The contribution to wildlife conservation of an Italian Recovery Centre

Wildlife recovery centres are widespread worldwide and their goal is the rehabilitation of wildlife and the subsequent release of healthy animals to appropriate habitats in the wild. The activity of the Genoese Wildlife Recovery Centre (CRAS) from 2015 to 2020 was analysed to assess its contribution to the conservation of biodiversity and to determine the main factors affecting the survival rate of the most abundant species. In particular, the analyses focused upon the cause, provenance and species of hospitalised animals, the seasonal distribution of recoveries and the outcomes of hospitalisation in the different species. In addition, an in-depth analysis of the anthropogenic causes was conducted, with a particular focus on attempts of predation by domestic animals, especially cats. Significantly, 96.8% of animals hospitalised came from Liguria, the region in north-western Italy where CRAS is located, with 44.8% coming from the most populated and urbanised areas of Genoa, indicating a positive correlation between population density and the number of recoveries. A total of 5881 wild animals belonging to 162 species were transferred to CRAS during the six years study period. The presence of summer migratory bird species and the high reproductive rates of most animals in summer resulted in a corresponding seasonal peak of treated animals. Birds represented 80.9% of entries; mammals accounted for 18.6% of hospitalisations; and about 0.5% of the entries were represented by reptiles and amphibians. Species protected by CITES and/or in IUCN Red List amounted to 8% of the total number of individuals. Consistent with results recorded elsewhere from Italy and other European countries, 53.9% of the specimens treated were released in nature; 4.7% were euthanised and 41.4% died. There was a significant difference between taxa in the frequency of individuals that were released, died or euthanised due to the intrinsic characteristics of species (more resistant or more adaptable to captivity than others) and/or to the types of debilitative occurrences common to each species (e.g. Nature Conservation 44: 1–20 (2021) doi: 10.3897/natureconservation.44.65528 https://natureconservation.pensoft.net Copyright Gabriele Dessalvi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. RESEARCH ARTICLE Launched to accelerate biodiversity conservation A peer-reviewed open-access journal Gabriele Dessalvi et al. / Nature Conservation 44: 1–20 (2021) 2 infections, wounds, traumas, fractures). A total of 14.2% of wildlife recovery was from injuries caused with certainty by people or domestic animals (human impact), with 54.3% of these hospitalised animals having been victims of predation attempts by domestic animals, mainly cats. The percentage of release in nature of animals hospitalised following human impact was significantly lower than overall cases (31.2% vs. 53.9%) due to the greater severity of the injuries. The percentage of animals released showed a further reduction to 27.1% amongst victims of predation attempts by pets. The work of Rehabilitation/Recovery Centres contributes to wildlife conservation. In particular, the CRAS in Genoa is a Centre with an increasing level of activity concerning the rehabilitation of species under CITES protection and/or included on the IUCN Red List. The contribution and experience of CRAS operators is critical for the success of ‘information campaigns’ aimed at limiting the number of stray dogs and cats because of their impact on wildlife. Therefore, the activity of a properly-managed CRAS can significantly contribute both directly and indirectly to wildlife conservation, resulting in important territorial safeguards for the protection of biodiversity.


Introduction
Wildlife Recovery Centres are widespread worldwide (Tribe and Brown 2000;Burton and Doblar 2004;Rouffignac 2008;Rodriguez et al. 2010;Wimberger et al. 2010;Molina-López et al. 2011;Grogan and Kelly 2013). In Italy, they are managed by public or private entities that conduct their work in close contact with the local administrations (Table 1). The goal of each Recovery Centre is the rehabilitation of wildlife, defined as "the treatment and temporary care of injured, diseased and displaced indigenous animals and the subsequent release of healthy animals to appropriate habitats in the wild" (Miller 2012).
