Scaling
=======

In this example, the time evolution of type **scaling** is demonstrated.
In the scenario, the emissions increase linearly from the year 2019 to the year 2039.
The emissions in 2039 are set to be twice as much as in 2019.

Imports
-------
If the openairclim package cannot be imported, make sure that you have installed the package with pip or added the oac source folder to ``PYTHONPATH``.

.. jupyter-execute::

    import xarray as xr
    import matplotlib.pyplot as plt
    import openairclim as oac

    xr.set_options(display_expand_attrs=False)


Input files
-----------

In order to be able to execute this example simulation, three types of input are required.

* Configuration file `scaling.toml`
* Emission inventories 

    * `ELK_aviation_2019_res5deg_flat.nc`
    * `ELK_aviation_2039_res5deg_flat.nc`

* Time evolution file for scaling: `time_scaling_linear_2019-2039.nc`

Emission inventories
^^^^^^^^^^^^^^^^^^^^

* Source: DLR Project EmissionsLandKarte (`ELK`_)
* Resolution down-sampled to 5 deg resolution
* Converted into format suitable for OpenAirClim
* Inventory years 

    * 2019 (original)
    * 2039 (same inventory as original, only year changed)

.. _ELK: https://elkis.dlr.de/

Time evolution
^^^^^^^^^^^^^^

* Time evolution with **scaling** of emissions
* Time period: 2000 - 2050
* Linear ramp-up between years 2019 and 2039

.. jupyter-execute::

    evo = xr.load_dataset("source/demos/input/time_scaling_linear_2019-2039.nc")
    display(evo)

    fig, ax = plt.subplots()
    evo.scaling.plot(ax=ax)
    ax.grid(True)


Simulation run
--------------

.. jupyter-execute::

    oac.run("source/demos/02_scaling/scaling.toml")


Results
-------

Time series
^^^^^^^^^^^

* Emission sums
* Concentrations
* Radiative forcings
* Temperature changes

.. jupyter-execute::

    results_ds = xr.load_dataset("source/demos/02_scaling/results/scaling.nc")
    display(results_ds)

.. jupyter-execute::

    # Plot Radiative Forcing and Temperature Changes

    ac = "TOTAL"
    rf_cont = results_ds.RF_cont.sel(ac=ac) * 1000
    rf_co2 = results_ds.RF_CO2.sel(ac=ac) * 1000
    rf_h2o = results_ds.RF_H2O.sel(ac=ac) * 1000
    dt_cont = results_ds.dT_cont.sel(ac=ac) * 1000
    dt_co2 = results_ds.dT_CO2.sel(ac=ac) * 1000
    dt_h2o = results_ds.dT_H2O.sel(ac=ac) * 1000

    fig, ax = plt.subplots(ncols=2, figsize=(10,5))
    ax[0].grid(True)
    ax[1].grid(True)
    rf_cont.plot(ax=ax[0], color="deepskyblue", label="cont")
    rf_co2.plot(ax=ax[0], color="k", label="CO2")
    rf_h2o.plot(ax=ax[0], color="steelblue", label="H2O")
    dt_cont.plot(ax=ax[1], color="deepskyblue", label="cont")
    dt_co2.plot(ax=ax[1], color="k", label="CO2")
    dt_h2o.plot(ax=ax[1], color="steelblue", label="H2O")
    ax[0].set_ylabel("Radiative Forcing [mW/m²]")
    ax[1].set_ylabel("Temperature Change [mK]")
    ax[0].legend()
    ax[1].legend()