Be. Law et al., Leaf area distribution and radiative transfer in open-canopy forests: implications for mass and energy exchange, TREE PHYSL, 21(12-13), 2001, pp. 777-787
Leaf area and its spatial distribution are key canopy parameters needed to
model the radiation regime within a forest and to compute the mass and ener
gy exchange between a forest and the atmosphere. A much larger proportion o
f available net radiation is received at the forest floor in open-canopy fo
rests than in closed-canopy forests. The proportion of ecosystem water vapo
r exchange (lambdaE) and sensible heat exchange from the forest floor is th
erefore expected to be larger in open-canopy forests than in closed-canopy
forests.
We used a combination of optical and canopy geometry measurements, and robu
st one- and three-dimensional models to evaluate the influence of canopy ar
chitecture and radiative transfer on estimates of carbon, water and energy
exchange of a ponderosa pine (Pinus ponderosa Dougl. ex Laws.) forest. Thre
e-dimensional model simulations showed that the average probability of diff
use and direct radiation transmittance to the forest floor was greater than
if a random distribution of foliage had been assumed. Direct and diffuse r
adiation transmittance to the forest floor was 28 and 39%, respectively, in
the three-dimensional model simulations versus 23 and 31%, respectively, i
n the one-dimensional model simulations. The assumption of randomly distrib
uted foliage versus inclusion of clumping factors in a one-dimensional, mul
ti-layer biosphere-atmosphere gas exchange model (CANVEG) had the greatest
effect on simulated annual net ecosystem exchange (NEE) and soil evaporatio
n. Assuming random distribution, NEE was 41% lower, net photosynthesis 3% l
ower, total lambdaE 10% lower, and soil evaporation 40% lower. The same com
parisons at LAI 5 showed a similar effect on annual NEE estimates (37%) and
lambdaE (12%), but a much larger effect on net photosynthesis (20%), sugge
sting that, at low LAI, canopies are mostly sunlit, so that redistribution
of light has little effect on net photosynthesis, whereas the effect on net
photosynthesis is much greater at high LAIs.