Neil D'Cruze ¹ Angie Elwin ¹ P E. Perez-Peña ² Roberto Vieto ¹ Alexander E. Asfaw ¹ Lauren A Harrington ^3*

• ¹ World Animal Protection, London, England, United Kingdom
• ² Instituto de Investigaciones de la Amazonia Peruana, Iquitos, Peru
• ³ University of Oxford, Oxford, United Kingdom

Domestic wildlife markets have important nutritional, medicinal, cultural, and financial significance for local communities, but the scale and diversity of wildlife trade that passes through them is also associated with negative impacts on biodiversity, poor animal welfare, and potential human health risk. To design, and monitor the effectiveness of, interventions to ameliorate such impacts, an understanding of the species sold at the markets and their purpose is required, together with a robust (and potentially flexible) baseline. Here we focus on Belén (the largest open wildlife market in the Peruvian Amazon) and Modelo market, in Iquitos, Peru. To obtain a broad overview of the nature of these markets, and their potential impacts on native wildlife, we surveyed wildlife products for sale approximately weekly over a year, using two different survey methods (open and discreet). Both markets sold predominantly wild meat, and some pets; at Belén market >30% observations were of decorative, spiritual, or medicinal products. At least 71 unique species (including mammals, reptiles, birds, and invertebrates) were observed in total. 27.7% of species were threatened or Near Threatened globally or nationally but there was no evidence that discreet surveys increased their detection. To provide a baseline to support future conservation monitoring, we present data on product availability, volume (observed per survey), and price, for the eight most frequently observed species 'groups' (lowland paca, peccaries, caiman, river turtles, boas, yellow footed tortoise, parrots, and brocket deer). Our aim was to quantify several different market metrics to maximize future utility of the dataset. To provide a complete understanding of the range of species involved, we also provide a description of all threatened species recorded at the markets, the products sold, and their uses, including those that were only observed occasionally. Beyond providing a comparative dataset, we suggest that simulations using these data could be used to optimize future monitoring efforts. Finally, our observations of correlations of per survey trade volumes of some species with daily river water levels in Iquitos (positive for lowland paca, caiman, yellow-footed tortoise and negative for parrots, river turtle eggs) may inform optimal time of year for species-specific surveys.