The ‘model’ used was simply the sum of in-season returns plus catch, with some assumptions made (based on recent historical proportions) about how to allocate returns and catch among the 5 different runs. The in-season catch data are incomplete, so run size was increased assuming Alaska fisheries account for 3% of the final run size. This approach is not very sophisticated, but it attempts to make use of the most proximal measurements.

This and the previous box appear to be limited to 500 characters, not 500 words, so see attachment for a description. Also, since I can't seem to upload xlsx files, I have uploaded pdfs of the spreadsheet, with and without formulas displayed.