Python Program for computing leaf conductance

Current version is: 1.5

NEW: you can now provide dry and fresh weight in the data frame to compute RWC: see sections program flow step3 and data format below

How to install?

Install Python version if needed

Anaconda

Miniconda

Download full folder from git

1. Direct download

From the green box named 'clone' in the right corner > download .zip

2. From the terminal

git clone https://github.com/xbouteiller/gminComputation.git

Install dependencies

pip install -r requirements.txt

Install package

Open a terminal in the DetectEvent folder, then :

python setup.py install

Program Execution

Copy the file gminExec.py in the desired folder

Then open a terminal

python gminExec.py

I also provided some additonal for simplifying execution in the exec folder

if you are on a windows platform:

• create a shorcut for the launch.bat file
• place the shortcut in an ampty folder and double click on it
• the program will be executed in the terminal and files & figures will be saved in the empy folder

if you are on a linux platform:

1. in the bash-ex-linux.sh file
   • replace the path /home/xavier/anaconda3/bin/python with the path linking to your python version
   • if you use anaconda you should have to replace only xavier with you user name
   • replace the path /home/xavier/Documents/development/gmin/gminExec.py with the correct path linking to the gminExec.py file
2. in the launch-linux.desktop
   • replace the path /home/xavier/Documents/development/gmin/exec/bash-ex-linux.sh with the correct path to bash-ex-linux.sh
   • you can now copy the launch-linux.desktop file in a new folder and double click on it

Installing updates

git pull origin main

python setup.py install

Program flow

Step 1

1. The program ask to chose:
   • A folder that will be parsed
   • Some unique ID from a single file

The program writes the number of files found

Step 2

2. You have to chose which method you will use:
   • Select time interval manually then compute gmin in the interval time
   • Filtering the data based on RWC and then compute gmin in the interval time

Step 3 - Option 1

3. If you chose the manual points selection:
   • You have to select two points on the curve
   • Gmin is computed based on a linear regression between these two points

Step 3 - Option 2

4. If you chose the method based on RWC:

   • the semi auto method will plot curve each time
   • the full auto will precede to the gmin computation automatically

5. The data are first filtered based on RWC:

Default values for the RWC filtering are 80% and 50%, but thsi can be changed manually:

python gminExec.py --rwc_sup 90 --rwc_inf 20 # Superior threshold : 90%, inferior : 20%

python gminExec.py -rs 90 -ri 20 # It is a shortcut for the code above

6. New if the columns Dry_weight and Fresh_weight are provided, he software use the provided values to compute RWC

7. Gmin is computed based on a linear regression between the two boundaries of the RWC filtered data

Step 5

8. Synthetic figures and data frames are saved within the output_fig and output_files folder

Data format

Data must be stored within files For a better files recognition, first row of the csv file should contain the string "conductance" otherwise all csv from a folder will be parsed

Columns should be named as follows:

Quantitative columns

• weight_g : leaf weight as a function of time (g)
• T_C : temperature (°C)
• RH : Relative Humidity
• Patm : atmospheric pressure (KPa)
• Area_m2 : area of the leaf (m2)
• Fresh_weight : fresh (saturated) weight of the leaf (g)
• Dry_weight : dry weight of the leaf (g)

Qualitative columns

• campaign : campaign name
• sample_ID : ID of the sample, should be unique for each sample

Date

• date_time : time (best with the format YEAR/MONTH/DAY HOUR:MINUTE )