Sampaio Ricardo, Morato Ronaldo G, Abrahams Mark I, Chiarello Adriano G, Peres Carlos A
Centro Nacional de Pesquisa e Conservação de Mamíferos Carnívoros (CENAP), Instituto Chico Mendes de Conservação da Biodiversidade (ICMBio), Atibaia, São Paulo, Brasil.
Pós Graduação em Biologia Comparada e Departamento de Biologia, Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto, Brasil.
Conserv Biol. 2025 Jun;39(3):e70029. doi: 10.1111/cobi.70029.
Assessing local anthropogenic impacts on tropical forest wildlife is central to conservation science in low-governance regions, particularly in the context of community-based resource management. Two methods frequently used to sample wildlife are based on local ecological knowledge (LEK), which is low cost and draws on substantial human experience, and camera trap sampling (CT), which is widely used due to its spatial replicability and the increasing diversity of analytical frameworks. We compared the effectiveness of both methods in assessing the impact of hunting on wildlife at 70 local community catchments across the Brazilian Amazon. We did this by assessing local occurrences of 17 focal species as recounted by 187 subsistence hunters in interviews and by assessing data from 631 CT deployments. We evaluated correlations among the distances from the nearest community at which species and species groups occurred, derived from either LEK or CT data. We also assessed how species' morphological and socioecological traits influenced estimates derived from LEK and CT. Estimates derived from LEK were more strongly correlated with species abundance than with occupancy estimates. Large-bodied, harvest-sensitive species, such as tapir (Tapirus terrestris), white-lipped peccary (Tayassu pecari), and curassow (Crax or Pauxi spp.), were spatially depleted more than 15 km away from human communities. These species showed particularly strong positive correlations across all estimates. The precision of LEK estimates increased with independent data on local game species abundance. These results underline the utility of data derived from LEK in assessing patterns of local wildlife abundance, especially for large-bodied species sensitive to hunting pressure. Local ecological knowledge-based methods are valuable in their own right, especially in areas where alternative sampling methods are lacking and human impact is not matched by conservation efforts.
评估当地人为活动对热带森林野生动物的影响，是治理水平较低地区保护科学的核心内容，特别是在基于社区的资源管理背景下。常用于野生动物采样的两种方法，一种基于当地生态知识（LEK），成本低且利用了大量人类经验；另一种是相机陷阱采样（CT），因其空间可复制性和分析框架的日益多样化而被广泛使用。我们比较了这两种方法在评估巴西亚马逊地区70个当地社区集水区狩猎对野生动物影响方面的有效性。我们通过评估187名自给性猎人在访谈中讲述的17种重点物种的当地出现情况，以及评估631次相机陷阱部署的数据来做到这一点。我们评估了从LEK或CT数据得出的物种和物种组出现的最近社区距离之间的相关性。我们还评估了物种的形态和社会生态特征如何影响从LEK和CT得出的估计值。从LEK得出的估计值与物种丰度的相关性比与占有率估计值的相关性更强。大型、对捕猎敏感的物种，如貘（Tapirus terrestris）、白唇西貒（Tayassu pecari）和凤冠雉（Crax或Pauxi spp.），在距离人类社区超过15公里的地方在空间上减少。这些物种在所有估计值中都显示出特别强的正相关。LEK估计值的精度随着当地猎物物种丰度的独立数据而提高。这些结果强调了从LEK得出的数据在评估当地野生动物丰度模式方面的实用性，特别是对于对捕猎压力敏感的大型物种。基于当地生态知识的方法本身就很有价值，特别是在缺乏替代采样方法且保护努力与人类影响不匹配的地区。
