The importance of canopy structure in controlling the interception loss of rainfall: Examples from a young and an old-growth Douglas-fir forest

The canopy water storage capacity ( S), direct throughfall fraction ( p), the ratio of evaporation to rainfall intensity ( E ¯ / R ¯ ) and interception loss ( I n), of a Douglas-fir forest are influenced by short (seasonal) and long-term (decades to centuries) changes in the forest canopy. Gross precipitation (P G) and net precipitation (P n) were measured in a young (25-year-old) Douglas-fir forest and the results compared with measurements previously made in a nearby old-growth (>450-year-old) Douglas-fir forest [Link, T.E., Unsworth, M.H., Marks, D., 2004. The dynamics of rainfall interception by a seasonal temperate rainforest. Agric. Forest Meteorol. 124, 171–191.]. Canopy rainfall variables were estimated using a regression-based method that estimates S, p and E ¯ / R ¯ for individual storms using the relationship between P G and P n. The individual storm estimates of S, p and E ¯ / R ¯ for the young forest were applied to a rainfall interception model (Gash model [Gash, J.H.C., 1979. An analytical model of rainfall interception by forest. Q. J. R. Meteorol. Soc. 105, 43–55.]) to determine the effect of seasonal changes in canopy hydrologic variables have on estimates of I n (young forest only). The Gash model was previously applied to the old-growth forest [Link, T.E., Unsworth, M.H., Marks, D., 2004. The dynamics of rainfall interception by a seasonal temperate rainforest. Agric. Forest Meteorol. 124, 171–191.]. The young forest had significantly different S (1.40 mm ± 0.27) and p (0.12 ± 0.07) relative to the old-growth forest ( S = 3.32 ± 0.35; p = 0.42 ± 0.07). Seasonal variation in canopy structure, such as deciduous leaf senescence and coniferous needle drop, were correlated with decreases in S. The differences in S and p between the two forests resulted in an I n that was only slightly larger in the old-growth forest because the E ¯ / R ¯ for the two forests were similar (young = 0.18 ± 0.06; old-growth = 0.17 ± 0.08). E ¯ / R ¯ in the young and old-growth forests may have been similar because developmental changes associated with old-growth forest may alter the external resistance ( r a) and the effective area for evaporation. The Gash model successfully predicted I n for the young forest on a seasonal basis (3.29% error), but experienced larger errors (range = −91 to 36% error) for individual storms. The seasonal error and the error for individual storms improved when seasonal variations in canopy characteristics were incorporated in the mo