Predicting fish weight using photographic image analysis: a case study of broad whitefish in the lower Mackenzie River watershed

Many small-scale fisheries are remote in nature, making data collection logistically difficult. Thus, there is a need for accessible solutions that address the gaps present these fisheries. One possible solution to incorporate photography into community- or harvest-based monitoring frameworks and employ images estimate biological data. Here, we test this approach using łuk dagaii, broad whitefish, Coregonus nasus (Pallus, 1776) Gwich'in Settlement Area, region Mackenzie River system Canada's Northwest Territories. We used photographs taken by collaborators simple, standardized set-up ask question: how accurately can weight be estimated from photo? Using random forest models based on morphometric photograph measurements as well season location of harvest, predicted whitefish within 13% true (257 g, fish weighing an average 2036 g). The model predictions were distributed their residuals most fish, though discuss biases at low high weights. Image analysis cost, method may contribute ongoing, community/harvest-based fishery where length (measured) (predicted) tracked through time.