Silvopasture is the deliberate integration of a woody component with grazed pastures as understorey. It is one of the most extended agroforestry practices all over the world. Silvopasture use is key to increase the sustainability of livestock farming systems as silvopasture reduces the use of concentrates since the woody component provides feed for animals. However, it is not an extensively used practice in Europe. This paper aims at evaluating, from Eurostat, LUCAS database and the 118 rural development programs, the current situation of permanent grasslands in the Mediterranean area of Europe as well as the rural development programmes fostering silvopasture to better understand how sustainable land use systems are promoted and provide insights to foster silvopasture across Europe. The results of this study show that most of the policy measures related to silvopasture are adapted to the local necessity. The already existing agroforestry managed land (dehesas/montado) are related to measures supporting regeneration and maintenance while in those areas where agroforestry does not exist the measures are related to forest fire prevention.

Pastoral systems based on grazing itineraries are very common along the Mediterranean region and provide an opportunity to manage the fuel load and reduce fire risk in the ecosystem. Daily and seasonal movements of flocks bring on different grazing pressure (GP) (sheep ha⁻¹) over the landscape. This study presents an approach to model sheep GP under a Mediterranean pastoral system in the Northeast of Portugal. The GP in a given location depends on the distance from the stable to the parish border, the distance to the stable, the heads of livestock and their preference for land use and cover (LUC). The geoprocessing we applied in this study integrated several spatial databases: the latest official Portuguese vector mapping of land use and cover (COS2015) and administrative boundaries (CAOP2018), the livestock stables location, and Sentinel-2 Multispectral Images. During the geoprocessing, the stocking density (SD) (sheep ha⁻¹) were calculated and spatially interpolated. Homogeneous LUC units (permanent crops; annual crops; forest; shrubland; grassland; water bodies) were obtained by Random Forest supervised classification algorithm (kappa = 89.3%; global accuracy = 91.2%). Boolean overlapping of the LUC classes obtained by the supervised classifier with the mask created from COS2015 provides the potentially grazed LUC classes. Integrating LUC preferences with SD allows calculating and mapping the GP. The most common GP class is 0–0.25 sheep ha⁻¹. Seeing the GP per LUC class, a value of 1.84 sheep ha⁻¹ was found in permanent crops, 1.73 in annual crops, and 1.25 in grassland, 0.88 in grazed forests and 0.84 in shrublands. The GP modelling and mapping could assist in the implementation of herding programmes aimed at reducing fire hazards at a parish or at a regional scale.

Machine learning models such as XGBoost show strong potential for predicting pasture quality metrics like crude protein (CP) content in tamani grass (Panicum maximum). However, their ‘black box’ nature hinders practical adoption. To address this limitation, this study applied SHapley Additive exPlanations (SHAP) to interpret an XGBoost model and uncover how management practices (grazing interval, nitrogen fertilization, and pre- and post-grazing heights) and environmental factors (precipitation, temperature, and solar radiation) jointly influence CP predictions. Data were divided into 80% for training/validation and 20% for testing. Model performance was assessed with stratified 5-fold cross-validation, and hyperparameters were tuned via grid search. The XGBoost model yielded a Pearson correlation coefficient (r) of 0.78, a mean absolute error (MAE) of 1.45, and a coefficient of determination (R²) of 0.57. The results showed that precipitation in the range of 100–180 mm increased the predicted CP content. Application of 240 kg N ha⁻¹ year⁻¹ positively affected predicted CP, whereas a lower dose of 80 kg N ha⁻¹ year⁻¹ had a negative impact, reducing predicted levels of CP. These findings highlight the importance of integrated management strategies that combine grazing height, nitrogen fertilization, and grazing intervals to optimize crude protein levels in tamani grass pastures.