A comparison of methods for estimating juvenile salmon habitat capacity to assist with restoration planning and evaluation.

This article compares seven analytical methods for estimating juvenile Chinook Salmon (Oncorhynchus tshawytscha) habitat rearing capacity, abundance, and habitat suitability within a 3.2-km reach of the Grande Ronde River, Oregon. The methods include statistical models such as quantile random forests (QRF) and the Fish Data Analysis Tool (FDAT), deterministic models like the habitat suitability index (HSI) and the unit characteristic method (UCM), and simpler habitat expansion approaches. Results showed wide variation in estimates, ranging from 1,048 to 24,530 juvenile salmon, with simple habitat expansion methods aligning more closely with snorkel survey counts (4,179 juveniles). The study concludes that habitat expansion methods are particularly suitable for reach-scale restoration planning, while model-based approaches like QRF and FDAT are better for watershed-scale assessments and prioritization, emphasizing the need for consistent, region-specific fish density and habitat data to improve accuracy.