts at various locations in the Chilliwack area. Goldin and Lavkulich (1988a) found that Mehlich II detected differences in extractable Mg in soils of different land use and parent material in the Fraser Lowland of British Colwnbia and Washington State, despite considerable variability at the scale used in the study (Goldin and Lavkulich (I 988b). For a field plot at Kam100ps Station where magnesium deficiency in forage com was doewnented, Broersma and van Ryswyk (1979) reported that although soiltest Mg was relatively high (in excess of 1000 kglha), a cation imbalance in the soil probably contributed to the Mg deficiency. The method of Mg extraction was not docwnented but ammoniwn acetate was being used for soil-testing at that time (Neufeld 1980). Neilsen and Edwards (1982) found that there was a good correlation between anunoniwn acetate extractable soil Mg and apple leaf tissue Mg, particularly when soil Mg was expressed as a percentage of exchangeable bases. Ammoniwn acetate extraction detected changes in soil Mg having vegetation cover and nitrogen fertilizer rate treatulents in an apple tree lysimeter trial (Neilsen and Stevenson 1983) and in a calciwn application field trial in an Okanagan apple orchard (Neilsen et al 1981). Kelowna soil extraction also appeared to be useful for determining the Mg status of interior British Colwnbia fields growing apples (Neilsen et al 1989). Similar to that found for K, the Kelowna extractant was able to detect leaching of Mg by drip fertigation of a gravelly sandy loam planted to McIntosh apples in the Okana"oan Valley (Parchomchuck et al1993). As for K, ammoniwn acetate has been used extensively on British Colwnbia soils to measure exchangeable Mg in relation to soil survey (e.g. Luttmerding 1981), and determining exchangeable cations of soils for other purposes (e.g. John 1971b, 1972, 1974; John and Gardner 1971; John et alI972, 1977), but were not interpreted in relation to plant response and often the values for Mg were not fully docwnented. Penney et al (1977) used 1 N KCl to determine exchangeable Mg in a study of acid soils of Alberta and northeastem British Colwnbia, but only percentage base saturation values were docwnented in the report. Basis for calcium method recommendation. Considerably less data has been docwnented on soil extractable Ca for soil test purposes than for K and Mg, despite the use of ammoniwn acetate extraction to the mid 1980's (Neufeld 1980). In his review of the development of soil testing in British Colwnbia, Bertrand (1981) did not present any data nor make any comments on Ca as a soil test. It is probable that some measurements of extractable Ca were made, since anunoniwn acetate was regularly used to determine exchangeable Ca for soil survey (e.g. Luttmerding 1981), and for purposes other than development of a soil test (e.g. John 1971b, 1972, 1974, John and Gardner 1971, Jolm et al1972, 1977), but have not been interpreted for Ca soil testing. Penney et al (1977), however, used 1 N KCl to extract exchangeable Ca in a study of acid soils of Alberta and the Peace River area of British Colwnbia. Correlations between soil extractable (carbon dioxide bubbled through a soil:water mixture) Ca and apple tree vigor and yield in the Okanagan were considered inconclusive since many other contributing factors, such as moisture holding capacity, are closely related to soil ea (Wilson 1949). John (1971 a) concluded that Ca was not deficient in British Colwnbia soils, especially if they were adequately limed. He did admit that information on which this conclusion was based was incomplete. It is probable that it was asswned that Ca, as a nutrient, would not be deficient in soils of high pH and that any problems with acidic soils would be automatically alleviated by proper liming, therefore the anunoniwn Ca extraction value was used only to modify Mg recommendations (Neufeld 1980). It is generally asswned that Ca is not deficient in soils that are managed to have a suitable pH for agricultural crop production (Habey et alI990). Lime recommendations in 1980 were based on pH measurements and more recently by a buffer pH test. With the availability of multi-element analysis capability by ICAP-AES in the early 1980's (van Lierop 1985) and the advantages of a multi-nutrient extraction, the Kelowna extractant was adopted for Ca analysis. It is not known whether any comparisons were made, similar to that for K, fur Ca extract by anunoniwn acetate and Kelowna solutions (van Lierop and Gough 1989). It is probable that the two extractants would extract proportionately similar but not necessarily the same amounts of Ca since the Kelowna extractant also contains anunoniwn that could displace exchangeable Ca (Habey et alI990). The presence of acetic acid in the Kelowna extractant would probably dissolve some calciwn carbonate that would be present in calcareous or recently limed soils. Yee and Broersma il989) found that a linear correlation between soil extractable Ca and barley Ca concentration was slightly better using ammonium acetate (r ~ 0.59) than using Kelowna (0.48) in a greenhouse study on British Colwnbia and Alberta soils with pH ranging from 5.2 to 8.4. However, the Ca was not a primary nutrient for which the study was

30

designed, therefore it is not known whether the soils included a range of Ca deficient to sufficient s9i1s. Similar to Mg, most research reports for British Columbia conditions where extractable Ca was assessed in relation to "availability", have used ammonium acetate as the soil extractant. It has been shown that ammonium acetate extractable ea varied in response to histories of manUIe applications on Fraser Valley soils (Bomke and Lavkulich 1975) and could also detect residual Ca from limestone applied five growing seasons previously at Agassiz (Kowalenko et aI 1980). A correlation between raspberry leaf Ca concentration and ammonium acetate extractable soil Ca in a K and Mg fertilizer trial at Abbotsford was not considered to be useful despite measurements being made because the Ca concentration in the leaf was not consistently stable for a meaningful correlation (Kowalenko 1981). Correlations between filbert leaf Ca and ammonimTI acetate extractable soil Ca were poor to non-significant in a fertilizer and lime application study (Kowalenko and Maas 1982) and in an orchard survey in the Fraser valley (Kowalenko 1984). Calcium, according to leaf concentrations, did not appear to have been deficient in the fertilizerllime study, but a number of orchards in the survey study may have been Ca deficient or at least somewhat below optimum. Similar to K and Mg, Goldin and Lavkulich (1988a) found that Mehlich II extractable soil Ca differed with land use and parent material in Fraser Lowland soils cleared at different times, even though variability of the measurements were relatively high (Goldin and Lavkulich 1988b). Under interior soil and crop conditions, Broersma and van Ryswyk (1979) used ammonimn acetate extractable soil Ca to help diagnose a potential Mg deficiency, which appeared to be influenced by the cation imbalance in the soil. Neilsen and Edwards (1982) did not detect a direct plant to soil (ammonimn acetate extraction) relationship for Ca in an Okanagan survey study, but they did conclude from leaf Ca and K correlations that soil cations should be snitably balanced for proper apple production. This supported the use of Ca:Mg ratios for Mg fertilizer recommendations (Neufeld 1980). Mason (1964), using an acid leached and reconstituted coarse textured Okanagan soil, showed that sesd1ing leaf Mg concentration and dry weights were influenced by Ca:Mg ratios. Amrnonimn acetate extraction detected a change in soil surface Ca for vegetation cover treatments in a Iysimeter growing apple trees (Neilsen and Stevenson 1983) and five years after calcium hydroxide and gypsmn were applied to an apple orchard field trial (Neilsen et aI 1981). Kelowna extraction of soil Ca was found to be responsive to fertilizer treatments and detected Ca leaching by drip fertigation in apple orchard soils in the interior (Neilsen et aI 1989, Parchomchuck et aI 1993). In the Peace River area, Hoyt and Hennig (1982) found that an ammonimn acetate extraction of soil Ca was able to detect changes in Ca in 1978 in soils that were limed in 1970. References.

Bertrand, R.A. 1981. Development of soil and tissue testing in B.C. In Soil fertility evaluation and research. 7th British Colmnbia Soil Science Workshop Report. B.C. Depart. of Agric., Victoria. pp. 3-28. Bomke, A.A. and Lavkullch, L.M. 1975. Composition of poultry manure and effect of heavy application on soil chemical properties and plant nutrition, British Colmnbia, Canada. In Managing livestock wastes. PROC-275,Am. Soc. Agric. Engineers, st. Joseph, Michigan. pp.614-617. Broersma, K. and van Ryswyk, A.L. 1979. Magnesimn deficiencies observed in forage com varieties. Can. J. Plant Sci. 59: 541-544. Goldin, A. and Lavkullch, L.M. 1988a. Historical land clearing in the Fraser Lowland of British Columbia and Washington State: I. Effects on soil genesis. Soil Sci. Soc. Am. J. 52: 467-473. Goldin. A. and Lavkullch, L.M. 1988b. Historical land e1eering in the Fraser Lowland of British Colmnbia and Washington State: II. Effects on soil variability. Soil Sci. Soc. Am. J. 52: 473-477. Habey, V.A., Russelle, M.P. and Skogley, E.O. 1990. Testing soils for potassium, calcimn and magnesium. In R.L. Westerman (ed.) Soil testing and plant analysis. Third edition. Soil Sci. Soc. Am. Book Series no. 3, Soil Sci. Soc. Am., Madison, Wise. pp. 181-227. Hoyt, P.B. and Hennig, A.M.F. 1982. Soil acidification by fertilizers and longevity of lime applications in the Peace River region. Can. J. Soil Sci. 62: 155-163. John, M.K. 197ta. Minor elements in soil. Canadex no. 531. John, M.K. 197tb. Soil properties affecting the retention of phosphorus from effluent. Can. J. Soil Sci. 51: 315-322. John, lVI.K. 1972. Factors affecting the adsorption of micro-amounts of tagged phosphorus by soils. Comm. Soil Sci. Plant Anal. 3: 197-205.

31

John, M.K. 1974. Extractable and plant-available zinc in horizons of several Fraser River aIIu.vial soils. Can. J. Soil Sci. 54: 125-132. John, M.K. and Gardner, E.IL 1971. Fonns of phosphorus in a sequence of soils developed on Fraser River alluvium. Can. J. Soil Sci. 51: 363-369. John, M.K., Case, V.W. and van Laerhoven, e. 1972. Liming of aIfalfu (Medicago sativa L.). 1. Effect on plant growth and soil properties. Plant and Soil 37: 353-361. John, M.K., Chnah, H.H. and van Laerhoven, e.J. 1977. Boron response and toxicity as alfected by soil properties and rates of boron. Soil Sci. 124: 34-39. Kowalenko, C.G. 1981. Effects of magnesium and potassium soil applications on yields and leaf nutrient concentrations of red raspberries and on soil analyses. Comm. Soil Sci. Plant Anal. 12: 795-809. Kowalenko, C.G. 1984. Derivation of nutrient requirements of filberts using orchard surveys. Can. J. Soil Sci. 64: 115-123. Kowalcnko, e.G and Maas, E.F. 1982. Some effects of fertilizer and lime application to filbert orchards in the Fraser Valley of British Columbia. Can. J. Soil Sci. 62: 71-77. Kowalenko, C.G., Maas, E.F. and van Laerhoven, C.I. 1980. Residual effects of high rates of limestone, P, K, and Mg applications: Evidence of induced Mn and Zn deficiency in oats. Can. J. Soil Sci. 60: 757-761. Luttmerding, ILA. 1981. Soils of the Langley - Vancouver map area. Volume 6 Technical Data -- Soil profile descriptions and analytical data. RAB Bulletin 18. Resource Analysis Branch, B.C. Min. Environ., Kelowna. 345 pp. Mason, J.L. 1964. Effect of exchangeable magnesium, potassium and calcium in the soil on magnesium content of apple seedlings. Proc. Am. Soc. Hart. Sci. 84: 32-38. Neilsen, G.IL and Edwards, T. 1982. Relationships between Ca, Mg, and K in soil, leaf; and fruits of Okanagan apple orchards. Can. J. Soil Sci. 62: 365-374. Neilsen, G.H. and Stevenson, D.S. 1983. Leaching of soil calcium, magnesium, and potassium in irrigated orchard Iysimeters. Soil Sci. Soc. Am. J. 47: 692-696. Neilsen, G.H., Hogue, E. and Drought, B.G. 1981. The effects of surface-applied calcium on soil and mature Spartan apple trees. Can. J. Soil Sci. 61: 295-302. Neilsen, G.H., Hoyt, P.B. and Hogue, E.J. 1989. Identification of K deficiency in British Columbia apple orchards. Can. J. Soil Sci. 69: 715·719. Neufeld, J.H. 1980. Soil testing methods and interpretations. Publication no. 80-2, B. C. Min. of Agric., Kelowna. 29 pp. Penney, D.C., Nyborg, M., Hoyt, P.B., Rice, W.A., Slemans, B. and Laverty, D.H. 1977. An assessment of the soil acidity problem in Alberta and northeastern British Columbia. Can. J. Soil Sci. 57: 157-164. van Llerop, W. 1985. Comparison of laboratory methods for evaluating plant-available soil phosphorus. In The role of soil analysis in resource management Proceedings of the Ninth British Columbia Soil Sci. Workshop. MOE Technical Report 16. B.C. Min. of Environ., Victoria. pp.90·95. van Lierop, W. and Gough, N.A. 1989. Extraction of potassium and sodium from acid and calcareous soils with the Kelownamultiple element extractant. Can. J. Soil Sci. 69: 235-242. van Ryswyk, A.L. 1985. Tentative fertilizer recommendations for irrigated com in B.C. southern interior. In Proceedings of North Okanagan - Shuswap Soil Seminar, 28 March 1985, Enderby. Wilcox, J.e. 1949. Some factors affecting apple yields in the Okanagan Valley. V. Available P, K and Ca in the soil. Sci. Agric. 29: 27-44. Woodbridge, C.G. 1955. Magnesium deficiency in apple in British Columbia. Can. J. Agric. Sci. 35: 350357. Yee, A. and Broersma, K. 1989. An evaluation of multi-element soil tests. Progress Report, Agriculture Canada, Prince George. 15 pp.

