Landscape and Ecological Engineering

, Volume 14, Issue 2, pp 221–229 | Cite as

The influence of time factors on the dynamics of roe deer collisions with vehicles

  • Gytautas IgnataviciusEmail author
  • Vaidotas Valskys
Original Paper


Environmentalists and authorities responsible for road safety are trying to reduce the number of wildlife collisions with vehicles (WCV) worldwide. Roe deer are the most common large animal involved in WCV in Europe. This article discusses the distribution of 2010 wildlife-vehicle collisions involving roe deer (WVRD) in Lithuania in 2013 and 2014. The collisions were analyzed in terms of monthly and daily data for each month separately, and the results are compared with the time of sunrise and sunset in Lithuania. By analyzing trends of natural factors that influence the number of collisions we show that the frequency of WVRD is strongly correlated with seasonal and yearly changes in sunrise and sunset. This research shows that these natural factors are extremely important for the dynamics of WVRD. Future analysis of these factors and application of appropriate preventative measures should significantly reduce the risk of collision between vehicles and roe deer.


Wildlife collisions Time factors Road Traffic Population dynamics 


Wildlife collisions with vehicles (WCV) are an important socioeconomic, road safety, human health and animal welfare problem (Groot-Bruinderink and Hazebroek 1996; Gunson et al. 2009; Joyce and Mahoney 2001; Romina and Bissonette 1996). As a result of these accidents, people may be killed or injured, material lost and wildlife numbers decline.

Roads negatively affect animals in a number of ways, and wildlife fatalities are one of their most important negative impacts (Litvaitis and Tash 2008; Smith-Patten and Patten 2008; Haigh 2012). Traffic is one of the main factors negatively affecting the abundance of animal populations (Cervinka et al. 2015). Scientists recognize that roads are very important for the stability of animal habitats, and with their increase, their importance will only increase (Findlay and Bourdages 2000; Gibbs and Shriver 2002). In Sweden, for example, WCV leads to a reduction in animal populations by 1–12% (Seiler et al. 2004).

The prevalence of WCV is strongly influenced by both anthropogenic and natural factors. The frequency of collisions with animals is directly related to the intensity of traffic on roads, thus the density of the road network is important (Damarad and Bekker 2003), as are several types of preventative measures.

The frequency of some large-animal WCV is strongly correlated with their population dynamics (Balčiauskas 2009). WCV is also influenced by the landscape, seasonality, time of day, weather conditions and other natural factors (Snow et al. 2014; Seo et al. 2015; Loro et al. 2016).

Natural environmental factors are very important for the dynamics of WCV, and analysis of their trends allow one to simulate and predict when and where the greatest risk of WCV occurs (Hubbard et al. 2000; Malo et al. 2004; Seiler 2005).

Globally, road safety authorities are trying to reduce the prevalence of and damage due to WCV. To address these challenges, authorities require detailed information about WCV, including informative and reliable data (Gunson et al. 2009; Snow et al. 2014).

WCV are analyzed in most countries through registers of data of traffic safety, to which police contribute by collecting and analyzing data and identifying the most dangerous roads, the so-called hot spots. Various preventative measures are implemented based on the results of these analysis, such as putting up signs on dangerous stretches of roads, erecting road barriers and other active and passive protection measures that are designed to reduce the frequency of collisions and resultant damage (Spellerberg 1998; Clevenger et al. 2001; Huijser et al. 2007; Fahrig and Rytwinski 2009). The synthesis of natural and ecological data also enable the identification of areas where it is appropriate to install various types of bypasses to allow animals to safely cross roads and thus reduce the negative impact of WCV on biodiversity (Cserkész et al. 2012; Evink 2002; Forman et al. 2003; Seiler 2004). Often, institutional data on WCV are scarce and difficult to analyze because of the fragmentation of data sets, lack of precision, negligent data collection, use of complicated legal terms and lack of competence of professionals who gather information and register incidents (Steiner et al. 2014). All of this affects the analysis of incidents, which due to errors, does not always lead to the implementation of the most effective preventative measures.

