Add trend examples

evaluations/scripts/trends-evaluation.py

@@ -0,0 +1,128 @@
+output_dir = <output_dir>
+exp_pi = <your name>
+
+data_dir = output_dir + "/data"
+coldata_dir = output_dir + "/coldata"
+proj_id = "tutorial"
+exp_id = "emep-trends"
+
+
+DEFAULT_RESAMPLE_CONSTRAINTS = dict(
+    yearly=dict(monthly=9),
+    monthly=dict(
+        daily=1,
+        weekly=1,
+    ),
+    daily=dict(hourly=1),
+)
+
+
+CFG = dict(
+    # Output directories
+    json_basedir=data_dir,
+    coldata_basedir=coldata_dir,
+    # Run options
+    reanalyse_existing=True,  # if True, existing colocated data files will be deleted
+    raise_exceptions=True,  # if True, the analysis will stop whenever an error occurs
+    clear_existing_json=False,  # if True, deletes previous output before running
+    # Map Options
+    add_model_maps=False,  # Adds a plot of the whole map. Very slow!!!
+    only_model_maps=False,  # Adds only plot above, without any other evaluation
+    filter_name="ALL-wMOUNTAINS",  # Regional filter for analysis
+    map_zoom="Europe",  # Zoom level. For EMEP, Europe is typically used
+    regions_how="country",  # Calculates statistics for different regions. Typically "country" is used, but that does not work for satellite data
+    # Time and Frequency Options
+    ts_type="monthly",  # Colocation frequency (no statistics in higher resolution can be computed)
+    freqs=["monthly", "yearly"],  # Frequencies that are evaluated
+    main_freq="monthly",  # Frequency that is displayed when opening webpage
+    periods=[
+        "2000-2010"
+    ],  # List of years or periods of years that are evaluated. E.g. "2005" or "2001-2020"
+    # Statistical Options
+    min_num_obs=DEFAULT_RESAMPLE_CONSTRAINTS,
+    obs_remove_outliers=False,
+    model_remove_outliers=False,
+    colocate_time=True,
+    zeros_to_nan=False,
+    weighted_stats=True,
+    annual_stats_constrained=True,
+    harmonise_units=True,
+    resample_how={
+        "vmro3max": {"daily": {"hourly": "max"}}
+    },  # How to handle Ozone. Used all the time in EMEP
+    # Experiment Metadata
+    exp_pi=exp_pi,
+    proj_id=proj_id,
+    exp_id=exp_id,
+    exp_name="Evaluation of EMEP data, with trends",
+    exp_descr=("Evaluation of EMEP data, with trends"),
+    public=True,
+)
+
+
+BASE_FILTER = {
+    "latitude": [30, 82],
+    "longitude": [-30, 90],
+    "altitude": [-200, 5000],
+}
+
+EBAS_FILTER = {
+    **BASE_FILTER,
+    "data_level": [None, 2],
+    "set_flags_nan": True,
+}
+
+CFG["obs_cfg"] = {
+    "EBAS-tc": dict(
+        obs_id="EBASMC",
+        web_interface_name="EBAS",
+        obs_vars=[
+            "concpm10",
+            "concpm25",
+        ],
+        obs_vert_type="Surface",
+        colocate_time=True,
+        ts_type="monthly",
+        obs_filters=EBAS_FILTER,
+    )
+}
+
+
+# Dir where emep data is found
+folder_EMEP = f"/lustre/storeB/project/fou/kl/emep/ModelRuns/2021_REPORTING/TRENDS/"
+
+# Setup for models used in analysis
+CFG["model_cfg"] = {
+    "EMEP": dict(
+        model_id="EMEP",
+        model_data_dir=folder_EMEP,
+        gridded_reader_id={"model": "ReadMscwCtm"},
+        model_use_vars={"vmro3": "vmro3max"},
+        model_rename_vars={"vmro3max": "vmro3"},
+        model_ts_type_read="daily",
+    ),
+}
+
+
+CFG.update(
+    dict(
+        add_trends=True,
+        avg_over_trends=True,
+        obs_min_yrs=7,
+        stats_min_yrs=7,
+        sequential_yrs=False,
+    )
+)
+
+
+if __name__ == "__main__":
+    from pyaerocom.aeroval import EvalSetup, ExperimentProcessor
+    from pyaerocom import const
+
+    print(
+        const.CACHEDIR
+    )  # Prints where to find the caching folder. Not needed but this folder should be emptied now and then, so I like to see where it is
+
+    stp = EvalSetup(**CFG)  # Makes a setup object from the dict, that PyAeroval can use
+    ana = ExperimentProcessor(stp)  # Makes an experiment object
+    res = ana.run()  # Runs the experiment