32

SULPHUR C.G. Kowalenko Research Scientist, Agriculture Canada Research Station, Agassiz

RECOMMENDATION: 1. Extraction with 0.1 M CaCI2' or any method that includes a weak solution of CaCI2. Snlphnr in the extract can be determined by hydrlodlc acid rednctlon and subsequent quantification of the sulphide produced (methylene blue or bismuth methods), by inductively coupled plasma (ICP) instrumentation or by a method (e.g. gravimetric, turbidimetric, Indirect colorimetry) based on precipitation of sulphate with barium. 2. Extraction with the Kelowna extracting solution (0.25 N acetic acId and 0.015 N ammonium fiuoride) at a 1:10 v/v soiJ.extractant ratio for 5 minutes and sulphur determined by ICP instrumentation. Interpretation of the values derived by these extractions should be used with caution because of limited support data.

The implementation of the first sulphur soil test in British Columbia (Neufeld 1980), which was a CaCl2 extraction, appears to have been based largely on research conducted in Alberta (Bertr~d 1981, Bentley et al 1955, Carson et al 1972, Nyborg 1968, Walker 1972, Walker and Doornenbal 1972, Wyatt and Doughty 1928). Neale (1974) conducted a correlation study on sulphur which included 19 field sites in the interior of the province, from Creston to Vanderhoof with alfuJfa as the.test crop. The results from the study were variable and included only one season of testing, making it difficult to make finn conclusions other than that the'6 ppm CaCl2·extractable S04·S critical value for sulphur recommendations had merit for interior British Columbia alfalfa and intensive crop production. No trials were conducted at the south coast. Bart (1969), in a study using phosphate buffer extraction solutions, showed that adsorbed sulphate-sulphur was present in south coast British Columbia soils and that adsorbed sulphate was not extracted by water (and hence with weak CaCI2l. Lowe and Eaton (Appendix XII) found, in a growth chamber study, that the highest correlation (r = 0.913) between soil extractable sulphate with sulphur uptake by barley was with water, and that including adsorbed sulphate (as occurs with a phosphate buffer extraction) did not improved the correlation very much. This study included soil samples from central and southern interior of the province and from the Fraser Valley. Kowalenko and Lowe (1975a,b) showed that sulphate-sulphur extracted by 0.15% CaCIZ (i.e. soil solution sulphate) was more closely correlated to microbial activity than with solutions (sodium acetate, phosphate buffer, sodium bicarbonate) that extracted adsorbed as well as solution sulphate. It was also shown that the correlation was much poorer when the soil was previously air dried. Air drying caused variable (increased and decreased) extraction of sulphate. Further, both Bart (1969) and Kowalenko and Lowe (1975a) produced data which showed that analytical limitations result in possibly incomplete interpretation of the results. The analytical limitation is th.t the hydriodic acid method for determining sulphate sulphur includes both organic and inorganic sulphate. Solutions that extract solution and adsorbed sulphate from soils also extract considerable organic sulphate. Little or maybe even none of the organic sulphate extracted is directly available to growing plants. The hydriodic acid reduction method for determining sulphur has a number of advantages for soil analyses, particularly over methods that are based on precipitation with barium (Beaton et al 1968). The hydriodic acid method has adequate sensitivity and is relatively free from interference by extraction solutions and constituents (organic and inorganic compounds) extracted from the soil along with sulphur. However, the method is time consuming, reagents are costly and has not been mechanized to date. The original method, which involves quantification of the sulphide produced by hydriodic acid reduction with methylene blue reagent, requires a one hour digestion/distillation prior to the colorimetric determination (Johnson and Nishita 1952). Kowalenko and Lowe (1972) showed that the hydriodic acid reduction/distill.tion step could be reduced from one hour to 20 minutes with • bismuth sulphide rather than methylene blue colorimetric finish. This reduced time was due to the elimination of a gas wash step that was possible with the bismuth method since it is less sensitive to interferences than the methylene blue

33

method. The time required for analyses was further reduced from 20 to 10 minutes by modifYing the design of the reduction/distillation apparatus (Kowalenko 1985). In 1984, the provincial laboratory adopted the Kelowna multiple-element extractant coupled with the multi-element analysis capability of an inductively coupled argon plasma atomic emission spectrophotometer (ICAP-AES) (van Lierop 1988). Data has been published on the effectiveness of this extractant/analysis combination for phosphorus, potassium, sodium and nitrate-nitrogen (van Lierop 1986, 1988, van Lierop and Gough 1989), but not for sulphur. As for phosphorus and potassium, the adoption of the Kelowna extraction with ICAP-AES analysis for sulphur was based on a correlation between a previously accepted method (CaCI2 with hydriodic acid, Neufeld 1980) and the new method on a wide range of soils (van Lierop 1985). The precise correlation data used as the basis for the change is not available, but unpublished data that was possibly used is shown in Figure I. The extractant used in the

Figure 1. Comparison of sulphur determined by inductively coupled plasma atomic emission in O.OIM CaCl2 and Kelowna extractions of 40 British Columbia soils.

illustrated comparison were Kelowna and probably 0.01 M CaCI2 with ICAP-AES measurement of sulphur for both. Although the concentration of CaCI2 in this comparison is considerably lower that 0.1 M previously used (Neufeld 1980) and ICAP-A.ES instead of hydriodic reduction method of quantification, sulphur by the two CaC12 extract systems should be very similar since theoretically only soil solution sulphate sulphur is extracted by both CaCl2 solutions. The correlation coefficient (r), which included 40 soil samples, was 0.78. The regression equation between these two methods of extraction was: CaCI2 = -1.46 + 0.56 (Kelowna). This would coincide reasonably well with the conversion of interpretations from the original (Neufeld 1980) to the Kelowna (van Lierop 1985) method of extraction and analysis (Table 1). van Lierop (1985) reported a correlation coefficient range of 0.95 and 0.99 between CaCI2 and Kelowna sulphur extraction methods. A similar comparison of sulphur extraction methods conducted by Broersma and Yee (personal communication) with 95 soil samples yielded a correlation coefficient of 0.83.

34

Table 1. Comparison of soil sulphur test recommendations used before and after 1985 for British C?lumbia soils. Rating VL

0.1 M CaCIZl

L

6

lused prior to 1985; with hydriodic acid reduction method for sulphate sulphur measurement 2used after 1985; with ICAP-AES total sulphur measurement.

In a greenhouse soil test correlation trial using 48 soils from British Columbia and Alberta, Yee and Broersma (1989) found that the Kelowna extraction with ICAP-AES analysis of sulphur compared very favorably with CaCI2 extracts in regard to correlations with barley dry matter yields and plant sulphur concentration and uptake (rable 2). The two CaCI2 methods were 0.001 M extraction with a barium based sulphate-sulphur analysis and 0.01 M CaCIZ with ICAP-AES total sulphur analysis.

Table Z. Coefficients of determination (R2) for inverse-logarithmic regressions between barley dry matter yield, sulphur concentration and sulphur uptake versus soil test sulphur values in a greenhouse correlation trial L . Soil test 0.001 M CaCIZ sulphate-S 0.01 M CaCIZ total S Kelowna total S

Dtymatter 0.29 0.33 0.29

S concentration 0.60 0.58 0.63

S uptake

0.72 0.70 0.69

1. From: Yee, A. and Broersma, K. 1989. An evaluation ofmuIti-element soil tests. Progress Report, Agriculture Canada, Prince George. 15 pp.

Preliminary sulphur analyses of soil samples (Kowalenko unpublished data) from field sulphur tertilizer response trials (Kowalenko 1984) showed that a 0.01 M CaCI2 extraction is not very promising as a soil test method for south coast British Columbia conditions (rable 3). It is evident from this comparison of CaCI2 and PO 4-buffer that the soils were capable of adsorbing sufficient sulphate from the soil solution to result in quite consistently low CaCIZ extractable values. The two extraction methods that would include adsorbed sulphate (pO 4-buffer and sodium acetate) and total sulphate-sulphur (hydriodic acid reduction directly on the soil sample) did not appear to be very promising for predicting plant available soil sulphur either. Part of the problem of poor correlation may be that the soils had been air dried and the sulphur quantification method (hydriodic acid reduction) included both organic and inorganic sulphate-sulphur. Further work is required.

35

Table 3. Comparison of soil (0-15 em depth) sulphate-sulphur analyses using hydriodic acid redduction method 1. to relative yield response to sulphur fertilizer in south coast field trials.

Site !.D. 79El 80E6 80E2 80E4 80El 80E3 80E7 80E5

Relative 2. yield response 1.74" 1.261.131.06 1.05 1.05 1.05 0.85-

0.01 M CaClZ_ 2.02.3

P04 buffer 25.3 49.5 19.3 36.7 37.9 30.8 17.2 87.0

1.5

1.0 1.3 1.9 3.3 3.0

Sodium acetate 7.1 13.6 4.9 11.7 13.1 9.2 7.0 25.0

Total S04-S

62.9 119.5 58.0 91.1 89.7 79.6 45.0 191.6

- Significant at P = 0.05 I. Kowalenko, C.G. unpublished data. 2. From: Kowalenko, C.G. 1984. Yield response offorage grass to sulphur applications on Fraser Valley soils. Proceedings of Sulphur-84. Sulphur Development Institute of Canada (SUDIC), Calgary. pp. 823827.

References. Bart, A.L. 1969. Some mctors -affecting the extraction of sulphate from selected Lower Fraser Valley and Vancouver'Island soils. M. Sc. Thesis, The University of British Columbia, Vancouver. Beaton, J.D., Burns, G.R. and Platon, J. 1968. Determination of sulphur in soils and plant material. Technical Bulletin No. 14. The Sulphur Institute, Washington, D.C. 56 pp. Bentley, C.F., Hoff, D.J. and Scott, n.B. 1955. Fertilizer studies with radioactive sulphur. II. Can. 1. Agilc. Sci. 35: 264-281. Bertrand, R.A. 1981. Development of soil and tissue testing in B.C. In Soil fertility evaluation and research. 7th British Columbia Soil Science Workshop Report. B.C. Depart. of Agric., Victoria. pp.3-28. Carson, J.A., Crepin, J.M. and Nemunis-SlugzdInis, P. 1972. A sulmte-sulfur method used to delineate the sulfur status of soils. Can. J. Soil Sci. 52: 278-281. Johnson, C.M. and Nishlta, H. 1952. Microestimation of sulphur in plant materials, soils, and irrigation waters. Anal. Chem. 24: 736-742. Kowalenko, C.G. 1894. Yield response of forage grass to sulphur applications on Fraser Valley soils. In Proceedings of Sulphur-84. Sulphur Development Institute of Canada (SUDIC), Calgary. pp. 823-827. Kowalenko, C.G. 1985. A modified apparatus for quick and versatile sulphate sulphur analysis using hydriodic acid reduction. Comm. Soil Sci. Plant Anal. 16: 289-300. Kowalenko, C.G. and Lowe, L.E. 1972. Observations on the bismuth sulfide colorimetric procedure for sulfate analysis in soil. Comm. Soil Sci. Plant Anal. 3: 79-86. Kowalenko, C.G. and Lowe, L.E. 1975a. Evaluation of several extraction methods and of a closed incubation method for soil sulfur mineralization. Can. J. Soil Sci. 55: 1-8. Kowalenko, C.G. and Lowe, L.E. 1975b. Mineralization of sulfur from four soils and its relationship to soil carbon, nitrogen and phosphorus. Can. J. Soil Sci. 55: 9-14. Neale, W.G. 1974. Sulpfur correlation project (D.A.T.E. Project #3). B.C. Department of Agriculture, Kelowna. (see Appendix XIII). Neufeld, J.H. 1980. Soil testing methods and interpretations. Publication no. 80-2, B.C. Min. of Agric., Kelowna. 29 pp. Nyborg, M. 1968. Sulphur deficiency in cereal grains. Can. J. Soil Sci. 48: 37-41. van Lierop, W. 1985. New soil sulphur interpretations. News circular, September 17, 1985. (see Appendix XIV)

36

van Lierop, W. 1986. Soil nitrate determination using the Kelowna multiple element extractant. Comm. Soil Sci. Plant Anal. 17: 1311-1329. ' van Lierop, W. 1988. Determination of available phosphorus in acid and calcareous soils with the Kelownamultiple-element extractant. Soil Sci. 146: 284-291. van L1erop, W. and Gougll, N.A. 1989. Extraction of potassium and sodium from acid and calcareous soils with the Kelowna element extractant. Can. 1. Soil Sci. 69: 235·242. Walker, D.R.1972. Soil sulfate I. Extraction and measurement. Can. J. Soil Sci. 52: 253·260. Walker, D.R. aud Doornenbal, G. 1972. Soil sulfate II. AI; an index of the sulfur available to legumes. Can. J. Soil Sci. 52: 261·266. Wyatt, F.A. and Doughty, J.L. 1928. The sulphur content of Alberta soils. Sci. AgUc. 8: 549·555. Vee, A. and Broersrna, K. 1989. An evaluation of multi· element soil tests. Progress Report, Agriculture Canada, Prince George. 15 pp.

