Merge branch 'master' of https://gitlab.dkrz.de/mipdata/tutorials-and-use-cases

8f343de0 · Fabian Wachsmann · 37b70176 · 28f77f87 · 8f343de0
Commit 8f343de0 authored 4 years ago by Fabian Wachsmann
--- a/notebooks/use-case_count_summer_days.ipynb
+++ b/notebooks/use-case_count_summer_days.ipynb
@@ -11,7 +11,7 @@
   "cell_type": "markdown",
   "metadata": {},
   "source": [
-    "In this Use Case you will learn how to access a data file from DKRZ's CMIP6 archive and calculate the annual summer of summer days for a particular location.\\\n",
+    "In this Use Case you will learn how to access a data file from DKRZ's CMIP6 archive and count the annual number of summer days for a particular location.\\\n",
    "\\\n",
    "You will be using:\n",
    "- [Intake](https://github.com/intake/intake) for finding the data the data\n",
@@ -26,22 +26,14 @@
   "outputs": [],
   "source": [
    "# get formating done automatically according to style `black`\n",
-    "%load_ext lab_black"
+    "#%load_ext lab_black"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": []
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
   "source": [
    "import intake\n",
    "import requests\n",
@@ -61,43 +53,16 @@
   "metadata": {},
   "source": [
    "## Choose Location and Year\n",
-    "If ambiguous the more likely location will be chosen"
+    "If ambiguous the more likely location will be chosen\n",
+    "\n",
+    "<a id='selection'></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "2bf9b0dc98b44051aeca85e44b792fa6",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Text(value='', description='Enter place:')"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "82413fc5564e47e1bd958028b2fc4e19",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Dropdown(description='Select year: ', options=(1850, 1851, 1852, 1853, 1854, 1855, 1856, 1857, 1858, 1859, 186…"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
   "source": [
    "place = widgets.Text(description=\"Enter place:\")\n",
    "display(place)\n",
@@ -249,10 +214,12 @@
   ]
  },
  {
-   "cell_type": "markdown",
+   "cell_type": "code",
+   "execution_count": null,
   "metadata": {},
+   "outputs": [],
   "source": [
-    "## Compare Model Grid Cell with chosen Location"
+    "selected_path"
   ]
  },
  {
@@ -261,21 +228,14 @@
   "metadata": {},
   "outputs": [],
   "source": [
-    "# Find nearest model coordinate\n",
-    "\n",
-    "# First, find the index of the grid point nearest a specific lat/lon.\n",
-    "abslat = np.abs(tasmax_year_xr[\"lat\"] - location.latitude)\n",
-    "abslon = np.abs(tasmax_year_xr[\"lon\"] - location.longitude)\n",
-    "c = np.maximum(abslon, abslat)\n",
-    "\n",
-    "([xloc], [yloc]) = np.where(c == np.min(c))"
+    "tasmax_xr"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
-    "Draw Map"
+    "## Compare Model Grid Cell with chosen Location"
   ]
  },
  {
@@ -284,6 +244,16 @@
   "metadata": {},
   "outputs": [],
   "source": [
+    "# Find nearest model coordinate\n",
+    "\n",
+    "# First, find the index of the grid point nearest a specific lat/lon.\n",
+    "abslat = np.abs(tasmax_year_xr[\"lat\"] - location.latitude)\n",
+    "abslon = np.abs(tasmax_year_xr[\"lon\"] - location.longitude)\n",
+    "c = np.maximum(abslon, abslat)\n",
+    "\n",
+    "([xloc], [yloc]) = np.where(c == np.min(c))\n",
+    "\n",
+    "# draw map\n",
    "m = folium.Map(location=[location.latitude, location.longitude], zoom_start=8)\n",
    "tooltip = location.latitude, location.longitude\n",
    "folium.Marker(\n",
@@ -314,6 +284,13 @@
    "m"
   ]
  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Draw Map"
+   ]
+  },
  {
   "cell_type": "markdown",
   "metadata": {},
@@ -333,7 +310,8 @@
    "tasmax_year_place_df.loc[:, 'Actual Temperature'] = tasmax_year_place_xr.values\n",
    "tasmax_year_place_df.loc[:, 'Summer Day Threshold'] = 25\n",
    "\n",
-    "#test"
+    "tasmax_year_place_df.hvplot.line(y=['Actual Temperature', 'Summer Day Threshold'], \n",
+    "                value_label='Temperature in °C', legend='bottom', title='Daily maximum Temperature near Surface for ' +place.value, height=500, width=620)"
   ]
  },
  {
@@ -342,18 +320,15 @@
   "metadata": {},
   "outputs": [],
   "source": [
-    "tasmax_year_place_df.hvplot.line(y=['Actual Temperature', 'Summer Day Threshold'], \n",
-    "                value_label='Temperature in °C', legend='bottom', title='Daily maximum Temperature near Surface for ' +place.value, height=500, width=620)"
+    "no_summer_days = tasmax_year_place_xr[tasmax_year_place_xr > 25].size\n",
+    "print(\"In \" +str(year.value) +\" \" +place.value +\" had \" + str(no_summer_days)+\" summer days.\")"
   ]
  },
  {
-   "cell_type": "code",
-   "execution_count": null,
+   "cell_type": "markdown",
   "metadata": {},
-   "outputs": [],
   "source": [
-    "no_summer_days = tasmax_year_place_xr[tasmax_year_place_xr > 25].size\n",
-    "print(\"In \" +str(year.value) +\" \" +place.value +\" had \" + str(no_summer_days)+\" summer days.\")"
+    "[Try another location and year](#selection)"
   ]
  }
 ],

