We examined agricultural land use changes and their impacts on the local livelihoods of upstream farmers after the Paunglaung hydropower dam construction in Myanmar. Landsat and GeoEye images, acquired in 2011 and 2018, were used to detect agricultural land use changes. A marked decrease in agricultural land (5.77% of farmland and 9.64% of swidden) was observed, whereas orchards and plantations were introduced as new farming practices. Using the household survey, the income strategies were categorized by different wealth groups, and the socioeconomic condition was compared among the groups. We also compared the annual household income before and after dam construction. A significant change in income strategies after the dam construction were observed in low‐income households, but not in medium and high‐income households. Local people engaged in diversified livelihood strategies after the dam construction, mainly in relocated villages. Besides, farmland lost due to dam construction caused the significant insufficiency of rice self‐consumption and consequently resulted in greater dependency on the market of rice and their products. This study highlighted land use changes due to dam construction and its impacts on the livelihood of local farmers, contributing to planning the land use and livelihood strategies for affected farmers to ensure their livelihood security.
Sensing technologies that are independent of berry-color variation across grape cultivars, particularly under varying environmental stress conditions, are crucial for advancing precision viticulture. This study developed a two-step cross-cultivar quality prediction model for three grape cultivars with red or white berries—Rondo, Zweigelt, and Kerner—using an outdoor hyperspectral camera (HSC). We introduce a seamless soft-clustering approach that groups spectral characteristics without prior cultivar labels, using a coefficient-boosting method integrated with two dimensionality-reduction techniques: principal component analysis (PCA) and uniform manifold approximation and projection (UMAP). For Brix prediction across the three grape cultivars, the R2 values were 0.72 (normal model), 0.75 (PCA-boosted), and 0.83 (UMAP-boosted); for pH prediction, the values were 0.70, 0.75, and 0.82, respectively. UMAP-boosted models consistently outperformed the normal model (without boosting), and PCA-boosted models also improved performance. UMAP-boosted models further outperformed the PCA-boosted model in predicting both Brix and pH, in single-cultivar and cross-cultivar settings, likely because UMAP better captures local and global structure in hyperspectral data across cultivars. Overall, the proposed framework enables automated clustering and improved prediction of major oenological parameters from outdoor hyperspectral data under challenging field conditions, supporting non-destructive quality assessment in real-world vineyards.