37

BORON G.H. Neilsen Research Scientist, Agriculture Canada Research Station, Summerland RECOMMENDATION: Extract with boiling water for 5 minutes using 1:2 soil weight to water volume ratio and determine boron in tbe filtered extract with a suitable colorimetric metbod. Azometbine-H or curcumin (based on rosocyanin) colored complexes are recommended because they have been found to be suitable for British Columbia soils. Precautions regarding boron contamination from glassware, etc. should be included. Boron quantHlcation using Inductively coupled plasma atomic emission spectroscopy (ICP-AES) has potential bul must be studied further.

The methods of boron extraction and measurement currently used in British Columbia (Neufeld 1980) are based on limited field calibration research (Kowalenko and Neilsen 1992). The research has included studies on raspberries, vegetables, forages and fruit trees. The method for extracting boron is boiling the soil in water for 5 minutes using 1 part (weight) soil to 2 parts (volume) water (Neufeld 1980). The duration of boiling is critical since longer boiling times can significantly increase the quantities of boron extracted, especially that held in organic matter (Gupta 1967). Care must also be taken with the type of container used in the extraction. For example, boron can be leached from new pyrex digestion tubes causing significant boron contamination (Gestring and Soltanpour 198Ib). Alternative extractants have recently been studied (Renan and yupta 1991), but data for British Columbia soil conditions is not available. The most commclllly-ti's-ed measurement of boron in hot water extracts is a colorimetric method involving .: yellow coloured complex after reaction with azomethine-H (Neufeld 1980). The coloured complex is measured at 416 - 420 nM, has a wide sensitivity (0.2 - 10 jlg B/ml) and pH stability range, and has been tested under British Columbia soil conditions (John et al 1975). Errors can occur with this method through spectral interferences via suspended materials or dissolved (coloured) organic matter, and if the desired pH range for colour development is not achieved. Clarification of the extract with carbon black or activated charcoal and pH buffering salts are an important part of this measurement method (Wolf 1974). An alternative measurement method that has been tested and used in British Columbia involves the use of a modified curcurnin method which utilizes rosocyanin as the coloured complex (Kowalenko and Lavkulich 1976). This colorimetric measurement may be less sensitive to interference by extraction of yellow coloured organic from the soil because the quantification is at 550 nM, and the method is quite sensitive (0 - 1 jlg B/ml). Boron measurement with ICP-AES has shown potentia! (Gestring and Soltanpour 198Ia), but there is insufficient local research to apply it to British Columbia conditions. The ICP-AES method is not influenced by coloured solutions, has very few chemical interferences and is capable of multi-element measurement. The method, however, may not be adequately sensitive for the concentrations of boron that need to be measured under local soil conditions. References. Gestring, W.D. and Soltmpour, P.N. 1981a. Boron analysis in soil extracts and plant tissue by plasma emission spectroscopy. Comm. Soil Sci PlantAna!. 12: 733-742. Gestring, W.D. and Soltmpour, P.N. 1981b. Evaluation of wet and dry digestion methods for boron deterrnination in plant samples by ICP-AES' Comm. Soil Sci. Plant Anal. 12: 743-753. Gupta, V.C. 1967. A simplified method for determining hot water-soluble boron in podzol soils. Soil Sci. 103: 424-428. John, M.K., Chuah, H.H. and Neufeld, J.H. 1975. Application of improved azomethine-H method to the determination of boron in soils and plants. Anal. Letters 8: 559-568, Kowaleuko, C.G. and Lavkullch, L.M. 1976. A modified curcurnin method for boron analysis of soil extracts. Can. J. Soil Sci. 56: 537-539.

38

Kowalenko, C.G. and NeUsen, G.H. 1992.· Assessment of the need for micronutrient applications for agricultural crop production in British Columbia. Agriculture Canada Research Branch Teclmical Bulletin

1992-5E. Neufeld, J.H. 1980. Soil testing methods and interpretations. Publication no. 80-2, B. C. Min. of Agric., Kelowna. 29 pp. Renan, L. and Gupta, U.c. 1991. Extraction of soil boron for predicting its availability to plants. Comm. Soil Sci. PJantAnal. 22: 1003-1012. Wolf, B. 1974. Improvements in the azomethine-H method for the determination of boron. Comm. Soil Sci. Plant Anal. 5: 39-44.

39

ZINC, MANGANESE, COPPER AND IRON

Denise Neilsen Researoh Scientist, Agrioulture Canada Researoh Station, Summerland RECOMMENDATION: Both DTPA and 0.1 N HCI extractions are currently being used as soil tests for zinc, copper, manganese and iron in British Columbia but are largely based on research conducted outside of the province. Further research Is required to evaluated their performance for British Columbia soli and crop conditions. The DTPA test Is relatively useful for these micronutrients but may bave to been supplemented with additional information (e.g. soli pH) to achieve maximum efficiency. Multielement extractants may be useful but should at least be correlated with other extractants and preferably with plant response before adoption.

Although boron soil testing has been oonduoted for a fairly long time (Neufeld 1980), tests for zino, manganese, copper and iron have been more recent. A recent review of micronutrient research in British Columbia (Kowalenko and Neilsen 1992) shows that little research data is available for British Columbia soil and crop conditions on which soils tests for zinc, manganese, copper and iron can be evaluated. Some research with British Columbia soils was initiated on multielement extractants that would be suitable for micronutrients as well as macronutrients i.e., Kelowna extractant modified with DTPA and EDTA (van Lierop 1985, 1989, van Lierop and Tran 1990), but no data on micronutrients were published, and none of these extractants were adopted for use. Both DTPA and 0.1 N HCI are used by local soil test laboratories' for zinc, copper, manganese and iron but appear to'be based OIL extrapolation of information from research conducted outside the province. An ideal soil test should have the following attributes: an extracting solution with a strong theoretical relationship to labile phases of the element, be rapid and convenient to use in the laboratory bave a large database relating extractable nutrient levels to crop response in the field for a large number of soil types and over many years. The database tor a soil test procedure for zinc, manganese, copper and iron in British Columbia is minimal, therefore an evaluation of the effeotiveness can only be done on a theoretical basis. General information on the extractants used are as follows: Lindsay and Norvell (978) DTPA extractant Procedure Extractant: 0.005 M DTPA + 0.01 M CaCI2 buffered at pH 7.3 by 0.1 M TEA. Shake 10 g soil with 20 rnI extractant in a 125 rnI corncal flask for 2 hours on a horizontal shaker with a stroke of8 em and a speed ofl20 cycles/minute. Advantages Has a strong theoretical basis i.e. synthetic chelates form complexes with free metal ions decreasing their activity in solution which provokes replenishment by desorption and dissolution. Extracted metals thus include solution and labile forms. Buffering and the presence of Ca in the extractant prevent the release of metals by dissolution of CaC03' Widely used. Relatively successful for zinc, manganese, copper and iron. Disadvantages Has a non· equilibrium extraction which can be affected by shaking time, shaking speed, temperature and the shape of the extraction vessel ( Sims and Jackson 1991). These problems may be overcome by strict attention to standardization ofprocedures. Was developed for neutral and calcareous soils, but zinc, manganese, copper and iron become increasingly available as pH decreases. This leads to two types of error. Firstly, the buffering of the extractant at pH 7.3 may mask the effect of soil pH on availability, and secondly, soil pH may affeot the buffering of the extractant so that the extraction pH becomes unpreclictable. TIlese difficulties may be overcome by buffering the extractant at a clifferent pH. It has been demonstrated that DTPA buffered at pH

40

5.3 provided greater chelation capacity and pH control for acid and waste-contaminated soils (Norvell 1984). This difficulty may also be overcome by including soil pH measurements in predictive equations. Misuse of the DTPA soil test has been identified by O'Connor (1988) as fulling into four categories: 1. Method alteration i.e. extracting solution, sample preparation, extracting procedure, applying critical values to crops without field calibration. 2. pH consideration (previously discussed). 3. Metal loading i.e. may not be acceptable for contaminated soils. 4. Other metals i.e. may not be extendible to other heavy metals (cadmium, chromium, nickel, lead). Wear and Sommer (1948) 0.1 M HCl extractant Procedure Soil is mixed with 0.1 M HCl extractant at a 1:5 wi.lvol. ratio and shaken for 30 minutes. Advantages Simple and rapid procedure. A large data base relating this soil test to plant response exists but not for British Columbia. Disadvantages There is no sound theoretical basis for this extractant. It extracts solution, exchangeable and some mineral forms of micronutrients which may not all be available to the plant. Its use is restricted to acidic soils as it is insufficiently buffered to extract nutrients from calcareous soils. Several other soil test methods have been used for zinc, manganese, copper and iron (Martins and Lindsay 1990), some are as follows: Mehlich I: 1:5 wlv or vlv soil to solution (0.05 M HCI + 0.0125 M H2S04> shaken for 5 minutes. Mehlich 2: 1:10 vlv soil to solution (0.2 M acetic acid + 0.2 M NH4CI + 0.015 M NH4F + 0.012 M HCl) shaken for 5 minutes. Mehlich 3: similar to Mehlich 2 except that the solution consists of 0.2 M acetic acid + 0.25 M NH4N03 + 0.015 M NH4F + 0.013 M HN03 + 0.001 M EDTA). Modified Kelowna: 1:10 vlv soil to solution (0.25 M acetic acid + 0.GI5 M NH4F with 0.005 M DTPA or 0.001 M EDTA) shaken for 5 minutes. Most of these extractants (the last three of the above list) have been intended for multielement applications. Advantages Cost-effective when used for multielement purposes. Rapid. Disadvantages These extractants have a smaller database than DTPA (although a number of studies have now been done to relate Mehlich 3 to plant performance and to other extractants). They require expensive equipment (i.e. inductively coupled plasma emission spectrograph for multielement measurement) to be effective. They may be less accurate for individual nutrients, particularly for manganese. None of these methods have been tested to any great degree under British Columbia soil and crop conditions.

References. Kowalenko, C.G. and Neilsen, G.H. 1992. Assessment of the need for micronutrient applications for agricultural crop production in British Columbia. Agriculture Canada Research Branch Technical Bulletin 1992-5E. Lindsay, W.L and Norvell, W.A. 1978. Development of a DTPA soil test for zinc, iron, manganese and copper. Soil Sci. Soc. Am. J. 42: 421-428. Martins, D.C. and Lindsay, W.L. 1990. Testing soils for copper, iron, manganese, and zinc. In R.L. Westerman (ed.) Soil testing and plant analysis. Third edition. Soil Sci. Soc. Am. Book Series no. 3, Soil Sci. Soc. Am., Madison, Wisc. pp.229-264.

41

Neufeld, J.H. 1980. Soil testing methods and interpretations. Publication no. 80-2, B. C. Min. of Agric., Kelowna. 29 pp. Norvell, W.A. 1984. Comparison of chelating agents as extractants for metals in diverse soil materials. Soil Sci. Soc. Am. J. 48: 1285-1292. O'Connor, G.A. 1988. Use and misuse of the OTPA soil test. J. Environ. Qual. 17: 715-718. Sims, J.T. and Johnson. G.V. 1991. Micronutrient soil tests. In J.J. Mordvedt et al (eds.) Micronutrients in agriculture. Second edition. Soil Sci. Soc. Am. Book Series no. 4, Soil Sci. Soc. Am., Madison, Wisc. pp. 427-476. van Lierop, W. 1985. Comparison of laboratory methods for evaluating plant-available soil phosphorus. In The role of soil analysis in resource management. Proceedings of the Ninth British Columbia Soil Sci. Workshop. MOE Technical Report 16. B.C. Min. of Environ., Victoria. pp. 90-95. van Lierop, W. 1989. Effect of EDTA and OTPA on available-P extraction with the Kelowna multiple element extractant. Can. J. Soil Sci. 69: 191-197. van Lierop, W. and Iran, I.S. 1990. Relationship between crop response and available phosphorus by the Kelown. and EOTA and DTPA-modified multiple-element extractants. Soil Sci. 149: 331-338. Wear, J.I. and Sommer, A.L. 1948. Acid-extractable zinc of soils in relation to the occurrence of zinc deficiency symptoms of com: A method of analysis. Soil Sci. Soc. Am. Proc. 12: 143-144.

.;

.

42

APPENDIX I Soil and Tissue Testing Council Technical Meeting, Nov. 24, 1992. 8:30- 9:00

Welcome and general comments - Chairman (Ron Bertrand) Introductory Comments - Tam Pringle (B.C.M.A.F.F.) - Jean Crepin (Norwest Labs.)

