Interannual variation of water balance and summer evapotranspiration in an eastern Siberian larch forest over a 7-year period (1998-2006)

Water vapor, energy fluxes, and environmental conditions were measured in an eastern Siberian larch forest for 7 water years, from 1998 to 2006, to understand the water-balance characteristics and interannual variation (IAV). The latent heat flux accounted for 38¿67% of the sum of turbulent heat fluxes in June, July, and August, a relatively moderate fraction was compared to values measured at mid- and low latitudes. More than 70% of the annual precipitation evaporated during May to September. Annual evapotranspiration, including interception loss, was relatively steady at 169¿220 mm compared with the wide range in annual precipitation (111¿347 mm year¿1). The evapotranspiration rate was 1.49¿2.30 mm day¿1 on a daily basis from May to September above a dry canopy. This feature is one of the remarkable characteristics of the water balance in eastern Siberian forests. The thawing depth of the permafrost has been rapidly deepening since 2004, such that the maximal thawing depth varied from 127 cm before 2003 to over 200 cm after 2004. At the same time, there was a very large increase in the moisture content of the surface soil. This increase could not be explained by the amount of annual precipitation alone and may have been due to inflow from the deeper thawing layer. The IAV of evapotranspiration was small, but the yearly evapotranspiration coefficient (the ratio of evapotranspiration to potential evaporation) ranged from 0.30 to 0.45. These results indicate that the IAV of evapotranspiration is controlled by regulation of the land surface rather than by atmospheric demand. Soil-moisture content was the most important variable among the factors determining the evapotranspiration coefficient at an interannual temporal scale. This result differs somewhat from previous satellite-based findings that air temperature was a major variable for plant activity. This difference might result from the fact that the IAV of soil water content did not correspond to that of the precipitation amount because of the presence of the permafrost. By contrast, the soil water content was strongly affected by precipitation in the previous summer