How do Teleconnection Patterns Affect Carbon Exchange in the Pacific Northwest?
Overview
This study looks at the year-to-year variability in three Northern Hemispheric, Pacific teleconnection patterns to determine if CO2 and H2O fluxes at the Wind River old-growth forest are affected by climatic changes associated with these periodic events. The three patterns examined are the Pacific Decadal Oscillation (PDO), Pacific/North American Oscillation (PNA) and El Nino-Southern Oscillation (ENSO). Cool ocean/atmospheric oscillation phase events typically bring cooler and wetter weather conditions than normal to the Pacific Northwest while warm phase events are usually associated with warmer and drier conditions.
I am interested in the following questions: (1) How do teleconnection patterns affect local meteorology at the Wind River site; (2) To what extent do local meteorological anomalies caused by teleconnection patterns affect mass exchange? (3) What are the important mechanistic variables during the positive and negative leading modes of teleconnection patterns that drive interannual NEE variability? and (4) Can differences in growth and canopy structure due to meteorological variability be found at the regional scale using MODIS remote sensing data?
Methodology
This project used a combination of datasets: (1) historical, local climatological data (1919-2004), (2) long-term eddy covariance flux data from the Wind River AmeriFlux site (1999-2004), (3) ocean/atmospheric oscillation index time series data (1950-2004), and (4) MODIS fPAR and EVI products (2000-2004). fPAR is the fraction of absorbed photosynthetically active radiation through the canopy and EVI is the Enhanced Vegetation Index.
First, I examined the influence of ENSO-related (Multivariate ENSO Index, MEI) and bidecadal oscillations (Pacific Decadal Oscillation, PDO, and Pacific/North American Oscillation, PNA) on local climate variability at the Wind River AmeriFlux site. To take advantage of the climate phase additive effects (i.e., when the PDO, PNA and MEI are all in-phase), I created a new, singular index called the "composite climate index" or CCI.
CCI = PDO + PNA + MEI
Meteorological driving mechanisms, including air temperature, precipitation, vapor pressure deficit were then related to ecosystem fluxes. Differences in carbon uptake and release were examined with respect to canopy structural changes (MODIS anomalies in fPAR and EVI).
Highlighted Results
The PDO, PNA and ENSO collectively explained 90% of the variance in annual net ecosystem production (NEP) over the six years. The forest transitioned from an annual net carbon sink (NEP = + 217 g C m-2 year-1) to a source (NEP = - 100 g C m-2 year-1) during two dominant teleconnection patterns between 1999 and 2004. The carbon sink year (1999) occurred during a strong La Niña while the source year (2003) occurred during El Niño. When climate indices are in-phase, i.e. all are negative (cool) or positive (warm), the greatest anomalies in carbon flux and mechanistic variables (light use-, water use-efficiency) were observed. Annual averages were + 0.63 g C m-2 day-1 (- 0.27 g C m-2 day-1) for NEP, 3.1 mg C / g H2O (4.1 mg C / g H2O) for WUE, and 1.7 g C MJ-1 (2.1 g C MJ-1 ) for LUE during in-phase cool (warm) years. Teleconnection patterns were linked to interannual variability in the MODIS Enhanced Vegetation Index (EVI) but not to MODIS Fraction of absorbed Photosynthetically Active Radiation (FPAR). This work suggests that any increase in the strength or frequency of ENSO events coinciding with in-phase, low frequency Pacific oscillations (PDO and PNA) will increase CO2 uptake variability in Pacific Northwest conifer forests.