9:00- 9:15

Review of basics of soil testing - C.G. Kowalenko

9:15- 9:45 Analysis on weight or volume of sample / air- or oven-dried basis (Speaker: T.F. Guthrie) 9:45-10:15

pH/lime requirement measurement (Speaker: R.O. Kline)

10:15-10:30

Break

10:30-11:00

salinity measurements: electrical conductivity, extractable sodium, sodium absorption ratio (Speaker: R.O. Kline)

11:00-11:30 Nitrogen measurements: nitrate, total nitrogen/organic matter (Speaker: C.G. Kowalenko) 11:30-12:00 Phosphorus measurement (Speaker: N.A. Gough) 12:00- 1.:00 'Lunch 1:00- 1:30

Potassium, magnesium and calcium measurements (Speaker: N.A. Gough)

1:30- 2:00

Sulphur measurement (Speaker: C.G. Kowalenko)

2:00- 2:30

Boron measurement (Speaker: G.H. Neilsen)

2: 30- 3: 00

Copper, zinc, manganese and iron measurements (Speaker: D. Neilsen)

3:00- 3:30

Break

3:30- 5:00 General discussion on methods General meeting of Soil and Tissue Testing Council - reports, motions, future meeting(s), ... Speakers presentations should be linrited to 10-12 nrinutes, with the remaining time left for discussion. Toward the end of each 30 minute time period, a formal decision/recommendation will be made on a(n) standard/accepted method of measurernent(s). A proceedings will be published after the meeting. I will edit/coordinate the content. Include a complete listing of research citations and as much "unpublished" data as you can find, so that we have a well documented publication for the future.

43

APPENDIX II

REQUIREMENTS FOR A SAMPLE-BASED PLANT NUTRIENT MANAGEMENT SYSTEM FOR BRITISH COLUMBIA by C.G. Kowalenko

Agriculture Canada Research Station Agassiz

Outline of a presentation at a Soil Fertility worker meeting held in Richmond on 6 Jtme 1989.

44

REQUIREMENTS FOR A SAMPLE-BASED PLANT NUTRIENT MANAGEMENT SYSTEM FOR BRITISH COLUMBIA I.

PURSUITS/ACTIVITIES of the system (what) 1.Development (i). Determination of basic principles -chemical and instrumentation methodologies -soil and plant processes and interactions (ii). Correlation (Determining the relationship between sample analysis and plant response) -empirical relationships (general mathematical relationship) -mechanistic relationship (basic biological/ chemical process relationship) (iii).Calibration (Extrapolation of specific relationships (correlations)to field scale situations) (iv). Interpretation (Development of the recommendations) This is influenced by the philosophy such as: (a).sufficiency (b).build-up and maintenance (c).basic cation saturation concept 2. Implementation (i). Field sampling (ii). Laboratory analyses (iii).Recommendation 3.Promotion (i). General education -of basic principles and concepts (ii). Specific advertising -of secific procedures, laboratories, etc. 4. Utilization Will vary with purpose ego (i). production (farmer) (ii). service/marketing (fertilizer dealer/consultant) (iii).protection (environmentalists) 5.Monitoring Data banks and general observations used for: (i). improving efficiency of the system (ii). improving quality of the system (iii).evaluating environmental implications

These all interact within the system

ego

Development Promotion

~

t

Implementation

H

t

Utilization

45

Monitoring

II. PARTICIPANTS/AGENCIES of the system (who) 1.Researchers -public and/or private 2. Laboratories -public and/or private 3.Consultants -public and/or private 4.Users -farmers, fertilizer dealers, researchers, environmentalists 5. Consumers -general benificiaries for: (i). low cost food of high quality (ii).environmental concerns All of these interact and may be involved in one or more of the pursuits/activities outlined above. This has changed with time ego 1.1960's and 1970's Researchers (government & university)

t Public laboratory

~X;OD'i.! Farmers

agents

~Fertilizer ~

E

dealers

*

General public 2.Early to mid 1980's Researchers (government, university & private)

t

Public lab. ~IPrivate lab.~Private lab.~ .....

\

t

Extension Private -:r--""""~-~ consultants agents

$

Farmer

.,

• Fertlilizer dealers

~

General public

.'

46

1

3.Since April 1988 Researchers (government, university & private)

t

IPrivate lab. ("

• Private lab. E

> .....

Extensio~~_~~~ ~~~~.rivate __-______

agents

~

~consultants

~

Falmer :-

"7!rtlilter dealers

t

General public III.PROCESS/ASSEMBLY for the system (how) There have been various committees/groups involved in an attempts to coordinate the various pursuits of the various participants 1.1920's - 1960's Field crops branch 2.1961 Soils Advisory Committee 3.Mid to late 1960's Soil Fertility Subcommittee of Soils Advisory Committee (also Western Canada Soil Test Committee, National Soil Fertility Committee .... ) 4.1970's Emphasis on environmental implications 5.Early 1980's ' First soil fertility workshop in over ten years. Soil Fertility Subcommittee changed to Soil Management Subcommittee and subsequently further divided into Soil Fertility and Soil Physics Workgroups

47

APPENDIX III

Selected pages of proceedings of "Meeting

No.5 of the British Columbia Subcommittee on Soil Testing Procedures held 22 November 1966 at the University of British Columbia" Printed in 1967 at Agassiz. British Columbia

48

-

~I.

':;:'"'"

-

to ni tro;;E.'n by alfalfa. Bl~t thi:r:k they get a res!,cnse so

Sc=e of t~e farmers S2";f go ailead and do

87.

Eu;;:hes. I ~'las more 'Horried about potash this year. so I ra."C 4 tests going fror.: 60 to 120 pounds of potash ,-lith various ~: and P combi."l.ations. There was no particular -oattern al'ld some of these sites ';;ere (lui te le':l i.7J. natash.. ~~o it is still very confusL"'1g 21'!.d yC"'J.. ·,aren' t gOing- to get ;:luch out of it until you get to the particular site you are i.7J.terested in. You can study it for that field but 80S soon as you move off it to another field you are in trouble again.

8B.

Carstea. PotassiUr:t is a tricky eleJI!ent because there is tile e:Cc!l8.."l.[;eable, slo~!ly availctble and someti!!les fi:~ed ,0tassiuE. These are different for different soils so the response is different.

89.

E'-l;;;hes. ~'!hat bothers me is that for very 10';; soil test values ';Ie have 200 and sometimes 250 pounds K20 and i-Ie can't get tr..a.t response from it. The fello~'!s across the line say they never go over 120 pounds KZO. I thinI-: ~t;e 1:.2.79 a long ~.!a.y to go before He ans\·... er all the questions that have been askec. here the~· :ge.st half hcur.

90.

77e·.;.-:-'e1d. 'ie will bO on to the next item no ..", number 4. This is pH interpretation for li~e reco~~e~dstions ~y N. L John. This has quite 2. history.

91.,

John. Before I begi."l. the paper, I '1lould IH:e to bring you ;.tIl to date . .'cs JoP..l1. Neufeld mentioned, the limi."Cg problea has a long history. T'.'!o years ago tnere \lere discussions on li::;i."l.g Fraser Valley soils in particular. .Some pecple thought we may be adding more lime thar. necessary. It '::as discussed further at the Scil Science '.!o::::.;:sho-o last yef:!" ar.d decided ,ie should look into it further because tne:-e ·,Ias a possibility \. re l,lere Gverlicing cur soils. I haven't tine to bring everybody up to date, es!,)ecially those ·,:ho are here for the first tine, so Sc:::e of t1:e disc;.tssion may not be as comlJlete for them. l.t one of the meetings at Abbotsford last year, they s~ggested ~hat there should be a write-up, at least te:::porarily, and suggestions as to hO\1 ue can lice our SC)ils. They sl.tgzested that Bob Fletchar, HUF~:h Ga::.~dr.. er cn:.,i m;yself do this ':!ri te-UT) and distribute it to the various dis trict ,:orciculturists and-agriculturists so they could use it '.tntil something better comes It"g. "hen I ste.rt~d. ·.:riti:::lg both HUGh G;"1.rdr.. er and Dob P-;Letche!" 1:2.d left, so ':lith Vern Case's help I tried to "'I!'ite some;:hing up. I Circulated this to SCeJ8 of the peo!,le '..;":10 :lre pri!!:ari:y

49

- 25 Fcr}:ing

C'"

cr

of the memcers of the Soils COGmittee.

so~~

this asyect of lining - either in extensic'n

I

di~~'t

se;::d it to all of the~ because I didn't think they 'iiculd ce too inter~sted L~ it. I had some cO~ents from seme cf these people and I tried to c;lange a bit of it. One reaSon I wG.."lted to circulate this ...'as that "'e don't .;ant -:0 prolcng this discussion too long.. I th:i.13.l-: t\·ro years is a lot of time. If there are certain objections I thL"lk they should be able to put them in black and ','-hi te and bring them today or have sent them previously. At_ the \'!orkshop Bob Fletcher shm-Ied that of about 46 . observations on pasture, hay and other crops there ',lere 26 _;-/hich shol-/ed a reduction in yield. Of the others there 11as no apparent increase or if tr~ere \-las an increase he could not say if it i-las significG.."lt. I tried to sUr'!:lf"t.rize the results which Bob Fletcher p~ese!lted .... """ +'-;1 ·)C.Len C e ji ....1_..... ""'-r.d a 1T, s~Co', ~;,,,, S o~_ "o~"s"op 1e basis. It is felt therefore that considera~ic~ of org~>1ic matter and textural class may be all tha""C is requ.ired at present. Do not be confused on the functions of organic ll!atter regard to lir: i 1'lg.. The higher the org81ic matter, ~l:e lcn":er the pH level at \"fhich the pl~"YJ.ts C~~ gro~:i~ 3i.:.t to raise the pH (if' necess3.ry) requires larger 9.DC\.:l:ts of lime.

\i'i.. th

'i'he reco!!";.[Jendecl rate of lise is for 3 to 4- years, a8S1.l£'.i.r..g that the lime ~Iill 1)e 11ell mixed '·/i.th soil. The C02rse textured soil may require ssaller but !!lore fre~~te~t app.lications of liure ~ The same applies to s3taclished past:lre crops, \"{here the efficiency of ::-.i:15% organic

matter.

*Raies above this level should be split, so that for annual crops it may be applied in consecutive years. Fer establishing perennials, insure thorough mixing of the cultivated layer. Mix half the amount into the soil and 'plow under; then mix other half into the plowed surface. For established creps, never apply more than of this rate at anyone year.

t

'"

APPENDIX VI

INCUBATION LIME REQUIREMENT TRIAL ON SIX B.C. CENTRAL INTERIOR SOILS

R. Kline Soil Specialist

July 12. 1984

Unpublished report to British Columbia Ministry of Agriculture, Fisheries and Food

75

Incubation Lime Requirement Study Objective To determine lime requirements (LR) of six Central Interior soils using an incubation method; to compare incubation LR results with the SMP buffer pH reSults; to evaluate the use of soil factors (% OM; % clay) in the determination of LR's. Materials and Methods Six representative soils were sampled in the fall of 1983 to a ZO em depth, dried and ground to pass a Z mm sieve and placed in unheated storage until early spring 1984. Incubation lime requirement tests were run using 100 gm soil and reagent grade CaC03 «400 mesh). The soils were limed at rates of 0, O.ZS, 0.33, 0.67, 1.0, 1.33, 1.67, Z.OO, and 3.00 times the amount of lime required to obtain a pH HZO of 6.5, as predicted by the SMP buffer pH method. Each treatment was replicated four times. The soils were incubated at room temperature and were wetted to field capacity three times during the 16-18 week incubation. Soils with high clay contents required,an extra two-week incubation period. The soils were crushed as finely' as possible prior to re-wetting. Upon completion of the incubation period, soil pH's were measured in distilled HZO and .01 M CaCIz, using 1:1 and J:~soil solution-rations respectively. A calomel reference electrode with ceramic junction wds used to determine soil pH. Equilibration times were ~ hour for water pH's and 1 hour for CaCIZ pH's. CaClz - pH's vary anywhere from 0.3 to 0.8 units lower than HZO - pH's depending upon the dilution rate of HZO - pH. In this study CaClz pH at Z:l was found to be approximately 0.3 units lower than HZO - pH at 1:1 ratio. The CaClZ - pH'swere found to be 0.5-0.6 units lower than a Z:l HZO:soil pH. Therefore, CaClZPH's of 5.4 and 5.9 correspond to the commonly reported HZO pH's of 6.0 and 6.5 using the 2:1 HZO : soil pH. Chemical and physical properties for the six soils were conducted by the soil laboratories of the Ministries of Agriculture and Food, and Environment. These properties are presented in Table 1. Results and Discussion The comparison of SMP-LR predictions to the incubation LR values were closely correlated (Tables Z and 3). Close correlations could be expected as incubating soils and lime is the standard method of referencing SMP-LR and other LR methods. The 1:1 HZO : soil ratio correlated better than the CaC12 Over both target pH's, 6.0 and 6.5.

. . . Z

76

- 2 -

The possibility that pH values drop with incubation was not part of this studies' objectives. However, a comparison or original p~s measured at 2:1 rates in H20 to incubated samples at 1:1 rates does not show any major decrease in pH beyond that expected from comparing different solution : soil ratios, with one exception being the Pineview soil (see Table 4). Other soil factors influence pH such as % OM and % clay. Comparisons of these factors to incubation LR as measured in H20 were made and presented in Table 5. Simple and multiple correlations of % OM and % clay were made. A comparison of the relationship 6pH (% OM) was made where ApH is represented by pH 6.5 minus original soil pH multiplied by % OM - i.e., (6.5 - pHorig) x % OM. In Table 5, it can be seen that when the original pH is used a r = 0.75 is obtained as compared to r = 0.96 when an unlimed incubated sample is used. Summary The SMP buffe·r LR method correlated closely with the incubated LR method for six Central Interior soils, whether soil pH was measured in H20 or CaC12 (r = 0.94 and 0.95 for pH H20 6.0 and· 6.5 respectively). Soil factors sud~··~s % clay [r = 0.99(pH 6.0) and 0.96(pH 6.5) % OM r = 0.8l(pH 6.0) and 0.84(pH 6.5J, and a multiple correlation of the two above factors [R = 0.99(pH 6.0) and 0.98(pH 6.5)] correlated well with incubation LR. The comparison of the ApH (% OM) function did not correlate as well as the other soil factors (r = 0.75 for pH 6.5) when original, unlimed unincubated soil pH's were used. When unlimed incubated pH values were used the correlation was good (r = 0.96 for pH 6.5). This suggests that incubating soils at warm temperatures may affect the soil acidity, and therefore raises questions about the use of incubation methods as a LR prediction tool. Field· studies have been initiated on several of the six soils used in this incubation LR study. Two were started prior to using SMP-LR predictions and four after. Not enough time has elapsed on three out of six soils to do a proper comparison of field pH's to SMP or incubated LR predictions at the time of writing this report.

