PhD Thesis
The study was carried out in Itigi thicket in Manyoni District to assess land cover
dynamics and estimate biomass and volume of thicket and tree (associate trees) species.
The assessment of land cover dynamics was based on data extracted from remote sensing
using the 1991, 2000 and 2011 satellite images and key informants interviews.
The estimation of biomass and volume of thicket and tree species, data from destructive
and non-destructive sampling was used. Sixty thicket clumps and thirty trees were
sampled for destructive sampling. The study covered two dominant thicket species:
Combretum celastroides Laws and Pseudoprosopsis fischeri (Tab) Harms and five
dominant tree species: Canthium burtii Bullock sensu R. B. Drumm, Cassipourea mollis
(R. E. Fr.) Alston, Haplocoelum foliolosum L, Lannea fulva (Engl.) England and
Vangueria madagascariensis J. F. Gmelin. All sampled thicket clumps and trees were
destructively for above- (AGB), belowground (BGB) biomass and volume. Analysis of
land cover dynamics was based on supervised image classification using maximum
likelihood classifier (MLC). For modelling biomass and volume of individual thicket
clump and tree, different nonlinear multiplicative model forms were tested. The final
models of biomass were selected based on Akaike Information Criterion (AIC) while the
final models of volume were selected based on coefficient of determination (R2) and
relative root mean square error (RMSEr). Then the final selected models were applied to
estimate biomass and volume of non-destructive sampling data. The results showed that
thicket occupies large area in Itigi thicket. For example in 1991, the area occupied by
thicket was 345 150.5 ha (67.85 %), in 2000 was 313 451 ha (61.62 %) and in 2011 was
293 444.8 ha (57.7 %). Apparently, this large area occupied by thicket declined during
1991 - 2000 and 2000 - 2011. The decline in thicket areas was attributed to increase in
anthropogenic activities such as wood extraction, clearing for agriculture, livestock grazing and fires. The model fitting showed that, large parts of the variation in biomass of
thicket clumps were explained by basal area weighed mean diameter at breast height
(dbhw) of stems in the clump and number of stems in the clump (stem count, i.e. st), i.e.
for AGB and BGB of C. celastroides Laws up to 89 % and 82 % respectively and for
AGB and BGB of P. fischeri (Tab) Harms up to 96 % and 95 % respectively. For tree
species most variation was explained by diameter at breast height (dbh) alone, i.e. up to
85 % and 69 % for ABG and BGB respectively. It was also noted that, for thicket, large
parts of the variation in volume of thicket clumps were explained by dbhw, height (ht) and
stem count, i.e. for C. celastroides Laws up to 69 % (R2) and for P. fischeri (Tab) Harms
up to 93 %. For trees most variation was explained by dbh and ht, i.e. up to 93 %.
Although there might be some uncertainties related to biomass and volume estimates for
large areas, for practical reasons, it is recommended the selected models to be applied to the entire area where Itigi thicket extends outside the study site, and also to those thicket
and tree species present that were not included in the data used for modelling. The
methods used in this study to assess land cover dynamics, biomass and volume stocks of thicket and tree species highlight the importance of integrating remote sensing and forest inventory in understanding the thicket resources dynamics and generating information that could be used to overcome the Itigi thicket problems for the sustainability of this unique vegetation.
Climate Change Impact, Adaptation and Mitigation
programme (CCIAM)