ArticlePDF Available

The effect of COVID-19 movement restriction on Korean expressway wildlife− vehicle collisions

Authors:
Global Ecology and Conservation 51 (2024) e02949
Available online 10 April 2024
2351-9894/© 2024 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Short communication
The effect of COVID-19 movement restriction on Korean
expressway wildlifevehicle collisions
Hyomin Park , Sangdon Lee
*
Department of Environmental Science & Engineering, College of Engineering, Ewha Womans University, Seoul 03760, South Korea
ARTICLE INFO
Keywords:
Wildlife management
Conservation strategy
Roadkill
Road ecology
Hotspots
ICT-Environmental Impact Assessment
ABSTRACT
The construction of roads obstructs animal movement, which directly results in fatal wild-
lifevehicle collisions (WVCs). Owing to the worldwide COVID-19 outbreak in 2020, movement
restrictions were imposed by the Korean government between February and March 2020, which
represented the beginning of the COVID-19 pandemic. During this period, this study examined the
effects of reduced trafc volume on WVCs. The study selected the DangjinYeongdeok, Jungang,
and Jungbu Expressways as study sites since they have the highest occurrence of WVCs in Korea.
The sections of these expressways with the highest densities of WVCs were denoted as WVC
hotspots. We categorized the period from February to March 2020 as ‘strict pandemic. In
addition, the period from February to March 20152019 was categorized as pre-strict
pandemic, and between April 2020 and December 2020 was classied as the pandemic period.
We analyzed the relationship between WVC and trafc volume for each period in the designated
hotspots. As a result, there were statistically signicant differences in WVCs and trafc volumes
across all routes for different periods. WVCs and trafc volume showed a strong negative cor-
relation during the strict pandemic period. WVC per trafc volume was positively correlated
between the strict pandemic period and the pandemic period, while it was negatively correlated
between pre-strict pandemic and strict pandemic periods. The deceleration in WVC continued
after all routes were out of strict pandemic. This study shows that WVC and trafc volume are
negatively correlated and that a strong, albeit temporary, trafc reduction can help reduce WVC.
Therefore, the overall quantitative conclusion of this study on the interaction can serve as a
valuable reference to develop practical strategies for preventing WVCs.
1. Introduction
The increasing number of new roadways has become a considerable threat to wildlife globally (Primack et al., 2021). With the
continuous expansion of towns and heavier trafc between cities, newly constructed roads have increasingly fragmented wildlife
habitats (Forman et al., 2003), eventually, isolating, and conning wildlife. Roadways obstruct animal movement, inuence species
diversity, and directly result in fatal wildlifevehicle collisions (WVCs) (Forman et al., 2003; Laurance et al., 2009; Van der Ree et al.,
2015). Due to the use of wider lanes and adjacency to forests, expressways have higher trafc speeds and volumes than other road
types, thereby making them more prone to WVCs, and thus, a signicant threat to wildlife (Park et al., 2021). This problem is
exemplied in South Korea, where persistent industrial growth necessitates the construction of additional expressways, which
* Corresponding author.
E-mail address: lsd@ewha.ac.kr (S. Lee).
Contents lists available at ScienceDirect
Global Ecology and Conservation
journal homepage: www.elsevier.com/locate/gecco
https://doi.org/10.1016/j.gecco.2024.e02949
Received 30 October 2023; Received in revised form 8 April 2024; Accepted 10 April 2024
Global Ecology and Conservation 51 (2024) e02949
2
increases the threat to wildlife.
Wildlifevehicle collisions (WVCs) are a signicant threat to many species, causing nancial loss, and posing a serious risk to driver
safety (Laube et al., 2023). Since WVCs have caused severe problems worldwide, researchers have actively investigated various
associated issues, including the socioeconomic costs of WVCs (Huijser et al., 2009) and mitigation measures (Lester, 2015). Several
Fig. 1. A map of study sites in South Korea showing sections of the surveyed expressways.
H. Park and S. Lee
Global Ecology and Conservation 51 (2024) e02949
3
factors affect the incidence of WVCs, including spatial characteristics (Park et al., 2021), trafc volume changes (Bencin et al., 2019),
and seasonal migrations of wildlife (Main and Allen, 2002). Among these, trafc volume is a critical predictor of wildlife roadkill rates
(Taylor and Goldingay, 2010; Visintin et al., 2017; Pagany, 2020). The presence of roadways generally affects wildlife (Fahrig and
Rytwinski, 2009), and trafc volume is strongly correlated to the frequency and potential risks of WVCs (Litvaitis and Tash, 2008). In
South Korea, the escalating incidence of WVCs has also become a nationwide problem, thereby prompting further research. However,
the relationship between WVCs on expressways and trafc volume in the country requires further investigation.
Owing to the worldwide COVID-19 outbreak in 2020, movement restrictions were imposed by the Korean government between
February and March 2020 to curb the spread of the virusthe beginning of the COVID-19 pandemic. These restrictions signicantly
reduced human activities (Primack et al., 2021), thereby providing opportunities to observe the human inuence on WVCs (Driessen,
2021). Several studies have reported that changes in human activity during the pandemic altered the behavior of wild animals
(Manenti et al., 2020; Rutz et al., 2020). Thus, during this period, this study examined the effects of reduced trafc volume on WVCs.
Based on nationwide WVC data from 2015 to 2020, this study selected the DangjinYeongdeok, Jungang, and Jungbu Expressways
as study sites since they have the highest occurrence of WVCs in Korea. The sections of these expressways with the highest densities of
WVCs were denoted as WVC hotspots, meaning that in these hotspots, the risk of a collision was statistically higher than in any other
section of the expressway (Seiler et al., 2016). The identication of the hotspot locations is important for the effective application of
mitigation measures (Bíl et al., 2016). Therefore, we analyzed the relationship between the WVCs and trafc volume in these hotspots.
To our knowledge, there has been no previous study conducted in Korea on the changes in the occurrence of WVCs during the
pandemic period. Therefore, this study seeks to establish important basic data necessary for the management plan of WVC occurrences
on expressways.
2. Materials and methods
2.1. Study sites
To choose the appropriate areas for this study, we analyzed the annual numbers of WVCs per kilometer of expressway in Korea
between 2015 and 2019. Expressways under 100 km in length were excluded owing to suggestions that the WVC rate for short
roadways is often overestimated (Lee et al., 2014). Based on this analysis, DangjinYeongdeok Expressway (DYE), Jungang
Expressway (JAE), and Jungbu Expressway (JBE) were included in the study (Fig. 1, Appendix 1). All expressways selected in this
study have maximum and minimum speed limits of 110 km/h and 50 km/h, respectively.
The DYE (route number 30) has a total length of 278.6 km and connects the eastern and western regions of South Korea, thereby
traversing the central inland regions of Chungcheongnam-do, Chungcheongbuk-do, and Gyeongsangbuk-do. Its rst section opened in
November 2007, with subsequent additional sections being completed in stages thereafter.
Next, the JAE (route number 55) passes through Gyeongsangbuk-do, Chungcheongbuk-do, and Gangwon-do, cutting across the
eastern inland area of the country, north to south. These regions are enclosed by the Taebaek and Sobaek Mountain Ranges. The JAE is
288.9 km long, while the rst section of the expressway opened to trafc in December 1994.
The JBE (route number 35) connects Tongyeong-si, Gyeongsangnam-do, Hanam-si, and Gyeonggi-do. Initially, it was 117.2 km
long between Hanam-si, Gyeonggi-do, Nami-myeon, Cheongwon-gun, and Chungcheongbuk-do. The TongyeongDaejeon Expressway
was built and joined to the original JBE; therefore, the current total length of the expressway is now 332.5 km, and it has been renamed
the TongyeongDaejeon Jungbu Expressway. Notably, it was rst opened in December 1987.