77

TABLE 1 Physical & Chemical Properties of Representative Soils Soil Name

Texture Class

Clay

Silt %

Sand

OM %

Bednesti

Si

5.7

80.6

13.7

Driftwood

SiL

24.9

52.7

Fraser

SiL

6.1

Saxton

SiL

Stellako Pineview

pH (H2O)

pH SMP

3.4

5.9

6.5

22.4

5.3

5.9

6.2

58.0

35.9

1.8

5.9

6.7

11. 6

56.8

31. 6

2.6

5.4

6.5

SiL

16.7

68.0

15.3

5.3

5.4

6.4

HC

76.1

18.9

5.0

6.4

5.5

5.6

TABLE 2 Lime requirements (LR to pH 6.0 and 6.5) as predicted by the SMP-buffer pH, and incubation LR method measured in CaC12 and H20. Soil Name

SMP-LR pH 6.0

Incubation LR

pH 6.S

H2 0 pH 6.0 pH 6.5

CaC12 pH 6.0* pH 6.5*

CaC03 as ------------------- TONS/ACRE+ ------------------------Bednesti

1.4

2.0

0.9

1.4

0.6

1.7

Driftwood

2.1

3.1

1.5

2.8

1.8

5.3

Fraser

0.9

1.0

0.7

1.3

0.7

1.2

Saxton

1.0

1.8

2.1

2.9

1.4

2.2

Stellako

1.6

2.3

1.4

2.2

0.9

1.8

Pineview

4.6

6.0

6.8

9.1

4.3

6.2

*

pH 5.4 and 5.9 in CaClz is assumed to correspond respectively to pH 6.0 and 6.5 in H2O,

+

Tons/acre can be converted to meq/100 gm by dividing by 0.5 (CaC03'as T/ac + 0.5 = CaC03 as meq/100 gm soil) 78

TABLE 3 Correlation coefficients of the comparison of incubation - LR with SMP-LR with incubation LR measured in H20 and Cacl2.

Target pH

Correlation coefficient (r) 6.0

0.94 +

6.5

0.95

H2 O

6.0

*

0.95

6.5

*

0.90

CaCl2

*

pH 5.4 + 5.9 in CaCl2 is assumed to correspond respectively to pH 6.0 and 6.5 in H20

+ Correlation coefficients calculated on a Ton/acre comparison rather than a meq/IOO gm basis.

79

TABLE 4

Comparison of original and unlimed incubated pH (H20)

Soil Name

Original pH (H2O) (J:t.) ( IS cre.. :

,J!.J

Incubated pH (H2O)

,,,,,J J

(1: 1) (So,"'.

Bednesti

5.9

5.6

Driftwood

5.9

5.4

Fraser

5.9

5.7

Saxton

5.4

4.9

Stellako

5.4

5.2

Pineview

5.5

4.5

80

';o'-o[1orl)

TABLE 5 Correlation of Incubated LR (pH 6.0 and 6.5) with twO soil factors i. OM and % Clay and with a ~pH (% OM) function.

Soil Factor

Correlation Coefficient (r)

Incubated - LR pH (H2O)

% OM

6.0

0.81

% OM

6.5

0.84

% Clay

6.0

0.99

% Clay

6.5

0.96 0.99 (R)

% OM

+

% Clay

6.0

% OM

+

% Clay

6.5

~pH(orig) ~

*

(% OM)

*

pH (incub) (% OM) +

"

0.98 (R)

6.5

0.75

6.5

0.96

Represents the equation (pH 6.5 - pHorig) x (% OM); the subtraction of unlimed original soil pH

+ Represents the equation (pH 6.5 - pHincub) x (% OM); the subtraction of unlimed incubated soil pH

81

APPENDIX VII

LIMING TRIALS IN

,

BRITISH COLUMBIA5CENTRAL INTERIOR

R. Kline Soil Specialist B.C. Ministry of Agriculture and Food

\l\ ~7

Unpublished report

82

LIMING TRIALS IN BRITISH COLUMBIA'S CENTRAL INTERIOR R. Kline - \ 1~? Introduction Liming agricultural soils in Central British Columbia is not a common practise, although early reports noted that soils were moderately acid. (1). Lime trials were established at the Prince George and Smithers experimental stations during the 1940's and 1950's, and were noted to give gradual improvements to both forage and cereal crop yields (2,3). However, it was concluded that the cost of lime outweighed the benefits derived from liming. Since then, only one lime trial has been conducted. It was designed to evaluate the impact of seed placed lime on rhizobuim bacteria, and was conducted by W. Rice, Agriculture Canada, Beaverlodge Research Station in 1972 (8). Upon reviewing the existing data on lime trials in Central British Columbia, it was evident that the knowledge that existed did not shed much light on the nature of crop responses to lime, nor the life expectency of elevated soil pH from liming soils under the climatic conditions of the region. In 1982, two lime quarries were being developed in an eighty kilometer radius of Prince George, and there was an e"xpected supply of

bulk agricultural lime. This provided an impetus to gather some crop response data from the central interior in a manner that could be seriously evaluated. At the same time, lime requirement recommendations were being changed at the B.C. Ministry of Agriculture and Food's Soil, Feed and Tissue Testing laboratory. The Shoemaker, MacLean, Pratt (SMP) buffer-pH lime requirement (LR) method was being introduced, and this was a chance to evaluate the SMP-LR under field conditions (10, 11).

Experimental Field site location and description Four sites were selected for field trial establishment across central British Columbia. Two sites were chosen close to Prince George, one site near Smithers, and a fourth south east of Quesnel. Table 1 briefly describes the soil classification and location of the four field trials. Full descriptions have been made by others (5, 6, 7, 9). The site south east of Quesnel was started in May 1982, prior to the use of the SMP-buffer method for predicting lime requirements. The other three sites were established in 1983; the sites near Prince George were established in May, and the site in Smithers was established in August. Soils in Central B.C. generally are acid in the surface horizons and gradually increase in pH with depth. Soil pH in layers below Bt horizons are often 6.0 or greater. Soil pH's through the horizons were not conducted on the soils in Table 1. However, from other data and experience, the above generalization is assumed to be applicable.

83

- 2 -

TABLE 1

CENTRAL INTERIOR LIME TRIAL SOIL DESCRIPTIONS

1.

Soil Association: Bednesti - Orthic Gray Luvisol Relief: Rolling glacio-lacustrine plain Slope and elevation: 3 E; 762 meters a.s.l. Horizons: Ap 0-15 em: Silt loam Ae 15-25 em: Silt loam Bt 25-40 em: Silty clay loam BC 40-45 em: Silt loam C 45+ em: Silty clay loam Location: Nunweiler Ranch Reid Lake NW of Prince George

2.

Soil Association: Driftwood - Dark Gray Luvisol Relief: Rolling glacial till plain Slope and elevation: 3 SE; 487 meters a.s.l. Horizons: Ap 0-18 em: Loam-clay loam AB 18-25 em: Clay loam Bt 25-36 em: Sandy clay C 36+ cm: Clay loam Location: Aspencroft Ranch, Evelyn - N of Smithers

3.

Soil Association: Pineview - Gleyed Gray Luvisol Relief: Flat glacio lacustrine plain Slope and elevation: 1 N; 670 meters a.s.l. Horizons Ap 0-10 em: Clay loam Bt 10-18 cm: Clay BC 18-25 em: Clay C 25+ em: Clay Locations: Johnson Dairy, Hart Highway - N of Prince George

4.

Soil Association: Saxton-Orthic Dystric Brunisol Relief: Fluvial fan overlaying bench in a narrow valley Slope and Elevation: 3 N; 560 meters a.s.l. Horizons: Ap 0-18 cm: Sandy clay loam Bm 18-28 cm: Very fine sandy loam IC 28-46 em: Silty clay loam IIC 46-66 cm: Sandy gravels lIIC 66+ cm: Fine sandy loam Location: Bell Ranch, Quesnel River - SE of Quesnel

Site Preparation; Experimental Methods and Treatments 1. Previous Crops - All sites had been in a grass legume crop and were cultivated in the fall prior to plot establishment. Grass weeds have proven to be a problem for both the barley and alfalfa that were estabished on these sites.

2. Experimental design - Plots measuring 2.8 X 5.5. meters, were arranged with lime treatments randomized within each of four replicate blocks on the Bednesti, Driftwood and Pineview soils. The plot size for the Saxton soil was 4.7 X 6.2 meters, also arranged in a randomized block design.

84

- 3 -

3.

Lime Hydrated lime (Ca(OH)2) was applied to the plots by hand. A neutralizing value of 120% that of pure calcium carbonate (Ca C03) was assumed. This assumption was made as the Ca (OH)2 had·been stored over winter in paper bags and some conversion to Ca C03 may

have taken place. After applying the lime, plots were roto tilled twice with an eight horsepower roto tiller. Incorporation depth was approximately 10 cm. Good incorporation was achieved on the Saxton and the Bednesti soil, average on the Driftwood, and poor to fair on

the Pineview clay. Soil moisture conditions were dry when the Pineview clay Was roto tilled, so that small clods were formed with cultivation. Observations at that time tended to indicate that lime would coat the clods. Lime rates varied according to the SMP buffer-pH LR's for target pH's 6.0 and 6.5 (11). The rates applied are noted in Table 2 for the Bednesti Driftwood and Pineview soils. Lime rates for the Saxton soil were derived from the previous lime requirement method which had been established through liming trials predominately at the coastal agricultural area of British Columbia, Lower Mainland. Soils were categorized on the basis of organic matter content, texture, and

upper limit of pH found on these soils. Central Interior soils were catagorized in this fashion as well, although little research had .. been conducted for effectiveness.

The lime rate for the Saxton soil

was estimated at 4.5 tonnes/hectare, using the method in effect till 1983. The rates applied were 0, 2.24, 4.48, and.,6.67 tonnes/hectare (4). SMP Buffer-pH test was conducted in 1984 and the LR were found to be 3.0 and 4.VTonnes/ha for target pH's 6.0 and 6.5 respectively. TABLE 2 LIME RATES TO REACH TARGET pH's OF 6.0 AND 6.5 PREDICTED BY THE SMP BUFFER pH METHOD FOR 3 CENTRAL BRITISH COLUMBIA SOILS

Soil Type

Lime Requirement (SMP-LR*) Tonnes/hectare Target pH 6.0

Target pH 6.5

4.5

Bednesti

Driftwood

4.8

6.9

Pineview

12.0

15.0

'-I. I

'3·0

*

The regression equations used for these LR'g were:

Y(6.0):179.8 + 3.387X - 49.22X Y(6.5):107.3 + 0.986X - 22.27X Current regression equations estimate mineral soil LR on a volume basis

over 20 cm depth are: Y(6.0):164 + 3.129X2 - 45.17X Y(6.5)=107 + 1.189X2 - 23.55X

85

- 4 -

4.

Fertilizers Blanket fertilizer rates were applied to the lime treatments for both the barley and alfalfa crops. The fertilizer types used for all years were: 34-0-0 - ammonium nitrate 11-51-0 - mono ammonium phosphate 0-0-60 - potassium chloride (potash) Na2B407 - Borate - 68 MgS04 - magnesium sulphate The fertilizers were varied according to the rates indicated in Table 3. TABLE 3 FERTILIZER NUTRIENTS APPLIED TO LIMED BARLEY AND ALFALFA CROPS FOR 4 CENTRAL BRITISH COLUMBIA SOILS

Fertilizer Nutrients Applied (Kg/ha) Soil Type

Bednesti Driftwood Pineview

Barley* 1983 N P K S Mg B

,. 1982

Saxton

N P K S Mg

45.0 20.0

B

3.5

Alfalfa 1984

1985-86

70 105 101 30

39.0 76.6 83.0 14.0 7.0 5.6

39.0 76.6 83.0 14.0

1983*

1984

1985

45.0 20.0 67.0

39.0 74.0 83.0 14.0 7.0 5.6

39.0 74.0 83.0 14.0

2.5

* Fertilizer rates drilled in with the barley seed was 10 Kg N/ha and 21 Kg P/ha. Crops Barley (var. Klondike) was spring seeded at a rate of 100 Kg/ha, with a push type V belt single row seeder on 23 cm row spacing, for the Bednesti, Driftwood, and Pineview and Saxton sites in 1983. In 1982, barley (var. Galt) was broadcast at the rate of 100 Kg/ha, raked and packed on the Saxton site. These yields are not reported due to the extreme variation. The 1983 seeding on the Saxton soil was late (June 16), due to weed control treatments. Alfalfa (var. Peace) was seeded at all four sites in May 1983, at the rate of 15 Kg/ha with a V belt single row seeder on a 23 cm row spacing. The alfalfa was coated, and included inocculum, but inocculum was added in a dry form at seeding time.