Quite often the selection of preventative measures and implementation of these for WCV is limited to hot spot measurements and analysis (Cserkész et al. 2013). Usually hot spots are determined based on statistical models that assess how the landscape, traffic and wildlife abundance affect the rate of WCV (Danks and Porter 2010; Gomes et al. 2009; Malo et al. 2004; Ramp et al. 2005; Snow et al. 2014). The analysis of places where WCV is usually recorded are the basis for the identification of physical environmental factors that pose the greatest risk to WCV frequency. However, this approach is not sufficient. The frequency of WCV is influenced by not only physical environmental factors but also by animal behavior, and the influence on these of climatic, seasonal and time of day factors (behavioral periodicity, sunrise etc.) (Steiner et al. 2014; De Vries 2015).

Large animals (over 30 kg), typically vertebrates, are most frequently involved in collisions with vehicles (Conover et al. 1995; Putman 1997). The most prevalent large vertebrate species involved in WCV are roe deer (Capreolus capreolus), elk (Cervus canadensis), wild boar (Sus scrofa) and others depending on the country in question. Wild boar are involved in as many as 39% of accidents in Belgium (Morelle et al. 2013), while in Lithuania WCV usually involve roe deer (WVRD). In other European countries WCV statistics on animals usually involve different species, but are still dominated by roe deer, elk, other deer species, wild boar and other large mammals.

Roe deer (C. capreolus) is the smallest but most common deer species in Europe (Apollonio et al. 2010). It is also one of the species most often involved in accidents with cars, and the number of traffic accidents including these animals in recent years has increased (Groot-Bruinderink and Hazebroek 1996; Langbein et al. 2011; De Vries 2015) in Europe. In Lithuania accidents including roe deer also represent a major part of the WCV associated with large animals (Balčiauskas 2009; Oškinis et al. 2013). Roe deer collisions with traffic are usually fatal to the animals. Though people are rarely killed in these road accidents, they often experience various degrees of physical damage and other effects on their health. Specialists have evaluated the average economic loss due to such collisions to be about 1900€ (Putman et al. 2004).

Roe deer collisions with vehicles are heavily influenced by seasonal behavioral characteristics (feeding, breeding and reproduction periods). Roe deer behavior changes cyclically during the year, which influences the rate of collisions with vehicles. There are three behavioral aspects of roe deer that may affect WCV. While it is known that roe deer have specific daily and seasonal peaks of activities, their behavioral changes have large fluctuations and they show high variability in their activity rhythms (Cagnacci et al. 2011; Mysterud 1999).

Models on the timing of roe deer activity in their life cycle and annual and daily cycles are some of the most important ways of understanding the intensity and patterns of their collisions with vehicles. A seasonal peak of WCV has been identified in many countries; however, the patterns and related models are not consistent between different natural regions. They show a large influence of animal behavior on WCV and its interaction with local environmental factors (Litvaitis and Tash 2008), climatic conditions (Compare Dal et al. 2007) as well as predators, hunting, and other interference effects (Rost and Bailey 1979).

Scientists admit that there is currently insufficient information on the patterns and frequencies of roe deer collisions with vehicles (Barthelmess and Brooks 2010; Steiner et al. 2014). Most of the analyses of roe deer collisions with vehicles have identified daily or yearly cyclical patterns with well-defined peaks, but so far little has been shown regards how the daily cyclical pattern is influenced by the cyclical nature of factors throughout the year. In this work we try to at least partially reduce the gap in our knowledge on the influence of time factors on vehicle collisions with roe deer.


