Land Cover Classification in Complex and Fragmented Agricultural Landscapes of the Ethiopian Highlands

Living with landslides: Land use on unstable hillslopes in a rural tropical mountainous environment in DR Congo

Landslides are processes that naturally occur on numerous hillslopes across the world. In inhabited regions, landslides are commonly seen as a threat and a land degradation process. Yet, in densely-populated rural mountainous regions in the tropics, local communities have often no choice but to live on steep terrains naturally impacted by landslides. Besides, landslides may also be a source of opportunities for these communities. However, little is known on the rationale underlying land use in landslides. The aim of this study was to assess the extent, modes of valorization and degree of satisfaction of famers exploiting landslides in a populated rural mountainous environment of DR Congo (territory of Kalehe). We interviewed 82 farmers living on 57 representative landslides, these mass movements having been selected according to their characteristics (size, type) and position along the hillslopes and taking into account accessibility or safety constraints. We show that almost all landslides are being exploited by farmers and that they adapt their land use to the type of landslide. Indeed, significant contrasts are observed between landslides and the surrounding hillslopes for subsistence crops, forests, eucalyptus plantations and pasture. Farmers also adapt land use according to local variations in slope or wetness within a given landslide. Nearly half of the farmers reported that their land was more valuable inside than outside landslides. Better soil fertility, higher soil moisture, lower sand or stone content, lower slopes are some of the main factors that increase the land value, offering more favorable conditions for cropping than on land outside landslides. Despite the perceived risk of landsliding, famers settlement on unstable slopes appears justified by the immediacy of the benefits that outweigh the potential dangers. Better understanding the reasons for the settlement of populations on unstable slopes may help devise better risk reduction strategies.

Monitoring Urban Deprived Areas with Remote Sensing and Machine Learning in Case of Disaster Recovery

Rapid urbanization and increasing population in cities with a large portion of them settled in deprived neighborhoods, mostly defined as slum areas, have escalated inequality and vulnerability to natural disasters. As a result, monitoring such areas is essential to provide information and support decision-makers and urban planners, especially in case of disaster recovery. Here, we developed an approach to monitor the urban deprived areas over a four-year period after super Typhoon Haiyan, which struck Tacloban city, in the Philippines, in 2013, using high-resolution satellite images and machine learning methods. A Support Vector Machine classification method supported by a local binary patterns feature extraction model was initially performed to detect slum areas in the pre-disaster, just after/event, and post-disaster images. Afterward, a dense conditional random fields model was employed to produce the final slum areas maps. The developed method detected slum areas with accuracies over 83%. We produced the damage and recovery maps based on change analysis over the detected slum areas. The results revealed that most of the slum areas were reconstructed 4 years after Typhoon Haiyan, and thus, the city returned to the pre-existing vulnerability level.

Sentinel-1 and Sentinel-2 Data for Savannah Land Cover Mapping: Optimising the Combination of Sensors and Seasons

Savannahs are heterogeneous environments with an important role in supporting biodiversity and providing essential ecosystem services. Due to extensive land use/cover changes and subsequent land degradation, the provision of ecosystems services from savannahs has increasingly declined over recent years. Mapping the extent and the composition of savannah environments is challenging but essential in order to improve monitoring capabilities, prevent biodiversity loss and ensure the provision of ecosystem services. Here, we tested combinations of Sentinel-1 and Sentinel-2 data from three different seasons to optimise land cover mapping, focusing in the Ngorongoro Conservation Area (NCA) in Tanzania. The NCA has a bimodal rainfall pattern and is composed of a combination savannah and woodland landscapes. The best performing model achieved an overall accuracy of 86.3 ± 1.5% and included a combination of Sentinel-1 and 2 from the dry and short-dry seasons. Our results show that the optical models outperform their radar counterparts, the combination of multisensor data improves the overall accuracy in all scenarios and this is particularly advantageous in single-season models. Regarding the effect of season, models that included the short-dry season outperform the dry and wet season models, as this season is able to provide cloud free data and is wet enough to allow for the distinction between woody and herbaceous vegetation. Additionally, the combination of more than one season is beneficial for the classification, specifically if it includes the dry or the short-dry season. Combining several seasons is, overall, more beneficial for single-sensor data; however, the accuracies varied with land cover. In summary, the combination of several seasons and sensors provides a more accurate classification, but the target vegetation types should be taken into consideration.

