No abstract

Effect of pan-sharpening multi-temporal Landsat 8 imagery for crop type differentiation using different classification techniques This study evaluates the potential of pan-sharpening multi-temporal Landsat 8 imagery for the differentiation of crops in a Mediterranean climate. Five Landsat 8 images covering the phenological stages of seven major crops types in the study area (Cape Winelands, South Africa) were acquired. A statistical pansharpening algorithm was used to increase the spatial resolution of the 30m multispectral bands to 15m. The pan-sharpened images and original multispectral bands were used to generate two sets of input features at 30 and 15 metre resolutions respectively. The two sets of spatial variables were separately used as input to decision trees (DTs), k-nearest neighbour (k-NN), support vector machine (SVM), and random forests (RT) machine learning classifiers. The analyses were carried out in both the object-based image analysis (OBIA) and pixel-based image analysis (PBIA) paradigms. For the OBIA experiments, three image segmentation scenarios were tested (good, over and under segmentation). The PBIA experiments were carried out at 30m and 15m resolutions. The results show that pan-sharpening led to dramatic (~15%) improvements in classification accuracies in both the PBIA and OBIA approaches. Compared to the other classifiers, SVM consistently produced superior results. When applied to the pansharpened imagery SVM produced an overall accuracy of nearly 96% using OBIA, while PBIA's overall accuracy was 1.63% lower. We conclude that pansharpening Landsat 8 imagery is highly beneficial for classifying agricultural fields whether an objector pixel-based approach is used.

No abstract

Wetlands provide vital ecosystem services such as water purification, flood control, and climate moderation among others, which enhance environmental quality, promote public health, and contribute to risk reduction. The biggest threat to wetlands is posed by human activities which transform wetlands, often for short-term consumptive benefits. This paper aimed to classify and map recent land cover and provide a multi-temporal analysis of changes from 2002 to 2014 in the Nakivubo wetland through which wastewater from Kampala city drains to Lake Victoria in Uganda. The paper contributes through spatially congruent change maps showing site-specific land cover conversions. In addition, it gives insight into what happened to the wetlands, why it happened, how the changes in the wetlands affect the communities living in them, and how the situation could be better managed or regulated in future. The analysis is based on very high resolution (50-62 cm) aerial photos and satellite imagery, focus group discussions, and key informant interviews. Overall, the analysis of losses and gains showed a 62 % loss of wetland vegetation between 2002 and 2014, mostly attributable to crop cultivation. Cultivation in the wetland buffering the lake shore makes it unstable to anchor. The 2014 data shows large portions of the wetland calved away by receding lake waves. With barely no wetland vegetation buffer around the lake, the heavily polluted wastewater streams will lower the quality of lake water. Furthermore, with increased human activities in the wetland, exposure to flooding and pollution will be likely to have a greater impact on the health and livelihoods of vulnerable communities. This calls for a multi-faceted approach, coordination of the various stakeholders and engagement of wetland-dependent communities as part of the solution, and might require zoning out the wetland and restricting certain activities to specific zones.

No abstract