[Meta] Seasonal dynamics in leaf area

[davidorme]

This is a meta issue to coordinate tasks needed to implement a new model of the seasonal dynamics in leaf area.

Theoretical background

The theoretical background of this issue builds on two recent papers from Lemontree:

1. Cai et al. (2023) "A unifying principle for global greenness patterns and trends" bioRxiv. This paper establishes a simple model for global greenness as the minimum of water and energy limitations to productivity. It generates a prediction of annual maximum fraction of absorbed photosynthetically active radiation (fAPAR, $f_{APAR}$) from the predictions of the P Model for a year, plus some additional data on precipitation and aridity.
2. Zhou et al (2024) "A general model for the seasonal to decadal dynamics of leaf area" bioRkiv. This paper takes the predicted annual maximum $f_{APAR}$ from above, along with the resulting predicted leaf area index LAI, and then uses this information to calculate daily time series within modelled years of the steady state and realised LAI to give seasonal patterns.

Implementation

Boya Zhou (@zhouboya), who is an author on both papers and lead on the second has provided the following draft implementation, which uses a single site to calculate the required values, and which we should be able to use as a golden dataset for regression tests. [TODO - I'm just sorting out some issues with the data but will upload shortly].

Phenology_methods.docx

The sub-issues below break out suggested steps in that word document describing the calculations, except that I've broken out the routine for estimating growing season.

• Create a golden dataset for use in testing and development of phenology module #354
• Add a converter function from precipitation (mm/m2) to moles. #378
• Resolve issues with golden dataset to duplicate calculation #367
• Implement a routine for calculating growing season from datetime and temperature series. #352
• Calculation of annual fAPAR and LAI values from a fitted PModel and precipitation data #348
• Calculate tunable parameter $m$ defined as the ratio of steady state annual LAI to GPP. #349
• Calculate the steady state LAI. #350
• Calculate the realised LAI given a lagged process. #351

[j-emberton]

How does this new LAI calculation method interact with the two leaf model which depends on LAI? I guess as LAI changes the sunlit/shaded balance also changes?

[davidorme]

This new approach can be used to calculate a daily LAI prediction for a site, so provides an alternative route to generating LAI for use in e.g. the two leaf model.

[j-emberton]

This new approach can be used to calculate a daily LAI prediction for a site, so provides an alternative route to generating LAI for use in e.g. the two leaf model.

ah ok, I get it now

[j-emberton]

Hi @MarionBWeinzierl , would you like to pick up this work stream?