86

- 5 -

6.

Weed control Broad leaved weeds were controlled in barley by using Embutox-E (2,4-DB) at 4.25 L/ha in June for Bednesti, Driftwood, and Pineview sites in 1983. Hand weeding of grasses was necessary as well. The Saxton site was sprayed with Round-up (glyphosate) at 5,0 L/ha in May 1983 prior to seeding barley; and the barley was sprayed in July to control broad leaved weeds. Weed control in alfalfa (1984) for all sites consisted of using Embutox-E at L/ha. Fusilade (fluazifop-butyl) was applied at 2.0 L/ha to control couchgrass (Agropyron repens) at the Pineview site in June 1985. Hand weeding for grass control was necessary at the other sites.

7.

Cultural After barley crops were harvested, plots were roto tilled in the fall once, and then in the spring twice before seeding alfalfa.

8.

Crop Measurements Barley was harvested with hand scythes at the soft dough stage by taking the center 8 rows of 12, over a length of 5.0 meters. Wet plot weights were recorded on site with a hanging scale and sub samples were collected for dry matter determination. Seedling year alfalfa was harvested with a Jari power sickle mower at a height of 10 cm over three rows by approximat~ly 4.0 meters in August 1983. In subsequent alfalfa growth was harvested with the same mower, but over 5 rows and a length of approximately 4.0 meters. Each length of the treatment plots was measured after cutting. Wet plot weights were determined at the site, and 500 g sub samples (approximate) collected for dry matter determination and tissue analysis. First growth alfalfa waS harvested usually at the end of June or beginning of July in the bud to 10% bloom stage. Second growth alfalfa is harvested from mid-September to early October. Results

Soil pH Soil pH from the 0-15 cm depth were taken each spring and autumn for 1984 and 1985. The autumn 1983 sampling was not conducted on the Driftwood site, as it has been limed in August. Some seasonal fluctuations can be seen in the soil pH's, even on the unlimed treatments, which may be due to seasonal soil moisture differences, biological activity etc. By the autumn of 1985, however, lime treatments have created a difference in pH's. The Bednesti and Pineview soils are under the target

pH of 6.5 by 0.32 and 0.15 pH units respectively, while the Driftwood soil is 0.28 pH units over the target pH of 6.5. The Saxton soil after 3.5 years of lime applications required a rate of 6.72 tonnes/hectare to stabilize at a soil pH of 6.5. This lime requirement is higher than 4.1 T/ha predicted by the SMP buffer pH. As noted, incorporation was not ideal, and this may explain some of

the differences found in the field trial data. summarized in Table 4.

87

Soil pH data is

- 6 -

Crop Yields Barley as a first crop following liming did not show significant yield increases due to liming (Table 5). This may be due in part, that barley will produce good yields when pH values are above 5.5, and when values drop below this point, it is primarily aluminum which restricts

growth. Liming recommendations for barley then attempt to bring the pH to or above 5.5. Only the Saxton and Pineview soils have pH's in this range, however, the yields did not prove statistically significantly different due to liming. The poor yields at the Saxton site are somewhat

puzzling, however, its felt that the combination of late seeding and heavy rains followed by a lengthy dry period may have resulted in nitrogen being leached below the root zone. The barley was stunted and pale green during the season. No additional nitrogen was added. Alfalfa crops were seeded in 1984, but only the Pineview site was harvested. There was an immediate growth, color, and vigor difference in alfalfa at this site. The yields in Table 6 show that the response difference was significant in 1984. In 1985, the same visual differences were present but statistically there was enough variance to not show a significant yield difference. However, there were statistically significant yield differences due to lime on the first cut of alfalfa'at the Bednesti and Saxton sites, and in the second cut on the Driftwood sites.

Discussion Despite weed control, the Pineview soil site still has extensive alsike clover and couchgrass infestation. Couchgrass is controlled with

Fusilade, but not eradicated. Alsike clover was hand weeded out but has re invaded the plot. It is particularly thick on the unlimed treatments and probably accounts for the lack of statistically different yields. However, it is also abundantly present on the limed treatments. Visual differences in alfalfa growth and health is seen at the Pineview and Saxton soils and to a lesser extent at the Bednesti site.

The soil pH at the Saxton site indicates that its liming requirement may be higher than that predicted by the SMP buffer pH method, which was similar to the amount predicted by the method in use prior to 1983 (ie. 4.1 T/ha for pH 6.5 via SMP buffer pH method v.s. 4.5 T/ha via the pre-1983 method.) Further evaluation of alfalfa response and soil pH changes are to be monitored for at least two more years. Future recommendations for

alfalfa response to lime in central B.C. will be based on these trials. Presently, it appears that the SMP buffer pH method is effective in predicting the lime requirement for central B.C. soils. The method used prior to 1983 would not have predicted the high lime requirements of the Pineview or even the Driftwood loam, although it appeared to be adequate

for the Saxton soil.

88

·7-

'--I

TABLE 4 EFFECT OF LIME TREATMENTS ON SOIL pH (H20) VALUES OVER;rHREE YEARS ON FOUR SOILS IN CENTRAL BRITISH COLUMBIA

SOIL TYPE

LIME APPLIED T/ha

DATE APPLIED

ORIGINAL SOIL pH

Bednesti - Silt Loam

0 4.5

May 1983

5.90

Driftwood

0 4.8 6.9

August

5.90

-

Loam

Pineview - Clay

0 12

Saxton

0 2.24 4.48 6.72

SOIL pH SPRING 1984

AUTUMN 1984

SPRING 1985

AUTUMN 1985

SPRING 1986

AUTUMN 1986

5.80 a 6.58 b

5.53 a 6.95 b

5.58 a 5.98 b

5.75 a 6.33 b

5.7 a 6.18 b

5.80 a 5.85 a

5.53 a 6.23 b

5.98 a 6.35 b 6.55 b

5.75 a 6.30 b 6.475 b

6.00 a 6.60 b 6.85 b

5.78 a 6.48 b 6.78 c

5.89 a 6.40 b 6.55 b

5.60 a 6.33 b 6.65 b

5.65 a 6.15 b 6.18 b

5.50 a 6.10 b 6.33 b

5.30 a 6.08 b 6.35 c

5.20 a 5.70 b 6.03 c

5.28 a 6.05 b 6.50 c

5.55 5.80 6.30 6.50

5.53 5.80 6.15 6.50

5.65 5.70 6.03 6.00

1983 May 1983

5.50

5,58 a 6.63 b 6.68 b

5.50+ 5.70 5.90

May 1982

5.30

5.63 5.93 6.58 6.60

5.40'a 5.75"b 5.98 bc 6.10 c

15 - Sandy Loam

AUTUMN 1983

a a b b

a b c d

a a b c

Means followed by a common letter are not significantly, different at the 0.05 P level by using the Student-Newman-Keuls l Test.

+ Mean of two replications.

a a a a

'" 00

8 -

TABLE 5

EFFECT OF LIME ON WHOLE CROP BARLEY

PRODUCTION FOR 4 CENTRAL BRITISH COLUMBIA SOILS

Soil

Lime Treatment T/ha

Type

Whole Barley

Mean Yield (Kg DM/ha) 1983 7663 a 8123 a

0

Bednesti

4.5 0

Drift;fJood

4.8 6.9 Pineview

0

5333 a 6499 a 6622 a

12 15 Saxton

0

618 a 615 a 720 a 732 a

2.24 4.48 6.67

Means followed by the same letter are not significantly different at the 0.05 level using the Student'Ne\

Y·0.13'0.965X r -0.985·· sy'x -0.80 10

,~

20

LR - scooped samples

Fig. 2. Relationship between the lime requirement (LR) of scooped soils, adjusted to 109, and 109 weighed samples.

119

APPENDIX X

THE RELATIONSHIP BETWEEN ELECTRICAL CONDUCTIVITY MEASURED ON A SATURATED PASTE EXTRACT AND ELECTRICAL CONDUCTIVITY MEASURED ON A 2:1 EXTRACT

SOIL SCIENCE 315 TERM PROJECT

EVELINE WOLTERSON

09512685

MARCH 1983

120

INTRODUCTION

Soluble salts are- defined as the inorganic soil constituents

appreciabl~Y

soluble (as ions) in water (Black 1957).

The

determination of their concentration is related to the ability

of

a!so~7w~ter

extract to conduct electricity. The current ,.J. '~o (generated i~ the extract is approximately proportional to the (,$",

,.1.'>/'

salt content in the soil. The term electrical conductivity (with units mhos.em- l ) is designated to define this phenomenon.

The correlation between plant growth and soluble salt concentration, or electrical conductivity, has been ... ell documented (Black 1957; Russell 1973).

Electrical conductivity

(E.C.) is used to predict seed germination, e~~rgence rates

and yield potentials of plants and crops.

As with many

diagnostic tools, methods of measurement can vary, which complica~es

interpretation.

on techniques and problems

Researchers have pUblished much encoun~ered

in assessing soil

salinity, both in the field and in the lab (Halvorson, et al. 1977; Yadav, et'al. 1979).

Generally, lab determination

requires a soil-water extraction.

The choice-;f a suitable

-2-

soil-water ratio depends upon the purpose in making the measurements, the number of samples to be handled, the time available for doing the work and the accuracy required. Traditionally, if a correlation was sought

between soluble

salt concentration and plant growth, the researcher extracted close to the soil water content at which plants grow {i.e. saturated paste

ext~action:

Black 1965; Hesse 1971}.

If monitoring soil salinity over time or if speed and efficiency were'the objective, an extraction with a higher soil-water ratio

could be made (i.e. 2:1 extraction, 5:1 extraction, etc.

Black 1965; Hesse 1971).

Applying the speed and efficiency

of the latter technique to the diagnostic

properties of the

former would make E.C. interpretations more convenient." If a correlation exists between E.C. determined by the saturated paste method (EC sat ) and E.C. determined by the 2:1 method (EC : 1 ), then one could take advantage of the assets of both 2 methods. It is the purpose of this- investigation to demonstr,ate that a correlation does exist, applies for different soils, and holds over a wide range of E.C. '5.

121

-3~~TERIALS

AND METHODS

Two soils. flooded by sea water during a storm in December, 1982, were chosen.

The flood period was approximately the

same for both sails.

They are deltaic deposits of the

Gleysolic order, with moderate to poor drainage.

One soil

(-\lestha») TS) and--r is from a Westham Island farm lying next to

an arm of the Fraser River where it meets the ocean. soil series is Hestham/Crescent. fine textured.

The

It is medium to moderately

The other soil tMua-Bayl was sampled near

Hud Bay where the Serpentine RiVer enters Boundary Bay.

It

is in the Sandel/Kittel:' soil series and has a medium texture. The samples were collected approximately one month after the salt water inundation.

They were air-dried, crushed and passed

through a 2 mm. sieve and stored in plastic bags. determined by two separate@ extractions.

E.C. was

Details on the

saturated paste method are outlined by Black (1965) and the method used forthe 2:1 extraction is given in Laboratory Methods for Soil Science 315 (Soil Fertility 1983).

The

Radiometer Type CDM2e conductivity meter and the SD-815 Solu Bridge Soil Tester were used to measure EC sat and EC 2 : 1 ,

resp~ct~vely.

-4-

The least squares linear regression analysis determined the relationship between EC sat and Ee 2 : 1 • Data was grouped according to the two soil series with a further grouping for the Mud Bay samples to allow for two ranges in E.C. values.

RESULTS

~~D

DISCUSSION

vs Ee : regression analyses are summarized sat 2 l in Figures 1-3. All correlation coefficients exceed 0.97

Results of EC

and standard errors for the three analyses are less than 0" 56. Slopes and y-intercepts for each data grouping yield three equations that relate EC sat with EC 2 : 1 " The equations for each correlation are given itLFigures 1-3. Notice that the equation for the Westham Island soil (finer texture)

has a

steeper slope than the two equations for the Mud Bay soil. This is contrary to the findings of Halvorson, et al. (1977) who established that increasing clay content decreased the slope.

Their samples ranged from 6% to 63% clay with slopes

from 12.99 to 3.06, respectively.

Because only textural

classes and not actual %c1ays are known for the soils in this study, it is difficult to say if differences in slope are due to clay content or whether or not the differences are significant.

122

-0-

Soils with a wider more clearly.

~ange

in texture would indicate this

Nevertheless, it appears possible to come up

' raa

ALfALFA

Obtective To detel1l1lne i f alfalfa will respond to the recommended rate ot zinc on a soil which h4S 0.5, 0.2 and 0.1 ppm OTPIt -Zn in the 0-15, 15-30 and 30-45 Cl!l soil depths respectively. Location:

Harper Ranch, Kamloops

Fertilizer Treatments. Zinc SUlphate was applied at 11.2 k&/ha Zn to thne plots and the three other plots of a similar size received no zinc. P. K. 8 and S were applied to all plots at rates of 336 kg P205Jha. 224 k20Jha. 4.5 k& B/ha and 22.4 kg S/ha respectively, Interim: Results The one harvest that was taken produced no significant yield difference between the averages of the two zinc treatments. Zinc concentrations in plant tissue were significantly different between the control plots and the zinc treated plots at p"'O.06. Increased tissue Zn concentrations because of Zn are likely to be statisticaUy significant in l~a9.

