ABSTRACT1. We report the results of the first systematic assessment of global patterns and research priorities emerging in the field of rodent pest management. We carried out an extensive literature review targeted towards identifying the most relevant rodent pests, their impacts and the most common methods used to control them. 2. We identified three disproportionally important pest species that are characterized by severe, generalist and geographically widespread impacts: the black rat Rattus rattus, the Norway rat Rattus norvegicus and the house mouse Mus musculus.Overall, only 7% of known rodent species may be considered pests. Scansorial (i.e. terrestrial and semi arboreal) and fossorial species are generally important as pests, while aquatic and arboreal species have only specific impacts. 3. Impacts of rodent pests on arable crops were studied most, followed by impacts on ecosystems. Studies on arable crops were typical in countries with low net income and health expenditure, while the opposite was observed for studies on ecosystems. Poisons were the most commonly used control method, followed by traps and habitat management. The need to control rodent species is expanding, especially to protect ecosystems and public health. Unlike in other fields of pest management (e.g. insect control), in rodent control we are approaching new problems with old solutions; control strategies and methods have not kept pace with emerging impacts. 4. The need to control a rodent pest species is higher when it is non-native than within its original geographical range. The impact of a rodent species in its native range is a good predictor of the impact it may have in areas of introduction. 5. Our review will contribute towards guiding researchers and stakeholders to focus research efforts and investments on a subset of species, and on new, less hazardous control techniques.

No abstract

Summary 1.Habitat loss must be distinguished from habitat fragmentation so that appropriate conservation management can be applied. Few studies have evaluated the independent effects of habitat loss and habitat fragmentation on the distribution of vertebrates, and none has evaluated the independent effect of changes in structural connectivity. We carried out a landscape-scale experiment to assess the independent contribution of these three processes and to examine what landscape scale factors affect the distribution of two forest-dependent arboreal rodents: the hazel dormouse Muscardinus avellanarius and the red squirrel Sciurus vulgaris. 2. Habitat loss, rather than habitat fragmentation per se, was the major driver of distribution patterns for both species. As predicted, structural connectivity (hedgerow networks) played an important role in determining the distribution of the hazel dormouse, but not of the red squirrel. 3. Our models predict that long lengths of hedgerows (>30 km) are unlikely to increase the probability of occurrence of hazel dormouse in landscapes where there are low levels of forest cover (<5%-10%). 4. Synthesis and applications. Our empirical findings indicate that structural connectivity and habitat loss may have additive effects on vertebrate distribution. For the hazel dormouse, improving structural connectivity will be ineffective if the amount of forest cover in the landscape is less than 5-10%. The key message from this study is that resources should not be invested in landscape linkages until their efficacy for the given level of suitable habitat has been assessed.

Observations of the rooting activity of wild boar in a holly-oak grove showed that in March±April the decrease of acorns in the diet (31%) was lower than their availability (82%). Moreover the occurrence of deep rooting events remains high despite the low occurrence of grass roots in the diet. These observations suggested that wild boar may exploit hoards of acorns collected by small mammals living in the study area (mainly wood mice Apodemus sp.). In order to test this hypothesis two experimental trials were set up to:(1) investigate whether wild boar were able to locate acorns buried in the ground (range 0±30 cm) and (2) establish if mouse burrows were more likely to be excavated than locations without burrows. The results clearly show that wild boar actively search for buried acorns, mainly in March (59% vs 31% in April and nothing in May) and that burrows are excavated signi®cantly more than locations without burrows (ratio 2:1, respectively). Moreover, locations with burrows are characterized by a decline of rooting activity as a function of the distance from their centre (P = 0.02), which is absent in the control locations (P = 0.74). Our results show that wild boar are able to partly compensate for a reduced above-ground availability of acorns by predating on hoards collected by small mammals. Since this occurs during a critical period for female wild boar when they are giving birth and lactating, this behaviour may strongly in¯uence the population dynamics of both wild boar and small mammals.

Aim Habitat loss and fragmentation are amongst the greatest threats to biodiversity world-wide. However, there is still little evidence on the relative influence of these two distinct processes on biodiversity, and no study, to date, has investigated the independent contribution of structural connectivity in addition to habitat loss and fragmentation. The aim of this study is to evaluate the independent effects of habitat loss (the decrease in total amount of habitat), habitat fragmentation per se (habitat subdivision) and structural connectivity (in the form of hedgerow networks) on the distribution of seven resident forest-dependent birds in central Italy. Location Central Italy. Methods We strategically selected 30 landscapes (each of 16 km2 in size) with decreasing total amount of forest cover and with contrasting configuration of patches and contrasting lengths of hedgerow networks. Presence/absence of birds in each landscape unit was studied through point counts. Results The amount of forest cover in the landscape had the strongest relative influence on birds' occupancy, whilst habitat subdivision played a negligible role. Structural connectivity and the geographic position of the landscape unit played a relatively important role for four species. Main conclusions Our study shows the importance of disentangling the contribution of different landscape properties in determining distribution patterns. Our results are consistent with the fact that halting habitat loss and carrying out habitat restoration should be conservation priorities, since habitat loss is the main factor affecting the distribution of the target species; implementation of structural connectivity through hedgerows, instead, should be evaluated with caution since its contribution is secondary to the predominant role of habitat loss