Tracking changes in soil organic carbon across the heterogeneous agricultural landscape of the Lower Fraser Valley of British Columbia

Increasing soil organic carbon (SOC) can improve the capacity of agricultural systems to both adapt to and mitigate climate change. Despite its importance, the current understanding of the magnitude or even the direction of SOC change in agricultural landscapes is limited. While changes in land use/land cover (LULC) and climate are among the main drivers of changes in SOC, their relative importance for the spatiotemporal assessment of SOC is unclear. This study evaluated LULC and SOC dynamics using archived and recent soil samples, remote sensing, and digital soil mapping in the Lower Fraser Valley of British Columbia, Canada. We combined both pixel- and object-based analysis of Landsat satellite imagery to assess LULC changes from 1984 to 2018. We achieved an overall accuracy of 81% and kappa coefficient of 0.77 for LULC classification using a random forest model. For predicting SOC for the same time period, we applied soil and vegetation indices derived from Landsat images, topographic indices, historic soil survey variables, and climate data in a random forest model. The SOC prediction of 2018 resulted in a coefficient of determination (R²) of 0.67, concordance correlation coefficient (CCC) of 0.76, and normalized root mean square error (nRMSE) of 0.12. For 1984, the SOC prediction accuracies were 0.46, 0.58, and 0.18 for R², CCC, and nRMSE, respectively. We detected SOC loss in 61%, gain in 12%, while 27% remained unchanged across the study area. Although we detected large losses of SOC due to LULC change, the majority of the SOC losses across the landscape were attributed to areas that were remained in the same type of agricultural production since 1984. Climate variability did not, however, have a strong effect on SOC changes. These results can inform decision making in the study area to support sustainable LULC management for enhancing SOC sequestration.

Assessing the Accuracy of Multiple Classification Algorithms for Crop Classification Using Landsat-8 and Sentinel-2 Data

The launch of Sentinel-2A and B satellites has boosted the development of many applications that could benefit from the fine resolution of the supplied information, both in time and in space. Crop classification is a necessary task for efficient land management. We evaluated the benefits of combining Landsat-8 and Sentinel-2A information for irrigated crop classification. We also assessed the robustness and efficiency of 22 nonparametric classification algorithms for classifying irrigated crops in a semiarid region in the southeast of Spain. A parcel-based approach was proposed calculating the mean normalized difference vegetation index (NDVI) of each plot and the standard deviation to generate a calibration-testing set of data. More than 2000 visited plots for 12 different crops along the study site were utilized as ground truth. Ensemble classifiers were the most robust algorithms but not the most efficient because of their low prediction rate. Nearest neighbor methods and support vector machines have the best balance between robustness and efficiency as methods for classification. Although the F1 score is close to 90%, some misclassifications were found for spring crops (e.g., barley, wheat and peas). However, crops with quite similar cycles could be differentiated, such as purple garlic and white garlic, showing the powerfulness of the developed tool.

Spatiotemporal Analysis of Land Cover Changes in the Chemoga Basin, Ethiopia, Using Landsat and Google Earth Images

Land cover change is a major environmental concern in the northwestern highlands of Ethiopia. This study detected land cover transitions over the past 30 years in the Chemoga basin (total area = 118,359 ha). Land cover maps were generated via the supervised classification of Landsat images with the help of the Google Earth (GE) images. A total of 218 unchanged land features sampled from GE images were used as the training datasets. Classification accuracy was evaluated by comparing classified images with 165 field observations during the 2017 field visit. The overall accuracy was 85.4% and the kappa statistic was 0.81, implying that the land classification was satisfactory. Agricultural land is the dominant land cover in the study basin, and increased in extent by 2,337 ha from 1987 to 2017. The second and third most dominant land cover types, grassland and woodland, decreased by 1.9% and 3.6%, respectively, over the past 30 years. The increase in agricultural lands was mostly due to the conversion of grasslands and woodlands, although some agricultural lands changed to Eucalyptus plantations and human settlements. The results revealed that the expansion of built-up space and agricultural lands was the major driver of fragmentation of the landscape, and degradation of natural resources in the Chemoga basin, Ethiopia.

Post-Disaster Recovery Assessment with Machine Learning-Derived Land Cover and Land Use Information

Post-disaster recovery (PDR) is a complex, long-lasting, resource intensive, and poorly understood process. PDR goes beyond physical reconstruction (physical recovery) and includes relevant processes such as economic and social (functional recovery) processes. Knowing the size and location of the places that positively or negatively recovered is important to effectively support policymakers to help readjust planning and resource allocation to rebuild better. Disasters and the subsequent recovery are mainly expressed through unique land cover and land use changes (LCLUCs). Although LCLUCs have been widely studied in remote sensing, their value for recovery assessment has not yet been explored, which is the focus of this paper. An RS-based methodology was created for PDR assessment based on multi-temporal, very high-resolution satellite images. Different trajectories of change were analyzed and evaluated, i.e., transition patterns (TPs) that signal positive or negative recovery. Experimental analysis was carried out on three WorldView-2 images acquired over Tacloban city, Philippines, which was heavily affected by Typhoon Haiyan in 2013. Support vector machine, a robust machine learning algorithm, was employed with texture features extracted from the grey level co-occurrence matrix and local binary patterns. Although classification results for the images before and four years after the typhoon show high accuracy, substantial uncertainties mark the results for the immediate post-event image. All land cover (LC) and land use (LU) classified maps were stacked, and only changes related to TPs were extracted. The final products are LC and LU recovery maps that quantify the PDR process at the pixel level. It was found that physical and functional recovery can be mainly explained through the LCLUC information. In addition, LC and LU-based recovery maps support a general and a detailed recovery understanding, respectively. It is therefore suggested to use the LC and LU-based recovery maps to monitor and support the short and the long-term recovery, respectively.

