Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Soil variability and the use of conventional and innovative methods for assessing soil fertility Bank, Gary Michael 1984

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1984_A6_7 B35.pdf [ 9.81MB ]
Metadata
JSON: 831-1.0096036.json
JSON-LD: 831-1.0096036-ld.json
RDF/XML (Pretty): 831-1.0096036-rdf.xml
RDF/JSON: 831-1.0096036-rdf.json
Turtle: 831-1.0096036-turtle.txt
N-Triples: 831-1.0096036-rdf-ntriples.txt
Original Record: 831-1.0096036-source.json
Full Text
831-1.0096036-fulltext.txt
Citation
831-1.0096036.ris

Full Text

SOIL VARIABILITY AND THE USE OF METHODS FOR ASSESSING By GARY MICHAEL B.Sc.(Agr.)» The University of CONVENTIONAL AND INNOVATIVE SOIL FERTILITY BANK British Columbia, 1979 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Soil Science We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March 1984 © Gary Michael Bank I n p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r a n a d v a n c e d d e g r e e a t t h e U n i v e r s i t y o f B r i t i s h C o l u m b i a , I a g r e e t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e a n d s t u d y . I f u r t h e r a g r e e t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may b e g r a n t e d b y t h e h e a d o f my d e p a r t m e n t o r b y h i s o r h e r r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l n o t b e a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . D e p a r t m e n t o f S & v l S c i e n c e . T h e U n i v e r s i t y o f B r i t i s h C o l u m b i a 1956 Main M a l l V a n c o u v e r , C a n a d a V6T 1Y3 ABSTRACT S o i l sampling plays a major role in the process which leads to f e r t i l i z e r recommendation. Conventionally, characterization of f i e l d chemistry is based on a single composite sample. This single value is used as the basis for a f e r t i l i z e r recommendations which often amounts to many hundreds of d o l l a r s per f i e l d . In order to test the v a l i d i t y of the conventional composite sampling method a sampling program involving three d i f f e r e n t sampling methods was undertaken. The sampling program was designed to make a comparison of the information provided by and the costs associated with the three d i f f e r e n t sampling methods. In addition, factors having the largest influence on v a r i a b i l i t y were determined. Characterization of the chemical v a r i a b i l i t y of several a g r i c u l t u r a l f i e l d s in the Lower Fraser Valley was also achieved. Four f i e l d s were sampled in the spring and three of them in the f a l l . Soils were sampled in the plow layer, at 30-60 cm, and at 60-90 cm. Random s t r a t i f i e d sampling, random s t r a t i f i e d composite and conventional sampling was carried out on each f i e l d . S o i l analyses included exchangeable bases, pH, %C, NO3-N, and Bray's extractable P. Analysis of variance showed that differences among f i e l d s and depths were the most important sources of variance. Contributions of time and interaction effects to variance were generally small. A comparison of mean values produced by the diffe r e n t sampling methods showed that mean values resulting from the use of conventional sampling for NO3-N, P, and K, were within i i 20% of the s t r a t i f i e d random mean 65% of the time, and within 50-125% of the s t r a t i f i e d random mean 17% of the time. Random s t r a t i f i e d composite mean values d i f f e r e d from the random s t r a t i f i e d mean by less than 50% in a l l cases. The value of s t r a t i f i e d random sampling (detailed sampling) becomes apparent when a comparison of the chemical value within each of the f i e l d strata is made with the conventional mean. In the worst case, only 22% of a f i e l d ' s area f e l l within plus or minus 15 ppm of the conventional mean for P. The high cost of analysis for detailed sampling, compared to the cost associated with conventional sampling, was shown to be o f f s e t by savings in the amounts of f e r t i l i z e r recommended. The results of detailed sampling could be used to provide optimum f e r t i l i z e r recommendations to a l l parts of the f i e l d and also to reduce f i e l d chemical v a r i a b i l i t y . Cluster analysis carried out on plow layer samples of individual f i e l d s was able to c l a s s i f y f i e l d s into units with s i g n i f i c a n t l y d i f f e r e n t chemistry. The multivariate approach was able to distinguish among areas of f i e l d s with s i g n i f i c a n t l y d i f f e r e n t levels of CEC, %C, pH, and the major f e r t i l i z e r variables P and K. The lack of contiguousness of unit members, for some f i e l d s , and the need for more than one P and K recommendation within many of the cluster units would reduce the possible p r a c t i c a l usefulness of these r e s u l t s . Cluster analysis run on plow layer data was highly successful in discriminating among three f i e l d s . This suggests a high degree of chemical identity of f i e l d s , even i f found on the same mapping unit. i i i TABLE OF CONTENTS PAGE ABSTRACT i i TABLE OF CONTENTS i v LIST OF TABLES v i LIST OF FIGURES ix LIST OF APPENDICES x i LIST OF ABBREVIATIONS USED IN THE TEXT. x i i ACKNOWLEDGEMENTS x i i i INTRODUCTION 1 DESCRIPTION OF STUDY SITES 3 SAMPLING TIMES 6 SAMPLING METHODS 6 LABORATORY METHODS 10 STATISTICAL METHODS 12 CHAPTER I. SOURCES OF VARIABILITY Introduction 16 Overall v a r i a b i l i t y 19 Determining s i g n i f i c a n t differences among sampling methods 22 Differences among f i e l d s : in the plow layer 25 Differences among f i e l d s : at the parent material l e v e l 32 Time as a source of v a r i a b i l i t y 36 Comparing the v a r i a b i l i t y of three d i f f e r e n t sampling depths 41 Summmary 50 iv I I . COMPARISON OF SAMPLING METHODS Introduction 54 Comparing mean values of 3 sampling methods 56 S t a t i s t i c a l sampling reqirements . 61 Comparison of costs associated with various sampling methods 63 E f f e c t of error i n mean value on f e r t i l i z e r recommendation 65 Possible alternatives to high analysis cost methods 69 Summary 74 III . AN INNOVATIVE APPROACH TO SOIL FERTILITY MANAGEMENT: A SINGLE PARAMETER APPROACH Introduction 76 Comparing f e r t i l i z e r requirements using mean value and plot values 78 Summary 87 IV. AN INNOVATIVE APPROACH TO SOIL FERTILITY MANAGEMENT: A MULTIPARAMETER APPROACH Introduction 89 Results of cluster analysis of plow layer samples: Individual f i e l d s 91 Clustering 3 f i e l d s on the basis of plow layer data... 103 A combined approach to f e r t i l i t y assessment 107 Summary 112 CONCLUSIONS 114 References 117 APPENDICES 122 v LIST OF TABLES Table Page 1-1 Sources of variance, F-probability values, and per-cent of t o t a l variance att r i b u t a b l e to each source of variance for 3 f i e l d s sampled at 3 depths in spring and f a l l 21 1-2 Sources of variance, (including sampling method) f-pr o b a b i l i t y values, homogeneity of variance r e s u l t s , and percent of total variance attributable to each source of variance (for 4 f i e l d s ' plow layer data in spring) . 24 1-3 ANOVA results showing p r o b a b i l i t y of s i g n i f i c a n t differences among f i e l d s and percent of variance attributable to differences among f i e l d s and error, at two d i f f e r e n t sampling times (plow layer data) 27 1-4 Student-Newman-Keuls range t e s t for four f i e l d s ' plow layer samples i n spring 28 1-5 Student-Newman-Keuls range test for three f i e l d s ' plow layer samples i n f a l l 29 1-6 ANOVA results showing p r o b a b i l i t y of a s i g n i f i c a n t difference among f i e l d s and percent of variance at-tri b u t a b l e to differences among f i e l d s at two d i f f -erent sampling times (60-90 cm samples) 33 1-7 Student-Newman-Keuls range test for four f i e l d s ' 60-90 cm samples i n spring 34 1- 8 Student-Newman-Keuls range test for three f i e l d s ' 60-90 cm samples i n f a l l 35 2- 1 Comparison of three sampling method's estimate of f i e l d mean values f o r five plow layer parameters in f a l l 58 2-2 Comparison of three sampling method's estimate of f i e l d mean values f o r f i v e plow layer parameters in spring 59 2-3 Percent difference of s t r a t i f i e d random composite and conventional sampling means from the s t r a t i f i e d random mean i n spring and i n the f a l l ( plus or minus indicates the d i r e c t i o n of dif f e r e n c e ) . . . 60 2-4 Sample numbers required to estimate mean value of plow layer parameters with a ±. 10% precision and 90% confidence 62 v i 2-5 Cost associated with analysis for P and K for d i f f -erent sampling methods, for one f i e l d 63 2-6 Quantifying error i n f e r t i l i z e r recommendation assoc-iated with high and low f e r t i l i t y estimates for P at two s o i l test l e v e l s (recommendations based on BCMAF guidlines for crop group three) 67 2-7 Quantifying error i n f e r t i l i z e r recommendation assoc-iated with high and low f e r i l i t y estimates f o r K at two s o i l test l e v e l s , recommendations based on BCMAF guidelines for crop group three 68 2-8 Correlation matrix for variables at three depths (data from 102 p r o f i l e s samples at three depths. 1, 2, or 3 indicate sampling depths of of plow l a y e r , 30-60 cm, and 60-90 cm respectively) 71 2-8b Corr e l a t i o n matrix for variables at three depths (data from 102 p r o f i l e s sampled at three depths; 1, 2, or 3 indicates sampling depth of of plow l a y e r , 30-60 cm, and 60-90 cm respectively) 72 2-9 r values for selected parameters in the plow layer i n spring and for combined spring and f a l l data 73 2- 10 r values for selected parameters at 60-90 cm, data from a l l f i e l d s combined 73 3- 1 Summary of percent of the area of four f i e l d s which would have phosphorus plot values above and below a + 15 ppm increment around the f i e l d arithmetic mean... 80 3-2 Comparing phosphorus f e r t i l i z e r recommendations using values obtained by detailed sampling to that r e s u l t -ing from the use of the conventional mean value, for HRST i n spring 82 3-3 Comparing potassium f e r t i l i z e r recommendations using values obtained by detailed sampling to that r e s u l t -ing from the use of the conventional mean value, for HRST i n spring 8 3 3-4 F e r t i l i z e r cost vs s o i l analysis cost for one, two, and three sampling units given that the f i e l d is 3.6 ha. and f e r t i l i z e r i s added at 220 kg/ha. for both P and K 85 4-1 Plot # and parameter values associated with cluster r e s u l t s for RN plow layer data i n spring. Plots are arranged i n dendrogram order, column one of plot ID i d e n t i f i e s the f i e l d , columns three and four i d e n t i f y p l o t number 94 v i i 4-2 Table showing s i g n i f i c a n t differences among RN cluster units 9 5 4-3 Plot # and parameter values associated with cluster results for RC plow layer data i n spring. Plots are arranged i n dendrogram order, column one of plot ID i d e n t i f i e s the f i e l d , columns three and four i d e n t i f y p l o t number 96 4-4 Table showing s i g n i f i c a n t differences among RC cluster units 97 4-5 Plot # and parameter values associated with cluster r e s u l t s for HRST plow layer data i n spring. Plots are arranged in dendrogram order, column one of plot ID i d e n t i f i e s the f i e l d , columns three and four i d e n t i f y p l o t number 99 4-6 Table showing s i g n i f i c a n t differences among HRST cluster units 100 4-7 Plot # and parameter values associated with cluster r e s u l t s for CLPRO plow layer data i n spring. Plots are arranged i n dendrogram order, column one of plot ID i d e n t i f i e s the f i e l d , columns three and four i d e n t i f y p l o t number 101 4-8 Table showing s i g n i f i c a n t differences among CLPRO cluster units 102 4-9 Plot # and parameter values associated with cluster unit members from plow layer data of three f i e l d s . Plots are arranged i n dendrogram order, column one ofplot ID i d e n t i f i e s the f i e l d , column three and four i d e n t i f y plot number 10 5 4-10 Table showing s i g n i f i c a n t differences among cluster units for plow data from three f i e l d s i n spring 106 4-11 Comparing a multivariate to a si n g l e parameter approach for assessing lime requirements for RN in spring 109 4-12 Assessing the usefulness of cluster units in recommending P for HRST i n spring 110 v i i i LIST OF FIGURES Figure Pag e 1.0 Site description 4 2.0 I l l u s t r a t i o n of 3 sampling methods with randomly selected p l o t locations and plot numbers for s t r a t i f i e d random, d i s t r i b u t i o n of 3 s t r a t i f i e d random composite plots (A,. B, & C) and sampling point, and recommended sampling pattern for con-ventional method 9 3.0 S t a t i s t i c a l methods 13 1-1 Comparison of f i e l d s ' mean values for Ca, Mg, K, and CEC i n the plow la y e r , error bars indicate range of observations, spring and f a l l data combined 30 1-lb Comparison of f i e l d s ' mean values for %C, pH, P, and NO5-N i n the plow layer , error bars indicate range of observations, spring and f a l l data combined 31 1-2 Changes in % CV from spring to f a l l for three f i e l d s ' plow layer samples 38 1-3 Changes in % CV from spring to f a l l for three f i e l d s ' 30-60 cm samples 39 1-4 Changes in % CV from spring to f a l l for three f i e l d s ' 60-9 0 cm samples 4 0 1-5 F i e l d mean value & %CV for Ca for 3 f i e l d s at three depths 44 1-6 F i e l d mean value & %CV for K for 3 f i e l d s at three depths 45 1-7 F i e l d mean value & %CV for CEC for 3 f i e l d s at three depths 46 1-8 F i e l d mean value & %CV for pH for 3 f i e l d s at three depths 47 1-9 F i e l d mean value & %CV for NO3-N for 3 f i e l d s at three depths 48 1-10 F i e l d mean value & %CV for P for 3 f i e l d s at three depths 49 ix 3- 1 Spacial d i s t r i b u t i o n of P levels for HRST in spring. C a l c u l a t i o n of % of f i e l d area corr-e c t l y f e r t i l i z e d based on the assumption that a ± 15 ppm increment around the arithmetic mean i s c o r r e c t l y f e r t i l i z e d 81 4- 1 Dendrogram from clustering of RN plow layer data i n spring, clustered on Ca, CEC , pH, %C, and P 94 4-2 Spacial d i s t r i b u t i o n of RN cluster unit members 95 4-3 Dendrogram from clustering of RC plow layer data i n spring, clustered on Ca, CEC, pH, %C, and P 96 4-4 Spacial d i s t r i b u t i o n of RC cluster unit members 97 4-5 Dendrogram from cl u s t e r i n g of HRST plow layer data i n spring, clustered on Ca, CEC, pH, %C, and P 99 4-6 Spacial d i s t r i b u t i o n of HRST cluster unit members 100 4-7 Dendrogram from clustering of CLPRO plow layer data i n spring, clustered on Ca, CEC, pH, %C, and P 101 4-8 Spacial d i s t r i b u t i o n of CLPRO cluster unit members.... 102 4-9 Dendrogram for clustering of three f i e l d s * data from the plow laye, l e t t e r s indicated cluster units...104 x LIST OF APPENDICES Appendix Page 1. S t r a t i f i e d random sampling data 122 2. S t r a t i f i e d random composite sampling data 130 3. Conventional sampling data 133 4. Mean, standard deviation, and % CV for plow layer samples for 3 f i e l d s (spring & f a l l data combined).. 134 5. Mean, standard deviation, and % CV for 60/90 cm samples for 3 f i e l d s (spring & f a l l data combined).. 135 6. Mean, standard deviation, and % CV for p r o f i l e data for 3 f i e l d s (spring & f a l l data combined) 136 7. BCMAF potassium recommendations 137 8. BCMAF phosphorus recommendations 138 9. Principal components for three f i e l d s : plow layer data i n spring 139 x i LIST OF ABBREVIATIONS USED IN TEXT RN = Reynolds RC = Reynolds clover HRST = Horstings CLPRO = CI over dale produce Description of sample i d e n t i f i c a t i o n 6 columns are used for sample i d e n t i f i c a t i o n , the description i s as follows: col 1 f i e l d ID col 2 season col 3 & 4 p l o t # col 5 sampling depth col 6 sample type F i e l d i d e n t i f i c a t i o n Season i d e n t i f i c a t i o n 1 = CLPRO 2 = HRST 1 = spring 3 = RN 4 = R C 2= f a l l Sampling depth 1 = 0 / plow layer 2 = 30/60 cm 3 = 60/9 0 cm Sample type 1 = s t r a t i f i e d random 2 = s t r a t i f i e d random composite 3 = convention composite i x i i ACKNOWLEDGEMENT S The p a r t i a l funding provided by the National Sciences and Engineering Council (Canada) i s acknowledged (Grant A-7411) . Many people i n the s o i l s department provided helpful ideas and discussion. I extend special thanks to those who helped out with my f i e l d work. The laboratory analysis could not have been completed without the advice of Ms. T. D. Nguyen, P a t t i Carbis, J u l i e Lansiquot, and Eveline Wolterson. The good humor of a l l the people in the lab made that part of my work a pleasant task. I would l i k e to thank my committee members Dr. Art Bomke, Dr. Grant Kowalenko, and Dr. Les Lavkulich for their advice and guidence. The guidence and encouragement of my thesis advisor Dr. Hans Schreier i s e s p e c i a l l y acknowledged. Thanks to Forest Johnston who provided many hours of valuable e d i t i n g . F i n a l l y , I would l i k e to thank a l l my family members for their encouragement. My wife Caroline was always patient and encouraging throughout my graduate program. My son James unknowingly provided the sort of encouragement that can not be provided by any adult. INTRODUCTION The v a r i a b i l i t y of s o i l chemical properties d i r e c t l y and i n d i r e c t l y exert a large influence on the practice of s o i l f e r t i l i z a t i o n . B a l l and Williams (1968) state that a knowledge of s o i l v a r i a b i l i t y i s as important as the mean value for pedological and ecological studies. Unfortunately, the magnitude and sources of v a r i a b i l i t y and i t s subsequent influence on the practice of f e r t i l i z a t i o n receive l i t t l e attention. V a r i a b i l i t y of f e r t i l i t y l e v e l s within f i e l d s can not only lead to patchy or variable crop yields but can also strongly a f f e c t the accuracy of f e r t i l i z e r recommendations based on s o i l samples. F e r t i l i z e r usage i s an expensive but integral part of intensive crop production. S o i l sampling provides the basis of the process of s o i l testing and resu l t i n g f e r t i l i z e r recommendations, therefore, i t s importance can not be under-estimated. In order for the farmer to determine the amount of f e r t i l i z e r required, whether to optimize or maximize crop production, he must have accurate characterization of f i e l d f e r t i l i t y . Many researchers have considered the problem of v a r i a b i l i t y i n forested s o i l , (Mader, 1963; Blyth and Macleod, 1978; S l a -v i n s k i , 1977). Hamond et al (1958) Peck and Dibb (1977) and Potash and Phosphate Inst. (197 8) have provided insight into s o i l v a r i a b i l i t y with respect to a g r i c u l t u r a l s o i l s . Beckett and Webster (1971), in an extensive review, provide a comparison of the v a r i a b i l i t y of forested and of a g r i c u l t u r a l s o i l s . 1 Though t h i s , and other research, provides a basis for further v a r i a b i l i t y research, i t l i k e l y can not answer the s p e c i f i c v a r i a b i l i t y problems associated with s o i l sampling and f e r t i l i z e r recommendations in the Lower Fraser Valley. In order to address the s p e c i f i c questions of sampling methods and sampling requirments for d i f f e r e n t f i e l d s in spring and in f a l l , a study of s o i l v a r i a b i l i t y involving three sampling methods was undertaken. Four a g r i c u l t u r a l s o i l s in the Lower Fraser Valley were sampled at three d i f f e r e n t depths and two d i f f e r e n t times. The aims of the research were: 1 . To examine three sources of v a r i a b i l i t y ( f i e l d s , depth,and time) to f a c i l i t a t e an understanding of v a r i a b i l i t y and i t s impact on s o i l sampling for f e r t i l i z e r recommendations. 2. To evaluate the effectiveness of conventional composite sampling and s t r a t i f i e d random composite sampling in predicting mean f i e l d f e r t i l i t y l e v e l s . 3 . To compare the o v e r a l l costs and recommended f e r t i l i z e r rates resulting from the use of detailed sampling to those res u l t i n g from the use of conventional sampling. 4. To explore the use of a multivariate c l a s s i f i c a t i o n tech-nique (cluster analysis) in predicting f i e l d management units. 5. To use cluster analysis to determine s i g n i f i c a n t differences in s o i l chemistry among f i e l d s on the basis of plow layer samples . 2 DESCRIPTION OF STUDY SITES Four study s i t e s were chosen to be representative of three d i f f e r e n t parent materials. Two f i e l d s were on the same pure mapping unit, one f i e l d was on an association with v i s u a l l y apparent and systematic v a r i a t i o n s , and the fourth f i e l d was on a map unit with randomly occurring variations in s o i l organic matter depth and content. The s o i l s of these four f i e l d s are representative of a large proportion of the c u l t i v a t e d land i n the Lower Fraser Valley. They also represent the range of mapping concepts on which commercial f i e l d s may be found. A detailed description of the f i e l d s is found in Figure 1.0. The f i r s t two f i e l d s which were chosen are located on Westham Island, about 50 km south west of Vancouver. These f i e l d s are found on a pure mapping unit represented by the Crescent s o i l s e r i e s . One studied f i e l d had peas grown on i t i n the previous year while the other f i e l d had barley under seeded with red clover grown on i t for one year prior to sampling. Both f i e l d s were drained and have been under c u l t i v a t i o n for greater than ten years. The next f i e l d was chosen to represent an association and i s found near the community of Deroche, located about 120 km east of Vancouver. Though plowing and clearing have modified the o r i g i n a l deposits the association members can s t i l l be recognized. Matsqui, the well drained member, i s found along the upper portions of the r o l l i n g swell and swale topography, which i s c h a r a c t e r i s t i c of the l a t e r a l accretion deposits in this area. Along the ridge tops the effect of management has, in many 3 FIELD (location) PREVIOUS CROP SO SERIES IL DESCRIPTION DRAINAGE CLASSI-FICATION PARENT MATERIAL DESCRIPTION Reynolds (Westham I s l . ) peas Crescent mod-poor poor O.G -medium to moderately f i n e textured d e l t a i c d e p o s t i t s Reynolds Clover (Westham I s l . ) clover barley Crescent mod-poor poor 0.G -medium to moderately f i n e textured d e l t a i c d e p o s t i t s Rorsting (Deroche) brussels sprouts Dewdney F a i r f i e l d Matsqui imperfect imperfect well to mod GLE.MB GLE.MB E.EB -15-50cm of med t e x t u r e d , l a t e r a l l y accreted f l o o d p l a i n deposits over sand -med to moderately f i n e textured, l a t e r a l l y accreted flood p l a i n deposits -15-50cm of med t e x t u r e d , l a t e r a l l y accreted f l o o d p l a i n deposits over sand Cloverdale Produce (Cloverdale) potatoes Lulu Richmond Vinod very poor very poor very poor T.M T.H R.Gsp -40-160cm of p a r t i a l l y decomposed organic m a t e r i a l over mod f i n e textured d e l t a i c deposits -40-160cm of well decomposed organic mmaterial over mod f i n e - t e x t u r e d d e l t a i c deposits -10-40cm of organic m a t e r i a l over moderately f i n e textured d e l t a i c deposits Figure 1.0 SITE DESCRIPTION instances, been to remove the thin covering of medium textured material and expose the underlying sandy s o i l . Dewdney i s found in the depressions between the ridges. Only a very small amount of F a i r f i e l d i s found on the f i e l d but i t too has been modified by management. The low l y i n g area of F a i r f i e l d i s modified by a mantle of coarser material from adjacent higher areas as the farmer f i l l e d the lowest part of the f i e l d . H i s t o r i c a l l y , this f i e l d was used as dairy pasture for over f i f t y years. Prior to sampling, the f i e l d had been used for brussels sprouts production for two seasons. The l a s t f i e l d i s found on the property of Cloverdale Produce in Cloverdale, about 40 km south east of Vancouver. The property is e f f e c t i v e l y t i l e drained through the growing season but i s flooded much of the winter. The study f i e l d occurs on two map units . To the north end of the f i e l d the Richmond and Vinod series are found, while the south three quarters of the f i e l d i s dominated by the Richmond and Lulu s e r i e s . These s o i l s occur i n a random manner across the f i e l d . There i s a strong contrast in f e r t i l i t y and manageability between the Vinod (Rego Gleysol) and the other two organic s o i l s . With time and management the organic layers have shown considerable subsidence (Wood, per-sonal communication) leading to the exposure of mineral s o i l s in many loca t i o n s . Plowing has served to mix the mineral s o i l with the organic p r o f i l e s through much of the f i e l d . In this way the f i e l d becomes more variable than the map unit description would indicate. 