Wildlife roadkill in Chitwan, Nepal: identifying affected species, potential drivers and hotspots
Data files
Jul 07, 2026 version files 8.38 KB
Jul 07, 2026 version files 8.37 KB
Abstract
Roadkill poses a significant threat to species conservation. However, there is scarce information on the roadkill of wild animals. Therefore, to address these research gaps, this study aimed to assess roadkill status of vertebrates, including their conservation status and possible drivers such as seasons, structure of road and habitat types, using direct observation methods over four seasons (autumn, winter, summer and spring) in 2024–2025 on two national highways: part of East–West Highway (H01) and Madan Ashrit Highway (H05). The status of road-killed vertebrates was evaluated by using relative frequency and encounter rate. The seasonal difference among the reported road-killed vertebrates was visualized using venn diagram. The generalized linear model was used to show the relationship between the total number of roadkill cases of vertebrates with different variables such as seasons, habitat types and condition of road. Similarly, this study also mapped the roadkill hotspots based on data using the Inverse Weighted Distance (IDW) algorithms. A total of 618 cases involving 36 species of wild vertebrates were reported on two highways, comparatively higher in H01. Among the reported species, 36.1% were mammals following 27.78% Aves, 25% reptiles and 11.1% amphibians. This study also recorded two globally Near Threatened (NT), four nationally Vulnerable (VU), and two nationally Near Threatened (NT) roadkill vertebrates. Roadkill risk is closely associated with habitat types. The road sections that cross between forests, forest edges, and residential areas comparatively have higher roadkill cases than croplands. River crossings are critical risk zones for wildlife vehicle encounters. Major roadkill hotspots on both national highways were identified, and recommendations were made for implementing speed limits and constructing underpasses and overpasses during road renovations.
Dataset DOI: 10.5061/dryad.51c59zwpq
Description of the data and file structure
The study was conducted from July 2024 to June 2025. Two national highways (H01 and H05) were selected for systematic roadkill data collection. The entire length of the two highways (approximately 36 km each) was monitored using motorbikes, with two personnel involved: one driver and another observer. The bike was driven at a speed of 20 ± 5 km/h to allow the observer to detect road-killed wild animals effectively. The survey was conducted twice a week at regular intervals, during early morning (between 06:00 and 09:00 h) and late afternoon (between 14:00 and 17:00 h). A total of 208 survey visits were completed across the two national highways, with 104 visits on each highway. Each highway was monitored twice a week during the study period. Besides systematic surveys, opportunistic data were recorded through the nonsystematic visits conducted in response to external information. These visits were conducted based on the information from media sources, local people, and government officials, and carried out at any time of day on both highways to document roadkill incidents. However, the data collected from these opportunistic observations were not included in the statistical analysis. The study period was divided into four seasons: summer or rainy (June–August), autumn (September–November), winter (December–February), and spring (March–May) to explore the seasonal variation. When roadkill was detected, the bike stopped, and the wild animal's carcass was observed carefully. The dead animals were identified by using the field guide books and identification keys (e.g., Herpetofauna of Nepal by Shah and Tiwari (2004), birds of Nepal by Inskipp et al. (2016) and Grimmett et al. (2016), mammals of Nepal Baral and Shah (2008)). The photographs of each roadkill incident were taken, and the geographic coordinates of the location were recorded using a Global Positioning System (GPS). To avoid multiple counts of the same individuals, the carcasses were removed from the road surface and deposited on the roadside so that the conservation authority can pick them up. The carcasses were not collected in the field, just recorded species. The zoological name, order, family along with IUCN and NRDB (National Red Data Book) threatened species were verified using the authentic sources such as http:// www. iucnr edlist. org/ , The Status of Nepal's Mammals: The National Red List Series, https:// www. himal ayann ature. org/ citiz en-science/ naton al-red-list-of-birds/ and https:// amphi biaweb. org/ . Variables such as season, habitat type, and road structure were recorded during the field study. Data were listed in two .csv files.
In the data 'one side forest' stands for one side forest and another side either cropland or resident areas of the road and s mode stands for sigmoid road.
Files and variables
File: Data_road_kill_animals_(East_west_highway_H01)_Adhikari_et_al._2026.csv
Description:
Variables
- SN: Serial number
- Name: Name of the vertebrate species reported
- Seasons: Summer or rainy (June-August), Autumn (September-November), Winter (December to February), Spring (March-May).
- Habitat types:
- Forest: Road crossing along the forest area (i.e. both side of the road have forest)
- Forest edge: One side has forest and other side has either settlement or cropland
- Cropland: Cultivated area
- Residential area: Settlement area
- River crossing: road that passes through the rivers or streams
- Road structure
- Straight
- S-mode or sigmoid mode: having bends or turns
File: Data_road_kill_animals(Madan_Ashrit_highway_H05)_Adhikari_et_al._2026.csv.
Description:
Variables
- SN: Serial number
- Species: Name of the vertebrate species reported
- Category: Different classes of vertebrates (i.e. Mammals, Birds, Reptiles and Amphibia)
- Seasons: Summer or rainy (June-August), Autumn (September-November), Winter (December to February), Spring (March-May).
- Habitat types:
- Forest: Road crossing along the forest area (i.e. both side of the road have forest)
- Forest edge: One side has forest and other side has either settlement or cropland
- Cropland: Cultivated area
- Residential area: Settlement area
- River crossing: road that passes through the rivers or streams
- Road structure
- Straight
- S-mode or sigmoid mode: having bends or turns
Code/software
R software version 4.2.0
Access information
Other publicly accessible locations of the data:
- na these are primary data, these are included in this article only
Data was derived from the following sources:
- Primary data
Changes after Jul 7, 2026: Edited README.md.