Dynamic Land Cover Mapping of Urbanized Cities with Landsat 8 Multi-temporal Images: Comparative Evaluation of Classification Algorithms and Dimension Reduction Methods

Uncontrolled and continuous urbanization is an important problem in the metropolitan cities of developing countries. Urbanization progress that occurs due to population expansion and migration results in important changes in the land cover characteristics of a city. These changes mostly affect natural habitats and the ecosystem in a negative manner. Hence, urbanization-related changes should be monitored regularly, and land cover maps should be updated to reflect the current situation. This research presents a comparative evaluation of two classification algorithms, pixel-based support vector machine (SVM) classification and decision-tree-oriented geographic object-based image analysis (GEOBIA) classification, in producing a dynamic land cover map of the Istanbul metropolitan city in Turkey between 2013 and 2017 using Landsat 8 Operational Land Imager (OLI) multi-temporal satellite images. Additionally, the efficiencies of the two data dimension reduction methods are evaluated as part of this research. For dimension reduction, built-up index (BUI) and principal component analysis (PCA) data were calculated for five images during the mentioned period, and the classification algorithms were applied on data stacks for each dimension reduction method. The classification results indicate that the GEOBIA classification of the BUI data set provided the highest accuracy, with a 91.60% overall accuracy and 0.91 kappa value. This combination was followed by the GEOBIA classification of the PCA data set, which highlights the overall efficiency of the GEOBIA over the SVM method. On the other hand, the BUI data set provided more reliable and consistent results for urban expansion classes due to representing physical responses of the surface when compared to the data set of the PCA, which is a spectral transformation method.

Landscape research in Ethiopia: misunderstood or lost synergy?

A full understanding of the concept of landscape plays a paramount role in sustainable management of natural resources and an increase of landscape studies. However, little is known about the concept of landscape, landscape research and its application in Ethiopia. Hence, the overall objective of this paper is to explore the concept of landscape and review available literatures on landscape research in Ethiopia and to identify research gaps. A questionnaire (n=30) was administered to explore the concept of landscape. A systematic review of available studies on landscape and related concepts has also been made. Out of the 398 papers in which the terms ‘landscape’ and ‘Ethiopia’ appeared in the title, keywords or abstract, 26 papers, having 10 or more keywords related to landscape research were included in this in-depth review. An exploratory study of art and media has been made to examine the perception of artists on landscapes. The results of the study show that the perception of Ethiopian artists on landscape is highly associated with concept of the landscape. The findings of the survey also reveal that the meaning of the term landscape differs semantically. The findings of the review also indicate that landscape studies carried out in Ethiopia do not fully cover the holistic concept of landscape; as they mostly focus more on physical features of the landscape. Moreover, the interdisciplinary approach that integrates landscape ecology, perception and history, which is important for understanding landscapes and landscape changes, is also lacking. Generally, the concept of landscape seems to be misconceived in most studies undertaken in Ethiopia, mainly because it is interchangeably used with land use and land cover. Hence, there is a need for a better understanding of the concept of landscape and the applications of a holistic landscape approach.

National-scale cropland mapping based on spectral-temporal features and outdated land cover information

The lack of sufficient ground truth data has always constrained supervised learning, thereby hindering the generation of up-to-date satellite-derived thematic maps. This is all the more true for those applications requiring frequent updates over large areas such as cropland mapping. Therefore, we present a method enabling the automated production of spatially consistent cropland maps at the national scale, based on spectral-temporal features and outdated land cover information. Following an unsupervised approach, this method extracts reliable calibration pixels based on their labels in the outdated map and their spectral signatures. To ensure spatial consistency and coherence in the map, we first propose to generate seamless input images by normalizing the time series and deriving spectral-temporal features that target salient cropland characteristics. Second, we reduce the spatial variability of the class signatures by stratifying the country and by classifying each stratum independently. Finally, we remove speckle with a weighted majority filter accounting for per-pixel classification confidence. Capitalizing on a wall-to-wall validation data set, the method was tested in South Africa using a 16-year old land cover map and multi-sensor Landsat time series. The overall accuracy of the resulting cropland map reached 92%. A spatially explicit validation revealed large variations across the country and suggests that intensive grain-growing areas were better characterized than smallholder farming systems. Informative features in the classification process vary from one stratum to another but features targeting the minimum of vegetation as well as short-wave infrared features were consistently important throughout the country. Overall, the approach showed potential for routinely delivering consistent cropland maps over large areas as required for operational crop monitoring.