5 SAMPLING TIMES Fields were sampled f i r s t i n the spring and then i n the f a l l of 1982. Spring sampling took place over a two week period s t a r t i n g during the l a s t week in A p r i l . F a l l sampling took place over about a three week period, s t a r t i n g in mid October. SAMPLING METHODS In order to test the eff e c t of sampling i n t e n s i t y and sampling method, three sampling methods were used. The methods were s t r a t i f i e d random, s t r a t i f i e d random composite, and the method currently recommended by the BCMAF (Anonymous c, 197 8), which w i l l be c a l l e d the conventional sampling method. Figure 2.0 i l l u s t r a t e s the d i f f e r e n t sampling methods. Within each f i e l d 18 square and equal size plots were layed out by marking corners with stakes and survey tape. A random number (one through 18) was then produced so that one plot could be eliminated leaving the number of plots to be sampled at 17. Next, two sets of random numbers were selected to give coordinates for s t r a t i f i e d random and s t r a t i f i e d random composite sampling points within each plot. For Reynolds (RN) , Reynolds Clover (RC) , and Horsting (HRST) the plot size was 45m x 45 m. For Cloverdale Produce (CLPRO) , the area with the same cropping history was much larger so the plot size was increased to 70m x 70 m. For a l l the f i e l d s except HRST the sampling area was layed out to cover about 80% of the f i e l d ' s 6 area with the same cropping h i s t o r y . In the case of HRST the sampling area cover about one f i f t h of the entire area i n brussels sprouts production. SAMPLING STRATIFIED RANDOM PLOTS With th i s method each f i e l d had 17 plots sampled and analysed i n d i v i d u a l l y . A pair of numbers selected from a random numbers table provided the coordinates for the sampling point, within any given p l o t . On a r r i v i n g at the randomly selected point for a given p l o t , the s o i l was sampled for bulk density at 2 depths and chemical properties at three depths. A detailed description i s as follows: f i r s t the bulk density was sampled at the surface, then a 30 cm deep p i t was dug to determine the depth of the plow layer. Sampling for chemical properties was then carried out using a 2.5 cm diameter Oakfield probe. Between f i v e and seven samples from a 30 cm x 30 cm area were taken to the depth of the plow layer . These were then composited to form the plow layer sample for that p a r t i c u l a r p l o t . The 30/60 cm sample for chemical analyses was taken from the bottom of the p i t . The sample consisted of a composite of 5 cores taken from a 30 cm x 30 cm area. The 60/90 cm samples were,then taken from the same locations as the 30/60 cm samples. F i n a l l y the bulk density sample for the 30 cm depth was taken from the bottom of the p i t . 7 STRATIFIED RANDOM COMPOSITE Three composite samples were taken from each f i e l d . From the 18 possible plots 3 groups of 5 plots were randomly selected to make up the 3 composite samples which were labeled 1,2, and 3. A new set of random coordinates gave the composite samples independence from the s t r a t i f i e d random samples. Composite samples were taken from the same depths as were the s t r a t i f i e d random samples. Each plot's contribution to a given composite sample con-sisted of f i v e cores taken at .2m interv a l s over a i m distance. CONVENTIONAL SAMPLING Conventional composite sampling was carried out as described in the BCMAF guidelines for s o i l sampling (Anonymous c , 197 8) . Ten to 15 samples were taken from each f i e l d . Samples were taken to avoid small low wet areas, dead furrows, and areas close to fences. Generally sampling avoided any "noncharacteristic" areas. Each sampling point was represented by a si n g l e auger sample to the depth of the plow la y e r . Plow layer depth was determined by the f e e l of h i t t i n g a more compact plow pan beneath. 8 CONVENTIONAL SAMPLING METHOD STRATIFIED RANDOM STRATIFIED RANDOM COMPOSITE 1 7 13 C B 2 8 14 B A A 3 9 15 B C A 4 10 16 A C 5 11 17 A C B 6 12 18 C B F i g . 2.0 I l l u s t r a t i o n of 3 sampling methods with randomly selected plot locations and plot number for s t r a t i f i e d random, d i s t r i b u t i o n of 3 s t r a t i f i e d random composite plots (A, B, & C) and sampling point, and recommended sampling pattern for conventional method. 9 LABORATORY METHODS Within 24 hours of sampling , s o i l samples were put out to air dry at about 22 degrees C. Samples were subsequently ground to pass a 2mm sieve and stored in a i r t i g h t containers. Coarse fragment content was neg l i g i b l e in a l l cases and was not measured. Bulk densities were determined using the core method (Blake, 19 65). In order to be consistent with BCMAF f e r t i l i z e r recommendations, chemical analysis methods outlined i n the BCMAF S o i l Testing and Interpretations manual (Neufeld,1980) were followed as c l o s e l y as was p r a c t i c a l . Organic carbon was determined using the Walkley Black method (Allison,1965) . T i t r a t i o n end points were determined coloro-me t r i c a l l y . Values for pH were determined using 15 g of s o i l in a 1:2 s o i l water mixture. Equilibrium time was 30 minutes. The pH meter used was a Radiometer PH M62 standard pH meter. Nitrate-N was extracted using .02N CaSO^. in a 1:10 s o i l to so l u t i o n r a t i o by weight. Shaking time was 10 minutes and a 2g sample was used. Rather than using the modified phenoldisulfonic acid method for color development, as outlined in the BCMAF methods, the method outlined by Technicon was used (Anonymous a,1977 ). In general, the extracted nit r a t e i s reduced to n i t r i t e by a copper-cadmium reductor column. The n i t r i t e ion then reacts with coloring reagents to form a reddish-purple dye and the n i t r i t e concentration i s determined color i m e t r i c a l l y by the autoanalyser. It was assumed that n i t r i t e content was neglible so there was no 10 concern that a s i g n i f i c a n t p o s i t i v e nitrate error would occur. Phosphorus was determined using Bray P-l extracting s o l u t i o n of 0.025 N HC1 and 0.03 NH^F. The s o i l solution r a t i o was 1:10 and a shaking time of 10 minutes was used. A 2 g s o i l sample was used. The determination of color d i f f e r s from that used by the BCMAF. A s l i g h t l y modified version of the ascorbic acid method described by Murphy and Riley (1962) was used for color development. Color i n t e n s i t y was determined by a G i l f o r d Stasar II spectrophotometer at 660 urn. Exchangeable Ca, Mg, Na, and K were determined by the ammonium acetate method at pH 7.0 (Chapman,1965). Total exchange capacity was determined on the same sample. The concentration of exchangeable bases was determined on a Perkin Elmer 306 atomic absorption spectrophotometer. Exchange capacity was determined by analysing for ammonium on the Technicon Autoanalyser II (Anonymous b, 1974) 11 STATISTICAL METHODS An outline of the s t a t i s t i c a l procedures used i n data analysis can be seen in the flow chart ( f i g . 3.0). In a l l , seven main kinds of calculations or tests were performed. Analysis of variance tests (ANOVA) were generally performed using UBC ANOVAR (Grieg,M and O s t e r l i n , 1978). The assumptions of ANOVA are that the samples are selected at random, the samples are homoscedastic (have equal variance), and that the samples are normally d i s t r i b u t e d . Fortunately the test i s robust enough to be used with considerable departure from normality and homoscedasticity (Zar, 1974). UBC GENLIN (Greig and Bjerring, 198 0) was used for analysis of variance in the case where sample numbers within l e v e l s were uneven. ANOVA only tests whether population means are s i g n i f i c a n t l y d i f f e r e n t and not which sample means are d i f f e r e n t . In order to establish which means were s i g n i f i c a n t l y d i f f e r e n t and the order of the magnitude of the differences the Student-Newman-Keuls (SNK) multiple range test was run when the ANOVA results indicated a s i g n i f i c a n t difference among means, at the 95% confidence l e v e l . The Mann-Whitney U-test (Siegal, 19 59) was the nonparametric test used to determine i f the cluster analysis groupings were s i g n i f i c a n t l y d i f f e r e n t at a 95% confidence l e v e l . Due to the d i f f i c u l t y in d i r e c t l y comparing the v a r i a b i l i t y of parameters expressed in d i f f e r e n t units and with d i f f e r e n t magnitudes of mean values i t i s common to express v a r i a t i o n as 12 CONCLUSIONS CONCLUSIONS CONCLUSIONS 1. Which f a c t o r s c o n t r i b u t e most t o v a r i a b i l i t y 2 . S i g n i f i c a n t d i f f e r e n c e s among f i e l d s 3. S i g n i f i c a n t d i f f e r e n c e s among d e p t h s 4 . S i g n i f i c a n t d i f f e r e n c e s among s e a s o n s 1 . B e s t s a m p l i n g method 2. Most c o s t e f f i c i e n t s a m p l i n g method 3 . R e q u i r e d s a m p l e I f o r g i v e n a c c u r a c y and p r e c i s i o n Most v a r i a b l e s e a s o n O r d e r o f v a r i a b i l i t y o f e l e m e n t s 1 . L e s s c o s t l y p r e d i c t i o n B e t h o d s 2. F i e l d s e p a r a t i o n I n t o management u n i t s 3 . S e p a r a t i o n o f f i e l d s ( u n i q u e n e s s o f f i e l d s ) F i g u r e 3.0 STATISTICAL METHODS COMPARISON OF CONCLUSIONS p a r a m e t r i c vS n o n - p a r a m e t r i c the c o e f f i c i e n t of v a r i a t i o n (CV). CV is a unitless value defined as %CV = standard deviation/mean x 100% (Zar, 1974). Three related parameters were obtained using the MIDAS program (Fox and Guire, 1976). Correlation matrices for combinations of time, f i e l d , and depth were calculated for a l l elements. Where c o r r e l a t i o n results showed an r value of greater than .6 a scattergram and regression equation were generated. The assumption for the determination of regression i s that the values of the dependent value are obtained without error or are small in comparison to the dependent variable. P r i n c i p a l component analysis (PCA) was also performed using MIDAS. Webster (1977) writes that PCA has proven to be one of the most valuable means of exploring relationships among s o i l p r o f i l e s . PCA is a non-parametric n-dimensional data analysis. Values were unweighted and an orthogonal axis rotation was used. The value of PCA is that i t allows the user to reduce the number of variables to be worked with. The variables chosen w i l l be the ones, which according to PCA, explain the largest amount of variance for each p r i n c i p a l component. Variables are selected according to the size of the contribution they make to the component axis. A latent vector value or eigenvalue with an absolute value close to one indicates a large contribution of that variable to the v a r i a b i l i t y of data points around that p r i n c i p a l axis. In other words i t i s c l o s e l y aligned and not nearly at right angle to the p r i n c i p a l axes. Parameters which make major contributions to the various p r i n c i p a l axes are selected to be used i n cluster analysis. The data output also indicates the percentage of tota l variance explained by each 14 nearly at right angle to the p r i n c i p a l axes. Parameters which make major contributions to the various p r i n c i p a l axes are selected to be used in cluster analysis. The data output also indicates the percentage of to t a l variance explained by each p r i n c i p a l component. I t i s generally desirable to chose as many components as required to explain 65-75% of the t o t a l variance. For this study, that number was generally less than f i v e . Cluster analysis was performed by UBC CGROUP (Patterson and Whitaker ,1978) . This program was used in combination with UBC CORDER. CORDER orders the o r i g i n a l data set i n a manner which i s comparable to the grouping order. Clustering i s a multivariate s t a t i s t i c a l method which numerically assesses which individuals or groups of individuals are most s i m i l a r , in order to create a hi e r a r c h i c a l c l a s s i f i c a t i o n . CGROUP is based on Ward's method (1963) which i s an average neighbour grouping which minimizes the error sums of squares of the distance between groups or ind i v i d u a l s . Output included a dendrogram of the grouping and error terms associated with each grouping step. The error term gives the value of the error sums of squares of the distance between the elements grouped. When the error term takes a jump in magnitude i t suggests that a jump to another hierarchy has been taken . 15 CHAPTER I SOURCES OF VARIABILITY INTRODUCTION The experiment was designed to determine whether there were s i g n i f i c a n t differences among f i e l d s , depths, seasons, sampling methods, and i n d i r e c t l y , among parent materials. Also, to quantify the r e l a t i v e contribution of each of these factors to ove r a l l v a r i a b i l i t y as an aid in understanding their impact on f e r t i l i z e r recommendations resulting from s o i l sampling. Analysis of variance (ANOVA) was the primary method used to determine these differences and the r e l a t i v e contributions of these factors to v a r i a b i l i t y . A l l analyses in this chapter used data from random s t r a t i f i e d sampling. In the study there was an opportunity to determine whether there are s i g n i f i c a n t differences between f i e l d s on the same s o i l series as well as among f i e l d s on d i f f e r e n t s o i l s e r i e s . The difference between f i e l d s on the same series was investigated i n two ways. The f i r s t method involved only samples from the plow l a y e r . This determined whether s i m i l a r i t i e s or differences inherited from the parent material would be overcome by management used on a p a r t i c u l a r f i e l d . The other involved the comparison of chemistry of the r e l a t i v e l y unaltered s o i l at the 60-90 sampling depth. This determined i f there were differences in chemistry of two f i e l d s within the same parent material. 16 Since there were s i g n i f i c a n t interactions among f i e l d s and depth the v a r i a b i l i t y of s o i l chemistry with depth was examined by p l o t t i n g mean f i e l d values against depth. The c o e f f i c i e n t of va r i a t i o n (CV) was also presented to compare the CVs of f i e l d s and depths. F i n a l l y , the influence of time of sampling on s o i l chemical v a r i a b i l i t y was considered. The eff e c t of sampling time on chemical v a r i a b i l i t y , in par t i c u l a r spring versus f a l l , should be elucidated since i t w i l l determine the preferred time of s o i l sampling of a g r i c u l t u r a l f i e l d s . Though sampling f i r s t i n spring and l a t e r i n the f a l l after cropping does not allow for a v a l i d comparison of f i e l d mean values, from a f e r t i l i t y point of view, i t does allow for a comparison of v a r i a b i l i t y during the two periods. The ef f e c t of time, and perhaps management, on s o i l chemical v a r i a b i l i t y i s i l l u s t r a t e d by graphically comparing the CV of f i e l d chemistry prior to cropping in spring and after cropping in the f a l l . Mclntyre (1967) as a resu l t of his review of l i t e r a t u r e on the subject concluded that time was not a serious factor i n s o i l sampling. This may be true i n a general sense but the same assumption should not be made on an area s p e c i f i c basis. In an area such as the Lower Fraser Valley where winter conditions are not l i k e l y to stop chemical reactions from proceding for prolonged periods, i t may be possible that s o i l chemical v a r i a b i l i t y w i l l change s i g n i f i c a n t l y over the winter period. The differences in v a r i a b i l i t y , i f great in magnitude, should be considered in the s e l e c t i o n of s o i l sampling time. 17 The effect of sampling method i s analysed in a separate section because the v a l i d i t y of the comparison i s questionable from a s t a t i s t i c a l point of view. 18 O V E R A L L V A R I A B I L I T Y Analysis of variance was f i r s t carried out using data from three f i e l d s , t h r e e depths, and two sampling times. Despite highly s i g n i f i c a n t differences among f i e l d s and depths for most variables and f i v e of the variable for seasons, inte r p r e t a t i o n of the main e f f e c t s , f i e l d s , depths, and seasons, could not proceed as normal, due to s i g n i f i c a n t i n t e r a c t i o n of f i e l d x season, f i e l d x depth, or season x depth, for most parameters (Table 1-1) . Only three of the f i e l d s were used in these analyses because the fourth, CLPRO, was sampled only in the spring. F a l l sampling was not carried out since the farmer had limed prior to the planned sampling period. When there are s i g n i f i c a n t i n teractions, s i g n i f i c a n t differences at the higher l e v e l can not be simply interpreted. For example, when there i s a s i g n i f i c a n t i n t e r a c t i o n among f i e l d s and seasons, an F-value which shows s i g n i f i c a n t differences among f i e l d s can not be interpreted as being e n t i r e l y due to differences among f i e l d s , since i t could also include a season contribution. An alternative to determining s i g n i f i c a n t differences through the use of F-values the r e l a t i v e importance of each factor can be determined by di v i d i n g the factor's sums of squares by the t o t a l sums of squares. The bottom portion of table 1-1 provides a summary of the percentage contribution to t o t a l sums of squares of each of the fact o r s , from the preceding ANOVA re s u l t s . 19 Fields plus depth explain 65% or more of the variance for Ca, K, pH, %C, and P. Depth i s espe c i a l l y important for P and K, explaining 60% and 65% of t o t a l v a r i a b i l i t y respectively. R e l a t i v e l y large percentages of the t o t a l variance f o r p H and %C being explained by f i e l d s (62% and 32%) indicates that differences among f i e l d s are large for these variables. Judging from the contribution of season to the t o t a l sums of squares (always less than 2.1%) i t can said that this main factor has a r e l a t i v e l y small contribution to t o t a l variance. The r e l a t i v e l y large percent sums of squares due to error for Mg, Na, CEC, and NO -N indicate that within f i e l d v a r i a b i -l i t y i s large compared to among f i e l d s . This does not mean that their CV w i l l be e s p e c i a l l y large but suggests that they would not be good d i f f e r e n t i a t i n g c r i t e r i a . 20 Table 1-1 Sources of v a r i a n c e , F - p r o b a b i l i t y v a l u e , and percent of t o t a l v a r i a n c e a t t r i b u t a b l e to each source of va r i a n c e f o r 3 f i e l d s sampled at 3 depths i n i n s p r i n g and f a l l . F - p r o b a b i l i t y values SOURCE Ca Mg Na K CEC pH %C NO3 -N p FIELD 0 .0000 0 .0000 0 .0000 0 .0000 0 .0000 0 .0000 0.0000 0.0000 0 .OOOO SEASON 0 .2475 0 .0095 0 .0086 0 .6292 0 .0007 0 .0000 0.0001 0.9441 0 .1484 FxS 0 .0046 0 .0000 0 .0000 0 .0001 0 .0073 0 .1149 0.0848 0.0000 0 .0003 DEPTH 0 .0000 0 .1955 0 .0005 0 .0000 0 .0000 0 .0000 0.0000 0.0000 0 .0000 FxD 0 .0221 0 .0000 0 .0000 0 .0000 0 .0021 0 .0000 0.0001 0.0001 0 .0000 SxD 0 .5517 0 .3065 0 .8133 0 .3005 0 .3439 0 .3015 0.5255 0.0000 0 .6795 FxSxD O.4130 0 .0618 0 .8421 0 .3839 0 .9336 0 .6123 0.7975 0.0000 0 .0000 PERCENT OF TOTAL SUMS OF SQUARES source of 4SS Ca Mg Na K CEC pH %C NO3-N P FIELD 21.8 24 .9 11.8 11. 3 25.8 62 . 0 32 . 0 7.7 10.3 SEASON 0.1 1.4 1.6 0.02 1.9 2.0 1.5 0.O1 0 .1 FxS 1.1 5 .1 6 .2 1.5 1.6 0.01 0.4 6 .0 0.9 DEPTH 47.8 0.7 3.8 60.4 22.8 21.0 39.5 15 . 0 65 .2 FxD 1.1 6.7 8 . 1 3.9 2.7 2.7 2 .1 3.8 5.4 SxD 0.1 0.5 1.0 0.2 0.3 1.1 0.1 5 . 9 0.04 FxSxD 0.4 0.6 0.4 0.4 0.1 0.01 0.1 19.4 2.2 ERROR 27.6 60 .0 68 .0 22.2 44.8 11.9 24 .3 42 . 1 15.8 T o t a l number of ob s e r v a t i o n i s 306, F - p r o b a b i l i t y values of 0.05 or l e s s are s i g n i f i c a n t at 0.05 l e v e l DETERMINING SIGNIFICANT DIFFERENCES AMONG SAMPLING METHODS Determination of s i g n i f i c a n t differences among the three sampling method's a b i l i t y to estimate f i e l d mean values could not be accomplished using a standard analyis of variance program. Instead, U.B.C. GENLIN (Greig and Bjerring, 1980) was used. GENLIN is a general l e a s t squares analysis program which i s designed to accommodate unbalanced sample numbers. GENLIN was run on plow-layer samples, from four f i e l d s , c o l l e c t e d i n spring. The results are summarized in Table 1-2. For the analysis of differences among sampling methods i t should be r e c a l l e d that from each f i e l d , 17 samples were co l l e c t e d for the random s t r a t i f i e d method, three samples for the random s t r a t i f i e d composite method, and one for the conventional method. I t should also be noted that the normality and homoscedasticity assumptions of analysis of variance become c r i t i c a l when the analysis is unbalanced. In order to check for homoscedasticity of the sampling methods, the B a r t l e t t chi-square test was used. In Table 1-2 the chi-square test results show that the sampling methods' variances are s i g n i f i c a n t l y d i f f e r e n t for Ca, Mg, and P. A Chi-square value of less than 0.05 indicates that the sampling methods do not have equal variances, therefore, the results of the F-test may not be v a l i d . 22 With regard to the other parameters, the r e l i a b i l i t y of the homogeneity of variance t e s t , when such low sample numbers are included, i s suspect. Departures from normality, which are expected, also diminish the v a l i d i t y of the variance t e s t , therefore, in t e r p r e t a t i o n of these r e s u l t s , which ultimately a f f e c t the in t e r p r e t a t i o n of the F-test results must be done cautiously. The F-probability values for sampling method ( Table 1-2) indicate that there are s i g n i f i c a n t differences among sampling methods for only Mg and P. However, due to s i g n i f i c a n t interactions of f i e l d x sampling method for Ca, K, and NO3-N, the sampling method for these variables may play an important role i n some f i e l d s while not i n others, so no d e f i n i t i v e statement can be made for these variables. The % sums of squares table (in Table 1-2) indicates that differences among f i e l d s account for the major part of the variance. Sampling methods accounted for 3.6% and 6% of the t o t a l variance for Mg and p respectively. For the other parameters, method accounted for less than 1.5% of the t o t a l variance. The large residual sums of squares for K suggests that within f i e l d v a r i a b i l i t y i s great, compared to among f i e l d v a r i a b i l i t y . 