PS 49-54
On quantifying net ecosystem carbon exchange from adjacent forests affected by drainage flow using process network approach

Thursday, August 14, 2014
Exhibit Hall, Sacramento Convention Center
Minseok Kang, National Center for AgroMeteorology, Seoul, Korea, Republic of (South)
Benjamin L. Ruddell, Fulton Schools of Engineering, Arizona State University, Mesa, AZ
Juyeol Yun, Department of Landscape Architecture and Rural Systems Engineering, Seoul National University, Seoul, Korea, Republic of (South)
Bindu Malla Thakuri, Department of Atmospheric Sciences, Yonsei University, Seoul, Korea, Republic of (South)
Sehee Kim, Interdisciplinary Program in Agricultural and Forest Meteorology, Seoul National University, Seoul, Korea, Republic of (South)
Chunho Cho, Forecast Research Division, National Institute of Meteorological Research, Jeju, Korea, Republic of (South)
Jung-Hwa Chun, Department of Forest Conservation, Korea Forest Research Institute, Seoul, Korea, Republic of (South)
Joon Kim, Landscape Architecture and Rural Systems Engineering, Seoul National University, Seoul, Korea, Republic of (South)
Background/Question/Methods

Observation of net ecosystem exchange (NEE) of carbon dioxide (CO2) by the eddy covariance (EC) technique in hilly terrain affected by drainage flow is a difficult problem because the lateral or advective CO2 flux is not negligible in these circumstances. For the flux tower sites on a hill slope there is a high probability that nighttime drainage of CO2 results in overestimation of respiration at the bottom of a hill slope, and underestimation at the top. A multi-level CO2 concentration vertical profile has been measured along with eddy covariance flux measurements at the middle and end of a hill slope i.e., the GDK and GCK flux sites of KoFlux (Korea Flux Monitoring Network), for 2008 and 2009. In order to test the hypothesis that drainage of CO2 at night links the upper to the lower hill slopes, we constructed an information flow process network based on the observed CO2 concentrations and fluxes. This test infers advective CO2 coupling at specific times of day and year by observing covariation in the dynamics of the concentrations and fluxes.

Results/Conclusions

We found evidence of CO2 drainage from the GDK to the GCK. The net information flow from the GDK to the GCK was positive during nighttime and negative during transition period around sunrise. Additionally, the magnitude of the net information flow increased with time lag, peaking at four hour timescale. We also identified the major paths of the CO2 flows between the different canopy layers of the hill slope which showed the primary information flows. We are applying the findings to develop a correction for NEE at the sites.

Acknowledgment. This research was supported by the Korea Meteorological Administration Research and Development Program under Grant CATER 2014-3030.