FIRMS — fire-season time-series¶

Count daily fire pixels and total daily FRP over a fire season for one region, then plot the activity curve. Shows the backend chunking a multi-week window into ≤10-day requests transparently.

Setup¶

The imports plus an output directory for the downloaded detections. FIRMS is a vector backend, so download() writes the active-fire table to OUT_DIR and returns it for plotting.

In [ ]:

import datetime as dt
import os
from pathlib import Path

from earthlens import EarthLens

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

import datetime as dt import os from pathlib import Path from earthlens import EarthLens OUT_DIR = Path('firms_output') OUT_DIR.mkdir(exist_ok=True)

Credentials¶

FIRMS needs a free MAP_KEY (https://firms.modaps.eosdis.nasa.gov/api/map_key/). Set FIRMS_MAP_KEY in your environment; the live cells below skip cleanly (nbval-lax-safe) when it is absent, so the notebook never fails offline.

In [ ]:

HAS_KEY = bool(os.environ.get('FIRMS_MAP_KEY'))
print('FIRMS_MAP_KEY set:', HAS_KEY)

HAS_KEY = bool(os.environ.get('FIRMS_MAP_KEY')) print('FIRMS_MAP_KEY set:', HAS_KEY)

Query a fire-season window¶

A several-week window is split into ≤10-day chunks internally and returned merged. Use a recent window so the NRT sensor has coverage — here the last 30 days.

In [ ]:

TODAY = dt.date.today()
START = (TODAY - dt.timedelta(days=30)).strftime('%Y-%m-%d')
END = TODAY.strftime('%Y-%m-%d')

TODAY = dt.date.today() START = (TODAY - dt.timedelta(days=30)).strftime('%Y-%m-%d') END = TODAY.strftime('%Y-%m-%d')

Download the detections¶

The request is built first, then download() is called on its own line so each step reads clearly. The whole fetch is wrapped in a guard so the notebook skips gracefully when the key is missing or the live query fails.

In [ ]:

fires = None
if HAS_KEY:
    try:
        firms = EarthLens(
            data_source='firms',
            variables=['VIIRS_SNPP_NRT'],
            start=START,
            end=END,
            aoi=[-124.0, 36.0, -118.0, 42.0],  # Northern California / Pacific NW
            path=str(OUT_DIR),
        )
        fires = firms.download(progress_bar=False)
        print('detections:', len(fires))
    except Exception as exc:
        print('skipped live query:', exc)

fires = None if HAS_KEY: try: firms = EarthLens( data_source='firms', variables=['VIIRS_SNPP_NRT'], start=START, end=END, aoi=[-124.0, 36.0, -118.0, 42.0], # Northern California / Pacific NW path=str(OUT_DIR), ) fires = firms.download(progress_bar=False) print('detections:', len(fires)) except Exception as exc: print('skipped live query:', exc)

Daily fire activity¶

Group the detections by acquisition date to get a daily fire-pixel count and the total daily FRP. These two series drive the activity plot.

In [ ]:

daily = None
if fires is not None and len(fires):
    daily = fires.assign(date=fires['acq_datetime'].dt.date).groupby('date')
    counts = daily.size()
    total_frp = daily['frp'].sum()

daily = None if fires is not None and len(fires): daily = fires.assign(date=fires['acq_datetime'].dt.date).groupby('date') counts = daily.size() total_frp = daily['frp'].sum()

Plot the activity curve¶

Bars show the daily fire-pixel count (left axis) and the line shows total FRP in MW (right axis); the busiest day is annotated as the peak.

In [ ]:

if daily is not None:
    import matplotlib.pyplot as plt

    fig, ax1 = plt.subplots(figsize=(9, 4))
    ax1.bar(counts.index, counts.values, color='tab:orange', alpha=0.6)
    ax1.set_ylabel('fire-pixel count', color='tab:orange')
    ax1.set_xlabel('date')
    ax2 = ax1.twinx()
    ax2.plot(total_frp.index, total_frp.values, color='tab:red', marker='o')
    ax2.set_ylabel('total FRP (MW)', color='tab:red')
    ax1.set_title(f'Daily fire activity, {START} to {END}')
    peak = counts.idxmax()
    ax1.annotate(
        f'peak: {peak}',
        xy=(peak, counts.max()),
        xytext=(0.6, 0.85),
        textcoords='axes fraction',
        arrowprops=dict(arrowstyle='->'),
    )
    fig.tight_layout()
    plt.show()

if daily is not None: import matplotlib.pyplot as plt fig, ax1 = plt.subplots(figsize=(9, 4)) ax1.bar(counts.index, counts.values, color='tab:orange', alpha=0.6) ax1.set_ylabel('fire-pixel count', color='tab:orange') ax1.set_xlabel('date') ax2 = ax1.twinx() ax2.plot(total_frp.index, total_frp.values, color='tab:red', marker='o') ax2.set_ylabel('total FRP (MW)', color='tab:red') ax1.set_title(f'Daily fire activity, {START} to {END}') peak = counts.idxmax() ax1.annotate( f'peak: {peak}', xy=(peak, counts.max()), xytext=(0.6, 0.85), textcoords='axes fraction', arrowprops=dict(arrowstyle='->'), ) fig.tight_layout() plt.show()