Search

• The relationship between farming systems and vegetation patterns was explored. • Five farm types were identified based on ranching and cropping activities. • Farming systems had little influence on farm-scale vegetation patterns. • Current crop expansion in the studied region mainly affected pastures and not forests. • High resolution land cover maps should be used in future research and policies. In addition to protected areas, sustainable working landscapes are key to successful biodiversity conservation. Yet such landscapes are threatened by rapid crop expansion, in particular in Brazil. In this context, this study explores the influence of farming systems on farm-scale vegetation patterns around the Serra da Bodoquena National Park in Mato Grosso do Sul. To collect data on farming systems and how they are evolving, we conducted interviews at 40 farms covering 120,000 ha, including eight farms with land within the national park. To assess vegetation patterns, we conducted pixel-wise and landscape analyses based on MapBiomas land cover maps from which we calculated seven metrics over the 2009–2019 period. Using multivariate methods, we identified the activities that differentiated farming systems, isolating five farm types with contrasting involvement in crop cultivation and ranching. We found that most farm-scale landscape metrics were only weakly influenced by farming systems. Temporal analyses and interviews suggested that biophysical and legislative contexts limit crop expansion, which mainly occurred at the expense of old pastures and did not directly impact forest proportion within farms. As a consequence, crop expansion in the region seemed to mainly affect small tree patches in pasture areas, making its effect on vegetation patterns barely detectable with 30-m resolution imagery. These findings suggest that rather than focusing solely on deforestation, monitoring the dynamics of wooded pastures with high-resolution images is crucial to assess the early effects of crop expansion on vegetation patterns and to ensure the conservation of biodiversity-friendly agricultural matrices around protected areas in Brazil. [ABSTRACT FROM AUTHOR]

We determined values of antioxidant biomarkers catalase (CAT), superoxide dismutase (SOD), glutathione peroxidase (GPx), glutathione S-transferase (GST) and malondialdehyde (MDA) in two bivalve species from a subtropical system in Brazil. Differences in the biomarkers activities measured in digestive glands and gills sampled at two seasons and in sites with different levels of contamination were tested using univariate and multivariate analyses. Seasonality was the main factor accounting for overall variability. Significant SOD and GPx activities and increased levels of lipid peroxides were detected in summer, while CAT activity was higher in winter. However, biomarkers respond to complex environmental settings as indicated by the significant interaction between season and contamination level. We propose A. flexuosa as a candidate for monitoring studies with sampling occurring at a yearly timescale, and using C. rhizophorae is better suited for designs that include both seasons. • Baseline levels for antioxidant enzymes are provided fortwotropical and subtropical bivalve species. • Seasonal and spatial variation in the biomarker response wasdescribed in bivalves'gills and digestive glands. • Multivariate approaches, such as redundancy analysis (RDA), are suggested for multi-species and multi-biomarker studies. • The multivariate antioxidant response was species-specific and strongly affected by seasonality. • A. flexuosa is an appropriate candidate for monitoring studies with sampling occurring at a yearly timescale. • C. rhizophorae is better suited for designs that include both seasons. [ABSTRACT FROM AUTHOR]