| dc.description.abstracteng | Understanding the dynamics of species interactions under the threats of habitat loss and fragmentation can be key to develop measures preventing further degradation of natural and agricultural systems. Agroecological knowledge and state of the art technologies can help to conciliate the often-discrepant objectives of biodiversity conservation and agricultural production. Specifically, information on the characteristics of plant-pollinator networks in agroecosystems can unveil the most efficient strategies to preserve ecosystem functionality and pollination services provision.
In chapter 1, I focused on the contributions of new technology to the objective of turning agricultural landscapes increasingly compatible with biodiversity. I reviewed the applications of unmanned aerial vehicles (UAVs) in ecology and precision agriculture. I 1) identified existing applications, 2) discussed limitations and advantages of the current technology, 3) highlighted knowledge gaps and 4) proposed new applications.
In chapter 2, I studied the characteristics of a plant-pollinator metanetwork of calcareous grasslands embedded in an agricultural matrix. I characterized and discussed the structural properties of the network that determine its stability and resilience to perturbations. Furthermore, I identified the traits of the most central nodes in the network. I found that the metanetwork was significantly more modular and less connected than expected by chance. This reflects the existence of many fragment-unique interactions and is an indication of poor metanetwork stability. I also demonstrated that habitat size and the diversity of land cover types in the surroundings of a grassland fragment are significant predictors of site centrality. Thus, these features can help to identify the most important fragments for metanetwork cohesiveness. Additionally, I found that the centrality of interactions depends on the pollinator size, species identity and also on the plant’s habitat specialization.
In chapter 3, I compared plant-pollinator networks constructed with flower visitation data to networks constructed with pollen transport data. The level of specialization of pollen transport networks was higher than that of visitation networks, as half of the interactions in the visitation networks did not occur in the pollen transport networks. This highlights the fact that visitation does not necessarily imply pollen transport, and I discussed its implications for the conservation of pollination. Considering that high specialization is known to be associated with low stability in mutualistic networks, this result has important implications for conservation. According to these findings, traditional studies on plant-pollinator networks, based on visitation data, would overestimate the stability of pollination networks. Additionally, I identified that almost a third of the total number of interactions found are difficult to spot given their low frequency, occurring only in the pollen transport networks. Finally, I found positive effects of landscape diversity on the total number and proportion of single-fragment interactions for pollen transport, but not for visitation networks.
In conclusion, the protection of large and small calcareous grasslands as well as the enhancement of landscape heterogeneity was found to be essential for the maintenance of the plant-pollinator metanetwork. Furthermore, the importance of interactions among habitat specialist plants and large-bodied generalist pollinators appeared to be fundamental to connect the plant-pollinator metanetwork. Nonetheless, small solitary bees and the habitat specialist butterfly Polyommatus coridon also played a central role for the plant-pollinator networks in calcareous grasslands. By identifying the most central plants, pollinators and interactions at the metacommunity level, the information reported in this work can inform tailored management measures to protect them. Among others, I suggest considering plant species’ roles in the metanetwork when applying conservation measures, such as flower strips, and landscape coordination among farmers to increase crop diversification. Moreover, I showed the great potential of UAVs to contribute to such conservation measures and to biodiversity management in agricultural landscapes. UAVs can assist in diverse tasks such as quantifying encroachment of calcareous grasslands and plant diversity monitoring. Additionally, they can contribute to farmer´s cropland management and agri-environmental schemes surveillance by governmental agencies. | de |