Abstract:
Kenya, facing a forest cover of only 7% compared to the recommended 10% for
sustainable development, contends with severe deforestation attributed to forest fires,
unsustainable human activities, and settlement expansion. Afforestation and
reforestation initiatives have been implemented by organizations and government
agencies to combat deforestation, enhance soil health, and sequester carbon. Kieni
forest encapsulates this crisis, exemplifying rampant deforestation's adverse impact on
ecosystems and local well-being. The ensuing extended droughts and unpredicted
rainfall patterns have led to depleted water sources, affecting agriculture and daily
sustenance. Notably, the decline in the African elephant population underscores the loss
of biodiversity. This research focuses on the utilization of a novel methodology that
leverages remote sensing data from Unmanned Aerial Vehicles (UAVs), in combination
with ground-based measurements. The primary objectives of this study include
estimating tree height using UAVderived Digital Terrain Model and Digital Surface
Model data, as well as tree species identification from UAV-derived images. The
methodology entails the derivation of Digital Terrain Model and Digital Surface Model
from aerial images. The estimation of tree height was done by the creation of a Canopy
Height Model (CHM) through the subtraction of DTM from DSM. To ensure the
precision of the CHM model, it was subjected to validation by comparing it against the
field measurements obtained during data collection. The accuracy of the estimated tree
heights was 0.9979 for R
and RMSE at 0.9748. This study underscores the effectiveness
of UAV-derived data in accurately estimating tree height. The results demonstrate the
potential of UAV technology to revolutionize forest monitoring and management,
offering a cost-effective and efficient solution for tree height estimation. The findings
presented in this abstract contribute to the growing body of knowledge supporting the
integration of UAVs in forestry and environmental research, paving the way for
improved resource management and ecological studies.