ABSTRACT 17
Bounding the Soil Respiration Models:
Lessons from Field Measurements
Ye Qi and Ming Xu
|
|
|
Soil carbon emission is a major source of uncertainty in estimation of terrestrial carbon budget. The simulation result of soil respiration generated with some major ecosystem models may vary substantially depending on the values of model parameters. We examined the uncertainty bounds of these ecosystem models based on the spatial and temporal variability in a key model parameter, the temperature sensitivity of soil respiration, as revealed by analyses of data from long-term, systematic field measurements. We first demonstrated the large ranges of the spatial and temporal variations of Q10, an index for the temperature sensitivity, based on our field measurement at a forest ecosystem in the Sierra Nevada, California. The Q10 values derived from soil temperatures at 10cm depth ranges of 1.21-2.63 with averages of 1.7. Q10 also showed a strong seasonal variation with the annual minimum occurring in mid summer and maximum in winter. Soil temperature and moisture explained 93% of the seasonal variation in Q10. The Q10 variations calculated from different locations and from soil temperature depths have significant effects on the simulation results. These variations tend to affect the seasonality more than on the annual average. Our simulations indicated that the variations of Q10 and its dependence on soil moisture and temperature had important implications for regional and global ecosystem carbon modeling, in particular for predicting the responses of terrestrial ecosystems to future global warming. We showed in this study how we should and could set a confidence bound on the results of the so-called process-based ecosystem models.
|
Contact Author: Ye Qi, University of California, Berkeley, Department of ESPM, 135 Giannini Hall, Berkeley, CA 94720-3312, (510) 643-0259, yqi@nature.berkeley.edu
|
|