The objective of this study was to evaluate forest cover change and forest degradation in Nyungwe-Kibira Park, a natural reserve straddling Rwanda and Burundi from 1986 to 2015. Landsat TM, ETM+ and 8OLI images of 30 m spatial resolution were used as primary datasets. Geographic Information System (GIS) techniques were used for forest cover mapping and landscape metrics were calculated by using FRAGSTATS software. Classification and change analysis of forest cover type and landscape patterns analysis were carried out. In addition, to analyze the correlated external disturbances, the buffer zone of 5 Km was delineated outside the boundary of Nyungwe-Kibira Park. The results revealed that in among 5 land cover classes considered within the Park, the dominant one was dense forest class covering over 70% of the entire Park area while in the buffer zone cultivated and open land dominated at over 90% between the years 1986 and 2015. Change detection highlighted that within Nyungwe-Kibira forest, approximately 0.27% (4.97 Km 2) of forest cover was cleared while 0.07% (1.22 Km 2) was regenerated annually. In the buffer zone, the annual cleared forest cover was about 0.76% (13.02 Km 2). The five landscape indices chosen at class level indicated a considerable fragmentation of forest inside the Park and the highest fragmentation in the buffer zone. Indeed, these results shed a bleak image over the future of the Nyungwe-Kibira forest that should be helpful for the policy-makers and managers of these natural parks to establish adequate policies to mitigate the forest loss and degradation by implementing quick and effective solutions.

The leaf area index (LAI), a valuable variable for assessing vine vigor, reflects nutrient concentrations in vineyards and assists in precise management, including fertilization, improving yield, quality, and vineyard uniformity. Although some vegetation indices (VIs) have been successfully used to assess LAI variations, they are unsuitable for vineyards of different types and structures. By calibrating the light extinction coefficient of a digital photography algorithm for proximal LAI measurements, this study aimed to develop VI-LAI models for pergola-trained vineyards based on high-resolution RGB and multispectral images captured by an unmanned aerial vehicle (UAV). The models were developed by comparing five machine learning (ML) methods, and a robust ensemble model was proposed using the five models as base learners. The results showed that the ensemble model outperformed the base models. The highest R2 and lowest RMSE values that were obtained using the best combination of VIs with multispectral data were 0.899 and 0.434, respectively; those obtained using the RGB data were 0.825 and 0.547, respectively. By improving the results by feature selection, ML methods performed better with multispectral data than with RGB images, and better with higher spatial resolution data than with lower resolution data. LAI variations can be monitored efficiently and accurately for large areas of pergola-trained vineyards using this framework.

No abstract