23 Table 1-2 Sources of v a r i a n c e , ( i n c l u d i n g sampling method) F - p r o b a b i l i t y v a l u e s , homogeneity of va r i a n c e r e s u l t s , and percent of t o t a l v a r i a n c e a t t r i b u t a b l e to each source of v a r i a n c e (for 4 f i e l d s ' plow l a y e r samples i n s p r i n g ) . F-PROBABILITY VALUES SOURCE Ca Mg Na K CEC pH %C NO3-N P FIELD O.OOOO 0.0000 0.0002 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 SAMP METH 0.1599 0.0096 0.2455 0.4800 0.5822 0.9254 0.8738 0.9188 0.0005 Fx SM 0.0000 0.0010 0.0570 0.0391 0.9995 0.9977 0.7482 0.0014 O.0014 HOMGENEITY OF VARIANCE ( B a r t l e t t Chi-square p r o b a b i l i t y values) SOURCE Ca Mg Na K CEC pH %C NO3-N P FIELD 0.000 0.000 0.000 0.027 0.000 0.175 0.000 0.000 0.000 SAMP METH 0.006 0.024 0.138 0.304 0.779 0.855 0.878 0.846 0.044 % SUMS OF SQUARES SOURCE Ca Mg Na K CEC pH %C N0 3 -N P FIELD 78 .1 60 .8 74.0 24.7 87.4 75 .6 86 .9 75 .1 63. 1 SAMP METH 0 .6 3.6 0.8 1.3 0.2 0 .05 0 .04 0 .05 6. 0 SMxF 9 .3 9.3 3.6 12.0 0.03 0 .09 0 .09 1 .13 8. 0 RESIDUAL 12 .0 26.2 20.78 62.0 12.0 24 .0 13 .0 23 .7 23. 0 Number of observations = 51, SAMP METH= sampling method, F = f i e l d , F - p r o b a b i l i t y values of 0.05 or l e s s i s s i g n i f i c a n t l y d i f f e r e n t at 0.05 l e v e l DIFFERENCES AMONG F I E L D S : IN THE PLOW LAYER D i f f e r e n c e s among f i e l d s were d e t e r m i n e d i n two w a y s . One was o n t h e b a s i s o f p l o w l a y e r d a t a and t h e o t h e r o n t h e b a s i s o f 60 /9 0 cm d a t a . I n b o t h c a s e s t h e d a t a f o r a n a l y s i s o f v a r i a n c e was f i r s t p a r t i t i o n e d by d e p t h a n d s e a s o n t o r e m o v e t h e e f f e c t o f i n t e r a c t i o n s . T a b l e 1-3 i n d i c a t e s h i g h l y s i g n i f i c a n t d i f f e r e n c e s among f i e l d means i n b o t h s p r i n g a n d f a l l f o r a l l n i n e p a r a m e t e r s . T h e S t u d e n t - N e w m a n - K e u l s r a n g e t e s t ( T a b l e 1-4 a n d 1-5) s h o w s w h i c h f i e l d s h a v e s i g n i f i c a n t l y d i f f e r e n t mean v a l u e s f o r a n y g i v e n p a r a m e t e r . I n s p r i n g P i s t h e o n l y p a r a m e t e r s h o w i n g a s i g n i f i c a n t d i f f e r e n c e among a l l f o u r f i e l d s . I n t h e f a l l ( T a b l e 1-5) t h e r e i s a s i g n i f i c a n t d i f f e r e n c e among a l l t h r e e f i e l d f o r K and p H . T h e two f i e l d s t h a t a r e f o u n d o n t h e same s o i l s e r i e s (RN & RC) show s i m i l a r mean v a l u e s f o r f i v e p a r a m e t e r s i n s p r i n g a n d o n l y f o r two p a r a m e t e r s i n t h e f a l l . A c o m p a r i s o n o f t h e % sums o f s q u a r e s i n s p r i n g a n d f a l l , f o r t h e v a r i o u s p a r a m e t e r s , c a n be u s e d t o p r o v i d e a s u g g e s t i o n f o r t h e b e s t s a m p l i n g p e r i o d . I d e a l l y , f i e l d s w o u l d be s a m p l e d w h i l e t h e w i t h i n f i e l d v a r i a n c e i s l o w , a s i n d i c a t e d by a r e l a t i v e l y s m a l l % sums o f s q u a r e s due t o e r r o r . T h e r e l a t i v e m a g n i t u d e s o f t h e e r r o r t e r m s , i n s p r i n g a n d f a l l , i n d i c a t e t h a t s p r i n g s a m p l i n g w o u l d be m o s t a p p r o p r i a t e f o r C E C , % C , N O 3 - N , a n d P , w h i l e f a l l s a m p l i n g w o u l d be r e c o m m e n d e d f o r K . T h e s e r e c o m m e n d a t i o n s a r e i n a g r e e m e n t w i t h t h o s e r e s u l t i n g f r o m c a l c u l a t i o n s of . s a m p l e n u m b e r s r e q u i r e d t o p r o v i d e a g i v e n l e v e l 25 of accuracy and confidence a t the two sampling times. These sampling requirements are provided i n chapter two. The p r o p o r t i o n s of sums of squares a t t r i b u t a b l e to the d i f f e r e n c e among f i e l d s and w i t h i n f i e l d s , f o r K, pH, %C, and P, are very s i m i l a r to those found by C i r p a et a l (1972). However, the c o n s i d e r a b l e d i f f e r e n c e s between the r e s u l t s of the two s t u d i e s i n f a l l i n d i c a t e the importance of i n d i c a t i n g sampling time i n order to make v a l i d comparisons of data s e t s . The graphs i n F i g u r e s 1-2 and l-2b p o r t r a y the mean and range of v a l u e s , f o r RN, RC, and HRST i n the plow l a y e r . Spring and f a l l data are combined i n order to provide a more gen e r a l p i c t u r e . The data used i n these graphs can be found i n the appendix. 26 Table 1-3 ANOVA r e s u l t s showing p r o b a b i l i t y of s i g n i f i c a n t d i f f e r e n c e s among f i e l d s and percent of v a r i a n c e a t t r i b u t a b l e to d i f f e r e n c e s among f i e l d s and e r r o r , a t two d i f f e r e n t sampling times (plow l a y e r samples). ANOVA RESULTS FOR 4 FIELDS IN SPRING F-PROBABILITY VALUES SOURCE Ca Mg Na K CEC pH %C N O 3-N P FIELD 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 % SUMS OF SQUARES SOURCE Ca Mg Na K CEC pH %C NO3-N P FIELD 79.9 58.4 76.2 18.8 86.3 72.3 85.1 72.2 63.7 ERROR 20.1 41.6 23.8 81.1 13.7 27.7 14.9 26.8 36.3 ANOVA RESULTS FOR 3 FIELDS IN FALL SOURCE Ca Mg Na K CEC pH %C NO3-N P FIELD 0.0000 0.0001 0.0000 0.0000 0.0016 0.0000 0.0000 0.0098 0.0015 % SUMS OF SQUARES SOURCE Ca Mg Na K CEC pH %C NO3-N P FIELD 44.4 31.8 63.0 72.5 23.7 80.5 54.5 17.5 24.0 ERROR 55.6 68.2 37.0 27.5 76.3 19.5 45.6 82.4 76.0 Number of observations i s 68 i n s p r i n g and 51 i n f a l l , F - p r o b a b i l i t y v alue of 0.05 or l e s s i s s i g n i f i c a n t at 0.05 l e v e l Table 1-4 Student-Newman-Keuls range test for four f i e l d s ' plow layer samples i n spring. ELEMENT FIELD * Ca RN RC HRST CLPRO Mg RN RC HRST CLPRO Na HRST RN CLPRO RC K HRST RC RN CLPRO CEC HRST RC RN CLPRO pH RN CLPRO RC HRST %C RC HRST RN CLPRO NO3-N RC CLPRO RN HRST P HRST RN RC CLPRO * underlined f i e l d means do not d i f f e r s i g n i f i c a n t l y at the 5% l e v e l , f i e l d s are arranged in order of increasing mean values. 28 Table 1-5 Student-Newman-Keuls range t e s t f o r three f i e l d s ' plow l a y e r samples i n f a l l . ELEMENT FIELD * Ca RN RC HRST Mg RN RC HRST Na HRST RC RN K HRST RN RC CEC HRST RC RN pH RN RC HRST %C HRST RC RN NO3-N HRST RN RC P HRST RN RC * u n d e r l i n e d f i e l d means do not d i f f e r s i g n i f i c a n t l y a t the l e v e l , f i e l d s are arranged i n order of i n c r e a s i n g mean v a l u e s . 29 co o - Z 1 8 O o e 12 03 g O o g 20 tr 10 S RC HRST RN RC HRST RN RC HRST RN Figure 1 -1 Comparison of three f i e l d s ' mean values for Ca, Mg, K, and CEC in the plow layer, error bars indicate range of observations, spring and f a l l data combined. co PH RC HRST RN RC HRST RN Figure 1-lb Comparison of three f i e l d s ' mean values for %C, pH, P, and NO -N in the plow layer, error bars indicate range of observations, spring and f a l l data combined. DIFFERENCES AMONG FIELDS: AT THE PARENT MATERIAL LEVEL The samples taken from 60/90 cm are thought to best represent the chemistry of the parent material, or the s o i l r e l a t i v e l y unaltered by management practices. As in the case of plow layer samples, data from spring and f a l l was analysed separately. The results of the F-test (Table 1-6) show that there are s i g n i f i c a n t differences among f i e l d s for a l l 9 parameters, in both spring and f a l l . Though differences in f i e l d chemistry at the 60/90 cm depth are of l i t t l e consequence to those interested in crop production, they may, nonetheless, be of some consequence to those people involved in s o i l c l a s s i f i c a t i o n . Also a comparison of differences among f i e l d s at the parent material depth and the plow layer depth w i l l indicate i f the eff e c t of management has been able to mask differences inherited from parent material, or i f i t has had l i t t l e e f f e c t . Results of the Newman-Keul•s range te s t (Table 1-7 & 1-8) show that no single parameter w i l l d istinguish among a l l f i e l d s in the spring and that only Ca and %C w i l l distinguish among a l l f i e l d s in the f a l l . While the range test indicates RN and RC (found on the same s o i l series) plow layer samples having f i v e parameters which have a si m i l a r mean value in spring and only two parameters with similar mean value in the f a l l , the 60/90 cm samples have s i x parameters with s i m i l a r mean values in both spring and f a l l . As be expected, this indicates that management tends to mask s i m i l a r i t i e s inherited from parent material. 32 Table 1-6 ANOVA r e s u l t s showing p r o b a b i l i t y of a s i g n i f i c a n t d i f f e r e n c e among f i e l d s and percent of v a r i a n c e a t t r i b u t a b l e to d i f f e r e n c e s among f i e l d s at two d i f f e r e n t sampling times (60-90 cm samples). ANOVA RESULTS FOR 4 FIELDS IN SPRING F-PROBABILITY VALUES SOURCE Ca Mg Na K CEC pH %C NO3-N P FIELD 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 % SUMS OF SQUARES SOURCE Ca Mg Na K CEC pH %C NO3-N P FIELD 47.6 55.1 41.8 79.2 46.6 86.1 40.6 58.9 25.3 ERROR 52.4 44.9 58.2 20.8 53.4 13.9 59.3 41.1 74.7 ANOVA RESULTS FOR 3 FIELDS IN FALL F-PROBABILITY VALUE SOURCE Ca Mg Na K CEC pH %C NO3-N P FIELD 0.0000 0.0046 0.0007 0.0000 0.0016 0.0000 0.0000 0.0005 0.0000 % SUMS OF SQUARES SOURCE Ca Mg Na K CEC pH %C NO3-N P FIELD 68.5 12.1 26.5 49.9 67.5 84.4 77.9 27.7 77.9 ERROR 31.5 87.9 73.5 50.1 32.5 15.6 22.1 72.3 22.1 Number of observations i s 68 i n s p r i n g and 51 i n f a l l , F - p r o b a b i l i t y v alue of 0.05 or l e s s i s s i g n i f i c a n t at 0.05 l e v e l . Table 1-7 Student-Newman-Kewls range te s t for four f i e l d s 60-90 cm samples i n spring. ELEMENT FIELD * Ca RN RC HRST CLPRO Mg RC RN HRST CLPRO Na HRST RC CLPRO RN K HRST RC RN CLPRO CEC HRST RC RN CPRO pH RN CLPRO RC HRST %C HRST RC RN CLPRO NO3-N CLPRO RC HRST RN P HRST RN RC CLPRO * underlined f i e l d means do not d i f f e r s i g n i f i c a n t l y at the 5% l e v e l , f i e l d s are arranged in order of increasing mean values. 34 Table 1-8 Student-Newman-kewls range test for three f i e l d s ' 60-90 cm samples i n f a l l . ELEMENT FIELD * Ca RN RC HRST Mg RC HRST RN Na HRST RC RN K HRST RC RN CEC HRST RC RN PH RN RC HRST %C HRST RC RN NO3-N RC RN HRST P HRST RN RC * underlined f i e l d means do not d i f f e r s i g n i f i c a n t l y at the 5% l e v e l , f i e l d s are arranged in order of increasing mean values. 35 TIME AS A SOURCE OF VARIABILITY I t is generally acknowledged that time may be a confounding factor in the int e r p r e t a t i o n of s o i l s data ( Haines and Cleveland, 1981; Peterson and Rolfe, 1982). Though experimental constraints on sampling did not allow for the ideal sequence of sampling times, i t did allow for the development of some insight into the effects of time on the v a r i a b i l i t y of s o i l chemistry and, therefore, on sampling r e l i a -b i l i t y . Ideally,sampling would have f i r s t taken place in the f a l l followed by a winter season and then spring sampling. This design would have allowed comparison of the a b i l i t y of two sampling periods to estimate the f i e l d ' s s o i l chemistry and f e r t i l i z e r re-quirements for the same cropping season. Due to the order of sampling times, a comparison of mean values for the plow layer samples would be of l i t t l e benefit. However,a comparison of the v a r i a b i l i t y at the two d i f f e r e n t times may provide some answers as to the preferred sampling time. S o i l chemistry would be best estimated when the v a r i a b i l i t y of the parameters to be measured was lowest. I f f e r t i l i t y parameters are found to be more variable in spring than i n f a l l , then an i d e n t i c a l number of spring samples w i l l provide a less accurate estimate of the f i e l d chemistry. Figures 1-2 through 1-3 show that, for any given parameter, the CV for a l l three f i e l d s does not always show the same d i r e c t i o n of change from spring to f a l l . For plow layer chemistry, the va r i a t i o n from spring to f a l l 36 i s generally of a larger magnitude for NO3, P, and K than for Ca, PH and CEC. Ca, NO3-N, K, and pH have a l l three f i e l d s showing the same d i r e c t i o n of v a r i a b i l i t y change from spring to f a l l while, for P and CEC, the d i r e c t i o n of v a r i a b i l i t y change over time i s not the same for a l l three f i e l d s . The CV for Ca and K i s lower i n f a l l than in spring for a l l three f i e l d s . For NO3-N and pH, v a r i a b i l i t y i s lower i n spring, though pH is only marginally so. CEC and P v a r i a b i l i t y i s considerably lower i n spring for two of the three f i e l d s . The 30/60 and 60/90 cm depths show trends similar to the plow layer. 37 Figure 1-2 Changes in percent CV from spring to f a l l for three f i e l d s ' plow layer samples. Figure 1-3 Changes in percent CV from spring to f a l l for three f i e l d s ' 30-60 cm samples. Figure 1-4 Changes in percent CV from spring to f a l l for three f i e l d s ' 60-90 cm samples. COMPARING THE VARIABILITY OF THREE DIFFERENT SAMPLING DEPTHS Because of a s i g n i f i c a n t f i e l d x depth i n t e r a c t i o n a simple comparison of depths, through the use of F - r a t i o s , could not be carri e d out. Instead, data results are compared graphically. F i g . 1-9 through 1-14 show mean values plotted against depth; along side are the respective CV values plotted against depth. The plots are for three f i e l d s and for s i x parameters. For the s i x variables shown in the Figures 1-5 through 1-10 the plow layer generally shows lower v a r i a b i l i t y than either the 30/60 or 60/90 cm depths. RN has the most exceptions, with CEC, pH, and P being more variable i n the plow layer, than at the other two depths. The higher v a r i a b i l i t y of p and pH in the plow layer for RN may be the resu l t of uneven d i s t r i b u t i o n of f e r t i l i z e r and lime. The high v a r i a b i l i t y at 30/60 and 60/90 cm may be a re s u l t of superimposing high l e v e l s of P, K, or Ca on lower background le v e l s Mclntyre (1979). The additions of K f e r t i l i z e r or lime, could result i n a large v a r i a t i o n of values through d i f f e r e n t i a l leaching or mixing into the p r o f i l e . The high v a r i a b i l i t y of highly immobile phophorus noted in the 30/60 cm depth could be a res u l t of plowing that has occasionally incorporated s o i l with high P levels into the 30/60 cm depth. Mixing of s o i l s with high and low P values would increase v a r i a b i l i t y . 41 NO3-N in RC is also most variable at the plow layer depth. Cameron et a l (1979) note that many factors influence the v a r i a b i l i t y of NO3-N but suggest that d i f f e r e n t i a l leaching rates may be one of the main causes of NO3 -N v a r i a b i l i l t y . The v a r i a b i l i t y at 60/90 cm is of intere s t since i t should be most c h a r a c t e r i s t i c of the parent material. Consideration of the r e l a t i v e l y unaltered samples from this depth enables one to determine i f high or low v a r i a b i l i t y of certain parameters i s ch a r a c t e r i s t i c of any of the studied parent materials. Since HRST is located on a mapping association, while RN and RC are located on a pure mapping unit, HRST may be expected to have higher v a r i a b i l i t y at the parent material l e v e l . Of the s i x plotted parameters, the %CV for CEC, NO3-N and P are within 3% of one another for RN and RC. These three parameters have a low CV for RN and RC when compared with those found on HRST. For the other 3 parameters there i s no clear pattern associated with expected map unit v a r i a b i l i t y . The description of changes in mean values with depth i s best divided into two parts: one for parameters whose difference between mean values of the plow layer and the 60/90 cm depth takes place almost e n t i r e l y between the plow layer and the 30/60 cm depths; and another for those parameters whose mean values decline at a r e l a t i v e l y even rate through the p r o f i l e . For P and K, 70 to 80% of the difference between plow layer mean values and 60/90 cm mean values takes place between the plow layer and the 30/60 cm depth. For Ca, CEC, pH, and NO3-N, the decline of mean values i s r e l a t i v e l y even from the plow layer through to the 60/90 cm depth. 42 Mclntyre (1979) also noted a dramatic decline of P and K values with depth. As a r e s u l t , he suggested that the depth of sampling should be consistent with the depth used to establish the s o i l test c a l i b r a t i o n . If sampling i s to a greater depth, than that used in the c a l i b r a t i o n , the available P w i l l be underestimated and the f e r t i l i z e r recommendation overestimated. 43 Figu re 1-5 F i e l d mean value and %CV for Ca for three f i e l d s at three depths o * 32 0/30 30/60 60/90 DEPTH 0/30 30/60 60/90 DEPTH Figure 1-6 F i e l d mean value and %CV for K for three f i e l d s at three depths 22 20 18 O O 16 6 O 14 UJ O 12 10 0/30 30/60 60/90 DEPTH 0/30 30/60 60/90 DEPTH Figure 1-7 F i e l d mean value and %CV for CEC for three f i e l d s at three depths Figure 1-8 F i e l d mean value and %CV for pH for three f i e l d s at three depths Figure 1-9 F i e l d mean value arid %CV for NO3 -N for three f i e l d s at three depths i g u r e 1 - 1 0 F i e l d mean v a l u e a n d %CV f o r P f o r t h r e e f i e l d s a t t h r e e d e p t h s SUMMARY P a r t i t i o n i n g of the sums of squares, among a l l factors, indicates that time contributes a r e l a t i v e l y small portion to t o t a l v a r i a b i l i t y . While Mclntyre (1967) suggest that time i s not a serious factor i n s o i l testing Anderson and Tiedman (1970) and Peterson and Rolfe (1982) found that there were s i g n i f i c a n t periodic variations in s o i l chemistry. Of the other two main fa c t o r s , depth and f i e l d s , depth i s the more important source of variance for Ca, K, %C, NO3-N, and P. The r e l a t i v e l y large error term for Mg, Na, CEC, and NO3-N suggest that within f i e l d variance i s large for these variables. Though time was not a major contributor to t o t a l variance, i t s influence may be great enough to cause consideration of i t s effect on sampling r e l i a b i l i t y . For the studied f i e l d s , i t was observed that the CV for P was lowest i n spring prior to cropping whereas the CV for K is lowest i n the f a l l after cropping. The magnitude of change in v a r i a b i l i t y for CEC and pH, from spring to f a l l , was much less than for P and K and i s , therefore, l i k e l y to be of no p r a c t i c a l s i g n i f i c a n c e . The eff e c t of changing CV on sampling requirement and r e l i a b i l i t y w i l l be addressed i n chapter two. The large contribution of depth to t o t a l v a r i a b i l i t y for K and P, as indicated by the high percentage of the t o t a l sums of squares (60.4 and 65.2% respectively) indicate that the depth to which s o i l samples are taken w i l l strongly influence the s o i l sample value. Although Mclntyre (1979) did not determine the percentage contribution of depth to the ov e r a l l v a r i a b i l i t y he did note the marked influence of sampling depth on P and K 50 v a r i a b i l i t y . This would suggest that care be taken to ensure that s o i l samples are taken to the same depth as was used to develop s o i l test correlations. An ANOVA program capable of handling unbalanced sample numbers was used to determine i f there were s i g n i f i c a n t differences among mean values predicted by the three d i f f e r e n t sampling methods, s t r a t i f i e d random, s t r a t i f i e d random compos-i t e , and conventional. The results indicate that there were s i g n i f i c a n t differences among methods for Mg, and P. Departures from the assumptions of ANOVA, added to very unbalanced sample numbers, require that the results indicating no s i g n i f i c a n t differences among methods for other variables be interpreted cautiously. B a l l and Williams (1971) made a comparison of several composite sampling methods but made no attempt to determine i f the differences among methods were s t a t i s t i c a l l y s i g n i f i c a n t . When spring data i s compared to the findings of Cirpa et a l (1971) the percent sums of squares attributable to differences among f i e l d s i s very s i m i l a r . S i g n i f i c a n t differences among f i e l d means as well as considerable differences in CV among f i e l d s , indicate that an i d e n t i c a l sampling program carried out on a l l f i e l d s w i l l produce highly variable estimates of f i e l d means. Lee et al (1975) also reported s i g n i f i c a n t differences among f i e l d s for most chemical properties. On an ov e r a l l basis ( data from three f i e l d s , two seasons, and three depths combined, n=306) the order of v a r i a b i l i t y for the studied variables i s as follows: 51 OVERALL VARIABILITY %CV 19 29 40 61 63 68 78 109 168 PARAMETER pH CEC Mg %C Ca N 0 3 K P Na On a depth basis with data from three f i e l d s and two seasons combined the order of v a r i a b i l i t y changes to the following (n=102) . VARIABILITY IN THE PLOW LAYER %CV 13 23 35 36 38 39 47 71 245 PARAMETER pH CEC Mg %C Ca K P NO3 Na VARIABILITY AT 60/90 cm %CV 21 23 35 36 38 39 47 71 245 PARAMETER PH CEC Mg K P %C Ca NO3 Na On both an ove r a l l basis and on a plow layer basis the v a r i a b i l i t y of NO3-N, P, and K i s greater than four of the other parameters studied. The high v a r i a b i l i t y of NO3-N w i l l cause considerable problems in attempts to characterize i t . However, i n Alberta, where s o i l sampling for NO3-N is carried out, Cameron et al 1979 reported CVs for NO3-N ranging for 40 to 72. The range of CV values for P reported in the l i t e r a t u r e , for ag r i c u l t u r a l f i e l d s , i s from 10 to 83 (Peck and Dibb, 1979 ; Nelson and McKracken, 1962; Cirpa et a l , 1972; Beckett and Webster ,1971; Cameron et a l , 1971). The majority of the CV values f a l l around 50, which compares well with the findings of this study. 52 Beckett and Webster (1971) as a result of their review report a mean CV for of 35 for K. Other reported values range from 19 to 66 (Peck and Dibb, 1979; Cameron et a l , 1971; Adams and Wilde , 198 0) . The variables Ca, Mg, %C, and pH also present no surprises when compared with the findings of the above authors. A CV of 245 for Na i s somewhat anomalous. While most of the above studies reported a CV for Na between about 10 and 50 Lewis (1976) reported a CV for Na as high as 80 while Brown (1979) reported a CV of 158. In part the large CV for Na may be due to the r e l a t i v e l y small value of the Na l e v e l s combined with f a i r l y large errors in measurement. The major reason for the large value i s l i k e l y due to the i n c l u s i o n of data from three f i e l d s and two seasons. The RC f i e l d i n particular had a high mean value and a large CV. This added to s i g n i f i c a n t seasonal variations and lower mean values in other f i e l d s would r e s u l t i n a large CV. Due to differences in the size of sampling areas, sampling methods, s o i l extraction methods, depth of sampling, land use, and s o i l parent materials i t i s d i f f i c u l t to make t r u l y v a l i d comparisons of these results with those reported i n the l i t e r a t u r e . Campbell (1979) states that the results of the review of s o i l v a r i b i l i t y compiled by Beckett and Webster (1971) are uninterpretable due to the great differences among studies. 53 CHAPTER II COMPARISON OF SAMPLING METHODS  INTRODUCTION Since the use of ANOVA in chapter one did not s a t i s f a c t o r i l y resolve the question of differences in the a b i l i t i e s of the three d i f f e r e n t sampling methods to estimate mean values, the aim of this chapter is to make further comparisons of mean values obtained from the three d i f f e r e n t sampling methods. The methods w i l l also be examined with respect to their merits and demerits in terms of the accuracy of their mean values and costs associated with each method. The comparisons w i l l center on the major f e r t i l i t y parameters, N,P, and K. The results of conventional sampling and s t r a t i f i e d random composite sampling w i l l be compared to the mean values obtained by the random s t r a t i f i e d method. S t r a t i f i e d random sampling estimates the population without bias (Webster, 1977) and, there-fore, is assumed to provide the best estimate of f i e l d f e r t i l i t y l e v e l s . The conventional method consists of a single composite of 10 to 12 samples which are chosen to be representative of the f i e l d . The s t r a t i f i e d random composite consists of three composites which are each made up of one random sample from each of f i v e p l o t s . The s t r a t i f i e d random method consists of 17 samples with one sample having been randomly taken from each of 17 p l o t s . 