Premium
Small Grain Response to Potassium Fertilizers as Related to Soil and Site Characteristics
Author(s) -
Veeh R. H.,
Skogley E. O.
Publication year - 1986
Publication title -
soil science society of america journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.836
H-Index - 168
eISSN - 1435-0661
pISSN - 0361-5995
DOI - 10.2136/sssaj1986.03615995005000020024x
Subject(s) - variables , fertilizer , mathematics , linear regression , regression analysis , statistics , yield (engineering) , soil fertility , soil test , grain yield , interaction , agronomy , soil science , soil water , environmental science , biology , materials science , metallurgy
The major objective of this study was to determine what environmental characteristics best related to increased small grain yield when fertilizer K was included in a balanced soil fertility management program. These results would help explain why contemporary soil test approaches for predicting crop response to applied K fertilizers do not work in this region, and help lead to alternative approaches to resolve this problem. Two hundred twenty‐two small grain experiments established over a 13 yr period on 127 site locations throughout the Great Plains of Montana were selected for this study. Two to five rates of K, ranging from 0 to 135 kg K/ha, had been applied in the field experiments. Both KCl and K 2 SO 4 were included as fertilizer sources in most experiments. Adequate rates of N and P were constant within an experiment, but varied from one experiment to another. Originally a total of 47 independent variables in various combinations were used in 45 multiple linear regression analyses. The dependent variables were (i) maximum, and (ii) average percent grain yield response to applied K. The statistical analysis was confounded considerably because of missing data for certain independent variables at some sites and because of intercorrelations between variables; however, significant relationships (simple r ) existed in many of the regression analyses between independent variables and the dependent variables. In the final analyses, 32 sites were examined using 27 independent variables with a universal data base. Variables with the highest correlation to percent yield response (in decreasing order) were: (i) dry consistence of the B k horizon; (ii) textural family; (iii) average soil temperature in May; (iv) growing season rainfall; and (v) stored spring soil moisture from 0 to 90 cm. Together these five variables produced a significant R 2 of 0.34 (P=0.05).