2.2. Data collection
The analysis was conducted based on the nationwide data for trafc volume and WVCs provided by the Korea Expressway Cor-
poration (KEC) (http://data.ex.co.kr) and spanned six years from January 2015 to December 2020. WVC investigation methods are
investigated in accordance with the ‘Guidelines for Roadkill Investigation and Managementset by the government as administrative
rules. WVC data are collected by the Safety Patrol of the KEC, which records the number of carcasses discovered along with other
related information during their expressway patrol, which occurs three times a day. Carcasses found are photographed by safety
patrols and sent to the ‘Animal Roadkill Information System, and the National Institute of Ecology uses the photos to identify the
animal species. The recorded items are as follows: the date; the jurisdictional authority; the route name; the distance from the
expressway starting point; the direction; the species of the animal killed in the WVC; the longitude and latitude of the WVC location.
Following the analysis, the carcass was removed from the site to prevent the formation of duplicate records. WVC investigations were
conducted in the same way as before, even when government movement restrictions due to COVID-19 were in place.
Toll collection system (TCS) data from the KEC was used to measure the trafc volume. TCS data contain detailed information
regarding the vehicles entering and leaving the expressways via the toll booths. Such data facilitates the easy and precise identication
of expressway usage patterns.
2.3. Data analysis
2.3.1. WVC hotspot modelling
Kernel density estimations (KDEs) were used to identify specic WVC hotspots in each expressway (Eq. 1). KDE calculates the
spatial density based on the distribution of individual points within a specic search radius (bandwidth) and generates kernel function
H. Park and S. Lee
Global Ecology and Conservation 51 (2024) e02949
4
K. In the estimation process, the larger the bandwidth, the greater the hotspot area (Da Silva et al., 2022). Thus, the spatial relevance
decreases as the distance between the center and an individual point increases.
f(x,y) = 1
nh2=
n
i=1
K(di
h)(1)
where f (x, y) is the density estimation at (x, y), n is the number of WVCs, h is the bandwidth, di is the location (x, y), and i is the
distance between occurrence points (Eq.1). A Gaussian kernel function is used to determine K (Eq. 2):
K(u) = 1

2
π
e1
2u2(2)
where u is the ratio of the distance at location (x, y) within the bandwidth.
To identify specic WVC hotspots in the expressways, the WVC density was estimated using KDEs based on the WVC data from
2015 to 2019. The WVC data used for the KDEs was based on the longitude and latitude of the WVC occurrence points to obtain
cumulative data for each WVC occurrence point. KDE was then used to estimate the WVC density for the entire route, which was then
categorized into ve sections (I, II, III, IV, and V) with equal ranges for WVC density using the equal interval method: Section I with a
WVC density ranging from 100% to 80%; Section II from 80% to 60%; Section III from 60% to 40%; Section IV from 40% to 20%;
Section V from 20% to 0%. Section I covers the areas with extremely high WVC occurrence densities (top 20%). By contrast, Section V
covers the areas with extremely low or zero WVC occurrence density (bottom 20%) (Park et al., 2021). Therefore, we classied Section
I as the highest WVC occurrence, Section II as high WVC occurrence, Section III as medium, Section IV as low, and Section V as very
low.
We identied the Section I regions with the highest WVC density in DYE, JAE, and JBE. We extracted the roads passing between the
toll booths located within the Section I as hotspot areas. This is to measure the trafc volume of hotspots accurately. Subsequently, the
trafc volume and WVCs within the hotspot zone between 2015 and 2020 were analyzed by period (before, during and after
restrictions).
2.3.2. Statistical Analysis
Owing to the severe risk of COVID-19, governments implemented strict restrictions during the initial stage of the pandemic to
contain the spread of the virus (Rilett et al., 2021). A wide range of studies has been conducted worldwide that have focused on the
early stages of the pandemic. Similarly, in Korea, strict regulations were implemented, particularly between February and March 2020.
Thus, this study focused on the initial period of the pandemic in Korea. We categorized the period from February to March 2020, when
strong measures were implemented to contain COVID-19 infections, as a ‘strict pandemic. In addition, the period from February to
March between 2015 and 2019 was categorized as pre-strict pandemic, and between April 2020 and December 2020, when strong
measures were lifted, was categorized as the pandemic period.
We conducted an analysis of ANOVA by a period (strict pandemic, pre-pandemic, and pandemic) on WVC data from all routes to
determine whether there were statistically signicant differences in WVC incidence due to the COVID-19 pandemic. Paired t-test were
also used to identify periods with statistically signicant differences in WVC. Additionally, a paired t-test was conducted for each
period of each route using WVC data. This derived the period when differences in WVC occurrence for each route were observed. In
addition, we derived the trafc volume for each route to determine whether there is a statistically signicant difference in the change
in trafc volume due to the COVID-19 pandemic by period (DYE, JAE, JBE). Then, we derived trafc volumes by period (strict
pandemic, pre-pandemic, pandemic) and conducted an ANOVA analysis.
Meanwhile, we analyzed the correlation between WVC and trafc volume during the strict pandemic using Pearsons correlation
coefcient. To examine the relationship between the changes in trafc volume and WVCs during the COVID-19 pandemic, we obtained
monthly data on the trafc volume and WVCs for the study routes for each period (strict pandemic, pre-strict pandemic, and
pandemic). To analyze the changes in WVCs per trafc volume, the percentage of WVCs per trafc volume was derived using monthly
data on trafc volume and WVCs. Then, a paired t-test was used to examine the difference in the percentage change of the percentage of
WVCs per trafc volume during each period. The correlation between WVC and trafc volume was analyzed for each period using the
Pearson correlation coefcient. All statistical analyses were performed using R (ver. 4.1.2).
3. Results and discussion
3.1. WVC hotspots
The roads passing between the toll booths located within the Section I derived using the KDE method were identied as hotspots.
The WVC hotspots for each study site were identied as follows: DYE: GongjuNamsejong IC; JAE: HoengseongNamwonju IC; JBE:
Gonjiam-Seoicheon IC (Fig. 2).
WVCs per kilometer have been decreasing steadily since 2015 throughout the total length of the DYE and JAE, whereas they have
increased since 2017 for the JBE. This trend was the same in the hotspots for each expressway. Despite this trend, the WVC/km of the
hotspots was observed to be higher than that of the WVC/km for the total length of the expressways. The percentage of WVCs that
occurred at each hotspot relative to the entire expressway has also grown (Table 1).
H. Park and S. Lee
Global Ecology and Conservation 51 (2024) e02949
5
Therefore, the WVCs in the hotspots have remained high, so we analyzed the WVC characteristics of each hotspot. We analyzed the
WVCs on the DYE, JAE, and JBE from 2015 to 2019 and our ndings revealed that water deer (Hydropotes inermis) were the most
frequently involved in collisions (Table 2), while April, May, and June were the months that exhibited the highest number of WVCs.
This suggests that WVCs exhibit seasonality, which is associated with the behavioral characteristics of water deer. According to Garrah
et al., (2015), WVCs are strongly seasonal, and this seasonality is extensively species-specic. The Korean water deer have a high
population density and are widely distributed throughout the country. During their breeding season in spring, they tend to expand
their home range (Park and Lee, 2013). Thus, during this time of year, young deer are more likely to be involved in vehicle accidents
compared to other seasons, thereby contributing to the high rate of WVCs in spring.