54 Though the cost of the s t r a t i f i e d random sampling method i s much higher than either of the other two methods, due to increased plot lay out time and increased analysis cost, i t does provide v a r i a b i l i t y information. The benefits of the various sampling methods are d i f f i c u l t to put into d o l l a r terms. Ideally, the information gained from each of the three sampling methods would be put in terms of f e r t i l i z e r costs and the the value of the crop obtained. However, such data for the Lower Fraser Valley i s not r e a d i l y a v a i l a b l e . 55 COMPARING MEAN VALUES OF THREE  SAMPLING METHODS Table 2-1 and 2-2 show the various estimates of mean values for NO3-N, P f K , %C, and pH for the plow layer i n spring and in f a l l . Table 2-3 shows the percentage difference of the s t r a t i f i e d random f e r t i l i t y estimates from the s t r a t i f i e d random composite and the conventional means. If spring and f a l l data for NO3-N, P and K are considered together, the s t r a t i f i e d random composite estimates of the mean values range from an overestimate of about 50% to an underestimate of about 40%, when compared to s t r a t i f i e d random r e s u l t s . The s t r a t i f i e d random composite samples overestimated the mean value 19 times, underestimated i t 15 times and determined i t exactly once. Conventional sampling estimates of NO3-N, P , and K mean values range from an overestimate of 125% to an underestimate of 78%. I t estimates the s t r a t i f i e d random mean within 20% of the mean 65% of the time, within 20-50% of the mean 18% of the time and 50-125% of the mean 17% of the time. Though the random s t r a t i f i e d composite tended to underestimate the mean value more often than the conventional samples, i t was not subject to such large differences in the estimate of the mean value. Mean values estimated by the s t r a t i -f i e d random composite method never d i f f e r e d from the s t r a t i f i e d 56 random mean by more than 50%. The conventional method d i f f e r e d from the s t r a t i f i e d random mean values by more than 50% about 20% of the time. As a result of the low v a r i a b i l i t y the estimates of mean pH values by either method i s very accurate (< + 5%). %C has moderate errors of estimate, which are generally less than + %15. 57 Table 2-1 Comparison of mean values from three sampl methods ( f o r five plow layer parameters in f a l l ) . SAMPLING METHOD & MEAN VALUE STRATIFIED STRATIFIED RANDOM FIELD PARAMETER RANDOM COMPOSITE CONVENTIONAL RC 14.1 15.1 3.1 HRST NO3 -N 7.7 8.8 8.5 RN (ppm) 9.9 27.3 15.5 CLPRO 30.8 36.0 RC 90.5 60.3 73.0 HRST P 53.7 48.0 49.0 RN (ppm) 84.4 66.7 74.0 CLPRO 244.0 272. 0 RC 0.86 0.86 0.86 HRST K 0.36 0.44 0.39 RN (me/10Og) 0.74 0.63 0.74 CLPRO 0.90 0. 90 RC 2.1 2.0 1.9 HRST C 1. 8 2.3 2.1 RN (%) 3.2 3.3 3.1 CLPRO 23.0 22. 8 RC 5.6 5.6 5.9 HRST PH 6.6 6.3 6.9 RN 4.9 5.0 5.1 CLPRO 5.5 5.7 58 Table 2-2 Comparison of mean values from three sampling methods ( f o r fi v e plow layer parameters in spring). FIELD PARAMETER SAMPLING METHOD & MEAN VALUE STRATIFIED STRATIFIED RANDOM RANDOM COMP OSITE CONVENTIONAL RC 5.3 5.5 4.5 HRST NO3 -N 28.0 26.0 29.0 RN (ppm) 13.0 12.0 12.0 CLPRO 16.0 21.0 18. 0 RC 128 125 288 HRST P 44 40 36 RN (ppm) 70 84 84 CLPRO 153 222 242 RC 0.72 0.43 0.77 HRST K 0.49 0.48 0.47 RN (me/10Og) 0.74 0.68 0.69 CLPRO 0.81 1. 21 1.45 RC 2.2 2.1 2.0 HRST C 2.2 2.0 2.2 RN % 3.5 3.3 3.0 CLPRO 31.9 29.2 30.4 RC 5.9 6.0 6.0 HRST pH 6.6 6.5 6. 5 RN 5.1 5.1 5.1 CLPRO 5.3 5.2 5.3 59 Table 2-3 Percent difference of s t r a t i f i e d random composite and convention sampling means from the s t r a t i f i e d random mean in the spring and in the f a l l ( plus or minus indicates d i r e c t i o n of difference). SAMPLING METHOD & %DIFFERENCE FROM STRATIFIED RANDOM MEAN VALUE FIELD PAR-|AMETER STRATIFIED RANDOM CONVEN-COMPOSITE TIONAL STRATIFIED RANDOM CONVEN-COMPOSITE TIONAL RC + 3.8 -15.1 „ + 7.1 -78.0 HRST NO -N - 7.1 + 3.6 +14 .3 + 10.4 RN (ppm) - 7.7 - 7.7 +27.3 +56.6 CLPRO + 31.3 +12.5 RC - 2.3 +125.0 -30.2 -19.3 HRST P - 9.1 -18.2 -10.6 - 8.8 RN (ppm) +20.0 +20.0 -21.0 -11.8 CLPRO +45.1 + 58.2 RC -40.1 + 7.2 + 0.5 + 0.1 HRST K - 2.3 - 3.7 +23.0 + 8.1 RN (me/lOOg) - 6.3 - 5.4 +14.9 - 0.3 CLPRO + 49.6 +79.2 RC - 4.5 - 9.1 - 4.8 - 9.5 HRST C - 9.1 0.0 + 27.8 + 16.7 RN % - 5.7 -14.3 +3.15 - 3.1 CLPRO - 8.5 - 4.7 RC + 1.7 + 1.7 - 0.2 + 3.9 HRST pH - 1.5 - 1.5 - 3.8 + 4.1 RN 0.0 0.0 + 2.7 + 4.3 CLPRO - 1.9 0.0 60 STATISTICAL SAMPLING REQUIREMENTS In order to make a probability statement about sampling requirements for NO3-N, P, and K the Husch equation (Husch, M i l l e r and Beers, 1972) was used to calculate the number of samples required to obtain an estimate of the mean. The equation is as follows: t (n-1) x CV n= AE Where n i s the number of samples needed to estimate the mean with a s p e c i f i e d allowable error and probability, t i s the value of Student's t - d i s t r i b u t i o n with n-1 degrees of freedom, CV i s the c o e f f i c i e n t of v a r i a t i o n , and AE is the allowable sampling error in percent. A 90% confidence l i m i t and a 10% allowable error was used to produce the sample numbers shown i n Table 2-4. Cameron et al (1971) recommended an 80% accuracy with a precision l e v e l of ± 20%, but preferably ± 10%. Though the sample numbers appear to be horrendous in some cases, the required sample numbers drop dramatically i f the allowable error i s increased. Using the highly variable parameter NO3-N as an example, i t can be determined that only ten samples are required to estimate the mean in f a l l , when the allowable error i s set at 40% and the confidence l e v e l at 90%. Cameron et al (1971) found sampling requirements for P and K similar to those shown i n Table 2-4 even though the p r a i r i e f i e l d s studied were generally much larger. 61 Table 2-4 Sample numbers required to estimate mean value of plow layer parameters with ± 10% precision and 90% confidence. COMPARING SPRING & FALL REQUIREMENTS PARAMETER & SAMPLE # REQUIRED SEASON FIELD NO3-N P K S RC 91 20 58 F RC 103 49 13 S HRST 35 55 79 F HRST 167 30 11 S RN 43 26 46 F RN 167 95 22 S CLPRO 43 65 68 - Sample numbers required to estimate mean value of plow layer parameters for individual f i e l d s using combined spring and f a l l data with ± 10/Cprecision and 90% confidence. PARAMETER AND # OF SAMPLES REQUIRED FIELD NO3-N P K RC 191 38 32 HRST 191 38 63 RN 88 38 30 * Mean value plus or minus 10% with 90% confidence. Sample numbers calculated using Husch equation (Husch et a l , 1972) 62 COMPARISON OF COSTS ASSOCIATED WITH VARIOUS SAMPLING METHODS Since the cost of using the random s t r a t i f i e d and random s t r a t i f i e d composite methods are high, in comparison to conventional sampling, their use in simply predicting mean values i s questionable. Table 2-5 shows the difference in s o i l analysis cost associated with the various methods. The costs r e f l e c t the cost of having a l l samples analysed for P and K at $18 per sample and include a 10% discount for the analysis of 17 samples. Costs were quoted by P a c i f i c S o i l s Analysis Inc. in Oct. 1983 (Herman, personal communication) The conventional method has one sample analysed, the random s t r a t i f i e d composite cost is for the analysis of three samples, and the s t r a t i f i e d random cost i s for the analysis of 17 samples, for the two elements. Table 2-5 Cost associated with analysis for P and K for d i f f e r e n t sampling methods, for one f i e l d . STRATIFIED STRATIFIED RANDOM RANDOM COMPOSITE CONVENTIONAL COST $244.80 $48.00 $16.00 In order to help put these analysis costs in perspective i t should be noted that a %50 underestimate in s o i l P l e v e l , for a 63 s o i l with a medium s o i l test, would result in an over-recommendation of 67 kg/ha of V^0S ( see table 2-6). This would re s u l t in the application of $30 worth of extra f e r t i l i z e r per ha., given a f e r t i l z e r cost of $450 per tonne of 11-55-0. This adds up to $108 for a 3.6 ha. f i e l d . The cost of such an error i n the estimate for P levels alone would be approximately equal to the extra analysis cost for the random s t r a t i f i e d sampling method. 64 EFFECT OF ERROR IN MEAN VALUE ON FERTILIZER RECOMMENDATION Tables 2-6 and 2-7 give some idea of the importance of the d i r e c t i o n and magnitude of error i n p and K estimates. The tables were compiled on the basis of information presented i n the B.C. Mini s t r y of Agriculture's S o i l Testing and Interpretation Handbook (Neufeld, 1981) . A copy of the recommendations can be found i n the appendix 7 and 8. The f e r t i l i z e r recommendations for Crop Group 3 were chosen for the sake of c a l c u l a t i o n . The compiled tables are based on two hypothetical cases, one where the mean true s o i l test l e v e l for P and K i s medium and one where the true s o i l test l e v e l is high. The re s u l t i n g error i n f e r t i l i z e r recommendation due to errors of 10 to 60% in estimate of the mean for both s o i l test l e v e l s , i s calculated i n kg/ha. If the true mean i s underestimated then the f e r t i l i z e r recommenda-tion w i l l be too high as indicated by the + sign. If the estimate i s high the recommendation w i l l be below the actual required amount. This i s indicated by a - sign. Tables 2-6 and 2-7 show that when a medium s o i l test i s obtained, for either P or K, a more co s t l y error i n recommended f e r t i l i z e r application rate occurs when the sampling method underestimates the true mean, than when i t overestimates. An underestimate of the K mean by 60% w i l l cause an extra 134 kg/ha 65 of K^ O to be prescribed. An overestimate of 60% w i l l mean an under-recommendation of 45 kg/ha. The seriousness of the u n d e r f e r t i l i zation in terms of reduced crop production i s not e a s i l y estimated without y i e l d curves being a v a i l a b l e . Errors in estimate of P levels produce a s i m i l a r r e s u l t . 66 Table 2-6 Quantifying error i n f e r t i l i z e r recommendation associated with high and low f e r t i l i t y estimates for P at two s o i l tests levels,(recommendations based on BCMAF guidlines for crop group three). SOIL TEST* % ERROR OF ESTIMATE | EXTREMES OF ESTIMATE RESULTING RECOMMENDATION ERROR (kg P^O^/ha) LOW HIGH ESTIMATE | ESTIMATE | 10 18-22 MED 20 16-24 — — 20 30 14-26 +44 — (ppm) 40 12-28 +44 — 50 10-30 +67 -23 60 8-32 +67 -23 10 45-50 20 40-60 +17 — HIGH 30 35-65 +17 — 50 40 30-70 +39 — (ppm) 50 25-75 +39 — 60 20-80 +62 * s o i l test as ppm P 67 Table 2-7 Quantifying error i n f e r t i l i z e r recommendation associated with high and low f e r t i l i t y estimates for K at two s o i l tests l e v e l s , recommendations based on BCMAF guidlines for crop group three. SOIL TEST* % ERROR OF ESTIMATE EXTREMES OF ESTIMATE RES ULTING RE COMMENDATION ERROR (kg K^O/ha) LOW HIGH ESTIMATE | ESTIMATE | 10 72- 88 __ MED 20 64- 96 + 12 — 80 30 56-104 + 12 -23 (ppm) 40 48-112 + 78 -23 50 40-120 + 96 -23 60 32-128 +134 -45 10 135-165 20 120-180 — — HIGH 30 10 5-19 5 +22 — 150 40 90-210 +22 — (ppm) 50 75-225 +45 — 60 60-240 +67 — * s o i l test as ppm K^0 68 POSSIBLE ALTERNATIVES TO HIGH ANALYSIS COST METHODS The obvious s o l u t i o n to the high cost of having many samples analysed i n the laboratory i s to reduce the number of samples. Unfortunately t h i s i s followed by a reduction i n the amount of information. Another alternative i s to analyse the same number of samples but for fewer parameters. If a strong c o r r e l a t i o n exists between certain parameters i t may then be possible that the value of unmeasured parameters can be predicted on the basis of measured parameters. Table 2-8 i s a c o r r e l a t i o n matrix presenting a l l possible correlations in the data set. Not only are parameters from the plow layer correlated with one another but they are also correlated with 30/60 and 60/90 cm parameters. The same i s true for the 30/60 and 60/90 cm parameter. The parameter's depth i s indicated by a 1 , 2, or 3 following the parameter name, with 1 corresponding to plow layer samples. Though there were many s i g n i f i c a n t correlations (Table 2-8) at the %95 l e v e l , there were few which could be considered moderately to highly correlated. Table 2-9 summarizes the best correlations for plow layer parameters. The "best" parameters were f i r s t chosen on the basis of data from three f i e l d s being analysed simultaneously. Where good correlations were found, they were further explored on an in d i v i d u a l f i e l d basis. It was determined that the r value from the combined data was usually a 69 poor indicator of the c o r r e l a t i o n between the same parameters in i n d i v i d u a l f i e l d s . The correlations from three f i e l d s and for pH and Ca in spring i s the best example of the v a r i a t i o n . Table 2-9 shows a high c o r r e l a t i o n for pH and Ca for RN (r=.96), however, for HRST r=.19. This sort of v a r i a t i o n among f i e l d s would l i m i t the p o s s i b l i t y of developing o v e r a l l predictive equations for other variables. In the 60/90 cm depth a greater number of high correlations exist (Table 2-10). The behavior of individual f i e l d s , though not explored, might be expected to be s i m i l a r to those i n the plow 1aye r . Though the usefulness of c o r r e l a t i o n and regression i s l i m i t e d by variations from f i e l d to f i e l d , i t may s t i l l be useful to pursue relationships on an i n d i v i d u a l f i e l d basis. If detailed sampling for individual f i e l d s was continued over a number of years a r e l i a b l e and useful r e l a t i o n s h i p might be discovered. The high degree of chemical identi t y of individual f i e l d s , which i s noted i n chapter 1 and 3 reinforces the l i k e l i h o o d that t h i s may be the case. 70 Table 2-8 Correlation matrix for variables at three depths, data from 102 pr o f i l e s samples at three depth. 1, 2, or 3 following variable name indicates sampling depths of plow layer , 30-60 cm, 60-90 cm respectively. R@ .05 = .1946, R@ .01 = .2540. VARIABLE 1 . CA 1 1 .0000 2 . MG1 .5450 1 .OOOO 3 . NA1 - .1586 - . 2 4 8 5 1 .0000 4 . KK1 - .0934 - .2918 .0624 1.0000 5 CEC1 .0986 - . 1208 .0161 .3871 1 .0000 6 . PHI .6711 .5198 - . 0 9 9 5 - . 5 1 4 5 - .4375 1 .0000 7 . C O - . 1 6 4 7 - .2798 - . 0 1 8 9 .3839 .5935 - . 6 3 3 5 1.0000 8 .N031 .4659 .4830 - . 1608 - .0111 .0806 .2589 . 1189 1.0000 9 . PP 1 - .1514 - . 3 7 9 0 .2386 .3921 .2121 - .2646 . 1648 - . 3 3 4 0 1 .0000 10 . CA2 - .1332 - .0998 .0156 .3480 .2660 - .3941 . 1976 .0329 . 1648 1 .OOOO 11 MG2 - . 0 8 3 8 - .0286 - .0891 . 1813 . 1714 - .2173 . 1483 .0945 .0547 .6079 1.0000 12 NA2 .0907 .0130 - .0766 - .1348 - .3151 .2063 - . 1 3 1 0 - . 1 0 2 9 - .0687 - . 2 1 1 5 - .2834 1.0000 13 . KK2 - . 0 2 5 6 .0443 .0023 - .1078 - .0311 .0783 - .0777 .0002 .0042 - . 2 5 1 0 - .2196 . 1325 14. CEC2 . 1679 .2333 .0546 - . 0 7 5 9 - .1866 .2617 - . 1 8 8 9 . 1070 - . 2 0 8 0 - . 0 9 5 9 - . 0 7 5 6 .2747 15 . PH2 - . 2 0 3 9 - . 1 9 4 9 - .0218 .3431 .2508 - . 4 1 0 6 .2388 - . 0 2 8 4 .2743 . 7440 .6113 - . 3 1 7 5 16. .CC2 .3184 .2744 .0500 - .2741 - . 3 6 4 9 .5061 - .3098 .2544 - . 2 8 9 5 - . 3 5 6 4 - .3697 .3540 17. .N032 .0974 .0710 .0185 .0112 .0937 - . 0 0 7 9 - .0528 - . 0 2 9 8 - . 0954 . 1843 .2084 - . 1 6 8 8 18 . PP2 .O0O8 .0670 .0615 .0174 - .0754 .0659 - .1374 - . 0 3 0 9 - . 0 5 5 0 - . 2 2 5 0 - . 4 7 3 0 .2957 19 . CA3 - .0927 - . 2 5 2 3 . 1752 .0695 - .1313 - .0502 .0772 - . 1 5 4 0 .2146 .0703 - . 0 8 6 8 . 1862 20. MG3 - . 2 4 5 2 - .2481 . 1394 .3031 .0329 - .2927 .0630 - . 3 3 3 8 . 1462 .2143 .0501 - .0054 21. NA3 .0180 .2340 - .1116 .0357 - .0061 .0324 - . 0 8 1 3 . 1396 - .1082 .0632 - .0871 - . 1 5 4 0 22. KK3 - .0131 .2382 - .1647 - . 1 4 9 5 . 1760 - . 1470 . 1526 .0792 - .2951 .2117 .3130 - .2481 23. CEC3 .0602 .2582 - . 1709 - .1903 .0384 .0596 - .0291 .0993 - .3199 .0321 . 1300 - .1857 24. PH3 - . 0 8 2 3 - .2772 . 1588 .2115 - .1291 - . 0 3 6 8 .0603 - . 1130 .2700 .0856 - .0561 . 1640 25. CC3 - . 0 3 6 7 .1139 - .2116 - .0567 . 1663 - . 0 3 1 2 .0955 . 1376 - . 2 4 9 5 .0672 . 1515 - . 2 6 7 6 26. N033 - . 2 3 3 9 - . 2 9 3 0 .2114 . 1371 .0573 - .1538 .0451 - . 3 1 4 1 .3495 .0364 .0697 - . 0 7 9 1 27. , PP3 .2642 .4010 - .2388 - .3014 .0031 .2471 - .0588 .3291 - . 3 9 2 0 - .1382 .0143 - . 1 4 7 0 1 . CA 1 2. MG1 3. NA 1 4. KK1 5. CEC 1 6. PHI 7. c c i 8 . N031 9. PP 1 10. CA2 11. MG2 12. NA2 Table 2-8t> 13 . KK2 1 .OOOO 14 . CEC2 .3892 1.0000 15 . PH2 - .3562 - .5108 1.0000 16 .CC2 .3926 .6585 - .6704 1.0000 17 .N032 - . 1 5 3 6 .0686 .0108 .0185 1.0000 18 . PP2 .2923 .3134 - .4453 .3098 - .1520 19 . CA3 . 1643 .0035 .0560 .1122 - .2331 20 . MQ3 .0038 .0784 .1114 - .0914 - .1103 21 NA3 - . 0 1 9 9 .0695 .0100 - .0348 - .0746 22 KK3 - .0321 .0196 .0967 - .1752 .2829 23. CEC3 - . 0 4 6 6 . 1026 - . 0 9 5 0 .0182 .2723 24. , PH3 .1139 - .0357 . 1286 .0490 ' - . 3 1 8 3 25 CC3 - . 1 1 2 3 - .0150 .0516 - .0984 . 1804 26. NOSS - . 0 5 7 4 - .0243 . 1041 - .1372 - . 1028 27. . PP3 - . 0 8 2 9 .0892 - .1641 .0980 . 1233 13. 14 . 15. 16. 17. KK2 CEC2 PH2 CC2 N032 25. CC3 1.OO0O 26. N033 .0241 1 .0000 27. PP3 .6678 - .1640 1.0000 25. 26. 27. CCS NOSS PP3 .0851 1.0000 .2019 .441 1 1 .OOOO . 1106 - . 3 9 6 0 .0565 1 .0000 .0814 - . 3 7 1 7 .0336 .2731 1 .0000 .0070 - .4088 .0688 .4636 .6211 1.0000 .0458 .8553 .3434 - . 4 2 3 6 - . 5 2 0 0 - .6614 1.0000 . 1343 - .6181 - . 1 4 8 6 .5151 .5340 .7660 - . 7 2 0 3 .0520 .0327 .0826 .0905 - . 1 4 3 5 - . 1363 .0304 .0828 - . 7 1 7 0 - . 4 2 6 8 .4074 .4319 .6127 - .7912 18. 19. 20. 21 . 22. 23. 24. PP2 CAS MCS3 NA3 KK3 CEC3 PH3 Table 2-9 r values for selected parameters in the plow layer i n spring and for combined spring and f a l l data. Table 2-9 FIELD 3 FIELDS SEASON PARAMETER HRST RN RC COMBINED pH:Ca .19 .96 .55 .66 SPRING pH:C -.41 -.66 -.13 -.63 C:CEC .65 .58 .76 .73 SPRING pH:Ca .20 .81 .42 .67 + pH:C -.38 -.44 -.10 -.63 FALL CtCEC .57 .25 .39 .39 Table 2-10 r values for selected parameters at 60-9 0 cm, data from a l l f i e l d s combined. PARMAMETERS pH:Ca C.CEC pH:CEC pH:C pH:P Ca:P r VALUE .86 .77 -.66 -.72 -.79 -.72 73 SUMMARY The combined estimates of s t r a t i f i e d random composite and conventional sampling methods for N, P, and K were within 10% of the s t r a t i f i e d random mean about 75% of the time. The remaining 25% of the time the estimates are generally within 20-50% of the mean but di f f e r e d by as much as 125%. Estimates of %C by the two methods generally d i f f e r e d by less than 15%. A l l pH estimates d i f f e r e d from the s t r a t i f i e d random mean by less than 5%. B a l l and Williams (1971) c a r r i e d out a comparison of various composite sampling techniques, however, differences in methods do not allow for a v a l i d comparison. The results from conventional, s t r a t i f i e d random and s t r a t i f i e d random composite suggest that the best estimates of the mean values for NO3-N are obtained i n the f a l l . For P there i s a very strong tendency for spring samples to underestimate the s t r a t i f i e d random mean value but the magnitude of the difference i s smaller in f a l l so that spring i s the preferred time for sampling. Spring sampling for K provides estimates closer to those provided by s t r a t i f i e d random sampling. If sampling i s carried out on a season and f i e l d s p e c i f i c basis then the number of samples required to estimate the mean value with + 10% p r e c i s i o n and 90% confidence i s highly variable. For K only two of a possible s i x f i e l d x season combinations have required sample numbers that coincide with the number suggested by the BCMAF (anomymous d, 1984) . The number of samples required to obtain the same confidence l e v e l s for P varies between 20 and 95. NO3-N sampling requirements are generally very high, ranging 74 from 43 to 167. Calculations of the sample numbers required to provide a s t a t i s t i c a l estimate of the mean show that spring i s c l e a r l y the prefered time to sample for NO3. The study of Cameron et al (1979) also found that spring would be a preferred time for sampling since the v a r i a b i l i t y of NO3-N was lower i n the spring than i n the f a l l , for two of the three study years. For two of the sampled f i e l d s , spring sampling for P also provides a better estimate. Potassium, however, i s best estimated i n the f a l l for a l l three f i e l d s . C l e a r l y the 15 samples suggested by BCMAF guidlines (anonymous, 1984) w i l l provide poor estimates of mean values for P and K. Often two to four times the recommended sample number would be required to provide the above l e v e l s of confidence. In order to accommodate r e a l i s t i c sample numbers and acceptable characterization of f e r t i l i t y variables Cameron et a l (1979) suggest 80% accuracy with 20% p r e c i s i o n . These are apparently r e a l i s t i c guidelines since greater errors could be very c o s t l y . A 50% underestimate of the mean P value, in a medium f e r t i l i t y f i e l d , would amount to an extra $30.50 worth of f e r t i l i z e r being applied per ha. For a 3.6 ha. f i e l d t h i s amounts to an extra $109 worth of f e r t i l i z e r . This i s just s l i g h t l y less than the cost of doing analysis for one parameter, using the s t r a t i f i e d random sampling method. 75 CHAPTER I I I AN INNOVATIVE APPROACH TO SOIL FERTILITY MANAGEMENT: A SINGLE PARAMETER APPROACH INTRODUCTION In the previous chapter i t was determined that i n comparison to s t r a t i f i e d random sampling r e s u l t s , conventional and composite sampling methods obtained variable, though generally r e l i a b l e , estimates of the mean value for f i e l d f e r t i l i t y parameters. In this section a comparison w i l l be made between the f e r t i l i z e r applications resulting from the use of a conventionally obtained mean value and that resulting from detailed sampling. The aim of the detailed approach i s to apply optimum le v e l s of f e r t i l i z e r , as recommended by the BCMAF guidelines, to a l l portions of the f i e l d . This assumes that the BCMAF f e r t i l i z e r recommendations lead to optimum f e r t i l i z a t i o n . Optimum f e r t i -l i z a t i o n to a l l parts of the f i e l d would be accomplished by recommending and subsequently applying variable amounts of f e r t i l i z e r to d i f f e r e n t portions of the f i e l d . This i s in contrast to the blanket application of f e r t i l i z e r based on a mean s o i l test value. In order to make a comparison of the information provided by deta i l e d sampling and that provided by a mean value, the P values 76 of individual plots within the f i e l d s are compared to a range around the f i e l d mean value. This results in a c a l c u l a t i o n of the percentage of the f i e l d which i s over-, under-, and c o r r e c t l y f e r t i l i z e d . In addition, BCMAF guidlines for f e r t i l i z a t i o n of brussels sprouts were followed on the HRST f i e l d i n order to compare the cost of s o i l analyis, plus P or K f e r t i l i z a t i o n , r e s u l t i n g from detailed sampling, to the costs associated with using the conventional approach. 