Temporal and spatial variability of soil respiration in a cool temperate broad-leaved forest 1. Measurement of spatial variance and factor analysis

Soil respiration rates and their relating micro-environmental factors were measured in September in 1999 and 2000 to clarify the controlling factors of soil respiration rates in a cool temperate broad-leaved deciduous forest. The study site was located on a southwestern slope of Mt. Norikura at about 1,400 m asl. The site contained an experimental area of one hectare, subdivided into 100 sub-plots, each 10 m by 10 m in size. The rate of soil respiration was measured by the closed chamber method. The micro-environmental factors examined included temperature (air, soil), soil water contents, CO₂ concentration, biomass (tree, understory vegetation), soil nitrogen and carbon contents, thickness of soil layers, and topographic conditions. Results of multiple regression analysis were as follows: 1) Rate of soil respiration was strongly affected by topographic conditions, being 5-12% higher at the summit of the hill and 2.5% lower at the bottom of the valley, compared with the average. 2) Multiple regression analysis of 2-year data using the stepwise method clarified that V_SR=27.94 X_ts-0.58 X_CO2+3.26 X_a+17.89 X_l-1.74 X_w+118.44 (R²=0.707) where V_SR is the soil respiration rate, X_ts the soil temperature at - 10 cm, X_CO2 the CO₂ concentration at 0 cm, X_a the relative altitude, X_l the thickness of litter layers and X_w the soil water content. 3) The measured soil respiration rate was high in 1999 (408.2 mgCO₂m^-2h^-1 on average) and low in 2000 (354.0 mgCO₂m^-2h^-1 on average), reflecting previous rainfall. The soil water contents in 1999 and 2000 were 33.6% and 20.8%, respectively. Environmental factors in 1999 were still unstable after plenty of rainfall. The high moisture content in 1999 seemed to conceal the effects of other micro-environmental factors.