3.2. The effect of the COVID-19 movement restriction on WVCs
To examine the impact on WVCs by the pandemic, we analyzed the relative change in the monthly number of WVCs during the
pandemic period versus that during the pre-pandemic period, focusing specically on the period from February to March 2020, when
the Korean government implemented movement restrictions. The WVCs recorded on all studied routes were analyzed by dividing them
into pre-COVID (20152019), strict pandemic (FebruaryMarch 2020), and pandemic (AprilDecember 2020). Thereafter, it was
conrmed that during the strict pandemic period, the number of WVCs increased compared to pre-COVID, and during the pandemic,
WVCs decreased compared to pre-COVID (Table 3). An ANOVA test was conducted to evaluate the differences in WVCs by period for all
Fig. 2. Categorization of the Wildlife density into ve sections (I, II, III, IV, V) for the (1) DangjinYeongdeok Expressway (DYE), (2) Jungang
Expressway (JAE), and (3) Jungbu Expressway (JBE), using kernel density estimations (KDEs). Sections with a high WVC density are marked in red
(I), while those with a low WVC density are marked in green (V). (I: 10080%, II: 8060%, III: 6040%, IV: 4020%, and V: 200%).
H. Park and S. Lee
Global Ecology and Conservation 51 (2024) e02949
6
study routes. As a result, the difference in the number of WVCs according to the period (pre-COVID, strict pandemic, and pandemic)
was statistically signicant (df=2, F value=4.772, p =0.012). Paired t-test were used to identify periods with statistically signicant
differences in WVC. There was a statistically signicant difference in the WVCs between the strict pandemic (FebruaryMarch 2020)
and the pandemic period (AprilDecember 2020) (t=2.988, df=5, p =0.031).
After the strict pandemic, the WVCs continued to decrease on all routes in this study compared to the previous year (Fig. 3). A study
from northwest Spain demonstrated results similar to our study, whereby an increase in WVCs was observed during the early months of
2020 before becoming reduced following the enforced COVID-19 lockdowns. Moreover, this trend persisted following the end of the
COVID-19 lockdown. According to that study, the number of WVCs was 30.22% lower in 2020 than in 2019, with a considerable
decline of 66.77% observed during the strictest lockdown period (García-Martínez-de-Alb´
eniz et al., 2022). This study also revealed a
signicant reduction in WVCs on the DYE, JAE, and JBE during 2020 (JanDec) compared to 2019 (JanDec), with declines of 50.09%,
42.55%, and 57.78%, respectively.
Meanwhile, the WVCs recorded for studied routes were analyzed by dividing them into pre-COVID (20152019), strict pandemic
(FebruaryMarch 2020), and pandemic (AprilDecember 2020). On DYE and JAE, the number of WVCs increased during the strict
pandemic (February to March 2020) compared to the pre-pandemic period. Specically, the DYE exhibited a 66.67% and 25% in-
crease, and the JAE exhibited a 100% and 42.86% increase. However, during this period, the number of WVCs on the JBE signicantly
decreased to 0. The number of WVCs showed a decreasing trend during the pandemic period (April to December 2020), compared to
the pre-COVID period, except for specic months, which were October for the DYE, November for the JAE, and August and December
for the JBE (Fig. 3).
Therefore, the results from the paired t-test to compare WVCs between the pre-strict pandemic and strict pandemic periods for each
study route revealed a statistically signicant difference in WVCs for the DYE (t=2.360, df=10, p=0.040) and JAE (t=2.562, df=10,
p=0.028). Contrastingly, for the JBE, the difference in WVCs during the same period was not statistically signicant (t=1.854, df=10,
p=0.093).
Numerous studies have investigated the potential impact of reduced human activities during various periods of the COVID-19
pandemic on worldwide incidences of WVCs. In this study, WVCs increased during the strict pandemic period on the DYE and JAE.
In Scotland, WVCs increased during the COVID-19 lockdown (Bíl et al., 2021), and in some states in the US, WVCs increased during the
pandemic (Abraham and Mumma, 2021). This suggests that wildlife may have increased their use of roads and areas near roads during
the pandemic in response to reduced trafc (Rutz et al., 2020).
While in this study, WVCs decreased during the strict pandemic period on the JBE. A study conducted in several US states reported a
decline in WVCs of 34% during the pandemic period, while some states showed a signicant relationship between the implementation
of COVID-19 restrictions and the change in WVCs (Shilling et al., 2021). Additionally, there was a 48% decrease in WVCs in Tasmania,
Australia, during the pandemic, with statistically signicant changes observed between WVCs during the pre-pandemic and pandemic
periods. Furthermore, a study conducted across 11 European countries revealed a 40% decrease in WVCs during the pandemic period
in Spain, Israel, Estonia, and the Czech Republic compared to previous periods (Driessen, 2021).
In this study, we found that WVCs differed by the duration of travel restrictions due to COVID-19, and the differences in WVCs by
duration were statistically signicant for each study route. Therefore, we can conrm that travel restrictions due to COVID-19 affect
the occurrence of WVCs.
3.3. Relationship between trafc volume and WVCs
To analyze the change in trafc volume due to the COVID-19 restriction, we compared the change in trafc volume in 2020 to the
average trafc volume from 2015 to 2019. We found that compared to pre-COVID-19, trafc volume on DYE and JAE decreased during
the COVID-19 restriction, while trafc volume on JBE increased. The decrease in trafc volume on JAE has been maintained even after
Table 1
Wildlifevehicle collision (WVC) analysis results in each study site and hotspot: WVC occurs per 1 km on the entire route, WVC occurs per 1 km in
each hotspot, and percentage of WVCs occurring at hotspots relative to total routes.
Year Entire expressway (WVC/km) Hotspot (WVC/km) WVC rate in hotspots (%)
DangjinYeongdeok Expressway
(DYE)
2015 2.83 4.24 17.96
2016 1.92 4.1 25.63
2017 1.06 2.9 21.38
2018 0.79 3.03 29.82
2019 0.85 2.72 24.9
Jungang Expressway
(JAE)
2015 0.35 4.74 10.86
2016 0.22 2.78 12.33
2017 0.18 1.76 15.38
2018 0.16 2.75 19.3
2019 0.16 2.33 17.54
Jungbu Expressway
(JBE)
2015 0.1 1.95 22.12
2016 0.08 1.67 17.52
2017 0.08 1.56 17.69
2018 0.09 1.37 19.9
2019 0.14 2.19 20.41
H. Park and S. Lee
Global Ecology and Conservation 51 (2024) e02949
7
Table 2
The status of wildlife involved in wildlifevehicle collisions (WVCs) in each hotspot zone.
Species DYE JAE JBE
2015 2016 2017 2018 2019 2015 2016 2017 2018 2019 2015 2016 2017 2018 2019
Water deer
(Hydropotes inermis)
91 88 59 66 57 92 55 41 40 40 33 24 27 30 40
Raccoon dog
(Nyctereutes procyonoides)
3 2 2 1 2 8 5 3 5 5 1 1 1
Wild boar
(Sus scrofa)
1 1 1 2 6 1 1 4
eopard cat
(Prionailurus bengalensis)
1 1
Asian badger
(Meles leucurus)
1 2 2 1 1 1
Korean hare
(Lepus coreanus)
2
Others 1
Total 95 92 65 68 61 101 63 53 45 47 34 26 27 30 45
H. Park and S. Lee
Global Ecology and Conservation 51 (2024) e02949
8
the restrictions were lifted (Fig. 4).
Therefore, we analyzed the trafc volume of each route during the COVID-19 restrictions by dividing it into pre-strict pandemic
(February March 20152019), strict pandemic (February-March 2020), and pandemic (April-December 2020) (Fig. 5). The results
show that trafc volume on DYE and JAE decreased signicantly during the strict pandemic compared to pre-strict pandemic trafc
volume. In contrast, after the strict pandemic, trafc volume increased signicantly. For JBE, trafc volume during the strict pandemic
tended to be higher than the pre-strict pandemic, and trafc volume continued to increase during the pandemic.