163

APPENDIX XII COMPARISON OF FOUR SULPHATE SULPffiJR EXTRACTANTS FOR PREDICTING AVAILABLE SOIL SULPffiJR FOR BARLEY GROwrn IN A POT STUDY

Unpublished report L.E. Lowe and G.W. Eaton University of British Columbia, Vancouver ABSTRACT

Sulphur uptake by barley in a growth chamber study was correlated with soil sulphate extracted by water and by three phosphate-containing extractants. The highest correlation coefficient (0.91) was observed for water extraction. Extractants having high phosphate concentration showed the poorest correlation with S uptake, even though they extracted the most sulphate. Water extractable sulphate accounted for 83% of the variability in S uptake, and 85% of the variation in sulphate-S content of tissue. These figures were not improved by inclusion of other variables. Inclusion of adsorbed sulphate and labile organic sulphate by extraction with concentrated phosphate solutions did not improve prediction of sulphur uptake and suggests these forms were not readily available to plants over a short growing period. INTRODUCTION

It has been shown (Ba!} 1969) that phosphate buffers of high concentration extract substantially more sulphate from some British Columbia soils, than is obtained by extraction with distilled water, especially from soils containing significant amounts of adsorbed sulphate. The extractive efficiency of the phosphate solutions was markedly pH -dependent both with respect to organic and inorganic sulphate. The ready availability of adsorbed sulphate to plants has been reported by a number of workers (Barrow 1967, Karnprath 1968, Williams and Steinbergs 1964). In these studies adsorbed sulphate has usually been defined in terms of sulphate extractable with relatively dilute phosphate solutions (0.01 M), in contrast to the study of Bart (1969) in which concentrations of up to 0.5 M were employed. The present study was undertaken to determine whether strong phosphate solutions yielded a better index of sulphur uptake by plants in a growth chamber study than extraction with water or dilute phosphate solutions, and hence detennine whether such phosphate extractants warranted further investigation in more extensive field trials. MATERIALS AND METHODS

Soils: Twenty two samples were collected from the central and southern interior of British Columbia in April 1970 and three additional samples from southwestern British Columbia were selected from those investigated by Bart (1969). All but the latter three samples were located in areas for which field responses to sulphur had previously been reported (fable I). All sites were under grass or forage crop, and elevations ranged from near sea level up to 4400 feet. Samples were air-dried and crushed to pass a 2-mm sieve. For some chemical analyses, sub-samples were further ground to pass a 60-mesh sieve. AnalytIcal methods: Total C and total S in soil were detennined with Leco Induction furnace and analyzers, total N by serni-micro-kjeJdahl, pH with g1ass electrode in 1:2.5 soil:water suspension and particle size distnbution by a hydrometer method (Black 1965). The samples represented a wide range of C content (0.87 - 6.22%), texture (7 - 80% clay), pH (4.8 -7.2) and total S content (130 - 1300 ppm) (fable 2). Extractable sulphate-S was detennined in extracts by the HI reduction-methylene blue colourimetric method of Johnson and Nishita (1952, Johnson and Ulrich 1959), which recovers both organic and inorganic sulphates. Extracts were obtained by I·hour extractions on a reciprocal shaker at a soil:extractant ratio of 1:5 using the following extractants: (1). distilled water, (2). KH2P04 solution containing 500 ppm P

164

Table I. Locations of and comments on samples used for a sulphur correlation pot study.

1

Town/citv Prinoton

Farm/source Allison

2 3

Hedley Cawston

Lawrence Indian reserve

4 5 6 7

Rock Creek Rock Creek Bridesville Bridesville

Smith Fair grounds Johansen Harfinan

8

11

Okanagan Falls Thomas Ranches Mitchell Armstrong Armstrong CDAplotof Summerland Grinrod Smaha

12 13 14

Salmon Valley Salmon Valley 136 Mile

Syme Freeze Downie

IS 16 17

Marguerite Kersley Soda Creek

McAllister Holly

18

Prince George

19

Vanderhoof

SVen Christensen C. Carpenter

20

Vanderhoof

I. Geamart

21 22

Fraser Lake Prince George

23 24 25

Van. Island Fraser Valley Fraser Valley

Poidevin Lot 1542 Nescarte(?) st. Saanichton Abbotsford Monroe

No.

9 10

Description!comments 2.5 miles E. ofjunotion with Highway 5 -- road along N. side of river (past log piles). 27.8 miles E. of Hedley -- No response. Upper bench; N.E. of Cawston on Lowe Avenue (2 stop signs). 5.1 miles N. of Rock Creek on Westbridge Road -- Response. Just across railway tracks. 1.5 miles E. of bridge, N. side of road -- Response. W. of town, Circle 2 Ranch 0.4 miles N. of highway at 4400' elevation, black -- Response. Going N.w. turn R. at ESSO station -- Response. Schubert Road 2.2 miles past "Frost"; upper levels --Response. Property adjacent to Parkinson, 0.5 mile N. of Armstrong, W. on Landsdowne to Rashdale Ar. - Response. R. at old coffee shop, between fences of cattle walk -Response. Mara clay loam -- Response. Approaching "Glenemma" on Mara loam. Past Lac Ia Hache, W. ofHillllway, Enterprise st. Rd. -Response. "7 Mile" Ranch, by rail crossing -- Response. In driveway -- Response. I mile S. of district boundary, by Gulf station, N. Horsefly turnoff Lot #2432; soil wet and near run-off creek; near hay field. IS - 27 Tp. 12; fence line; fairly dry; near original plots; near hay field. IS - 27 Tp. 12; re-broken hay field; near original plots; soil fairly moist. Lot 2013?; near fence line; soil wet; under sod. Soil dry on top and moist I inoh down; sununer fallowed, 1969. Bart (1969) M. Sc. thesis, p. 48. Bart (1969) M. Sc. thesis, p. 48. Bart (1969) M. Sc. thesis, p. 48.

(Ensminger 1954), (3)_ 0.1 N NaH2P04 in 2 N acetic acid (Bardsley and Lancaster 1960), (4). 0.5 M sodium phosphate at pH 7 (Bart 1969). Sulphate-S in plant tissue was determined by direct application of the Johnson-Nishita procedure (1952, Johnson and Ulrich 1959) and total S by the same procedure following Mg(N03l2 ashing (Chapman and Pratt 1961). Cropping conditions: The test crop was barley (Hordeum vulgare var. Vantage) and growing conditions were selected on the basis of an extensive investigation of the sulphur nutrition of this crop by Herath (1970). The seed was washed and germinated in distilled water. Seven seedlings/pot were planted in 6" (diam) x 5 112" pots at 5-cm plumule length, using a single pot for each soil. After seven days each was thinned to 4 plants/pot of uniform height. Day/night temperatures were 24/16 C with a day/night regime of 16/8 hours, employing fluorescent and incandescent light sources for the day period. The pots were cultured for 35 days watering with

165

Table 2. Selected chemical properties for soil samples used for a growth chamber correlation study' among soil sulphate extractions and barley growth.

No.

M

%C

CfN

% clay

I 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 17 18 19

6.8 7.2 7.1 6.5 5.3 5.9 5.7 6.0 5.5 5.7 5.9 6.1 7.1 6.4

11.6 6.2 7.4 18.4 15.8 10.8 13.8 21.2 17.3 16.3 14.3 10.9 10.7 12.2 14.2 11.8 10.3 20.5 11.9 15.3 14.9 9.9 15.1 15.1 9.8

13 12 19 6 10 17 12 7 13 17 27 42

5.7 6.1 5.3 5.6 5.1 5.7 4.7 5.2 4.8 4.6

2.56 1.39 0.94 1.82 2.49 3.04 6.22 1.19 3.02 2.56 2.95 5.04 4.12 4.64 1.67 5.34 2.97 1.99 3.62 0.87 1.82 4.39 2.47 2.84 1.71

27 14 26 25 24 27 9 34 80 33 20 21

0.045 0.031 0.029 0.019 0.020 0.042 0.103 0.017 0.023 0.020 0.027 0.054 0.015 0.065 0.023 0.062 0.031 0.027 0.037 0.013 0.026 0.054 0.022 0.031 0.027

5.9 0.7

2.87 1.42

13.4 3.7

22 IS

0.035 0.020

2fT

21 22 23

24 25

Av. S.D.

6A

11

%S

a nutrient solution (Meyer et al1955) containing all nutrients except sulphur. Tops were harvested at 1 ern above the soil surface, dried for 7 days at 80 C and ground in a Wiley mill. After harvesting, the soils were air dried and crushed, and the major roots removed by hand. A second crop was then grown under the same conditions as previously. On two soils, the crop failed completely, and these samples were eliminated from the analysis with suitable corrections. The remaining pots were thinned to 3 plants/pot and grown and harvested as before. Statistical methods: The results were analyzed by simple linear regression and correlation procedures and by a stepwise multiple linear regression procedure, using programs on file at the University of British Columbia computing centre. Simple correlation coefficients were determined from fresh wt. and dry wt, Suptake oftops and sulphate-S content oftops with all measurements. Values for r above OAO were significant at the 5% level, and greater than 0.50 at the 1% level. Values for non-significant correlations are not presented. RESULTS AND DISCUSSION Soil extractable sulpbate-S: Water (Extract I) and dilute phosphate (500 ppm P; Extract 2), which have been used in soil testing, generally extracted similar amounts of sulphate with mean values of9.8 and 9.3 ppm S respectively (fable 3). However, the most notable exception (sample #24) was a podzolic soil, known to be

166

Table 3, Soil extractable sulphate-S, and barley crop growth and sulphur measurements for growth,chamber correlation study,

Soil extractable S04-S (l1l1m}

NO,

I

2 3 4 5 6 7 g

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Av, S.D,

Water 9,1 10,3 42,8 15,3 8.4 8,4 10,0 5,9 3.4 3,8 7,8 9,4 13,8 10,3 1.9 6,6 8.4 9,7 8.1 5,6 5,0 21.3 9.1 3,0 7,2 9,8 8,0

0,1 N P04 Ml2P04 500 l1l1m P in HAc at l1H 7 5,6 26,9 21.3 16,6 18,8 4.4 33,1 40,0 39.4 6,9 10,6 12,5 5,0 15,6 10.3 8,4 19,1 12.5 6,6 12,8 37,2 2,8 3,4 3.4 3,4 5,9 12,2 5,0 12,2 3.1 9,4 9,7 10,9 13,4 10,6 26,8 33,8 27,8 11.6 12,9 14,1 26,9 5,6 6,6 7.5 5,6 8.8 14.4 ,,1:2 6,9 12.5 7,2 5,3 25,9 3,8 19,3 8.4 7,8 3,1 7.5 6.3 16.3 3.4 66,3 20,6 20.3 10,0 11.3 31.3 19,1 40,0 11.3 9,7 9,7 39,1 9,3 6,7

12,2 9,2

First crop (g DWt. /l1 ot) 5,20 3,65 5.10 3,40 2.35 3.35 2,85 2,80 3,75 2,55 3,95 6,70 7,10 5,60 1.88 5.35 3,95 2.55 2,40 1.80 0,60 4,25 3,00 1.75 4,15

22,6 14,0

3,60 1.58

Barley crol1 measurements Sulphur Uptake S04-S content (mg S/ content (l1l1ml ----12Q!2 (l1l1m) 1.41 8 271 13 1.27 347 6,68 1190 1310 40 1.20 353 0,48 16 206 1.16 15 346 0,09 5 33 25 206 0.58 35 393 1.47 0.58 52 227 30 255 1.01 264 1.77 25 2,36 62 332 45 205 1.15 0.51 20 272 0,20 18 37 270 1.07 30 0,48 28 190 0,48 15 200 0,40 220 13 0,08 12 133 3,54 525 833 0,95 30 318 0,81 28 465 1.67 85 403 324 257

1.26 1.36

95 250

Second crop (g D,Wtll1otl nil 0.50 1.75 nil 0,40 0,40 0,30 0,20 0.55 0,25 0,90 1,45 2,50 1.00 0,25 0,60 0.55 0,20 0,70 0,30 nil 1.90 0,20 0.55 1.10 0,66 0,64

high in hydrous oxides, For these two extractants, values for extracted sulphate-S ranged from 1,9 to 42,8 ppm with over 70% of the values below 10 ppm S, indicating that many of the samples might be expected to give responses to sulphur, Acidic 0,1 N phosphate (Extract 3) and neutral 0.5 M phosphate generally extracted somewhat higher levels with mean values of 12,2 and 22,6 ppm S respectively (Extract 4), Sulphate-S extracted by the four extractants were significantly correlated among each other except between water (Extract I) and neutral 0.5 M phosphate (Extract 4) (fable 4), Regressions between the extracts that were correlated were: Extract 1 = 2,70 + 0,68 (Extract 2) Extract I = 3.56 + 0,88 (Extract 3) Extract 2 = 3,83 + 0,89 (Extract 3) Extract 2 = 9,29 + 1.42 (Extract 4) Extract 3 = 13,2 + 0,77 (Extract 4),

167

Total uptake of S by tops: Sulphur uptake varied considerably between soils and ranged from 0.08 to 6.68 mg S/pot (fable 3). However, only for three soils (No.'s 3, 13 and 22) did uptake exceed 2 mg/pot.· Water extractable sulphate was more closely correlated with S-uptake (r = 0.91) than were any of the phosphatecontaining extractants (fable 4, Figure 1). More concentrated phosphate extractants showed the lowest

Table 4. Significant simple correlation coefficients among soil properties, extracted sulphate, and the weights, sulphur uptake and sulphur content of barley.