Wildlife "rehabilitation and relocation" is a traditional management practice, defined by Begg and Brown (1998) as "taking wild animals that are injured, sick or orphaned and providing basic veterinary support, with the aim of bringing them back in their natural state, in the habitat from which they come". Wildlife rescue, rehabilitation and relocation are arguably the most intimate, intense and costly interaction that most people can have with wildlife. Such activities naturally involve human intervention in the life of wild animals by raising important emotional, political and ethical issues (Tribe and Brown 2000). Moreover, they need to be based on a comprehensive approach that takes into account the needs of animal welfare, its eco-ethological basis and advances in veterinary science. Wild animals are in a sentinel position as biological indicators of environmental conditions in every habitat, even in urban and suburban areas. Some of their causes of debilitation and mortality can be described as "unnatural" and directly related to human activities. The activity of Wildlife Recovery Centres, therefore, can take on a triple importance: acting as a compensation system of anthropogenic impact, contributing to the conservation of animal biodiversity and guaranteeing a continuous monitoring of the general health status of wildlife (Burton and Doblar 2004).

The Centre
In Genoa, CRAS has operated since 2015 and is the only organisation authorised to recover, rehabilitate and relocate injured wild animals throughout the Genoa hinterland and the whole Liguria Region. The Centre is managed by ENPA Onlus (National Animal Protection Agency), the oldest Italian animal rights association (established in 1871) that manages many Recovery Centres and other specialised structures throughout Italy (www.enpa.it).
The Genoese CRAS is located in the Municipality of Campomorone in the Polcevera Valley, in the Liguria Region (north-western Italy) and consists of a central office where the main activities are carried out and of a rehabilitation area located about two kilometres away. The main complex includes a veterinary clinic, an area for welcoming the public and a series of rooms where animals are housed with 'batteries' for small animals and 'housings' for larger animals, as well as a series of bags and coolers for food storage. The final rehabilitation area for animals before their release is located in an isolated woodland away from the public. It is a fenced area that includes six aviaries of different sizes, a small stable and two internal enclosures. Larger animals can also be accommodated here. The animal's release can only take place after clearance from a veterinarian. This internationally-accepted procedure helps maximise the success of re-introducing an injured wild animal back into its natural habitat (Kirkwood 1993;McDougall et al. 2006;Harrington et al. 2013;Mullineaux 2014).
The recovery of injured or displaced wild animals is often carried out by citizens, especially in the case of small to medium-sized animals. In other instances, When a debilitated animal arrives at the Centre, staff records its date of entry, species, age, provenance and assess its apparent injury or disability. The animal then receives appropriate first aid and housing, before undergoing a thorough examination by a veterinarian, the results of which do not always agree with the initial diagnosis. Following the examination, the veterinarian may recommend that the animal be euthanised when it cannot be saved or when the injuries would result in an animal living with severe pain and/or loss of independence. On the other hand, if the veterinarian's evaluation is positive, the animal will receive the necessary treatment(s) for its recovery from the veterinarian and other appropriately qualified staff at the Centre.
At Campomorone, the Centre has only three employees professionally trained in zoology; the remaining staff is made up of volunteers trained in the treatment of wild animals who primarily feed and clean the animals and their enclosures. When an animal's treatment regime has finished and the observations on its behaviour are positive, the veterinarian will approve its transfer to a designated rehabilitation area.
During the last stage of its rehabilitation at the Centre, the animal is visited only once a day to be fed, in order to minimise human contact. The final rehabilitation phase varies due to factors such as the species and age of the animal, the cause(s) of debilitation and the timing of its release due to the availability of suitable habitat and favourable climatic conditions. All releases take place in accordance with the Italian law 157/1992 governing the management and protection of wildlife. The most common problem faced by the Centre is its inability to free some animals due to their overpopulation in the wild (e.g. in the case of Wild boar Sus scrofa) or because they belong to exotic species (e.g. the Rose-collared parakeet Psittacula krameri and the Greek tortoise Testudo graeca), which are often victims of neglect or object of seizures by the authorities, the latter being a common problem in other countries (Trendler 1995;Kirkwood and Sainsbury 1996;Kirkwood and Best 1998;Kirkwood 2000).

Materials and methods
The data recorded on the Centre's registers (2015-2020) were analysed in order to obtain information relevant to the management and conservation of wild animals.
An ordinary least square regression, followed by a Mann-Kendall Trend Test was performed on data about annual recoveries in order to verify the actual increase of the Centre's activity.
The geographic provenance of each specimen was attributed to the following areas: West and East Liguria, Genoa City and those external to Liguria.