Lithuania covers an area of 65.3,000 km2. It is located in the middle latitudes near the East European Plain in the western part of the transition from a maritime to continental climate, in the geographical area of Eastern European mixed forests. Lithuania lies in the temperate northern zone and has moderately warm summers and moderately cold winters. More than half of Lithuanian territory is covered by farmland, meadows and pastures, comprising 52.18% of the country. Forests cover about 33.56% of the country (Fig. 1), and since 2003 these have mostly comprised coniferous trees (56.1%); deciduous trees comprise 40.3% of all Lithuanian forests. As a consequence of the forested land, roe deer are found throughout the country.
Fig. 1

The main types of land use in Lithuania

Roads in Lithuania are divided into state and local roads. The total network of Lithuanian roads is 92,000 km long, with state roads in 2015 comprising 21,252 km. The Lithuanian road network is currently spread fairly evenly, and the road density is about 328.4 km 1000 km−2 (Fig. 2). Nearly 64% of state roads are covered with asphalt and about 36% with gravel; very few have other coatings. In Lithuania there are 521 km of four-lane roads, of which 417 km are highways.
Fig. 2

Road network in Lithuania

According to the Lithuanian Road Administration (2015), thousands of cars were registered in 2014 in Lithuania. The average daily traffic on roads was 8274 vehicles, 2044 of which were on national roads, and 357 on regional roads.

Currently roe deer (Capreolus capreolus) is the most abundant hoofed animal in Lithuania. Their number has been growing rapidly in recent decades; in 2006, there were about 86,000 roe deer, and in 2014 their abundance reached almost 120,000.

This article is based on the 2013 and 2014 official records of animal collisions with motor vehicles, covering the entire territory of Lithuania and all collisions with cars, provided the by Lithuanian Road Administration (2015).

This register includes all the incidents recorded in Lithuania, so it can be stated that the present study evaluates all officially recorded collisions with roe deer for this period.

During the period of analysis, 3788 vehicle collisions with animals were registered in Lithuania. Records relating to roe deer contributed 2010 units to the statistics, i.e. 53% of all registered events that involved animals. These data were grouped according to the calendar of registered events and the time of day. Data for 2013 and 2014 have been analyzed, and further analysis was based on statistical analysis and geographic information system distance analysis.

Since roe deer behavior is characterized by both annual and daily rhythms, in order to better visualize the collision dynamics and understand the data radar diagrams were used. This type of chart is the best way of visualizing cyclical processes.

This study also analyses the distribution dynamics of the data on collisions of these 2 years in months and within days of each month separately. The results are compared with sunrise and sunset times. The dependence of collisions on landscape, traffic or other factors is not analyzed here.


The death of roe deer on roads as a result of WCV varies throughout the year and appears to be related to specific seasonal behaviour of the animals. The percentage distribution of roe deer collisions with vehicles for different months is shown in Fig. 3.
Fig. 3

Percentage of total monthly car accidents and wildlife collisions with vehicles involving roe deer (WVRD) in Lithuania for 2013 and 2014 (n = 2010)

It is important to evaluate the impact of fluctuations in traffic intensity on seasonal roe deer collisions with cars. Although we did not have data on traffic intensity for any given season in Lithuania, we did have the total number of car accidents in Lithuania for separate months in 2013 and 2014, which reflect seasonal and daily changes of traffic intensity, albeit indirectly.

In 2013 and 2014 the total number of accidents was distributed quite evenly over the months. There was only a weak correlation between the total number of accidents and collisions with roe deer (r = 0.041). The total percentage of car accidents (8.90%) remained close to the average number of collision with roe deer (8.33%) in May, when collisions between cars and roe deer are at their peak (14.44% of all collisions). Therefore, it can be assumed that fluctuations in traffic intensity do not have a significant impact on accidents involving roe deer in Lithuania, thus this relationship was not further studied.

Regards the distribution of collisions during the day (in hours) in Lithuania in 2013 and 2014, there were 2010 roe deer collisions with vehicles that showed a different distribution during the day. There were three peaks during the day (Fig. 4) when the seasonal differences in collisions between different months were disregarded.
Fig. 4

Distribution of total number of car accidents and WVRD over the course of the day in Lithuania in 2013 and 2014 (n = 2010)

Only a very weak correlation (r = 0.24) was found between the hourly fluctuation in the number of accidents during the day and the number of car collisions with roe deer. The total number of accidents increased in the morning (0800 hours) and evening (1800 hours) peak hours. Increases in roe deer collisions with cars were recorded at 0700, 1700 and 2200 hours, times when traffic intensity had not yet reached its peak or was already decreasing. It should be noted that in Lithuania, due to the type of road network of the country and relatively small number of cars, significant peaks of increase in traffic are recorded only within large cities and their neighborhoods. Meanwhile, in the rest of the country, these fluctuations are minimal.

