Ecosystem change and respective conservation needs were at top priority areas in the globe especially at biodiversity hotspot zones and places of endemism like Bale eco-region. In this aspect, this study had tried to assess historical landscape changes of afro-alpine forest ecosystem with its respective driving forces and effectiveness of different institutional arrangements to protect this ecosystem. The study utilized satellite imagery from Landsat series starting from 1986 to 2016, ground truth data from GPS-based field survey, focus group discussions, and reconnaissance field survey. Historical landscape maps were prepared with a mix of supervised and unsupervised image classification methods. Future landscape change simulation was undertaken by using cellular automata modeling on Dynamic Environment for Geoprocessing Objects (Dinamica EGO) software. Beyond areal change and transition of landscape elements, level of ecosystem fragmentation was assessed with FRAGSTAT software. Results have shown that there is encroachment of human-induced land use patterns especially agriculture and grazing while closed canopy forest and woodland ecosystems have decreased in their extent. Simulation outputs for the coming three decades have also shown similar patterns with historical trends. Beyond changes in areal extent, landscape matrix analysis results indicated that there is increasing class and landscape level fragmentation both in its historical and future probable likelihoods. Effectiveness of different management strategies at different institutional arrangement were differ in their effectiveness that federal and regional government were not able to curve landscape change. In conclusion, increasing population pressure and weak policy implementation and legal enforcement has caused threats for alpine forest ecosystem in Bale eco-region. Except participatory forest management, management strategies at different institutional arrangements were failed to conserve forest. These calls re-innovated participatory forest management strategies that base community participation while simultaneously keeps livelihoods demands of local community.

Earth-observation-based mapping plays a critical role in humanitarian responses by providing timely and accurate information in inaccessible areas, or in situations where frequent updates and monitoring are required, such as in internally displaced population (IDP)/refugee settlements. Manual information extraction pipelines are slow and resource inefficient. Advances in deep learning, especially convolutional neural networks (CNNs), are providing state-of-the-art possibilities for automation in information extraction. This study investigates a deep convolutional neural network-based Mask R-CNN model for dwelling extractions in IDP/refugee settlements. The study uses a time series of very high-resolution satellite images from WorldView-2 and WorldView-3. The model was trained with transfer learning through domain adaptation from nonremote sensing tasks. The capability of a model trained on historical images to detect dwelling features on completely unseen newly obtained images through temporal transfer was investigated. The results show that transfer learning provides better performance than training the model from scratch, with an MIoU range of 4.5 to 15.3%, and a range of 18.6 to 25.6% for the overall quality of the extracted dwellings, which varied on the bases of the source of the pretrained weight and the input image. Once it was trained on historical images, the model achieved 62.9, 89.3, and 77% for the object-based mean intersection over union (MIoU), completeness, and quality metrics, respectively, on completely unseen images.

Northern Ethiopian Highlands, including Guna-Tana watershed, have experienced profound natural resources degradation which are resulted from coupled natural and anthropogenic factors. To mitigate this problem, Ethiopian government has launched various soil and water conservation programs at different watersheds. Overall objective of this study was to analyze impacts of soil and water conservation programs on vegetation regeneration and ecosystem productivity at in Guna-Tana watershed. As prime data source, the study has utilized Moderate Imaging Spectrometer satellite bi-monthly Enhanced Vegetation Index, 8-day land surface temperature and annual Net Primary Productivity products of the past 17 years starting from 2000. Imagery was processed by using various image preprocessing and analytical techniques. Long-term trend was tested by using Sens slope estimator and Mann–Kendall’s monotonic trend test. Analyzed trend was also segregated into slope and agroecology classes. More importantly, to supplement trend analysis, Vegetation Disturbance Index was developed. Results have showed that despite of long-term soil and water conservation programs, except small patches, vast expanses of the watershed have showed decrease in vegetation regeneration and primary productivity trend. This observed trend has also spatial variability across slope gradient and agroecological classes of the watershed. Though there is tendency of increasing vegetation regeneration and productivity, its observed that significant positive change as a result of watershed conservation programs was very little. This indicates that for better regeneration of vegetation and maintenance of ecosystem health in a watershed, intervention programs should be revised and constraints should be assessed. Taking these into consideration, the study calls further implementation strategies which have accounted agroecology and livelihoods production system.

The prime objective of this study was evaluating six satellite precipitation products by using standard statistical techniques to assess its capability to provide reliable rain rate (amount) and detect rainfall event correctly. High‐resolution precipitation products, Climate Hazards Group InfraRed Precipitation with Station (CHIRPS), Tropical Applications of Meteorology using Satellite data and ground‐based observations (TAMSAT), Tropical Rainfall Measuring Mission (TRMM 3B42RT v7), Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks (PERSIAN CDR), African Rainfall Climatology (ARC v2) and Climate Prediction Center Morphing technique (CMORPH v1.0) were utilized. Evaluation includes both numerical and categorial metrices. Results indicated that in terms of rainfall amount estimation, TAMSAT has relatively better capability whilst for detecting rain event, ARC v2 was found capable in Eastern Ethiopian landscape. All precipitation products underestimate precipitation amount with profound bias level. These needs a thorough bias correction before utilization of these satellite precipitation products.

Refugee-dwelling footprints derived from satellite imagery are beneficial for humanitarian operations. Recently, deep learning approaches have attracted much attention in this domain. However, most refugees are hosted by low- and middle-income countries where accurate label data are often unavailable. The Object-Based Image Analysis (OBIA) approach has been widely applied to this task for humanitarian operations over the last decade. However, the footprints were usually produced urgently, and thus, include delineation errors. Thus far, no research discusses whether these footprints generated by the OBIA approach (OBIA labels) can replace manually annotated labels (Manual labels) for this task. This research compares the performance of OBIA labels and Manual labels under multiple strategies by semantic segmentation. The results reveal that the OBIA labels can produce IoU values greater than 0.5, which can produce applicable results for humanitarian operations. Most falsely predicted pixels source from the boundary of the built-up structures, the occlusion of trees, and the structures with complicated ontology. In addition, we found that using a small number of Manual labels to fine-tune models initially trained with OBIA labels can outperform models trained with purely Manual labels. These findings show high values of the OBIA labels for deep-learning-based refugee-dwelling extraction tasks for future humanitarian operations.