The phenology of the entries was analysed for dominant species (at least 2% of the total individuals) or homogeneous groups of species, dividing the year into half-month periods (marked with the numbers 1 and 2). The recoveries were examined focusing on those species that represented at least 1% of the total individuals, while the remaining species were grouped homogeneously with a focus on the causes of debilitation and the final outcomes. Overall, differences amongst frequencies of species' outcomes (adding together the numbers of animals that died in care or were euthanised) were analysed using the Chi-square Test. Additional in-depth analysis was undertaken on data concerning animals hospitalised for injuries resulting from "anthropogenic causes". The frequencies of release of such individuals and of those admitted to the Centre following predation attempts by domestic animals was compared with the frequency of releases on the total of hospitalisations using the Chi-square Test.
PAST Software (Paleontological Statistics version 4.02, Hammer et al. 2001) was used to perform the statistical analyses mentioned above.

Results
There was a total of 5881 wild animals transferred to the Recovery Centre between 2015 and 2020 with the majority coming from the Liguria Region (Table 2, Fig. 1). Only a small percentage of animals came from outside Liguria (Piedmont, Emilia-Romagna, Aosta Valley). Despite its relatively small territory and high population density, the Genoese urban area resulted in the highest percentage of wild animal recoveries (44.8% of the total). High numbers of wildlife animal recoveries were also associated with the Polcevera Valley (18.8%), where CRAS is located. The Polcevera Valley is rich in natural and semi-natural patches of vegetation that lie adjacent to the southern border of the Capanne di Marcarolo Natural Park, which has been a protected area since 1979 ( Fig. 1).
A total of 162 species were housed in the Centre during the six-year study (Appendix 1) which recorded an associated increase in its level of activity over the same period (Fig. 2). The linear regression analysis performed on the annual number of recoveries confirmed a highly significant increasing trend (a = 253.6 ± 11.6; p << 0.01; r 2 = 0.97), which was also evidenced by the Mann-Kendall Test (p no trend = 0.0014) (box in Fig. 2).
The number of wildlife admissions to CRAS during the year (Fig. 3) showed a close relationship with the seasons, with a maximum peak occurring from early spring to early autumn and, upon close examination of the dominant species, patterns were evident in their hospitalisation. The Yellow-legged gull Larus michahellis, the Collared dove Streptopelia decaocto and the Blackbird Turdus merula are very common sedentary species, well adapted to the urban environment and their frequency of admission was almost constant throughout the year, with an increase during the breeding season. The Common swift Apus apus, is a widespread migratory species whose presence in Liguria occurs only during its breeding period, which correlated closely with its peak in admissions during spring and summer. European hedgehog Erinaceus europaeus and Roe deer Capreolus capreolus are common mammals whose admissions to CRAS are fairly frequent and evenly distributed throughout the year, except in the late spring-early summer when there is an increase in admissions due to births.  houses numerous tortoises and snakes resulting from abandonment or seizures by the Finance Police due to trade irregularities. Species protected by CITES (Convention on International Trade of Endangered Species) and/or included on the IUCN Red List , amounted to 8% of total admissions. Some of these species are very or quite common, such as the Little owl Athene noctua, Sparrowhawk Accipiter nisus, Buzzard Buteo buteo and Italian sparrow Passer italiae, while others represent rarer species, such as the Eagle owl Bubo bubo, Peregrine falcon Falco peregrinus and Italian wolf.