In fact, precise evaluation of the data showed the changing dynamics of collisions within a day in certain months, and that actually no more than two peaks existed during the day: one in the morning and one in the evening (Fig. 5). The correlation coefficient of the morning collision peak is r2 = 0.71093 and for the evening peak r2 = 0.89073. Both correlations are significant, indicating that the hours of sunrise and sunset have a very important influence on WVRD (Fig. 6).
Fig. 5

Daily dynamics of roe deer collisions with vehicles during different months of 2013 and 2014 (n = 2010)

Fig. 6

Cyclical nature of roe deer collisions with vehicles in spring (a) and autumn (b) during their annual activity cycle

The morning WVRD peak is not as intense as the evening one. Furthermore, in the autumn and early winter the morning peak is more noticeable than in the months of spring and July and August. Roe deer mate in autumn and early winter, and their activity increases in evening hours. On the other hand, July–August is the holiday season in Lithuania, and although we do not have data on daily changes in Lithuanian traffic intensity during these months, it is reasonable to assume that, especially in the morning hours, the traffic is not that intense.

Throughout the year, monthly WVRD morning peaks largely begin 2 h before sunrise and last until sunrise (Fig. 7).
Fig. 7

The influence of sunrise (a) and sunset (b) on WVRD

The evening collision peaks of WVRD show a similar frequency to morning ones with the exception that they begin half an hour before sunset and last up to 2 h afterwards.

About 75% of all car accidents in Lithuania occurred during the daytime in 2013 and 2014 (Fig. 8), and only about 25% in the dark. This is apparently related to the behavior of drivers as there are fewer vehicles on the road, the average speed is lower in the dark, and cars and pedestrians are more visible due to the use of car lights and reflectors, etc.
Fig. 8

The impact of daytime and nighttime on the proportion of car accidents and WVRD in Lithuania in 2013 and 2014 (n = 2010)

The distribution of WVRD in the nighttime and daytime shows a completely different pattern: 72% WVRD were recorded at night and only about 28% from sunrise to sunset. This could be explained by the fact that during the day, at least in the case of a potential WVRD, drivers are better able to observe animals in advance and, if necessary, reduce their speed or even stop their car. However, during the night on darker sections of road, roe deer may be invisible due to their color and because they are outside the direct field of car lights. Unfortunately, in such cases, the driver’s reaction time is often too slow and a collision occurs. In addition, roe deer may add to the chance of a collision because of their unpredictable behavior under stress, and they may become harder to avoid.


This study confirmed that WVRD is determined by behavioral rhythms of roe deer and changes in this during the year and with time of day. This work has only considered WVRD associated with roe deer density, but, if collision trends were to be analyzed including sex and age of roe deer, it is possible that certain deviations from these results would be obtained for different groups. Unfortunately, there are no relevant records in Lithuania on roe deer age and gender.

The number of vehicle collisions with roe deer showed an increase at certain times of the year. The first peak was in May, reaching 13.5% of all analyzed collisions, and 62.39% above the annual average (8.33%). The second peak of collisions was in November and December, at 11.44 and 10.80%, respectively. These trends are essentially the same as in other European countries (Langbein et al. 2011; Morelle et al. 2013; Schoon 2011; De Vries 2015).

Each month is characterized by morning and evening peaks in WVRD (Fig. 4). There are two evening peaks (Fig. 4), one at 1700 and and one at 2200 hours that can be linked to peaks in roe deer activity cycles in spring and the end of autumn/beginning of winter. The peak at 2200 hours is associated with the spring roe deer activity peak, and the peak at 1700 hours with that in October and November (Fig. 6). The morning peaks are more significant in late autumn and in winter, when the sunrise time is similar, and form a single peak. In the summer months, when the sun rises several hours earlier than at the end of autumn or in the winter, a second peak of morning accidents was expected. However, during the summer months there was a lower increase in WVRD that did not form a separate morning peak.

