GDACS global hazard dashboard¶

A 90-day snapshot of every GDACS hazard type worldwide: a faceted world map (one panel per hazard) coloured by alert level, plus a table of the current Red alerts.

GDACS is public — no credentials, no extra to install. download() returns a pyramids FeatureCollection (a geopandas.GeoDataFrame).

Setup¶

First the imports. matplotlib and pandas are used by the plotting and summary cells; earthlens provides the unified EarthLens entry point. We also create the output directory the backend writes into.

In [ ]:

import datetime as dt
from pathlib import Path

import matplotlib.pyplot as plt
import pandas as pd

from earthlens import EarthLens

OUT_DIR = Path('gdacs_output')
OUT_DIR.mkdir(exist_ok=True)

import datetime as dt from pathlib import Path import matplotlib.pyplot as plt import pandas as pd from earthlens import EarthLens OUT_DIR = Path('gdacs_output') OUT_DIR.mkdir(exist_ok=True)

The request window¶

We look back 90 days from today and list the six GDACS hazard codes we want, with friendly names for the map titles.

In [ ]:

TODAY = dt.date.today()
START = (TODAY - dt.timedelta(days=90)).strftime('%Y-%m-%d')
END = TODAY.strftime('%Y-%m-%d')

HAZARDS = ['EQ', 'TC', 'FL', 'VO', 'WF', 'DR']
HAZARD_NAMES = {
    'EQ': 'Earthquake',
    'TC': 'Tropical cyclone',
    'FL': 'Flood',
    'VO': 'Volcano',
    'WF': 'Wildfire',
    'DR': 'Drought',
}
print('window:', START, 'to', END)

TODAY = dt.date.today() START = (TODAY - dt.timedelta(days=90)).strftime('%Y-%m-%d') END = TODAY.strftime('%Y-%m-%d') HAZARDS = ['EQ', 'TC', 'FL', 'VO', 'WF', 'DR'] HAZARD_NAMES = { 'EQ': 'Earthquake', 'TC': 'Tropical cyclone', 'FL': 'Flood', 'VO': 'Volcano', 'WF': 'Wildfire', 'DR': 'Drought', } print('window:', START, 'to', END)

Pull every hazard for the window¶

A single request returns all six hazard types. We build the request first, then call download() on its own line so each step is easy to read and re-run. The call is wrapped so the notebook stays safe under nbval-lax when run offline.

In [ ]:

alerts = None
try:
    dashboard = EarthLens(
        variables=HAZARDS,
        data_source='gdacs',
        start=START,
        end=END,
        aoi=[-180, -90, 180, 90],
        alert_level=['Green', 'Orange', 'Red'],
        path=str(OUT_DIR),
    )
    alerts = dashboard.download()
    print(f'{len(alerts)} alerts')
    print(alerts['hazard_type'].value_counts().to_dict())
except Exception as exc:
    print(f'live GDACS query skipped (offline?): {type(exc).__name__}: {exc}')

alerts = None try: dashboard = EarthLens( variables=HAZARDS, data_source='gdacs', start=START, end=END, aoi=[-180, -90, 180, 90], alert_level=['Green', 'Orange', 'Red'], path=str(OUT_DIR), ) alerts = dashboard.download() print(f'{len(alerts)} alerts') print(alerts['hazard_type'].value_counts().to_dict()) except Exception as exc: print(f'live GDACS query skipped (offline?): {type(exc).__name__}: {exc}')

Faceted world map — one panel per hazard¶

Each panel shows that hazard's alerts coloured by level (green / orange / red).

In [ ]:

if alerts is not None and len(alerts):
    colours = {'Green': 'green', 'Orange': 'orange', 'Red': 'red'}
    fig, axes = plt.subplots(2, 3, figsize=(15, 7), sharex=True, sharey=True)
    for ax, code in zip(axes.ravel(), HAZARDS):
        subset = alerts[alerts['hazard_type'] == code]
        ax.set_title(f'{HAZARD_NAMES[code]} ({len(subset)})')
        ax.set_xlim(-180, 180)
        ax.set_ylim(-90, 90)
        if len(subset):
            subset.plot(
                ax=ax,
                color=subset['alert_level'].map(colours).fillna('grey'),
                markersize=18,
                alpha=0.6,
            )
    fig.suptitle(f'GDACS alerts by hazard, {START} to {END}')
    fig.tight_layout()
    plt.show()

if alerts is not None and len(alerts): colours = {'Green': 'green', 'Orange': 'orange', 'Red': 'red'} fig, axes = plt.subplots(2, 3, figsize=(15, 7), sharex=True, sharey=True) for ax, code in zip(axes.ravel(), HAZARDS): subset = alerts[alerts['hazard_type'] == code] ax.set_title(f'{HAZARD_NAMES[code]} ({len(subset)})') ax.set_xlim(-180, 180) ax.set_ylim(-90, 90) if len(subset): subset.plot( ax=ax, color=subset['alert_level'].map(colours).fillna('grey'), markersize=18, alpha=0.6, ) fig.suptitle(f'GDACS alerts by hazard, {START} to {END}') fig.tight_layout() plt.show()

The current Red alerts¶

Red is GDACS's highest impact level. These are the alerts most likely to need a coordinated response.

In [ ]:

if alerts is not None and len(alerts):
    red = alerts[alerts['alert_level'] == 'Red'].sort_values(
        'from_date', ascending=False
    )
    print(f'{len(red)} Red alerts in the window')
    cols = ['from_date', 'hazard_type', 'name', 'country', 'alert_score', 'glide']
    display(red[cols].head(20))

if alerts is not None and len(alerts): red = alerts[alerts['alert_level'] == 'Red'].sort_values( 'from_date', ascending=False ) print(f'{len(red)} Red alerts in the window') cols = ['from_date', 'hazard_type', 'name', 'country', 'alert_score', 'glide'] display(red[cols].head(20))

Alert-level breakdown per hazard¶

A compact cross-tab of how many alerts of each level each hazard produced.

In [ ]:

if alerts is not None and len(alerts):
    crosstab = pd.crosstab(alerts['hazard_type'], alerts['alert_level'])
    for level in ['Green', 'Orange', 'Red']:
        if level not in crosstab.columns:
            crosstab[level] = 0
    display(crosstab[['Green', 'Orange', 'Red']])

if alerts is not None and len(alerts): crosstab = pd.crosstab(alerts['hazard_type'], alerts['alert_level']) for level in ['Green', 'Orange', 'Red']: if level not in crosstab.columns: crosstab[level] = 0 display(crosstab[['Green', 'Orange', 'Red']])