Table 3
Monthly WVC average across the entire study route. The gray blocks represent the period when strict restrictions began in Korea during the COVID-
19 pandemic (FebruaryMarch 2020).
20152019 2020
month average variance average variance
Jan 1.867 1.310 2.333 0.943
Feb 0.800 0.980 1.333 0.943
Mar 1.067 1.236 1.333 0.943
Apr 4.667 2.936 1.667 1.247
May 17.533 9.142 6.000 0.816
Jun 13.667 5.861 4.000 0.816
Jul 5.533 3.964 2.000 0.816
Aug 2.467 2.306 1.667 1.247
Sep 1.867 1.454 0.667 0.471
Oct 2.000 1.461 1.333 0.471
Nov 2.667 1.776 1.000 1.414
Dec 2.667 2.119 1.333 1.247
Fig. 3. Monthly percentage change in wildlifevehicle collisions (WVCs) from the pre-pandemic period to the pandemic period on the Dang-
jinYeongdeok Expressway (DYE), Jungang Expressway (JAE), and Jungbu Expressway (JBE). The area within the black dotted lines refers to the
period from February to April 2020, when strict COVID-19 restrictions were implemented to prevent the spread of the virus.
H. Park and S. Lee
Global Ecology and Conservation 51 (2024) e02949
9
In the case of JAE, trafc volume has declined during and after the strict pandemic. This is due to the continuous dispersion of
vehicles due to the opening of the Jungbu Naeryuk Expressway near JAE. For this reason, it can be seen that the JAE route shows a
decreasing trafc volume trend every year (Appendix 2). On the other hand, it can be seen that the JBE has seen a decrease in the peak
value of trafc volume during the strict pandemic but an increase in the median and minimum values. The JBE is an expressway
connecting the capital region, and the hotspot section is located close to Seoul, which tends to show a signicant increase in trafc
volume every year (Appendix 2). In addition, the JBE is an expressway with high freight trafc volume, with more than 30% of the
trafc volume being trucks (https://stat.molit.go.kr/). While human movement has decreased during the COVID-19 lockdown, freight
transportation has expanded globally. JBE is also an expressway with a high proportion of freight vehicles, which may explain the
increase in trafc volume on JBE during this period.
Therefore, we used ANOVA analysis to determine if there was a statistically signicant difference in the trafc volume of each route
by period. As a result, we found that the difference in trafc volume by period was statistically signicant for DYE, JAE, and JBE routes
(DYE: df=2, F value=9.679, p=0.001**, JAE: df=2, F value=10.696, p=0.001**, JBE: df=2, F value=16.183, p=0.000***). There-
fore, the study found that COVID-19 restrictions impacted trafc volume.
To analyze the impact of trafc volume changes on WVCs due to movement restrictions during the COVID-19 pandemic, this study
analyzed the Pearson correlation between WVCs and trafc volume for all routes during the strict pandemic. The results showed a
strong negative correlation between WVCs and trafc volume during the strict pandemic (Pearson correlation coefcient =-0.873).
To analyze the inuence of trafc volume changes due to the COVID-19 pandemic on WVCs, our study calculated the WVCs by
trafc volume for each period over the entire route. This analysis found that the mean of the pre-strict pandemic was 0.329 (SD =
0.642), the strict pandemic was 0.541 (SD =0.675), and the pandemic was 0.499 (SD =0.784). Thus, the strict pandemic period
exhibited the highest mean as well as a greater range of change in WVC occurrences per trafc (Fig. 6). This is because trafc volume
and WVCs are strongly negatively correlated, with the decrease in trafc volume during the strict pandemic period leading to an
increase in WVCs.
Fig. 4. Monthly percentage changes in trafc volume from the pre-pandemic period to the pandemic period on the DangjinYeongdeok Expressway
(DYE), Jungang Expressway (JAE), and Jungbu Expressway (JBE). The area within the black dotted lines refers to the period from February to April
2020, when strict COVID restrictions were implemented to curb the spread of the virus.
H. Park and S. Lee
Global Ecology and Conservation 51 (2024) e02949
10
Fig. 5. Trafc volume from 2015 to 2020 on each route, broken down by pre-strict COVID (February-March 20152019), strict pandemic
(February-March 2020), and pandemic (April-December 2020) (Left is DangjinYeongdeok Expressway (DYE), center is Jungang Expressway (JAE)
and right is Jungbu Expressway (JBE).
Fig. 6. Wildlifevehicle collision (WVC) occurrence data per trafc volume by period for all study routes. (pandemic: April 2020 to December 2020,
Pre-strict pandemic: February to March between 2015 and 2019, strict pandemic: February to March 2020).
H. Park and S. Lee
Global Ecology and Conservation 51 (2024) e02949
11
Therefore, a paired t-test was employed to examine the statistical signicance of the mean difference in WVC occurrence per trafc
volume between periods. There was a statistically signicant difference in the incidence of WVCs per trafc volume between the strict
pandemic and the pandemic periods (t=4.2939, df=5, p =0.008). There is a positive correlation between the strict and pandemic
periods and a negative correlation between the pre-strict pandemic and the strict pandemic (Table 4).
During the COVID-19 pandemic, social distancing measures and associated restrictions considerably decreased human activities
worldwide. Several researchers predicted that the reduced trafc volume would lead to fewer WVCs (Primack et al., 2021; Coelho
et al., 2008; DAmico et al., 2015; Lin, 2016). A study on trafc volume and WVCs on the Anuran highway in Ontario, Canada, found a
negative relationship between the two variables. Saeki and Macdonald (2004) also discovered collisions in Japan and trafc volume
whereby WVCs occurred more frequently when trafc volume decreased.
During the pandemic period, certain sections of the expressways exhibited a higher incidence of WVCs than in any other period.
This can be attributed to wildlife behavioral characteristics, including breeding, foraging, dispersal patterns, and their preferred routes
for crossing in search of better habitats (Dean et al., 2019; Saint-Andrieux et al., 2020). In this study, the behavioral characteristics of
water deer are among the most critical factors inuencing WVCs on the DYE, JAE, and JBE routes. These collisions mostly occurred in
the spring, while winter and spring are the known breeding seasons for water deer. During this period, the water deer mainly used
wetlands and farmland; however, they can also expand their home range (Park and Lee, 2013). Thus, the lockdown period coincided
with the water deers breeding season, thereby increasing the risk of a vehicle collision. Some wildlife restrains from roads when the
trafc volume increases (van Langevelde and Jaarsma, 2005). According to Seiler (2005), as wildlife tends to intrude on less crowded
roadways, WVCs would occur more frequently in sections where the trafc volume is moderate compared to high-trafc sections. For
example, WVCs involving moose (Alces alces) occurred most frequently in areas with moderate trafc volumes. Similarly, Rendall
(Rendall et al., 2021) discovered that the WVC rate was highest on roadways with moderate trafc volume.
WVC increased on sections with reduced trafc during strict COVID, and the reduction in WVC was maintained after strict COVID.
Other studies also demonstrated similar results. In a study of an expressway in northwestern Spain, the WVC occurrence rate decreased
in number after the lockdown (García-Martínez-de-Alb´
eniz et al., 2022). The sharp decrease in trafc volume, similar to that during
the strict pandemic, can temporarily increase the occurrence of WVCs. However, it can also have a lasting impact on reducing such
incidences. A signicant temporary reduction in trafc volume can help to reduce WVCs.
