IMPROVING RESTORATION OF DEGRADED LANDS USING AUTOCHTHONOUS PLANT MATERIAL FROM SPECIES-RICH GRASSLANDS

The last decades have witnessed an increased request for restoration of degraded habitats worldwide. This necessity is due to an increasing degradation of natural ecosystems, for example through mining, conversion to agricultural land and urban development. Once the causes of degradation have ceased, the restoration of the degraded areas to an ecosystem that resembles the original one is advisable. This process does not create a new natural ecosystem, but puts the site on a trajectory towards an artificial ecosystem more similar to a natural one. The amount of research articles on the topic are also increasingly being published on scientific journals. However, knowledge is still lacking on several parts of the process. One of the main unresolved issues is the effective harvesting and use of seeds. While cultivars are usually selected for having a high biomass production and a reduced dormancy, wild ecotypes have evolved to cope with the harsh environmental conditions. The use of selected cultivars can cause hybridization with native ecotypes of grasses, reducing their suitability to survive in their own habitat. Thus, the most suitable plant material that should be used to ensure restoration success is a mixture of seeds of native species directly harvested in habitats in good conservation status, developing in similar ecological conditions, and with a plant species composition suitable for the restoration purposes. The assessment of the suitability of a certain seed mixture to a restoration site is usually expert-based, leading to high subjectivity. User-friendly modeling tools using multi-species approach are needed to make this process more objective and technically affordable to practitioners. The supply of seeds of wild ecotypes of herbaceous species is often challenging. Indeed, their supply is generally scarce and the knowledge on native seed behavior in restoration ecology is highly heterogeneous. Unlike commercial seeds, which are readily available for purchase, the availability of native seeds is directly limited by the methods used for their harvesting in defined source areas, usually called donor sites. Each method has its own pros and cons, which contribute in structuring the supply chain of native seeds. Differently from cultivated cultivars, native seeds usually show a variable degree of dormancy. Each species has its own physiological characteristics, which should be taken into account whenever they are used as part of a seed mixture. Dormancy varies significantly across the globe, being often correlated with the environmental conditions limiting each ecosystem. The most complex is the dormancy of a species, the most knowledge on how to manage it are necessarily required, through species-specific studies. The knowledge of each species germination features is of practical use to understand which treatments can be adopted to overcome germination constraints leading to a satisfactory achievement of restoration projects. Also environmental conditions affect restoration outcomes. They vary from site to site and include for example: predation of seeds, lack of soil nutrients, excessively high or low temperatures, drought, short vegetative season. These limitations are due to the local climatic and lithological conditions, and also depend on the main natural disturbances driving ecosystem dynamics. The current lack of knowledge on the topic of grassland ecological restoration led to the scientific objectives of this PhD thesis. The aim of the research was to understand possible ways to reduce some of the above-mentioned constraints to allow a better outcome of restoration projects. The scientific questions included: (1) Is it possible to assess the compatibility between a seed mixture sourced from a donor grassland and a restoration site? (2) Is it possible to reduce the germination constraints of native species? (3) Is it possible to artificially reduce environmental constraints?