In Fig. 4, the percentages of the final outcomes of hospitalisation for the dominant species are shown. In general, 53.9% of the specimens treated were released back into nature; 4.7% of the animals were euthanised; and the remaining 41.4% died. The Chisquare Test evidenced a significant overall difference amongst dominant species/group of species in terms of frequencies of released and euthanised/died animals (χ 2 = 408.32, 21 d.f., Monte Carlo p = 0.0001). Focusing on the causes of debilitation, 14.2% of hospitalisations was attributable to man and domestic animals (human impact). It is very likely that this percentage is underestimated, since, for almost all the other hospitalisations, the specific cause of the injuries found in the animals is unknown. Fig. 5A shows the relative incidences of the various anthropogenic causes of hospitalisation. Based on reports of people who have rescued the animals or on the examination of injuries by the veterinarian, 54.3% of them were related to attempts of predation by domestic animals, mainly cats (Felis catus). The percentage of release back into nature of animals hospitalised following an anthropogenic impact (Fig. 5B) was 31.2%. This value, due to the greater severity of the injuries, is significantly lower than 53.9% recorded on the total of recoveries (χ 2 = 106.35, 1 d.f., Monte Carlo p = 0.0001 -Fisher's Exact Test: P (no assoc.) <<0.01). From analysing data on the victims of predation attempts by domestic animals, the release rate resulted in 27.1%, which did not differ significantly from that of releases based on the total hospitalisations for anthropogenic causes (χ 2 = 1.7544, 1 d.f., Monte Carlo p = 0.1855 -Fisher's Exact Test: P (no assoc.) = 0.19191).

Discussion
The Genoese CRAS received annually some hundreds of debilitated wild animals that mainly came from the Liguria Region; a high percentage of admissions came from the relatively small and urbanised area of Genoa. This indicates the essential contribution of the general public to the recovery of wild animals and a positive correlation between density of the human population and the probability of wildlife recovery. The linear increase of activity recorded over the years seems to indicate an increase in the degree of public awareness on the role of the Centre in the rehabilitation of wild animals and on the importance of wildlife conservation, but also a greater anthropogenic impact on nature.
The seasonal distribution of wildlife admissions recorded at CRAS is the result of two main factors: the presence of summer migrants (especially birds) and the reproduction of most animals during the hot season. Both these factors led not only to a numerical increase of individuals in the wild, but also to an increase in the percentage of juveniles (more than 70% of the entries during the hot season) that are at greater risk of capture than adults. For example, Roe deer fawns, according to the ethology of the species, are frequently found alone crouching in the woods by people who take them thinking they have been abandoned (Marsan and Spanò 1999). The return of these animals by CRAS back into the wild can directly enhance the reproductive success of many wildlife species by increasing the survival rate of newborns and juveniles.
Birds accounted for 80.9% of entries; mammals 18.6%; and amphibians and reptiles 0.5%. This representation of animal type by percentage is quite common in recovery centres worldwide (Molina-López et al. 2017;Romero et al. 2019). The low number of recoveries of amphibians and reptiles is presumably the result of two concomitant factors: the difficulty of seeing them, especially when debilitated and the little empathy that these species arise in the community. Some of the species/group of species recovered in Genoa are so frequent amongst those hospitalised in European Recovery Centres that they are the subject of monographs or specific papers in the veterinary and conservation literature relating to recovery and rehabilitation. This is the case for the European hedgehog (Bunnell 2001;Bullen 2002;Martinez et al. 2014;Yarnell et al. 2019), Red fox (Matesic and Finegan 2016;Tolhurst et al. 2016), owls (Couper and Bexton 2012), Sparrow hawk (Kelly and Bland 2006) and birds of prey in general (Keran 1981;Molina-López et al. 2011). The latter species include those protected by CITES and included in IUCN Red List, which further highlights the important contribution of Recovery Centres in general and of the Genoese CRAS in particular, to wildlife conservation.
In Genoa, the percentage of released animals from CRAS was 53.9%, which can be considered as a successful outcome when compared with those known for other Italian CRAS Centres (http://www.recuperoselvatici.it/) that typically range between 35% and 65%. The lower percentages were recorded in the smaller Centres without veterinary assistance, while the highest percentage was recorded in the CRAS at Monte Adone (Bologna Province), a privately funded Centre. Similar results have been recorded in other countries. A study of four Royal Society for the Prevention of Cruelty to Animals (RSPCA) Recovery Centres, located in England, showed an overall release rate of about 40% of casualties (Grogan and Kelly 2013). The analysis of data from a long-term study (19 years), based on the activity of a Recovery Centre, located at Torreferrussa in the Spanish Region of Catalonia, achieved an overall percentage of released animals slightly over 50% (Molina-López et al. 2017). Tribe and Brown (2000), through an analysis of the outcomes in Australian Wildlife Rehabilitation Centres, showed an overall release rate ranging from 38 to 45%. The significant differences recorded amongst different species' release rates can be attributed to the intrinsic characteristics of the species (animals more resistant or more adaptable to captivity than others) and to the types of debilitation suffered by the different species (e.g. infections, wounds, traumas, fractures) (Molina-López et al. 2017;Hanson et al. 2019).
