Christensen, T.K., Elmeros, M. & Lassen, P. 2010. Forekomst af antikoagulante rodenticider i danske rovfugle, ugler og små rovpattedyr. En basisundersøgelse. National Environmental Research Institute. 84 pp. - NERI Technical Report no. 788.
In Denmark, control of rats, mice and voles is almost exclusively undertaken using anticoagulant rodenticides, slow-acting poisons that block the blood clotting ability of target animals. Rodents usually die through spontaneous internal bleeding within 3-8 days following consumption of a single or multiple smaller doses of anticoagulant rodenticides, dependant on type. Anticoagulants bind strongly to fatty tissue and break down only slowly, exposing avian and other vertebrate rodent predators and scavengers to a high risk of secondary poisoning through foraging on poisoned rodents.
Secondary fatal poisoning from anticoagulants has been documented in a wide range of species, both in the wild and from experimental trials. Most studies describe specific incidents or relate to suspicious deaths; few studies have been based on systematic collection of individuals to determine the extent of such secondary poisoning and its impact on the health and population dynamics of predators and scavengers.
The present study is the first systematic survey to establish the prevalence of anticoagulant rodenticides in raptors, owls and small mustelids in Denmark. The study focussed on species that strongly or periodically depend on rodents as food, i.e. those most at risk of poisoning and those most likely impacted at the population level.
The presence of anticoagulant rodenticides in raptors, owls and mustelids was studied in specimens obtained from zoological museums, wildlife management stations, taxidermists, airports, the National Environmental Research Institute and various private persons. Specimens were derived primarily from 2000-2009, supplemented by a few individuals from 1983-1999.
Contamination by five anticoagulant rodenticides (Coumatetralyl, Bromadiolone, Difenacoum, Flocoumafen and Brodifacoum) was measured in specimen liver tissue using high performance liquid chromatography (HPLC) and a combination of a fluorescence detector and a photodiode array (PDA) detector.
Manipulative experiments were conducted (i) to assess the risk of poisoning to Tawny Owl Strix aluco chicks in the nesting period by supplementary feeding with poisoned mice, and (ii) to determine the levels of accumulation of poison in mice in areas with standard rodent control activity (poison depots) around a game feeding station. Small sample sizes of owl nests and mice allow no firm conclusions to be drawn from these studies, and the results are only indicative of expected scenarios of poisoning.
Anticoagulant rodenticides were detected in 84-100% of all avian and mustelid specimens examined. Difenacoum, bromadiolone and brodifacoum were most commonly present. Coumatetralyl and flocoumafen occurred in fewer individuals. The average number of different anticoagulants found per individual ranged between 1.7 in marsh harrier Circus aeruginosus and 3.0 in eagle owl Bubo bubo. Overall more than 60% of birds and 80% of mustelids had two or more anticoagulants in their body.
For the kestrel Falco tinnunculus, common buzzard Buteo buteo, tawny owl, barn owl Tyto alba and little owl Athene noctua the average cumulative concentration of all anticoagulants in the liver ranged between 35 and 58 ng/g. For the red kite Milvus milvus and the eagle owl, the average concentration was 152 and 162 ng/g, respectively, while concentrations in rough-legged buzzard Buteo lagopus, marsh harrier, long-eared owl Asio otus and short-eared owl Asio flammeus ranged between 9 and 24 ng/g. There was no obvious explanation for the observed differences, but these may be due to the degree to which the species exploit urbanised areas and farmland. The higher concentrations in red kite and eagle owl may relate to their predation of carrion and rats, respectively. The average concentrations of anticoagulant rodenticides in stoat Mustela nivalis and weasel Mustela erminea were 58 and 63 ng/g, respectively. The highest levels of anticoagulants were measured in stoats and weasels.
Studies suggest that 200 ng/g anticoagulants in the liver represent a critical contamination level for raptors and owls. Between 5.7% and 22% of the examined kestrel, common buzzard, tawny owl, barn owl, little owl, stoat and weasel had concentrations higher than 200 ng/g. Amongst red kite and the eagle owl the proportion of individuals with more than 200 ng/g liver were 66% and 70%, respectively, while there were no critically high concentrations recorded in rough-legged buzzard, marsh harrier, long-eared owl and short-eared owl, although small sample size for the last species prevents firm conclusions.
Overall no differences in frequency and level of contamination were detected between sex and age groups, between time of year, or in relation to the registered cause of death, which could indicate a systematic pattern in the risk of poisoning. For stoat and weasel, concentrations of anticoagulants increased from spring to winter, a trend reflected amongst some of the bird species. Higher concentrations were also present in stoats and weasels found dead, compared to other causes of death, and for both species, body condition were weakly, but significantly correlated with rodenticide burdens.
The recorded frequencies of anticoagulant poisoning in Danish birds of prey, owls and mustelids, were markedly higher for most species than those from British studies. The number of anticoagulants recorded per individual was also higher in Danish birds and mustelids than found in Great Britain. Higher sensitivity of the HPLC analysis used in the present study cannot explain this difference. Possible explanations for the differences between Denmark and Great Britain may include differences in landscape structure, in application policies and methods for authorized rodent control and in private use of the anticoagulants available on the open market.
The use of anticoagulants over the past 10-20 years to control rodents in Denmark has not had any documented adverse effect on the populations of raptors and owls, as most species have increased in abundance during this period. However, it can not be ruled out that the use of anticoagulants locally or even nationally has elevated chick mortality in some birds of prey and owls, which has slowed their population growth. The experiments reported here showed that chicks of tawny owl fed poisoned mice constituting 35% of their food requirements concentrated critically high levels of poison above 200 ng/g in the liver (even for old birds). The study showed that free ranging mice in areas with standard applications of anticoagulants for rodent control may attain concentrations similar to those used in the feeding experiments. Hence, it is not unlikely that elevated chick mortality amongst raptors and owls may occur as a result of the current levels of use of anticoagulants for rodent control in Denmark.
The higher concentration of anticoagulant rodenticides in stoats and weasels dying from unknown causes and lower body condition in individuals with high rodenticides burdens indicate that the present anticoagulant rodenticide exposure have adverse effects on individual fitness and survival. Population development and status of stoats and weasels in Denmark are unknown, but both species are presumed to be in decline. Adverse effects of the massive exposure to anticoagulant rodenticides have the potential of contributing negatively to population developments.
Full report in pdf (2,84 MB)
