Estimation of the soil heat flux net radiation ratio based on spectral vegetation indexes in high-latitude Arctic areas

The vegetation communities in the Arctic environment are very sensitive to even minor climatic variations and therefore the estimation of surface ener gy fluxes from high-latitude vegetated areas is an important subject to be pursued. This study was carried out in July-August and used micrometeorolog ical data, spectral reflectance signatures, and vegetation biomass to estab lish the relation between the soil heat flux/net radiation (G/Rn) ratio and spectral vegetation indices (SVIs). Continuous measurements of soil temper ature and soil heat flux were used to calculate the surface ground heat flu x by use of conventional methods, and the relation to surface temperature w as investigated. Twenty-seven locations were established, and six samples p er location, including the measurement of the surface temperature and net r adiation to establish the G/Rn ratio and simultaneous spectral reflectance signatures and wet biomass estimates, were registered. To obtain regional r eliability, the locations were chosen in order to represent the different A rctic vegetation communities in the study area; ranging from dry tundra veg etation communities (fell fields and dry dwarf scrubs) to moist/wet tundra vegetation communities (snowbeds, grasslands and fens). Spectral vegetation indices, including the simple ratio vegetation index (RVI) and the normali zed difference vegetation index (NDVI), were calculated. A comparison of SV Is to biomass proved that RVI gave the best linear expression, and NDVI the best exponential expression. A comparison of SVIs and the surface energy f lux ratio G/Rn proved that NDVI gave the best linear expression. SPOT HRV i mages from July 1989 and 1992 were used to map NDVI and G/Rn at a regional scale.