 %% Cell type:markdown id: tags:

 # Count Annual Summer Days for a particular Location

 %% Cell type:markdown id: tags:

-In this Use Case you will learn how to access a data file from DKRZ's CMIP6 archive and calculate the annual summer of summer days for a particular location.\
+In this Use Case you will learn how to access a data file from DKRZ's CMIP6 archive and count the annual number of summer days for a particular location.\
 \
 You will be using:
 - [Intake](https://github.com/intake/intake) for finding the data the data
 - [Xarray](http://xarray.pydata.org/en/stable/) for loading and processing the data
 - [hvPlot](https://hvplot.holoviz.org/index.html) for visualizing the data

 %% Cell type:code id: tags:

 ``` python
 # get formating done automatically according to style `black`
-%load_ext lab_black
+#%load_ext lab_black
 ```

 %% Cell type:code id: tags:

 ``` python
 import intake
 import requests
 import folium
 import xarray as xr
 import matplotlib
 from IPython.display import display
 from ipywidgets import widgets
 from geopy.geocoders import Nominatim
 import numpy as np
 import pandas as pd
 import hvplot.pandas
 ```

-%% Output
-
-
 %% Cell type:markdown id: tags:

 ## Choose Location and Year
 If ambiguous the more likely location will be chosen

+<a id='selection'></a>
+
 %% Cell type:code id: tags:

 ``` python
 place = widgets.Text(description="Enter place:")
 display(place)

 x = range(1850, 2014)
 year = widgets.Dropdown(
    options=x,
    description="Select year: ",
    disabled=False,
 )
 display(year)
 ```

-%% Output
-
-
-
 %% Cell type:markdown id: tags:

 ### Check if your Selection is correct

 %% Cell type:code id: tags:

 ``` python
 geolocator = Nominatim(user_agent="any_agent")
 location = geolocator.geocode(place.value)

 m = folium.Map(location=[location.latitude, location.longitude])
 tooltip = location.latitude, location.longitude
 folium.Marker([location.latitude, location.longitude], tooltip=tooltip).add_to(m)
 display(m)

 print(location.address)
 print((location.latitude, location.longitude))
 print("Chosen year: " + str(year.value))
 ```

 %% Cell type:markdown id: tags:

 ## Load Intake Catalogue
 The catalog files are uploaded to the cloud. There is a gsv.gz file and a json file under
 https://swiftbrowser.dkrz.de/public/dkrz_a44962e3ba914c309a7421573a6949a6/intake-esm/
 We load the catalog descriptor with intake:

 %% Cell type:code id: tags:

 ``` python
 # col_url = "https://swift.dkrz.de/v1/dkrz_a44962e3ba914c309a7421573a6949a6/intake-esm/mistral-cmip6.json"
 # or
 col_url = "/work/ik1017/Catalogs/mistral-cmip6.json"

 col = intake.open_esm_datastore(col_url)
 ```

 %% Cell type:markdown id: tags:

 Let's see what is in the intake catalog. The underlying data base is given as a panda dataframe which we can access with 'col.df'. col.df.head() shows us the first rows of the table:

 %% Cell type:code id: tags:

 ``` python
 col.df.head()
 ```

 %% Cell type:markdown id: tags:

 ## Browse Intake Catalogue and open Variable

 %% Cell type:code id: tags:

 ``` python
 query = dict(
    source_id="MPI-ESM1-2-HR",
    variable_id="tasmax",
    table_id="day",
    experiment_id="historical",
    member_id="r10i1p1f1",
 )
 cat = col.search(**query)
 ```

