Training dataset for Rock ptarmigan

Abstract
Monitoring vulnerable species inhabiting mountain environments is crucial to track population trends and prioritize conservation efforts. However, the challenging nature of these remote areas poses difficulties in implementing effective and consistent monitoring programs. To address these challenges, we examined the potential of passive acoustic monitoring (PAM) on a cryptic high mountain bird species, the Rock Ptarmigan (Lagopus muta). We deployed 38 autonomous recording units spanning the Swiss Alps in areas where the birds are followed by a national monitoring program and built a machine-learning algorithm to automatize song recognition. We focused on studying the daily and seasonal call phenology of the species and relate it to meteorological and climatic data. Our results revealed that Rock Ptarmigans were vocally active from March to July, with a peak of activity occurring between mid-March and late April, one or two months earlier than the conventional count in the second half of May. The calling frequency peaked at dawn before dropping rapidly until sunrise. Daily vocal activity demonstrated a consistent dependency on daily weather and moon phase, while the timing of seasonal vocal activity was dependent on temperature and snow conditions.  We found that the peak of vocal activity occurred when the snowpack was still thick, and snow cover was close to 100% but with a local peak of high temperatures. Between our two study years, the peak of vocal activity occurred with 30 days delay in the colder year, highlighting the species' phenological plasticity in relation to environmental conditions. PAM has the potential to complement conventional acoustic counts of the cryptic birds by highlighting periods of higher detectability of the individuals or following small populations where individuals often remain undetected. Moreover, our case study supports the idea that PAM can provide valuable data over large spatial and temporal scales, allowing it to decrypt hidden ecological patterns and assist conservation efforts.

摘要
监测栖息于山地生境的脆弱物种，对于追踪种群动态、合理配置保护优先级至关重要。然而，这类偏远区域的复杂环境特征，给高效且统一的监测方案落地带来了显著挑战。为应对上述难题，我们针对一种隐秘性高山鸟类——岩雷鸟（Lagopus muta），探究了被动声学监测（Passive Acoustic Monitoring, PAM）的应用潜力。我们在瑞士阿尔卑斯山区的岩雷鸟国家级监测区域内布设了38台自主录音单元，并构建了机器学习算法以实现鸣声识别自动化。本研究重点分析了该物种的日、季鸣声物候特征，并将其与气象及气候数据进行关联分析。

研究结果表明：岩雷鸟的鸣唱活跃期为3月至7月，活动峰值出现在3月中旬至4月下旬，较常规的5月中下旬调查时段提前1至2个月；其鸣唱频率在黎明时分达到峰值，随后迅速回落至日出后水平。日间鸣唱活动始终与当日天气状况及月相存在稳定相关性，而季候鸣唱活动的时序则受温度与积雪条件调控。我们发现，鸣唱活动峰值出现时，积雪层仍较深厚、积雪覆盖率接近100%，但局部存在高温时段。在两个研究年度中，较寒冷年份的鸣唱活动峰值延迟了30天，这体现了该物种针对环境变化的物候可塑性。

被动声学监测有望通过标记个体高可检测性的时段，或是追踪通常难以被发现的小型种群，来补充针对隐秘性鸟类的常规声学调查。此外，本案例研究证实，被动声学监测能够在大时空尺度上获取高价值数据，从而揭示隐藏的生态模式并为保护工作提供支撑。