Soill1roQerties Total C TotalN Total S CIN

pH Clay % Sand % S04- Sz -Extract I -Extract 2 -Extract 3 -Extract 4

Fresh wt. n.s. 0.51' n.s. -0.53' -0.52* n.s. n.s. -0.52' -0.50 -0.68n.s.

Plant tissue (toQs} Soil sull1hate extracte~ Dry wt. S-ul1take S04 content Extract I Extract 2 Extract 3 Extract 4 0.46 n.s. n.s. n.s. n.s. n.s. n.s 0.62' n.s. n.s. n.s. n.s. n.s. 0.45 n.s, n.s. n.s. n.s n.s. n.s. 0.42 -0.50' -0.52" -0.62" -0.42 n.s. -0.44 -0.41 0.52* -0.46 n.s. n.s. n.s. n.s. n.s. 0.64* n.s n.s. n.s. n.s. n.s. n.s. n,s. n.s. -0.47 n.s n.s. n.s. n.s. -0.40* -0.41 -0.56' n.s.

-0.91* -0.88* -0.79" -0.57"

-0.92' -0.84* -0.66* -0.50

0.80* 0.760.50

0.65* 0.68"

0.50"

• Significant at P= 0.01 level, alfothers values significant at p = 0.05 (n.s. = not significant). Extract f ;"water; Extract 2 = dil. P04; Extract 3 = acidic 0.1 N P0 4; Extract 4= neutral 0.5 M P04'

Z

correlations. Using a stepwise multiple regression analysis, it was found that S-uptake could be predicted from water extractable S04-S (Extract I), with this single factor accounting for 83% of the variation in Suptake: S-uptake = -18.1 + 15.3 (Extract 1). Of the other extractants, only the dilute phosphate solution (Extract 2) approached the predictive efficiency of the water extract, and then only when the soil pH was included in the regression: S-uptake = -252 + 38.4 pH + 17.1 (Extract 2) (R2 = 0.82) For the remaining two extractants, namely 0.1 N phosphate (Extract 3) and neutral 0.5 M phosphate (Extract 4), the regression equations at best accounted for 65% of the variation in sulphur uptake. S-uptake = -7.64 + 11.4 (Extract 3) (R2 = 0.62) S-uptake = -503 + 31.3 C% + 93.1 pH + 7.93 (Extract 4) (R2 = 0.65) The inclusion of pH in the regression equation for the dilute (unbuffered) phosphate extractant, indicates that the extractive efficiency of the latter is dependent on pH, which is consistent with previous observations (Bart 1969). The inclusion of soil C content in the equation for Extract 4, presumably reflects the latter's reported ability (Bart 1969) to extract organic forms of sulphate, not readily available to plants over a short growing period. Sulphur-uptake was not found to be significantly correlated with total C, N or S content of the soil or with particle size distribution (fable 4). A negative correlation was however observed with CIN ratio (r = 0.52), i.e. soils with a wide CIN ratio were more likely to be S deficient. Sulphur-uptake was also correlated to sulphate-S content of tissue (r = 0.93) and inversely related to severity of visual S-deficiency symptoms as estimated by ranking on the basis of subjective evaluation of onset and intensity of symptoms (r = 0.49). Sulpbate-S content oftops: Sulphate levels in tissue have also been shown by a number of workers (Herath 1970, Ulrich and Hylton 1968) to be an index of sulphur supply to various crops. In this study sulphate content of tops ranged from 0 - 1190 ppm S, with only two (#3 and 22) exceeding 100 ppm S and only one (#3) exceeding 1000 ppm S, suggesting that nearly all of the samples were subject to some degree

168

o

500 ppm P t:. P04-HAc

water

X 0.5 M P04

II

70 X

60 '[ 50

~ 40

,

CI.l

"0

CI.l

""

t:. X t:.

X

20

X

10 Xx 0



X

X X

30

0

CI.l

o

X

X

X

X

XXX

.d< t:.~OX rP ~O ,

am! to Ted NniLu, my .-;ummer a~si5t{lllt. ,t;

173

, ..

-t~

-

i

_

"111\1.1": of ('O~'I'I':N'l'S

l'ot'el\·uJ>d AcknowledgmenLs.

i

Table of Contents List uf Oia~l'mH, Tables and Appendix I. JI ,

Ill,

IV. V,

VI. VlI.

VIII.

ii

JNl'lWOUCTION

1

OB,1 EfT 1 vt: fir Slll.FlII{ cm:I{E tAT ION I'IW.Jl:(,T (nate I'l'ujl;ct 113)

I)

FXI'EIUr.IE~'l'Al. DES Ita" AND III1RVES'l'lNG l>IETIlOIlS

I)

('IlHIICIIl. ANALYSIS

12

OH~ERVAT IONS ANIl R:':SlJI.TS

13

DlSCUSSIUN

16

CHNSIllERATIONS FUR FURTIII'R STUDY

20

REFERENCES

22

- ii -

Diagram r.

I'lot I>t's 19n

I.

Soil SuJraLe l.evl·ls

11

Avcl'agl' Yield \'/t'll4hls ami rlant. Sulfale Lends II.

14

Yield Response and Pl[tIiL Sulphate 1.t'\"I~1s fur In('rcnsiuA Soil ~04 l.tJ\'l~ls

17

Appf'lId j '(

1. 2.

Site

I.{)(~atiow·;

Compll~te

27

Analysis of

Site Suil Samples

2H

3.

Yield Data

29

4.

i'1'oteili i.evel:::;

5.

TisSUl' An I''''''

sulfate - aulful' 1501 - ';J but at deplhs gr"at".'

175

th~n

-

-

;

~0 1""""", ~!ll - :; i""",,"s0",,

cur~c

Ihl ... requirement h,",

at times h"~,, attributed to a high

",,1 fur

"equl.,.,.. e"t

of the fl)mbiotic nitrngull fi .. in!;\: rhizolll~. how{lvt'r

unab!.. LO ("OIl\p~t" "lthout added ""tfUL',

r"'s .. r~e_,> at .1 dCplh, U",

ha"e high sulfu" .'equir"lIletlt,..

ru,> bnth tI".

"mj ",,,die H"I'u simi'l1l'.

illl

L,·Y;, i. lu the adequate supply uf nitru"en

""Spon,."

provide,] Ill' the rhizullla.

Th:!s 0l.lnio" ha", been

In gr"sfi-1L'gumc mlxtut'c,. g,'own un "nIl',,,' d,'ficicnt soils, th" grass, with 11,.,. low,,,,. sulfur ""4"il'''",,,,,t$

g,'ass species to sulfur folluwing al'!,lh- .. t!u" of hillh

"i.l1 ",.,tc;:"mpetf· the legume.

le~"ls uf "it,'.-,~en.15

,\,. s1Ilrur it; ",dd"..

the problem oj" detcr~ining thc sulfur requircO>."nts

A sulfur IIp[>liciltiun

A ",·itlc,..l 1""cl uf 2-3 ppm

ur

of d"iffer..,nt suila for lIIa"i .. wn crop gruwth.

snlublc

soil sulfilt"-~lIlfur hilS been fonnd

These methou" ot' aoil a"alysio. inclUde organic 5ul("l'dete"Cliuiltions, total "ulful' d .. terminati",..;

and suIt ate sull'ur extraction

.ility fo,' U1 ~alul -;1, l'T14 ,", ..::~" " "'''' ,flc"tion lo " en·; lcyel "rilP,·.i3 _ d,'Cd'T.t ac,l r.c" .... lefid""t., " ~'pa .'~l c:r ], ..,,-1.

176

_ 10 _

,

,\""lysis of ""lfor in plant tl"'8ul; is nol..,

In/a'"

at th" present tIme, " runli"" pn,C",·d",·c in the

lb/"".

Ill" .''1"1 H'K ("uIHa:! AT!i\~ !'IWolfer (n"t .. P"U

011.111"1 HI

j;ulful'

The [iI·"t thirt.",,,, (110.'" 1-13)

).inel"""" (19) pluts, un scroced" .. e for

sulf,,1;~ dete,· .. lnlltion ... on .sotl e.,t,·acta wali mudifled

" "

all follo ...~.

1).1 It of pla!!t lII.te,'ial w"a added

di .. ectly to the boiling fl" .. k .10n/: wIth 2 lilt of 'IS

9

ll\
178

• 18 •

another had" possible resp",,,,,e.

§Si R

I" "

looki"~ at Tabie II Wol find that all plots

~

wlth plant ( .. if"Ifa) S04 _ M levels beI"w 700 PI''" Seither g4V, ruund VIt.

CONSTTlf:RAnO'lS FUR rURTllrR STlihY

increased respon" .. to sulfur as til" s""son adv .. nced. A ,.tudy of possible sulfur dufl"i"ncies 11\

lJ"ke,' "t ai, (1973) did so",,, ....,,·k with urch",·d· til .. 1.0w"r Fr .... er Vlliley lind Vancouv"r 1... 1I1nd, u"lng grass in Washingtun Stllte, "wd fo""d "" r""p''''''e in crops sueh as gra ..s-.,loy"r .. i:>.tul'llitle ... of this study is th .. L~ct th~t on (, o! th" 13 sit" .. ,

179

COl'li

"'"

tho test c,'op, sh"uld 1>" 1;0""Is

trat.ion, and Total S"lfur Uptake of FiYe Ann"al 1966

~utriellt

Ryegrass in Englalld

of Englaud and

114,1 0 4-113

Jones. M.D., and Quagliato, J.I.•

19i3

Response of I'ou .. Tropical Legum",s and Alfalfa ttl Varying

180

Leyels of Sulfur

Sull"u.· Institute J.

\o,'intc"r/:-'p"in'l;

1"73

Z'l.

Jl,'yl"nd luct::.'nt' in the I:"",le'-" Queensland

Au"tr"lian

,I.

1)" ..

!:~p.

Sci", I:.t:., ('"ld>i

Sell Smith

Rock Creek

Olher

"

Rick na"pu,'

Rock Creek

Oliver

En,;!"rby-Gr im'oo

Salmun A.. m Creston

'"

l",beau Ruge .. s

( .. cston Flat"

'"'

P iucutt

Ilurfalo Crcck

",-

[J.H.

,11

Salb.

2.0 7." O.7!:i

5

5.9 0.i6

5

4 .~ U.ll 0.:-'8

"" , • ,• ••• , ,.. , ,

U.S 0.Z6

5

5.' 6. ;; 0.36

6

, .1

4

9

Mclure

" " "

'0.

8

Albert Piva

" "

JsUe

"11

.. ,6 C.'} 0.54

.

'.4 , .4 0.46

3.5

'.9 0.42

,..

3 ., '.1 2.50 2.3

0.66

15

, .3 6 ••

16

3.6 7.1 0.32

17

18

5

19

3

WUI i.am" l./lkc

181

I

10

3

13

1

, "

•• 7 6. , 0.2.6

'.7 6.() 0.34

" , , , , •

9

I

, 4

I' t~~,~s li,/A

0,16

5.4 '.5 0.42

3

n!l~

let.' Analy"fs of ,site Soil Sa .. Ie"

U.~O

3.6 6.7 0.34

"

265 2;3

271 1300

,,,

J600

'" 490

"' ~,~"

'" SI 3.

1.80 0.40

8.1i 1.60

'HO

,,'

"

360

7.b 0.33

"5 1400

"0

133

190 4800

600

'90 4100

5"

"

...'.,

0.40

0.64 0.52

2.1 0.60

56

m 3400 6" 3.6 0.,0 6" 4150 1000 t 11.0 0.40

160

605 1620 looo! •• 1 0.84

115

1

6" 4350

'"

3.0 0.48

7000

"0

120

225 5900

125

3.5 0.40

190

'SO

2400

250

A. 6

370 2600

'70

6.3 0.44

'"

720

6.6 0.54

'"

4.Z

"

"

'SO

10

17

10.0 6.0 0.48

9

';00

8.~

3200

S

20..: 4 6 00

0.30

3. i

'"194 '"' 1~7A 310

130 4300

108

6

I",

34

•, • '" • • " •

17

,

'SO

3300

90 7700

5800 5800

7SO

6.5 0.26

0.4:;

0._~2

, .4 l.,10

I

.\prnmtx TABU:

- 30 -

,"SOll TA8U: 3. /"ontld)

'(ield nata ",thout

~hJ,'d

h'ith t1thlC'd

Site and R,. .

~o.

26.5

I

7.1 2

32

34

0.23

0.63

';

Hi

0.24

22

18

0.26

0.6

t6 _lOS

1.06

O.H

51

1,)

20

0.27

0,6]

17

,9

59 Ii6

21

19

0.25

0.65

26

21

0.28

0.6J 0.5

1.14

0.31

35

1.Z6

0.30

17

6s

1.09

0.3~

",8

61

31

24

0.28

1.15

(\,32

56

12

61

24

25

0.2,)

0.6

52

14

167

20

24

0.32

1.0

66

14

145

19

23

0.30

1.1,

36

16

162

20

25

0.31

l.~,

58

13

140

21

23

0.32

1.0,