The causes for WVC are extensive and their interplay is complex. Previous studies have suggested that changes in trafc volume
might impact the behavior of certain wildlife species, particularly large mammals, which might complicate the relationship between
trafc volume and WVCs (Seiler and Helldin, 2006; Jaarsma and Willems, 2002; Alexander et al., 2005). The frequency of wildlife
crossings is determined by the animals behavioral patterns (K¨
ammerle et al., 2017), and the surrounding environment often inuences
behavioral patterns. The occurrence of WVCs follows a very specic spatiotemporal pattern (Park et al., 2021; García-Martí-
nez-de-Alb´
eniz et al., 2022), so the characteristics of the road and the distribution of trafc must also be considered (García-Martí-
nez-de-Alb´
eniz et al., 2022). Thus, it is essential to understand the ecological characteristics of wildlife to comprehend their
relationship with human activities and, consequently, the causes of WVCs.
4. Conclusion
Newly constructed expressways increase the risk of WVCs, leading to signicant socioeconomic consequences. Thus, the rela-
tionship between trafc volume and WVCs has been widely studied. Human interactions with nature were fundamentally changed by
the COVID-19 lockdowns, introduced globally to mitigate the risks associated with the disease (Bates et al., 2021). During the
pandemic, containment measures restricted human activities, thereby offering opportunities to quantitatively assess the human in-
uence on WVCs.
During the strict COVID period, when movement restrictions were in place due to the COVID-19 outbreak, we found a negative
correlation between trafc and WVC. After strict COVID, WVC maintained a downward trend. Therefore, this study is of great sig-
nicance in quantitatively revealing the relationship between WVC and trafc volume on Korean expressways due to movement
restrictions caused by the COVID-19 outbreak. It provides a basis for establishing alternatives to WVC occurrence on expressways.
It is important to note that this study primarily focused on the impact of trafc volume, among several factors contributing to
wildlifevehicle collisions, and only utilized data from the initial phase of the pandemic. Therefore, there may be certain limitations to
the generalizability of the studys ndings. Subsequently, further studies that examine the relationship between trafc restrictions and
WVCs for a longer period of time, from the beginning to the end of the pandemic, are needed to provide ample information on WVC
mitigation (García-Martínez-de-Alb´
eniz et al., 2022). Additional research on various causes of WVCs, such as land cover, landscape
structure, speed, and road characteristics, can also help to expand the understanding of the human inuence on WVCs on expressways.
WVCs are the result of an interaction between humans and wild animals, and thus, it is essential to comprehend the relationship and
Table 4
Correlation of ratio (WVC/trafc volume) analysis results for each period using Pearson correlation (pandemic: April 2020 to December 2020;
pre-strict pandemic: February to March between 2015 and 2019; strict pandemic: February to March 2020). **P<.01.
Pre-strict pandemic Strict pandemic Pandemic
Pre-strict pandemic 1.000 - -
Strict pandemic -0.543 1.000 -
Pandemic 0.246 0.740** 1.000
H. Park and S. Lee
Global Ecology and Conservation 51 (2024) e02949
12
relevance of the interactions between humans and wildlife. In this sense, it is noteworthy that this study drew a quantitative conclusion
on the interaction between humans and animals. Moreover, it can serve as a reference to establish feasible strategies for WVC
prevention.
CRediT authorship contribution statement
Hyomin Park: Methodology, Data curation. Sangdon Lee: Supervision, Investigation, Funding acquisition, Conceptualization.
Declaration of Competing Interest
We conrm that there is no conict among authors.
Data availability
Data will be made available on request.
Acknowledgements
The nancial support was from KRF-2021R1A2C1011213, MOE-2020002990006 and 2021003360002.
Appendix 1
Wildlifevehicle collision (WVC) occurrence on the expressways in Korea between 2015 and 2019: the three routes with the highest
WVC occurrence were selected as study sites.
Route no. Route km WVC
30 DangjinYeongdeok Expressway 278.6 1665
55 Jungang Expressway 288.8 1560
35 Jungbu Expressway 332.5 1364
1 Gyeongbu Expressway 416.05 911
50 Yeongdong Expressway 234.4 739
15 Seohaean Expressway 340.8 713
45 Jungbunaeryuk Expressway 301.7 404
151 SeocheonGongju Expressway 61.4 370
25 Honam Expressway 194.2 317
65 Donghae/UlsanPohang Expressway 175.86 244
20 IkanPohang Expressway 130.3 233
251 Honam Branch Expressway 54 226
27 Suncheon Wanju Expressway 117.8 209
12 MuanGwangju /Gwangju Daegu Expressway 223.2 188
10 Namhae Expressway 273.1 160
40 Pyeongtaek Jecheon Expressway 109.4 130
37 Second Jungbu Expressway 31.1 119
100 Capital Region First Ring Expressway 91.7 88
102 Namhae 1st Branch Expressway 17.9 57
300 Daejeon Southern Belt Expressway 13.3 39
60 SeoulYangyang Expressway 88.8 36
253 Gochang-Damyang Expressway 42.5 29
104 Namhae 2nd Branch Expressway 20.6 19
451 Jungbunaeryuk Expressway 30 18
16 Ulsan Expressway 14.3 15
600 Busan Ring Expressway 48.8 6
110 Second Gyeongin Expressway 26.7 5
551 Jungang Branch Expressway 8.2 2
Appendix 2
Average monthly trafc volume from 2015 to 2020 by study route (vehicles/day). The gray blocks represent the period when strict
restrictions began in Korea during the COVID-19 pandemic (FebruaryMarch 2020).
H. Park and S. Lee
Global Ecology and Conservation 51 (2024) e02949
13
DYE JAE JBE
2015 2016 2017 2018 2019 2020 2015 2016 2017 2018 2019 2020 2015 2016 2017 2018 2019 2020
Jan 131.74 148.84 158.90 125.10 133.55 213.29 1065.68 1082.32 1126.58 1050.97 1047.10 1017.42 4259.29 4426.77 4379.35 5296.06 6049.35 6047.03
Feb 151.36 157.10 166.43 171.07 157.93 140.21 1159.57 1175.59 1139.21 1120.71 1051.07 930.41 4327.36 4521.24 4729.43 5578.50 5923.86 5538.00
Mar 152.06 162.84 158.32 157.94 151.55 144.84 1218.32 1234.65 1267.68 1229.48 1145.81 885.23 4601.87 4986.26 5152.90 5802.84 6159.61 5704.39
Apr 172.20 184.07 181.07 167.13 184.53 163.67 1302.20 1309.60 1387.60 1308.27 1239.47 1048.67 4848.00 5356.53 5374.80 6246.20 6277.07 6330.00
May 167.94 190.32 168.32 170.58 173.81 188.77 1366.00 1424.90 1433.29 1331.29 1318.39 1112.58 5123.03 5215.55 5526.45 6318.84 6528.13 6385.10
Jun 158.27 185.00 178.60 173.67 172.20 187.80 1312.00 1426.60 1435.47 1350.73 1274.73 1277.33 4883.60 5054.33 5484.27 6235.33 6099.27 6533.53
Jul 169.29 168.71 156.65 164.97 126.84 176.19 1362.58 1413.10 1403.35 1374.26 1253.35 1249.55 5008.32 5038.06 5358.32 6006.39 6099.61 6304.13
Aug 178.97 171.74 153.10 155.55 123.42 178.32 1458.84 1569.35 1430.26 1365.81 1300.32 1106.06 5182.32 5383.23 5618.71 6032.71 6356.32 6018.90
Sep 207.60 199.00 184.73 241.47 163.20 193.67 1439.93 1495.80 1460.67 1322.33 1251.07 1123.47 5565.80 5519.73 6318.93 6804.20 6449.87 6818.60
Oct 196.39 178.65 233.94 178.45 182.90 225.03 1460.90 1587.10 1433.03 1366.84 1344.90 1195.35 5547.42 5313.23 6551.42 6726.52 6415.61 6837.42
Nov 174.53 176.40 169.73 164.47 215.93 186.60 1348.07 1450.93 1344.80 1334.20 1295.53 1169.80 5091.40 5915.60 6797.33 6688.67 6438.07 6731.60
Dec 164.26 151.94 135.94 129.48 174.77 148.84 1259.81 1285.03 1204.00 1142.84 1165.61 1011.61 4826.26 4632.52 6160.77 6011.29 6059.29 6281.94
H. Park and S. Lee
Global Ecology and Conservation 51 (2024) e02949
14
References
Abraham, J.O., Mumma, M.A., 2021. Elevated wildlife-vehicle collision rates during the COVID-19 pandemic. Sci. Rep. 11 (1), 20391 https://doi.org/10.1038/
s41598-021-99233-9.
