# Integration of the landscape of fear concept in grassland management: An experimental study on subtropical monsoon grasslands in Bardia National Park, Nepal

## TL;DR

This study explores how deer species in Nepal's grasslands respond to habitat management practices, revealing that larger plots are preferred due to a perceived sense of safety from predators.

## Contribution

The study experimentally integrates the 'landscape of fear' concept into grassland management, revealing species-specific responses to predation risk and habitat treatments.

## Key findings

- Chital and swamp deer prefer larger plots, showing a higher pellet group density in the center of these plots.
- Hog deer did not show significant responses to the experimental treatments, indicating species-specific differences in behavior.
- Larger plots with mowing and fertilization treatments were more utilized by deer, highlighting the importance of perceived safety.

## Abstract

The ‘landscape of fear’ concept offers valuable insights into wildlife behaviour, yet its practical integration into habitat management for conservation remains underexplored. In this study, conducted in the subtropical monsoon grasslands of Bardia National Park, Nepal, we aimed to bridge this gap through a multi‐year, landscape‐scale experimental investigation in Bardia National Park, Nepal. The park has the highest density of tigers (with an estimated density of ~7 individuals per 100 km2) in Nepal, allowing us to understand the effect of habitat management on predation risk and resource availability especially for three cervid species: chital (Axis axis), swamp deer (Rucervus duvaucelii) and hog deer (Axis porcinus). We used plots with varying mowing frequency (0–4 times per year), size (ranging from small: 49 m2 to large: 3600 m2) and artificial fertilisation type (none, phosphorus, nitrogen) to assess the trade‐offs between probable predation risk and resources for these cervid species, which constitute primary prey for tigers in Nepal. Our results showed distinct responses of these deer to perceived predation risk within grassland habitats. Notably, these deer exhibited heightened use of larger plots, indicative of a perceived sense of safety, as evidenced by the higher occurrence of pellet groups in the larger plots (mean = 0.1 pellet groups m−2 in 3600 m2 plots vs. 0.07 in 400 m2 and 0.05 in 49 m2 plots). Furthermore, the level of use by the deer was significantly higher in larger plots that received mowing and fertilisation treatments compared to smaller plots subjected to similar treatments. Of particular interest is the observation that chital and swamp deer exhibited greater utilisation of the centre (core) areas within the larger plots (mean = 0.21 pellet groups m−2 at the centre vs. 0.13 at the edge) despite the edge (periphery) also provided attractive resources to these deer. In contrast, hog deer did not display any discernible reaction to the experimental treatments, suggesting potential species‐specific variations in response to perceived predation risk arising from management interventions. Our findings emphasise the importance of a sense of security as a primary determinant of habitat selection for medium‐sized deer within managed grassland environments. These insights carry practical implications for park managers, providing a nuanced understanding of integrating the ‘landscape of fear’ into habitat management strategies. This study emphasises that the ‘landscape of fear’ concept can and should be integrated into habitat management to maintain delicate predator–prey dynamics within ecosystems.

An aerial view of experimental plots in one of the open grasslands locally known as ‘Bagauraha Phanta’ in Bardia National Park, Nepal.