alphaA - PIK-LPJmL/LPJmL GitHub Wiki

scales leaf-level photosynthesis as computed by the simplified Farquhar model (see Collatz et al. 1991 and 1992) to the biome level (or agricultural stand level). It basically describes the fraction of the total area that is like a leaf, but at the same time it is the main carbon tuning factor of the model. However, Haxeltine and Prentice 1996a (see mainly page 556, therein) suggest that the scaling of light_limited J_E is not needed for agricultural systems, see also Haxeltine and Prentice 1996b.

is 0.5 for natural vegetation and can reach values up to 1.0 for agricultural stands.
The current R-script-for-LAI-max-calibration can be used to calibrate LAI_max to data of agricultural production intensity. It uses a combined calibration of LAI_max to describe plant performance and density, to describe the homogeneity of the stand and the harvest index, where the latter focuses on high-yielding varieties in intensely managed stands only.

Haxeltine and Prentice 1996a hypothesize that this scaling is necessary because natural systems are limited by water and nutrients, but we do model water stress and still need that factor of 0.5. Also after implementing nitrogen limitation, this factor is still needed (even though adjusted to 0.7 in LPJ-GUESS and 0.6 in LPJmL). However, after implementing the GSI phenology for PFTs cf. Forkel et al. 2014 a parameter optimization against remotely sensed fAPAR data was run and found PFT-specific values, ranging from 0.3 for the polar herbaceous PFT to 0.65 for the tropical broadleaved evergreen PFT. Alternative hypothesis is that natural ecosystems are in part different from a big leaf (with various aspects of heterogeneity) as we’re missing all kinds of other disturbances such as herbivores, whereas intensely managed cropland may in fact be very close to one homogeneous big leaf. Therefore, for each major model development such as the implementation of the nitrogen cycle, a new parameter optimization including is recommended.

Haxeltine and Prentice 1996b (page 695, text and table 2) describe as an ’empirical parameter that accounts for reductions in PAR utilization efficiencies in natural ecosystems’ and assign the value of 0.5 based on data summarized by Landsberg (1986). It is thus a scaling factor for the quantum efficiency parameters and but is implemented to scale APAR in the model, which basically does the same (see photosynthesis).

• Haxeltine, A, and IC Prentice. 1996a, A General Model for the Light-Use Efficiency of Primary Production, Functional Ecology, 10, 551-561, doi: 10.2307/2390165.
• Haxeltine, A, and IC Prentice. 1996b, BIOME3: An equilibrium terrestrial biosphere model based on ecophysiological constraints, resource availability, and competition among plant functional types, Global Biogeochemical Cycles, 10, 693-709.
• Landsberg, JJ. 1986, Physiological Ecology of Forest Production, Academic Press.