High nature value (HNV) grassland systems are increasingly important for the ecosystem services they provide and for their socio-economic impact in the current constant-changing context. The aim of our paper is to evaluate the long-term effect of organic fertilizers on HNV systems in the Apuseni Mountains, Romania. As an objective we want to identify the optimal intensity of conservation management and its recognition based on indicator value plant species. The experiments were established in 2001 on the boreal floor and analyze the effect of a gradient of four organic treatments with manure. Fertilization with 10 t ha⁻¹ manure ensures an increase in yield and has a small influence on diversity, and could be a real possibility for the maintenance and sustainable use of HNV. Each fertilization treatment determined species with indicator value that are very useful in the identification and management of HNV. The indicator species are useful in forecasting the fluctuations and successions in grasslands, determined by the modification of the dominance-codominance ratio and the real coverage of each species.

Grassland ecosystems occupy significant areas worldwide and represent a reservoir for biodiversity. These areas are characterized by oligotrophic conditions that stimulate mycorrhizal symbiotic partnerships to meet nutritional requirements. In this study, we selected Festuca rubra for its dominance in the studied mountain grassland, based on the fact that grasses more easily accept a symbiotic partner. Quantification of the entire symbiosis process, both the degree of colonization and the presence of a fungal structure, was performed using the root mycorrhizal pattern method. Analysis of data normality indicated colonization frequency as the best parameter for assessing the entire mycorrhizal mechanism, with five equal levels, each of 20%. Most of the root samples showed an intensity of colonization between 0 and 20% and a maximum of arbuscules of about 5%. The colonization degree had an average value of 35%, which indicated a medium permissiveness of roots for mycorrhizal partners. Based on frequency regression models, the intensity of colonization presented high fluctuations at 50% frequency, while the arbuscule development potential was set to a maximum of 5% in mycorrhized areas. Arbuscules were limited due to the unbalanced and unequal root development and their colonizing hyphal networks. The general regression model indicated that only 20% of intra-radicular hyphae have the potential to form arbuscules. The colonization patterns of dominant species in mountain grasslands represent a necessary step for improved understanding of the symbiont strategies that sustain the stability and persistence of these species.

Assessing the efficiency of nitrogen application is a valuable first step in establishing an optimum N fertilizer management plan. The aim of this study was to elaborate a sustainable N management system for sown grassland installed in temperate regions, through manipulation of nitrogen use efficiency and nitrogen response efficiency (NRE) indicators. Research was performed in western Romania. The biological material consists in two forage species of major importance for the forage base worldwide, namely Trifolium pratense L. and Lolium multiflorum Lam., in pure stands and in mixture. Both forage species were sown at two different spacings between rows (12.5 and 25.0 cm) and were fertilized with nitrogen (N) in three doses (N0-control, N50-fertilized with 50 kg N ha⁻¹ and N100-fertilized with 100 kg N ha⁻¹). The studies were made in 2016 and 2017, years very different from each other in matter of rainfalls. The results revealed a different behavior of the three sward types analyzed: while in Trifolium pratense monoculture and in the mixture of the two species studied, less N was removed than was applied; on the other hand Lolium multiflorum monoculture used a high proportion of the N applied, being at the limit to access N from the organic pool or residual N from fertilizer applications to a preceding crop.

Our aim is to evaluate the changes in arbuscular mycorrhiza characteristics of winter wheat in a three-year experiment. Study results show that fertilizers produce strong variations in arbuscular mycorrhiza extension, with colonization frequency values within 76–98%. The intensity of colonization is only 12% when phosphorus (P) exceeds nitrogen (N) in autumn, but reaches 38% when the N:P ratio is equal. Root colonization shows no consistency from one experimental year to another, with the largest fluctuations recorded in colonization intensity (22–65%) and arbuscules abundance (0–5%). Arbuscules are maintained below 1% by fertilizer with more P than N. Colonization forecasting models indicate P as a factor for the reduction of symbiosis. Each kg of applied P can reduce the colonization frequency by 0.28% and intensity by 0.37%. The maximum of the colonization degree is 61% due to the synergy of equal N and P doses in autumn and ammonium nitrate applied in spring. The application of multiple moderate doses acts as a stimulant for the development of a large root-fungal interface.

Ecosystems with high natural value (HNV) have generally been maintained by agricultural practices and are increasingly important for the ecosystem services that they provide and for their socio-economic impact in the ever-changing context. Biodiversity conservation is one of the main objectives of the European Green Deal, which aims to address biodiversity loss, including the potential extinction of one million species. The aim of this research was to trace the effects of organic and mineral fertilizers on the floristic composition, but also on the number of species, of the grasslands with high biodiversity (HNV) from the Transylvanian Plain, Romania. The experiments were established in 2018 on the nemoral area and analyzed the effect of a gradient of five organic and mineral treatments. Fertilization with 10 t ha⁻¹ manure or N₅₀ P₂₅K₂₅ ensures an increase in yield and has a small influence on diversity, and it could be a potential strategy for the maintenance and sustainable use of HNV grasslands. Each fertilization treatment determined species with indicator value that are very useful in the identification and management of HNV grasslands. The dry matter biomass increases proportionally as the amounts of fertilizer applied increase and the number of species decreases.

Arnica montana L. is a medicinal plant with significant conservation importance. It is crucial to monitor this species, ensuring its sustainable harvesting and management. The aim of this study is to develop a practical system that can effectively detect A. montana inflorescences utilizing unmanned aerial vehicles (UAVs) with RGB sensors (red–green–blue, visible light) to improve the monitoring of A. montana habitats during the harvest season. From a methodological point of view, a model was developed based on a convolutional neural network (CNN) ResNet101 architecture. The trained model offers quantitative and qualitative assessments of A. montana inflorescences detected in semi-natural grasslands using low-resolution imagery, with a correctable error rate. The developed prototype is applicable in monitoring a larger area in a short time by flying at a higher altitude, implicitly capturing lower-resolution images. Despite the challenges posed by shadow effects, fluctuating ground sampling distance (GSD), and overlapping vegetation, this approach revealed encouraging outcomes, particularly when the GSD value was less than 0.45 cm. This research highlights the importance of low-resolution image clarity, on the training data by the phenophase, and of the need for training across different photoperiods to enhance model flexibility. This innovative approach provides guidelines for mission planning in support of reaching sustainable management goals. The robustness of the model can be attributed to the fact that it has been trained with real-world imagery of semi-natural grassland, making it practical for fieldwork with accessible portable devices. This study confirms the potential of ResNet CNN models to transfer learning to new plant communities, contributing to the broader effort of using high-resolution RGB sensors, UAVs, and machine-learning technologies for sustainable management and biodiversity conservation.