Alexander, S.M., Waters, N.M., Paquet, P.C., 2005. Trafc volume and highway permeability for a mammalian community in the Canadian Rocky Mountains. Can.
Geogr. 49 (4), 321331. https://doi.org/10.1111/j.0008-3658.2005.00099.x.
Bates, A.E., Primack, R.B., Biggar, B.S., et al., 2021. Global COVID-19 lockdown highlights humans as both threats and custodians of the environment. Biol. Conserv.
263, 109175 https://doi.org/10.1016/j.biocon.2021.109175.
Bencin, H.L., Prange, S., Rose, C., Popescu, V.D., 2019. Roadkill and space use data predict vehicle-strike hotspots and mortality rates in a recovering bobcat (Lynx
rufus) population. Sci. Rep. 9 (1), 15391 https://doi.org/10.1038/s41598-019-50931-5.
Bíl, M., Andr´
aˇ
sik, R., Svoboda, T., Sedoník, J., 2016. The KDE+software: a tool for effective identication and ranking of animalvehicle collision hotspots along
networks. Land. Ecol. 31 (2), 231237. https://doi.org/10.1007/s10980-015-0265-6.
Bíl, M., Andr´
aˇ
sik, R., Cícha, V., Arnon, A., Kruuse, M., Langbein, J., N´
ahlik, A., Niemi, M., Pokorny, B., Colino-Rabanal, V.J., Rolandsen, C.M., Seiler, A., 2021. COVID-
19 related travel restrictions prevented numerous wildlife deaths on roads: a comparative analysis of results from 11 countries. Biol. Conserv. 256, 109076
https://doi.org/10.1016/j.biocon.2021.109076.
Coelho, I.P., Kindel, A., Coelho, A.V.P., 2008. Roadkills of vertebrate species on two highways through the Atlantic Forest Biosphere Reserve, southern Brazil. Eur. J.
Wildl. Res. 54, 689699. https://doi.org/10.1007/s10344-008-0197-4.
DAmico, M., Rom´
an, J., De los Reyes, L., Revilla, E., 2015. Vertebrate road-kill patterns in Mediterranean habitats: who, when and where. Biol. Conserv. 191,
234242. https://doi.org/10.1016/j.biocon.2015.06.010.
Da Silva, A.C.F.B., De Menezes, J.F.S., Santos, L., G.R.O, 2022. Roadkill risk for capybaras in an urban environment. Landsc. Urban Plan 222, 104398. https://doi.org/
10.1016/j.landurbplan.2022.104398.
Dean, W.R.J., Seymour, C.L., Joseph, G.S., Foord, S.H., 2019. A review of the impacts of roads on wildlife in semi-arid regions. Diversity 11 (5), 81. https://doi.org/
10.3390/d11050081.
Driessen, M.M., 2021. COVID-19 restrictions provide a brief respite from the wildlife roadkill toll. Biol. Conserv. 256, 109012 https://doi.org/10.1016/j.
biocon.2021.109012.
Fahrig, L., Rytwinski, T., 2009. Effects of roads on animal abundance: an empirical review and synthesis. Ecol. Soc. 14, 21.
Forman, R.T., Sperling, D., Bissonette, J.A., Clevenger, A.P., Cutshall, C.D., Dale, V.H., Fahrig, L., France, R., Goldman, C.R., Heanue, K., Jones, J.A., Swanson, F.J.,
Turrentine, T., Winter, T.C., 2003. Road ecology. Science and solutions. Island Press, Washington.
García-Martínez-de-Alb´
eniz, ´
I., Ruiz-de-Villa, J.A., Rodriguez-Hernandez, J., 2022. Impact of COVID-19 lockdown on wildlife-vehicle collisions in NW of Spain.
Sustainability 14 (8), 4849. https://doi.org/10.3390/su14084849.
Garrah, E., Danby, R.K., Eberhardt, E., Cunnington, G.M., Mitchell, S., 2015. Hot spots and hot times: wildlife road mortality in a regional conservation corridor.
Environ. Manag 56 (4), 874889. https://doi.org/10.1007/s00267-015-0566-1.
Huijser, M.P., Dufeld, J.W., Clevenger, A.P., Ament, R.J., McGowen, P.T., 2009. Cost-benet analyses of mitigation measures aimed at reducing collisions with large
ungulates in the United States and Canada: a decision support tool. Ecol. Soc. 14 (2), 15.
Jaarsma, C.F., Willems, G.P., 2002. A Reducing habitat fragmentation by minor rural roads through trafc calming. Landsc. Urban Plan. 58, 125135. https://doi.org/
10.1016/S0169-2046(01)00215-8.
K¨
ammerle, J.L., Brieger, F., Kr¨
oschel, M., Hagen, R., Storch, I., Suchant, R., 2017. Temporal patterns in road crossing behaviour in roe deer (Capreolus capreolus) at
sites with wildlife warning reectors. PLoS One 12 (9), e0184761. https://doi.org/10.1371/journal.pone.0184761.
Laube, P., Ratnaweera, N., Wr´
obel, A., Kaelin, I., Stephani, A., Reier-Baechtiger, M., Graf, R.F., Suter, S., 2023. Analysing and predicting wildlifevehicle collision
hotspots for the Swiss road network. Land. Ecol. 38, 17651783. https://doi.org/10.1007/s10980-023-01655-5.
Laurance, W.F., Goosem, M., Laurance, S., 2009. Impacts of roads and linear clearings on tropical forests. Trends Ecol. Evol. 24, 659669. https://doi.org/10.1016/j.
tree.2009.06.009.
Lee, G., Tak, J., Pak, S., 2014. Spatial and temporal patterns on wildlife road-kills on highway in Korea. J. Vet. Clin. 31, 282287.
Lester, D., 2015. Effective wildlife roadkill mitigation. J. Trafc Transp. Eng. 3, 4251. https://doi.org/10.17265/2328-2142/2015.01.005.
Lin, S.C., 2016. Landscape and trafc factors affecting animal road mortality. J. Environ. Eng. Landsc. 24, 1020. https://doi.org/10.3846/16486897.2015.1098652.
Litvaitis, J., Tash, J., 2008. An approach toward understanding wildlife-vehicle collisions. Environ. Manag. 42, 688697. https://doi.org/10.1007/s00267-008-9108-
4.
Main, M.B., Allen, G.M., 2002. Landscape and seasonal inuences on roadkill of wildlife in southwest Florida. Fla Sci. 65, 149158.
Manenti, R., Mori, E., Di Canio, V., Mercurio, S., Picone, M., Caf, M., Brambilla, M., Ficetola, G.F., Rubolini, D., 2020. The good, the bad and the ugly of COVID-19
lockdown effects on wildlife conservation: insights from the rst European locked down country. Biol. Conserv. 249, 108728 https://doi.org/10.1016/j.
biocon.2020.108728.
