1999 Volume 12 Issue 1 Pages 28-38
Characteristicso f spectral reflectanceh ave been studied for snow and vegetation, w hich are important elements for heatand water budget at ground surfaces.U sing this characteristicsw, e have investigated methods to estimate the vegetation density and to distinguish the snow existence. The spectral reflectance was measured under various conditions of the vegetation density and snow cover with movable potted seedlings, a ssuming high latitudalf orests with snow cover. We found that multiple scatteringi n the forest must be considered to simulate the spectralr eflectance, a fter the investigationw ith simple models. Next, we proposed vegetation indices to know the status of the vegetation and snow indices to decide the snow existence using remote sensing.P revious representativev egetation indicesa re using one visiblea nd one near infrared wavelength band, and they can estimate the activityo f the vegetation.H owever, the value of these indices is variable for the same vegetation if the forest floort ype changes, especiallys now cover. We found that the proposed vegetation indices could show the vegetation density (LAI<3) using three wavelength bands (one visiblea nd two near infrared bands), because they depended not on surface elements (soil type and snow) but on only LAI. Moreover, the snow indices can distinguish the snow existence on the forest floor when LAI<3.
