A synoptic approach for crop loss assessment used to study wheat. VI. The pathogen data and their relationship to soil and cultural data

1983 ◽  
Vol 34 (2) ◽  
pp. 167 ◽  
Author(s):  
BA Stynes ◽  
LG Veitch

As a part of a general investigation of factors that affect wheat crops on Yorke Peninsula, S. Aust., pathogens were measured at 3 and 8 weeks after emergence of the crop and at anthesis. A first study for regression work showed that, except on the coronal root system at anthesis, the pathogen data were independent of the soil and cultural practice data. Further, it showed that patterning within the pathogens was quite weak and so the value of alternative representations was marginal in this situation. In a second study, significant relationships were found between logarithmically transformed pathogen data, first within the pathogens between sampling times and root systems, and second between the pathogens and the soil and cultural practice data. The former relationships were mostly of the expected type in that high levels of a particular pathogen tended to persist over sampling times and extend over both root systems. The latter relationships were generally weaker, with the two most marked being (i) a tendency in sandy soils for coronal roots at anthesis to have high levels of Rhizoctonia solani, and (ii) for seminal roots at anthesis to have high levels of Heterodera avenae when plants were grown in shallow soils on calcareous rocks, particularly where the farmers applied high levels of superphosphate.

1981 ◽  
Vol 32 (1) ◽  
pp. 1
Author(s):  
BA Stynes ◽  
LG Veitch

From an initial set of 13 measurements taken to define previous land use and the cultural practices adopted by farmers on Yorke Peninsula, S.A., an alternative set of eight variables was derived. The reduced number of variables retained almost all of the information present in the original set and were close to being statistically independent. This simplification is desirable for the later study, when these variables will be considered for inclusion in regression models to predict the growth and yield of wheat. A weak association between these variables and the soil variables specified in Paer III indicated that farmers were not strongly influenced by soil type and properties when deciding which cultural practices to adopt.


1981 ◽  
Vol 32 (1) ◽  
pp. 9
Author(s):  
BA Stynes ◽  
LG Veitch ◽  
HR Wallace

As part of a general study of factors that affect wheat crops on Yorke Peninsula, S.A., growing and mature wheat plants were described in terms of morphological parameters and concentrations of elements in the tops. The measurements were taken 3 and 8 weeks after emergence of the crops, at anthesis and at maturity. The analyses of the morphological variables showed consistent patterns of variation in that, firstly, plants varied widely in their overall size, and secondly, that plants differed in the relative sizes of their tops and roots except at 8 weeks and at anthesis when the seminal root systems did not relate to the remainder of the plant. These two patterns of variation accounted for major portions of the variability, on average 88 % at 3 weeks, 95 % at 8 weeks and 88 % at anthesis, when the seminal root system was separated off in the latter two groups. The seminal root system and the element concentrations at 8 weeks and at anthesis were far less coherent sets of data. The results of these analyses provide the basis of the decisions concerning the information to go into the subsequent regression analyses for growth and yield of wheat.


1981 ◽  
Vol 32 (1) ◽  
pp. 9
Author(s):  
BA Stynes ◽  
LG Veitch ◽  
HR Wallace

As part of a general study of factors that affect wheat crops on Yorke Peninsula, S.A., growing and mature wheat plants were described in terms of morphological parameters and concentrations of elements in the tops. The measurements were taken 3 and 8 weeks after emergence of the crops, at anthesis and at maturity. The analyses of the morphological variables showed consistent patterns of variation in that, firstly, plants varied widely in their overall size, and secondly, that plants differed in the relative sizes of their tops and roots except at 8 weeks and at anthesis when the seminal root systems did not relate to the remainder of the plant. These two patterns of variation accounted for major portions of the variability, on average 88 % at 3 weeks, 95 % at 8 weeks and 88 % at anthesis, when the seminal root system was separated off in the latter two groups. The seminal root system and the element concentrations at 8 weeks and at anthesis were far less coherent sets of data. The results of these analyses provide the basis of the decisions concerning the information to go into the subsequent regression analyses for growth and yield of wheat.


1981 ◽  
Vol 32 (1) ◽  
pp. 1
Author(s):  
BA Stynes ◽  
LG Veitch

From an initial set of 13 measurements taken to define previous land use and the cultural practices adopted by farmers on Yorke Peninsula, S.A., an alternative set of eight variables was derived. The reduced number of variables retained almost all of the information present in the original set and were close to being statistically independent. This simplification is desirable for the later study, when these variables will be considered for inclusion in regression models to predict the growth and yield of wheat. A weak association between these variables and the soil variables specified in Paer III indicated that farmers were not strongly influenced by soil type and properties when deciding which cultural practices to adopt.


Soil Research ◽  
1979 ◽  
Vol 17 (2) ◽  
pp. 227 ◽  
Author(s):  
LG Veitch ◽  
BA Stynes

Study of the relationship between soil properties and traditional soil classifications, using the canonical analysis procedure, showed that both a highly significant and a substantial relationship exists. Classification data on average accounts for about 35% of the total variation of the soil variable data, individual values ranging from about 8% for average log salinity to 74% for the -0.1 bar water content of the soil. The analysis showed that classification data in five soil groups could be substantially retained in two dimensions. The soil variables most involved in the first dimension were per cent clay and the -0.1 bar water content of the soil, and in the second dimension, the available water content of the subsoil (30-100 cm), average pH and average log salinity. These results should be useful in other studies where it is desired to incorporate soil information in a model designed to account for variation in one or more measures of interest.


Sign in / Sign up

Export Citation Format

Share Document