New article
The relationship between remotely-sensed spectral heterogeneity and bird diversity is modulated by landscape type. Article written by Dominika Prajzlerová.
Dive into the fascinating world of bird diversity modeling with our latest study, recently published in Elsevier. Our research delves into the intricate relationship between bird species richness and landscape features, aiming to uncover the most effective predictors for conservation efforts.
Traditionally, classified land cover data has been the go-to for explaining bird species richness patterns. However, our study challenges this norm by showcasing the potential of unclassified remote-sensed images, particularly multispectral data from Landsat 8. We investigated whether these unclassified data could outperform traditional methods in modeling bird diversity.
What did we find? Surprisingly, unclassified remote-sensed data, especially spectral heterogeneity metrics, emerged as superior predictors of bird species richness compared to classified land cover data. Our models, enriched with interactions between unclassified data and landscape types, revealed nuanced relationships that vary across different landscapes.
This breakthrough underscores the importance of considering landscape types and spatial scales when assessing bird diversity. By leveraging spectral heterogeneity from unclassified multispectral data, we offer a powerful tool for conservationists to prioritize efforts and protect avian biodiversity across the Czech Republic.