Abstract: Aims Soil respiration may have distinct dynamic patterns at different temporal scales since it is affected by diverse
abiotic and biotic factors. Seasonal variation in soil respiration is largely controlled by abiotic factors such as
temperature and soil moisture, whereas the regulation of diurnal variation is likely physiological rhythms of plants. Our
objectives were to compare seasonal and diurnal patterns of soil respiration and to evaluate relationships between soil
respiration and temperature at annual and diurnal scales. 
Methods We examined seasonal variations of soil respiration using infrared gas analyzers at monthly or bimonthly intervals
from April 2004 to March 2005, and diurnal variations in July, September and November 2004 as well as in January, March and
May 2005 in a montane evergreen broad-leaved forest in Ailao Mountains, China. Soil temperature, air temperature, soil water
content and air humidity were measured at the same time. We evaluated Q10 values of soil respiration and correlations between
soil respiration and soil temperature. 
Important findings Soil respiration fluctuated with distinct seasonal and diurnal patterns. Soil respiration was higher in
the wet season (May through October) than in the dry season (November through April). Diurnal patterns of soil respiration
varied among seasons. The mean rate of soil respiration was higher in nighttime than in daytime in July, September, January
and March, but lower in November and May. On the whole year basis, soil respiration correlated strongly with soil temperature
and soil water content. However, on a diurnal scale, these regressions were not significant. Q₁₀ values were 4.48, 7.17 and
2.34 for the whole year, dry season and wet season, and their corresponding soil temperature ranges were 5.9-16.6, 5.9-11.0
and 10.3-16.6 ℃, respectively. Our results demonstrate that biotic and abiotic factors have distinct impacts on soil
respiration at different temporal scales in the forest. Estimation on daily and annual car bon fluxes based on instantaneous
measurements of soil respiration, rather than 24 hour measurements, may cause severe deviation from actual values because of
the lack of diurnal correlation between soil respiration and temperature.