Pagany, R., 2020. Wildlife-vehicle collisions - inuencing factors, data collection and research methods. Biol. Conserv. 251, 108758 https://doi.org/10.1016/j.
biocon.2020.108758.
Park, H., Lee, S., 2013. Habitat use pattern of Korean waterdeer based on the land coverage map. J. Wetl. Res. 15 (4), 567572. https://doi.org/10.17663/
JWR.2013.15.4.567.
Park, H., Kim, M., Lee, S., 2021. Spatial characteristics of wildlife-vehicle collisions of water deer in Korea Expressway. Sustainability 13 (24), 13523. https://doi.org/
10.3390/su132413523.
Primack, R.B., Bates, A.E., Duarte, C.M., 2021. The conservation and ecological impacts of the COVID-19 pandemic. Biol. Conserv. 260, 109204 https://doi.org/
10.1016/j.biocon.2021.109204.
Rendall, A.R., Webb, V., Sutherland, D.R., White, J.G., Renwick, L., Cooke, R., 2021. Where wildlife and trafc collide: roadkill rates change through time in a
wildlife-tourism hotspot. Glob. Ecol. Conserv. 27, e01530 https://doi.org/10.1016/j.gecco.2021.e01530.
Rilett, L.R., Tufuor, E., Murphy, S., 2021. Arterial roadway travel time reliability and the COVID-19 pandemic. J. Transp. Eng. A Syst. 147 (7), 04021034 https://doi.
org/10.1061/JTEPBS.0000559.
Rutz, C., Loretto, M.C., Bates, A.E., Davidson, S.C., Duarte, C.M., Jetz, W., Johnson, M., Kato, A., Kays, R., Mueller, T., Primack, R.B., Ropert-Coudert, Y., Tucker, M.A.,
Wikelski, M., Cagnacci, F., 2020. COVID-19 lockdown allows researchers to quantify the effects of human activity on wildlife. Nat. Ecol. Evol. 4, 11561159.
https://doi.org/10.1038/s41559-020-1237-z.
Saeki, M., Macdonald, D.W., 2004. The effects of trafc on the raccoon dog (Nyctereutes procyonoides viverrinus) and other mammals in Japan. Biol. Conserv. 118,
559571. https://doi.org/10.1016/j.biocon.2003.10.004.
Saint-Andrieux, C., Calenge, C., Bonenfant, C., 2020. Comparison of environmental, biological and anthropogenic causes of wildlifevehicle collisions among three
large herbivore species. Popul. Ecol. 62, 6479. https://doi.org/10.1002/1438-390X.12029.
Seiler, A., 2005. Predicting locations of moose-vehicle collisions in Sweden. J. Appl. Ecol. 42, 371382. https://doi.org/10.1111/j.1365-2664.2005.01013.x.
Seiler, A., Helldin, J.O., 2006. Mortality in wildlife due to transportation. The ecology of transportation: Managing mobility for the environment. Springer
Netherlands, Dordrecht.
Seiler, A., Olsson, M., Rosell, C., Van Der Grift, E.A., 2016. SAFEROAD safe roads for wildlife and people: cost-benet analyses for wildlife and trafc safety.
SAFEROAD Tech. Rep. 4, 40.
Shilling, F., Nguyen, T., Saleh, M., Khant Kyaw, M., Tapia, K., Trujillo, G., Bejarano, M., Waetjen, D., Peterson, J., Kalisz, G., Sejour, R., Croston, S., Ham, E., 2021.
A reprieve from US wildlife mortality on roads during the COVID-19 pandemic. Biol. Conserv. 256, 109013 https://doi.org/10.1016/j.biocon.2021.109013.
Taylor, B.D., Goldingay, R.L., 2010. Roads and wildlife: impacts, mitigation and implications for wildlife management in Australia. Wildl. Res. 37 (4), 320331.
https://doi.org/10.1071/WR09171.
H. Park and S. Lee
Global Ecology and Conservation 51 (2024) e02949
15
Van der Ree, R., Smith, D.J., Grilo, C., 2015. The ecological effects of linear infrastructure and trafc: challenges and opportunities of rapid global growth. Handb.
Road. Ecol. 19.
van Langevelde, F., Jaarsma, C.F., 2005. Using trafc ow theory to model trafc mortality in mammals. Land. Ecol. 19, 895907. https://doi.org/10.1007/s10980-
005-0464-7.
Visintin, C., Van Der Ree, R., McCarthy, M.A., 2017. Consistent patterns of vehicle collision risk for six mammal species. J. Environ. Manag. 201, 397406. https://doi.
org/10.1016/j.jenvman.2017.05.071.
H. Park and S. Lee
Article
Full-text available
Context Wildlife–vehicle collisions (WVCs) are a significant threat for many species, cause financial loss and pose a serious risk to motorist safety. Objectives We used spatial data science on regional collision data from Switzerland with the objectives of identifying the key environmental collision risk factors and modelling WVC risk on a nationwide scale. Methods We used 43,000 collision records with roe deer, red deer, wild boar, and chamois from 2010 to 2015 for both midlands and mountainous landscape types. We compared a fixed-length road segmentation approach with segments based on Kernel Density Estimation, a data-driven segmentation method. The segments’ environmental properties were derived from land-cover geodata using novel neighbourhood operations. Multivariate logistic regression and random forest classifiers were used to identify and rank the relevant environmental factors and to predict collision risk in areas without collision data. Results The key factors for WVC hotspots are road sinuosity, and two composite factors for browsing/forage availability and traffic noise—a proxy for traffic flow. Our best models achieved sensitivities of 82.5% to 88.6%, with misclassifications of 20.14% and 27.03%, respectively. Our predictions were better in forested areas and revealed limitations in open landscape due to lack of up-to-date data on annual crop changes. Conclusions We illustrate the added value of using fine-grained land-cover data for WVC modelling, and show how such detailed information can be annotated to road segments using spatial neighbourhood functions. Finally, we recommend the inclusion of annual crop data for improving WVC modelling.
Article
Full-text available
Wildlife–vehicle collisions (WVCs) in many places have a significant impact on wildlife management and road safety. The COVID-19 lockdown enabled the study of the specific impact that traffic has on these events. WVC variation in the Asturias and Cantabria regions (NW of Spain) because of the COVID-19 lockdown reached a maximum reduction of −64.77% during strict confinement but it was minimal or nonexistent during “soft” confinement. The global average value was −30.22% compared with the WVCs registered in the same period in 2019, but only −4.69% considering the average throughout the period 2010–2019. There are huge differences between conventional roads, where the traffic reduction was greater, and highways, where the traffic reduction was lesser during the COVID-19 lockdown. The results depend on the season, the day of the week and the time of day, but mainly on the traffic reduction occurring. The results obtained highlight the need to include the traffic factor in WVC reduction strategies.