77 COMPARING FIELD FERTILIZER REQUIREMENTS USING FIELD MEAN VALUE AND PLOT VALUES The f i r s t method chosen to i l l u s t r a t e the effectiveness of the information provided by detailed sampling i s to determine the area of a f i e l d which f a l l s within various ranges of the mean value. Though th i s method does not compare sampling methods d i r e c t l y i t does demonstrate how a knowledge of spacial v a r i a b i l i t y could play an important role i n f e r t i l i z e r rec-ommendation and ap p l i c a t i o n . Table 3-1 summarizes results of this comparison by showing the percentage of the f i e l d ' s t o t a l area which i s over-, under-, or c o r r e c t l y f e r t i l i z e d . These calculations are based on the assumption that a + 15 ppm increment of P around the mean value i s the correct estimate of f i e l d f e r t i l i t y . This comparison demonstrates what proportions of a f i e l d w i l l be over- or u n d e r - f e r t i l i z e d , when compared to the f i e l d mean. The assumption that the plot value represents the plot's chemistry i s made i n order to make these c a l c u l a t i o n s . The second method of comparison goes one step further and calculates the amount of f e r t i l i z e r that would be applied per plot , using r e s u l t s of detailed sampling, and compares that value to one obtained from the use of a convention mean. In addition, the t o t a l cost of f e r t i l i z e r and sample anlysis for the two methods i s compared. By considering the f i e l d information summarizerd i n Table 3-1 i t can be seen that the usefulness of a f i e l d mean value for 78 predicting f e r t i l i z e r requirements w i l l be variable. I f one accepts the assumptions that the ± 15 ppm increment around the mean s o i l test provides the correct recommendation, i t can be seen that between 20% to 60% of the area within the four studied f i e l d s have s o i l P levels f a l l i n g within the defined range of the mean and are ,therefore, c o r r e c t l y f e r t i l i z e d . Eighteen to 44% of the f i e l d w i l l be under f e r t i l i z e d and 12 to 35% over f e r t i l i z e d . F i g . 3-1 shows the spac i a l d i s t r i b u t i o n of P values for the HRST f i e l d . Though one might expect a random occurrence of P values they often are, in f a c t , clustered i n groups with s i m i l a r P l e v e l s . Table 3-2 shows that,when conventional and detailed methods are compared, there i s an i n s i g n i f i c a n t difference i n the cost of s o i l analysis and subsquent P f e r t i l i z a t i o n for HRST. The cost of analyis and f e r t i l i z a t i o n i s $231 for the conventional and $251 for the detailed method. The benefit of the detailed sampling i n comparison to the conventional method i s i n preventing the under f e r t i l i z a t i o n of 5 of the 17 plots or about 30% of the f i e l d . Four of these plots would be under f e r t i l i z e d by more than about 30 kg/ha. According to the f e r t i l i z a t i o n guidelines the other 12 plots would be over f e r t i l i z e d , f i v e of these by 35 kg/ha. Using HRST as an example again, i t can be seen that the cost versus information evaluation for K i s quite d i f f e r e n t from that for P. Using the information provided by det a i l e d sampling provides f e r t i l i z e r recommendation which r e s u l t i n an addition 79 Table 3 - 1 Summary of percentage of the area of four f i e l d s which would have phosphorus plots values above and below a ± 15 ppm increment around the f i e l d arithmetic mean. PERCENT AREA IN CATEGORY FIELD CORRECTLY FERTILIZED UNDER FERTILIZED OVER FERTILIZED RN 59 24 18 HRST 60 18 12 RC 29 35 35 CLPRO 22 44 34 * spring data 80 Figure 3 - 1 Spacial d i s t r i b u t i o n of P levels for HRST in spring. Calculation of % of f i e l d area co r r -e c t l y f e r t i l i z e d based on the assumption that a ± 15 ppm increment around the arithmetic mean is c o r r e c t l y f e r t i l i z e d . 58 18 42 17 38 16 58 15 18 14 24 13 38 12 50 11 36 10 44 9 8 20 7 64 6 76 5 56 4 48 3 36 2 48 1 P l e v e l (ppm P) plo t # % AREA CORRECTLY FERTILIZED =60% % AREA OVER FERTILIZED = 12% % AREA UNDER FERTILIZED = 18% under f e r t i l i z e d i f P le v e l < 30 ppm co r r e c t l y f e r t i l i z e d i f P l e v e l between 30 & 60 ppm over f e r t i l i z e d i f P le v e l > 60 ppm 81 Table 3-2 Comparing phosphorus f e r t i l i z e r recommendations using values obtained by detailed sampling to that r e s u l t i n g from the use of the conventional mean value, for HRST in spring. CONVENTIONAL RESULTS Conventional mean value for s o i l P = 36 ppm P Recommended f e r t i l i z e r rate = 80 kg/ha P2.O5 or 16.2 kg/plot Have to add 145 kg/ha of 11-55-0 Cost to f e r t i l i z e 17 plots = $225 Analysis cost = $ 6 Total cost $231 RANDOM STRATIFIED (DETAILED) SAMPLING RESULTS PLOT # P (ppm) RECOMMENDATION kg P, 0^  / p lot (kg/ha) / P 20 5 1 48 67 13.4 2 36 78 15.6 3 48 67 13.4 4 56 45 9.0 5 76 45 9.0 6 64 45 9.0 7 20 134 26.8 * 9 44 78 15.6 10 36 112 22.4 * 11 50 45 9.0 12 38 90 18.0 * 13 24 112 22.4 * 14 18 145 29.0 * 15 58 45 9.0 16 38 67 13.4 17 42 67 13.4 18 58 45 9.0 * Plots which are u n d e r f e r t i l i zed Total f e r t i l i z e r applied to 17 plots = 257 kg F e r t i l i z e r cost = $115 Analysis cost = $136 Total cost $251 82 Table 3-3 Comparing potassium f e r t i l i z e r recommendations using values obtained by detailed sampling to that r e s u l t i n g from the use i n spring. of the conventional mean value, for HRST CONVENTIONAL RESULTS Conventional mean f i e l d value = .488 me/100 g Recamended f e r t i l i z e r rate = 45 kg/ha of K^O or 9 kg/plot Cost to f e r t i l i z e 17 plots = $54 Analysis cost = $ 8 Total cost $62 RANDOM STRATIFIED (DETAILED) SAMPLING RESULTS PLOT # K (ppm) RECOMMENDATION Ka.0 K^ O (kg/ha) kg/plot 1 262 45 9 2 197 45 9 3 225 45 9 4 248 45 9 5 318 45 9 6 379 45 9 7 98 112 22.4 * 9 14 280 56.0 * 10 20 5 45 9 11 234 45 9 12 150 67 13.4 * 13 234 45 9 14 168 45 9 15 243 45 9 16 271 45 9 17 243 45 9 18 412 45 9 * Plots which are un d e r f e r t i l i z e d Total f e r t i l i zer applied to f i e l d = 348 kg/ F e r t i l i z e r cost = $115.00 Analysis cost = $ 76.70 Total cost $212.70 83 $68 worth of K f e r t i l i z e r being applied (Table 3-3). When the f e r t i l i z e r cost i s added to the cost of analysis the t o t a l cost difference i s about $150 extra for the detailed method. If the recommendation, for K, based on the conventional sample, is followed, about 80% of the f i e l d w i l l be f e r t i l i z e d c o r r e c t l y . The remaining 20% of the f i e l d w i l l be under f e r t i l i z e d . The magnitude of the under f e r t i l i z a t i o n ranges from about 10 kg/ha to 65 kg/ha. About 10% of the f i e l d w i l l be under f e r t i l i z e d by greater than 50 kg/ha. A similar case i s stated i n Crops and S o i l s Magazine (Anonymous, 1984). The magazine provides an example showing how a single composite sample used to represent a large area l e d to u n d e r f e r t i l i z a t i o n of 75% of the f i e l d . The expected loss of y i e l d , on the u n d e r f e r t i l i z e d areas, was 15 bushels of corn per acre and half a ton of a l f a l f a per acre. The a r t i c l e quotes a s o i l s c i e n t i s t as saying "saving on sampling i s penny-wise and pound-foolish". A f i n a l comparison of sampling cost, which could also be ca l l e d information cost, and f e r t i l i z e r cost w i l l help to put the question of detailed sampling into perspective. In table 3-4 the cost of analysis for samples in groups of 17 i s given. Costs are quoted by P a c i f i c S o i l s Analysis Inc. in Oct. 1983 (Herman, personal communication) The analysis cost for up to 3 sampling units at discounted prices of 10%, 15%, and 20% for 17, 34, and 51 samples respectively are given . Analysis cost i s for P and K only, at the base price of $8 per sample. The term sampling unit rather than f i e l d is used since some f i e l d s may contain two or 84 more v i s u a l l y separable units, which would i d e a l l y be sampled i n d i v i d u a l l y . The assumptions made in calculating the f e r t i l i z e r costs are that each sampling unit i s 3.6 ha in size and the rate of f e r t i l i z e r application i s 220 kg/ha for both P^CV and K-^ O. Though th i s l e v e l is high by the BCMA guidelines i t does seem to be a "normal" , i f perhaps low, a p p l i c a t i o n rate for farmers producing cash crops in the Lower Fraser Valley (Reynolds, Wood, personal communications). F e r t i l i z e r costs are calculated using 11-55-0 at $450/tonne and 0-0-60 at $220 a tonne. Though the r a t i o of sampling to f e r t i l i z e r cost i s high on a s i n g l e year basis the r a t i o of analyis cost to f e r t i l i z e r cost drops to about 8% i f intensive sampling i s done once i n three years and i f three units are sampled. Table 3-4 F e r t i l i z e r cost vs s o i l analysis cost for one, two and three sampling units given that the f i e l d is 3.6 ha. and f e r t i l i z e r i s added at 220 kg/ha. for both P and K. ANALYSIS COST # SAMPLING FERTILIZER COST ANALYSIS X 100% UNITS (P + K) COST FERTILIZER COST 1 925 245 26 2 1850 462 25 3 2775 653 23 85 I t might be suggested that detailed sampling be carried out once every three to four years so that areas of the f i e l d which are poorly represented by the mean value could be recognized. Once recognized these areas of the f i e l d could be f e r t i l i z e d accordingly, so that v a r i a b i l i t y i s reduced. In this way the use of conventional sampling would become more appropriate. 86 SUMMARY T h e v a l u e o f d e t a i l e d s a m p l i n g becomes a p p a r e n t when P f e r t i l i t y l e v e l s o f t h e v a r i o u s p l o t s a r e c o m p a r e d t o a 30 ppm i n c r e m e n t a r o u n d t h e f i e l d m e a n . T h e r e s u l t s o f t h e c o m p a r i s o n show t h a t i n t h e b e s t s i t u a t i o n , f o r t h e f i e l d s s t u d i e d , t h a t 60% o f t h e f i e l d f a l l s w i t h i n t h e + 15 ppm i n c r e m e n t a r o u n d t h e m e a n . A t w o r s t o n l y 22% o f t h e f i e l d f e l l w i t h i n t h e mean r a n g e . P e r c e n t a g e o f t h e f i e l d s ' a r e a b e i n g u n d e r e s t i m a t e d r a n g e d f r o m 18 t o 44%, w h i l e 12 t o 35% was o v e r e s t i m a t e d . P e c k a n d D i b b (1979) a l s o s h o w e d t h e a r i t h m e t i c m e a n , o r c o m p o s i t e d s a m p l e , p o o r l y r e p r e s e n t i n g f i e l d f e r t i l i t y . I n t h e t h e i r e x a m p l e , 75% o f t h e f i e l d w o u l d be u n d e r f e r t i i i z e r f o r b o t h P a n d K , i f t h e mean v a l u e was u s e d f o r a f e r t i l i z e r r e c o m m e n d a t i o n . P u t t i n g a d o l l a r v a l u e o n i n t e n s i v e s a m p l i n g a n d r e s u l t i n g o p t i m u m f e r t i l i z a t i o n v e r s u s t h e c o s t o f c o n v e n t i o n a l s a m p l i n g a n a l y s i s and s u b s e q u e n t f e r t i l i z e r a p p l i c a t i o n s h o w s , f o r t h e o n e K e x a m p l e s t u d i e d , t h a t t h e c o s t o f a n a l y s i s f o r d e t a i l e d s a m p l i n g and s u b s e q u e n t f e r t i l i z a t i o n c o u l d c o v e r t h e c o s t o f more t h a n t r i p l i n g t h e f e r t i l i z e r r a t e r e c o m m e n d e d by t h e c o n v e n t i o n a l m e t h o d . H o w e v e r , e v e n t h i s t r i p l e d r a t e w o u l d s t i l l l e a v e o n e p l o t u n d e r f e r t i i i z e d . F o r t h e P e x a m p l e t h e d e t a i l e d s a m p l i n g a n d s u b s e q u e n t f e r t i l i z a t i o n c o s t o n l y $20 more t h a n use o f t h e c o n v e n t i o n a l m e t h o d . M o r e i m p o r t a n t l y t h e use o f d e t a i l e d s a m p l i n g p r e v e n t e d t h e u n d e r f e r t i i i z a t i o n o f a l m o s t 30% o f t h e 87 f i e l d , 12% of the t o t a l area would have been u n d e r f e r t i l i z e d by-greater than 65 kg/ha of P-2O5. Cameron et al (1971) noted that , due to large variations in s o i l chemistry throughout studied f i e l d s , a sampling scheme that would depict nutrient l e v e l trends within a f i e l d with only a few composite samples would be desirable. The periodic use of detailed sampling would appear to be that scheme. The periodic use of intensive sampling could be used to i d e n t i f y f e r t i l i t y patterns, which would serve as the basis for the composite sampling units within the f i e l d . I t would also serve as a guide for reducing v a r i a b i l i t y of f e r t i l i t y l e v e l s . In this way the use of the less expensive composite sampling techniques could be made more appropriate. Crops and S o i l s Magazine (Anonymous e, 1984) also povides an example j u s t i f y i n g t h i s sort of approach. In this example the s t r a t i f i c a t i o n of a f i e l d into four smaller parcels, prior to s o i l sampling, would have provided appropriate f e r t i l i z e r recommendation to a much greater portion of the f i e l d and provide an o v e r a l l increased y i e l d . If intensive sampling and analysis i s used on a yearly basis i t s cost i s about 23-26% of a " t y p i c a l " f e r t i l i z e r b i l l but i f i t i s done once i n 3 years i t becomes about 8% of the f e r t i l i z e r cost. 88 CHAPTER, IV AN INNOVATIVE APPROACH TO SOIL FERTILITY MANAGEMENT: A MULTIPARAMETER APPROACH INTRODUCTION S o i l s c i e n t i s t s have r e c o g n i z e d the dependence of crop y i e l d on more than one s o i l v a r i a b l e and have o f t e n attempted to e x p l a i n y i e l d i n terms of two or more s i t e f a c t o r s by d e v e l o p i n g m u l t i p l e r e g r e s s i o n equations (Hoffman, 1971; Nelson and Anderson, 1980; Nelson and McCraken, 1962). Since the g r e a t e r p r e d i c t i v e value of equations i n v o l v i n g two or more n u t r i e n t s or s i t e f a c t o r s has been r e c o g n i z e d , i t then seems reasonable t h a t c l u s t e r a n a l y s i s , which i s an n-dimensional approach, c o u l d be used to determine management u n i t s w i t h i n f i e l d s with p o t e n t i a l l y d i f f e r e n t f e r t i l i z e r responses. M u l t i v a r i a t e a n a l y s i s has t r a d i t i o n a l l y been used as a c l a s s i f i c a t i o n method i n the b i o l o g i c a l s c i e n c e s (Sokal and Sneath, 1963) but more r e c e n t l y has been u t i l i z e d i n l a n d c l a s s i f i c a t i o n Otfebster and Borrough, 1972) . S c h r e i e r and Z u l k i f l i (1983) were able to use f a c t o r and c l u s t e r a n a l y s i s to c l a s s i f y s o i l s with d i f f e r e n t y i e l d p o t e n t i a l s . I t s use as an a i d to improving f e r t i l i z e r recommendations of a g r i c u l t u r a l f i e l d s i s , however, an a p p a r e n t l y new a p p l i c a t i o n . The p r i n c i p a l aim of using m u l t i v a r i a t e t echniques, s p e c i f i c a l l y p r i n c i p a l component a n a l y s i s (PCA) and c l u s t e r 89 analysis, was to determine i f plow layer samples from individual f i e l d s would be separated into s i g n i f i c a n t l y d i f f e r e n t and contiguous units on the basis of several parameters. The multivariate approach was also used to determine i f plow layer data could be used to successfully discriminate among several f i e l d s . As with any c l a s s i f i c a t i o n system, the aim of cluster analysis i s to assign similar individuals to the same group. In the case of this study, i t i s hoped that these groupings, aside from being mathematically similar e n t i t i e s w i l l also be amenable to s i m i l a r management practices. The parameters used in cluster analysis were chosen on the basis of PCA r e s u l t s . Ca, CEC, %C, P, and pH were the parameters used most frequently in cluster analysis . 90 RESULTS OF CLUSTER ANALYSIS OF PLOW LAYER SAMPLES: INDIVIDUAL FIELDS The custer analysis progam was run on plow layer data, c o l l e c t e d i n the spring, for each of four f i e l d s . Results are given for the CLPRO f i e l d only. The success of c l a s s i f i c a t i o n was judged i n two ways. One was by testing supposed taxonomic units for s t a t i s t i c a l y s i g n i f i c a n t differences of the parameters between units. The Mann-Whitney U-test was used to determine which parameters had s i g n i f i c a n t l y d i f f e r e n t l e v e l s beween plots at the 95% l e v e l . Another was to consider the contiguity of the unit members. If unit members tended to be contiguous then they could be managed as a group rather than i n d i v i d u a l p l o t s . This could then be considered more successful than having unit members which were not contiguous. The success of c l u s t e r i n g i n d i v i d u a l f i e l d s was variable. Clustering was most successful using data from f i e l d s located on a s o i l complex or association. These were CLPRO and HRST. For RN and RC which were located on pure mapping units, the success of h i e r a r c h i c a l c l a s s i f i c a t i o n was marginal in that unit members tended not to be contiguous and there were fewer variables with s i g n i f i c a n t d i f f e r c e s between cluster units. A figure showing the s p a c i a l d i s t r i b u t i o n of c l a s s i f i c a t i o n units and a s i g n i f i c a n c e difference table i s found following the dendrogram for the various f i e l d s . There was often no s t a t i s t i c a l l y s i g n i f i c a n t differences among parameters within the various units, within the f i e l d , due to the small number of plots 91 within each unit. When the number of samples being compared was three versus four or fewer, the difference was i n s i g n i f i c a n t at the 95% l e v e l , even though there was no overlap of data values. Though there i s often no s t a t i s t i c a l l y s i g n i f i c a n t difference among uni t s , the difference could often be of consequence to management. Differences of this nature are noted i n the s i g n i f i c a n t difference tables. The lack of even one parameter which consistently has a s i g n i f i c a n t difference among a l l units for RN and RC (Table 4-2 and 4-4) shows the need for a multivariate approach i n order to distinguish among chemically d i f f e r e n t areas within the f i e l d . I t can be seen from the s i g n i f i c a n t difference tables for RN and RC that pH, C, or CEC account for most of the s i g n i f i c a n t differences among units, though Ca does play a s i g n i f i c a n t role in RC. Recalling how these parameters interact and using input from the c o r r e l a t i o n matrix one can draw a diagram to represent what might be c a l l e d the carbon-CEC factor. Nelson and Anderson (1980) state that CEC and %C are among the variables contributing to poor c o r r e l a t i o n of s o i l test with crop y i e l d . I n d i r e c t l y Ca and organic matter content w i l l influence s o i l structure; these were also stated to be among the C a CEC C 92 factors a f f e c t i n g c o r r e l a t i o n . Delineating areas of the f i e l d which have similar l e v e l s of these variables may be a step to improving the c o r r e l a t i o n of s o i l test and crop y i e l d and f e r t i l i z e r e f f i c i e n c y . In general terms some statement can be made about the units produced by cluster analysis of RN data (Table 4-1) . Units A and C are similar i n a l l the d i f f e r e n t i a t i n g parameters except P, where A has higher l e v e l s . Unit B has higher %C and CEC than either A or C so i t may have a higher f e r t i l i t y potential when the pH is managed favorably. Unit D which occupies a r e l a t i v e l y small area i s characterized by the lowest %C and CEC. RC was also separated into 3 d i f f e r e n t units (Fig. 4-3 & Table 4-3). In terms of Ca, P and %C units A and C are si m i l a r (Table 4-4). However, B has a lower pH than the other two units and usually has lower Ca values as well. Unit C is characterized by the highest %C and CEC values. Generally unit C also has higher P values than either A or B. If a farmer i s interested i n reducing f i e l d v a r i a b i l i t y as a means of optimizing or increasing o v e r a l l productivity the less than desirable spacial contiguousness of these unit members (Fig. 4-4) may present some mechanical problems. However, deter-mining management units within the f i e l d i s the f i r s t step to working with th e i r i n d i v i d u a l requirements. 93 ITEMS STEP I 1 3 2 7 3 9 4 7 5 12 6 2 7 4 8 6 9 2 10 6 11 4 12 2 13 6 14 2 15 2 16 1 F i g u r e 4-1 Dendrogram from c l u s t e r i n g of RN plow l a y e r data i n s p r i n g , C l u s t e r e d on Ca, CEC, pH, %C, and P. Table 4-1 P l o t # and parameter val u e s a s s o c i a t e d with c l u s t e r u n i t members of RN plow l a y e r data i n s p r i n g . P l o t s are arranged i n dendrogram o r d e r , column one of p l o t ID i d e n t i f i e s the f i e l d , c o l u m n s three and f o u r i d e n t i f y p l o t number. PARAMETER AND VALUE PLOT JNIT ID Ca Mg Na K CEC pH %C NO3-N P 31 111 13 9 3 o 8 7 0 . 11 0 6 0 2 0 0 6 3 2 5 13 1 4 8 O 31 2 1 1 ~f 3 8 0 8 9 0 . 12 0 6 7 2 2 4 5 3 3 7 13 0 84.0 31 9 1 1 S 2 5 0 8 3 O . 10 O 91 2 2 0 5 1 3 7 9 2 8 0 . 0 A 3 1 1 7 1 1 5 6 3 0 9 6 0 . 0 9 0 7 2 2 3 4 5 0 3 9 10 3 8 8 . O 3 1 1 1 1 1 5 15 1 17 0 . 15 O 9 5 2 5 0 4 7 4 C 11 5 8 0 . 0 31 3 1 1 5 15 0 8 9 0 . 13 0 7 7 21 9 4 9 3 4 9 8 9 5 . 0 3 1 1 6 1 1 5 5 9 1 14 0 . 13 0 0 3 21 8 5 0 3 4 7 3 100.0 31 4 1 1 8 4 3 1 0 6 0 . 12 0 6 7 2 8 1 5 3 3 9 16 3 7 2 . 0 ' B 31 5 1 1 7 4 1 1 2 7 0 . 13 o 6 6 2 6 0 5 0 3 9 16 5 84 . 0 3 1 1 8 1 1 5 p 8 2 0 . 0 6 9 7 8 2 9 2 4 9 3 9 11 2 5 2 . O 31 6 1 1 8 3 7 1 3 6 0 . 0 8 1 0 8 24 3 5 4 3 2 16 6 56.6 3 1 1 2 1 1 7 13 1 14 0 . 0 9 0 6 0 2 2 1 5 1 3 6 13 8 4 8 0 c 31 7 1 1 6 13 0 9 7 0 . 0 7 o 9 2 21 4 5 0 3 6 2 4 0 6 4 . 0 3 1 1 5 1 1 5 2 7 1 0 8 0 . 10 0 5 9 2 2 5 4 9 3 5 14 3 6 4 . 0 31 8 1 1 5 5 7 0 9 8 0 . 10 0 71 21 o 5 0 3 2 12 2 6 4 . O 3 1 1 3 1 1 3 9 4 1 0 7 0 . 0 5 0 7 8 16 3 4 8 3 3 8 0 5 2 . 0 0 3 1 1 4 1 1 5 6 4 1 14 0 0 6 0 9 8 16 O 5 O 3 4' 1 0 3 6 4 . 0 GROUPED 0 ERROR 15 14 16 8 13 9 5 1 1 IO 7 17 3 12 4 6 2 0 . 0 7 S 2 9 2 4 O . 1 8 8 8 2 8 2 0 . 4 1 2 0 2 7 4 0 . 6 8 2 8 6 5 0 0 . 7 3 7 0 3 4 8 0 . 9 0 0 1 4 0 8 0 . 9 7 0 9 5 5 8 1 . 3 6 4 3 6 2 7 1 . 7 8 6 7 2 7 0 2 . 7 1 2 3 7 1 8 3 . 1 6 3 3 2 0 5 3 . 2 6 2 2 1 0 8 6 . 0 6 1 2 7 1 7 8 . 8 4 5 0 0 5 0 1 7 . 1 2 7 0 7 5 3 6 . 7 0 4 2 3 9 1 10 5 7 13 2 3 17 14 9 15 6 8 16 4 11 12 1. I LI A T B T _ £ « * » * * » * • * * » * * * * • * 94 UNIT PLOT# Fiqure 4-2 SPACIAL DISTRIBUTION OF RN CLUSTER UNIT MEMBERS Table 4-2 TABLE SHOWING SIGNIFICANT DIFFERENCES BETWEEN CLUSTER UNITS * (only clustering parameters checked for s i g . difference) B C D A \ CEC P P * B \ CEC \ C \ p CEC* %C* P* C \CEC* that values are not s i g n i f i c a n t l y d i f f e r e n t at the (by Mann-Whitney U-test) but there is not overlap of B A A C D D 18 17 16 15 14 13 C A A C C 12 11 10 9 8 7 C B B A A 6 5 4 3 2 1 * indicates 95% l e v e l , data 'values. 95 ITEMS GROUPED STEP I J ERROR 1 1 2 1 5 0 8 5 5 9 8 1 3 2 9 1 1 0 . 9 1 7 6 0 7 3 3 1 3 1 7 1 1 6 8 0 1 6 4 4 8 1 4 1 3 3 6 2 7 3 2 5 4 1 0 1 . 3 4 3 3 5 9 0 6 7 1 2 1 7 8 1 4 3 7 9 7 2 6 2 . 0 6 5 3 6 7 7 8 5 9 2 6 9 8 1 5 3 5 9 7 8 2 9 7 6 3 7 0 8 1 0 2 3 3 1 2 3 0 2 6 8 1 1 4 7 3 6 6 9 7 2 7 3 1 2 5 1 6 4 5 6 1 3 0 6 0 1 3 1 1 3 6 7 6 2 4 3 8 8 1 4 1 4 9 2 4 7 3 0 9 7 1 5 1 2 1 7 . 0 3 4 9 4 3 1 6 1 5 2 5 . 4 5 1 1 7 2 1 1 0 8 3 1 6 1 3 7 1 4 5 1 7 1 2 4 1 5 "T. 2 9 6 1 1 1. T B * * * * * * * * Figure 4-3 Dendrogram from cl u s t e r i n g of RC plow layer data in spring, Clustered on Ca, CEC, pH, %C, and P. Table 4 -3 Plot # and parameter values associated with cluster unit members of RC plow layer data i n spring. Plots are arranged in dendrogram order, column one of plot ID i d e n t i f i e s the field,columns three and four i d e n t i f y plot number. PARAMETER AND VALUE PLOT UNIT ID Ca Mg Na K CEC pH %c NO 3-N P 4 1 1 1 1 7 . 0 8 0 . 9 0 0 5 2 0 5 C 1 3 5 6 1 1 . 4 3 3 1 1 2 C 4 1 1 4 1 1 6 . 0 3 1 . 0 6 0 4 0 0 4 6 1 7 7 5 9 1 . 8 5 . 2 1 5 6 0 4 1 1 8 1 1 6 . 4 3 O . 9 4 O 2 2 0 5 1 1 8 1 6 3 1 . 9 4 4 1 6 8 0 4 1 4 1 1 8 . 2 3 1 . 5 5 0 2 7 0 8 4 1 7 7 5 8 2 . 4 5 3 9 6 o A 4 1 1 0 1 1 9 . 2 1 0 . 9 8 0 4 2 0 6 2 2 0 2 5 S 2 . 3 3 3 1 2 0 0 4 1 7 1 1 8 . 4 2 1 . 1 2 0 3 0 0 7 8 1 7 2 5 9 2 . 2 5 .C 1 4 8 0 4 1 1 3 1 1 9 . 6 3 1 . 2 3 0 2 9 0 9 7 1 7 7 6 0 2 . 0 7 . 2 1 2 0 0 4 1 1 6 1 1 1 0 . 2 7 1 . 6 1 0 4 2 0 8 9 1 6 0 6 2 2 . 0 7 . 2 1 2 0 0 4 1 8 1 1 8 . 1 1 O . 8 2 O 3 6 0 6 0 1 8 6 6 2 1 . 9 3 . 8 1 0 8 .o 4 1 1 5 1 1 7 . 1 7 1 . 2 3 0 3 0 0 4 6 1 6 8 5 8 1 . 7 4 . 7 1 1 2 . 0 4 1 2 1 1 5 . 2 3 0 . 8 6 0 4 3 0 2 1 1 7 3 5 5 2 . 0 4 ,C 1 2 8 .o B 4 1 6 1 1 6 . 1 3 0 . 7 0 4 5 5 0 6 9 2 1 2 5 5 2 . 1 8 . 3 1 0 8 0 4 1 3 1 1 6 . 1 9 1 . 2 7 0 2 3 0 8 5 1 6 8 5 2 2 . 1 2 . 3 7 6 . 0 4 1 5 1 1 1 1 . 7 8 0 . 9 4 0 2 5 1 1 0 2 1 9 6 1 2 . 6 5 . 0 1 1 6 . 0 r 4 1 9 1 1 9 . 8 7 0 . 8 2 0 4 4 1 0 7 2 0 6 6 0 2 . 8 3 . 8 1 6 8 . 0 4 1 1 1 1 1 1 0 . 4 4 1 . 9 5 0 2 9 1 1 2 2 1 2 6 O 2 . 4 1 2 . 3 1 5 2 o 4 1 1 7 1 1 7 . 6 0 0 . 8 2 0 . 3 5 0 5 4 2 0 6 5 6 2 . 5 5 . 3 1 6 0 . 