Changes between Version 5 and Version 6 of DevelopmentActivities/ORCHIDEE-hydraulicArch


Ignore:
Timestamp:
10/09/21 16:30:13 (4 months ago)
Author:
peylin
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • DevelopmentActivities/ORCHIDEE-hydraulicArch

    v5 v6  
    11= Hydraulic architecture improvements = 
    22 
    3 == Meeting the 07 October 2021: comparison of Julien and Yitong schemes =  
     3== Meeting the 07 October 2021: comparison of Julien and Yitong schemes==  
    44 
    55The purpose of the meeting was to compare Yitong Yao’s model with Julien Alléon’s one in order to i) Understand the differences between both models and ii) prepare a strategy in order to implement only one “mixed” model in ORCHIDEE_trunk. 
     
    3434 
    3535'''Discussion:''' 
    36 Both models answer to a precise need which led to the previous choices. 
    37 Both models seem to have good results on the site studied. 
    38 Several options are possible in ORCHIDEE, one would be a flag to activate one scheme or the other. This, at the end, should be avoid in order to have only one working model. 
    39 The best option would be to implement a mixed version of both models keeping a resolution without iterations and introducing the dynamic resistances. 
    40 Looking at the previous table: 
    41 •       Root absorption: Best option would be to activate a flag to choose between the three models; 
    42 •       Transport: Best option would be to have oth dynamic resistances and capacitances ; 
    43 •       Stomatal conductance: Best option would be to have only one stomatal conductance (this option leads to instabilities as mentioned in Julien’s model; 
    44 •       Resolution method: Best option would be to avoid iterations as much as possible. Moreover, recalculating the energy budget has to be avoid in order to remove iterations in mleb. 
    45 To avoid instability, several options were given: 
    46 •       Look more precisely at the storage pools that should act as a buffer; 
    47 •       Look at Krinner et al. (2005) were a method to smooth the curves is described; 
    48 •       Improve the estimation method. 
     36 
     37Both models seem to have good results on the site studied. Several options are possible in ORCHIDEE, one would be a flag to activate one scheme or the other (but hard to maintain). The best option would be to implement a mixed version of both models keeping a resolution without iterations and introducing the dynamic resistances: 
     38 - Root absorption: Best option would be to activate a flag to choose between the three models; 
     39 - Transport: Best option would be to have oth dynamic resistances and capacitances ; 
     40 - Stomatal conductance: Best option would be to have only one stomatal conductance (this option leads to instabilities as mentioned in Julien’s model; 
     41 - Resolution method: Best option would be to avoid iterations as much as possible. Moreover, recalculating the energy budget has to be avoid in order to remove iterations in mleb. 
     42 - To avoid instability, several options were given: 
     43  - Look more precisely at the storage pools that should act as a buffer; 
     44  - Improve the "predictor / corrector" approach 
     45   
    4946Dynamic resistances are crucial for the mortality calculation (which is wanted in the next version of the trunk. 
    50 Organisation : 
    51 Yitong is entering the final stretch of her PHD and will not have that much time to allocate to the project before March 2022. It would be interesting that she only does a simulation with a precise configuration in order to look at the time calculation added by the iterations. This can be done over Hesse. 
    52 Julien will look closer on how to implement the model and compare th 
     47 
     48'''Organisation:''' 
     49Yitong is entering the final stretch of her PHD and will not have much time to allocate to the project before March 2022. It would be interesting that she only does a simulation with a precise configuration in order to look at the time calculation added by the iterations. This can be done over Hesse. 
     50Julien will look closer on how to implement the two models and compare them. 
    5351 
    5452