In the Genoese CRAS, a significant percentage of hospitalisations (14.2%) were due to anthropogenic causes. Mortality of wildlife (especially birds) due to such causes has been previously analysed in depth both globally and locally (e.g. Galuppo and Borgo 2006;de Lucas et al. 2007;Loss et al. 2014Loss et al. , 2015Janssen et al. 2020). These studies highlight that at least for birds, which are the most frequently hospitalised animals in the CRAS we studied, predation by cats alone accounts for a higher mortality than the other main anthropogenic causes of death combined (agricultural chemicals, electrocution and collisions against buildings, windows, vehicles, communication towers, power-lines and wind turbines). In our study, more than half (54.3%) of hospitalisations due to anthropogenic causes were due to predation attempts by domestic animals (mainly cats), with a release percentage significantly lower than the overall one (27.1%). Similar results were demonstrated by Loyd et al. (2017) who reviewed data from 82 North American wildlife Rehabilitation Centres. Domestic animals were found to be responsible for 14% of the hospitalisations and 78% of the attacked animals did not survive. Many papers have highlighted and quantified the impact on reptiles, small birds and mammals due to feral cats (Churcher and Lawton 1987;Woods et al. 2003;Kays and DeWan 2004;Baker et al. 2005;Dickman 2009;Legge et al. 2017;. In the collection of the Museum of Natural History of Genoa, for example, there are 14 birds, 13 mammals and one reptile from Liguria and Piedmont Regions that were killed by cats between 1997 and 2019 (Table 3). Furthermore, one of the authors (L. Galli, pers. comm.) saw an apparently docile dog killing Cats and dogs, left free to roam, significantly alter the natural balances that regulate wildlife, resulting in significant levels of mortality for many species, including endemic ones (Churcher and Lawton 1987;Woods et al. 2003;Kays and DeWan 2004;Baker et al. 2005;Ancillotto et al. 2013;Loss et al. 2013Loss et al. , 2015.

Conclusions
The work of rehabilitation/recovery centres contributes to wildlife conservation and the one in Genoa is growing in its activity concerning recovery and release of species under protection from CITES and/or the IUCN Red List, which now stands at 8% . The Genoese CRAS's level of efficiency is above average with 53.9% of animals released back into their natural habitats. However, this Centre needs the continuous support from the community (most of the funds come from private donations) and from the government authorities in order to continue and expand its work. Future projects of the Centre include the enlargement of facilities, hiring and training of new staff and the purchase of a radio-tracking system for monitoring the animals once released. These systems have been in use in some Centres for several years and allow staff to more accurately evaluate success rate of animals that manage to reenter their natural ecosystem (Kenward 1993;Griffiths et al. 2010;Grogan and Kelly 2013;Mullineaux 2014;Musto et al. 2020). The CRAS in Genoa recorded a certain percentage (14.2%) of cases hospitalised because of human direct or indirect impact. Moreover, animals which recovered from injuries due to predation attempts by domestic animals were those at highest risk of death. This result leads us to believe that, in addition to the Centre's activity, the adoption of preventative measures and a greater disclosure concerning the cause(s) of injury is fundamental. For example, the importance of animal over-and underpasses in enhancing connectivity between habitats of wild animals and in reducing the risk of animal-vehicles collisions is well known (Burton and Doblar 2004;Misłajek et al. 2020;Ważna et al. 2020). However, information campaigns designed to raise the level of awareness of politicians and the general public on the destructive impact of stray cats and dogs can also be effective in protecting wildlife populations (Loss et al. 2013;Mori et al. 2019;Trouwborst and Somsen 2020). It is critical that cat and dog owners prevent their pet animals from straying into wildlife habitats and the staff at CRAS can play a lead role in educating the general public on how best to curb the hunting of wildlife by cats and dogs and how pet owners can play a pivotal role in the protection and conservation of our wildlife, especially the conservation of our critically endangered species.