0 96 A 1 B 2 B 3 A 4 C 5 B 6 A 7 A 8 C 9 A 10 C 11 12 A 13 A 14 A 15 A 16 C 17 A 18 UNIT •PLOTt Fig u r e 4-4 SPACIAL DISTRIBUTION OF RC CLUSTER UNIT MEMBERS Table 4-4 TABLE SHOWING SIGNIFICANT DIFFERENCES BETWEEN CLUSTER UNITS * (only c l u s t e r i n g parameters checked f o r s i g . d i f f e r e n c e ) A B C A pH CEC B CEC* pH* C* Ca* C * i n d i c a t e s t h a t v a l u e s are not s i g n i f i c a n t l y d i f f e r e n t at th 95% l e v e l , (by Mann-Whitney U-test) but there i s not o v e r l a p o data v a l u e s . 97 For HRST and CLPRO, located on a mapping association and complex respectively, c l u s t e r i n g units had more contiguous members (Fig. 4-6 & 4-8) . Management, from the mechanical point of view would be less of a problem for these f i e l d s . For HRST (Table 4-5) unit C is a unit with high C and high CEC as well as high P and K l e v e l s . B is a unit with low organic C, f a i r l y low CEC and P. While unit A occupies the largest portion of the f i e l d i t has moderate C, CEC, P, and K l e v e l s . By considering values for %C, CEC, and p for plot number 13 one can see that i t i s l i k e l y m i s s c l a s s i f ied and should belong to B rather than A. This corrected c l a s s i f i c a t i o n i s also i n agreement with f i e l d observations. The clustering of CLPRO data (Fig. 4-1) also produces contiguous units but here the c l a s s i f i c a t i o n i s not dominated by differences in C or CEC (Table 4-8). There are no s i g n i f i c a n t differences among units for K but NO3-N in B is s i g n i f i c a n t l y higher than i n A or C. Also the P levels in B are s i g n i f i c a n t l y lower while the plot has very high %C and CEC. The unit d i f f e r e n t i a t i o n would serve as an aid to determining requirements for P and lime applications rates for d i f f e r e n t parts of the f i e l d . The data suggest that unit B has a greater a b i l i t y than at l e a s t unit C to produce NO3-N, so add-it i o n s of NO to this unit i n par t i c u l a r might be reduced without affe c t i n g crop production l e v e l s . Though P levels for B are s i g n i f i c a n t l y lower than either of the other two units, the unit s t i l l does not require any addition of P, according to BCMA guidelines. 98 ITEMS GROUPED STEP Figure 4 -5 ERROR 1 2 1 5 0 . 2 2 4 1 2 2 2 * 2 3 1 0 O . 3 2 5 9 8 3 7 • 3 8 9 0 . 4 7 0 0 0 6 3 * 4 7 1 3 0 . 4 8 5 8 8 6 8 * 5 1 2 1 6 1 . 1 0 8 4 1 6 6 * 6 8 1 2 1 . 1 1 8 5 0 8 3 * 7 7 1 1 1 . 1 7 0 4 3 9 7 8 5 1 4 1 . 2 3 5 4 1 2 6 * 9 2 3 1 . 3 3 3 7 8 4 1 * 1 0 1 2 2 . 7 7 8 0 7 7 1 * 1 1 4 6 3 . 6 2 3 6 9 8 2 * 1 2 1 8 4 . 3 5 7 5 8 9 7 * 1 3 4 5 6 . 3 6 2 6 9 4 7 * 1 4 1 7 1 1 . 3 7 8 7 7 1 * 1 5 4 1 7 1 7 . 7 3 8 6 7 8 * . 1 6 1 4 3 1 . 2 8 7 5 2 1 * 1 0 1 6 1 7 1 5 1 2 T' LI 1 3 1 1 1 4 T_ Dendrogram from cl u s t e r i n g of HRST plow layer data in spring. Clustered on Ca, CEC, pH, %C, and P. Table 4-5 Plot # and parameter values associated with cluster unit members of HRST plow layer data i n spring. Plots are arranged in dendrogram order, column one of plot ID i d e n t i f i e s the field,columns three and four i d e n t i f y plot number. P A R A M E T E R A N D V A L U E P L O T U N I T I D C a M q N a K C E C P H %c N O ^ - N p 2 1 1 1 1 9 . 4 0 1 . 9 8 0 0 5 0 5 5 1 7 6 5 8 2 3 4 5 3 4 8 . 0 2 1 2 1 1 1 2 . 3 6 2 . 0 9 0 0 5 O 4 2 1 8 1 6 3 1 8 2 1 3 3 6 . 0 2 1 1 6 1 1 1 2 . 2 3 1 . 7 8 0 0 4 0 5 8 1 9 6 6 3 2 5 2 8 2 3 8 . 0 2 1 3 1 1 1 3 O O 1 . 9 1 0 0 4 0 4 7 1 8 5 6 6 2 5 2 4 0 4 8 . 0 A 2 1 1 1 1 1 1 3 . 1 4 1 . 8 0 0 0 4 0 4 9 2 1 1 6 4 2 8 2 1 2 5 0 . 0 2 1 9 1 1 7 . 3 7 3 . 5 7 0 1 0 0 0 2 1 2 4 6 4 2 4 3 0 3 4 4 . 0 2 1 1 0 1 1 1 0 . 0 1 1 . 6 4 0 0 3 0 4 4 1 3 9 6 4 1 7 2 2 0 3 6 . 0 2 1 1 3 1 1 8 . 2 3 1 . 8 7 0 0 3 0 4 9 1 5 7 6 4 1 4 2 6 2 2 4 . 0 2 1 1 7 1 1 5 . 8 1 1 . 7 2 O 0 6 O 5 1 1 8 3 6 5 1 9 3 5 3 4 2 . O 2 1 7 1 1 9 . 7 7 2 . 5 4 0 0 5 0 2 0 1 5 5 7 1 0 7 1 2 2 2 0 . 0 B 2 1 1 4 1 1 8 . 8 0 1 . 5 8 0 0 5 0 3 5 1 3 3 6 8 1 1 2 4 8 1 8 . 0 2 1 1 2 1 1 8 5 4 1 . 1 7 0 0 3 O 3 2 1 4 2 6 9 1 3 2 5 3 3 8 . 0 2 1 4 1 1 2 2 . 1 . 6 2 . 0 3 0 0 3 0 5 2 2 2 0 7 1 3 6 3 2 0 5 6 . 0 C 2 1 6 1 1 2 3 5 1 2 . 2 8 0 0 4 0 8 0 2 6 4 6 9 0 7 3 5 3 6 4 . O 2 1 5 1 1 1 2 . 6 9 1 . 8 7 0 0 4 0 6 8 1 8 5 6 8 2 0 2 8 7 7 6 . 0 2 1 1 5 1 1 1 3 4 2 1 . 9 5 P 0 5 P 5 C 1 5 9 7 1 1 6 2 9 P 5 8 O 2 1 1 8 1 1 2 0 7 8 2 . 7 9 0 0 5 0 8 8 3 1 8 6 1 6 0 3 9 0 5 8 O 99 A A C B A 18 17 16 15 14 13 B A A A B 12 11 10 9 8 7 C C C A A A 6 5 4 3 2 1 UNIT PLOT# Fig u r e 4-6 SPACIAL DISTRIBUTION OF HRST CLUSTER UNIT MEMBERS TABLE SHOWING SIGNIFICANT DIFFERENCES Table 4-6 BETWEEN CLUSTER UNITS * (only c l u s t e r i n g parameters checked f o r s i g . d i f f e r e n c e ) A B C A pH %C Ca pH P B Ca* P* %C* CEC* C * i n d i c a t e s t h a t v a l u e s are not s i g n i f i c a n t l y d i f f e r e n t a t the 95% l e v e l , (by Mann-Whitney U-test) but there i s not o v e r l a p of data v a l u e s . 100 ITEMS GROUPED STEP I J ERROR 1 6 17 0.0392054 2 9 12 0.4077622 3 5 8 0.5288830 4 6 14 ' O.7189637 5 7 16 0.7554817 6 6 1 1 1.3469353 7 9 15 1.3748283 8 7 13 1.5037069 9 9 10 2.6959743 10 3 4 2.7054052 1 1 2 7 3.2535639 12 5 6 4.0867004 13 1 2 6.5103121 14 3 9 11.455968 15 1 5 15.739188 16 1 3 31.873734 1 13 17 4 10 2 5 14 9 16 8 11 12 6 3 15 Figure 4-7 Dendrogram from cl u s t e r i n g of CLPRO plow layer data in spring, Clustered on Ca, Na, pH, %C, and NO3-N. 0 Table 4-7 Plot # and parameter values associated with cluster unit members of CLPRO plow layer data in spring. Plots are arranged in dendrogram order, column one of plot I D i d e n t i f i e s the f i e l d , columns three and four i d e n t i f y plot number. PLOT PARAMETER AND VALUE UNIT ID Ca Mg Na K CEC pH %C NO3 -N P 11 111 21 3? 2 21 0 09 0 45 121 4 3 9 4 1 5 20 2 138 0 11 211 14 ? 1 1 21 0 04 0 48 65 1 5 1 19 4 12 3 144 .0 A 11 811 19 63 1 76 0 05 0 77 106 9 4 6 37 8 14 8 204 .O 111911 21 72 2 30 0 10 0 76 94 7 5 0 30 8 14 2 204 .0 111611 22 03 1 50 0 06 0 76 89 4 5 1 28 1 20 0 240.0 11 511 26 68 4 10 0 20 0 83 1 18 8 5 0 40 8 24 0 150.0 11 911 24 45 2 48 0 12 0 61 129 7 4 9 42 4 24 . 0 126 . 0 11 611 28 53 3 59 0 14 0 80 12 1 0 5 3 42 2 17 . 5 128 .0 B 112011 28 27 4 36 0 12 0 88 121 8 5 3 42 0 18 O 136 .0 111711 28 76 1 99 0 06 0 64 112 2 5 3 36 9 19. 8 20.0 111311 33 96 4 18 0 15 0 69 134 1 5 5 46 6 17 . 7 66.0 i\ $\\ 29 66"'" 2 56 0 4 1 0 77 91 2 6 0 28 5 10. 0 156 .0 11 4 11 26 96 3 57 0 25 0 75 93 3 5 5 37 7 8 . 8 164 .O 111011 31 12 2 75 0 16 0 83 45 8 5 2 17 3 10. 5 102 .0 C 111511 30 62 2 19 0 13 1 58 60 1 5 3 16 8 14 . 3 225 .0 111811 35 85 3 18 0 13 1 58 64 0 5 8 18 8 1 1 . 2 226 .O 111111 24 21' 2 07 0 08 0 57 81 5 5 7 14 6 9. 3 178 O 101 A A C C B B 1 2 3 4 5 6 A B C C B 8 9 10 11 12 13 C A B C A B 15 16 17 18 19 20 UNIT PLOT# F i g u r e 4-8 SPACIAL DISTRIBUTION OF CLPRO CLUSTER UNIT MEMBERS Table 4-8 TABLE SHOWING SIGNIFICANT DIFFERENCES BETWEEN CLUSTER UNITS * A B C A Ca Na pH,%c CEC P Ca ,Na pH,%C B pH %C CEC P NO 3 C * i n d i c a t e s t h a t v a l u e s are not s i g n i f i c a n t l y d i f f e r e n t a t the 95% l e v e l , (by Man-Whiteney U-test) but there i s no o v e r l a p of data v a l u e s . 102 CLUSTERING THREE FIELDS ON THE BASIS OF PLOW LAYER DATA Plow layer samples col l e c t e d from 3 f i e l d s in spring were clustered to determine (a) whether the e f f e c t of intensive management was so great as to mask the differences that would be expected to be inherited from parent material differences and (b) i f f i e l d s within the same s o i l series have unique chemistry. The parameters chosen for c l u s t e r i n g , on the basis of p r i n c i p a l components analysis, were Ca, K, CEC, pH, and NO3-N. The clus-tering dendrogram is shown in F i g . 4-9 while the accompanying data i s shown in Table 4-9. Table 4-10 shows s i g n i f i c a n t d i f f e r -ences among cl u s t e r i n g units. The clustering of plow layer data was highly successful with only 4 out of 51 samples c l a s s i f i e d as part of a d i f f e r e n t f i e l d . Three of RN plots (unit D) were c l a s s i f i e d as part of RC (unit E & F) and one RN plot was c l a s s i f i e d as a HRST (unit A, B & C). In addition to separating f i e l d s that had d i f f e r e n t parent materials ( ie HRST vs RN or RC ) the method was able to separate two f i e l d s which are located on the same parent material and s o i l series . While intensive management might be expected to ameliorate differences among f i e l d s , the r e s u l t of this analysis shows that the plow layers of i n d i v i d u a l f i e l d s have a unique chemical i d e n t i t y . 103 ITEMS GROUPEO STEP ERROR O t 28 30 0 0446773 2 38 41 0 0519814 3 47 48 0 065164 1 4 2 11 0 0733829 5 3 8 0 0843768 e 3 16 0 1619698 7 42 51 0 1657859 e 39 43 O 1833399 9 19 20 0 2054O05 10 46 49 0 2239828 11 44 50 0 2278172 12 4 5 0 2622739 13 2 14 0 267 3090 14 26 29 0 2860777 15 19 27 0 3545980 16 22 32 0 3713120 17 39 45 0 3751 130 18 3 9 0 4066071 19 12 13 0 4411944 20 42 47 0 5002070 21 40 44 0 5140747 22 12 37 0 5393366 23 38 46 0 5487280 24 4 17 0 7831124 25 26 28 0 7896174 26 1 19 0 8369961 27 35 42 0 9069808 28 3 10 0 9094539 29 6 7 1 0013952 30 22 31 1 0138388 31 15 36 1 1495047 32 21 23 1 2975655 33 18 33 1 3794603 34 2 3 1 5086489 35 1 22 1 9292316 36 24 26 1 9951601 37 24 25 2 5536032 38 35 40 2 6278038 39 2 6 2 93405O6 40 38 39 3 34 10921 41 21 34 4 1343699 42 2 4 4 1586361 43 15 35 6 5504570 44 1 18 6 8017826 45 12 38 7 9264555 46 1 24 9 8556862 47 12 15 It ) 663940 48 2 12 27 474991 49 1 21 39.749237 50 1 2 96.331116 B T L Figure 4-9 Dendrogram for c l u s t e r i n g of three f i e l d s ' data from the plow l a y e r , l e t t e r s i n d i c a t e c l u s t e r u n i t s . Table 4-9 p l o t # and parameter va lues a s s o c i a t e d w i t h c l u s t e r u n i t members from plow l a y e r data of th ree f i e l d s . P l o t s are arranged i n dendrogram o r d e r , column one of p l o t ID i d e n t i f i e s the f i e l d , columns three and four i d e n t i f y p l o t number. PARAMETER AND VALUE PLOT UNIT ID Ca Mq Na K CEC pH %c NO3-N p 31 111 13 9 3 0 . 8 7 0 11 0 6 0 2 0 0 6 3 2 5 1 3 . 1 48 0 21 2 1 1 12 3 6 2 . 0 9 0 0 5 0 4 2 18 1 6 3 1 8 2 1 . 3 36 0 21 3 1 1 13 .00 1 . 9 1 0 0 4 0 4 7 18 5 6 . 6 2 5 24 0 48 0 2 1 1 1 1 1 13 14 1 . 8 0 0 04 0 4 9 21 1 • 6 . 4 2 8 2 1 . 2 5 0 0 A 21 5 1 1 12 6 9 1 . 8 7 0 0 4 0 6 8 18 5 6 . 8 2 C 28 . 7 76 0 2 1 1 6 1 1 12 . 2 3 1 . 7 8 0 0 4 0 5 8 19 6 6 . 3 2 5 2 8 . 2 38 0 2 1 1 5 1 1 13 4 2 1 . 9 5 0 0 5 0 5 0 15 9 7 . 1 1 6 2 9 . 0 5 8 0 21 111 9 . 4 0 1 . 9 8 0 0 5 0 5 5 17 6 5 . 8 2 3 4 5 . 3 48 0 2 1 1 7 1 1 5 81 1 72 0 0 6 0 51 18 3 6 . 5 1 9 3 5 3 4 2 0 31 7 H § 7 7 2 . 5 4 0 0 5 O 2 0 1S 5 7 . 1 6 7 12 . 2 2 0 6 2 1 1 0 1 1 10 0 1 i . 6 4 0 0 3 0 44 13 9 6 . 4 1 7 2 2 . 0 36 0 B 2 1 1 3 1 1 8 2 3 i . 8 7 0 0 3 0 4 9 15 7 6 . 4 i 4 2 6 . 2 ' 24 0 2 1 1 2 1 1 8 5 4 i . 17 0 0 3 0 3 2 14 2 6 . 9 1 3 2 5 . 3 38 0 2 1 1 4 1 1 8 8 0 1 . 5 8 0 0 5 0 3 5 13 3 6 . 8 i 1 24 . 8 18 0 21 9 1 1 7 3 7 3 . 5 7 0 10 0 0 2 12 4 6 . 4 2 4 3 0 . 3 44 0 ii iii 2 2 5 6 2 . 0 3 0 0 3 0 5 2 22 0 7 . 1 3 6 32 O 5 6 0 C 2 1 6 1 1 2 3 51 2 . 2 8 0 0 4 0 8 0 2 6 4 6 . 9 0 7 3 5 . 3 6 4 0 2 1 1 8 1 1 2 0 7 8 2 . 7 9 0 0 5 0 8 8 31 8 6 . 1 6 0 3 9 . 0 58 0 31 iii 1 38 0 8 9 0 12 0 6 7 2 2 4 5 . 3 3 7 1 3 . 0 84 0 3 1 1 2 1 1 7 13 1 14 0 0 9 0 6 0 2 2 1 5 . 1 3 6 1 3 . 8 4 8 0 3 1 1 5 1 1 5 2 7 1 . 0 8 0 10 0 5 9 22 5 4 . 9 3 5 1 4 . 3 64 0 31 3 1 1 5 15 O . 8 9 o 13 0 7 7 21 9 4 . 9 3 4 9 . 8 9 5 0 3 1 8 1 1 5 57 0 . 9 8 0 10 0 71 21 O 5 . 0 3 . 2 1 2 . 2 6 4 0 D 3 1 1 7 1 1 5 6 3 0 9 6 0 0 9 0 7 2 2 3 4 5 . 0 3 9 1 0 . 3 8 8 0 31 9 1 1 5 2 5 0 . 8 3 0 10 0 91 2 2 0' 5 . 1 3 7 9 . 2 8 0 0 3 1 1 1 1 1 5 15 1 . 17 0 15 0 9 5 2 5 O 4 . 7 4 0 1 1 . 5 8 0 0 31 6 1 1 8 3 7 1 . 3 6 0 0 8 1 0 8 24 3' 5 . 4 3 2 16 . 0 5 6 0 31 7 1 1 6 13 0 9 7 0 0 7 0 9 2 21 4< 5 . 0 3 . 6 24 O 6 4 0 31 4 1 1 8 4 3 1 0 6 o 12 0 6 7 2 8 1< 5 . 3 3 9 16 . 3 72 0 31 5 1 1 7 41 1 2 7 0 13 0 6 6 2 6 O 5 . 0 3 9 1 6 . 5 84 0 3 1 1 8 1 1 5 5 2 0 8 2 0 Q6 0 7 8 2? 2' 4 . 9 9 9 1 1 . 2 52 0 3 1 1 3 1 1 3 9 4 1 0 7 0 0 5 0 7 8 16 3i 4 . 8 3 . 3 S O 52 0 3 1 1 4 1 1 5 6 4 1 14 0 0 6 0 9 8 16 0< 5 . 0 3 .4 1 0 . 3 6 4 0 41 3 1 1 6 19 1 2 7 0 2 3 0 8 5 16 8< 5 . 2 2 1 2 . 3 7 6 0 41 4 1 1 8 2 3 1 5 5 0 2 7 0 8 4 17 7t 5 . 8 2 . 4 5 . 3 ' 9 6 0 E 41 7 1 1 8 4 2 1 12 0 3 0 0 7 8 17 2« 5 . 9 2 . 2 5 . 0 148 0 4 1 1 3 1 1 9 6 3 1 2 3 0 2 9 0 9 7 17 7i 6 . 0 2 .0 7 . 2 120 0 4 1 1 6 1 1 IO 2 7 1 61 o 4 2 0 8 9 16 CX 6 . 2 2 0 7 . 2 1 2 0 0 41 5 1 1 1 1 7 8 0 9 4 0 2 5 1 10 21 9< 6 . 1 2 6 5 . 0 1 16 0 4 1 9 1 1 9 8 7 0 8 2 0 4 4 1 0 7 2 0 6< 6 . 0 2 8 3 . 8 168 0 4 1 1 1 1 1 10 4 4 1 9 5 0 2 9 1 1? 21 2" 6 . 0 2 4 1 2 . 3 152 0 3 1 1 6 1 1 5 5 9 1 . 14 0 13 0 0 3 21 8' 5 . 0 3 . 4 7 . 3 100 0 4 1 2 1 1 5 2 3 o 8 6 o 4 3 0 21 17 3' 5 . 5 2 O 4 . 0 128 0 41 111 7 0 8 0 9 0 0 5 2 0 5 0 13 5' 6 . 1 1 4 3 . 3 112 0 4 1 8 1 1 8 1 1 0 . 8 2 0 3 6 0 6 0 18 6' 6 . 2 1 . 9 3 . 8 108 0 F 4 1 1 8 1 1 6 4 3 0 . 9 4 o 2 2 0 51 18 1' 6 . 3 1 9 4 . 4 168 0 41 141 1 6 0 3 1 . 0 6 0 4 0 0 4 6 17 7' 5.9 1 . 8 5 . 2 156 0 41 1511 7. 17 1 2 3 0 3 0 o 4 6 16 8< 5.8 1 7 4 . 7 1 12 0 41 6 1 1 6 i i 0 . 7 0 4 5 5 0 6 9 21 2' 5.5 2 1 8 . 3 108 0 4 1 1 0 1 1 9 21 o 9 8 0 4 2 0 6 2 2 0 2' 5 . 9 2 3 3 . 3 120 0 * 1 1 7 1 1 7 6 0 0 . 8 2 0 3 5 0 54 2 0 6 5.6 2 . 5 5.3 160 0 105 Table 4-10 S i g n i f i c a n t differences among cluster units for plow layer data from 3 f i e l d s in spring. A B C D E F A K CEC Ca Ca CEC Ca CEC pH N 0 3 K pH N O 3 Ca pH NO3 B Ca K CEC NO3 Ca K CEC N O 3 K CEC NO3 Ca K CEC N O 3 C Ca N 0 3 Ca CEC NO3 Ca CEC N 0 3 D N CEC PH N O 3 K pH CEC NOj E K F * l i s t e d parameters are s i g n i f i c a n t l y d i f f e r e n t at the 95% l e v e l , only cl u s t e r parameter checked 106 A COMBINED APPROACH TO FERTILITY ASSESSMENT A combined approach i s one where the results of detailed sampling and cluster analysis are used i n combination. The cost of a combined approach i s considerably higher than that of detailed sampling followed by a singl e parameter assessment for P and K. Laboratory costs for obtaining CEC f %C, and NO3 -N measurements account for most of the additional expense. The cost of running the cluster analyis program and having the interpretations done, would add another small amount. Costs not included i n this evaluation are the cost of sampling as well as additional labor and management costs. The analyses performed for the conventional approach are only for P and K. The analyses performed for the combined approach are for the exchangeable bases, t o t a l CEC, pH, %C, NO3-N, and p. The cost for 17 samples i s $3 44 for the combined approach compares with $245 for the detailed approach. Though the cost of a combined approach i s about 40% higher than that of the detailed approach for P and K, i t i s not excessive i f sampling i s carried out once every three years. The additional information so gained would allow the farmer to reduce some of the causes, other than f e r t i l i z e r l e v e l s , which might be contributing to variable or other than optimum crop y i e l d s . The occasional use of a combined method would allow an on-going check to determine i f additional management practices were having the desired e f f e c t of reducing v a r i a b i l i t y . 107 Though there are often s i g n i f i c a n t differences i n nutrient l e v e l s among clustered units, the use of unit designation as a basis for f e r t i l i z e r recommendation i s often inappropriate. The reason i s that the cluster r e s u l t s , which might be considered a natural c l a s s i f i c a t i o n , generally do not coincide with the s o i l test l i m i t s set by the BCMAF. In order that information obtainable by cluster results i s not forgone, a combined approach, including information from cluster analysis and a more conventional use of detailed sampling, would be most appropriate. The farmer would thus be able to optimize f e r t i l i z e r usage while working to ameliorate differences among units, other than P and K l e v e l s , that might cause uneven yields throughout the f i e l d . The unit designation and lime requirement for RN is an instance in which the BCMAF and cluster analysis c l a s s i f i c a t i o n s do coincide (Table 4-11). Three of the RN units encompass only one lime requirement and the fourth encompasses two. This degree of agreement between the range of values found i n a unit and the range for s p e c i f i c recommendations does not, however, hold for unit P values found i n RN and HRST. Table 4-12 shows that each HRST unit generally encompasses more than one s o i l test range and, therefore, more than one f e r t i l i z e r recommendation. Plow layer data, c o l l e c t e d i n spring, for HRST and RN indicate that the %carbon-CEC factor , mentioned i n the previous section, i s a n o n - f e r t i l i z e r factor p o t e n t i a l l y causing y i e l d differences among units. In addition to having a positive e f f e c t on the s o i l ' s physical properties this factor i s important i n the retention of added K. High l e v e l s of exchangeable K at 60/90 cm 108 Table 4-11 Comparing a multivariate to a single parameter approach for assessing lime requirement for RN i n spring DETAILED SAMPLING pH RESULTS 5.4 5.2 5.0 5.1 4.8 5.0 5.4 5.0 5.0 5.1 5.0 5.3 5.0 5.0 6.4 5.1 4.9 MULTIVARIAT ANALYSIS UNIT DISTRIBUTION c C B B A A B A A A C A C D C D pH lime requirement (tonnes/ha) 4.0- 4.5 4.6-5.0 5.1- 5.5 5.6-6.0 unit pH range 9.0 7.0 4.5 2.0 lime recommendation A 4.8-5.3 B 5.0-5.4 C 5.0-5.4 D 4.9-5.0 4.5-7.0 4.5 4.5 7.0 Additional information gained from multivariate method Range of observations UNIT CEC N03 -N P COMMENTS  A 21.8-25 7.3-13 80-100 mod. CEC, low NO3-N, high P B 26-29.2 11.2-16.5 52-84 high CEC, high NO3-N, mod P C 21-24 .3 12.2-24 48-64 mod. CEC, high NO3-N, low P D 16-16.3 8-10.3 52-64 low CEC, low NO3-N, low P 109 Table 4-12 Assessing the usefulness of cluster units in recommending P for HRST i n spring. c B 6 64 12 38 18 58 C A A 5 76 11 50 17 42 C A A 4 56 10 36 16 38 A A C 3 48 9 44 15 58 A B 2 36 8 14 18 A B A 1 48 7 20 13 24 Plot# Unit P level SUMMARY OF P LEVELS AND UNIT DESIGNATION FOR HRST APPROXIMATE P2O5-P BCMA RECOMMENDATION UNIT RANGE RATING kg/ha j I I I A 24-50 M H 45-134 B 18-38 M M + 90-134 C 56-76 H H + 45 * * starter effect i n some areas; recommendation for crop group 4 110 depth for a l l f i e l d s suggests that the plow layer s o i l ' s a b i l i t y to f i x K or retain i t on the exchange capacity has been exceeded. In the cluster analysis of HRST data, three plots were i d e n t i f i e d as a cluster with low organic matter and low CEC (unit B). A fourth plot (# 13), was apparently m i s c l a s s i f i e d ; i t should also be part of unit B. Associated with the low organic matter and CEC of unit B, are K le v e l s that are lower than those i n most other parts of the f i e l d (Table 4-5). The low suface K values do not, however, coincide with equivalently low concentrations of K at 60/90 cm. Plot # 7 has the highest K value of a l l samples taken from 60/90 cm for this f i e l d . The f a i l u r e for low K values at 60/90 cm to coincide with low surface K values indicates that excessive leaching has occurred. Unit D of RN (Table 4-1) has a problem similar to that described for HRST in that the unit has low CEC. The difference i s that the low CEC does not appear to be related to low organic matter content. While the %C is r e l a t i v e l y high, the CEC of these plots is about 16 me/lOOg compared to a CEC of 20-25 me/lOOg for unit A and C and 26-29 me/lOOg for unit B. I l l SUMMARY Cluster analysis of f i e l d s ' plow layer data, on an indi v i d u a l f i e l d basis, indicates that multivariate analysis can be useful for c l a s s i f y i n g f i e l d s into s i g n i f i c a n t l y d i f f e r e n t units. Though units may be s i g n i f i c a n t l y d i f f e r e n t on the basis of several of the examined parameters the u t i l i t y of this c l a s s i f i c a t i o n is l i m i t e d from a P and K f e r t i l i z a t i o n point of view, since c l a s s i f i c a t i o n units often contain two or more P and K f e r t i l i z e r recommendatios. Multivariate c l a s s i f i c a t i o n would be very useful in elucidating management units which may be c o n t r i -buting to uneven crop y i e l d s , due to factors other than insuf-f i c i e n t P and K l e v e l s . Burrough (1982) points out that "multivariate analysis techniques can be very valuable for revealing how a s o i l varies s p a c i a l l y as a multivariate phen-onenon. Differences among units has l a r g e l y been attributed to differences in what has been described as the Carbon-CEC factor. The Carbon-CEC factor i s made up of CEC, %C, pH, and Ca. Nelson and Anderson (1980) note the actual yields usually do correlate well with the s o i l test results because of the influence of factors other than the measured variable. Among the factors causing the poor c o r r e l a t i o n are s o i l texture, structure, CEC, and organic matter. These factors are either included in or are influenced by the carbon-CEC factor. Optimization of f e r t i l i z e r application could possibly be brought about by amelioration of f i e l d units which have been 112 recognized as having a less than desirable l e v e l s of those parameters contributing to the %carbon-CEC factor. In part this may be brought about by additions of organic matter. An approach to s o i l f e r t i l i t y management which combines a single-parameter approach and a multivariate assessment, based on detailed sampling r e s u l t s , may provide the best means of optimizing f e r t i l i z e r usage. Generally, the single parameter method i s able to provide the best method of assessing P and K f e r t i l i z e r requirements but the multivariate approach provides a means of determining management units which require more than just d i f f e r e n t f e r t i l i z e r , applications to reduce the v a r i a b i l i t y of factors contributing to variable crop y i e l d s . Determination of management units also means the farmer can treat groups of plots in the same way, rather than treating a l l plots as in d i v i d u a l s . The cost of the combined method i s about 40% higher than the single-parameter detailed sampling method, when only s o i l analysis costs are considered. In part, the possible success of detailed sampling and a combined approach, involving multivariate analysis, i s due to the scale f a c t o r . While the f i e l d i s generally used as a c l a s s i -f i c a t i o n unit for s o i l sampling purposes, the u t i l i t y of a c l a s s i s f i c a t i o n system is l i m i t e d by the precision with which the units are mapped (Hoffman, 1971). Detailed sampling provides a more precise characterization of the f i e l d due to the much larger scale of characterization. 113 CONCLUSIONS An experiment involving intensive sampling through the use of random s t r a t i f i e d , random s t r a t i f i e d composite, and conventional composite sampling was carried out on four f i e l d s . The f i e l d s were located on three d i f f e r e n t mapping units so that a comparison of f i e l d s on the same as well as d i f f e r e n t mapping units could be made. Three of the f i e l d s were sampled in both spring and f a l l and one was sampled in the spring only. Soils were sampled at three d i f f e r e n t depths and analysed for exchangeable Ca, Mg, Na, K, t o t a l CEC, pH, organic carbon, NO3-N, and Bray P-l phosphorus. The project was designed to determine and quantify the major factors which influence the effectiveness of s o i l sampling for f e r t i l i z e r p r e s c r i p t i o n of a g r i c u l t u r a l f i e l d s . The major factors examined were d i f f e r e n t f i e l d s , depth, time, and sampling method. Effectiveness was measured on the basis of a b i l i t y to estimate f e r t i l i z e r requirements for a l l parts of the f i e l d as well as the cost of the method. In addition, the usefulness of numerical techniques in distinguishing among f i e l d s and determining s i g n i f i c a n t l y d i f f e r e n t management units within the f i e l d s was explored. As a resu l t of the experimental work and data analysis the following conclusions were made. 114 1. ANOVA results showed, for most variables, that depth and f i e l d s explained the largest portion of to t a l variance. For P and K depth explained 65 and 60% respectively of to t a l variance. Time explained 2% or less of to t a l v a r i a b i l i t y for a l l variables. The contribution of interactions was generally small. The error term, which can l a r g e l y be attributed to within f i e l d variance, was 60, 68, 42, and 45% for Mg, Na, CEC, and NO3-N respectively but was considerably smaller for the other variables. On an o v e r a l l basis there are s i g n i f i c a n t differences among f i e l d s , depths, and seasons, but their i n t e r p r e t a t i o n i s complicated by s i g n i f i c a n t interactions of the main f a c t o r s . 