Has methane increased near this oil/gas infrastructure?

What you can answer

• Whether NASA detected a methane plume over a specific lat/lon during a date window.
• Plume source strength (kg/hr) from L2B Methane Enhancement data inverted with PBL height + wind.
• Repeat detection rate at a given facility (how often EMIT overhead caught a plume).
• Whether the plume cluster has changed in size or frequency over the operational period.

What you can NOT answer with these alone

• 24/7 continuous monitoring — EMIT revisit at a given point is ~16 days, sometimes longer; you see snapshots, not surveillance.
• Sub-60m attribution to a specific well/tank when multiple are co-located within an EMIT pixel.
• Total facility emissions per year without auxiliary models filling EMIT data gaps.

Code template

import earthaccess
import geopandas as gpd
from shapely.geometry import box

earthaccess.login(strategy="netrc")

# 1. Define facility location + buffer
facility = (32.65, -103.05)  # example Permian Basin coords
buffer_deg = 0.05  # ~5 km
aoi = (facility[1]-buffer_deg, facility[0]-buffer_deg,
       facility[1]+buffer_deg, facility[0]+buffer_deg)

# 2. Search EMIT L2B methane plume product
plumes = earthaccess.search_data(
    short_name="EMITL2BCH4PLM",  # confirm short_name via CMR
    bounding_box=aoi,
    temporal=("2022-08-01", "2026-05-01"),
)

# 3. Open as COG, mask to AOI, compute total plume mass per detection
# 4. Pull MERRA-2 PBL height + 10m wind for the detection timestamps to invert source rate
# 5. Tabulate: date · plume mass · estimated source rate (kg/hr)
# 6. Plot: time-series of source rate at facility; map of plume polygons

Expected output

• Time-series chart: estimated source rate (kg/hr) at the facility across EMIT detections
• Map: overlaid plume polygons colored by detection date
• Histogram: source-rate distribution

Caveats

• EMIT is hyperspectral imagery, not a methane sensor per se — the L2B methane product is derived from radiance via matched filter; false positives exist (look for the QA flag).
• Sub-pixel mixing of multiple plume sources within 60m is common in dense oilfield basins.
• EMIT does not see at night (it’s a passive optical instrument).
• The plume vector product (L2B PLM) lags the raw enhancement product by weeks; use L2B Enhancement (L2BCH4ENH) for the most recent data.
• UNEP-IMEO MARS workflow validates EMIT detections manually before public release of plume polygons; if you’re using the public plume catalog you’re already getting QA’d data, but if you’re processing the enhancement product yourself, you’re not.

Cross-DAAC composition

This is a 2-DAAC + 1-external join: LP DAAC (EMIT) + GES DISC (MERRA-2) + TROPOMI (external, ESA but mirrored via Earthdata Search). Auth uniform via Earthdata Login. See recipes/r02-emit-merra2-fusion.mdx.

Sources

• EMIT L2B products: https://lpdaac.usgs.gov/products/emitl2bch4plmv001/
• UNEP-IMEO MARS dataset on Hugging Face: https://huggingface.co/guides/UNEP-IMEO/MARS-Hyperspectral
• SpaceML plume-hunter: https://github.com/spaceml-org/methane-detection
• EMIT science: https://earth.jpl.nasa.gov/emit/