Are temporal variations of leaf traits responsible for seasonal and inter‐annual variability in ecosystem CO₂ exchange?

Summary 1. Seasonal and inter‐annual variability in ecosystem carbon dioxide (CO 2 ) exchange is attributed to numerous climate drivers. However, climate effects on metabolism often override ecological functions. This study seeks insight into which biological and ecological processes influence temporal patterns of ecosystem productivity in natural ecosystems. 2. The specific objectives of this study are to (i) identify seasonal and inter‐annual patterns of ecosystem‐level photosynthesis in relation to climatic conditions, (ii) examine and compare seasonal and inter‐annual variations in leaf traits for annual grasses and oak trees across multiple years, and (iii) explore interactions among leaf traits and ecosystem‐level photosynthesis across multiple seasons and years. 3. We conducted this study in a woody savanna and open grassland in California, USA. Ecosystem‐level photosynthetic rates of annual grasses ( A grass ) and oak tree canopy ( A canopy ) were deduced from eddy covariance measurements over a 7‐year period (2001 and 2007). In conjunction, we sampled grass and oak leaves at weekly to monthly intervals and constructed a multi‐year time series of leaf nitrogen concentration (N), leaf mass per unit area (LMA), leaf carbon concentration (C), and leaf carbon stable isotope discrimination (Δ). 4. Given the same grass community age or tree canopy age, inter‐annual variations of the photosynthetic rates were up to 1–2 gC m −2 day −1 for annual grasses and oak trees while the two types of vegetation were exposed to different, wide ranges of inter‐annual climate fluctuations: up to 5 °C in daily mean soil temperature, 15% in soil moisture, and 10 mol m −2 day −1 in photosynthetically active radiation. 5. While both grass and oak leaf traits varied seasonally and inter‐annually, they experienced temporal patterns and seasonal peaks that were distinct from one another. Multi‐year means of grass leaf N, C, Δ, and LMA were 2·3%, 40·8%, 22·6‰ and 71·3 g m −2 , respectively; multi‐year means of oak leaf N, C, Δ, and LMA were 1·9%, 45·1%, 20·5‰ and 132 g m −2 , respectively. 6. Based on the analysis of variance, seasonal and inter‐annual terms were associated with A grass or A canopy up to 90% or 81%. On the other hand, variations in leaf N, LMA, C, Δ, and their interactions could statistically explain about 53% and 26% of variations in A grass and A canopy, respectively. 7. We discussed possible biological and ecological processes involved in regulating seasonal and inter‐annual variability in ecosystem‐level photosynthesis. Clearly, seasonal and inter‐annual variation in ecosystem photosynthesis was strongly associated with the dynamics of leaf traits.