2. Of the major f e r t i l i t y parameters studied the v a r i a b i l i t y of NO3-N, CEC, pH, and to a large extent P was lower before the growing season than after i t . The v a r i a b i l i t y of Ca, and K was lower i n the f a l l , a fter the cropping season. 3. On the basis of plow layer data taken from three f i e l d s in spring and in f a l l the CV for NO3-N, P and K i s 39, 47, and 71% respectively. The CV for Mg, %C, and Ca i s 23, 35, 36 and 38% respectively. pH is the least variable parameter, with a CV of 13%. Na had a CV of 245%. For K the plow layer was generally the least variable l a y e r . For P the 30/60 cm depth was the most variable depth for two of the three f i e l d s , with the plow layer and the 60/90 cm depths having approximately equal v a r i a b i l i t y . 115 4. Using the detailed sampling mean as standard for comparison, conventional sampling estimated the means for N, P, and K are as follows: 65% of the time conventional estimates were within 20% of the detailed sampling mean, 18% of the time the estimates are within 21-50% of the detailed mean, and about 2% of the time d i f f e r by more than 100%. S t r a t i f i e d random composite sampling estimates for N, P, and K never d i f f e r e d from the s t r a t i f i e d random mean by more than 50%, so that as a means of predicted mean f i e l d values i t may be considered as a worth while a l t e r n a t i v e . The estimates of mean pH values by either random s t r a t i f i e d composite and conventional methods were very accurate (< ± 5%). %C had moderate errors of estimate, which are generally less than ± 15%. 5. I f sample numbers required to estimate the f i e l d mean (plus or minus 10% with 90% confidence) are calculated i n d i s c r i m i n a t l y of sampling time, Reynolds Clover (RC) and Horstings (HRST) require 191 samples and Reynolds (RN) requires 88 samples to determine the mean NO3-N value. For P, a l l three f i e l d s require 38 samples. For potassium, RC and RN have approximately equal sampling requirements, 30 and 32 samples respectively. HRST requires 63 samples. 6. I t was shown that, as a resu l t of skewed or a wide range of s o i l f e r t i l i t y values, the use of mean s o i l f e r t i l i t y l e v e l s results in a f e r t i l i z e r recommendations, which leave considerable portions of f i e l d s o v e r f e r t i l i z e d and under f e r t i l i z e d . The additional cost of analyis for detailed s o i l sampling, which 116 provided accurate f e r t i l i z e r recommendations for a l l parts of the f i e l d , was shown to be o f f s e t by savings in the resulting f e r t i l i z e r a pplication. 7. The use of cluster analysis on plow layer samples from individual f i e l d s was successful in c l a s s i f y i n g s i g n i f i c a n t l y d i f f e r e n t units within the f i e l d s . The units would be of limited use as an aid to f e r t i l i z e r recommendation but are p o t e n t i a l l y useful in distinguishing among management units which have s i g n i f i c a n t l y d i f f e r e n t l e v e l s of n o n - f e r t i l i z e r variables. The most important of these variables are %C, pH, Ca, and CEC. 8. ANOVA showed that f i e l d s often had s i g n i f i c a t l y d i f f e r e n t mean values but also showed that a single variable was seldom able to discriminate among a l l f i e l d s . The multivariate approach of cluster analysis was highly successful in discriminating among f i e l d s on the of plow layer data alone. 9. S i g n i f i c a n t l y d i f f e r e n t mean values and i n d i v i d u a l ranges of CV for the various f i e l d s as well as the high degree of c l a s s i f i c a t i o n success shown by c l u s t e r analysis indicates a high degree of chemical uniqueness among f i e l d s . Since f i e l d s have a unique chemical v a r i a b i l i t y associated with them i t follows that conventional sampling seldom provides adaquate characterization of f i e l d f e r t i l i t y . This ultimately leads to less than optimum f e r t i l i z e r use. 117 REFERENCES Adams, J.A. and Wilde, R.H. 1980. Comparison of the variability in soil taxonomic units with that of associated soil mapping unit. Aust. Journ. of Soil Sci. 18 (3) 285-297. Allison, L.E. 1965. Organic matter determination; Walkley-Black method. In: C.A. Black (Ed.), Methods of Soil Analysis. Part 2. Chemical and microbiological properties. ASA monograph 9, Madison, Wisconsin, p. 1372-1376. Anonymous3. 1977. Technicon Autoanalyzer II Methodology: Determination of nitrate and nitrite in soils. Industrial Methods, No. 487-77A. Anonymous0. 1974. Technicon Autoanalyzer II Methodology: Individual/ simultaneous determination of nitrogen and phosphorus BD acid digests. Industrial Methods, No. 334-74A. Anonymous0. 1978. Soil Sampling. Soils Branch, Ministry of Agriculture, Province of B.C. 4 pp. Anonymous^ . 1978. What about the quality of soil samples? Better Crops With Plant Food, Potash and Phosphate Inst., Atlanta. Anonymouse. 1984. Crops and Soils Magazine. Vol. 36, p. 23-24. Ball, D.F. and Williams, W.M. 1968. Variability of soil chemical proper-ties in two uncultivated Brown Earths. J. Soil Sci., Vol. 19, No. 2, p. 379-391. Ball, D.F. and Williams, W.M. 1971. Further studies on variability of soil chemical properties: Efficiency of sampling programmes on an unculti-vated Brown Earth. J. Soil Sci., Vol. 22, No. 1, p. 60-68. Beckett, P.H.T. 1967. Lateral changes in soil variability. J. Austr. Inst. Agri. Sci. 33: 172-179. Beckett, P.H.T. and Webster, R. 1971. Soil variability: A review. Soil and Fert. 34 (1): 1-15. Blake, G.R. 1965. Bulk density; core method. In: C.A. Black (Ed.), Methods of Soil Analysis. Part I. Physical and mineralogical proper-ties. ASA monograph, Madison, Wisconsin, p. 375-377. Blyth, J.D. and Macleod, D.A. 1978. The significance of soil variability for forest soil studies in north-east Scotland. J. Soil Sci. 29: 419-430. Bracewell, J.M., G.W. Robertson and J. Logan. 1979. Variability of organic matter and exchangeable cations within the A2 horizon of an iron podzol. J. Soil Sci. 30: 327-332. 118 Brown, J.L. 1979. Systematic study of the v a r i a b i l i t y of podzolic s o i l along a trench i n a sugar maple-yellow b i r c h stand. Can. J . S o i l S c i . 59(2): 131-146. Burrough, P.A. 1982. Computer assistance for s o i l survey and land evaluation. S o i l Survey and Land Evaluation, Vol. 2, No. 2, 25-36. Cameron, D.R., Kowalenko, CG. and Campbell, C.A. 1979. Factors a f f e c t i n g n i t r a t e nitrogen and chloride leaching v a r i a b i l i t y i n a f i e l d p l o t . S o i l S c i . Soc. Am. J . , Vol. 43, 455-460. Cameron, D.R., Nyborg, M., Toogood, J.A. and Laverty, D.H. 1971. Accuracy of f i e l d sampling for s o i l t e s t s . Can. J . S o i l S c i . 51: 165-175. Campbell, J.B. 1979. S p a t i a l v a r i a b i l i t y of s o i l s . Association of American Geographers. 69: 544-556. Chapman, H.D. 1965. Cation-exchange capacity. In: C.A. Black (Ed.), Methods of s o i l a n a l y s i s . Part 2. Chemical and m i c r o b i o l o g i c a l properties. ASA monograph 9, Madison, Wisconsin, p. 891-901. Cirpa, J.E., O.W. Bidwell, D.A. Whitney, and A.M. Feyerherm. 1972. V a r i a -tions with distance i n selected f e r t i l i t y measurements of pedons of a western Kansas u s t o l l . S o i l S c i . Soc. Amer. P r o c , V o l. 36, 111-114. Drees, L.R. and Wilding, L.P. 1973. Elemental v a r i a b i l i t y within a sampl-ing u n i t . S o i l S c i . Soc. Amer. Proc. 37: 82-87. Fox, D.J. and Guire, K.E. 1976. Documentation for MIDAS. S t a t i s t i c a l research laboratory, The University of Michigan, 3rd e d i t i o n . Greig, M. and O s t e r l i n , D. 1978. UBC ANOVAR - Analysis of variance and covariance. Computing Centre, University of B r i t i s h Columbia, Vancouver, B.C., 69 pp. Greig, M. and Bj e r r i n g , B. 1980. U.B.C. Genlin: A general least squares analysis program. Computing Centre, U.B.C, Vancouver, B.C., Canada, 55 pp. Haines, S.G. and Cleveland, G. 1981. Seasonal v a r i a t i o n i n properties of f i v e forest s o i l s i n S.W. Georgia. S o i l S c i . Soc. Am. J . 45: 139-143. Hammond, T.C, T r i t c h e t t , W.L. and Chew, Y. 1958. S o i l sampling i n r e l a -t i o n to s o i l heterogeneity. S o i l S c i . Soc. Am. Proc. 22: 548-552. Heminway, R.C 1955. S o i l sampling errors and advisory analyses. J . A g r i c . S c i . 46: 1-7. Heyne, P.T. 1973. The economic way of thinking. Science research a s s o c i -ates, Inc. 119 Hoffman, D.W. 1971. The assessment of s o i l p r o d u c t i v i t y for a g r i c u l t u r e . Dept. of Land Resource Science, Un i v e r s i t y of Guelph, Guelph, Ontario, 51 pp. Husch, B., M i l l e r , C.I. and Beers, T.W. 1972. Forest Mensuration, 2nd Ed., The Ronald Press Co., New York, N.Y. 410 pp. Ike, A.F. and Cl u t t e r , J.L. 1968. The v a r i a b i l i t y of forest s o i l s on the Georgia Blue Ridge Mountains. S o i l S c i . Soc. Amer. Proc. 32: 284-288. Lee, R., Bailey, J.M., Northly, R.D., Barber, P.R. and Gibson, E.J. 1975. Va r i a t i o n i n some chemical and physical properties of three related s o i l types: Dannervirke s i l t loam, K i r v i t e a s i l t loam, and Morton s i l t loam. N.Z. Jour, of Agric. Res. 18: 29-36. Lewis, T. 1976. The t i l l - d e r i v e d podzols of Vancouver Island. Ph.D. Thesis, University of B r i t i s h Columbia, Vancouver, B.C., 159 pp. Luttermerding, H.A. 1980. S o i l s of the Langley-Vancouver map area. RAB B u l l e t i n 18, Report No. 15, V o l . 1. Mclntyre, G.A. 1967. S o i l sampling for s o i l t e s t i n g . J . Austr. Inst. Agr. Sc. 33: 309-320. Mader, D.L. 1963. S o i l v a r i a b i l i t y - A serious problem i n s o i l - s i t e studies i n the northeast. S o i l S c i . Sco. Amer. Proc. 27: 707-709. M o l l i t o r , A.V., Leaf, A.L. and Morris, L.A. 1980. Forest s o i l v a r i a b i l i t y on northeastern flood p l a i n s . S o i l S c i . Soc. Amer. Jour. 44 (3) 617-620. Murphy, J . and Ril e y , J.P. 1962. A modified single s o l u t i o n method for the determination of phosphate i n natural waters. Anal. Chem. Acta. 27: 31-36. Nelson, L.A. and Anderson, R.L. 1980. P a r t i t i o n i n g of s o i l test - crop response p r o b a b i l i t y . In: S o i l Testing: c o r r e l a t i o n and int e r p r e t i n g the a n a l y t i c a l r e s u l t s . ASA Special P u b l i c a t i o n No. 29, Madison, Wisconsin 53711, p. 19-38. Nelson, L.A. and McCracken, R.J. 1962. Properties of Norfolk and Ports-mouth s o i l s : s t a t i s t i c a l summarization and influence on corn y i e l d . Proc. S o i l S c i . Soc. Amer. 26: 497-502. Neufeld, J.H. 1980. S o i l testing methods and i n t e r p r e t a t i o n s . B.C. Min i s t r y of A g r i c u l t u r e . 29 pp. Patterson, J.M. and Whitaker, R.A. 1978. H i e r a r c h i c a l grouping analysis with optional contiguity constraint. UBC-CGR0UP, University of B r i t i s h Columbia, 21 pp. 120 Peck, T.R. and Dibb, D.W. 1979. Are you getting f u l l value from your s o i l test? Better Drops With Plant Food. Potash and Phosphate Inst., Atlanta, F a l l . Peterson, D.L. and Rolfe, G.L. 1982. Seasonal v a r i a t i o n i n nutrients of floodplains and upland forest s o i l s of central I l l i n o i s . S o i l S c i . Soc. Amer. Jour., V o l . 46, p. 1310. Peterson, D.L. and G.L. Rolfe. 1982. Seasonal v a r i a t i o n i n nutrients of f l o o d p l a i n and upland forest s o i l s of central I l l i n o i s . S o i l S c i . Soc. Am. J . 46: 1310-1315. Schreier, H. and Z u l k i f l i , M.A. 1983. A numerical assessment of s o i l survey data for a g r i c u l t u r a l management and planning. S o i l Survey and Land Evaluation, Vol. 3, No. 2 : 41-53. Sie g e l , S. 1956. Nonparametric s t a t i s t i c s : For the behavioural sciences. McGraw-Hill Book Co., New York, Toronto, London, p. 116-136. S l a v i n s k i , H.C. 1977. S o i l v a r i a b i l i t y along a topographic sequence -U.B.C. Endowment Lands. M.Sc. Thesis, University of B r i t i s h Columbia, Vancouver, B.C., Canada, 203 pp. Sokal, R.R. and Sneath, P.H.A. 1963. P r i n c i p a l s of numerical taxonomy. Freeman and Co., San Francisco and London. Webster, R. and Burrough, P.A. 1972. Computer-based s o i l mapping of small areas from sample data. I. Multi a r i a t e c l a s s i f i c a t i o n and ordina-t i o n . J . S o i l S c i . 23, 210-221. Webster, R. 1977. Quantitative and numerical methods i n s o i l c l a s s i f i c a -tion and survey. Clarendon Press, Oxford. Winter, G.R. 1981. Management of forage systems. Olds College, Olds, Alberta, Canada, p. 185-222. Zar, J.H. 1974. B i o s t a t i s t i c a l a n a l y s i s . P r e n t i c e - H a l l , Inc., Eglewood C l i f f s , N.J., 620 pp. 121 A p p e n d i x 1 RANDOM STRATIFIED SAMPLING DATA j 1 2 2 ID C « Kg Na K CEC pH XC N O j - N P B01 %l»01St BD2 22 131 3. 2 7 0 1 . 2 6 0 0 . 0 8 0 0 106 9 . 8 0 0 5 . 22 331 S . 3 9 0 1 9 4 0 0 . 086 0 . 106 1 1 5 0 0 6 . 22 431 5 . . 0 5 0 1 . 3 1 0 0 . 103 0 093 7 5 0 0 5 . 22 931 2. 6 9 0 0 6 3 0 0 . 054 0 . 138 5 . 8 0 0 5 . 22 631 6. 2 3 0 2 . 370 0 . 0 8 0 0 067 10 700 5 . 22 731 6 8 2 0 3 4 7 0 0 133 0 . 086 7 . 700 6. 22 B31 7 . 2 2 0 3 . 0 0 0 0 . 124 0 . 096 18 800 6 . 22 931 4 . 2 4 0 1 . 2 9 0 0 . 134 0 . 0 8 6 10. 8 0 0 5 2 2 1 0 3 1 2 . 8 4 0 0 730 0 . 1 17 0 . 036 7 . 200 5 2 2 1 1 3 1 5 . 8 4 0 1 9 8 0 0 . 147 0 . 0 8 0 13 700 6 2 2 1 2 3 1 3 . . 6 8 0 1 2 0 0 0 0 5 5 0 . 061 10. 20O 5 . 2 2 1 3 3 1 5 . 5 3 0 3 . 3 3 0 0 0 8 6 0 . 144 17 . 300 6 221431 4 . 170 3 0 0 0 0 064 0 . 077 10. 7O0 6 221531 3 . 7 1 0 0 . 6 5 0 0 0 5 9 0 . 042 6 . 6O0 5 2 2 1 6 3 1 3 .7 10 0 . 9 6 0 0 . 052 0 . 04 2 7 . 8 0 0 6 2 2 1 7 3 1 5 . 0 1 0 1 2 4 0 0 . 052 0 . 0 8 6 9 . 700 5 2 2 1 8 3 1 3 4 6 0 1 0 6 0 0 0 4 5 0 . 0 3 5 6 . 700 5 22 121 7 . 7 6 0 1 . 9 4 0 0 . 0 9 5 0 195 11 400 5 22 321 3 . 4 3 0 0 . 9 0 0 0 . 0 4 0 0 . 157 7 100 5 22 421 5 . 3 6 0 1 . 4 3 0 0 . 087 0 . 102 13 8 0 0 5 22 52 1 3 8 8 0 0 9 0 0 0 . 0 5 0 0 . 0 9 9 8 6 0 0 5 22 621 4 0 8 0 2 3 8 0 0 . 081 0 086 17 . 40O 5 22 721 8 . 6 8 0 4 . 260 0 145 0 150 3 1 400 6 22 821 4 . 9 2 0 1 4 10 0 131 0 0 5 1 8 9 0 0 6 22 931 6 . 6 8 0 1 . 8 4 0 0 165 0 093 14 8 0 0 5 . 221021 3 . 8 7 0 0 . 8 6 0 0 108 0 . 064 9 9 0 0 6 2 2 1 1 2 1 7 . 150 2 . 180 0 144 0 . 125 18 6 0 0 5 2 2 1 2 2 1 6 5 0 0 2 . 0 4 0 0 14 1 0 . 150 18 100 5 2 2 1 3 2 1 5 . 3 4 0 1 9 6 0 0 0 8 3 0 . 1 3 1 14 8O0 6 2 2 1 4 2 1 4 . 400 1 . 9 0 0 0 0 5 3 0 109 9 . 3 0 0 6 221521 5 . 6 3 0 1 . 0 4 0 0 058 0 007 12 30O 6 3 2 1 6 2 1 6 . 760 1 . 7 1 0 0 06 9 0 093 14 100 6 2 2 1 7 2 1 8 . 1 10 2 160 0 0 8 0 0 . 1 15 15 30O 5 2 2 1 8 2 1 6 . 2 1 0 1 8 4 0 0 . 0 5 9 0 .07 4 1 1 . 2 0 0 5 32 111 10 . 0 3 0 2 OOO 0 059 0 . 5 1 2 19 . too 6 22 31 1 8 . 170 1 . 160 0 . 0 4 8 0 . 3 2 3 13 5O0 6 22 411 11 . 6 2 0 1 . 4 9 0 0 056 0 . 449 22 100 6 22 51 1 15 . 180 1 9 2 0 0 . 0 7 4 0 . 3 8 6 25 700 6 22 61 1 10 . 2 8 0 1 7 7 0 0 . 0 4 6 0 4 18 6 300 6 22 71 1 10 . 2 5 0 2 3 3 0 0 . 0 7 8 0 . 266 i r . 700 7 22 81 1 6 . 9 7 0 1 . 4 5 0 0 . 0 7 8 0 . 2 4 0 8 100 6 22 91 1 9 . 8 6 0 1 5 3 0 0 . 128 0 4 11 22 6 0 0 6 2 2 1 0 1 1 9 . 5 1 0 1 . 120 0 lOO 0 349 2 1 . 5 0 0 6 2 2 1 1 1 1 11 . 4 7 0 1 7 5 0 0 . 101 0 . 4 1 1 25 . 100 6 221211 12 . 8 7 0 1 8 8 0 0 . 0 ? 1 0 317 25 . 8 0 0 6 221311 8 . 3O0 1 7 3 0 0 . 0 4 6 0 . 259 14 6O0 6 2 2 1 4 1 1 9 . 8 5 0 1 . 4 3 0 0 .04 3 0 . 2 6 2 15 . 0 0 0 6 2 2 1 5 1 1 1 1 . 7 2 0 1 . 6 3 0 0 0 5 3 0 .411 22 OOO 6 2 2 1 6 1 1 13 . 2 6 0 1 . 7 7 0 0 . 0 5 5 0 . 3 5 0 2 1 . 9 0 0 6 22 17 1 1 11 . 4 0 0 1 . 3 5 0 0 0 2 5 0 . 355 20 OOO 6 231811 1 1 . 4 0 0 1 . 7 7 0 0 . 0 3 2 0 . 3 5 5 20 . 5 0 0 6 32 131 1 . 9 5 0 1 4 2 0 0 318 o . 209 12 . 6 0 0 3 32 231 1 . 8 5 0 1 . 489 1 026 0 23<i 17 . too 3 4 2 0 0. 2 9 0 2 2 . 8 0 0 3 . 000 0. 0 0. 0 0 0 5 5 0 0. 320 11 300 3 . 0 0 0 0. 0 0 . 0 0. 0 6 2 0 0 . 190 18. 8 0 0 5 . 0 0 0 0. 0 0 0 0. 0 3 4 0 0. 130 7 . 200 7 . ooo 0. 0 0. 0 0. 0 270 0. 300 12 . 700 6 ooo 0. 0 0 . 0 0 . 0 6 7 0 0 . 310 2 . 8 0 0 2 . ooo 0. 0 0. 0 0. 0 5 0 0 0. 330 2 . 700 3 . 0 0 0 0. 0 0. 0 0 0 6 8 0 0. 350 9 . 3 0 0 5 . 0 0 0 0. 0 0 . 0 0 0 710 0 130 6 2 0 0 7 . ooo 0 0 0 0 0 . 0 0 3 0 0. 350 9 . OOO 3 ooo 0 0 0 . 0 0. 0 6 0 0 0. 390 11 8 0 0 5 ooo 0. 0 0. 0 0. 0 3 8 0 0. 7 5 0 6 8O0 4 0 0 0 0 0 0 . 0 0 0 5 9 0 0. 160 2 8 0 0 3 ooo 0 0 0 0 0. 0 780 0. 4 2 0 13 300 4 ooo 0 0 0 0 0 0 0 0 0 0. 190 3 300 4 . 0 0 0 0 . 0 0 . 0 0. 0 8 1 0 0 270 26 300 4 ooo 0 0 0 0 0 0 8 2 0 0. 130 1 1 5O0 4 . 0 0 0 0 0 0. 0 0. 0 9 6 0 1. 0 3 0 29 300 2 ooo 1 . 3 1 0 0 . 0 0 0 6 0 0 0. 190 3 700 3 ooo 1 . 2 9 0 0 0 0 0 4 4 0 0 310 13 700 6 0 0 0 1 . 2 9 0 0 0 0 0 8 3 0 0 330 10 710 8 . 0 0 0 1 . 3 4 0 0 0 0 0 190 0 5 4 0 14 300 4 . 0 0 0 1 . 4 2 0 0 0 0 0 8 1 0 0 . 4 10 3 5O0 2 . 0 0 0 I . 3 8 0 0 0 0 o 4 3 0 0. 160 3 . 5 0 0 4 . 0 0 0 1 . 5 3 0 0 0 0 0 7 6 0 0 5 2 0 5 3 0 0 3 ooo 1 . 3 4 0 0 0 0 . 0 0 5 0 0. 300 4 2 0 0 6 . 0 0 0 1 . 3 9 0 0 0 0 . 0 6 4 0 0 . 6 0 0 1 1 BOO 3 ooo 1 . 2 5 0 0 0 0 0 8 6 0 0 . 6 3 0 13 . 3 0 0 4 . 0 0 0 1 . 260 0 . 0 0 . 0 010 0 . 7 0 0 7 . 8 0 0 7 . 0 0 0 1 . 4 5 0 0 0 0 0 20O 0 . 2 6 0 3 OOO 4 ooo 1 . 3 8 0 0 . 0 0 . 0 0 6 0 0 . 4 3 0 9 . 6 0 0 4 . 0 0 0 1 . 3 6 0 0 . 0 0 0 300 0 . 3 0 0 3 . 700 3 . 0 0 0 1 . 3 9 0 0 . 0 0 . 0 6 5 0 0 . 4 5 0 29 300 3 . 0 0 0 1 . 2 4 0 0 . 0 0 . 0 . 7 1 0 0 . 2 7 0 28 . 3 0 0 3 . 0 0 0 1 . 390 o 0 0 . 0 0 7 0 2 160 3 . 700 6 ? ooo 1 310 0 . 0 1 . 1O0 3 4 0 1 8O0 5 . 8 0 0 66 0 0 0 1 . 4 2 0 0 . 0 0 . 0 . 8 9 0 1 . 6 8 0 5 20O 52 0 0 0 1 . 3 3 0 0 0 1 . too 7 5 0 2 . 7 0 0 17 . 7 0 0 65 . 0 0 0 1 . 0 6 0 0 . 0 0 . 0 . 2 6 0 3 . 0 6 0 4 7 0 0 62 . 0 0 0 1 . 2 1 0 o 0 1 . 270 . 120 0 . 6 8 0 4 70O 29 .ooo 1 . 5 2 0 0 0 1 . 3 5 0 360 0 . 6 8 0 3 7O0 23 ooo 1 . 3 6 0 0 0 1 . 3 7 0 . 6 3 0 1 . 8 4 0 4 BOO 63 0 0 0 1 . 190 0 . 0 1 . 120 . 5 9 0 1 . 8 2 0 8 . 8 0 0 64 0 0 0 1 . 0 9 0 0 . 0 0 . 0 . 170 2 . 5 9 0 13 OOO 73 ooo 1 . 2 1 0 0 0 1 . 0 2 0 3 9 0 3 . 4 4 0 15 . 300 54 ooo 1 . 130 0 . 0 0 . 0 . 6 6 0 1 . 2 7 0 4 6O0 44 0 0 0 1 4 3 0 0 0 1 . 4 2 0 . 9 9 0 1 . 0 8 0 5 . 0 0 0 61 0 0 0 1 . 490 0 . 0 1 . 4 2 0 6 0 0 1 . 6 6 0 5 300 51 .ooo 1 . 300 0 . 0 1 . 180 . 7 9 0 1 . 9 8 0 5 . 7 0 0 39 ooo 1 . 2 8 0 0 0 1 . too . 8 5 0 2 . 180 16 OOO 61 ooo 1 . 2 3 0 0 0 1 . 130 . 4 3 0 1 . 8 0 0 7 . 700 45 ooo 1 . 2 2 0 0 0 0 . 0 . 9 2 0 1 . 4 8 0 7 60O 13 ooo 0 . 0 0 0 0 0 . 9 1 0 1 . 450 12 . 6 0 0 1 1 ooo 0 . 0 0 0 0 . 0 1 2 3 O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O M - - « O N > O M O O O O O O O O o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o • - O O - O - c 8 8 8 8 8 § 8 8 8 8 8 8 8 § 8 8 ^ §§§§§§8§§§§§§§§§§§I§§§l!§§I§8§§§§§§§§§§§§i§§§l§§§§§§l§§§ o o o o o o o - - - o - - o - o o o o o o o - - * - o - - o - » — — - - * - o — - - » o - * o o o o o o o o o o o o o o o 0 0 0 0 0 0 0 0 0 0 0 < > - a O < D « l D < g i O C 9 M M O l D O - > d l b O - ' - 0 ' * < D - 0 - I D - * 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 '§§3 833 §§83 S8 3 ^ o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o b o o o b o i j o o o o o o o o o o o o o o o o o o o o o c o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o b o o o o o o o b b o o b b o o o o o b b o o o o o o o o o o b b o o I D C a Mg N a K C E C p H %C N O j - N P B 0 1 % m o 1 s t B D 2 4 2 9 3 1 2 3 8 0 0 8 7 7 0 1 5 3 0 . 1 8 0 i ; 2 0 0 4 OOO 1 . 0 5 0 3 . TOO 1 6 . 0 0 0 0 . 0 0 0 0 0 4 2 1 0 3 1 2 9 0 0 1 4 5 0 0 . 1 5 2 0 1 9 1 18 1 0 0 4 4 3 0 0 . 9 5 0 1 . 0 0 0 16 OOO 0 0 0 . 0 0 0 4 2 1 1 3 1 2 . 8 4 0 1 0 4 0 0 1 5 4 0 . 15 1 18 . 100 4 4 3 0 1 . 3 3 0 3 . 3 0 0 18 ooo 0 . 0 0 0 0 0 4 2 1 2 3 1 2 . 1 6 0 1 . 3 0 8 0 . 1 5 4 0 . 1 8 9 2 0 9 0 0 4 0 4 0 1 1 9 0 2 . 8 0 0 18 0 0 0 0 . 0 0 0 0 . 0 4 2 1 3 3 1 1 . 7 1 0 1 6 6 1 0 1 9 7 0 . 2 0 2 16 7 0 0 3 9 1 0 0 8 9 0 3 . OOO 15 0 0 0 0 . 0 0 0 0 0 4 2 1 4 3 1 2 3 3 0 1 . 4 8 3 0 1 9 4 0 . 1 8 0 18 6 0 0 4 0 6 0 1 . 1 5 0 3 . 2 0 0 18 ooo 0 0 0 0 0 0 4 2 1 5 3 1 1 1 5 0 1 1 5 7 0 1 9 0 0 . 1 9 4 17 2 0 0 3 5 5 0 1 1 2 0 3 OOO 15 o o o 0 . 0 0 0 0 0 4 2 1 6 3 1 2 3 9 0 1 6 7 9 0 2 5 8 0 . 2 5 3 17 . 2 0 0 4 3 3 0 0 9 4 0 5 . 5 0 0 1 1 ooo 0 0 0 0 0 0 4 2 1 7 3 1 2 . 7 9 0 0 8 9 1 0 1 4 2 0 . 1 3 2 15 3 0 0 4 3 9 0 0 6 3 0 4 5 0 0 12 ooo 0 . 0 0 0 0 0 4 2 1 8 3 1 2 1 1 0 1 2 1 0 0 1 6 4 0 . 181 2 0 . OOO 3 9 4 0 1 0 8 0 9 OOO 15 ooo 0 0 0 0 0 0 4 2 2 2 1 2 8 6 0 0 9 0 6 0 0 9 5 0 . 2 2 5 2 2 100 4 0 4 0 1 . 6 0 0 7 . 6 0 0 2 4 ooo 1 2 1 0 0 0 0 0 4 2 3 2 1 0 1 0 0 1 . 2 4 4 0 0 B 6 0 . 3 4 1 19 1 0 0 4 1 5 0 1 3 8 0 3 9 0 0 15 ooo 1 . 1 2 0 0 0 0 0 4 2 4 2 1 3 . 7 2 0 1 5 9 9 0 0 6 0 0 . 7 C 2 17 . 2 0 0 4 5 4 0 1 1 6 0 7 . BOO 15 0 0 0 1 . 2 4 0 0 0 0 0 4 2 5 2 1 7 . 3 0 0 2 0 8 3 0 1 0 6 0 . 2 2 3 7 1 9 0 0 5 2 5 0 1 1 8 0 9 0 0 0 13 ooo 1 . 1 8 0 0 0 0 . 0 4 2 6 2 1 4 7 4 0 1 . 4 8 5 0 1 4 4 0 1 8 7 18 eoo 4 4 0 0 o . 9 6 0 5 7 0 0 16 ooo 1 . . 2 3 0 0 0 0 0 4 2 7 2 1 4 8 1 0 0 7 2 8 0 0 8 9 0 . 2 3 2 2 1 4 0 0 4 6 3 0 2 1 4 0 1 1 3 O 0 19 . 0 0 0 1 0 8 0 0 0 0 . 0 4 2 8 2 1 7 6 0 0 0 2 1 0 0 . 1 0 1 0 3 7 5 2 3 3 0 0 5 . 5 7 0 1 . 6 1 0 12 9 0 0 19 . 0 0 0 1 1 6 0 0 0 0 . 0 4 2 9 2 1 S . 2 5 0 0 951 0 1 5 2 0 2 10 1 9 100 4 6 6 0 1 4 2 0 4 OOO 34 ooo 1 2 3 0 0 0 0 0 4 2 1 0 2 1 4 . . 7 6 0 1 . 2 5 0 0 . 1 3 8 0 . 2 1 8 18 100 4 8 2 0 1 0 2 0 2 6 0 0 13 . 0 0 0 1 1 9 0 0 0 0 0 4 2 1 1 2 1 7 2 5 0 1 . 6 4 6 0 1 6 8 0 2 0 0 19 5 0 0 5 1 5 0 1 . 0 9 0 16 0 0 0 15 0 0 0 1 0 6 0 0 0 0 0 4 2 1 2 2 1 5 . 6 0 0 1 . 8 8 3 0 1 7 8 0 . 1 9 8 2 0 5 0 0 5 0 5 0 1 . 1 8 0 1 0 2 0 0 15 . 0 0 0 1 , . 0 9 0 0 0 0 0 4 2 1 3 2 1 4 . 1 3 0 1 1 6 9 0 1 5 8 0 2 5 2 2 3 . 7 0 0 4 . 5 2 0 0 . 8 4 0 9 OOO t 7 . 0 0 0 1 2 7 0 0 0 0 0 4 2 1 4 2 1 6 8 3 0 1 5 5 9 0 1 5 0 0 2 1 1 2 0 0 0 0 4 9 1 0 1 . 3 3 0 16 7 0 O 2 1 o o o 1 2 0 0 0 o 0 0 4 2 1 5 2 1 3 . 5 6 0 0 9 9 3 0 1 9 7 0 . 1 6 2 19 . 1 0 0 4 . 1 1 0 1 . 4 8 0 8 . 3 0 0 17 . 0 0 0 1 . 3 5 0 0 . 0 0 . 0 4 2 1 6 2 1 5 8 2 0 1 1 3 2 0 1 7 9 0 2 0 0 1 9 . IOO 4 6 5 0 1 . 3 0 0 5 2 0 0 2 4 0 0 0 1 2 5 0 0 . 0 0 0 4 2 1 7 2 1 5 8 9 0 0 9 8 3 0 1 4 8 0 . 2 7 6 2 2 3 0 0 4 7 7 0 1 . 1 6 0 12 . 7 0 0 9 0 ooo t 0 5 0 0 0 0 0 4 2 1 8 2 1 6 6 8 0 1 6 3 0 0 1 4 3 0 2 2 8 1 9 . 5 . 0 6 0 1 . 1 2 0 13 8 0 0 2 4 0 0 0 1 . 2 2 0 0 0 0 0 4 2 2 1 1 7 0 8 0 0 9 6 5 0 0 6 3 0 8 9 1 2 2 . 100 5 o e o 2 0 8 0 1 3 . t o o 1 0 9 ooo 1 . 1 8 0 0 0 0 0 4 2 3 1 1 5 8 4 0 2 . 1 7 5 0 0 6 5 0 7 9 6 22 100 4 9 O 0 2 . 2 0 0 1 1 1 0 0 6 1 . 0 0 0 1 . 2 2 0 0 0 0 0 4 2 4 1 1 6 0 3 0 1 . 9 2 2 0 0 7 0 0 8 7 2 2 1 4 0 0 5 0 5 0 1 . 7 8 0 5 . 9 0 0 4 5 ooo 1 . 1 9 0 0 . 0 0 0 4 2 5 1 1 12 . 6 7 0 1 9 9 1 0 0 5 0 0 . 8 4 7 2 1 /10O 6 . 3 5 0 2 . 0 6 0 1 1 0 0 0 9 7 ooo 1 . 2 4 0 0 0 0 0 4 2 6 1 1 1 0 . 1 1 0 1 5 1 0 0 0 8 5 0 5 8 8 19 5 O 0 5 . 7 7 0 1 . 4 4 0 7 . 0 1 0 9 1 . 0 0 0 1 . 3 1 0 0 0 0 0 4 2 7 1 1 1 0 2 3 0 1 2 4 4 0 . 0 7 8 0 8 0 3 2 3 ^ 0 0 5 . 5 8 0 1 . 9 9 0 1 1 . 0 0 0 1 0 3 . 0 0 0 1 . 2 4 0 0 0 0 0 4 2 8 1 1 1 0 . 9 2 0 2 0 2 B 0 . 0 7 7 0 7 5 8 2 1 . 4 0 0 5 . 9 3 0 1 . 7 5 0 2 0 . 2 0 0 8 0 . 0 0 0 1 . 2 2 0 0 . 0 0 0 4 2 9 1 1 1 0 8 10 1 5 4 6 0 1 1 7 0 916 2 1 . 4 1 0 5 . 7 1 0 2 . 3 6 0 9 OOO 1 16 000 0 . 8 3 0 0 . 0 0 0 4 2 1 0 1 1 1 0 . 6 3 0 1 . 6 0 5 0 1 13 0 7 2 1 3 3 3 0 0 5 . 6 3 0 2 . 2 5 0 3 . 8 0 0 1 3 5 . 0 0 0 1 . 2 1 0 0 .0 0 . 0 4 2 1 1 1 1 1 0 . 9 0 0 1 9 1 1 0 . 1 0 5 0 9 6 1 2 3 3 0 0 5 . 7 9 0 2 1 3 0 17 OOO 1 6 6 . 0 0 0 1 . 2 5 0 0 0 0 0 4 2 1 2 1 1 11 . 0 1 0 1 4 6 7 0 1 2 3 0 . 7 5 8 2? . 3 0 0 5 . 5 6 0 2 4 2 0 2 1 . 5 1 0 8 7 0 0 0 1 2 5 0 0 0 0 0 4 2 1 3 1 1 9 . 9 3 0 1 . 2 0 4 0 1 0 5 1 0 Q 3 2 6 .000 5 5 8 0 2 . 2 1 0 9 . 9 0 0 8 3 . 0 0 0 1 . 2 4 0 0 0 0 0 4 2 14 1 1 1 0 . 2 5 0 1 . 7 9 5 0 1 0 6 0 9 3 5 2 3 . 7 0 0 5 . 8 0 0 2 . 0 0 0 2 2 . 1 0 0 6 5 0 0 0 1 . 3 1 0 0 . 0 0 0 4 2 1 5 1 1 1 0 . 2 7 0 2 . 0 3 0 0 . 1 1 2 0 7 3 3 20 C O O 6 0 8 0 1 . 7 1 0 10 2 0 O 5 4 ooo 1 . 3 1 0 0 0 0 . 0 4 2 1 6 1 1 9 . 8 2 0 1 . 0 6 3 0 1 1 3 0 9 7 3 3 4 . 200 5 . 5 6 0 2 . 2 1 0 15 7 0 0 8 5 0 0 0 1 . 2 4 0 0 . 0 0 . 0 4 2 1 7 1 1 1 0 . 4 0 0 1 4 18 0 1 0 1 1 , 2 0 O 2 2 . 8 0 0 5 . 5 5 0 2 . 5 6 0 2 6 . 7 0 0 9 1 . 0 0 0 1 1 5 0 0 0 0 0 4 2 1 8 1 1 8 . 7 7 0 .. 2 . 0 3 0 0 1 1 2 0 7 5 8 19 . 5 0 0 5 . 8 7 0 1 . 7 8 0 19 8 0 0 7 1 . 0 0 0 1 . 