Yearling seroprevalence and relation to exposure. (a) Time series of estimated yearling seroprevalence, estimated as in Figure 1. (b) Linear regression of the relative change in high-titer seroprevalence for the whole population against yearling seroprevalence. Squares show data points for particular years (labeled by the final two digits of the calendar year t). The dashed line shows the regression including all points; the dotted line shows the regression with the point for the year 2000 excluded.