Premium
The use of CO 2 flux time series for parameter and carbon stock estimation in carbon cycle research
Author(s) -
Hill Timothy Charles,
Ryan Edmund,
Williams Mathew
Publication year - 2012
Publication title -
global change biology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 4.146
H-Index - 255
eISSN - 1365-2486
pISSN - 1354-1013
DOI - 10.1111/j.1365-2486.2011.02511.x
Subject(s) - series (stratigraphy) , covariance , estimation theory , equifinality , statistics , time series , data assimilation , eddy covariance , environmental science , mathematics , algorithm , atmospheric sciences , econometrics , meteorology , ecosystem , ecology , physics , biology , paleontology
Data assimilation ( DA ) is increasingly being employed to estimate the parameters and states of terrestrial ecosystem models from eddy covariance measurements of net carbon ( C ) fluxes. The length of the observation time series used varies for each study. The impact of these differences has not been quantified explicitly. Therefore, in this study, we investigate the importance of the time series length relative to observation noise and data gaps. Different length synthetic time series are used to determine the parameter and C stocks of a simple ecosystem C model. Two commonly used DA schemes are tested: the sequential E nsemble K alman F ilter (En KF ) and a batch M etropolis M arkov chain M onte C arlo algorithm. Longer time series improve both the parameter and C pool estimates of the E n KF , while adversely affecting those of the M etropolis algorithm. For both DA approaches, the length of the time series has more influence on the parameter and pool estimates than the level of random noise or amount of data. In this study, the E n KF provides more robust parameter and C pool estimates than the M etropolis algorithm. Optimized parameters and states are often used as the basis for forecasting future responses. Despite having better parameter and C pool estimates, E n KF forecasts estimates have much larger uncertainties than the M etropolis algorithm forecast estimates. Finally, we suggest that the structure of simple box models, as used in this study, introduces a large degree of equifinality into DA . Neither DA scheme correctly accounts for the equifinality, but our results suggest that it is particularly problematic for the batch M etropolis algorithm.