3 5 0 0 . 0 0 0 3 1 1 3 1 1 . 8 7 0 1 1 4 0 0 2 9 1 0 1 9 3 15 OOO 3 . 9 5 0 2 . 1 0 0 15 . 5 0 0 1 0 . 0 0 0 0 . 0 0 . 0 0 0 3 1 2 3 1 0 . 9 4 0 1 OOO 0 . 8 1 0 0 2 5 0 15 . 4 0 0 3 . 9 7 0 1 . 5 0 0 9 . 2 0 0 2 7 ooo 0 . 0 o . 0 0 . 0 3 1 3 3 1 1 . 2 4 0 1 . 3 7 0 1 . 0 1 5 0 2 6 4 13 . 9 0 0 4 . 1 7 0 1 . 3 O 0 7 . 3 0 0 1 9 ooo 0 0 0 . 0 0 . 0 3 1 4 3 1 1 . 2 6 0 1 . 3 6 0 1 . 4 9 0 0 2 2 7 19 OOO 4 1 3 0 1 . 5 0 0 13 . 0 0 0 19 . 0 0 0 0 . 0 0 . 0 0 . 0 3 1 5 3 1 1 . 3 4 0 1 . 6 6 0 1 . 5 0 1 0 . 3 1 5 17 . too 4 . 0 6 0 1 . 7 0 0 1 1 . 8 0 0 2 4 ooo 0 0 0 0 0 0 3 1 6 7 1 1 . 5 7 0 1 1 4 0 0 . 8 7 5 0 . 2 16 17 9 0 0 4 . 0 8 0 1 . 9 0 0 16 . 7 0 0 19 . 0 0 0 0 . 0 0 . 0 0 0 3 1 7 3 1 1 . 3 O 0 0 . 6 3 0 0 . 1 6 4 0 . 1 8 7 17 OOO 4 . 0 4 0 1 9 O 0 12 . 3 0 0 2 1 ooo 0 0 0 . 0 0 0 3 1 8 3 1 1 . 5 0 0 0 7 6 0 0 . 1 7 8 0 . 2 3 4 14 . 8 0 0 4 2 9 0 2 . 3 0 0 4 5 0 O 1 0 . 0 0 0 0 . 0 0 . 0 0 . 0 3 1 9 1 1 1 . 1 10 1 . 4 5 0 1 . 2 6 4 0 2 5 5 17 BOO 3 . 8 1 0 2 . 2 0 0 7 . 0 0 0 2 1 ooo 0 . 0 0 0 0 0 3 1 1 1 3 1 1 . 4 5 0 1 . 4 9 0 1 . 0 1 5 0 . 22-1 16 J O O .1 . 0 5 0 1 . 7 O 0 8 .200 16 ooo 0 . 0 0 . 0 0 0 1 2 5 uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu to (TV o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o c o o o o o o o o o o o o o o o o o o o • - -OOOOO--OOO--OOO-OOOOO--OO--Oa3OOCDO^O^(T>!»CBt»UO3<0—O^O^0^3(C(^  OGOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 0 0 0 0 - - 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 1 <X> CD CD — L O uu< OOCOO-OOOOOOOOOOOOOOOOOO-OOOOOOOOOOOOOOOOOOOOOOOOOOOOO X — i» — U 0*1 O tfJ <TI «J 05 CJ '0 H <T) O CP — — "O -B 71 "/^•JOLKTUC'JJlOI'JU'lfl'JOC — -7^C/»OQT) £» T> 3> <D OOOOOOOOOOOO0^ y'^ 03cTiy)Att(Tt'Jl(nai'JlK)U0BKJfO t* — CD — Jl O -J1<J.&(0>T>3>'J1<J10B»J ikKjyfl-J n OU'0'CDCIU110-»J1^<OOK)*0 b CD Ul ( 88; ) O '-J ' !83S; S el OCOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO CJ — — (J ^/ o o o o o o o a o o o o o o o o o o o o o o o o o o o o o o o • U 0 0 ' 0 ' 0 ' ^ U U U ' 0 > ' ^ 0 O U ' l f l ^ U J l J I 4 ) l 0 U U U ^ O ^ > l ^ A O l 0 J k « U < > l U U ) A I D U A r j l < J 0 - > U ) ' J U M W « l l ' W U M U g U U U U U O O I U I M \ j O O U , U | B O - * t ' l O U I J I | > V O >IUUO)U - J ^ O D O U t f l O t J I U I f l O l I I 5 8 8 8 8 8 8 8 8 3 8 8 8 8 8 8 8 8 8 8 3 8 8 8 8 8 8 8 3 8 8 8 8 8 8 8 8 8 8 S S S 8 8 8 8 8 8 8 8 8 8 8 8 !3§! ) o 5 < \m o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o - - — o o o — o o o o - o — o o o o o o o OOOOOOOOOOOOO^<DU)<O<0(9<D(DtocO<OOD(BaDaB<O^a(O^-*(DtOU}OCDU)a}U)O(DQOOOOOOO au(OMUiObm^Ooso<T)OOOOB'MU)ui(D^>i(o^ocoij)NUiaioO OOOOOOOOOOOOOUlUima'<iIi(OWUbftil>J(j>i<<lOO^U>>'*8l( OOOOOOOOOC • tn -» KJ !t> CP -. _> CD CP ji cP ( > O O O O O O ( • a t o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o b o o b o o o o o o o o o o b b o b b o o o b o o b b b o o o o o b o o o o b o b o o o o o o o b o o ID Ca Mg Na K Cf c pH %C NOj -N P BD 1 Xmo 1 s 1 BD2 111631 13. 140 12 . 800 0. 910 0. 830 57 . 900 4 . 350 20 OOO 0 300 21 000 0. 0 0 0 0. 0 1 1 1731 6. 110 6. 640 0 860 0. 740 32 OOO 3 700 7 . 1B0 0. 100 16. OOO 0. 0 0. 0 0. 0 1 1 1831 3. SOO 2 690 0. 450 0. 500 23 200 3. 590 2 . 950 2 . 200 2? OOO 0. 0 0 0 0. 0 111931 3. 650 4 . 820 0 420 0. 560 19 300 3 450 1 . 920 0. 300 15 000 0. 0 0 0 0. 0 1 12031 15 550 6. 310 0. 210 0. 620 83 300 4 . 940 37. COO 3 000 0. 0 0. 0 0. 0 0. 0 11 121 9 860 3. 400 0 390 0. 770 59. OOO 4 010 12 . 700 7 500 93 000 0. 206 79 330 0 0 11 221 6. 1 10 1 . 190 0. 260 0. 500 65. 100 4 . 130 1 . 680 6 500 19 000 1. 040 35 . 700 0. 0 11 321 8 750 2 . 810 0. 320 0. 740 49 300 4 . 330 8 . 170 2 000 21 000 0. 294 69. 680 0 0 1 1 421 4 . 920 2 . 340 0. 450 0. 470 25 100 3 860 6 . 050 1 BOO 24 000 1. 030 35. 790 0. 0 1 1 521 4 530 3 510 0 490 0. 4B0 37 4O0 3 780 6. 770 0 500 17 000 0. 933 38 BOO 0. 0 1 1 621 24 700 4 890 0. 450 o. 660 144 . 6O0 4 520 46. 590 5. 700 60 000 0. 216 77 7B0 0. 0 1 1 821 4. 220 1 290 0. 140 0. 6 10 31 600 3 . 730 4 . 580 4 500 28 ooo 0 788 45 490 0 0 1 1 92 1 9. 670 4 690 0. 540 0. 880 70 600 .1, 630 14 . 380 3 700 40 000 0. 198 80. 050 0. 0 111021 1 1 . 820 3. 790 0. 390 0. 340 1 15. 300 4 200 35. 7 30 5 . 300 16 .000 0. 352 64 . 760 0. 0 111121 6. 090 2 460 0. 160 0. 450 25 . 000 4 360 3 . 600 3 300 16 000 0 677 54 . 880 0 0 111221 5. 480 2. 670 0. 280 0. 930 43 400 3 880 4 . 740 0 300 26 .000 0 769 44 830 0. 0 111321 22 480 10. 330 0. 590 0. 4 10 162 200 * 630 47 . 260 9 '00 96 ooo 0. 187 80 390 0 0 11152 1 7 . 360 1 820 0. 130 0. 840 57 900 3 570 12 340 7 . 500 2 1 ooo 0 255 7 1 460 0 0 11162 1 9 450 2 -580 0. 170 0. 430 80. 600 4 240 23 . 170 5 500 49 ooo 0. 220 76 170 0 0 11172 1 9. 490 2 990 0 280 0. 7 30 68 400 3 680 14 . 640 5 800 36 000 0. 210 78 500 0 .0 1 1 182 1 6 060 1 . 330 0. 180 0. 810 26 700 4 130 3 700 3 700 27 000 0. 926 37 7 30 0 0 11192 1 3 860 2 320 0 240 0. 500 18 400 3 760 2. 070 0 100 26 000 0 881 40 070 0 .0 112021 19 010 4 570 0 130 0. 620 104 300 4 770 31 . 620 to 700 0 0 0. 396 62 240 0 .0 11 111 21 320 2 . 210 0 090 0. 450 121 4O0 3 910 41 500 70 2O0 138 ooo 0 466 46 100 0 .0 11 211 14 210 1 2 10 0. 040 0 480 65 10O 5 140 19 400 13 300 144 ooo 0 614 36 BOO 0 .0 11 311 29 060 2 560 0 4 10 0. 770 91 200 6 030 28 .500 10 000 156 000 0 572 42 300 0 .0 11 4 11 26 960 3. 570 0 250 0 750 93 300 5 .560 37. 770 8 BOO 164 ooo 0 .519 47 . 950 0 .0 11 511 26 680 4 100 0 200 0 830 1 18 . BOO 5 010 40 820 24 . OOO 150 ooo 0 449 54 440 0 .0 11 611 28 .530 3 590 0 140 0 800 12 1 .000 5 .370 42. 280 17 . 500 128 .000 0 398 53 660 0 .0 1 1 8 11 19 630 1 760 0 050 0 . 770 106 . 900 4 680 37 B20 14 800 204 ooo 0 453 47 .740 0 0 11 911 24 .450 2 480 0 120 0 610 129 . 700 4 970 42 400 24 000 176 000 0 . 434 50 4 10 0 .0 11lOI1 31 . 120 2 750 0 160 0 8"<0 4* 800 5 .260 17 370 10 50O 102 000 0 .789 36 810 0 .0 111111 24 . 2 10 2 .070 0 080 o .570 8 1 , 5O0 5 7 10 14 .610 9 3O0 17B ooo 0 617 36 . 700 0 .0 111311 33 .960 4 180 0 150 o .690 134 10O 5 500 46 .6 10 i ; 70O 66 .ooo 0 .368 58 290 0 .0 111511 30 .620 2 190 0 130 1 . SRO 80 too 5 .390 16 800 1-> 300 235 ooo 0 .702 37 .540 0 .0 1 1 16 1 1 22 .830 1 500 0 060 0 . 760 89 .400 5 . 150 28 1B0 20 000 240 ooo 0 .538 46 .240 0 .0 1117 11 28 .760 1 .990 0 O60 0 640 1 12 200 5 . 340 36 .980 19 800 70 ooo 0 434 5 1 220 o .0 1 1 18 1 1 35 .850 3 180 0 no 1 . 5*0 6 1 . OOO 5 810 18 8B0 1 1 2O0 226 ooo 0 862 36 .020 0 0 1119 11 21 .720 2 . 300 0 too 0 . 7S0 91 . 700 5 030 30 . 830 14 .200 204 .000 0 517 45 .860 0 o 112011 28 . 270 4 .960 0 120 0 .880 12 1 800 .320 42 .050 18 000 136 .000 0 .351 57 .900 0 .0 1112 11 25 .440 3 .570 0 . 130 0 . 790 t 14 BOO 5 . 240 37 .430 70 000 104 000 0 .447 54 .880 0 .0 21 131 6 .670 1 700 0 .065 0 139 10 .900 6 .080 0 360 5 BOO 3 ooo 0 .0 0 .0 0 .0 2 1 23 1 5 .950 1 . 1 10 0 068 0 .083 7 . 90O 6 . 260 0 . 260 3 . 5O0 2 ooo 0 0 0 .0 0 .0 21 331 8 . 460 2 400 0 .057 0 107 15 .800 6 .090 1 120 5 BOO 3 .000 0 0 0 .0 0 .0 2 1 43 1 4 .9 10 1 .450 0 .056 0 07O 5 . 300 6 . 340 0 . 300 5 . 200 3 ooo 0 .0 0 .0 0 o 21 531 3 . 720 0 . 840 0 038 0 . 202 6 .500 5 950 0 240 3 .300 7 .000 0 .0 0 .0 0 .0 21 631 7 . 940 2 . 4BO 0 .07 4 0 139 15 000 6 260 0 . 740 4 800 4 ooo 0 .0 0 0 0 .0 2 1 73 1 1 1 . 970 2 .050 0 .073 0 385 19 OOO 6 .500 0 220 3 .500 3 ooo 0 0 0 .0 0 .0 21 931 6 .OOO 1 . 4 10 0 096 0 08 2 1 1 . 700 6 OOO 0 . 330 3 OOO 4 .000 0 .0 0 o 0 .0 211031 4 . 370 1 . 130 0 .057 0 . 083 7 . 7O0 .990 0 .240 5 . 300 6 ooo 0 .0 0 0 0 .0 211131 3 .690 1 . 1O0 0 046 0 095 8 .800 5 770 0 . 230 4 . 200 7 ooo 0 0 0 .0 0 0 211231 4 . 190 1 .520 0 . OF8 o 0R9 6 OOO 6 070 0 . 150 2 . 200 3 ooo 0 0 0 .0 0 .0 2 1 1331 4 . 390 1 . 4 10 o . OSR 0 025 6 500 6 . 370 0 .370 3 BOO 3 ooo 0 .0 0 0 0 0 2 1143 1 7 . 060 3 120 o . 101 0 .12 1 12 OOO 6 . 540 O . 360 4 OOO 2 .ooo 0 0 0 .0 0 0 127 I D C a M o N a K C € C p H %C HOyN P B D 1 t m o l s t B D 2 2 1 1 5 3 1 4 5 6 0 1 2 9 0 0. 0 7 0 0 . 1 4 0 7 4 0 O 6 1 3 0 0. 3 3 0 3 7 0 0 4 0 0 0 0. 0 0. 0 0. 0 2 1 1 6 3 1 6 2 6 0 1 . 6 6 0 0 . 0 7 1 0. 1 7 2 1 0 . 8 0 0 6 . 0 1 0 0. 3 O 0 3 5 0 0 5 . 0 0 0 0. 0 0. 0 0. 0 2 1 1 7 3 1 2 3 1 0 1 . 4 B 0 0. 0 7 1 0 1 4 6 8 8 0 0 6 . 2 1 0 0 3 8 0 4 . 3 0 0 2 OOO 0 0 0 . 0 0 0 2 1 1 8 3 1 4 . 1 6 0 2 4 8 0 0 0 6 5 0. 3 2 1 1 3 . 3 0 0 6 . 4 4 0 0. 6 3 0 5 . 5 0 0 5. OOO 0. 0 0 . 0 0. 0 2 1 1 2 1 7 . 7 3 0 2 . 2 0 0 0. 0 5 3 0 2 5 9 I S 0 0 0 5. 9 2 0 0. 8 1 0 5 . 2 0 0 4 0 0 0 1 . 2 5 0 3 4 7 7 0 0. 0 2 1 2 2 1 9. 4 6 0 1 . 7 2 0 0. 0 7 7 0. 0 9 5 14 . 1 0 0 6 . 4 4 0 0. 7 7 0 6 . 7 0 0 2 . OOO 1 . 2 0 0 3 4 . 0 5 0 0. 0 2 1 3 2 1 1 2 . 6 2 0 2 . 6 2 0 0. 0 8 7 0. 1 14 2 3 4 0 0 5. 9 6 0 2 . 7 7 0 1 0 . 3 0 0 4 . OOO 1 . 1 5 0 3 4 . 9 5 0 0 0 2 1 4 2 1 7 . 8 5 0 2 . 4 2 0 0 . 0 6 4 0. 1 0 1 1 5 OOO 6 . 5 4 0 0. 7 1 0 1 5 . 8 0 0 2 OOO 1 . 2 0 0 3 5 . 9 7 0 0 0 2 1 5 2 1 3. 7 2 0 0. 8 4 0 0 . 0 4 0 0 . 1 4 5 4 9 0 0 6 . 1 5 0 0. 3 7 0 2 7 0 0 2 0 OOO 1 . 3 1 0 7 . 6 0 0 0. 0 2 1 6 2 1 11 . 1 6 0 2 . 5 6 0 0. 0 8 6 0. 3 0 3 2 1 . 6 0 0 6 . 0 7 0 2 . 6 9 0 9 7 0 0 5 0 0 0 0. 9 3 8 3 1 . 0 9 0 0. 0 2 1 7 2 1 8. 1 0 0 3. 8 5 0 0. 1 2 0 0. 1 2 6 16 4 0 0 6 . 4 7 0 0. 3 9 0 4 . 0 0 0 4 . ooo 1 . 2 6 0 2 5 . 4 3 0 0. 0 2 1 9 2 1 8. 0 2 0 2 . 0 5 0 0. 1 0 8 0. 1 3 3 17 7 0 0 5 . 9 5 0 1 . 1 0 0 8 7 0 0 5 ooo 2 4 0 2 5 . 5 5 0 0 0 2 1 1 0 2 1 4 . 3 6 0 0 9 6 0 0. 0 4 3 0. 1 5 9 7 3 0 0 6 . 0 5 0 0. 2 5 0 5 7 0 0 7 0 0 0 1 . 3 4 0 11 . 2 3 0 0. 0 2 1 1 1 2 1 s. 8 0 0 1 . 6 0 0 0. 0 6 2 0. 134 15 1 0 0 5 8 9 0 0 7 4 0 6 8 0 0 B . 0 0 0 1 . 1 2 0 2 7 , 8 9 0 0 0 2 1 1 2 2 1 4 . 6 1 0 1 . 3 5 0 0. 0 4 9 0. 1 6 4 8 . ooo 6 2 1 0 0. 3 3 0 4 . 7 0 0 3 . 0 0 0 1 . 3 5 0 12 . 8 6 0 0 0 2 1 1 3 2 1 7 . 7 8 0 2 4 4 0 0 0 8 1 0. 1 9 7 12 2 0 0 6 4 7 0 0. 8 4 0 6 7 0 0 2 ooo 1 . 1 2 0 2 6 2 2 0 0 0 2 1 1 4 2 1 6 2 4 0 2 . 0 5 0 0 0 8 9 0. 1 2 7 ' 15 5 0 0 6 3 3 0 0. 3 9 0 3 3 0 0 3 ooo 1 . 4 0 0 2 1 2 0 0 0 0 2 1 1 5 2 1 6 . 6 5 0 1 . 7 0 0 0. 0 5 2 0 . 3 1 2 15 9 0 0 6 4 2 0 0 8 9 0 8 0 0 0 1 3 000 1 . 2 9 0 2 2 8 1 0 0. 0 2 1 1 6 2 1 8. 4 4 0 1 . 9 5 0 0. 0 6 3 0 2 7 9 15 9 0 0 6 . 0 4 0 0 9 6 0 8 OOO 6 . 0 0 0 0. 9 1 0 2 5 3 2 0 0 .0 2 1 1 7 2 1 3 . 7 3 0 2 1 5 0 0 0 6 9 0 3 8 2 1 5 9 0 0 6 3 8 0 0. 5 8 0 8 OOO 5 . 0 0 0 0. 9 0 0 3 0 8 7 0 0 0 2 1 1 8 2 1 4 . 7 6 0 2 . 2 3 0 0. 0 6 1 0 3 0 0 19 4 0 0 6 1 9 0 1 . 5 1 0 1 0 2 0 O 5 ooo 1 . 0 4 0 3 0 OOO 0 0 2 1 1 1 1 9 4 0 0 1 . 9 8 0 0. 0 5 6 0 . 5 5 6 17 6 0 0 5 . 8 5 0 2 3 3 0 4 5 3 0 0 4 8 ooo 0. 9 7 6 2 3 1 0 0 0 0 2 1 2 1 1 1 2 . . 3 6 0 2 0 9 0 0. 0 5 9 0 . 4 7 3 18 1 0 0 6 . 3 0 0 1 , 8 6 0 2 1 3 0 0 3 6 ooo 0 9 1 0 2 2 4 0 0 0 . 0 2 1 3 1 1 1 3 . 0 0 0 1 . 9 1 0 0 0 4 6 0 4 7 7 18 . 5 0 0 6 . 6 6 0 2 5 4 0 2 4 OOO 4 B .000 1 . 0 1 0 3 1 . 1 2 0 0 . 0 2 1 4 1 1 2 2 . 5 6 0 2 0 3 0 0 0 3 8 0 . 5 2 8 2 2 OOO 7 . 1 0 0 3 6 8 0 3 2 ooo 5 6 ooo 0 8 5 0 2 6 1 6 0 0 0 2 1 5 1 1 1 2 6 9 0 1 8 7 0 0 0 4 9 0 6 8 3 18 . 5 0 O 6 . 8 5 0 2 . 0 3 0 2 8 7 0 0 7 6 ooo 0 . 9 9 2 18 6 1 0 0 . 0 2 1 6 1 1 2 3 5 1 0 2 2 8 0 0 0 4 6 0 8 0 9 2 6 . 4 O 0 6 . 9 3 0 0 7 8 0 3 5 . 3 0 0 6 4 .000 0 7 3 3 2 7 . 3 7 0 0 .0 2 1 7 1 1 9 . 7 7 0 2 5 4 0 0 0 5 8 0 2 0 8 15 . 5 0 0 7 . 1 5 0 0 . 7 7 0 12 2 0 0 2 0 ooo 1 . 1 0 0 18 0 6 0 0 . 0 2 1 9 1 1 7 . 3 7 0 3 . 5 7 0 0 1 0 5 0 0 2 5 12 4 O 0 6 . 4 3 0 2 4 1 0 3 0 . 3 0 0 44 ooo 0 . 9 5 1 2 2 . 1 4 0 0 .0 2 1 1 0 1 1 1 0 . 0 1 0 1 6 4 0 0 . 0 3 8 o 4 4 3 13 . 9 1 0 6 . 4 5 0 1 . 7 7 0 2 2 . 0 0 0 3 6 . 0 0 0 0 9 9 5 2 1 6 4 0 0 . 0 2 1 1 1 1 1 13 . 1 4 0 1 . 8 0 0 0 . 0 4 1 0 . 4 9 6 2 1 . 1 0 0 6 . 4 8 0 2 . 8 9 0 2 1 . 2 0 0 S O ooo 0 . 8 8 2 2 6 . 5 6 0 0 . 0 2 1 1 2 1 1 8 . 5 4 0 1 . 1 7 0 0 . 0 3 3 0 3 2 2 14 7 0 0 6 . 9 6 0 1 3 9 0 2 5 3 O 0 3 8 ooo 1 . 0 9 0 17 OOO 0 . 0 2 1 1 3 1 1 B . 2 3 0 1 . 8 7 0 0 . 0 3 5 0 . 4 9 6 15 7 0 0 6 . 4 6 0 1 . 4 8 0 7 6 . 2 0 0 2 4 ooo 1 . 0 4 0 1 9 OOO 0 .0 2 1 1 4 11 8 . BOO 1 5 8 0 0 . 0 5 1 0 . 3 5 6 13 . 3 0 0 6 . 8 7 0 1 . 1 4 0 7 4 . 8 0 0 18 . 0 0 0 1 . 0 8 0 17 . 3 2 0 0 . 0 2 1 1 5 1 1 1 3 . 4 2 0 1 . 9 5 0 0 0 5 6 0 . 5 0 O 15 9 0 0 7 . 1 4 0 1 6 B 0 2 9 OOO 5 8 . 0 0 0 0 . 9 1 0 2 0 O 4 0 0 . 0 2 1 1 6 1 1 1 2 . 2 3 0 1 . 7 8 0 0 . 0 4 7 0 5 8 3 19 . 6 0 0 6 . 3 6 0 2 . 5 2 0 2 8 . 2 0 0 3 8 ooo 0 . 8 8 0 2 3 . 8 5 0 0 . 0 2 1 1 7 11 5 8 1 0 1 7 2 0 0 0 6 4 0 5 1 6 1B . 3 0 0 6 . 5 8 0 1 . 9 6 0 3 5 3 O 0 4 2 . 0 0 0 0 9 O 0 2 3 8 6 0 0 . 0 2 1 1 8 11 2 0 . 7 8 0 2 . 7 9 0 0 . 0 5 8 0 . 8 8 4 3 1 8 0 0 6 . 1 7 0 6 . 0 0 0 3 9 . 0 0 0 S B . 0 0 0 0 . 7 9 0 2 5 . 9 4 0 0 . 0 4 1 1 3 1 2 . 0 1 0 0 . 6 3 0 0 3 9 0 0 7 1 0 14 OOO 4 2 0 0 0 . 9 1 0 3 5 0 0 1 0 0 0 0 0 .0 0 .0 0 0 4 1 2 3 1 1 . 6 3 0 0 . 2 1 0 0 2 0 0 0 . 1 5 0 16 3 0 0 4 . 4 2 0 0 6 8 0 1 8 0 0 2 7 . 0 0 0 0 .0 0 0 0 . 0 4 1 3 3 1 1 . 7 8 0 1 . 0 3 0 0 3 0 0 0 . 3 6 0 13 . 5 0 0 4 . 2 0 0 0 . 8 8 0 1 . 8 0 0 19 ooo 0 . 0 0 . 0 0 . 0 4 1 4 3 1 1 . B 7 0 0 . 6 1 0 0 . 3 4 0 , 0 . 3 0 0 17 . 3 0 0 4 5 0 O 1 . 2 0 0 2 . 2 0 0 19 .ooo 0 . 0 0 . 0 0 . 0 4 1 5 3 1 2 . 2 3 0 0 . 5 5 0 0 . 3 2 0 0 . 1 7 0 15 . 0 0 0 4 6 1 0 0 . 9 6 0 1 7 0 0 2 4 . 0 0 0 0 0 0 0 o . 0 4 1 6 3 1 1 . 2 4 0 0 . 4 0 0 0 3 7 0 0 . 1 8 0 13 . 7 0 0 4 0 8 0 0 . 8 6 0 2 . 2 0 0 19 ooo 0 .0 0 0 0 . 0 4 1 7 3 1 1 . 7 8 0 0 . 3 3 0 0 . 3 2 0 0 . 1 5 0 13 . 7 0 0 4 . 0 4 0 0 . 8 6 0 1 . 8 0 0 2 1 ooo 0 . 0 0 . 0 0 . 0 4 1 8 3 1 6 . 4 2 0 1 . 3 3 0 0 . 8 9 0 0 . 1 3 0 17 . 2 0 0 5 7 2 0 0 . 8 0 0 3 . 8 0 0 1 0 .ooo 0 . 0 0 . 0 0 . 0 4 1 9 3 1 1 . 9 1 0 0 4 9 0 0 . 4 8 0 0 . 1 9 0 ' 16 OOO 4 . 1 3 0 1 . 2 9 0 3 . 2 0 O 2 1 . 0 0 0 0 O 0 . 0 0 . 0 4 1 1 0 3 1 2 . 2 8 0 0 8 9 0 0 5 2 0 0 1 1 0 16 . 4 0 0 4 . 5 4 0 1 . 9 9 0 1 5 0 0 16 . 0 0 0 0 . 0 0 . 0 0 . 0 4 1 1 1 3 1 2 . 2 3 0 0 . 7 2 0 0 4 8 0 0 . 1 4 0 16 OOO 4 4 4 0 1 OOO 0 . 8 0 0 2 1 ooo 0 . 0 0 . 0 0 . 0 4 1 1 3 3 1 2 . 7 3 0 1 . 3 5 0 0 5 2 0 0 . 7 5 0 15 . 9 0 0 4 5 5 0 0 . 9 1 0 4 . 0 0 0 16 ooo 0 . 0 0 . 0 0 0 4 1 1 4 3 1 1 . 8 5 0 1 . 5 2 0 0 . 5 5 0 0 . 2 1 0 16 BOO 4 2 5 0 1 . 0 4 0 4 . 5 0 0 17 ooo 0 . 0 0 . 0 0 . 0 4 1 1 5 3 1 1 . 6 8 0 0 . 4 4 0 0 . 4 0 0 0 . 1 6 0 14 . 3 0 0 4 2 8 0 1 . 0 4 0 3 . 4 0 0 15 ooo 0 0 0 . 0 0 . 0 4 1 1 6 3 1 1 . 4 9 0 1 . 0 9 0 1 1 9 0 0 . 1 1 0 7 5 . 3 0 0 4 . 7 0 0 3 . 7 5 0 5 . 3 0 0 18 ooo 0 . 0 0 . 0 0 . 0 4 1 1 7 3 1 1 . 2 7 0 0 . 3 2 0 0 . 3 3 0 0 . 1 4 0 14 3 O 0 4 0 3 0 1 . 2 1 0 3 . 5 0 0 18 ooo 0 . 0 0 . 0 0 o 128 h-1 00-----000-00--00-0-000-00 — 000-0000 o o o o o o o o o o o < » o o o o o o o o o o o o o o o o o o o o o o o 83§8§S8 5 i ^ 0 0 0 0 0 0 - 0 - 0 0 0 - 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 88§888888888888888888^ sS8S§§lsi§S«5SS8§88SSl §8335§c183SS§5^^ . . . _ _. . .. . _ 5 8 8 8 8 8 8 8 8 8 8 § 8 § 8 ^ ± 8§8§i333S88^^ o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o < b o b o o o o b o o o o b o o b o o b b b o b b b o o o o o o o o o i o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o b b b o o b o b o b o b o o b o o o o b b b o o o o b o o o o o o o o Appendix 2 RANDOM STRATIFIED COMPOSITE SAMPLING DATA 130 ID Ca Mg Na K CFC pH 12 1 12 10 000 12 212 10 370 12 312 10 130 22 132 5 750 22 232 4 970 22 332 5 690 22 122 7 250 22 222 6 580 22 322 6 730 22 1 12 9 010 22 212 12 700 22 312 9 910 32 132 1 510 32 232 1 310 32 332 2 030 32 122 2 380 32 222 1 750 32 322 2 22C 32 1 12 8 950 32 212 6 060 3? 312 6 670 42 132 1 700 42 232 2 1 10 42 332 2 490 42 122 3 530 42 222 4 340 42 322 5 120 42 1 12 8 950 42 212 9 240 42 312 9 720 3 1 132 0 460 31 232 2 490 3 t 332 1 620 31 122 2 050 3 1 222 1 590 31 322 2 '930 3 1 1 12 5 150 31 2 12 6 25G 31 312 7 890 1 1 132 8 2 10 1 1 232 6 130 1 1 332 8 400 1 1 122 8 810 1 1 222 8 240 1 1 322 1 1 850 1 1 112 38 840 1 1 212 39 460 1 1 312 43 290 21 132 2 790 2 1 232 2 850 2 1 332 3 060 2 1 122 3 480 2 1 222 3 240 5 389 0 242 6 314 0 264 3 619 0 155 1 650 0 072 1 770 0 074 1 450 0 074 1 840 0 070 1 900 0 064 1 670 0 073 1 590 0 045 2 020 0 051 1 430 0 055 1 285 0 081 1 958 1 976 1 510 1 156 0 877 O 364 0 102 0 680 1 102 0 479 1 265 0 161 1 081 0 191 1 224 0 199 1 07 3 0 138 1 510 0 170 1 889 0 152 0 973 0 137 1 4 16 0 168 2 024 0 14 1 1 404 0 120 1 518 0 105 1 863 0 078 0 380 0 240 0 910 0 21 1 1 880 1 361 0 870 0 152 1 420 0 8 10 0 820 0 178 o 910 0 102 1 050 0 101 1 440 0 097 3 940 0 390 3 830 0 400 5 130 0 750 2 640 0 220 2 850 0 220 4 200 0 4 10 4 410 0 1 10 3 960 0 1 10 5 250 0 160 1 820 0 090 1 780 o 089 1 920 0 091 2 130 0 099 1 760 0 088 0 834 107 000 0 897 1 19 000 0 973 77 200 0 099 10 200 0 08 3 9 500 0 122 13 500 0 163 13 000 0 202 12 300 0 173 14 200 0 386 17 200 0 575 24 600 0 355 12 300 0 221 20 500 0 3 17 2 1 400 0 231 20 900 0 170 20 900 0 22 1 25 600 0 186 23 300 0 545 2 1 900 0 664 30 200 0 687 25 100 0 2GG 14 900 0 202 19 100 0 232 14 000 0 291 23 700 0 174 19 100 0 282 20 900 0 853 23 700 0 726 2 1 900 1 01 1 20 500 0 266 15 500 0 258 18 000 0 3-10 19 400 0 244 16 100 o 301 17 800 0 356 20 500 0 636 19 900 0 572 2 1 4 00 0 827 2 1 000 0 6-40 30 300 0 520 32 000 0 600 43 400 0 700 40 300 0 5 10 32 900 0 640 49 100 1 230 75 900 1 120 87 900 1 290 94 700 0 12 1 10 200 0 134 1 1 100 o 165 1 1 100 0 146 14 400 0 324 13 300 5 520 25 460 5 320 25 560 5 750 18 070 5 920 0 580 5 940 0 300 5 970 0 550 5 860 0 680 6 010 0 880 5 890 . 1 030 6 230 1 680 6 400 2 820 6 370 2 34 4 030 1 500 3 750 1 770 3 850 1 970 4 350 1 990 4 090 2 150 4 2 10 3 030 5 320 2 850 4 880 3 760 4 900 3 340 3 660 0 970 3 890 1 020 4 080 1 040 4 260 1 280 4 460 1 220 4 7 10 1 260 5 400 2 040 5 760 2 050 5 740 2 010 4 040 1 700 4 290 2 300 4 150 1 600 4 290 1 300 3 910 1 800 4 380 2 300 5 040 3 200 5 060 3 500 5 250 3 200 4 080 5 190 4 OOO 3 670 3 950 6 900 4 170 9 170 4 170 3 820 4 140 8 790 5 350 24 580 5 1 10 29 460 5 050 33 500 6 180 0 320 6 160 0 3 10 6 300 0 530 6 340 0 600 6 130 0 740 27 200 232 000 27 600 224 000 37 700 276 000 18 300 5 000 9 800 4 000 19 000 6 000 17 700 6 000 7 800 7 000 14 700 9 000 5 300 48 000 10 700 48 000 10 300 48 000 10 300 13 000 6 000 16 000 6 500 19 000 1 1 OOO 13 000 7 700 15 000 9 800 19 000 2 1 000 52 000 9 700 70 000 7 200 78 000 5 200 15 000 4 300 17 OOO 4 800 14 000 1 1 500 16 000 12 200 12 000 12 200 12 000 1 1 700 51 000 17 500 7 1 OOO 16 200 58 OOO 7 000 16 000 9 800 14 000 10 300 14 000 7 000 18 000 9 BOO 12 000 10 300 16 000 10 300 6-1 000 10 5O0 84 000 14 900 1C4 000 3 000 28 000 3 700 35 000 2 800 30 000 4 000 4 1 000 2 500 29 000 4 700 50 000 2 1 300 224 000 17 000 2 18 000 24 300 224 000 4 200 4 000 4 OOO 6 000 4 000 0 .0 7 800 5 000 6 300 7 .000 131 I-1 CO I O -j-UMUOUU-'-.-.U UMUMUUUMUUUUU o o o o o o o o o o o o o UO-U'lOUIDUIOJUOUO o o o o o o o o o o o o o o o o o o o o o o o o o o uuuuiuiuiuiinuiOOOO cn~Jcooi-«-jcjoiuit'Oicn-j OOOOOOOOOuimo^  OOOOOOOOOOOOO O^-^cn^-Jcjcj-kOio-J-* OOOOOOOOOCDio — 31 -I OOOOOOOOOOOOO OOOOOOOOOOOOO looouio-'uo&biimo -JOCB-4-glDO~J00OJCD(D l^ OOOOOOOOOOOOO O M KJ o^^u-u-.oofojxn-J 0-J-»oiuicj--iroOOcncnO OOOOOOOOOOOOO M M U U1U1U1UIUIUIUUIMOUIUU OOOOOOOOOOOOO OOOOOOOOOOOOO uOU-«->-'W^-.U6Jk tn&cn-u-b&O-^cncnrorocD OOOOOOOOOOOOO OOOOOOOOOOOOO OOOOOOOOOOOOO o 0) z 01 I z Appendix 3 CONVENTIONAL SAMPLING DATA I D C a Mg Na K C E C p H %C N O 3 - N P 12 23 .79 4 .66 0 . 198 0. 89 89 . 2 5 , .67 22 .83 36 .0 272 22 12 .60 1 . 53 0 .049 0. 38 21 . 2 6 . 85 2 . 10 8 .5 49 32 7 .71 1 . 32 0 . 142 6. .74 25 . 1 5, . 1 1 3 . 12 15 .5 74 42 9 .01 1 .58 0 077 0. 86 24 . 2 5 , 86 1 .93 3 .0 73 31 6 . 76 1 .25 0. , 105 0. 68 22. . 1 5. , 13 3. .0 1 1 .6 84 1 1 43 , . 78 5 .00 0. , 16 1 . 45 82 . 4 5 . 31 30. . 38 1.7 .5 242 21 5. 88 2 .07 0. .060 0. 47 19, .4 6. ,51 2. .'19 29 . 3 36 41 9. .94 1 . . 58 0. . 10 0. 77 14 . 7 5. .95 1 .99 4 .5 288 133 Appendix 4 Mean, standard deviation, and %CV for plow layer samples from three f i e l d s , spring and f a l l data combined. Ca Mg Na K CEC FIELD X sd CV X sd CV x sd CV x sd CV x sd CV RC 8.93 2.00 22 1 .37 0.45 32 0.34 0.76 221 0.79 0.22 28 20.3 2.7 13 HRST 11.6 3.99 34 1 .84 0.45 26 0.06 0.02 39 0.42 0.17 39 18.7 5.3 28 RM 6.68 2.07 31 1 .11 0.25 22 0.15 0.07 46 0.74 0.20 27 23.7 4.4 47 pH %C NO3-N P FIELD X sd CV x sd CV x sd CV X sd CV RC 5.78 0.36 6 2 .11 0.32 15 9.7 6.6 68 109.0 33.7 31 HRST 6.61 0.33 5 2 .02 1.00 50 18.0 12.2 68 49.0 15.3 31 RN 5.02 0.39 8 3 .35 0.50 15 11.3 5.2 46 77.4 30.4 39 134 Appendix 5 Mean, s tandard d e v i a t i o n , and %CV for 60-90 cm samples from three f i e l d s , s p r i n g and f a l l data combined. Ca Mg Na K CEC FIELD X sd CV X sd CV x sd CV x sd CV x sd CV RC 2.24 1.11 50 0.96 0.45 47 0.31 0.24 78 0.21 0.11 51 16.8 2.5 15 HRST 5.22 1.91 37 1.65 0.72 44 0.08 0.03 36 0.11 0.07 67 10.2 3.7 3fe RN 1.33 0.30 22 1.45 0.52 36 0.74 0.68 92/) 0.27 0.07 25 17.7 2.3 13 pH %C NO3-N P FIELD X sd CV X sd CV x sd CV x sd CV RN 4.27 0.47 11 1.11 0.41 37 3.4 1.8 53 16.5 4.0 24 HRST 5.22 1.91 6 0.33 0.21 67 7.4 5.8 78 4.1 1.5 37 RN 3.91 0.24 6 1.63 0.41 25 8.2 4.4 54 16.0 4.3 27 135 Appendix 6 Mean, standard deviation, and %CV for p r o f i l e data. Data taken from three depths, three f i e l d s and two sampling seasons combined. Ca Mg Na K CEC FIELD X sd CV X sd CV x sd CV X sd CV x sd CV RC 5.55 3.21 59 1.14 0.49 43 0.32 0.47 146 0.42 0.30 72 18.7 2.9 16 HRST 7.56 3.95 51 1.80 0.66 37 0.07 0.03 42 0.29 0.18 79 14.3 5.7 40 RN 3.50 2.64 75 1.17 0.42 36 0.41 0.49 120 0.42 0.26 63 20.6 4.3 21 pH %C NO3-N P FIELD X sd CV X sd CV x sd CV X sd CV RC 5.01 0.77 15 1.51 0.56 37 6.6 5.3 80 49.3 47.4 96 | HRST 6.23 0.44 7 1.01 1.00 99 11.6 9.9 85 19.4 22.9 118 | RN 4.40 0.54 12 2.37 0.85 36 9.3 4.8 52 36.3 34.1 94 | 136 Appendix 7 BCMAF POTASSIUM RECOMMENDATIONS Table 2 Area — Vancouver Island, Lower Mainland, Okanagan, Kootenay, Kamloops and Williams Lake (Zones 1, 2, 3, 4, and Subzone 5.01) Recommended Potassium (K 20) Applications for Selected Crops Based on Soil Test Values Potassium (K^O) S o i l Test Value ppm (lb/ac) Rating Potassium (K^O) Required kg/ha (lb/ac) Crop Group 1 Crop Group 2 Crop Group 3 Crop Group 4 25 (50) 35 (70) 50(100) 65 (130) 80 (160) 100 (200) 125 (250) 150 300 175 (350) 175+ (350+) L-L L + M-M M + H-H H + H + 168 (150) 112 (100) 90 (80) 67 (60) 45 (40) 22 (20) 0- 22* (0- 20*) 0 - 22* (0 - 20*) 0- 22* (0- 20*) 0- 22* (0- 20*) 224 (200) 168 (150) 112(100) 90(80) 67 (60) 45 (40) 0- 22* (0- 20*) 0- 22* (0- 20*) 0- 22* (0- 20*) 0- 22* (0- 20*) 224 (200) 224 (200) 168 (150) 112 (100) 90(80) 67 (60) 45 (40) 0- 45* (0-40*) 0-45* (0-40*) 0-45* (0-40*) 280 (250) 280 (250) 280 (250) 224 (200) 168 (150) 112 (100) 90 ( 80) 67 ( 60) 45 (40) 0 - 45* (0 40*) * Starter e f f e c t i n some areas (Neu f e l d , 198 0) Appendix 8 BCMAF PHOSPHORUS RECOMMENDATIONS SOIL T E S T I N T E R P R E T A T I O N S T a b l e 1 Area — Vancouver Island, Lower Mainland, Okanagan, Kootenay, Kamloops and Williams Lake (Zones 1, 2, 3, 4, and Subzone 5.01) Recommended Phosphorus (P 2O s) Applications for Selected Crops Based on Soil Test Values Phosphorus (P) Soil Test Value ppm (lb/ac) Rating Phosphorus (P z0 5) Required kg/ha (lb/ac) Crop Group 1 Crop Group 2 Crop Group 3 Crop Group 4 5 (10) 10(20) 15 (30) 20(40) 30(60) 40 (80) 50(100) 70(140) 70+ (140+) L -L M-M M + H-H H + H + 134 (120) 90(80) 67 (60) 45 (40) 28(25) 0- 17* (0- 15) 0- 17* (0- 15) 0- 17* (0- 15*) 0 - 17* (0 - 15*) 157 (140) 134 (120) 90(80) 67 (60) 45 (40) 28 (25) 0- 22* (0- 20*) 0 - 22* (0 - 20*) 0 - 22* (0 - 20*) 190(170) 157(140) 134 (120) 90 (80) 67 (60) 45 (40) 0 - 28* (0 - 25*) 0- 28* (0 - 25*) 0 - 28* (0 - 25* 224 (200) 190(170) 157 (140) 134 (120) 90 (80) 67 (60) 0- 45* (0-40*) 0 - 45* (0 - 40*) 0 - 45* (0 - 40*) • Starter effect in some areas (Neufeld, 1980) Appendix 9 P R I N C I P A L C O M P O N E N T S : 3 F I E L D S AT O / P L I N S P R I N G T E S T S T A T I S T I C DF S I G N I F N» 51 OUT OF 51 I N D E P E N D E N C E 2 4 8 . 2 1 3 6 0 . E Q U I C O R R E L A T I O N 2 5 7 . 4 4 3 5 0 . ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) COMPONENT 3 . 3 9 4 8 2 . 1 6 3 6 1 . 2 1 7 2 . 8 4 0 9 2 . 5 3 7 0 5 % V A R I A N C E 3 7 . 7 2 61 . 7 6 7 5 . 2 8 8 4 . 6 3 9 0 . 6 0 I N D E P E N D E N C E 1 7 0 . 7 5 1 0 7 . 6 3 71 . 2 5 8 4 2 . 1 7 7 2 3 . 9 8 7 DF 3 5 27 2 0 14 9 S I G N I F 0 . . 0 0 0 0 . 0 0 0 0 . 0 0 0 1 . 0 0 4 3 1 . C A - . 3 4 6 9 6 . 2 6 9 6 1 . 5 2 7 6 1 - . 8 7 4 0 7 - 1 . 1 6 3 2 9 2 . M G - . 4 5 2 9 7 . 1 4 2 0 O . 9 9 1 6 9 - 2 . 4 1 1 7 9 - 1 . 5 7 9 6 3 - 1 3 . N A . 1 6 2 5 2 - . 2 0 4 5 9 . 3 7 2 4 7 . 8 7 0 9 6 - . 1 1 5 8 8 4 . KK . 2 6 6 1 4 . 2 8 7 5 8 . 4 0 5 3 4 - . 2 7 1 7 0 - . 7 5 7 4 7 5 . C E C . 1 8 6 1 9 . 5 5 8 0 8 . 1 8 1 6 1 . 1 4 7 2 1 . 3 1 6 6 3 G . P H - . 4 5 3 1 3 - . 1 8 2 5 4 . 3 5 7 3 8 - . 1 0 7 5 6 . 4 6 1 8 6 - 1 7 . C C . 2 3 5 7 9 . 5 4 3 4 1 - . 1 6 9 0 8 . 8 6 7 8 0 - 1 . 2 2 2 6 2 8 . N 0 3 - . 4 1 9 8 6 . 3 0 7 4 7 - . 8 1 0 1 2 - 1 . 2 1 8 9 5 - . 1 5 5 4 9 9 . P P . 3 2 4 5 2 - . 2 1 9 0 3 . 4 7 1 8 5 - . 2 6 3 7 6 . 4 5 4 6 0 139 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0096036/manifest

Comment

Related Items