Monthly WVRD shows an increase in April and June, and a peak in May. This can be explained by the fact that roe deer are very active during this period. During estrus, roe deer leave their birthplace and adults create new territories (Pokorny 2006; Groot Bruinderink and Hazebroek 1996). Though roe deer are considered sedentary, remaining in their territories even when starving, in the spring mature males will defend their territory from other roebucks. The females, except during birth, do not have a permanent territory and thus remain where conditions for breeding and eating are best. During breeding, roe deer males, especially young ones, are aggressive (Worm 2014). In addition, if the density of animals is too high, young roebuck (1–2 years) are forced to retreat to less favorable areas. In order to avoid aggression or search for safe habitats, roe deer may cross roads and thus often end up in traffic collisions.

Another, smaller collision peak in Lithuania is observed in October and November, and has also been recorded in other European countries. This may be due to the fact that during this period young roe deer of at least 4–5 months of age become more active, more mobile and therefore more likely to become victims of a road accident. Moreover, during October and November the second estrus period occurs in roe deer that were not inseminated during the first estrus period of May-June. As a result, adult roe deer, especially those looking for or chasing other deer, often cross roads (De Vries 2015). In late autumn and early winter roe deer form herds in which total activity of the population increases. In autumn and winter, roe deer live in small groups that usually consist of a female with the last of the summer-born juveniles. Young males keep apart in groups of two or three, while older males are usually solitary. During the cold season, feeders in the forests create herds of dozens of animals. During this period, roe deer spend more time in search of food and thus migrate between different areas, and are more likely to be killed by collisions with vehicles. Leading roe deer of herds, which try to be the first to cross roads, have a greater chance of being involved in a collision.

Autumn hunting may also affect the occurrence of WVRD. In Lithuania roe deer hunting is carried out in late autumn–early winter, causing disturbance of animals and encouraging them to move away from safer areas. Hunting is carried out during daylight hours, thus there should be a rise in accidents involving roe deer at this time of day. However, this trend was not established in our study. It can be assumed that during daylight hours roe deer fleeing from hunters and crossing roads are easier to spot. Meanwhile, in the evening or at night, when they return to their territories, it is harder for drivers to spot them, which increases the possibility of WVRD.

Climatic factors also cannot be ignored, especially the effect of fog on the number of collisions. However, according to Lithuanian Hydrometeorological Service data, fog is least likely in May–July in most of Lithuania, thus there should not be a direct effect of fog on the number of collisions. In the autumn–winter period, when the probability of fog is highest, it may affect the number of collisions. However, the foggy season is much longer than the period of the measured increase in accidents. Thus, we cannot be sure that fog has a significant effect on the number of incidents, and additional research is required.

Annual fluctuations of WVRD in Lithuania identified in this study are essentially similar to those in some other European countries (Langbein et al. 2011; Rodríguez-Morales et al. 2013; De Vries 2015). Slight seasonal peaks of accident rate variation or shifts of this to one or the other half of a month have been recorded for different countries, and are probably associated with different climatic conditions, e.g. southern European countries record an earlier spring peak of WVRD, thus peaks are probably related to different climatic conditions, such as seen in June, which is associated with estrus (Diaz-Varela et al. 2011; Lagos et al. 2012). For northern European countries, including Lithuania, the two peaks converge.

WVRD is characterized by an annual peak and also daily peaks. Roe deer are more active after dark when the two peaks were observed: one in the evening and one at night (De Vries 2015). Roe deer are sensitive to human activities, which influence them directly or indirectly during the day, thus they avoid open spaces with human presence. They tend to migrate and forage in the dark (Putman 1997), and when this coincides with morning or evening increases in traffic an increase in the number of collisions with vehicles is seen. In addition, in the evening, when light is poor, it is difficult for drivers to clearly observe the road and migrating animals. As a result, most collisions with roe deer occur at night (Schoon 2011; Groot-Bruinderink and Hazebroek 1996).

