ASSESSMENT OF HETEROGENITY IN SOIL PROPERTIES OF NIMBIA FOREST RESERVEIN SOUTHERN GUINEA SAVANNAH OF NIGERIAFOR SITE SPECIFIC MANAGEMENT
ASSESSMENT OF HETEROGENITY IN SOIL PROPERTIES OF NIMBIA FOREST RESERVEIN SOUTHERN GUINEA SAVANNAH OF NIGERIAFOR SITE SPECIFIC MANAGEMENT
No Thumbnail Available
Date
2011-06
Authors
OBIDIKE, EVELYN OMELEBERE
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Spatial analysis of soil heterogeneity is a vital prerequisite for site specific forest
management. This study was aimed at describing the field-scale horizontal spatial
variability and developing models for predicting soil aggregate stability from other soil
propertiesin Nimbia Forest Reserve of the Southern Guinea Savanna zone of Nigeria. Georeferenced
topsoil samples (0-0.15 m) and field infiltration were obtained systematically
from block NF80 along 300 m transect at 3 m interval (n=100). These samples were
analyzed for both physical and chemical properties. Soil data were analyzed using classical
and spatial statistical tools. Results revealed large differences between minimum and
maximum values of the investigated soil properties. The results showed high degree of
variabilityin the majorityof the soil properties with coefficients of variability (CV) ranging
from 0.74% to 90.83%. Using the students’ t-test statistics at 5 % level of probability,
autocorrelation (ACF) and crosscorrelation functions (CCF) were calculated. Calculated
ACF revealed strong spatial dependence between adjacent observations (19 lags) of large
macroaggregate-associated organic carbon (OCa) and no autocorrelation in total iron.
Based on the result of the CCF of wet sieved mean weight diameter (MWDw) versus other
soil properties; correlated soil properties such as clay, total phosphorus (TP), dithionite iron
(Fed), pyrophosphate-extractable iron (Fep), moisture content and aggregate fraction
associated organic carbon (OCa, OCb, OCc, OCd, OCe) were selected and used in various
combinations for multiple linear regression (MLR) and state-space analyses. The results
showed that state-space models described the variation in MWDw better than the
equivalent MLR models. The best model performance explained 99.85 % of the MWDw
variation when clay, Total P, Fed, OCa, OCc, OCd and OCe were used in prediction. The
relationship of clay, Fed and OCa to all the selected best performed models of each
scenarios suggested that these soil properties have effective contribution to the spatial
variation of MWDw. Semivariograms were used to quantify the spatial structure of soil
properties. All properties, except total iron, exhibited a definable spatial structure, which
were described by spherical, exponential, gaussian or linear models. Majority of the soil
properties showed strong spatial dependence. Alternative analytical tools such as the statespace
and semivariogram under forest conditions; considered the underlying processes of
soil properties along transect, identify the local relations between MWDw and selected
properties and; quantify these relationships taking measurement and model errors into
account. Analysis of variation in soil properties is complex. Nevertheless, spatial statistical
tools provided concise information about spatial variation and relationships between soil
properties than the classical statistics tools and; this information is crucial for precision in
site-specific soil management planning in such forest ecosystems.
Description
A THESIS SUBMITTED TO THE POSTGRADUATE SCHOOL,
AHMADU BELLO UNIVERSITY, ZARIA NIGERIA
IN PARTIAL FULFILMENT FOR THE AWARD OF MASTERS IN
SOIL SCIENCE
DEPARTMENT OF SOIL SCIENCE AHMADU BELLO UNIVERSITY,
ZARIA, NIGERIA
JUNE, 2011
Keywords
ASSESSMENT,, HETEROGENITY,, SOIL,, PROPERTIES,, NIMBIA,, FOREST,, RESERVE,, SOUTHERN,, GUINEA,, SAVANNAH,, NIGERIA, SITE,, SPECIFIC,, MANAGEMENT,