Life sciences · the experiment
Watch mammals through birds
H5N1 bird flu in wild birds and in mammals, four years of US records, a month-by-month model. Four questions, from beginner to expert.
1 · Say what is shown
Beginner
What does the study show?
2 · Quadruple the birds
The model is multiplicative: doubling (1 + the month's bird detections) multiplies expected mammal detections by 1.579, that is +57.9%.
Intermediate
If (1 + bird detections) quadruples in the same month, by how much do expected mammal detections rise?
3 · Separate association from cause
The model accounts for the yearly cycle, for the previous month's mammal detections and for lasting differences between states.
Expert
Why does the paper still speak of association rather than transmission?
4 · Measure the overprediction
Over February–July 2026, the model with birds predicted 167.3 mammal detections, the model without birds 220.3. 101 were observed.
Expert
By how much does the model with birds overpredict the observed total?
Surveillance targeted
—