This study showed that the time of year is very important for the prediction of WVRD peaks. Daily peaks are directly related to the change of sunrise and sunset time during the year, and therefore peaks at different times of the year may be significantly different.

In conclusion, season and time of the day are very important factors for the dynamics of WVRD. Therefore, to apply optimal preventive measures for WVRD, particularly those related to the improvement of traffic or restrictions of traffic intensity, more attention needs to be paid to environmental and road safety factors.


  1. Apollonio M, Andersen R, Putman R (2010) European ungulates and their management in the 21st century. Cambridge University Press, CambridgeGoogle Scholar
  2. Balčiauskas L (2009) Distribution of species-specific wildlife—vehicle accidents on Lithuanian roads, 2002–2007. Est J Ecol 58:157–168CrossRefGoogle Scholar
  3. Barthelmess EL, Brooks MS (2010) The influence of body-size and diet on road-kill trends in mammals. Biodivers Conserv 19:1611–1629CrossRefGoogle Scholar
  4. Cagnacci F, Focardi S, Heurich M, Stache A, Hewison AJM, Morellet N, Kjel-lander P, Linnell JDC, Mysterud A, Neteler M, Delucchi L, Ossi F, Urbano F (2011) Partial migration in roe deer: migratory and resident tactics are endpoints of a behavioural gradient determined by ecological factors. Oikos 120:1790–1802CrossRefGoogle Scholar
  5. Cervinka J, Riegert J, Grill S, Salek M (2015) Large-scale evaluation of carnivore road mortality: the effect of landscape and local scale characteristics. Mamm Res 60:233–243CrossRefGoogle Scholar
  6. Clevenger AP, Chruszcz B, Gunson KE (2001) Highway mitigation fencing reduces wildlife–vehicle collisions. Wildl Soc Bull 29:646–653Google Scholar
  7. Conover M, Pitt W, Kessler K, DuBow T, Sanborn W (1995) Review of human injuries, illnesses, and economic losses caused by wildlife in the United States. Wildl Soc Bull 23:407–414Google Scholar
  8. Cserkész T, Farkas J, Ottlecz B (2012) Complex wildlife–vehicle collision research on the SMMC’s highway network. In: Report to Hungarian State Motorway Management Company, Eötvös University Road Ecological Work Group, BudapestGoogle Scholar
  9. Cserkész T, Ottlecz B, Cserkész-Nagy Á, Farkas J (2013) Interchange as the main factor determining wildlife–vehicle collision hotspots on the fenced highways: spatial analysis and applications. Eur J Wildl Res 59:587–597CrossRefGoogle Scholar
  10. Dal Compare L, Sturaro E, Cocca G, Ramanzin M (2007) An analysis of roe deer (Capreolus capreolus) traffic collisions in the Belluno province, eastern Italian Alps. Ital J Anim Sci 6:848–850Google Scholar
  11. Danks ZD, Porter WF (2010) Temporal, Spatial, and Landscape Habitat Characteristics of Moose– Vehicle Collisions in Western Maine. J Wildl Manage 74:1229–1241. CrossRefGoogle Scholar
  12. Damarad T, Bekker GJ (2003) COST 341—habitat fragmentation due to transportation infrastructure: findings of the COST Action 341. Office for Official Publications of the European Communities, LuxembourgGoogle Scholar
  13. De Vries M (2015) Road kills of roe deer (Capreolus capreolus) in the Netherlands: assessment of impacts and mitigation measures. In: MSc internship report, Copernicus Institute of Sustainable Development, Utrecht University, UtrechtGoogle Scholar
  14. Diaz-Varela ER, Vazquez-Gonzalez I, Marey-Pérez MF, Álvarez-López CJ (2011) Assessing methods of mitigating wildlife–vehicle collisions by accident characterization and spatial analysis. Transp Res Part D Transp Environ 16:281–287CrossRefGoogle Scholar
