E: Number of different ecosystems where the model has been applied.

N / A: Non applicable, for models that did not pass the cut-off score of 5.0, the number of countries and ecosystems was not assessed.

**Table 1.** Ranking and scores of the models included in the comparative study (with 3 or more documents in the database).

The simulation of nutrient status effects on energy exchange is based on coupling nutrient (N and P) uptake from the soil through the root to the canopy, with nutrient assimilation in the root and canopy. This coupling determines nutrient concentrations in the leaf, which in turn determines leaf carboxylation rates and hence leaf conductance.

Growth respiration is linked to expansive growth of vegetative and reproductive organs at different nodes of each shoot branch, using data on biochemistry of growth and yield to estimate coefficients to partition mobilized C, N and P. Such coefficients also depend on phenological stages. Estimated growth is then allocated to different stem internodes, leaves, and sheaths, changing their lengths, areas and volumes [30, 31]. Then, leaf and stem surfaces (heights and areas) are estimated and used to calculate irradiance interception and aerody‐ namic conductance. Root and mycrorrizhal axes (both primary and secondary) extensions are driven by growth respiration, mobilizing stored C, N and P [32].

Microbial activity in ECOSYS is represented as a parallel set of substrate-microbe complexes, which includes the rhizosphere, plant residues and animal manure, and native organic matter [32-34]. To simulate microbial growth (facultative and obligated aerobic and anaerobic heterotrophs) at an hourly step, the temperature and water contents of the litter and soil layers are used [34-36]. Temperature and moisture are derived from the energy balance calculations described above.

ECOSYS is a highly complex model with substantial calibration requirements. The strength of its approach is its flexibility, provided by a detailed representation of ecophysiological processes that allow the exploration of the ecological consequences of modifying many different environmental factors. The main weakness of this approach is that validating the accuracy of its simulation algorithms and verifying output are significant challenges, due to the difficulty of finding independent values of many ecophysiological values. In addition, its management capabilities appear limited suggesting that the model is best catagorized as a research tool.

### *3.2.2. 3-PG*

**Model Model References Time elapsed Time since**

152 10 Biodiversity in Ecosystems - Linking Structure and Function
