Background

Forests play a major role in the terrestrial carbon cycle through CO₂ assimilation and respiration, as well as on the Earth climate by influencing atmosphere CO₂ concentration (Luyssaert et al., 2007; Bonan, 2008; Canadell and Raupach, 2008; Reichstein et al., 2013). Gross primary productivity (GPP), plants’ carbon uptake through photosynthesis, is the ultimate source of organic material in land biosphere in general and in food production in particular.

GPP at the ecosystem scale is mainly derived using the eddy covariance (EC) technique, as the difference between EC-measured net ecosystem CO₂ exchange (NEE) and the daily inferred ecosystem respiration (Re). The latter one is obtained by extrapolating measured night-time NEE, which equals to ecosystem respiration, Re, to daytime based on empirical equations of Re response to temperature and soil humidity (Aubinet et al., 2000; Baldocchi, 2003; Reichstein et al., 2005; Grunzweig et al., 2009). However, the EC approach has several limitations and uncertainties. Its application is limited to relatively large homogeneous and flat terrains. EC technique is critically dependent on factors such as the system deployment in field (e.g., height above the canopy), on atmospheric conditions (e.g., turbulence conditions, advections) (Aubinet et al., 2000), on corrections applied to data processing programs and algorithms. A common way to assess EC measurements reliability is through the evaluation of the ‘energy closure’ over the measured ecosystem. This test indicates in most sites an energy gap of 20% or more (Foken, 2008). Empirical extrapolation of the night-time NEE measurement to approximate daytime Re has by itself significant uncertainties (e.g., Van Gorsel et al., 2009). NEE measurements provide a whole ecosystem flux only, thus explicit carbon uptake by trees cannot be distinguished from others ecosystem layers. EC measurements are also relatively expensive.

The current protocol describes an alternative method for calculation of forest trees GPP based on measurements of air temperature and humidity, trees’ sap flow (SF) rate and intrinsic water use efficiency (WUEi, the ratio A/g_s), where A is the rate of photosynthetic CO₂ assimilation and g_s is stomatal conductance. This protocol is based on the recently published study of Klein et al., 2016. WUEi is a parameter characterizing plant species, with a seasonal variation (Seibt et al., 2008; Klein et al., 2013 and 2016). It can be calculated from the carbon isotope ratio (δ¹³C) in the assimilated carbon of plant tissues (Farquhar and Richards, 1984). This is combined with sap flow measurements, which can be obtained easier and at a lower cost than EC measurements, and often have less spatial limitations (independent on ecosystem footprint, topography, homogeneity, etc.), and enable GPP estimation in complex ecosystems. The method is applicable to calculate GPP of woody vegetation and can be considered as total ecosystem GPP if the contribution of understory layer to the total ecosystem GPP can be neglected (e.g., dry environments) or independently assessed. Note that using this approach can also be applied to archived SF and ¹³C data to reconstruct past variations in GPP, as long as these data are available.