Secondary Dispersal of Tree Seeds on Snow

Abstract

1 We studied secondary seed dispersal on snow using a wind tunnel, experimental releases in clearings, and observations of the dispersal curve of an isolated Betula alleghaniensis in a large clearing. We derived a model for wind dispersal which couples the primary and secondary movement for a point source, and then used this to modify an area source model dispersal curve. 2 The wind tunnel results showed that the probability of entrainment of the seed was directly proportional to wind speed and inversely proportional to the square root of wing loading. Field releases showed that secondary distance traveled was, in turn, proportional to the probability of entrainment and to travel time. 3 The point source model overestimated seed density in the distance interval of rising entrainment probability and underestimated seed density at subsequent distances. The area source model adequately predicted dispersal for autumn vs. winter fractions of a B. alleghaniensis stand dispersing seeds into a clearing. The isolated B. alleghaniensis indicated that travel time was 10 h. 4 According to the model, secondary dispersal for trees on snow is inconsequential when the fraction of seeds that abscise in winter is < 0.15 (conifers and most angiosperms). However, it will have an important role for genera such as Betula which have large winter-abscising crop fractions. 5 Further, we note that the relative importance of secondary dispersal is almost solely an inverse function of release height, and therefore for very small statured plants (e.g. herbs), secondary dispersal should display a dramatic effect even if only a small portion of the crop is involved. Finally, we argue that wind speeds near the surface within a forest are so low that the effect of secondary dispersal within forests will be minor.