Establishment of a black cottonwood plantation on an exposed site in Iceland: Plant growth and site energy balance

An experimental site for monitoring plant-atmosphere interactions during development of a black cottonwood stand was established in Gunnarsholt, south Iceland in 1989. The main objectives are to study the long-term modification of microclimate in this developing tree canopy, located on an exposed site, and the effects of microclimate on tree physiology and growth. The experimental site in Gunnarsholt meets the most stringent criteria of energy balance measurement. It is a 14.5-ha single-clone plantation of black cottonwood (Populus trichocarpa) situated within a flat, treeless landscape with homogeneous aeolian soils. Instrumentation on a tower located in the center of the plantation includes a range of sensors whose outputs are monitored by a datalogger for continuous recording of the site's microclimate. Plant responses have been measured at both annual and within-season time scales. This paper presents early results on the biometeorological characteristics and energy balance conditions of the site, and on the plant responses. The radiation efficiency at the site was conservative at a value of 0.50 (1991 through 1994), but the reflection coefficient increased from 1993 through 1995 as the canopy and understorey developed. Seasonal pattern of the reflection coefficient was characterized by steadily increasing values up to mid-August, at which point the trend turns downwards during the senescence phase on the site. The maximum reflection coefficient was 0.22 in 1994 and 1995, but only 0.18 in 1993. Net radiation in 1993 and 1994 was highly variable with maximum daily totals close to 15 MJ m(-2) day(-1) and a minimum around 5 MJ m(-2) day(-1). Bowen ratio values for July 1993 ranged from 0.5 to over 2, but during most of the 1994 growing season the Bowen ratio was less than 1, i.e. latent heat flux dominated the energy balance. This pattern was probably the result of higher and more frequent precipitation during the measuring period in 1994 than in 1993. Height growth of the black cottonwood saplings was slow and staggered until 1994 when they doubled their height, from 30 cm to 62 cm. The growth increase in 1994 coincides with a relatively warm growing season (mean air temperatures in July and August 1994 were 12.1 degrees C and 10.7 degrees C, respectively). Above-ground biomass of the plants increased exponentially from 4.5 g m(-2) in the fall of 1991 to 83 g m(-2) in 1994. Maximum leaf area index (LAI) at the site was 1.5 in August 1994, but leaf area of the black cottonwood at the end oo the 1994 growing season was only 0.3 m(2) m(-2), indicating that black cottonwood contributed relatively little to the site's LAI. Cover estimates support this, and black cottonwood plants covered less than 10% of the total surface area until late in the 1994 growing season. Thus. the period described in this paper reflects the starting point of canopy development before the trees start to significantly alter tile site energy balance.