Page 130 - AJWEP-22-5

Nkinda, et al.

mixed–use landscapes. Classification accuracy was land use policies are essential. Developing scenario-
affected by cloud cover, atmospheric conditions, and based spatial models that account for multiple variables
seasonal variability, despite efforts to minimize these will enhance
through image preprocessing and ground–truthing.
Misclassifications were particularly common between Acknowledgments
spectrally similar classes, such as sparse vegetation
and shrubland, cropland and barren land, or plantation We are grateful to Joshua Saitoti, Lusekelo Silabi,
forests and orchards. Ground-truthing was also Geofrey Mgini, and local communities living adjacent
constrained by limited spatial coverage and access to to the Ngerengere River whose efforts made this
remote areas, which may have introduced sampling bias. research successful. We thank the Mbeya University
In addition, the study focused mainly on biophysical of Science and Technology in Tanzania of providing
drivers of change and did not fully incorporate complex in-kind support, including working space and internet
socio-economic, institutional, or political factors that access to facilitate this research. Moreover, we would
influence land use decisions. The 2034 projections were like to thank all of our friends who helped us during the
based on historical trends and assumptions, without design and writing of this article by offering insightful
accounting for future uncertainties such as policy and practical suggestions. We also thank them for
shifts, technological developments, or climate change taking the time to polish the manuscript and provide an
impacts – which, although recognized, were treated as incisive critique that greatly improved the work.
a constant rather than modeled directly. Finally, the use
of secondary data for historical validation may have Funding
introduced inconsistencies or reduced the precision of
results. None.

6. Conclusion Conflict of interest

Overall, the study revealed that between 2004 and 2024, The authors declare they have no competing interests.
extensive areas of forested land within the Ngerengere
River catchment were cleared and converted into Author contributions
croplands for cultivating crops such as bananas, cassava,
and maize. This land conversion reflects growing Conceptualization: Rose Nkinda, Fredrick Ojija,
pressure from subsistence farming and population Gisandu Malunguja
expansion. The observed increase in shrubland and Formal analysis: All authors
sparse vegetation, alongside changes in bare land Investigation: Rose Nkinda
and built-up areas, indicates ongoing ecological Methodology: All authors
degradation, habitat fragmentation, and shifting land Supervision: Fredrick Ojija, Giovanni Bacaro, Gisandu
use patterns. The application of GEE enabled high- Malunguja
resolution spatial and temporal analysis, making it an Writing – original draft: Rose Nkinda, Fredrick Ojija
effective tool for monitoring and projecting land cover Writing – review & editing: All authors
changes. These findings highlight the urgent need for
sustainable land management, reforestation efforts, and Availability of data
active community engagement to restore ecological
integrity and support long-term environmental and Data are available from the corresponding author upon
socioeconomic resilience. Despite these valuable reasonable request.
insights, the study had several shortcomings, such as
limited detection of fine-scale changes due to the use References
of medium-resolution imagery and under-exploration
of socioeconomic and institutional drivers. Therefore, 1. Abera W, Tamene L, Kassawmar T, et al. Impacts of land
future research should utilize high-resolution satellite use and land cover dynamics on ecosystem services in
data, integrate socioeconomic and policy analysis, and the Yayo coffee forest biosphere reserve, southwestern
incorporate climate change modeling. In addition, long- Ethiopia. Ecosyst Serv. 2021;50:101338.
term hydrological monitoring and evaluation of current doi: 10.1016/j.ecoser.2021.101338

Volume 22 Issue 5 (2025) 124 doi: 10.36922/AJWEP025180137