 %% Cell type:markdown id: tags:

 show query

 %% Cell type:code id: tags:

 ``` python
 cat.df
 ```

 %% Cell type:markdown id: tags:

 open selected year

 %% Cell type:code id: tags:

 ``` python
 query_result_df = cat.df
 query_result_df["start_year"] = query_result_df["time_range"].str[0:4].astype(int)
 query_result_df["end_year"] = query_result_df["time_range"].str[9:13].astype(int)
 query_result_df["selection"] = (year.value - query_result_df["start_year"] >= 0) & (
    year.value - query_result_df["end_year"] <= 0
 )
 query_result_df["path"].where(query_result_df["selection"] == True)
 selected_path_index = query_result_df.loc[query_result_df["selection"] == True][
    "path"
 ].index[0]
 selected_path = query_result_df["path"][selected_path_index]

 ds_tasmax = xr.open_dataset(selected_path)
 tasmax_xr = ds_tasmax["tasmax"]

 time_start = str(year.value) + "-01-01T12:00:00.000000000"
 time_end = str(year.value) + "-12-31T12:00:00.000000000"
 tasmax_year_xr = tasmax_xr.loc[time_start:time_end, :, :]
 ```

+%% Cell type:code id: tags:
+
+``` python
+selected_path
+```
+
+%% Cell type:code id: tags:
+
+``` python
+tasmax_xr
+```
+
 %% Cell type:markdown id: tags:

 ## Compare Model Grid Cell with chosen Location

 %% Cell type:code id: tags:

 ``` python
 # Find nearest model coordinate

 # First, find the index of the grid point nearest a specific lat/lon.
 abslat = np.abs(tasmax_year_xr["lat"] - location.latitude)
 abslon = np.abs(tasmax_year_xr["lon"] - location.longitude)
 c = np.maximum(abslon, abslat)

 ([xloc], [yloc]) = np.where(c == np.min(c))
-```
-
-%% Cell type:markdown id: tags:
-
-Draw Map
-
-%% Cell type:code id: tags:

-``` python
+# draw map
 m = folium.Map(location=[location.latitude, location.longitude], zoom_start=8)
 tooltip = location.latitude, location.longitude
 folium.Marker(
    [tasmax_year_xr["lat"][yloc], tasmax_year_xr["lon"][xloc]],
    tooltip=tooltip,
    popup="Model Grid Cell Center",
 ).add_to(m)
 folium.Marker(
    [location.latitude, location.longitude],
    tooltip=tooltip,
    popup="Location selected by You",
 ).add_to(m)

 # Define rectangle
 rect_lat1 = (tasmax_year_xr["lat"][yloc - 1] + tasmax_year_xr["lat"][yloc]) / 2
 rect_lon1 = (tasmax_year_xr["lon"][xloc - 1] + tasmax_year_xr["lon"][xloc]) / 2
 rect_lat2 = (tasmax_year_xr["lat"][yloc + 1] + tasmax_year_xr["lat"][yloc]) / 2
 rect_lon2 = (tasmax_year_xr["lon"][xloc + 1] + tasmax_year_xr["lon"][xloc]) / 2
 #

 folium.Rectangle(
    bounds=[[rect_lat1, rect_lon1], [rect_lat2, rect_lon2]],
    color="#ff7800",
    fill=True,
    fill_color="#ffff00",
    fill_opacity=0.2,
 ).add_to(m)
 m
 ```

 %% Cell type:markdown id: tags:

+Draw Map
+
+%% Cell type:markdown id: tags:
+
 ## Draw Temperature Time Series and count Summer days

 %% Cell type:code id: tags:

 ``` python
 tasmax_year_place_xr = tasmax_year_xr[:, yloc, xloc] - 273.15
 tasmax_year_place_df = pd.DataFrame(index = tasmax_year_place_xr['time'].values, columns = ['Temperature', 'Summer Day Threshold'])
 #
 tasmax_year_place_df.loc[:, 'Actual Temperature'] = tasmax_year_place_xr.values
 tasmax_year_place_df.loc[:, 'Summer Day Threshold'] = 25

-#test
-```
-
-%% Cell type:code id: tags:
-
-``` python
 tasmax_year_place_df.hvplot.line(y=['Actual Temperature', 'Summer Day Threshold'],
                value_label='Temperature in °C', legend='bottom', title='Daily maximum Temperature near Surface for ' +place.value, height=500, width=620)
 ```

 %% Cell type:code id: tags:

 ``` python
 no_summer_days = tasmax_year_place_xr[tasmax_year_place_xr > 25].size
 print("In " +str(year.value) +" " +place.value +" had " + str(no_summer_days)+" summer days.")
 ```
+
+%% Cell type:markdown id: tags:
+
+[Try another location and year](#selection)