Abstract: As a top predator, the leopard cat (Prionailurus bengalensis) plays an important role in maintaining the structure of the food web and ecosystem stability. For the study of the diet composition of predators, a relatively simple method is the fecal sample contents analysis, while the DNA metabarcoding technology on fecal residues has the advantages of being more accurate and detailed, but it also has shortcomings. Therefore, exploring the complementary advantages of different methods will help to improve their application effectiveness. In this study, 71 fecal samples of leopard cats collected from four areas in Beijing were analyzed by using the DNA metabarcoding technique and fecal content analysis with the aims of understanding the food resources utilization of leopard cats and assessing the complementary results of these two approaches. The results showed that the DNA metabarcoding identified 36 prey species, belonging to 22 families and 10 orders. The small mammals were the main prey, among which the rats comprise the highest percentage and birds in second place in Baihuashan, Songshan, and Yunmengshan sampling areas. In Yunfengshan area, the highest percentage of prey was birds followed by rats and mice in second place. The fecal content analysis identified 9 types of prey, including insects and plants which were not detected by the DNA metabarcoding. Furthermore, this approach detected the prey numbers of three rats and two birds in scat samples, respectively, and the secondary food is plants and insects. Both methods showed that Niviventer confucianus, Apodemus draco, and A. agrarius were the most frequently preyed rodents. This study identified more prey species by DNA metabarcoding while the fecal content analysis revealed the numbers of the prey, indicating complementary results between these two methods. In addition, the diet composition characteristics of leopard cats provid baseline information for further detecting the inter-species coexistence among the sympatric predators.

Key words: Leopard cat, Non-invasive sampling, DNA metabarcoding, Fecal content analysis, Diet composition

摘要： 豹猫(Prionailurus bengalensis)作为北京地区的顶级食肉动物，对于维持食物网结构和生态系统稳定性起到重要的生态作用。对于捕食动物食物构成研究，较为简便的方法是粪样内容物检视法，而粪样残余物DNA鉴定技术具有更为准确细致的优势，但也存在不足，探索不同方法的优势互补，将有助于提高技术应用成效。本研究利用DNA宏条形码技术与粪样内容物分析法，对采集自北京市4个自然保护区的71份豹猫粪样进行食物构成分析，比较两种分析方法的特点，了解豹猫的食物资源利用状况。结果显示，DNA宏条形码技术共鉴别出36种猎物，来自10目22科，4个保护区的豹猫食性具有显著差异，百花山、松山、云蒙山保护区的豹猫食物种类的出现比率均以小型哺乳类为主，其中对鼠类的捕食比例最高，对鸟类的捕食次之，而分布于云峰山保护区的豹猫对鸟类捕食比例最高，对鼠类的捕食次之。粪样内容物分析法鉴别出9类猎物，其中包括昆虫和植物两种DNA宏条形码技术未检出的食物，4个保护区的豹猫食物均以鼠类和鸟类为主，且最多检测出鼠类数量为3只、鸟类2只，次要食物则为植物和昆虫。两种方法均显示北社鼠(Niviventer confucianus)、中华姬鼠(Apodemus draco)、黑线姬鼠(A. agrarius)被取食的频率最高。本研究基于DNA宏条形码技术鉴定出动物性猎物种类较为齐全，粪样内容物分析法能够区分猎物个体数量，实现了在豹猫食性分析中的互补效应，同时，本研究发现了豹猫的食物组成特点，为后续深入开展捕食动物种间共存提供了基础信息。

关键词: 豹猫, 非损伤性取样, DNA宏条形码, 粪样内容物分析法, 食物组成