Does a general temperature-dependent Q10 model of soil respiration exist at biome and global scale?

Soil respiration (SR) is commonly modeled by a Q(10) (an indicator of temperature sensitivity) function in ecosystem models. Q(10) is usually treated as a constant of 2 in these models, although Q(10) value of SR often decreases with increasing temperatures. It remains unclear whether a general temperature-dependent Q(10) model of SR exists at biome and global scale. In this paper, we have compiled the long-term Q(10) data of 38 SR studies ranging from the Boreal, Temperate, to Tropical/Subtropical biome on four continents. Our analysis indicated that the general temperature-dependent biome Q(10) models of SR existed, especially in the Boreal and Temperate biomes. A single-exponential model was better than a simple linear model in fitting the average Q(10) values at the biome scale. Average soil temperature is a better predictor of Q(10) value than average air temperature in these models, especially in the Boreal biome. Soil temperature alone could explain about 50% of the Q(10) variations in both the Boreal and Temperate biome single-exponential Q(10) model. Q(10) value of SR decreased with increasing soil temperature but at quite different rates among the three biome Q(10) models. The k values (Q(10) decay rate constants) were 0.09, 0.07, and 0.02/degrees C in the Boreal, Temperate, and Tropical/Subtropical biome, respectively, suggesting that Q(10) value is the most sensitive to soil temperature change in the Boreal biome, the second in the Temperate biome, and the least sensitive in the Tropical/ Subtropical biome. This also indirectly confirms that acclimation of SR in many soil warming experiments probably occurs. The k value in the "global" single-exponential Q(10) model which combined both the Boreal and Temperate biome data set was 0.08/degrees C. However, the global general temperature-dependent Q(10) model developed using the data sets of the three biomes is not adequate for predicting Q(10) values of SR globally. The existence of the general temperature-dependent Q(10) models of SR in the Boreal and Temperate biome has important implications for modeling SR, especially in the Boreal biome. More detail model runs are needed to exactly evaluate the impact of using a fixed Q(10) vs a temperature-dependent Q(10) on SR estimate in ecosystem models (e. g., TEM, Biome-BGC, and PnET).Soil respiration (SR) is commonly modeled by a Q(10) (an indicator of temperature sensitivity) function in ecosystem models. Q(10) is usually treated as a constant of 2 in these models, although Q(10) value of SR often decreases with increasing temperatures. It remains unclear whether a general temperature-dependent Q(10) model of SR exists at biome and global scale. In this paper, we have compiled the long-term Q(10) data of 38 SR studies ranging from the Boreal, Temperate, to Tropical/Subtropical biome on four continents. Our analysis indicated that the general temperature-dependent biome Q(10) models of SR existed, especially in the Boreal and Temperate biomes. A single-exponential model was better than a simple linear model in fitting the average Q(10) values at the biome scale. Average soil temperature is a better predictor of Q(10) value than average air temperature in these models, especially in the Boreal biome. Soil temperature alone could explain about 50% of the Q(10) variations in both theBoreal and Temperate biome single-exponential Q(10) model. Q(10) value of SR decreased with increasing soil temperature but at quite different rates among the three biome Q(10) models. The k values (Q(10) decay rate constants) were 0.09, 0.07, and 0.02/oC in the Boreal, Temperate, and Tropical/Subtropical biome, respectively, suggesting that Q(10) value is the most sensitive to soil temperature change in the Boreal biome, the second in the Temperate biome, and the least sensitive in the Tropical/Subtropical biome. This also indirectly confirms that acclimation of SR in many soil warming experiments probably occurs. The k value in the "global" single-exponential Q(10) model which combined both the Boreal and Temperate biome data set was 0.08/degrees C. However, the global general temperature-dependent Q(10) model developed using the data sets of the three biomes is not adequate for predicting Q(10) values of SR globally. The existence of the general temperature-dependent Q(10) models of SR in the Boreal and Temperate biome has important implications for modeling SR, especially in the Boreal biome. More detail model runs are needed to exactly evaluate the impact of using a fixed Q(10) vs a temperature-dependent Q(10) on SR estimate in ecosystem models (e.g., TEM, Biome-BGC, and PnET).