Simulation of land surface albedo—a case study

Surface albedo determines the distribution of solar radiation between the earth’s surface and the atmosphere. It affects the global climate directly by altering surface energy balance, and indirectly by controlling ecosystem processes and greenhouse gas exchange. In this study, a land surface albedo model was constructed based on the probability approach for ray tracing and the basic optical parameters of ecosystem elements. The model was applied to a boreal aspen forest and results were compared with field measurements. Results show that seasonal and diurnal albedo dynamics were well simulated by the model. The standard deviation between the simulated and measured reflected radiation was 2.5-5.0 Wm-2 in different seasons. The model results also indicate that (1) the diffuse radiation albedo in winter is higher than the direct radiation albedo. As a result, the average albedo under clear conditions tends to be lower than overcast conditions; (2) the impact of SZA on surface albedo is higher during the growing season than in spring and fall; (3) the visible band albedo in winter is lower than the NIR albedo in the early morning and late afternoon, but higher around noontime. Daily averages of the two broad bands are close. However, in spring and fall, the visible band albedo is less than half of the NIR albedo. Furthermore, in summer, the surface albedo is mainly dominated by the NIR albedo and the contribution of the visible band albedo is very small. These seasonal variations in wavelength-albedo relationships could have profound influences in remote sensing applications such as radiance ratio based vegetation indices.

Model sensitivity analyses show that the surface albedo of this boreal aspen forest is less sensitive to the canopy architecture parameters, such as stand crown shape, stand density, and leaf inclination angle. This is mainly because this mature aspen forest has a high stand density with a large LAI. These ecosystem characteristics also result in the low sensitivity of summer surface albedo to variations in LAI, bark reflectance, and soil reflectance. However, wood area index is found to greatly affect surface albedo during leafless seasons, particularly in winter when the underlying ground is covered by snow. The model results suggest that accurate estimates of WAI are essential to improve the accuracy of surface albedo simulation in leafless seasons. Variations in the reflectance of plant branches and ground can significantly affect the surface albedo in leafless seasons, and variations in the leaf optical parameters can significantly affect that in the growing season. Since leaf optical features can change with ecosystem conditions, such as plant water status, the higher albedo observed from satellite over the Canadian prairies in the early summer of 2001, which experienced very dry conditions, can be partially attributed to the water stressed canopy. This means that the impact of climate on ecosystems could result in significant feedbacks through albedo change.