Fanjinshan National Nature Reserve (FNNR) is a biodiversity hotspot in China that is part of a larger, multi-use landscape where farming, grazing, tourism, and other human activities occur. The steep terrain and persistent cloud cover pose challenges to robust vegetation and land use mapping. Our objective is to develop satellite image classification techniques that can reliably map forest cover and land use while minimizing the cloud and terrain issues, and provide the basis for long-term monitoring. Multi-seasonal Landsat image composites and elevation ancillary layers effectively minimize the persistent cloud cover and terrain issues. Spectral vegetation index (SVI) products and shade/illumination normalization approaches yield significantly higher mapping accuracies, compared to non-normalized spectral bands. Advanced machine learning image classification routines are implemented through the cloud-based Google Earth Engine platform. Optimal classifier parameters (e.g., number of trees and number of features for random forest classifiers) were achieved by using tuning techniques. Accuracy assessment results indicate consistent and effective overall classification (i.e., above 70% mapping accuracies) can be achieved using multi-temporal SVI composites with simple illumination normalization and elevation ancillary data, despite the fact limited training and reference data are available. This efficient and open-access image analysis workflow provides a reliable methodology to remotely monitor forest cover and land use in FNNR and other mountainous forested, cloud prevalent areas.
With rapid urbanization worldwide, anthropogenic impacts such as human settlements and invasive carnivores (dogs Canis familiaris, cats Felis catus) are altering spatial distributions and temporal activity patterns of native species. In this study, we focused on spatiotemporal responses of native mammals to anthropogenic impacts in a protected area surrounded by a large metropolis (i.e. Yangmingshan National Park inside Taipei-Keelung metropolis in northern Taiwan). We collected site use data of 11 mammal species (i.e. dogs, cats, nine native species) between 2012 and 2017 with a camera system comprising 121 camera sites. We quantified anthropogenic disturbances as distance to human settlements and activity levels of free-roaming dogs and cats. Species richness and occurrences of the native mammals increased with increasing distances to human settlements and decreasing activity level of dogs, with the latter having a stronger effect than the former. Diel activity overlap between native mammals and dogs was lower during April–July season, coinciding with the breeding season for several native mammals. In contrast, activity level of cats showed no relationships with species richness, occurrences or diel activities of the native mammals. This study demonstrated negative impacts of human settlements and free-roaming dogs on native mammal communities for protected areas in urban environments, and highlights dog activity as a major anthropogenic threat to wildlife.
Globally, biodiversity has declined at an unprecedented rate, challenging the viability of ecosystems, species, and ecological functions and their corresponding services. Payments for ecosystem services (PES) programs have been established and implemented worldwide to combat the degradation or loss of essential ecosystems and ecosystem services without sacrificing the well-being of people. With an overarching goal of reducing soil erosion, China’s Grain-to-Green program (GTGP) converts cropland to forest or grassland. As one of the largest PES programs in the world, GTGP has great potential to offer biodiversity conservation co-benefits. To consider how GTGP may influence biodiversity, we measured forest structure and plant and wildlife species diversity at both GTGP forest and natural forest sites in Fangjingshan National Nature Reserve, China. We also evaluated the relationship between canopy cover and biodiversity measures to test whether forest cover, the most commonly measured and reported ecological metric of PES programs, might act as a good proxy for other biodiversity related parameters. We found that forest cover and species diversity increased after GTGP implementation as understory and overstory plant cover, and understory and midstory plant diversity at GTGP sites were similar to natural forest. Our results suggest that GTGP may also have been associated with increased habitat for protected and vulnerable wildlife species including Elliot’s pheasant (Syrmaticus ellioti), hog badger (Arctonyx collaris), and wild boar (Sus scrofa). Nevertheless, we identified key differences between GTGP forest and natural forest, particularly variation in forest types and heterogeneity of overstory vegetation. As a result, plant overstory diversity and wildlife species richness at GTGP forest were significantly lower than at natural forest. Our findings suggest, while forest cover may be a good proxy for some metrics of forest structure, it does not serve as a robust proxy for many biodiversity parameters. These findings highlight the need for and importance of robust and representative indicators or proxy variables for measuring ecological effects of PES programs on compositional and structural diversity. We demonstrate that PES may lead to biodiversity co-benefits, but changes in program implementation could improve the return on investment of PES programs to support conservation of biodiversity.