Abstract
Aim:
To assess the soil quality indices and its impact on rice yield in Upper
Brahmaputra Valley Zone of Assam.
Methodology: Seventy-three numbers of geo referenced soil samples
were collected from the rice ecosystems and analysed for twenty-one soil
physical, chemical and biological parameters. The soil quality indices (SQI)
were developed using statistical tools like principal component analysis
(PCA) techniques and expert opinion (EO). Relative soil quality index (RSQI)
was also developed for grouping the soils into categories. Correlation matrices
were drawn between different soil quality indices. The optimum values of soil
quality indices were computed to sustain 80% or more of the existing in field
maximum rice yield (5.20 t ha-1).
Results:
Multivariate statistics showed that four biological parameters viz.,
fluorescein di-acetate activity, phosphate solubilising bacteria, total
bacterial population and collembolan population and three chemical parameters
viz., cation exchange capacity, electrical conductivity? and
diethylene tri amine penta acetic acid-Zinc could explain 70.2% of the
cumulative variance. RSQI demonstrated that >50% and >30% of soils
belonged to medium and good category. The regression of percent relative rice
yield obtained from farmers field, illustrated that soil functions based
EO-SQI could explain high degree of relationship (R2=0.289;
r=0.537*), followed by RSQI (R2=0.284;r=0.532*) and PCA-SQI (R2=0.143;
r=0.378*) to explain the variability of soils. The optimum value indicates
that the rice soils having PCA-SQI value >0.55 were likely to give 80% or
more of the maximum yield of UBVZ of Assam.
Interpretation: Approaches of rating of soil quality
based on PCA-SQI may be a useful tool, and there is need of more extensive
investigations to validate its usefulness for assessment of soil quality in
different cropping sequences of Assam.
Key
words:
Principal component analysis, Rice ecosystem, Soil quality index, Soil
enzymes?
|