  15. Evink G (2002) Interaction between roadways and wildlife ecology: a synthesis of highway practice. In: National Cooperative Highway Research Program Synthesis 305, Transportation Research Board, Washington, DCGoogle Scholar
  16. Fahrig L, Rytwinski T (2009) Effects of roads on animal abundance: an empirical review and synthesis. Ecol Soc 14:1–21CrossRefGoogle Scholar
  17. Findlay CS, Bourdages J (2000) Response time of wetland biodiversity to road construction on adjacent lands. Conserv Biol 14:86–94CrossRefGoogle Scholar
  18. Forman RTT, Sperling D, Bissonette JA, Clevenger AP, Cutshall CD, Dale VH, Fahrig L, France R, Goldman CR, Heanue K, Jones JA, Swanson FJ, Turrentine T, Winter TC (2003) Road ecology: science and solutions. Island Press, Washington, DCGoogle Scholar
  19. Gibbs JP, Shriver G (2002) Estimating the effects of road mortality on turtle populations. Conserv Biol 16:1647–1652CrossRefGoogle Scholar
  20. Gomes L, Grilo C, Silva C, Mira A (2009) Identification methods and deterministic factors of owl roadkill hotspot locations in Mediterranean landscapes. Ecol Res 24:355–370. CrossRefGoogle Scholar
  21. Groot-Bruinderink GWTA, Hazebroek E (1996) Ungulate traffic collisions in Europe. Conserv Biol 10:1059–1067CrossRefGoogle Scholar
  22. Gunson KE, Clevenger AP, Ford AT, Bissonette JA, Hardy AA (2009) Comparison of data sets varying in spatial accuracy used to predict the occurrence of wildlife–vehicle collisions. Environ Manage 44:268–277CrossRefPubMedGoogle Scholar
  23. Haigh A (2012) Annual patterns of mammalian mortality on Irish roads. Hystrix Ital J Mammal 23:58–66Google Scholar
  24. Hubbard MW, Danielson BJ, Schmitz RA (2000) Factors influencing the location of deer-vehicle accidents in Iowa. J Wildl Manage 64:707–713CrossRefGoogle Scholar
  25. Huijser MP, Mc Gowen P, Fuller J, Hardy A, Kociolek A, Clevenger AP, Smith D, Ament R (2007) Wildlife–vehicle collision reduction study. Report to US Congress. US Department of Transportation, Federal Highway Administration, Washington, DCGoogle Scholar
  26. Joyce TL, Mahoney SP (2001) Spatial and temporal distributions of moose–vehicle collisions in Newfoundland. Wildl Soc Bull 29:281–291Google Scholar
  27. Lagos L, Picos J, Valero E (2012) Temporal pattern of wild ungulate-related traffic accidents in northwest Spain. Eur J Wildl Res 58:661–668CrossRefGoogle Scholar
  28. Langbein J, Putman R, Pokorny B (2011) Traffic collisions involving deer and other ungulates in Europe and available measures for mitigation. Ungulate management in Europe: problems and practices. Cambridge University Press, CambridgeGoogle Scholar
  29. Lithuanian Road Administration (2015) Statistics of fatal and injury road accidents in Lithuania, 2011–2014. Ministry of Transport and Communications of the Republic of LithuaniaGoogle Scholar
  30. Litvaitis JA, Tash JP (2008) An approach to understanding wildlife vehicle collisions. Environ Manage 42:688–697CrossRefPubMedGoogle Scholar
  31. Loro M, Ortega E, Arce RM, Geneletti D (2016) Assessing landscape resistance to roe deer dispersal using fuzzy set theory and multicriteria analysis: a case study in Central Spain. Landsc Ecol Eng 12:41–60CrossRefGoogle Scholar
  32. Malo JE, Suarez F, Diez A (2004) Can we mitigate animal-vehicle accidents using predictive models? J Appl Ecol 41:701–710CrossRefGoogle Scholar
