A non-destructive method for three-dimensional characterizing plateau pika's burrow system

The plateau pika's burrow system exerts a significant influence on the alpine ecosystem. However, current methods are insufficient for accurately characterizing its spatial structure. In this study, we utilized drone imagery and ground-penetrating radar (GPR) scanning in alpine grasslands on the eastern edge of the Qinghai-Tibetan Plateau (QTP). We developed a novel signal processing workflow for trajectory detection and reconstruction of the three-dimensional structure of pika's burrow systems. The results indicate that, within a 10 × 10 m study area, eight burrows were identified in moderately degraded grassland, including six simple burrows and two complex burrows. The longest tunnel measured 830 cm, with an average depth and diameter of 23.2 cm and 7.0 cm, respectively. These burrows accounted for 2.5% of the pika affected soil volume. In severely degraded grassland, twelve burrows were identified, with the longest tunnel measuring 580 cm, and an average depth and diameter of 15.6 cm and 7.0 cm, accounting for 5.2% of the pika affected soil volume. Compared to the extraction method, the values for length, depth, and orientation were largely consistent, although the diameter of burrows was slightly lower in GPR data. Regression analysis further confirmed the feasibility of proposed method. This non-destructive detection method provides critical technical support for studying the ecological impacts of burrow systems of fossorial animals on the QTP.