The presence or absence of each pond was then tabulated for each of the 10 dates. Temporal changes in pond density were related to concurrent population density changes, adjusting population estimates using linear interpolation of the two censuses bracketing each of the 10 pond evaluation dates.