Article
Full-text available
In recent decades, rapid industrial growth has accelerated the construction of new roads, which has led to the destruction and isolation of wildlife habitats. Newly constructed roads affect wildlife in many ways. In particular, fatal wildlife-vehicle collisions (WVCs) have a direct impact on wildlife. A substantial number of WVCs occur every year on expressways, where vehicle speeds and vehicle traffic are significant. However, our understanding of the relative importance of the factors associated with areas in which large numbers of WVCs occur on the expressway remains poor. Therefore, herein, we analyze the spatial characteristics of WVCs. The effect of spatial distribution on the occurrence of WVCs was analyzed using the types of land cover in the areas where water deer appear (Cheongju, Boeun, and Sangju) and the areas in which WVCs occur along the Cheongju–Sangju Expressway (CSE). We identified the WVC hotspots by using CSE patrol data recorded between January 2008 and December 2019, and we analyzed the corresponding distribution patterns and land cover characteristics. Along the CSE, a total of 1082 WVCs occurred, out of which collisions involving water deer (Hydropotes inermis argyropus) accounted for 91%. Water deer appear frequently in Forested Areas and Agricultural Land, but the WVC distribution in the Hotspots followed a highly clustered pattern, with a higher proportion of WVCs occurring in Used Areas (areas including buildings such as residential facilities, commercial and industrial facilities, and transportation facilities). Used Areas have a smaller cut slope compared to Forested Areas, and Used Areas are open terrains. Therefore, the occurrence of WVCs will be high given that wildlife can easily access the expressway. Based on these results, we can infer that the landscapes near the expressway influence the occurrence of WVCs. To establish an effective policy for reducing WVCs on a road, the WVC characteristics and spatial distribution of the road should be considered together. Further research on the wildlife ecology and land-use status of WVC hotspots is required to mitigate WVCs on expressways and protect human and animal life. Therefore, if the characteristics of WVC hotspots are analyzed considering the characteristics of various ecosystems, an appropriate WVC reduction plan can be established.
Article
Full-text available
Wildlife-vehicle collisions threaten both humans and wildlife, but we still lack information about the relationship between traffic volume and wildlife-vehicle collisions. The COVID-19 pandemic allowed us to investigate the effects of traffic volume on wildlife-vehicle collisions in the United States. We observed decreased traffic nationwide, particularly in densely populated states with low or high disease burdens. Despite reduced traffic, total collisions were unchanged; wildlife-vehicle collisions did decline at the start of the pandemic, but increased as the pandemic progressed, ultimately exceeding collisions in the previous year. As a result, nationwide collision rates were higher during the pandemic. We suggest that increased wildlife road use offsets the effects of decreased traffic volume on wildlife-vehicle collisions. Thus, decreased traffic volume will not always reduce wildlife-vehicle collisions.
Article
Full-text available
The global lockdown to mitigate COVID-19 pandemic health risks has altered human interactions with nature. Here, we report immediate impacts of changes in human activities on wildlife and environmental threats during the early lockdown months of 2020, based on 877 qualitative reports and 332 quantitative assessments from 89 different studies. Hundreds of reports of unusual species observations from around the world suggest that animals quickly responded to the reductions in human presence. However, negative effects of lockdown on conservation also emerged, as confinement resulted in some park officials being unable to perform conservation, restoration and enforcement tasks, resulting in local increases in illegal activities such as hunting. Overall, there is a complex mixture of positive and negative effects of the pandemic lockdown on nature, all of which have the potential to lead to cascading responses which in turn impact wildlife and nature conservation. While the net effect of the lockdown will need to be assessed over years as data becomes available and persistent effects emerge, immediate responses were detected across the world. Thus initial qualitative and quantitative data arising from this serendipitous global quasi-experimental perturbation highlights the dual role that humans play in threatening and protecting species and ecosystems. Pathways to favorably tilt this delicate balance include reducing impacts and increasing conservation effectiveness.
Article
Full-text available
Millions of wild animals are killed annually on roads worldwide. During spring 2020, the volume of road traffic was reduced globally as a consequence of the COVID-19 pandemic. We gathered data on wildlife-vehicle collisions (WVC) from Czechia, Estonia, Finland, Hungary, Israel, Norway, Slovenia, Spain, Sweden, and for Scotland and England within the United Kingdom. In all studied countries WVC statistics tend to be dominated by large mammals (various deer species and wild boar), while information on smaller mammals as well as birds are less well recorded. The expected number of WVC for 2020 was predicted on the basis of 2015–2019 WVC time series representing expected WVC numbers under normal traffic conditions. Then, the forecasted and reported WVC data were compared. The results indicate varying levels of WVC decrease between countries during the COVID-19 related traffic flow reduction (CRTR). While no significant change was determined in Sweden, where the state-wide response to COVID-19 was the least intensive, a decrease as marked as 37.4% was identified in Estonia. The greatest WVC decrease, more than 40%, was determined during the first weeks of CRTR for Estonia, Spain, Israel, and Czechia. Measures taken during spring 2020 allowed the survival of large numbers of wild animals which would have been killed under normal traffic conditions. The significant effects of even just a few weeks of reduced traffic, help to highlight the negative impacts of roads on wildlife mortality and the need to boost global efforts of wildlife conservation, including systematic gathering of roadkill data.
Article
Full-text available
Understanding when and where roadkill is most likely to occur is vital to reducing wildlife-vehicle collisions. However, little is known about how roadkill rates change through time and whether or not the key influences on roadkill also change. Understanding changes in roadkill will facilitate the best implementation of mitigation measures. We aimed to determine how roadkill rates have changed between two distinct time periods and assess whether the spatial and temporal drivers of roadkill rates may have changed: with a view to informing taxon-specific mitigation strategies. We assess the spatial and temporal factors that influence road mortalities in two periods (1998-99 and 2014) at the same site for multiple taxa. Bi-weekly surveys were undertaken from February to May 1998 and 1999 and again from February to June 2014. In total 2 479 individual roadkill were recorded throughout the surveys, with 1.59 roadkill per km per month in the 1990s, increasing to 2.39 per km per month in 2014. Roadkill rates increased primarily with road speed limit with mortalities peaking at moderate (60 – 80 km/h) speeds, however, the structural complexity of roadside vegetation and traffic volume influenced roadkill rates for some taxa but not others. We show that roadkill rates have changed through time with shifts in both the temporal and spatial influences on these roadkill rates. These changes are likely associated with changes in the abundance of taxa and increased vehicle traffic. The spatial and temporal drivers of roadkill rates were found to be taxon specific, and although mitigation measures exist, assessment of their efficacy remains a priority.
Article
Wildlife vehicle-collision is one of the most visible negative effects that roads exerts on animals and has increased dramatically across the world. Despite its conspicuousness, studies about roadkills in cities have been neglected lacking in road ecology. We developed new approach for estimating capybara roadkill hotspots in Campo Grande, a big city in Brazil. We also investigated potential driving factors correlated with roadkill occurrence, to build a predictive roadkill map for the entire city and propose mitigation measures. We monitored capybara roadkills for thirteen years and found hotspots using a graph-based kernel density estimation. We tested four predictors to identify which characteristics influence roadkill occurrence: distance from water bodies, distance from parks, vegetation cover, and traffic flow. We used a generalized linear mixed model to test for significant effects and to predict roadkill occurrences. Hotspots analysis showed four hotspots surrounding large green areas and water bodies, probably due to capybara habitat and physiological requirements. The predictive map shows latent hotspots, locations that have the characteristics necessary for a capybara to live but where we do not have observed deaths. To mitigate risk, we recommend using speed reduction tools around parks. A decrease in capybaras roadkills could positively impact human population welfare and material damage caused by these collisions.
Article
This paper evaluated the effect of the COVID-19 preventive orders on arterial roadway travel time reliability (TTR). A comparative analysis was conducted to examine average travel time distributions (TTD), and their associated TTR metrics, before and during the COVID-19 pandemic. Travel time data for four urban arterial corridors in Nebraska, disaggregated by peak period and direction, were analyzed. It was found that in 2020, the average TTD mean and standard deviation values for all 16 scenarios were reduced by an average of 14.0% and 43.4%, respectively. The travel time index, the planning time index, the level of travel time reliability (LOTTR), and the buffer index metrics associated with these TTDs were reduced, on average, by 14.0%, 19.7%, 3.5%, and 35.0%, respectively. In other words, whether the test corridors were more reliable during the pandemic was a function of which TTR metric was used. The paper concludes by arguing for a fundamental change in how arterial TTR is measured and reported to different user groups.