  33. Morelle K, Lehaire F, Lejeune P (2013) Spatio-temporal patterns of wildlife–vehicle collisions in a region with a high-density road network. Nat Conserv 5:53–73CrossRefGoogle Scholar
  34. Mysterud A (1999) Seasonal migration pattern and home range of roe deer (Capreolus capreolus) in an altitudinal gradient in southern Norway. J Zool 247:479–486CrossRefGoogle Scholar
  35. Oškinis V, Ignatavičius G, Vilutienė V (2013) An evaluation of wildlife-vehicle collision pattern and associated mitigation strategies in Lithuania. Environ Eng Manage J 12:2323–2330CrossRefGoogle Scholar
  36. Pokorny B (2006) Roe deer-vehicle collisions in Slovenia: situation, mitigation strategy and counter measures. Vet Arhiv 76:177–187Google Scholar
  37. Putman RJ (1997) Deer and road traffic accidents: options for management. J Environ Manage 51:43–57CrossRefGoogle Scholar
  38. Putman RJ, Langbein J, Staines BW (2004) Deer and road traffic accidents: a review of mitigation measures—costs and cost-effectiveness. Rep Deer Comm Scotl 3:1–6Google Scholar
  39. Ramp D, Caldell J, Edwards KA (2005) Modelling of wildlife fatality hotspots along the Snowy Mountain Highway in New South Wales, Australia. Biol Conserv 126:474–490. CrossRefGoogle Scholar
  40. Rodríguez-Morales B, Díaz-Varela ER, Marey-Pérez MF (2013) Spatiotemporal analysis of vehicle collisions involving wild boar and roe deer in NW Spain. Accid Anal Prev 60:121–133CrossRefPubMedGoogle Scholar
  41. Romin LA, Bissonette JA (1996) Deer–vehicle collisions: status of state monitoring activities and mitigation efforts. Wildl Soc Bull 24:276–283Google Scholar
  42. Rost GR, Bailey JA (1979) Distribution of mule deer and elk in relation to roads. J Wildl Manage 43:634–641CrossRefGoogle Scholar
  43. Schoon R (2011) Pas op: aanrijdingen overstekend met reeën wild! In Rtrecht. Terra salica. Bur Faunabeheer 5:14–48Google Scholar
  44. Seiler A (2004) Trends and spatial patterns in ungulate–vehicle collisions in Sweden. Wildl Biol 10:301–313CrossRefGoogle Scholar
  45. Seiler A (2005) Predicting locations of moose–vehicle collisions in Sweden. J Appl Ecol 42:371–382CrossRefGoogle Scholar
  46. Seiler S, Helldin JO, Seiler C (2004) Road mortality in Swedish mammals: results of a drivers’ questionnaire. Wildl Biol 10:183–191Google Scholar
  47. Seo C, Thorne JH, Choi T, Kwon H, Park CH (2015) Disentangling roadkill: the influence of landscape and season on cumulative vertebrate mortality in South Korea. Landsc Ecol Eng 11:87–99CrossRefGoogle Scholar
  48. Smith-Patten B, Patten M (2008) Diversity, seasonality, and context of mammalian roadkills in the southern Great Plains. Environ Manage 41:844–852CrossRefPubMedGoogle Scholar
  49. Snow NP, Williams DM, Porter WF (2014) A landscape-based approach for delineating hotspots of wildlife–vehicle collisions. Landsc Ecol 29:817–829CrossRefGoogle Scholar
  50. Spellerberg IF (1998) Ecological effects of roads and traffic: a literature review. Glob Ecol Biogeogr Lett 7:317–333CrossRefGoogle Scholar
  51. Steiner W, Leisch F, Hackländer K (2014) A review on the temporal pattern of deer-vehicle accidents: impact of seasonal, diurnal and lunar effects in cervids. Accid Anal Prev 66:168–181CrossRefPubMedGoogle Scholar
  52. Worm B (2014) Het ree: observeren en herkennen. IPC groene ruimte, ArnhemGoogle Scholar

Copyright information

© International Consortium of Landscape and Ecological Engineering and Springer Japan KK, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Vilniaus UniversitetasVilniusLithuania

Personalised recommendations