title: "PLPE simulations"

output:

html_document:

toc: yes

toc_float: yes

toc_depth: 4

code_folding: show

pdf_document:

toc: yes

editor_options:

chunk_output_type: console

```{r setup, include=FALSE}

The objective is to simulate the different planting designs from the density experiment in Palapa(*i.e.* PLPE) using ARCHIMED. We will then compare the light interception, carbon assimilation and transpiration at plot scale for each design to compare them with the measured yield.

The main hypothesis is that fruit yield is strongly related to carbon assimilation.

```{r echo=T, message=FALSE}

Only the ephemeris of march will be simulated for this pre-study, at half-hourly time-step, clearness of 75% (*i.e.* sunny day), and mean temperature of 24:

The time steps will start from 05:00 to 18:30 by 30 minutes time-step. A dynamic of air temperature is generated from a minimum and maximum daily temperature (18 and 30°C resp.).

The file is saved into the ARCHIMED input folders:

* Setting the path to ARCHIMED to make the simulations:

This path has to be adapted to where ARCHIMED is located in the computer.

* Saving the file:

To simulate the different designs, we need first the 3D mock-ups of the palm trees (OPF files) and their planting design (OPS files).

For the moment, the data about these trees are quite incomplete, so we compute manually the main parameters from the field data (`param_Vpalm_PLPE_design.csv`), and use the other parameters from an average palm tree of the DA1 progeny from the SMSE trial.

The scene is made using VPalmr according to the data from each design. A less advanced user may prefer to make the scenes using VPALM-IDE instead.

Three files are imported :

* The `param_Vpalm_PLPE_design.csv` data.frame. It was made from computations made on the still uncomplete PLPE data. It is why we use it as follows instead of using the built-in import functions from VPalmr or VPALM-IDE.

* The `vpalm_template.rds` file, which is a VPALM input computed from an average tree from the DA1 progeny, and which is used as a template to fill the new values, and as default values for the data that is not yet available.

* The designs files located in "1-data/PLPE_plot_design" that are used to make the OPS files.

Creating and writing the OPF and OPS:

```{r eval=TRUE}

The palm tree that is in-between the triple density row in the "TripleInRow" design is actually more similar to the ones from the control, because it experiments aproximately the same density. Hence, we change the OPS file to correspond to it.

Here are the plots of the different designs, made from repeated voronoïs:

An ARCHIMED simulation is done for each design on the same day.

We have to set the path to ARCHIMED on the computer:

Setting the meteo file to use for all simulations:

NB: you can check the value afterwards to be sure:

The general properties of ARCHIMED can be optionnally set if needed.

Here we ask for:

* 46 sky sectors for the turtle to compute the diffuse light

* 200.000 pixels for the pixel table to compute light interception

We also need to set the latitude and altitude of the plots to compute the sun position:

And finally the output directory:

We have to set the output variables we are interested in. Here we will ask for:

* stepNumber : the step ID, to compute variables for each time-step;

* stepDuration: the step duration in seconds, to integrate over each time-step;

* plantId: the plant ID to be able to compute per-plant variables;

* nodeId: the node ID, which is the id of each object in the scene (*e.g.* leaflet);

* type: the node type (*e.g.* leaflet);

* meshArea: the mesh area of each node, *e.g.* to get the palm tree leaf area;

* enb_netrad_W_m2: the net radiation of the node;

* enb_latentheat_W_m2: the latent heat of the node;

* enb_sensibleheat_W_m2: the sensible heat of the node;

* enb_transpir_kg_s: the transpiration of the node;

* temperature_C: the temperature of the node;

* photo_assimilation_rate_umol_m2_s: net photosynthesis (An) by surface unit (see photo_assimilation_umol_s for photosynthesis at the node scale);

* photo_stomatal_conductance_mol_m2_s: stomatal conductance (Gs) by surface unit.

The plot boundaries and the target OPS files are set for each simulation, and then ARCHIMED is called to make the simulation.

```{r, eval= FALSE}

Finding the output files:

Importing the meteorology file from the first simulation (all are the same):

Finding the plots area:

Importing the node values (main output):

Computing the intercepted radiation of the leaves, the assimilation, transpiration, leaf area and LAI for each design and orientation:

* At time-step scale:

* At tree scale:

```{r, message=FALSE,warning=FALSE}

* At plot scale:

* At daily scale:

* At tree scale:

* At plot scale:

```{r include=FALSE}

Setting the same colors for all plots:

Intercepted global radiation per time-step and per planting design at plot scale:

The same at tree scale:

```{r, message= FALSE, warning=FALSE, fig.height= 12}

At tree scale:

```{r, message= FALSE, warning=FALSE}

At plot scale:

```{r, message= FALSE, warning=FALSE}

The daily values at plot scale for all variables are as follow:

```{r, message= FALSE, warning=FALSE, fig.height = 12}