# Livestock Dung Proxies Provide Insights into Grazing Density Quantification and Distribution

## TL;DR

This study uses livestock dung as a proxy to measure grazing intensity using drone surveys, offering a new way to monitor and manage grasslands.

## Contribution

A novel method using yak dung density as a proxy for grazing intensity, assessed via drone imagery.

## Key findings

- Yak dung is clearly identifiable in UAV images, with detection accuracy decreasing at higher altitudes.
- Yak dung density effectively captures the temporal and spatial variability of grazing intensity.
- The method is high in frequency, accuracy, and efficiency for monitoring livestock-grassland interactions.

## Abstract

Grazing is the predominant utilization of grasslands worldwide. Nevertheless, accurately quantifying grazing intensity across spatial and temporal dimensions remains challenging. In this study, we introduced a novel approach to estimate grazing intensity by utilizing livestock dung as a proxy indicator in combination with repeated drone-based surveys. The feasibility of this method was demonstrated through a case study conducted in a household pasture on the Qinghai–Tibetan Plateau, China. The proposed approach demonstrates high frequency, accuracy, and operational efficiency, making it well suited for investigating grazing intensity and analyzing livestock-resource interactions.

Managed grazing is the most widespread and economically significant form of grassland utilization worldwide. Accurate quantification of the spatiotemporal distribution of grazing intensity (GI) is crucial for promoting sustainable management of livestock-grassland ecosystems. However, a reliable method for dynamically monitoring GI and quantifying key proxies under real-world grazing conditions is still lacking. In this study, we developed a practical approach to estimate GI using sequential unmanned aerial vehicle (UAV) monitoring and evaluated its feasibility in a typical household pasture on the Qinghai–Tibetan Plateau, China. Our findings show that: (1) yak dung is clearly identifiable in UAV image, although detection accuracy decreases with increasing flight altitude (from 100% at 2 m to 93.16% at 20 m); (2) yak dung density serves as a feasible proxy for GI, effectively capturing its temporal and spatial variability; (3) yak dung density reflects cumulative GI from May to September, and its representativeness increases with the length of accumulation. The proposed approach is characterized by high frequency, accuracy, and efficiency. It is well-suited for studying animal behavior and evaluating livestock–resource relationships, thereby providing valuable insights for sustainable grassland ecosystem management.